Guest

Cisco IOS Software Releases 12.1 E

IOS Server Load Balancing, 12.1(8a)E

Table Of Contents

IOS Server Load Balancing

Overview of the IOS SLB Feature

Functions and Capabilities

Algorithms for Server Load Balancing

Weighted Round Robin

Weighted Least Connections

Alternate IP Addresses

Automatic Server Failure Detection

Automatic Unfail

Avoiding Attacks on Server Farms and Firewall Farms

Backup Server Farms

Client-Assigned Load Balancing

Content Flow Monitor Support

Delayed Removal of TCP Connection Context

Dynamic Feedback Protocol for IOS SLB

Firewall Load Balancing

Maximum Connections

Multiple Firewall Farm Support

Network Address Translation (NAT) and Session Redirection

Port-Bound Servers

Probes

Probes in Server Load Balancing

Probes in Firewall Load Balancing

Protocol Support

Redundancy Enhancements

Route Health Injection

Slow Start

Sticky Connections

SynGuard

TCP Session Reassignment

Transparent Webcache Load Balancing

WAP Load Balancing

Benefits

Restrictions

Supported Platforms

Supported Standards, MIBs, and RFCs

Required Configuration Tasks

Configuring the Server Farms

Verifying the Server Farms

Configuring the Virtual Servers

Verifying the Virtual Servers

Configuring the Restricted Clients

Verifying the Restricted Clients

Verifying IOS SLB Connectivity

Optional Configuration Tasks

Specifying a Server Load Balancing Algorithm

Specifying a Bind ID

Configuring Real Server Attributes

Adjusting Virtual Server Values

Configuring IOS SLB Firewall Load Balancing

Configuring One or More Probes

Configuring the Firewall Farm

Verifying the Firewall Farm

Verifying Firewall Connectivity

Configuring HTTP Probes

Configuring Ping Probes

Configuring WSP Probes

Configuring IOS SLB Dynamic Feedback Protocol

Configuring IOS SLB as a DFP Manager

Configuring a Client Subsystem as a DFP Agent

Configuring IOS SLB NAT

Implementing IOS SLB Stateless Backup

How IOS SLB Stateless Backup Works

Configuring IOS SLB Stateless Backup

Enabling HSRP

Customizing Group Attributes

Verifying the IOS SLB Stateless Backup Configuration

Sample IOS SLB Stateless Backup Configuration

Configuring IOS SLB Stateful Backup

Configuring Stateful Backup for Server Load Balancing

Configuring Stateful Backup for Firewall Load Balancing

Monitoring and Maintaining the IOS SLB Feature

Configuration Examples

Complete Example Configuration

Example of a Layer 3 Switch with ISL, VLAN, and BVI with GEC

Example of IOS SLB with Firewall Load Balancing

Internal Firewall Load-Balancing Device

External Firewall Load-Balancing Device

Example of IOS SLB with Server Load Balancing and Firewall Load Balancing

Internal Server and Firewall Load-Balancing Device

External Firewall Load-Balancing Device

Example of IOS SLB with Multiple Firewall Farms

Internal Firewall Load-Balancing Device

External Firewall Load-Balancing Device

Example of IOS SLB with Probes

Example of a Layer 3 Switch Configured with IOS SLB

Example of IOS SLB with NAT

Switch A Configuration Statements

Switch B Configuration Statements

Switch C Configuration Statements

Example of an IOS Layer 3 Switch with HSRP

Device A (Active) Configuration Statements

Device B (Standby) Configuration Statements

Description of Configuration

Examples of IOS SLB with Stateless Backup

Example with Dynamic Routing and Trunking

Example with Dynamic Routing and No Trunking

Example with Static Routing and Trunking

Example with Static Routing and No Trunking

Example of IOS SLB with Stateful Backup

Switch SLB1 Configuration Statements

Switch SLB2 Configuration Statements

Example of IOS SLB with Active Standby

SLB 1 Configuration Statements

SLB 2 Configuration Statements

Access Router Configuration Statements

Example of IOS SLB with Redistribution of Static Routes

Routing Information Protocol (RIP)

Open Shortest Path First (OSPF)

Interior Gateway Routing Protocol (IGRP)

Enhanced Interior Gateway Routing Protocol (Enhanced IGRP)

Examples of IOS SLB with WAP Load Balancing

Example with WAP Load Balancing

Example with UDP Load Balancing

Examples of IOS SLB with Route Health Injection

Example with Two Distributed Sites with One Web Server Each

Example with Two Distributed Sites with Two Web Servers Each

Example with Two Distributed Sites with One Web Server and a Backup IOS SLB Switch Each

Example of IOS SLB with Sticky Connections

Command Reference

access

address (HTTP probe)

address (ping probe)

address (WSP probe)

advertise

agent

bindid

clear ip slb

client (virtual server)

credentials

delay (firewall farm TCP protocol)

delay (virtual server)

expect

faildetect (ping probe)

faildetect (real server)

header

idle (firewall farm TCP protocol)

idle (firewall farm UDP protocol)

idle (virtual server)

inservice (DFP agent)

inservice (firewall farm)

inservice (firewall farm real server)

inservice (server farm real server)

inservice (server farm virtual server)

interval (DFP agent)

interval (HTTP probe)

interval (ping probe)

interval (WSP probe)

ip dfp agent

ip slb dfp

ip slb entries

ip slb firewallfarm

ip slb natpool

ip slb probe (HTTP probe)

ip slb probe (ping probe)

ip slb probe (WSP probe)

ip slb serverfarm

ip slb vserver

manager

maxconns (firewall farm TCP protocol)

maxconns (firewall farm UDP protocol)

maxconns (server farm)

mls aging slb normal

mls aging slb process

mls ip slb search wildcard

nat

password

port (DFP agent)

port (HTTP probe)

predictor (server farm)

predictor hash address (firewall farm)

probe (firewall farm real server)

probe (server farm)

real (firewall farm)

real (server farm)

reassign

replicate casa (firewall farm)

replicate casa (virtual server)

request method, request url

retry

serverfarm

show ip dfp

show ip slb conns

show ip slb dfp

show ip slb firewallfarm

show ip slb natpool

show ip slb probe

show ip slb reals

show ip slb replicate

show ip slb serverfarms

show ip slb stats

show ip slb sticky

show ip slb vserver

standby authentication

standby name

standby priority, standby preempt

standby timers

standby track

sticky (firewall farm TCP protocol)

sticky (firewall farm UDP protocol)

sticky (virtual server)

synguard (virtual server)

tcp

udp

url (WSP probe)

virtual (virtual server)

weight (firewall farm real server)

weight (server farm)

Debug Commands

debug ip dfp agent

debug ip slb

FAQ (Frequently Asked Questions)

Glossary


IOS Server Load Balancing


Feature History

Release
Modification

12.0(7)XE

This feature was introduced with support for the following platforms:

Catalyst 6000 Family Switches with Supervisor Engine 1

Cisco 7200 Series Routers

The following functions were provided:

Algorithms for Server Load Balancing

Automatic Server Failure Detection

Automatic Unfail

Client-Assigned Load Balancing

Delayed Removal of TCP Connection Context

Dynamic Feedback Protocol for IOS SLB

Maximum Connections

Port-Bound Servers

Slow Start

Sticky Connections

SynGuard

TCP Session Reassignment

12.1(1)E

The following functions were added:

Alternate IP Addresses

Content Flow Monitor Support

Network Address Translation (NAT) and Session Redirection—Server NAT

Redundancy Enhancements—Stateless Backup

Transparent Webcache Load Balancing

12.1(2)E

The following functions were added:

Probes—HTTP Probes

Network Address Translation (NAT) and Session Redirection—Server and Client NAT

Redundancy Enhancements—Stateless and Stateful Backup

12.1(3a)E

The following functions were added:

Firewall Load Balancing

Probes—HTTP and Ping Probes

Protocol Support

Redundancy Enhancements—Stateless and Stateful Backup, and Active Standby

WAP Load Balancing

12.1(5a)E

The following functions were added:

Avoiding Attacks on Server Farms and Firewall Farms

Probes—HTTP, Ping, and WSP Probes

12.1(5)T

The Cisco IOS Release 12.1(1)E feature was migrated to Cisco IOS Release 12.1(5)T, supporting Cisco 7200 Series Routers only.

12.1(7)E

Support for the following platform was added:

Cisco 7100 Series Routers

The following functions were added:

Multiple Firewall Farm Support

Route Health Injection

12.1(8a)E

Support for the following platform was added:

Catalyst 6000 Family Switches with Supervisor Engine 2

The following functions were added:

Backup Server Farms

Dynamic Feedback Protocol for IOS SLB—DFP Agent Subsystem


This document describes the Cisco IOS Server Load Balancing (SLB) feature in Cisco IOS Release 12.1(8a)E. It includes the following sections:

Overview of the IOS SLB Feature

Functions and Capabilities

Benefits

Restrictions

Supported Platforms

Supported Standards, MIBs, and RFCs

Required Configuration Tasks

Optional Configuration Tasks

Monitoring and Maintaining the IOS SLB Feature

Configuration Examples

Command Reference

Debug Commands

FAQ (Frequently Asked Questions)

Glossary

Overview of the IOS SLB Feature

The IOS SLB feature is an IOS-based solution that provides IP server load balancing. Using the IOS SLB feature, you can define a virtual server that represents a group of real servers in a cluster of network servers known as a server farm. In this environment, the clients connect to the IP address of the virtual server. When a client initiates a connection to the virtual server, the IOS SLB function chooses a real server for the connection based on a configured load-balancing algorithm.


Note IOS SLB does not support load balancing of flows between clients and real servers that are on the same local area network (LAN) or virtual LAN (VLAN). The packets being load balanced cannot enter and leave the load-balancing device on the same interface.


IOS SLB also provides firewall load balancing, which balances flows across a group of firewalls called a firewall farm.

Figure 1 illustrates a logical view of a simple IOS SLB network.

Figure 1 Logical View of IOS SLB

Functions and Capabilities

This section describes the following functions and capabilities provided by IOS SLB.


Note Some IOS SLB functions are specific to one platform and are not described in this feature module. For information about those functions, refer to the appropriate platform-specific documentation.


Algorithms for Server Load Balancing

Alternate IP Addresses

Automatic Server Failure Detection

Automatic Unfail

Avoiding Attacks on Server Farms and Firewall Farms

Backup Server Farms

Client-Assigned Load Balancing

Content Flow Monitor Support

Delayed Removal of TCP Connection Context

Dynamic Feedback Protocol for IOS SLB

Firewall Load Balancing

Maximum Connections

Multiple Firewall Farm Support

Network Address Translation (NAT) and Session Redirection

Port-Bound Servers

Probes

Protocol Support

Redundancy Enhancements

Route Health Injection

Slow Start

Sticky Connections

SynGuard

TCP Session Reassignment

Transparent Webcache Load Balancing

WAP Load Balancing

Algorithms for Server Load Balancing

IOS SLB provides the following load-balancing algorithms:

Weighted Round Robin

Weighted Least Connections

You can specify one of these algorithms as the basis for choosing a real server for each new connection request that arrives at the virtual server.

Weighted Round Robin

The weighted round robin algorithm specifies that the real server used for a new connection to the virtual server is chosen from the server farm in a circular fashion. Each real server is assigned a weight, n, that represents its capacity to handle connections, as compared to the other real servers associated with the virtual server. That is, new connections are assigned to a given real server n times before the next real server in the server farm is chosen.

For example, assume a server farm comprised of real server ServerA with = 3, ServerB with = 1, and ServerC with = 2. The first three connections to the virtual server are assigned to ServerA, the fourth connection to ServerB, and the fifth and sixth connections to ServerC.


Note Assigning a weight of n=1 to all of the servers in the server farm configures the IOS SLB device to use a simple round robin algorithm.


Weighted Least Connections

The weighted least connections algorithm specifies that the next real server chosen from a server farm for a new connection to the virtual server is the server with the fewest active connections. Each real server is assigned a weight for this algorithm, also. When weights are assigned, the server with the fewest connections is based on the number of active connections on each server, and on the relative capacity of each server. The capacity of a given real server is calculated as the assigned weight of that server divided by the sum of the assigned weights of all of the real servers associated with that virtual server, or n1/(n1+n2+n3...).

For example, assume a server farm comprised of real server ServerA with = 3, ServerB with = 1, and ServerC with = 2. ServerA would have a calculated capacity of 3/(3+1+2), or half of all active connections on the virtual server, ServerB one-sixth of all active connections, and ServerC one-third of all active connections. At any point in time, the next connection to the virtual server would be assigned to the real server whose number of active connections is farthest below its calculated capacity.


Note Assigning a weight of n=1 to all of the servers in the server farm configures the IOS SLB device to use a simple least-connection algorithm.


Alternate IP Addresses

IOS SLB enables you to telnet to the load-balancing device using an alternate IP address. To do so, use either of the following methods:

Use any of the interface addresses to telnet to the load-balancing device.

Define a secondary IP address to telnet to the load-balancing device.

This function is similar to that provided by the LocalDirector (LD) Alias command.

Automatic Server Failure Detection

IOS SLB automatically detects each failed Transmission Control Protocol (TCP) connection attempt to a real server, and increments a failure counter for that server. (The failure counter is not incremented if a failed TCP connection from the same client has already been counted.) If a server's failure counter exceeds a configurable failure threshold, the server is considered out of service and is removed from the list of active real servers.

Automatic Unfail

When a real server fails and is removed from the list of active servers, it is assigned no new connections for a length of time specified by a configurable retry timer. After that timer expires, the server is again eligible for new virtual server connections and IOS SLB sends the server the next qualifying connection. If the connection is successful, the failed server is placed back on the list of active real servers. If the connection is unsuccessful, the server remains out of service and the retry timer is reset.

Avoiding Attacks on Server Farms and Firewall Farms

IOS SLB relies on a site's firewalls to protect the site from attacks. In general, IOS SLB is no more susceptible to direct attack than is any switch or router. However, a highly secure site can take the following steps to enhance its security:

Configure real servers on a private network to keep clients from connecting directly to them. This ensures that the clients must go through IOS SLB to get to the real servers.

Configure input access lists on the access router or on the IOS SLB device to deny flows from the outside network aimed directly at the interfaces on the IOS SLB device. That is, deny all direct flows from unexpected addresses.

To protect against attackers trying to direct flows to real or nonexistent IP addresses in the firewall subnet, configure the firewalls in a private network.

Configure firewalls to deny all unexpected flows targeted at the firewalls, especially flows originating from the external network.

Backup Server Farms

A backup server farm is a server farm that can be used when none of the real servers defined in a primary server farm is available to accept new connections. When configuring backup server farms, keep in mind the following considerations:

A server farm can act as both primary and backup at the same time.

The same real server cannot be defined in both primary and backup at the same time.

Both primary and backup require the same NAT configuration (none, client, server, or both). In addition, if NAT is specified, both server farms must use the same NAT pool.

Client-Assigned Load Balancing

Client-assigned load balancing allows you to limit access to a virtual server by specifying the list of client IP subnets that are permitted to use that virtual server. With this feature, you can assign a set of client IP subnets (such as internal subnets) connecting to a virtual IP address to one server farm or firewall farm, and assign another set of clients (such as external clients) to a different server farm or firewall farm.

Content Flow Monitor Support

IOS SLB supports the Cisco Content Flow Monitor (CFM), a web-based status monitoring application within the CiscoWorks2000 product family. You can use CFM to manage Cisco server load-balancing devices. CFM runs on Windows NT and Solaris workstations, and is accessed using a web browser.

Delayed Removal of TCP Connection Context

Because of IP packet ordering anomalies, IOS SLB might "see" the termination of a TCP connection (a finish [FIN] or reset [RST]) followed by other packets for the connection. This problem usually occurs when there are multiple paths that the TCP connection packets can follow. To correctly redirect the packets that arrive after the connection is terminated, IOS SLB retains the TCP connection information, or context, for a specified length of time. The length of time the context is retained after the connection is terminated is controlled by a configurable delay timer.

Dynamic Feedback Protocol for IOS SLB

With IOS SLB Dynamic Feedback Protocol (DFP) support, a DFP manager in a load-balancing environment can initiate a TCP connection with a DFP agent. Thereafter, the DFP agent collects status information from one or more real host servers, converts the information to relative weights, and reports the weights to the DFP manager. The DFP manager factors in the weights when load balancing the real servers. In addition to reporting at user-defined intervals, the DFP agent sends an early report if there is a sudden change in a real server's status.

You can define IOS SLB as a DFP manager, as a DFP agent for another DFP manager (such as DistributedDirector), or as both at the same time. In such a configuration, IOS SLB sends periodic reports to DistributedDirector, which uses the information to choose the best server farm for each new connection request. IOS SLB then uses the same information to choose the best real server within the chosen server farm.

This capability enables a single device to perform both global load balancing, as managed by DistributedDirector, and local load balancing, as managed by IOS SLB.

DFP also supports the use of multiple DFP agents from different client subsystems (such as IOS SLB) at the same time.

As part of the implementation of the DFP agent subsystem, the manager command has been removed. Its function is now provided by the ip dfp agent global configuration command, and by the following DFP agent configuration commands:

inservice (DFP agent)

interval (DFP agent)

password

port (DFP agent)

Firewall Load Balancing

As its name implies, firewall load balancing enables IOS SLB to balance flows to firewalls. Firewall load balancing uses a load-balancing device on each side of a group of firewalls (called a firewall farm) to ensure that the traffic for each flow travels to the same firewall, ensuring that the security policy is not compromised.

You can configure more than one firewall farm in each load-balancing device.

Layer 3 firewalls, which have IP-addressable interfaces, are supported by IOS SLB firewall load balancing if they are subnet-adjacent to the firewall load-balancing device and have unique MAC addresses. The device does not modify the IP addresses in the user packet. To send the packet to the chosen firewall, the device determines which interface to use and changes the Layer 2 headers accordingly. This is the standard dispatched routing used by IOS SLB.

Layer 2 firewalls, which do not have IP addresses, are transparent to IOS SLB firewall load balancing. IOS SLB supports Layer 2 firewalls by placing them between two IP-addressable interfaces.

Whereas many Layer 3 firewalls might exist off a single Layer 3 interface on the load-balancing device (for example, a single LAN), only one Layer 2 firewall can exist off each interface.

When configuring the load-balancing device, you configure a Layer 3 firewall using its IP address, and a Layer 2 firewall using the IP address of the interface of the device on the "other side" of the firewall.

To balance flows across the firewalls in a firewall farm, IOS SLB firewall load balancing performs a route lookup on each incoming flow, examining the source and destination IP addresses (and optionally the source and destination TCP or User Datagram Protocol [UDP] port numbers). Firewall load balancing applies a hash algorithm to the results of the route lookup and selects the best firewall to handle the connection request.


Note IOS SLB firewall load balancing must examine incoming packets and perform route lookup. On Catalyst 6000 Family Switches, some additional packets might need to be examined. Firewall load balancing will impact internal (secure) side routing performance and must be considered in the complete design.


To maximize availability and resilience in a network with multiple firewalls, configure a separate equal-weight route to each firewall, rather than a single route to only one of the firewalls.

IOS SLB firewall load balancing provides the following capabilities:

Connections initiated from either side of the firewall farm are load-balanced.

The load is balanced among a set of firewalls—the firewall farm.

All packets for a connection travel through the same firewall. Subsequent connections can be "sticky," ensuring that they are assigned to the same firewall.

Probes are used to detect and recover from firewall failures.

Redundancy is provided. Hot Standby Router Protocol (HSRP), stateless backup, and stateful backup are all supported.

Multiple interface types and routing protocols are supported, enabling the external (Internet side) load-balancing device to act as an access router.

Proxy firewalls are supported.

Maximum Connections

IOS SLB allows you to configure maximum connections for server and firewall load balancing.

For server load balancing, you can configure a limit on the number of active connections that a real server is assigned. If the maximum number of connections is reached for a real server, IOS SLB automatically switches all further connection requests to another server until the connection number drops below the specified limit.

For firewall load balancing, you can configure a limit on the number of active TCP or UDP connections that a firewall farm is assigned. If the maximum number of connections is reached for the firewall farm, new connections are dropped until the connection number drops below the specified limit.

Multiple Firewall Farm Support

You can configure more than one firewall farm in each load-balancing device.

Network Address Translation (NAT) and Session Redirection

Cisco IOS NAT, RFC 1631, allows unregistered "private" IP addresses to connect to the Internet by translating them into globally registered IP addresses. Cisco IOS NAT also increases network privacy by hiding internal IP addresses from external networks.

IOS SLB can operate in one of two session redirection modes:

Dispatched mode—the virtual server address is known to the real servers; you must configure the virtual server IP address as a loopback address, or secondary IP address, on each of the real servers. IOS SLB redirects packets to the real servers at the media access control (MAC) layer. Since the virtual server IP address is not modified in dispatched mode, the real servers must be Layer 2-adjacent to IOS SLB, or intervening routers might not be able to route to the chosen real server.


Note IOS SLB supports FTP and firewall load balancing only in dispatched mode. Therefore, FTP and firewall load balancing cannot use NAT.


Directed mode—the virtual server can be assigned an IP address that is not known to any of the real servers. IOS SLB translates packets exchanged between a client and real server, translating the virtual server IP address to a real server IP address through NAT.

IOS SLB supports the following types of NAT:

Server NAT—By replacing the virtual server IP address with the real server IP address (and vice versa):

Servers can be many hops away from the load-balancing device.

Intervening routers can route to them without requiring tunnelling.

Loopback and secondary interfaces are not required on the real server.

The real server need not be Layer 2-adjacent to IOS SLB.

A less common form of server NAT is server port translation, which involves replacement of a virtual server port. Server port translation does not require server IP address translation, but the two translations can be used together.


Note If an IP address is configured as a real IP address for a NAT virtual server, you cannot balance connection requests from that address to a different virtual server (whether NAT or dispatched) on the same load-balancing device.


Client NAT—If multiple load-balancing devices are used, replacing the client IP address with an IP address associated with one of the devices results in proper routing of outbound flows to the correct device. Client NAT also requires that the ephemeral client port be modified since many clients can use the same ephemeral port. Even in cases where multiple load-balancing devices are not used, client NAT can be useful to ensure that packets from load-balanced connections are not routed around the device.

In both dispatched and directed modes, IOS SLB must track connections. Therefore, you must design your network so that there is no alternate network path from the real servers to the client that bypasses the load-balancing device.


Note Both server NAT and client NAT are supported for the same connection.


Port-Bound Servers

When you define a virtual server, you must specify the TCP or UDP port handled by that virtual server. However, if you configure NAT on the server farm, you can also configure port-bound servers. Port-bound servers allow one virtual server IP address to represent one set of real servers for one service, such as Hypertext Transfer Protocol (HTTP), and a different set of real servers for another service, such as Telnet.

Packets destined for a virtual server address for a port that is not specified in the virtual server definition are not redirected.

IOS SLB supports both port-bound and non-port-bound servers, but port-bound servers are recommended.


Note You cannot configure NAT on a firewall farm, therefore you cannot configure port-bound servers for IOS SLB firewall load balancing.


Probes

IOS SLB supports HTTP probes, ping probes, and WSP probes.

HTTP and ping probes are a simple way to verify connectivity for devices being server load-balanced, and for firewalls being firewall load-balanced (even devices on the other side of a firewall).

HTTP probes also enable you to monitor applications being server load-balanced. With frequent probes, the operation of each application is verified, not just connectivity to the application.

WSP probes detect failures in the Wireless Application Protocol (WAP) stack on port 9201.

You can configure more than one probe, in any combination of types (HTTP, ping, or WSP), for each server farm, or for each firewall in a firewall farm.

Probes in Server Load Balancing

Probes determine the status of each real server in a server farm. All real servers associated with all virtual servers tied to that server farm are probed.

If a real server fails for one probe, it is failed for all probes. After the real server recovers, all probes must acknowledge its recovery before it is restored to service.

Probes in Firewall Load Balancing

Probes detect firewall failures. All firewalls associated with the firewall farm are probed.

If a firewall fails for one probe, it is failed for all probes. After the firewall recovers, all probes must acknowledge its recovery before it is restored to service.

Make sure you configure the HTTP probe to expect status code 401, to eliminate password problems. See the expect command for more details.

Use the ip http server command to configure an HTTP server on the device. See the description of the ip http server command in the Cisco IOS Configuration Fundamentals Command Reference for more details.

HTTP probes do not support HTTP over Secure Socket Layer (HTTPS).

In a transparent webcache load-balancing environment, an HTTP probe uses the real IP address of the webcache, since there is no virtual IP address configured.

Protocol Support

IOS SLB supports the following protocols:

Domain Name System (DNS)

File Transfer Protocol (FTP)

Hypertext Transfer Protocol (HTTP)

Hypertext Transfer Protocol over Secure Socket Layer (HTTPS)

Internet Message Access Protocol (IMAP)

Mapping of Airline Traffic over IP, Type A (MATIP-A)

Network News Transport Protocol (NNTP)

Post Office Protocol, version 2 (POP2)

Post Office Protocol, version 3 (POP3)

RealAudio/RealVideo via HTTP

Remote Authentication Dial-In User Service (RADIUS)

Simple Mail Transport Protocol (SMTP)

Telnet

X.25 over TCP (XOT)

Redundancy Enhancements

An IOS SLB device can represent a single point of failure, and the servers can lose their connections to the backbone, if either of the following occurs:

The IOS SLB device fails.

A link from a switch to the distribution-layer switch becomes disconnected.

To reduce that risk, IOS SLB supports two redundancy options, both based on HSRP:

Stateless backup—provides high network availability by routing IP flows from hosts on Ethernet networks without relying on the availability of a single Layer 3 switch.

Stateful backup—enables IOS SLB to incrementally backup its load-balancing decisions, or "keep state," between primary and backup switches.

IOS SLB also supports active standby, in which two IOS SLBs can load-balance the same virtual IP address while at the same time acting as backups for each other. If a site has only one virtual IP address to load balance, an access router is used to direct a subset of the flows to each IOS SLB using policy-based routing.

IOS SLB firewall load balancing does not support active standby. That is, you cannot configure two pairs of firewall load balancing devices (one pair on each side of the firewalls), with each device in each pair handling traffic and backing up its partner.

Route Health Injection

By default, a virtual server's IP address is advertised (added to the routing table) when you bring the virtual server into service (using the inservice command). If you have a preferred host route to a website's virtual IP address, you can advertise that host route, but you have no guarantee that the IP address is available. However, you can use the advertise command to configure IOS SLB to advertise the host route only when IOS SLB has verified that the IP address is available. IOS SLB withdraws the advertisement when the IP address is no longer available. This function is known as route health injection.


Note When route health injection is configured, probes require a default route to the virtual server (specified using the ip route 0.0.0.0 0.0.0.0 command, for example). The route is not used, but it must exist to enable the sockets code to verify that the destination can be reached, which in turn is essential for route health injection to function correctly.


Slow Start

In an environment that uses weighted least connections load balancing, a real server that is placed in service initially has no connections, and could therefore be assigned so many new connections that it becomes overloaded. To prevent such an overload, slow start controls the number of new connections that are directed to a real server that has just been placed in service.

Sticky Connections

When you use sticky connections, new connections from a client IP address or subnet are assigned to the same real server (for server load balancing) or firewall (for firewall load balancing) as were previous connections from that address or subnet.

IOS SLB creates sticky objects to track client assignments. The sticky objects remain in the IOS SLB database after the last sticky connection is deleted, for a user-defined period. New connections from a client are sticky if the following conditions are met:

The real server is in either OPERATIONAL or MAXCONNS_THROTTLED state.

The sticky timer is defined on a virtual server or on a firewall farm.

The amount of time between the end of a previous connection from the client and the start of the new connection is within the sticky timer duration.
OR
A connection for the same client already exists. (That is, the connection is not the first for this client.)

Sticky connections allow you to create a sticky object for a subnet, ensuring that all flows from the subnet are sent to the same real server. (Make sure the volume of flows is not so large that it overwhelms the real server.)

Sticky connections also permit the coupling of services that are handled by more than one virtual server or firewall farm. This allows connection requests for related services to use the same real server. For example, web server (HTTP) typically uses TCP port 80, and HTTPS uses port 443. If HTTP virtual servers and HTTPS virtual servers are coupled, connections for ports 80 and 443 from the same client IP address or subnet are assigned to the same real server.

SynGuard

SynGuard limits the rate of TCP start-of-connection packets (SYNchronize sequence numbers, or SYNs) handled by a virtual server to prevent a type of network problem known as a SYN flood denial-of-service attack. A user might send a large number of SYNs to a server, which could overwhelm or crash the server, denying service to other users. SynGuard prevents such an attack from bringing down IOS SLB or a real server. SynGuard monitors the number of SYNs handled by a virtual server at specific intervals and does not allow the number to exceed a configured SYN threshold. If the threshold is reached, any new SYNs are dropped.

IOS SLB firewall load balancing does not support SynGuard.

TCP Session Reassignment

IOS SLB tracks each TCP SYN sent to a real server by a client attempting to open a new connection. If several consecutive SYNs are not answered, or if a SYN is replied to with an RST, the TCP session is reassigned to a new real server. The number of SYN attempts is controlled by a configurable reassign threshold.

IOS SLB firewall load balancing does not support TCP session reassignment.

Transparent Webcache Load Balancing

IOS SLB can load balance port 80 flows across a cluster of transparent webcaches. To set up this function, configure the subnet IP addresses served by the transparent webcaches, or some common subset of them, as virtual servers. Virtual servers used for transparent webcache load balancing do not answer pings on behalf of the subnet IP addresses, and they do not affect traceroute.

In some cases, such as when its cache does not contain needed pages, a webcache might need to initiate its own connections to the Internet. Those connections should not be load balanced back to the same set of webcaches. To address this need, IOS SLB allows you to configure client exclude statements, which exclude connections initiated by the webcaches from the load-balancing scheme.

IOS SLB firewall load balancing does not support transparent webcache load balancing.

In the following sample configuration, virtual server WEBCACHE examines all web flows passing through the load-balancing device and dispatches them to server farm WEBCACHE-FARM. The client exclude statement excludes flows originating from subnet 80.80.7.0, enabling the real servers 80.80.7.188 and 80.80.7.189 to communicate with the Internet as needed.

ip slb serverfarm WEBCACHE-FARM
  real 80.80.7.188
   inservice
  real 80.80.7.189
   inservice
ip slb vserver WEBCACHE
  virtual 0.0.0.0 0.0.0.0 tcp www
  serverfarm WEBCACHE-FARM
  client 80.80.7.0 255.255.255.0 exclude
  inservice

WAP Load Balancing

You can use IOS SLB to load balance Wireless Session Protocol (WSP) sessions among a group of WAP gateways or servers on an IP bearer network. WAP runs on top of UDP on a set of well known ports, with each port indicating a different WAP mode:

Connectionless WSP mode (IP/UDP [9200]/WSP). In connectionless WSP mode, WSP is a simple one-request/one-response protocol in which a single server-bound packet results in a server response of one or more packets.

Connection-oriented WSP mode (IP/UDP [9201]/WTP/WSP). In connection-oriented WSP mode, WTP handles retransmissions of WDP events, and WSP operates using a defined session bring-up/tear-down sequence. IOS SLB uses a WAP-aware finite state machine (FSM), driven by events in WSP sessions, to reassign sessions. This FSM operates only on port 9201, where the WSP sessions are not encrypted and WTP handles retransmissions.

Connectionless secure WSP mode (IP/UDP [9202]/WTLS/WSP). This mode functions the same as connectionless WSP mode, but with security provided by WTLS.

Connection-oriented secure WSP mode (IP/UDP [9203]/WTLS/WTP/WSP). This mode functions the same as connection-oriented WSP mode, but with security provided by WTLS.

IOS SLB uses ping probes to detect failures in the WAP load-balancing device, and WSP probes to detect failures in the WAP stack on port 9201.

Benefits

IOS SLB shares the same software code base as Cisco IOS and has all the software features sets of Cisco IOS software. IOS SLB is recommended for customers desiring complete integration of SLB technology into traditional Cisco switches and routers.

On the Cisco Catalyst 6500 switch, IOS SLB takes advantage of hardware acceleration to forward data packets at very high speed when running in dispatched mode.

IOS SLB assures continuous, high availability of content and applications with proven techniques for actively managing servers and connections in a distributed environment. By distributing user requests across a cluster of servers, IOS SLB optimizes responsiveness and system capacity, and dramatically reduces the cost of providing Internet, database, and application services for large-, medium-, and small-scale sites.

IOS SLB facilitates scalability, availability, and ease of maintenance:

The addition of new physical (real) servers, and the removal or failure of existing servers, can occur at any time, transparently, without affecting the availability of the virtual server.

IOS SLB's slow start capability allows a new server to increase its load gradually, preventing failures caused by assigning the server too many new connections too quickly.

IOS SLB supports fragmented packets and packets with IP options, buffering your servers from client or network vagaries that are beyond your control.

IOS SLB firewall load balancing enables you to scale access to your Internet site. You can add firewalls without affecting existing connections, enabling your site to grow without impacting customers.

Administration of server applications is easier. Clients know only about virtual servers; no administration is required for real server changes.

Security of the real server is provided because its address is never announced to the external network. Users are familiar only with the virtual IP address. You can filter unwanted flows based on both IP address and TCP or UDP port numbers. Additionally, though it does not eliminate the need for a firewall, IOS SLB can help protect against some denial-of-service attacks.

In a branch office, IOS SLB allows balancing of multiple sites and disaster recovery in the event of full-site failure, and distributes the work of load balancing.

Restrictions

IOS SLB has the following restrictions:

Does not support load balancing of flows between clients and real servers that are on the same local area network (LAN) or virtual LAN (VLAN). The packets being load balanced cannot enter and leave the load-balancing device on the same interface.

Operates in a standalone mode and currently does not operate as a MultiNode Load Balancing (MNLB) Services Manager. The presence of IOS SLB does not preclude the use of the existing MNLB Forwarding Agent with an external Services Manager (such as the LocalDirector) in an MNLB environment.

Does not support coordinating server load-balancing statistics among different IOS SLB instances for backup capability.

Supports FTP and firewall load balancing only in dispatched mode.

Does not support IOS SLB and Cisco Applications and Services Architecture (CASA) configured with the same virtual IP address, even if they are for different services.

Does not support both IOS server load balancing and firewall load balancing on the same flow, nor on the same server port. You can configure both server load balancing and firewall load balancing on the same device at the same time, but they must apply to different flows (different client-server pairs). These functions can run on the same EPIF (for example, server load balancing on port 1 and firewall load balancing on port 2). Load-balancing the server farm after a packet exits the load-balanced firewall farm requires a separate load-balancing device.

Does not support running both IOS SLB and the Content Switching Module (CSM) on the same switch.

When operating in dispatched mode, real servers must be Layer 2-adjacent, tag-switched, or via generic routing encapsulation (GRE) tunnel.

When operating in directed mode with server NAT, real servers need not be Layer 2-adjacent to IOS SLB. This allows for more flexible network design, since servers can be placed several Layer 3 hops away from the IOS SLB switch.

The DFP agent requires a delay between hello messages of at least 3 seconds. Therefore, if your DFP manager provides a timeout specification, you must set the timeout to at least 3 seconds.

For firewall load balancing:

No longer limited to a single firewall farm in each load-balancing device.

Limited to a single active firewall load-balancing device on each side of the firewall farm. Each firewall must have its own unique MAC address and must be Layer 2-adjacent to each device. The firewalls can be connected to individual interfaces on the device, or they can all share a VLAN and connect using a single interface.

Requires Ethernet between each firewall load-balancing device and each firewall.

On each firewall load-balancing device, requires that each Layer 2 firewall be connected to a single Layer 3 (IP) interface.

Flows with a destination IP address on the same subnet as the configured firewall IP addresses are not load-balanced. (Such flows could be a firewall console session or other flows on the firewall LAN.)

Does not support the following IOS SLB functions:

- Active standby

- Network Address Translation (NAT)

- Port-bound servers

- SynGuard

- TCP session reassignment

- Transparent webcache load balancing

For backup server farm support:

Does not support defining the same real server in both primary and backup server farms.

Requires the same NAT configuration (none, client, server, or both) for both primary and backup server farms. In addition, if NAT is specified, both server farms must use the same NAT pool.

Does not support HTTP redirect load balancing. If a primary server farm specifies a redirect virtual server, you cannot define that primary as a backup, nor can you define a backup for that primary.

For the Catalyst 6000 Family Switches:

Requires the Multilayer Switched Feature Card (MSFC) and the Policy Feature Card (PFC). When using redundant MSFCs in the same Catalyst 6000 Family switch, stateful backup between the two MSFCs is not supported, but stateless backup between the two MSFCs is supported.

The term "MSFC" refers to either an MSFC1 or an MSFC2, except when specifically differentiated.

The term "PFC" refers to either a PFC1 or a PFC2, except when specifically differentiated.

Requires that the Multilayer Switching (MLS) flow mode be set to full. For more information about how to set the MLS flow, refer to the "Configuring IP Multilayer Switching" section in the Catalyst 6000 Family MSFC (12.0) & PFC Configuration Guide, Release 5.4.

When operating in dispatched mode, real servers must be Layer 2-adjacent to IOS SLB (that is, not beyond an additional router), with hardware data packet acceleration performed by the PFC. All real servers that can be reached by a single IOS SLB device (that is, all real servers in a given server farm) be on the same VLAN. The loopback address must be configured in the real servers.

Requires that all firewall interfaces be on the same VLAN.

Provides no hardware data packet acceleration in directed mode. (Hardware data packet acceleration is performed by the PFC, and in directed mode the data packets are handled by the MSFC, not the PFC.)

Supports Native IOS only.

For the Cisco 7100 Series and Cisco 7200 Series:

Provides no hardware acceleration for the IOS SLB function for either dispatched mode or directed mode.

Supports Cisco IOS NAT in directed mode with no hardware data packet acceleration.

Does not support route health injection in 12.1(8a)E.

Supported Platforms

Catalyst 6000 Family Switches with Supervisor Engine 1

Catalyst 6000 Family Switches with Supervisor Engine 2

Cisco 7100 Series Routers

Cisco 7200 Series Routers

Supported Standards, MIBs, and RFCs

Standards

No new or modified standards

MIBs

CISCO-SLB-MIB


Note Although the objects in this MIB are defined as read-create, you cannot use the SNMP SET command to modify them. Instead, you must use the command line to set the associated command line keywords, after which the new values are reflected in SNMP.


RFCs

Cisco IOS NAT, RFC 1631

Required Configuration Tasks

This section describes the tasks required to configure a basic IOS SLB network.

Configuring IOS SLB involves identifying server farms, configuring groups of real servers in server farms, and configuring the virtual servers that represent the real servers to the clients. See the following sections for required configuration tasks for the IOS SLB feature.

Configuring the Server Farms

Verifying the Server Farms

Configuring the Virtual Servers

Verifying the Virtual Servers

Configuring the Restricted Clients

Verifying the Restricted Clients

Verifying IOS SLB Connectivity

Figure 2 shows a sample IOS SLB network with the following components:

Two server farms—one configured to allow access by the public and named PUBLIC, one configured to allow limited access and named RESTRICTED.

Five real servers configured as follows:

Three real servers in the PUBLIC server farm with IP addresses 10.1.1.1, 10.1.1.2, and 10.1.1.3

Two real servers in the restricted server farm with IP addresses 10.1.1.20 and 10.1.1.21

Two virtual servers—one configured to allow access by the public and named PUBLIC_HTTP and one configured to allow limited access and named RESTRICTED_HTTP.

Virtual server PUBLIC_HTTP is configured with IP address 10.0.0.1 load balancing TCP connections on the WWW port (80).

Virtual server RESTRICTED_HTTP is configured with IP address 10.0.0.2 load balancing TCP connections on the WWW port (80) and allows access only from clients from network 10.4.4.0 255.255.255.0.

Figure 2 Example IOS SLB Network

To configure the IOS SLB network shown in Figure 2, use the following commands beginning in global configuration mode:

Command
Purpose
Router(config)# ip slb serverfarm serverfarm-name
Router(config-slb-sfarm)#

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

Router(config-slb-sfarm)# real ip-address

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

Router(config-slb-real)# inservice

Enables the real server for use by IOS SLB. See the inservice (server farm real server) command for more details.

Router(config-slb-real)# exit

Returns to server farm configuration mode.

Router(config-slb-sfarm)# end

Returns to global configuration mode.

Router(config)# ip slb vserver virtual_server-name

Identifies a virtual server and initiates virtual server configuration mode. See the ip slb vserver command for more details.

Router(config-slb-vserver)# virtual ip-address [network-mask] 
{tcp | udp} [port-number | wsp | wsp-wtp | wsp-wtls | 
wsp-wtp-wtls] [service service-name]

Specifies the virtual server IP address, type of connection, TCP or UDP port number, WSP mode, and optional service coupling. See the virtual (virtual server) command for more details.

Router(config-slb-vserver)# serverfarm primary-serverfarm-name 
[backup backup-serverfarm-name [sticky]]

Associates a real server farm with a virtual server, or configures a backup server farm. See the serverfarm command for more details.

Router(config-slb-vserver)# inservice

Enables the virtual server for use by IOS SLB. See the inservice (server farm virtual server) command for more details.

Router(config-slb-vserver)# client ip-address network-mask

Specifies which clients are allowed to use the virtual server. See the client (virtual server) command for more details.


The following sections include examples of the configuration commands used to configure and verify the IOS SLB network shown in Figure 2:

Configuring the Server Farms

Verifying the Server Farms

Configuring the Virtual Servers

Verifying the Virtual Servers

Configuring the Restricted Clients

Verifying the Restricted Clients

Verifying IOS SLB Connectivity

Configuring the Server Farms

The following commands configure the server farm PUBLIC and associate the three real servers:

Router# config t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.1.1.1
Router(config-slb-real)# reassign 2
Router(config-slb-real)# faildetect numconns 4 numclients 2
Router(config-slb-real)# retry 20
Router(config-slb-real)# inservice
Router(config-slb-real)# exit
Router(config-slb-sfarm)# real 10.1.1.2
Router(config-slb-real)# reassign 2
Router(config-slb-real)# faildetect numconns 4
Router(config-slb-real)# retry 20
Router(config-slb-real)# inservice
Router(config-slb-real)# exit
Router(config-slb-sfarm)# real 10.1.1.3
Router(config-slb-real)# reassign 2
Router(config-slb-real)# faildetect numconns 4
Router(config-slb-real)# retry 20
Router(config-slb-real)# inservice
Router(config-slb-real)# end

The following commands configure the server farm RESTRICTED and associate the two real servers:

Router# config t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)# ip slb serverfarm RESTRICTED
Router(config-slb-sfarm)# real 10.1.1.20
Router(config-slb-real)# reassign 2
Router(config-slb-real)# faildetect numconns 4
Router(config-slb-real)# retry 20
Router(config-slb-real)# inservice
Router(config-slb-real)# exit
Router(config-slb-sfarm)# real 10.1.1.21
Router(config-slb-real)# reassign 2
Router(config-slb-real)# faildetect numconns 4
Router(config-slb-real)# retry 20
Router(config-slb-real)# inservice
Router(config-slb-real)# end
Router#

Verifying the Server Farms

The following show ip slb reals command displays the status of server farms PUBLIC and RESTRICTED, the associated real servers, and their status:

Router# show ip slb real
real                    farm name        weight   state          conns
---------------------------------------------------------------------
10.1.1.1                 PUBLIC           8       OPERATIONAL      0
10.1.1.2                 PUBLIC           8       OPERATIONAL      0
10.1.1.3                 PUBLIC           8       OPERATIONAL      0
10.1.1.20                RESTRICTED       8       OPERATIONAL      0
10.1.1.21                RESTRICTED       8       OPERATIONAL      0
Router#

The following show ip slb serverfarm command displays the configuration and status of server farms PUBLIC and RESTRICTED:

Router# show ip slb serverfarm
server farm      predictor    nat   reals   bind id
---------------------------------------------------
PUBLIC           ROUNDROBIN   none  3       0
RESTRICTED       ROUNDROBIN   none  2       0
Router#

Configuring the Virtual Servers

The following commands configure the virtual servers PUBLIC_HTTP and RESTRICTED_HTTP:

Router#
Router# config t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# virtual 10.0.0.1 tcp www
Router(config-slb-vserver)# serverfarm PUBLIC
Router(config-slb-vserver)# idle 120
Router(config-slb-vserver)# delay 5
Router(config-slb-vserver)# inservice
Router(config-slb-vserver)#
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# exit
Router(config)# ip slb vserver RESTRICTED_HTTP
Router(config-slb-vserver)# virtual 10.0.0.2 tcp www
Router(config-slb-vserver)# serverfarm RESTRICTED
Router(config-slb-vserver)# idle 120
Router(config-slb-vserver)# delay 5
Router(config-slb-vserver)# inservice
Router(config-slb-vserver)#
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# end
Router#

Verifying the Virtual Servers

The following show ip slb vserver command verifies the configuration of the virtual servers PUBLIC_HTTP and RESTRICTED_HTTP:

Router# show ip slb vserver
slb vserver      prot  virtual               state         conns
-------------------------------------------------------------------
PUBLIC_HTTP      TCP   10.0.0.1:80           OPERATIONAL     0
RESTRICTED_HTTP  TCP   10.0.0.2:80           OPERATIONAL     0
Router#

Router#

Configuring the Restricted Clients

The following commands remove the virtual server RESTRICTED_HTTP from service, configure the restricted client access to the virtual server, then enable the virtual server again:

Router# config t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)# ip slb vserver RESTRICTED_HTTP
Router(config-slb-vserver)# no inservice
Router(config-slb-vserver)#
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# client 10.4.4.0 255.255.255.0
Router(config-slb-vserver)# inservice
Router(config-slb-vserver)#
src = 0 - 0
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# end
Router#

Verifying the Restricted Clients

The following show ip slb conns command verifies the restricted client access and status:

Router# show ip slb conns
vserver         prot client                real                  state     nat
-------------------------------------------------------------------------------
RESTRICTED_HTTP TCP  10.4.4.0:80           10.1.1.20             CLOSING   none
Router#

The following show ip slb conns command displays detailed information about the restricted client access status:

Router# show ip slb conns client 10.4.4.0 detail
VSTEST_UDP, client = 10.4.4.0:80
  state = CLOSING, real = 10.1.1.20, nat = none
  v_ip = 10.0.0.2:80, TCP, service = NONE
  client_syns = 0, sticky = FALSE, flows attached = 0
Router#

Verifying IOS SLB Connectivity

To verify that the IOS SLB feature has been installed and is operating correctly, ping the real servers from the IOS SLB switch, then ping the virtual servers from the clients.

The following show ip slb stats command displays detailed information about the IOS SLB network 
status:
Router# show ip slb stats
Pkts via normal switching:  0
Pkts via special switching: 6
Connections Created:        1
Connections Established:    1
Connections Destroyed:      0
Connections Reassigned:     0
Zombie Count:               0
Connections Reused:         0
Router#

Normal switching is when IOS SLB packets are handled on normal IOS switching paths (CEF, fastswitching, and process level switching). Special switching is when IOS SLB packets are handled on hardware-assisted switching paths.

See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB networks and connections.

Optional Configuration Tasks

This section describes the following optional tasks you can use to fine tune the IOS SLB configuration:

Specifying a Server Load Balancing Algorithm

Specifying a Bind ID

Configuring Real Server Attributes

Adjusting Virtual Server Values

Configuring IOS SLB Firewall Load Balancing

Configuring HTTP Probes

Configuring Ping Probes

Configuring WSP Probes

Configuring IOS SLB Dynamic Feedback Protocol

Configuring IOS SLB NAT

Implementing IOS SLB Stateless Backup

Configuring IOS SLB Stateful Backup

Monitoring and Maintaining the IOS SLB Feature

Specifying a Server Load Balancing Algorithm

To determine which real server to use for each new connection request, the IOS SLB feature uses one of the following load-balancing algorithms: weighted round robin (the default) or weighted least connections. See one of the following sections for more details:

The "Weighted Round Robin" section

The "Weighted Least Connections" section


Note You can configure a real server with a weight relative to other real servers in the server farm, using the weight (server farm) real server configuration command. If you use DFP, the static weights you define using the weight (server farm) command are overridden by the weights calculated by DFP. If DFP is removed from the network, IOS SLB reverts to the static weights.


To specify the load-balancing algorithm, use the following command in server farm configuration mode:

Command
Purpose
Router(config-slb-sfarm)# predictor [roundrobin | leastconns]

Specifies the algorithm to be used to determine how a real server is selected. See the predictor (server farm) command for more details.


The following example shows how to configure weighted least-connections algorithm:

Router(config)# ip slb serverfarm RESTRICTED
Router(config-slb-sfarm)# predictor leastconns

See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB network connections and the "Complete Example Configuration" section for an example of an IOS SLB network configuration.

Specifying a Bind ID

The bind ID allows a single physical server to be bound to multiple virtual servers and report a different weight for each one. Thus, the single real server is represented as multiple instances of itself, each having a different bind ID. DFP uses the bind ID to identify for which instance of the real server a given weight is specified. The bind ID is needed only if you are using DFP.

To configure a bind ID on the server farm for use by DFP, use the following command in server farm configuration mode:

Command
Purpose
Router(config-slb-sfarm)# bindid [bind_id]

Specifies a bind ID on the server farm for use by DFP. See the bindid command for more details.


The following example shows how to configure a bind ID of 309 on server farm RESTRICTED:

Router(config)# ip slb serverfarm RESTRICTED
Router(config-slb-sfarm)# bindid 309

See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB network connections and the "Complete Example Configuration" section for an example of an IOS SLB network configuration.

Configuring Real Server Attributes

You can configure any of the following real server attributes, by using the following real server commands beginning in global configuration mode:

Command
Purpose
Router(config)# ip slb serverfarm serverfarm-name
Router(config-slb-sfarm)#

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

Router(config-slb-sfarm)# real ip-address

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

Router(config-slb-real)# faildetect numconns number-conns 
[numclients number-clients]

Specifies the number of consecutive connection failures and, optionally, the number of unique client connection failures, that constitute failure of the real server. See the faildetect (real server) command for more details.

Router(config-slb-real)# maxconns number-conns

Specifies the maximum number of active connections allowed on the real server at one time. See the maxconns (server farm) command for more details.

Router(config-slb-real)# reassign threshold

Specifies the threshold of consecutive unacknowledged synchronizations that, if exceeded, result in an attempted connection to a different real server. See the reassign command for more details.

Router(config-slb-real)# retry retry-value

Specifies the interval, in seconds, to wait between the detection of a server failure and the next attempt to connect to the failed server. See the retry command for more details.

Router(config-slb-real)# weight weighting-value

Specifies the real server's workload capacity relative to other servers in the server farm. See the weight (server farm) command for more details.

Router(config-slb-real)# inservice

Enables the real server for use by IOS SLB. See the inservice (server farm real server) command for more details.


The following example shows how to configure the consecutive connection failures to 16 that constitute the failure of real server 10.1.1.1:

Router(config)# ip slb serverfarm RESTRICTED
Router(config-slb-sfarm)# real 10.1.1.1
Router(config-slb-real)# faildetect numconns 16 

The following example shows how to configure maximum number of connections to 1000:

Router(config-slb-real)# maxconns 1000

The following example shows how to configure to 4 the number of consecutive unacknowledged SYNs that initiates assignment of the connection to a different real server:

Router(config-slb-real)# reassign 4 

The following example shows how to configure the retry interval to 120 seconds between the detection of a server failure and the next attempt to connect on real server 10.1.1.1:

Router(config-slb-real)# retry 120 

The following example shows how to configure workload capacity to 16, relative to other servers in the server farm:

Router(config-slb-real)# weight 16 

The following example shows how to enable the real server back into service after making changes to its configuration:

Router(config-slb-real)# inservice 

See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB network connections and the "Complete Example Configuration" section for an example of an IOS SLB network configuration.

Adjusting Virtual Server Values

To change the default settings of the virtual server values, use the related virtual server command beginning in global configuration mode:

Command
Purpose
Router(config)# ip slb vserver virtual_server-name

Identifies a virtual server and initiates virtual server configuration mode. See the ip slb vserver command for more details.

Router(config-slb-vserver)# advertise

Controls the installation of a static route to the Null0 interface for a virtual server address. See the advertise command for more details.

Router(config-slb-vserver)# client ip-address network-mask

Specifies which clients are allowed to use the virtual server. See the client (virtual server) command for more details.

Router(config-slb-vserver)# delay duration

Specifies the amount of time IOS SLB maintains TCP connection context after a connection has terminated. See the delay (virtual server) command for more details.

Router(config-slb-vserver)# idle duration

Specifies the minimum amount of time IOS SLB maintains connection context in the absence of packet activity. See the idle (virtual server) command for more details.

Router(config-slb-vserver)# sticky duration [group group-id] 
[netmask netmask]

Specifies that connections from the same client use the same real server, as long as the interval between client connections does not exceed the specified duration. See the sticky (virtual server) command for more details.

Router(config-slb-vserver)# synguard syn-count interval

Specifies the rate of TCP SYNs handled by a virtual server in order to prevent a SYN flood denial-of-service attack. See the synguard (virtual server) command for more details.

Router(config-slb-vserver)# inservice

Enables the virtual server for use by IOS SLB. See the inservice (server farm virtual server) command for more details.


The following commands remove the virtual server RESTRICTED_HTTP from service and then configure the restricted client access to the virtual server:

Router(config)# ip slb vserver RESTRICTED_HTTP
Router(config-slb-vserver)# no inservice
Router(config-slb-vserver)#
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# client 10.4.4.0 255.255.255.0

By default, virtual server addresses are advertised. That is, static routes to the Null0 interface are installed for the virtual server addresses. To advertise these static routes using the routing protocol, you must configure redistribution of static routes for the routing protocol. To prevent the installation of a static route, use the no form of the advertise command:

Router(config-slb-vserver)# no advertise

The following command configures the delay timer to 20 seconds after the termination of the TCP connection to the virtual server:

Router(config-slb-vserver)# delay 20

The following command configures the idle time to 180 seconds (3 minutes) that the IOS SLB maintains connectivity to the virtual server in the absence of packet activity:

Router(config-slb-vserver)# idle 180

The following command configures the time to 60 seconds for connections from the same client to use the same real server:

Router(config-slb-vserver)# sticky 60 group 1

The following command configures the rate of TCP SYNs to 3600000 handled by the virtual server:

Router(config-slb-vserver)# synguard 3600000

In the following example, the virtual server is enabled again after modification:

Router(config-slb-vserver)# inservice
Router(config-slb-vserver)#
src = 0 - 0
.
(Information Deleted)
.
index = 1
Router(config-slb-vserver)# 

See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB network connections and the "Complete Example Configuration" section for an example of an IOS SLB network configuration.

Configuring IOS SLB Firewall Load Balancing

This section describes the tasks required to configure a basic IOS SLB firewall load-balancing network.

IOS SLB firewall load balancing uses probes to detect and recover from failures. You must configure a probe on each real server in the firewall farm. Ping probes are recommended; see the "Configuring Ping Probes" section for more details. If a firewall does not allow ping probes to be forwarded, use HTTP probes instead. See the "Configuring HTTP Probes" section for more details.

This section describes the following IOS SLB firewall load-balancing configuration tasks:

Configuring One or More Probes

Configuring the Firewall Farm

Verifying the Firewall Farm

Verifying Firewall Connectivity

Configuring One or More Probes

To configure an IOS SLB probe for firewall load balancing, refer to one of the following sections:

Configuring HTTP Probes

Configuring Ping Probes

Configuring WSP Probes

You can configure more than one probe, in any combination of types (HTTP, ping, or WSP), for each server farm, or for each firewall in a firewall farm.

Configuring the Firewall Farm

To configure an IOS SLB firewall load-balancing network, enter the following commands in order, beginning in global configuration mode:

 
Command
Purpose

Step 1 

Router(config)# ip slb firewallfarm 
firewallfarm-name
Router(config-slb-fw)#

Adds a firewall farm definition to the IOS SLB configuration and initiates firewall farm configuration mode. See the ip slb firewallfarm command for more details.

Step 2 

Router(config-slb-fw)# real ip-address

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode. See the real (firewall farm) command for more details.

Step 3 

Router(config-slb-fw-real)# probe name

Associates a probe with the firewall. See the probe (firewall farm real server) command for more details.

Step 4 

Router(config-slb-fw-real)# weight weighting-value

(Optional) Specifies the firewall's workload capacity relative to other firewalls in the firewall farm. See the weight (firewall farm real server) command for more details.

Step 5 

Router(config-slb-fw-real)# inservice

Enables the firewall for use by the firewall farm and by IOS SLB. See the inservice (firewall farm real server) command for more details.

Step 6 

Router(config-slb-fw)# access [source 
source-ip-address network-mask] [destination 
destination-ip-address network-mask]

(Optional) Routes specific flows to a firewall farm. See the access command for more details.

Step 7 

Router(config-slb-fw)# predictor hash address [port]

(Optional) Specifies whether the source and destination TCP or UDP port numbers, in addition to the source and destination IP addresses, are to be used when selecting a firewall. See the predictor hash address (firewall farm) command for more details.

Step 8 

Router(config-slb-fw)# replicate casa listening-ip 
remote-ip port-number [interval] [password [0|7] 
password [timeout]]

(Optional) Configures a stateful backup of IOS SLB firewall load balancing decision tables to a backup switch. See the replicate casa (firewall farm) command for more details.

Step 9 

Router(config-slb-fw)# tcp

(Optional) Initiates TCP protocol configuration mode. See the tcp command for more details.

Step 10 

Router(config-slb-fw-tcp)# delay duration

(Optional) For firewall farm TCP protocol configuration mode, specifies the amount of time IOS SLB firewall load balancing maintains TCP connection context after a connection has terminated. See the delay (firewall farm TCP protocol) command for more details.

Step 11 

Router(config-slb-fw-tcp)# idle duration

(Optional) For firewall farm TCP protocol configuration mode, specifies the minimum amount of time IOS SLB firewall load balancing maintains connection context in the absence of packet activity. See the idle (firewall farm TCP protocol) command for more details.

Step 12 

Router(config-slb-fw-tcp)# maxconns number-conns

(Optional) For firewall farm TCP protocol configuration mode, specifies the maximum number of active connections allowed on the firewall farm at one time. See the maxconns (firewall farm TCP protocol) command for more details.

Step 13 

Router(config-slb-fw-tcp)# sticky duration [netmask 
netmask]

(Optional) For firewall farm TCP protocol configuration mode, specifies that connections from the same IP address use the same firewall if either of the following conditions is met:

As long as any connection from that IP address exists.

For a period, defined by duration, after the last connection is destroyed.

See the sticky (firewall farm TCP protocol) command for more details.

Step 14 

Router(config-slb-fw)# udp

(Optional) Initiates UDP protocol configuration mode. See the udp command for more details.

Step 15 

Router(config-slb-fw-udp)# idle duration

(Optional) For firewall farm UDP protocol configuration mode, specifies the minimum amount of time IOS SLB firewall load balancing maintains connection context in the absence of packet activity. See the idle (firewall farm TCP protocol) command for more details.

Step 16 

Router(config-slb-fw-udp)# maxconns number-conns

(Optional) For firewall farm UDP protocol configuration mode, specifies the maximum number of active connections allowed on the firewall farm at one time. See the maxconns (firewall farm TCP protocol) command for more details.

Step 17 

Router(config-slb-fw-udp)# sticky duration [netmask 
netmask]

(Optional) For firewall farm UDP protocol configuration mode, specifies that connections from the same IP address use the same firewall if either of the following conditions is met:

As long as any connection from that IP address exists.

For a period, defined by duration, after the last connection is destroyed.

See the sticky (firewall farm UDP protocol) command for more details.

Step 18 

Router(config-slb-fw)# inservice

Enables the firewall farm for use by IOS SLB. See the inservice (firewall farm) command for more details.

Step 19 

Router(config-slb-fw-real)# exit

Returns to firewall farm configuration mode.

Step 20 

Router(config-slb-fw)# end

Exits configuration mode.

Verifying the Firewall Farm

The following show ip slb reals command displays the status of firewall farm FIRE1, the associated real servers, and their status:

Router# show ip slb real

real                  farm name        weight   state          conns
--------------------------------------------------------------------
10.1.1.2              FIRE1            8        OPERATIONAL    0
10.1.2.2              FIRE1            8        OPERATIONAL    0

The following show ip slb firewallfarm command displays the configuration and status of firewall farm FIRE1:

Router# show ip slb firewallfarm
firewall farm    hash        state         reals
------------------------------------------------
FIRE1            IPADDR      INSERVICE     2

Verifying Firewall Connectivity

To verify that IOS SLB firewall load balancing has been configured and is operating correctly:


Step 1 Ping the external real servers (the ones outside the firewall) from the IOS SLB firewall load-balancing switch.

Step 2 Ping the internal real servers (the ones inside the firewall) from the clients.

Step 3 Use the show ip slb stats command to display detailed information about the IOS SLB firewall load-balancing network status:

Router# show ip slb stats
Pkts via normal switching:  0
Pkts via special switching: 0
Connections Created:        1911871
Connections Established:    1967754
Connections Destroyed:      1313251
Connections Reassigned:     0
Zombie Count:               0
Connections Reused:         59752
Connection Flowcache Purges:1776582
Failed Connection Allocs:   17945
Failed Real Assignments:    0

Normal switching is when IOS SLB packets are handled on normal IOS switching paths (CEF, fastswitching, and process level switching). Special switching is when IOS SLB packets are handled on hardware-assisted switching paths.

Step 4 Use the show ip slb real detail command to display detailed information about the IOS SLB firewall load-balancing real server status:

Router# show ip slb real detail
10.1.1.3, FIRE1, state = OPERATIONAL, type = firewall
  conns = 299310, dummy_conns = 0, maxconns = 4294967295
  weight = 10, weight(admin) = 10, metric = 104, remainder = 2
  total conns established = 1074779, hash count = 4646
  server failures = 0
  interface FastEthernet1/0, MAC 0010.f68f.7020

Step 5 Use the show ip slb conns command to display detailed information about the active IOS SLB firewall load-balancing connections:

Router# show ip slb conns

vserver         prot client                real                  state     nat 
-------------------------------------------------------------------------------
FirewallTCP     TCP  80.80.50.187:40000    10.1.1.4              ESTAB     none
FirewallTCP     TCP  80.80.50.187:40000    10.1.1.4              ESTAB     none
FirewallTCP     TCP  80.80.50.187:40000    10.1.1.4              ESTAB     none
FirewallTCP     TCP  80.80.50.187:40000    10.1.1.4              ESTAB     none
FirewallTCP     TCP  80.80.50.187:40000    10.1.1.4              ESTAB     none

Step 6 See the "Monitoring and Maintaining the IOS SLB Feature" section for additional commands used to verify IOS SLB networks and connections.


Configuring HTTP Probes

HTTP probes verify connectivity for devices being server load-balanced, and for firewalls being firewall load-balanced. For a detailed description of HTTP probes, see the "Probes" section.

To configure an HTTP probe, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb probe name http

Configures the IOS SLB probe name and changes to HTTP probe configuration submode. See the ip slb probe (HTTP probe) command for more details.

Step 2 

Router(config-slb-probe)# address [ip-address]

(Optional) Configures an IP address to which to send the HTTP probe. See the address (HTTP probe) command for more details.

Step 3 

Router(config-slb-probe)# credentials {username 
[password]}

(Optional) Configures header values for the HTTP probe. See the credentials command for more details.

Step 4 

Router(config-slb-probe)# expect [status number] 
[regex regular-expression]

(Optional) Configures the expected HTTP status code or regular expression. See the expect command for more details.

Step 5 

Router(config-slb-probe)# header {name field-name 
[field-value]}

(Optional) Configures header values for the HTTP probe. See the header command for more details.

Step 6 

Router(config-slb-probe)# interval seconds

(Optional) Configures the HTTP probe transmit timers. See the interval (HTTP probe) command for more details.

Step 7 

Router(config-slb-probe)# port port-number

(Optional) Configures the port to which the HTTP probe is to connect. See the port (HTTP probe) command for more details.

Step 8 

Router(config-slb-probe)# request [method {get | 
post | head | name name}] [url path]

(Optional) Configures the URL path to request from the server, and the method used to perform the request to the server. See the request method, request url command for more details.

Step 9 

Router(config-slb-probe)# exit

Returns to global configuration mode.

Step 10 

Router(config)# ip slb serverfarm serverfarm-name



or


Router(config)# ip slb firewallfarm 
firewallfarm-name

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

or

Adds a firewall farm definition to the IOS SLB configuration and initiates firewall farm configuration mode. See the ip slb firewallfarm command for more details.

Step 11 

Router(config-slb-sfarm)# real ip-address


or


Router(config-slb-fw)# real ip-address

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

or

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode. See the real (firewall farm) command for more details.

Step 12 

Router(config-slb-sfarm)# probe name


or

Router(config-slb-fw-real)# probe name

Specifies an HTTP probe on the real server. See the probe (server farm) command for more details.

or

Specifies an HTTP probe on the real server. See the probe (firewall farm real server) command for more details.

Step 13 

Router(config-slb-real)# end
or
Router(config-slb-fw-real)# end

Exits configuration mode.

In addition, HTTP probes require a route to the virtual server. The route is not used, but it must exist to enable the sockets code to verify that the destination can be reached, which in turn is essential for HTTP probes to function correctly. The route can be either a host route (advertised by the virtual server) or a default route (specified using the ip route 0.0.0.0 0.0.0.0 command, for example).

To verify that the HTTP probe is configured correctly, use the following show ip slb probe command:

Router# show ip slb probe

Server:Port            State        Outages  Current  Cumulative
----------------------------------------------------------------
10.1.1.1:80            OPERATIONAL        0  never    00:00:00
10.1.1.2:80            OPERATIONAL        0  never    00:00:00
10.1.1.3:80            OPERATIONAL        0  never    00:00:00

Configuring Ping Probes

Ping probes verify connectivity for devices being server load-balanced, and for firewalls being firewall load-balanced. For a detailed description of ping probes, see the "Probes" section.

To configure a ping probe, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb probe name ping

Configures the IOS SLB probe name and changes to ping probe configuration submode. See the ip slb probe (ping probe) command for more details.

Step 2 

Router(config-slb-probe)# address [ip-address]

(Optional) Configures an IP address to which to send the ping probe. See the address (ping probe) command for more details.

Step 3 

Router(config-slb-probe)# faildetect number-of-pings

(Optional) Specifies the number of consecutive unacknowledged pings that constitute failure of the real server or firewall. See the faildetect (ping probe) command for more details.

Step 4 

Router(config-slb-probe)# interval seconds

(Optional) Configures the ping probe transmit timers. See the interval (ping probe) command for more details.

Step 5 

Router(config-slb-probe)# exit

Returns to global configuration mode.

Step 6 

Router(config)# ip slb serverfarm serverfarm-name



or


Router(config)# ip slb firewallfarm 
firewallfarm-name

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

or

Adds a firewall farm definition to the IOS SLB configuration and initiates firewall farm configuration mode. See the ip slb firewallfarm command for more details.

Step 7 

Router(config-slb-sfarm)# real ip-address


or


Router(config-slb-fw)# real ip-address

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

or

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode. See the real (firewall farm) command for more details.

Step 8 

Router(config-slb-real)# probe name


or

Router(config-slb-fw-real)# probe name

Specifies a ping probe on the real server. See the probe (server farm) command for more details.

or

Specifies a ping probe on the firewall. See the probe (firewall farm real server) command for more details.

Step 9 

Router(config-slb-real)# end
or
Router(config-slb-fw-real)# end

Exits configuration mode.

To verify that the ping probe is configured correctly, use the following show ip slb probe command:

Router# show ip slb probe

Server:Port            State        Outages  Current  Cumulative
----------------------------------------------------------------
13.13.13.13:80         OPERATIONAL        0  never    00:00:00

Configuring WSP Probes

WSP probes detect failures in the WAP stack on port 9201. For a detailed description of WSP probes, see the "Probes" section.

To configure a WSP probe, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb probe name wsp

Configures the IOS SLB probe name and changes to WSP probe configuration submode. See the ip slb probe (WSP probe) command for more details.

Step 2 

Router(config-slb-probe)# address [ip-address]

(Optional) Configures an IP address to which to send the WSP probe. See the address (WSP probe) command for more details.

Step 3 

Router(config-slb-probe)# interval seconds

(Optional) Configures the WSP probe transmit timers. See the interval (WSP probe) command for more details.

Step 4 

Router(config-slb-probe)# url [path]

(Optional) Configures the WSP probe URL path. See the url (WSP probe) command for more details.

Step 5 

Router(config-slb-probe)# exit

Returns to global configuration mode.

Step 6 

Router(config)# ip slb serverfarm serverfarm-name

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

Step 7 

Router(config-slb-sfarm)# real ip-address

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

Step 8 

Router(config-slb-sfarm)# probe name

Specifies a WSP probe on the real server. See the probe (server farm) command for more details.

Step 9 

Router(config-slb-real)# end

Exits configuration mode.

To verify that the WSP probe is configured correctly, use the following show ip slb probe command:

Router# show ip slb probe

Server:Port            State        Outages  Current  Cumulative
----------------------------------------------------------------
10.1.1.13:80         OPERATIONAL        0  never    00:00:00

Configuring IOS SLB Dynamic Feedback Protocol

The IOS SLB Dynamic Feedback Protocol (DFP) enables DFP agents to report relative weights for real host servers to DFP managers. The DFP managers factor in the weights when load balancing the real servers.

The weights calculated by DFP override the static weights you define using the weight (server farm) command. If DFP is removed from the network, IOS SLB reverts to the static weights.

See the following sections for more details about configuring DFP:

Configuring IOS SLB as a DFP Manager

Configuring a Client Subsystem as a DFP Agent


Note Remember that you can define IOS SLB as a DFP manager, as a DFP agent for another DFP manager (such as DistributedDirector), or as both at the same time. Depending on your network configuration, you might enter the commands for configuring IOS SLB as a DFP manager and the commands for configuring IOS SLB as a DFP agent on the same device, or on different devices.


Configuring IOS SLB as a DFP Manager

To identify a DFP agent with which IOS SLB can initiate connections, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb dfp [password [0|7] password 
[timeout]]

Configures DFP, supplies an optional password, and initiates DFP configuration mode. See the ip slb dfp command for more details.

Step 2 

Router(config-slb-dfp)# agent ip_address port-number 
[timeout [retry_count [retry_interval]]]

Identifies a DFP agent to which IOS SLB can connect. See the agent command for more details.

Use the following commands to:

Set the DFP password to Cookies (to match the DFP agent's password) and the timeout to 360 seconds, and then change the CLI to DFP configuration mode.

Enable IOS SLB to connect to the DFP agent with IP address 10.1.1.1 and port number 2221, and sets a timeout of 30 seconds, an infinite number of retries, and a retry interval of 10 seconds:

Router(config)# ip slb dfp password Cookies 360
Router(config-slb-dfp)# agent 10.1.1.1 2221 30 0 10

Configuring a Client Subsystem as a DFP Agent

To define the port number to be used by the DFP manager to connect to the IOS SLB DFP agent to receive DFP reports, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip dfp agent subsystem-name

Identifies a DFP agent subsystem and initiates DFP agent configuration mode. See the ip dfp agent command for more details.

Step 2 

Router(config-dfp)# interval seconds

(Optional) Configures a DFP agent weight recalculation interval. See the interval (DFP agent) command for more details.

Step 3 

Router(config-dfp)# password [0|7] password [timeout]

(Optional) Configures a DFP agent password for MD5 authentication. See the password command for more details.

Step 4 

Router(config-dfp)# port port-number

Defines the port number to be used by the DFP manager to connect to the DFP agent. See the port (DFP agent) command for more details.

Step 5 

Router(config-dfp)# inservice

Enables the DFP agent for communication with a DFP manager. See the inservice (DFP agent) command for more details.

Use the following commands to:

Identify DFP agent subsystem slb and change the CLI to DFP agent configuration mode.

Set the DFP agent weight recalculation interval to 11 seconds.

Set the unencrypted DFP agent password to Cookies (to match the DFP manager's password) and the timeout to 180 seconds.

Set the DFP communication port number for to 2221.

Enable the DFP agent for communication with the DFP manager.

Router(config)# ip dfp agent slb
Router(config-dfp)# interval 11
Router(config-dfp)# password Cookies 180
Router(config-dfp)# port 2221
Router(config-dfp)# inservice

Configuring IOS SLB NAT

Cisco Network Address Translation (NAT), RFC 1631, allows unregistered "private" IP addresses to connect to the Internet by translating them into globally registered IP addresses. NAT also increases network privacy by hiding internal IP addresses from external networks. For a detailed description of NAT and the difference between "client" and "server" mode, see the "Network Address Translation (NAT) and Session Redirection" section.

To configure IOS SLB NAT for client mode, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb natpool pool-name start-ip 
end-ip [netmask netmask | prefix-length 
leading_1_bits] [entries init-addr [max-addr]]

Configures the client address pool.

Step 2 

Router(config-slb-sfarm)# nat {server | client 
pool-name}

Configures which NAT mode to use. See the nat command for more details.

The following commands configure a NAT server on server farm PUBLIC:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# nat server

To configure IOS SLB NAT client mode for a specific server farm, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb natpool pool-name start-ip 
end-ip [netmask netmask | prefix-length 
leading_1_bits] [entries init-addr [max-addr]]

Configures the client address pool.

Step 2 

Router(config)# ip slb serverfarm serverfarm-name

Adds a server farm definition to the IOS SLB configuration and initiates server farm configuration mode. See the ip slb serverfarm command for more details.

Step 3 

Router(config-slb-sfarm)# nat {server | client 
pool-name}

Configures which NAT mode to use. See the nat command for more details.

Step 4 

Router(config-slb-sfarm)# real ip-address 
[port_number]

Identifies a real server as a member of a server farm and initiates real server configuration mode. See the real (server farm) command for more details.

Step 5 

Router(config-slb-real)# inservice

Enables the real server for use by IOS SLB. See the inservice (server farm real server) command for more details.

Step 6 

Router(config-slb-real)# exit

Returns to server farm configuration mode.

Step 7 

Router(config-slb-real)# end

Returns to global configuration mode.

Step 8 

Router(config)# ip slb vserver virtual_server-name

Identifies a virtual server and initiates virtual server configuration mode. See the ip slb vserver command for more details.

Implementing IOS SLB Stateless Backup

Stateless backup, based on the Hot Standby Router Protocol (HSRP), provides high network availability by routing IP flows from hosts on Ethernet networks without relying on the availability of any single Layer 3 switch. Stateless backup is particularly useful for hosts that do not support a router discovery protocol (such as the Intermediate System-to-Intermediate System [IS-IS] Interdomain Routing Protocol [IDRP]) and do not have the functionality to shift to a new Layer 3 switch when their selected Layer 3 switch reloads or loses power.

This section discusses the following topics in detail:

How IOS SLB Stateless Backup Works

Configuring IOS SLB Stateless Backup

Enabling HSRP

Customizing Group Attributes

Verifying the IOS SLB Stateless Backup Configuration

Sample IOS SLB Stateless Backup Configuration

How IOS SLB Stateless Backup Works

A Layer 3 switch running the HSRP detects a failure by sending and receiving multicast User Datagram Protocol (UDP) hello packets. When the IOS SLB switch running HSRP detects that the designated active Layer 3 switch has failed, the selected backup Layer 3 switch assumes control of the HSRP group MAC and IP addresses. (You can also select a new standby Layer 3 switch at that time.) Both the primary and the backup Layer 3 switch must be on the same subnet.

The chosen MAC and IP addresses must be unique and must not conflict with any others on the same network segment. The MAC address is selected from a pool of Cisco MAC addresses. Configure the last byte of the MAC address by using the HSRP group number. When the HSRP is running, it selects an active Layer 3 switch and instructs its device layer to listen on an additional (dummy) MAC address.

IOS SLB switching software supports HSRP over 10/100 Ethernet, Gigabit Ethernet, FEC, GEC, and BVI (Bridge-Group Virtual Interface) connections.

HSRP uses a priority scheme to determine which HSRP-configured Layer 3 switch is to be the default active Layer 3 switch. To configure a Layer 3 switch as active, you assign it a priority higher than that of all other HSRP-configured Layer 3 switches. The default priority is 100, so if you configure just one Layer 3 switch to have a higher priority, that switch becomes the default active switch.

HSRP works by the exchange of multicast messages that advertise priority among HSRP-configured Layer 3 switches. When the active switch fails to send a hello message within a configurable period, the standby switch with the highest priority becomes the active switch. The transition of packet-forwarding functions between Layer 3 switches is completely transparent to all hosts accessing the network.

HSRP-configured Layer 3 switches exchange the following types of multicast messages:

Hello—The hello message conveys the switch's HSRP priority and state information. By default, an HSRP switch sends hello messages every three seconds.

Coup—When a standby Layer 3 switch assumes the function of the active switch, it sends a coup message.

Resign—The active Layer 3 switch, sends this message when it is about to shut down or when a switch that has a higher priority sends a hello message.

At any time, HSRP-configured Layer 3 switches are in one of the following states:

Active—The switch is performing packet-transfer functions.

Standby—The switch is prepared to assume packet-transfer functions if the active router fails.

Speaking and listening—The switch is sending and receiving hello messages.

Listening—The switch is receiving hello messages.

Configuring IOS SLB Stateless Backup

Configuring stateless backup requires the following:

You must configure IOS SLB switches to run HSRP between interfaces on the server or firewall side.

You can configure multiple IOS SLB switches that share a virtual IP address as long as the client ranges are exclusive and you use policy routing to forward the flows to the correct IOS SLB switch.

To configure stateless backup over VLANs between IOS SLB switches, perform the following tasks in order:


Step 1 For server load balancing, configure the server farms, real servers, and virtual servers—See the "Required Configuration Tasks" section.


Note When you use the inservice (server farm virtual server) command to configure the virtual server as "in-service" you must use the optional standby command and configure an HSRP group name.


For firewall load balancing, configure the firewall farms—See the "Configuring IOS SLB Firewall Load Balancing" section.


Note When you use the inservice (firewall farm) command to configure the firewall farm as "in-service" you must use the optional standby command and configure an HSRP group name.


Step 2 Configure the IP routing protocol—See the "IP Routing Protocols" chapter of the Cisco IOS IP and IP Routing Configuration Guide.

Step 3 Configure the VLAN between the switches—See the "Virtual LANs" chapter of the Cisco IOS Switching Services Configuration Guide.

Step 4 Enable HSRP—See the "Enabling HSRP" section.

Step 5 Customize group attributes—See the "Customizing Group Attributes" section.

Step 6 Verify the IOS SLB HSRP configuration—See the "Verifying the IOS SLB Stateless Backup Configuration" section.


A sample stateless backup configuration is shown in the "Sample IOS SLB Stateless Backup Configuration" section.

Enabling HSRP

To enable HSRP on an IOS SLB interface, enable the protocol, then customize it for the interface. Enter the following command in interface configuration mode:

Command
Purpose
Router(config-if)# standby [group-number] ip [ip-address 
[secondary]]

Enables HSRP.


Customizing Group Attributes

To customize "hot standby" group attributes, use one or more of the following commands in interface configuration mode:

Command
Purpose
Router(config-if)# standby [group-number] authentication 
string

Selects an authentication string to be carried in all HSRP messages.

Router(config-if)# standby [group-number] name group-name

Specifies an HSRP group name with which to associate an IOS SLB interface.

Router(config-if)# standby [group-number] priority priority 
[preempt [delay [delay] [minimum [min-delay] | sync 
[sync-period]]]]

Configures the HSRP priority, preemption, and preemption delay.

Router(config-if)# standby [group-number] timers hellotime 
holdtime

Configures the time between hello packets and the hold time before other routers declare the active router to be down.

Router(config-if)# standby [group-number] track type-number 
[interface-priority]

Configures the interface to track other interfaces, so that if one of the other interfaces goes down, the hot standby priority for the device is lowered.


Verifying the IOS SLB Stateless Backup Configuration

For server load balancing, to verify that stateless backup has been configured and is operating correctly, use the following show ip slb vserver commands to display information about the IOS SLB virtual server status:

Router# show ip slb vserver

slb vserver      prot  virtual               state         conns   
-------------------------------------------------------------------
VS1              TCP   10.10.10.12:23        OPERATIONAL     2 
VS2              TCP   10.10.10.18:23        OPERATIONAL     2 

Router# show ip slb vserver detail
VS1, state = OPERATIONAL, v_index = 10
  virtual = 10.10.10.12:23, TCP, service = NONE, advertise = TRUE
  server farm = SERVERGROUP1, delay = 10, idle = 3600
  sticky timer = 0, sticky subnet = 255.255.255.255
  sticky group id = 0 
  synguard counter = 0, synguard period = 0
  conns = 0, total conns = 0, syns = 0, syn drops = 0
  standby group = None
VS2, state = INSERVICE, v_index = 11
  virtual = 10.10.10.18:23, TCP, service = NONE, advertise = TRUE
  server farm = SERVERGROUP2, delay = 10, idle = 3600
  sticky timer = 0, sticky subnet = 255.255.255.255
  sticky group id = 0 
  synguard counter = 0, synguard period = 0
  conns = 0, total conns = 0, syns = 0, syn drops = 0
  standby group = None

For firewall load balancing, to verify that stateless backup has been configured and is operating correctly, use the following show ip slb firewallfarm commands to display information about the IOS SLB firewall farm status:

Router# show ip slb firewallfarm
firewall farm    hash        state         reals
------------------------------------------------
FIRE1            IPADDR      INSERVICE     2

Router# show ip slb firewallfarm details
FIRE1, hash = IPADDRPORT, state = INSERVICE, reals = 2
  FirewallTCP:
   sticky timer = 0, sticky subnet = 255.255.255.255
   idle = 3600, delay = 10, syns = 1965732, syn drop = 0 
   maxconns = 4294967295, conns = 597445, total conns = 1909512
  FirewallUDP:
   sticky timer = 0, sticky subnet = 255.255.255.255
   idle = 3600
   maxconns = 1, conns = 0, total conns = 1
  Real firewalls:
    10.1.1.3, weight = 10, OPERATIONAL, conns = 298823
    10.1.1.4, weight = 10, OPERATIONAL, conns = 298622
  Total connections = 597445

Sample IOS SLB Stateless Backup Configuration

The following commands enable the HSRP standby group 100 IP address, priority, preempt, timers, configure a name and authentication for Device A in Figure 9:

Router(config-if)# standby 100 ip 172.20.100.10
Router(config-if)# standby 100 priority 110
Router(config-if)# standby 100 preempt delay sync 20
Router(config-if)# standby 100 timers 5 15
Router(config-if)# standby 100 name Web-group1
Router(config-if)# standby 100 authentication Secret
Router(config-if)# exit
Router# 

Configuring IOS SLB Stateful Backup

Stateful backup enables IOS SLB to incrementally back up its load-balancing decisions, or "keep state," between primary and backup switches. The backup switch has its virtual servers in a dormant state until HSRP detects failover; then the backup (now primary) switch begins advertising virtual addresses and filtering flows. You can use HSRP to configure how quickly the failover is detected.

This enhancement provides IOS SLB with a one-to-one stateful or idle backup scheme. This means that only one instance of IOS SLB is handling client or server flows at a given time, and that there is at most one backup platform for each active IOS SLB switch.

Configuring Stateful Backup for Server Load Balancing

To configure stateful backup to keep state across primary and backup Layer 3 switches in a server load-balancing environment, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb vserver virtual_server-name

Configures a virtual server and initiates virtual server configuration mode.

Step 2 

Router(config-slb-vserver)# replicate casa 
listening-ip remote-ip port-number [interval] 
[password [0|7] password timeout]

Configures a stateful backup of IOS SLB decision tables to a backup switch. See the replicate casa (virtual server) command for more details.

The following commands configure stateful backup for virtual server RESTRICTED_HTTP using listening IP 10.10.3.132 and remote IP 10.10.99.3 over port 1032 and configures the password as PASS for Device A in Figure 14:

Router(config)# ip slb vserver RESTRICTED_HTTP
Router(config-slb-vserver)# virtual 10.10.10.12 tcp telnet
Router(config-slb-vserver)# replicate casa 10.10.3.132 10.10.99.3 1032 password PASS
Router(config-slb-vserver)# inservice standby virt 
Router(config-slb-vserver)#
.
(Information Deleted)

Configuring Stateful Backup for Firewall Load Balancing

To configure stateful backup to keep state across primary and backup Layer 3 switches in a firewall load-balancing environment, enter the following commands in order, beginning in global configuration mode:

 
Command
Description

Step 1 

Router(config)# ip slb firewallfarm 
firewallfarm-name

Configures a firewall farm and initiates firewall farm configuration mode.

Step 2 

Router(config-slb-fw)# replicate casa listening-ip 
remote-ip port-number [interval] [password [0|7] 
password timeout]

Configures a stateful backup of IOS SLB decision tables to a backup switch. See the replicate casa (firewall farm) command for more details.

Monitoring and Maintaining the IOS SLB Feature

To obtain and display runtime information about IOS SLB, use the following commands in EXEC mode:

Command
Purpose
Router# show ip dfp [agent subsystem_name] [detail]

Displays information about DFP agents. See the show ip dfp command for more details.

Router# show ip slb conns [vserver virtual_server-name | 
client ip-address | firewall firewallfarm-name] [detail]

Displays all connections handled by IOS SLB, or, optionally, only those connections associated with a particular virtual server or client. See the show ip slb conns command for more details.

Router# show ip slb dfp [agent agent_ip_address port-number | 
manager manager_ip_address | detail | weights]

Displays information about DFP and DFP agents, and about the weights assigned to real servers. See the show ip slb dfp command for more details.

Router# show ip slb firewallfarm [detail]

Displays information about firewall farms. See the show ip slb firewallfarm command for more details.

Router# show ip slb natpool [name pool_name] [detail]

Displays information about the IOS SLB NAT configuration. See the show ip slb natpool command for more details.

Router# show ip slb probe [name probe_name] [detail]

Displays information about HTTP and ping probes defined to IOS SLB. See the show ip slb probe command for more details.

Router# show ip slb reals [vserver virtual_server-name] 
[detail]

Displays information about the real servers defined to IOS SLB. See the show ip slb reals command for more details.

Router# show ip slb replicate

Displays information about the IOS SLB replication configuration. See the show ip slb replicate command for more details.

Router# show ip slb serverfarms [name serverfarm-name] 
[detail]

Displays information about the server farms defined to IOS SLB. See the show ip slb serverfarms command for more details.

Router# show ip slb stats

Displays IOS SLB statistics. See the show ip slb stats command for more details.

Router# show ip slb sticky [client ip-address]

Displays information about the sticky connections defined to IOS SLB. See the show ip slb sticky command for more details.

Router# show ip slb vserver [name virtual_server-name] 
[detail]

Displays information about the virtual servers defined to IOS SLB. See the show ip slb vserver command for more details.


Configuration Examples

This section provides real-world examples of IOS SLB configurations and includes the following sections:

Complete Example Configuration

Example of a Layer 3 Switch with ISL, VLAN, and BVI with GEC

Example of IOS SLB with Firewall Load Balancing

Example of IOS SLB with Server Load Balancing and Firewall Load Balancing

Example of IOS SLB with Multiple Firewall Farms

Example of IOS SLB with Probes

Example of a Layer 3 Switch Configured with IOS SLB

Example of IOS SLB with NAT

Example of an IOS Layer 3 Switch with HSRP

Examples of IOS SLB with Stateless Backup

Example of IOS SLB with Stateful Backup

Example of IOS SLB with Active Standby

Example of IOS SLB with Redistribution of Static Routes

Examples of IOS SLB with WAP Load Balancing

Examples of IOS SLB with Route Health Injection

Example of IOS SLB with Sticky Connections


Note The IP and network addresses in these examples are generic, so you must replace them with the actual addresses for your network.


Complete Example Configuration

The following example provides a complete configuration using the commands described in this feature module:

Router# show running-config
Building configuration...

Current configuration:
!
.
(Information Deleted)
.
ip slb probe PROBE2 http
 request method POST url /probe.cgi?all
 header Cookie Monster
 header Authorization Basic U2VtaXN3ZWV0OmNoaXBz
! 
ip slb serverfarm PUBLIC
 nat server
 real 10.1.1.1
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
 real 10.1.1.2
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
 real 10.1.1.3
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
 probe PROBE2
!
ip slb serverfarm RESTRICTED
 predictor leastconns
 bindid 309
 real 10.1.1.1
  weight 32
  maxconns 1000
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
 real 10.1.1.20
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
 real 10.1.1.21
  reassign 4
  faildetect numconns 16
  retry 120
  inservice
!
ip slb vserver PUBLIC_HTTP
 virtual 10.0.0.1 tcp www
 serverfarm PUBLIC
 no inservice
!
ip slb vserver RESTRICTED_HTTP
 virtual 10.0.0.2 tcp www
 serverfarm RESTRICTED
 no advertise
 sticky 60 group 1
 idle 120
 delay 5
 client 10.4.4.0 255.255.255.0
 synguard 3600000
 inservice
!

Example of a Layer 3 Switch with ISL, VLAN, and BVI with GEC

This example configuration focuses on both the Inter-Switch Link (ISL) and virtual LANs (VLANs), and on integrated routing and bridging (IRB) using a bridge-group virtual interface (BVI) over Gigabit EtherChannel (GEC). The Cisco proprietary ISL allows any Fast Ethernet port to be configured as a trunk. The Spanning-Tree Protocol detects and breaks loops on all the VLANs carried across the trunk.

The Gigabit Ethernet interface information applies to both two-port and eight-port Gigabit Ethernet interfaces for a Catalyst 8540 campus Layer 3 switch. This example also includes port snooping and Network Time Protocol (NTP) configurations.

!
ip subnet-zero
no ip domain-lookup
ip name-server 171.69.2.132
ip name-server 198.92.30.32
ip multicast-routing
ip dvmrp route-limit 20000
bridge irb
!
interface FastEthernet1
no ip address
no ip directed-broadcast
no keepalive
!
interface FastEthernet1.128
ip address 172.68.16.10 255.255.255.0
ip helper-address 172.68.16.15
no ip redirects
no ip directed-broadcast
ip pim dense-mode
ip multicast ttl-threshold 1
encapsulation isl 128
!
interface FastEthernet1.199
ip address 172.68.17.15 255.255.255.0
ip helper-address 172.68.16.16
ip helper-address 172.68.16.17
ip helper-address 172.68.16.18
no ip redirects
no ip directed-broadcast
ip pim dense-mode
ip multicast ttl-threshold 1
encapsulation isl 199
!
interface FastEthernet1.201
ip address 172.68.18.10 255.255.255.0
ip helper-address 172.68.16.16
ip helper-address 172.68.16.17
ip helper-address 172.68.16.18
no ip redirects
no ip directed-broadcast
ip pim dense-mode
ip multicast ttl-threshold 1
encapsulation isl 201
!
interface FastEthernet2
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet3
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet4
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet5
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet6
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet7
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet8
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet9
ip address 172.68.19.10 255.255.255.0
ip helper-address 172.68.16.16
ip helper-address 172.68.16.17
ip helper-address 172.68.16.18
no ip redirects
no ip directed-broadcast
ip pim dense-mode
ip multicast ttl-threshold 1
ip sdr listen
no keepalive
!
interface FastEthernet10
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
interface FastEthernet11
no ip address
no ip directed-broadcast
no keepalive
shutdown
!
.
(Information Deleted)
.
interface GigabitEthernet41
 snoop interface FastEthernet3 direction both
 snoop interface FastEthernet5 direction both
 snoop interface FastEthernet6 direction both
ip address 172.68.21.10 255.255.255.0
ip helper-address 172.68.16.19
ip helper-address 172.68.16.20
ip helper-address 172.68.16.21
!
interface GigabitEthernet42
    ip address 172.68.1.1 255.255.255.0
    no ip directed-broadcast
    ip pim sparse-dense-mode
!
interface BVI1
    ip address 171.201.1.2 255.255.255.0
    no ip directed-broadcast
    ip pim dense-mode
    no ip route-cache cef
!
interface Ethernet0
ip address 172.68.20.10 255.255.255.0
no ip directed-broadcast
!
router eigrp 170
 network 171.200.0.0
 network 171.201.0.0
 network 172.68.0.0
 network 172.69.0.0
 no auto-summary
!
router bgp 180
 network 172.68.1.0
 network 172.69.1.0
 no auto-summary
!
ip classless
!
bridge 1 protocol ieee
bridge 1 route ip
!
ip http server
!
line con 0
line aux 0
line vty 0 4
login
!
ntp clock-period 17181168
ntp update-calendar
ntp server 171.71.150.52
ntp server 171.69.4.143
ntp server 171.69.5.10
end

Example of IOS SLB with Firewall Load Balancing

Figure 3 shows a sample IOS SLB firewall load-balancing network with the following components:

Two firewalls with IP addresses as shown

An internal firewall load-balancing device on the secure side of the firewalls

An external firewall load-balancing device on the Internet side of the firewalls

One firewall farm named FIRE1, containing both firewalls

Figure 3 IOS SLB with Layer 3 Firewalls in Different Subnets

When you configure IOS SLB firewall load balancing, the load-balancing devices use route lookup to recognize flows destined for the firewalls. To enable route lookup, you must configure each device with the IP address of each firewall that will route flows to that device.

In the following firewall farm configuration samples:

The internal (secure side) firewall load-balancing device is configured with firewall IP addresses 10.1.3.1 and 10.1.4.1.

The external (Internet side) firewall load-balancing device is configured with firewall IP addresses 10.1.1.2 and 10.1.2.2.

Internal Firewall Load-Balancing Device

The following commands configure ping probe PROBE1, HTTP probe PROBE2, and firewall farm FIRE1, and associate the two real servers for the load-balancing device on the internal (secure) side of the firewall:

Router(config)# ip slb probe PROBE1 ping	; Ping probe
Router(config-slb-probe)# address 10.1.1.1	; IP address of other load-balancing device
Router(config-slb-probe)# faildetect 4
Router(config-slb-probe)# ip slb probe PROBE2 http	; HTTP probe
Router(config-slb-probe)# address 10.1.2.1	; IP address of other load-balancing device
Router(config-slb-probe)# expect status 401
Router(config-slb-probe)# ip slb firewallfarm FIRE1	; Firewall farm FIRE1
Router(config-slb-fw)# real 10.1.4.1	; First firewall
Router(config-slb-fw-real)# probe PROBE1
Router(config-slb-fw-real)# inservice	; Enable first firewall
Router(config-slb-fw-real)# real 10.1.3.1	; Second firewall
Router(config-slb-fw-real)# probe PROBE2
Router(config-slb-fw-real)# inservice	; Enable second firewall
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# inservice	; Turn on firewall load-balancing device

External Firewall Load-Balancing Device

The following commands configure ping probe PROBE1, HTTP probe PROBE2, and firewall farm FIRE1, and associate the two real servers for the load-balancing device on the external (Internet) side of the firewall:

Router(config)# ip slb probe PROBE1 ping	; Ping probe
Router(config-slb-probe)# address 10.1.4.2	; IP address of other load-balancing device
Router(config-slb-probe)# faildetect 4
Router(config-slb-probe)# ip slb probe PROBE2 http	; HTTP probe
Router(config-slb-probe)# address 10.1.3.2	; IP address of other load-balancing device
Router(config-slb-probe)# expect status 401
Router(config-slb-probe)# ip slb firewallfarm FIRE1	; Firewall farm FIRE1
Router(config-slb-fw)# real 10.1.1.2	; First firewall
Router(config-slb-fw-real)# probe PROBE1
Router(config-slb-fw-real)# inservice	; Enable first firewall
Router(config-slb-fw-real)# real 10.1.2.2	; Second firewall
Router(config-slb-fw-real)# probe PROBE2
Router(config-slb-fw-real)# inservice	; Enable second firewall
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# inservice	; Turn on firewall load-balancing device

Example of IOS SLB with Server Load Balancing and Firewall Load Balancing

Figure 4 shows a sample IOS SLB load-balancing network with server load balancing and firewall load balancing running together, and the following components:

Two real servers with IP addresses as shown

One server farm named PUBLIC, containing both real servers

Two firewalls with IP addresses as shown

One firewall farm named FIRE1, containing both firewalls

An internal IOS SLB device on the secure side of the firewalls, performing server load balancing and firewall load balancing

An external firewall load-balancing device on the Internet side of the firewalls

Figure 4 IOS SLB with Server Load Balancing and Firewall Load Balancing

In the following firewall farm configuration samples:

The internal (secure side) firewall load-balancing device is configured with firewall IP addresses 10.1.3.1 and 10.1.4.1.

The external (Internet side) firewall load-balancing device is configured with firewall IP addresses 10.1.1.2 and 10.1.2.2.

Internal Server and Firewall Load-Balancing Device

The following example shows the configuration for ping probes ABCPROBE and XYZPROBE, firewall farm FIRE1, and server farm PUBLIC for the load-balancing device on the internal (secure) side of the firewalls:

ip slb probe ABCPROBE ping
  address 10.1.1.1
ip slb probe XYZPROBE ping
  address 10.1.2.1
!
ip slb firewallfarm FIRE1
  real 10.1.4.1
    probe ABCPROBE
    inservice
  real 10.1.3.1
    probe XYZPROBE
    inservice
  inservice
!
ip slb serverfarm PUBLIC
  nat server
  real 10.2.1.1
    inservice
    real 10.2.1.2
    inservice
    real 10.2.1.2
    inservice
!
ip slb vserver HTTP1
  virtual 128.1.0.1 tcp www
  serverfarm PUBLIC
  idle 120
  delay 5
  inservice

External Firewall Load-Balancing Device

The following example shows the configuration for ping probes ABCPROBE and XYZPROBE and firewall farm FIRE1 for the load-balancing device on the external (Internet) side of the firewalls:

ip slb probe ABCPROBE ping
  address 10.1.4.2
  ip slb probe XYZPROBE ping
  address 10.1.3.2
  ip slb firewallfarm FIRE1
  real 10.1.1.2
    probe ABCPROBE
    inservice
    probe XYZPROBE
    inservice
  inservice

Example of IOS SLB with Multiple Firewall Farms

Figure 5 shows a sample IOS SLB load-balancing network with multiple firewall farms and the following components:

Four firewalls with IP addresses as shown

An internal firewall load-balancing device on the secure side of the firewalls

An external firewall load-balancing device on the Internet side of the firewalls

One firewall farm named ABCFARM, containing the two firewalls on the left.

One firewall farm named XYZFARM, containing the two firewalls on the right.

Figure 5 IOS SLB with Multiple Firewall Farms

In the following firewall farm configuration samples:

The internal (secure side) firewall load-balancing device is configured with firewall IP addresses 10.1.3.1 and 10.1.4.1.

The external (Internet side) firewall load-balancing device is configured with firewall IP addresses 10.1.1.2 and 10.1.2.2.

Internal Firewall Load-Balancing Device

The following commands configure ping probes ABCPROBE and XYZPROBE and firewall farms ABCFARM and XYZFARM for the load-balancing device on the internal (secure) side of the firewalls:

Router(config)# ip slb probe ABCPROBE ping
Router(config-slb-probe)# address 10.1.2.1
Router(config-slb-probe)# ip slb probe XYZPROBE ping
Router(config-slb-probe)# address 10.1.1.1
Router(config-slb-probe)# ip slb firewallfarm ABCFARM
Router(config-slb-fw)# access source 10.1.6.0 255.255.255.0
Router(config-slb-fw)# inservice
Router(config-slb-fw)# real 10.1.4.2
Router(config-slb-fw-real)# probe ABCPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# real 10.1.4.3
Router(config-slb-fw-real)# probe ABCPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# ip slb firewallfarm XYZFARM
Router(config-slb-fw)# access source 10.1.5.0 255.255.255.0
Router(config-slb-fw)# inservice
Router(config-slb-fw)# real 10.1.3.2
Router(config-slb-fw-real)# probe XYZPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# real 10.1.3.3
Router(config-slb-fw-real)# probe XYZPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# inservice

External Firewall Load-Balancing Device

The following commands configure ping probes ABCPROBE and XYZPROBE and firewall farms ABCFARM and XYZFARM for the load-balancing device on the external (Internet) side of the firewalls:

Router(config)# ip slb probe ABCPROBE ping
Router(config-slb-probe)# address 10.1.4.1
Router(config-slb-probe)# ip slb probe XYZPROBE ping
Router(config-slb-probe)# address 10.1.3.1
Router(config-slb-probe)# ip slb firewallfarm ABCFARM
Router(config-slb-fw)# access destination 10.1.6.0 255.255.255.0
Router(config-slb-fw)# inservice
Router(config-slb-fw)# real 10.1.2.2
Router(config-slb-fw-real)# probe ABCPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# real 10.1.2.3
Router(config-slb-fw-real)# probe ABCPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# ip slb firewallfarm XYZFARM
Router(config-slb-fw)# access destination 10.1.5.0 255.255.255.0
Router(config-slb-fw)# inservice
Router(config-slb-fw)# real 10.1.1.2
Router(config-slb-fw-real)# probe XYZPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# real 10.1.1.3
Router(config-slb-fw-real)# probe XYZPROBE
Router(config-slb-fw-real)# inservice
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# inservice

Example of IOS SLB with Probes

Figure 6 shows an example configuration with IOS SLB real server connections configured as part of a server farm, focusing on using ping and HTTP probes to monitor applications being server load-balanced.

Figure 6 Sample Ping and HTTP Probe Topology

:

The topology shown in Figure 6 is a heterogeneous server farm servicing a single virtual server. Following are the configuration statements for this topology, including a ping probe named PROBE1 and an HTTP probe named PROBE2:

! Configure ping probe PROBE1, change CLI to IOS SLB probe configuration mode
Router(config)# ip slb probe PROBE1 ping
! Configure probe to receive responses from IP address 13.13.13.13
Router(config-slb-probe)# address 13.13.13.13
! Configure unacknowledged ping threshold to 16
Router(config-slb-probe)# faildetect 16
! Configure ping probe timer interval to transmit every 11 seconds
Router(config-slb-probe)# interval 11
! Configure HTTP probe PROBE2
Router(config-slb-probe)# ip slb probe PROBE2 http
! Configure request method as POST, set URL as /probe.cgi?all
Router(config-slb-probe)# request method post url /probe.cgi?all
! Configure header Cookie
Router(config-slb-probe)# header Cookie Monster
! Configure basic authentication username and password
Router(config-slb-probe)# credentials Semisweet chips
! Exit to global configuration mode
Router(config-slb-probe)# exit
! Enter IOS SLB server farm configuration mode for server farm PUBLIC
Router(config)# ip slb serverfarm PUBLIC
! Configure NAT server and real servers on the server farm
Router(config-slb-sfarm)# nat server
Router(config-slb-sfarm)# real 10.1.1.1
Router(config-slb-sfarm)#  inservice
Router(config-slb-sfarm)# real 10.1.1.2
Router(config-slb-sfarm)#  inservice
Router(config-slb-sfarm)# real 10.1.1.3
Router(config-slb-sfarm)#  inservice
Router(config-slb-sfarm)# real 10.1.1.4
Router(config-slb-sfarm)#  inservice
Router(config-slb-sfarm)# real 10.1.1.5
Router(config-slb-sfarm)#  inservice
! Configure ping probe on the server farm
Router(config-slb-sfarm)# probe PROBE1
! Configure HTTP probe on the server farm
Router(config-slb-sfarm)# probe PROBE2
Router(config-slb-sfarm)# end

Example of a Layer 3 Switch Configured with IOS SLB

Figure 7 shows an example configuration with IOS SLB server connections configured as part of a server farm, using real and virtual servers over Fast Ethernet interfaces.

Figure 7 Network Configuration for IOS SLB

As shown in the following sample configuration, the example topology has three public web servers and two restricted web servers for privileged clients in subnet 10.4.4.x. The public web servers are weighted according to their capacity, with server 10.1.1.2 having the lowest capacity and having a connection limit imposed on it. The restricted web servers are configured as members of the same sticky group, so that HTTP connections and Secure Socket Layer (SSL) connections from the same client use the same real server.

The network configuration to provide the previously described IOS SLB functionality follows:

! Unrestricted web server farm
ip slb serverfarm PUBLIC
! Use weighted least connections algorithm
  predictor leastconns
! First real server
  real 10.1.1.1
    weight 16
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Second real server
  real 10.1.1.2
    weight 4
!  Restrict maximum number of connections
    maxconns 1000
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Third real server
  real 10.1.1.3
    weight 24
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Restricted web server farm
ip slb serverfarm RESTRICTED
! Use weighted least connections algorithm
  predictor leastconns
! First real server
  real 10.1.1.20
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Second real server
  real 10.1.1.21
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
!
! Unrestricted web virtual server
ip slb vserver PUBLIC_HTTP
! Handle HTTP requests
  virtual 10.0.0.1 tcp www
! Use public web server farm
  serverfarm PUBLIC
  idle 120
  delay 5
  inservice
!
! Restricted HTTP virtual server
ip slb vserver RESTRICTED_HTTP
! Handle HTTP requests
  virtual 10.0.0.1 tcp www
! Use restricted web server farm
  serverfarm RESTRICTED
! Only allow clients from 10.4.4.x
  client 10.4.4.0 255.255.255.0
! Couple connections with RESTRICTED_SSL
  sticky 60 group 1
  idle 120
  delay 5
  inservice
!
! Restricted SSL virtual server
ip slb vserver RESTRICTED_SSL
! Handle SSL requests
  virtual 10.0.0.1 tcp https
! Use restricted web server farm
  serverfarm RESTRICTED
! Only allow clients from 10.4.4.x
  client 10.4.4.0 255.255.255.0
! Couple connections with RESTRICTED_WEB
  sticky 60 group 1
  idle 120
  delay 5
  inservice

Example of IOS SLB with NAT

Figure 8 shows an example configuration with IOS SLB real server connections configured as part of a server farm, focusing on the configuration of the NAT server and address pool of clients.

Figure 8 Sample IOS SLB NAT Topology

The topology in Figure 8 has four web servers, configured as follows:

Servers 1, 2, and 3 are running single HTTP server applications listening on port 80.

Server 4 has multiple HTTP server applications listening on ports 8080, 8081, and 8082.

Server 1 and Server 2 are load-balanced using Switch A, which is performing server address translation.

Server 3 and Server 4 are load-balanced using Switch B and Switch C. These two switches are performing both server and client address translation since there are multiple paths between the clients and the servers. These switches also must perform server port translation for HTTP packets to and from Server 4.

Switch A Configuration Statements

ip slb serverfarm FARM1
! Translate server addresses
  nat server
! Server 1 port 80
  real 10.1.1.1
    reassign 2
    faildetect numconns 4 numclients 2
    retry 20
    inservice
! Server 2 port 80
  real 10.2.1.1
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
!
ip slb vserver HTTP1
! Handle HTTP (port 80) requests
  virtual 128.1.0.1 tcp www
  serverfarm FARM1
  idle 120
  delay 5
  inservice

Switch B Configuration Statements

ip slb natpool web-clients 128.3.0.1 128.3.0.254
! NAT address pool for clients
ip slb serverfarm FARM2
! Translate server addresses
  nat server
! Translate client addresses
  nat client web-clients
! Server 3 port 80
  real 10.3.1.1
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8080
  real 10.4.1.1 port 8080
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8081
  real 10.4.1.1 port 8081
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8082
  real 10.4.1.1 port 8082
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
!
ip slb vserver HTTP2
! Handle HTTP (port 80) requests
  virtual 128.2.0.1 tcp www
  serverfarm FARM2
  idle 120
  delay 5
  inservice

Switch C Configuration Statements

ip slb natpool web-clients 128.5.0.1 128.5.0.254
! NAT address pool for clients
ip slb serverfarm FARM2
! Translate server addresses
  nat server
! Translate client addresses
  nat client web-clients 
! Server 3 port 80
  real 10.3.1.1
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8080
  real 10.4.1.1 port 8080
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8081
  real 10.4.1.1 port 8081
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
! Server 4 port 8082
  real 10.4.1.1 port 8082
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
!
ip slb vserver HTTP2
! Handle HTTP (port 80) requests
  virtual 128.4.0.1 tcp www
  serverfarm FARM2
  idle 120
  delay 5
  inservice

Example of an IOS Layer 3 Switch with HSRP

This example configuration for an IOS Layer 3 switch focuses on the HSRP, which provides high network availability. HSRP makes network topology changes transparent to the host. The active router is monitored by other standby routers, and as soon as an active router becomes unavailable, the standby router takes its place. Helper addresses facilitate connectivity by forwarding certain broadcasts to a target server.

Figure 9 shows the topology of an IP network with two Layer 3 switches configured for HSRP.

Figure 9 HSRP Example Network Topology

In this network:

An IP helper address identifies the Dynamic Host Configuration Protocol (DHCP) server IP address. This configuration also includes configuration for IP multicast, Distance Vector Multicast Routing Protocol (DVMRP), tunneling, and Protocol Independent Multicast (PIM) in sparse mode.

Device A is the active HSRP Layer 3 switch.

All hosts accessing the network use the IP address of the virtual servers (in this case, 10.10.10.12 or 10.10.10.18).

The configurations shown use the RIP routing protocol, but HSRP can be used with any other routing protocol supported by the Cisco IOS software, such as Open Shortest Path First (OSPF).


Note Some configurations that use HSRP still require a routing protocol for convergence when a topology change occurs. The standby Layer 3 switch becomes active, but connectivity does not occur until convergence occurs.


If the connection between Device A and the client accessing virtual server IP address 10.10.10.12 tcp 23 or 10.10.10.18 tcp 23 fails, fast-converging routing protocols, such as OSPF and the Enhanced Interior Gateway Routing Protocol (Enhanced IGRP), can respond within seconds, ensuring that Device B is prepared to transfer packets that would have gone through Device A.

Device A (Active) Configuration Statements

hostname Device A
!
ip slb serverfarm ServerGroup1
   real 172.20.100.3
   inservice
   real 172.20.100.4
   inservice
!
ip slb serverfarm ServerGroup2
   real 172.20.200.3
   inservice
   real 172.20.200.4
   inservice
!
ip slb vserver VS1
   virtual 10.10.10.12 tcp 23
   serverfarm ServerGroup1
   in-service standby Web-Group1
!
ip slb vserver VS2
   virtual 10.10.10.18 tcp 23
   serverfarm ServerGroup2
   in-service standby Web-Group2
!
ip routing
router rip
network 172.20.0.0
!
interface vlan100
ip address 172.20.100.1 255.255.255.0
standby 100 ip 172.20.100.10
standby 100 priority 110
standby 100 preempt delay sync 20
standby 100 timers 5 15
standby 100 name Web-Group1
standby 100 authentication Secret
!
interface vlan200
ip address 172.20.200.1 255.255.255.0
standby 200 ip 172.20.200.10
standby 200 priority 110
standby 200 preempt delay sync 20
standby 200 timers 5 15
standby 200 name Web-Group2
standby 200 authentication Covert
!

Device B (Standby) Configuration Statements

hostname Device B
!
ip slb serverfarm ServerGroup1
   real 172.20.100.3
   inservice
   real 172.20.100.4
   inservice
!
ip slb serverfarm ServerGroup2
   real 172.20.200.3
   inservice
   real 172.20.200.4
   inservice
!
ip slb vserver VS1
   virtual 10.10.10.12 tcp 23
   serverfarm ServerGroup1
   in-service standby Web-Group1
!
ip slb vserver VS2
   virtual 10.10.10.18 tcp 23
   serverfarm ServerGroup2
   in-service standby Web-Group2
!
ip routing
router rip
network 172.20.0.0
!
interface vlan100
ip address 172.20.100.2 255.255.255.0
standby 100 ip 172.20.100.10
standby 100 preempt delay sync 20
standby 100 timers 5 15
standby 100 name Web-Group1
standby 100 authentication Secret
!
interface vlan200
ip address 172.20.200.2 255.255.255.0
standby 200 ip 172.20.200.10
standby 200 preempt delay sync 20
standby 200 timers 5 15
standby 200 name Web-Group2
standby 200 authentication Covert

Description of Configuration

The standby ip interface configuration command enables HSRP and establishes 10.10.10.12 and 10.10.10.18 as the IP addresses of the virtual servers. The configurations of both Layer 3 switches include this command so that both switches share the same virtual IP address. The numbers 100 and 200 establish Hot Standby groups 100 and 200. (If you do not specify a group number, the default is group 0.) The numbers 100 and 200 in the following commands indicate that they apply to Hot Standby groups 100 and 200, respectively. The configuration for at least one of the Layer 3 switches in the Hot Standby group must specify the IP address of the virtual server; specifying the IP address of the virtual router is optional for other routers in the same Hot Standby group.

The standby preempt interface configuration command allows the Layer 3 switch to become the active switch when its priority is higher than all other HSRP-configured switches in this Hot Standby group. The configurations of both switches include this command so that each can be the standby Layer 3 switch for the other switch. If you do not use the standby preempt command in the configuration for a Layer 3 switch, that switch cannot become the active Layer 3 switch.

The standby priority interface configuration command sets the Layer 3 switch's HSRP priority to 110, which is higher than the default priority of 100. Only the configuration of Device A includes this command, which makes Device A the default active Layer 3 switch.

The standby timers interface configuration command sets the interval (in seconds) between hello messages (called the hello time) to 5 seconds, and sets the interval (in seconds) that a Layer 3 switch waits before it declares the active Layer 3 switch to be down (called the hold time) to 8 seconds. (The defaults are 3 seconds and 10 seconds, respectively.) To modify the default values, you must configure each Layer 3 switch to use the same hello time and hold time.

The standby name interface configuration command associates the IOS SLB interface with an HSRP group name (in this case, Web-Group1 or Web-Group2), previously specified on an inservice (server farm virtual server) command.

The standby authentication interface configuration command establishes an authentication string whose value is an unencrypted eight-character string that is incorporated in each HSRP multicast message. This command is optional. If you choose to use it, each HSRP-configured Layer 3 switch in the group should use the same string so that each switch can authenticate the source of the HSRP messages that it receives.

Examples of IOS SLB with Stateless Backup

There are several different ways in which you can configure IOS SLB stateless backup. The differences between the configurations depend on the networking capabilities of your load balancing devices, and on the capabilities of the distribution devices that direct client traffic to those load balancing devices.

If a load balancing device is capable of Layer 2 switching and VLAN trunking (such as the Catalyst 6000 Family Switch), you can wire the device directly to its real servers, and it can handle outbound flows from the real servers while acting as a standby for IOS SLB. HSRP is used on the server-side VLANs of the load balancing device, with the real servers routing to the HSRP address.

If a load balancing device is not capable of both Layer 2 switching and VLAN trunking, you must connect it and its real servers to a Layer 2 switch. This configuration is required in order to use HSRP on the server-side VLANs.

If a distribution device is capable of Layer 3 switching, it can use route redistribution to direct flows to the active load balancing device.

If a distribution device is capable of Layer 2 switching, it can use client-side HSRP on the load balancing device to direct flows to the active load balancing device.

While HSRP offers faster failover times, routing converges quickly enough for most configurations. If you use both client-side and server-side HSRP on the load balancing devices, you must use HSRP interface tracking and priorities to synchronize the client-side and server-side HSRP groups.

This section contains the following examples, illustrating several different IOS SLB stateless backup configurations:

Example with Dynamic Routing and Trunking

Example with Dynamic Routing and No Trunking

Example with Static Routing and Trunking

Example with Static Routing and No Trunking


Note Stateful backup is omitted from these examples in the interest of simplicity. To see an example that uses stateful backup, see the "Example of IOS SLB with Stateful Backup" section.


Example with Dynamic Routing and Trunking

Figure 10 shows a sample IOS SLB stateless backup configuration with the following characteristics:

The IP address for real server 1 is 10.10.1.3, and for real server 2 is 10.10.1.4, routed to clients through 10.10.1.100.

The IP address for the virtual server is 10.10.14.1.

The IP address for VLAN 1 is 10.10.1.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 2 is 10.10.2.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 3 is 10.10.3.0, with a subnet mask of 255.255.255.0.

The distribution device uses EIGRP to learn the route to 10.10.14.1 via either 10.10.2.1 or 10.10.3.1, depending on which IOS SLB is active.

Figure 10 Stateless Backup with Layer 3 and Trunking

SLB 1 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4 numclients 2
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  switchport
  switchport vlan 1
int Eth2
  ip address 10.10.2.1 255.255.255.0
int vlan 1
  ip address 10.10.1.1 255.255.255.0
  standby ip 10.10.1.100
  standby priority 10 preempt delay sync 20
  standby name SERVER
  standby track Eth2
router eigrp 666
  redistribute static
  network 10.0.0.0

SLB 2 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Gig1
  no ip address
  switchport
  switchport trunk encapsulation isl
int Eth1
  switchport
  switchport vlan 1
int Eth2
  ip address 10.10.3.1 255.255.255.0
int vlan 1
  ip address 10.10.1.2 255.255.255.0
  standby ip 10.10.1.100
  standby priority 5 preempt delay sync 20
  standby name SERVER
router eigrp 666
  redistribute static
  network 10.0.0.0

Example with Dynamic Routing and No Trunking

Figure 11 shows a sample IOS SLB stateless backup configuration with the following characteristics:

The IP address for real server 1 is 10.10.1.3, and for real server 2 is 10.10.1.4, routed to clients through 10.10.1.100.

The IP address for the virtual server is 10.10.14.1.

The IP address for Subnet 2 is 10.10.2.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 3 is 10.10.3.0, with a subnet mask of 255.255.255.0.

The distribution device uses EIGRP to learn the route to 10.10.14.1 via either 10.10.2.2 or 10.10.3.2, depending on which IOS SLB is active.

Figure 11 Stateless Backup with Layer 3 and No Trunking

SLB 1 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  ip address 10.10.1.1 255.255.255.0
  standby ip 10.10.1.100
  standby priority 10 preempt delay sync 20
  standby name SERVER
  standby track Eth2
int Eth2
  ip address 10.10.2.1 255.255.255.0
router eigrp 666
  redistribute static
  network 10.0.0.0

SLB 2 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  ip address 10.10.1.2 255.255.255.0
  standby ip 10.10.1.100
  standby priority 5 preempt delay sync 20
  standby name SERVER
int Eth2
  ip address 10.10.3.1 255.255.255.0
router eigrp 666
  redistribute static
  network 10.0.0.0

Example with Static Routing and Trunking

Figure 12 shows a sample IOS SLB stateless backup configuration with the following characteristics:

The IP address for real server 1 is 10.10.1.3, and for real server 2 is 10.10.1.4, routed to clients through 10.10.1.100.

The IP address for the virtual server is 10.10.14.1.

The IP address for VLAN 1 is 10.10.1.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 2 is 10.10.2.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 3 is 10.10.3.0, with a subnet mask of 255.255.255.0.

The configuration uses static routing to the HSRP route on the distribution device.

Figure 12 Stateless Backup with Layer 2 and Trunking

SLB 1 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  switchport
  switchport vlan 1
int Eth2
  ip address 10.10.2.1 255.255.255.0
  standby ip 10.10.2.100
  standby priority 10 preempt delay sync 20
  standby track vlan1
int vlan 1
  ip address 10.10.1.1 255.255.255.0
  standby ip 10.10.1.100
  standby priority 10 preempt delay sync 20
  standby name SERVER
  standby track Eth2

SLB 2 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Gig1
  no ip address
  switchport
  switchport trunk encapsulation isl
int Eth1
  switchport
  switchport vlan 1
int Eth2
  ip address 10.10.2.2 255.255.255.0
  standby ip 10.10.2.100
  standby priority 5 preempt delay sync 20
int vlan 1
  ip address 10.10.1.2 255.255.255.0
  standby ip 10.10.1.100
  standby priority 5 preempt delay sync 20
  standby name SERVER

Distribution Device Configuration Statements

int Eth1
  switchport
  switchport distribution vlan 2
int Eth2
  switchport
  switchport distribution vlan 2
int vlan2
  ip address 10.10.2.3 255.255.255.0
  no shut
ip route 10.10.14.1 255.255.255.255 10.10.2.100

Example with Static Routing and No Trunking

Figure 13 shows a sample IOS SLB stateless backup configuration with the following characteristics:

The IP address for real server 1 is 10.10.1.3, and for real server 2 is 10.10.1.4, routed to clients through 10.10.1.100.

The IP address for the virtual server is 10.10.14.1.

The IP address for Subnet 2 is 10.10.2.0, with a subnet mask of 255.255.255.0.

The IP address for Subnet 3 is 10.10.3.0, with a subnet mask of 255.255.255.0.

The configuration uses static routing to the HSRP route on the distribution device.

Figure 13 Stateless Backup with Layer 2 and No Trunking

SLB 1 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  ip address 10.10.1.1 255.255.255.0
  standby ip 10.10.1.100
  standby priority 10 preempt delay sync 20
  standby name SERVER
  standby track eth2
int Eth2
  ip address 10.10.2.1 255.255.255.0
  standby ip 10.10.2.100
  standby priority 10 preempt delay sync 20
  standby track eth1

SLB 2 Configuration Statements

ip slb serverfarm SF1
  real 10.10.1.3
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
  real 10.10.1.4
    reassign 2
    faildetect numconns 4
    retry 20
    inservice
ip slb vserver VS1
  virtual 10.10.14.1 tcp www
  serverfarm SF1
  idle 120
  delay 5
  inservice standby SERVER
...
int Eth1
  ip address 10.10.1.2 255.255.255.0
  standby ip 10.10.1.100
  standby priority 5 preempt delay sync 20
  standby name SERVER
int Eth2
  ip address 10.10.2.2 255.255.255.0
  standby ip 10.10.2.100
  standby priority 5 preempt delay sync 20

Distribution Device Configuration Statements

int Eth1
  switchport
  switchport distribution vlan 2
int Eth2
  switchport
  switchport distribution vlan 2
int vlan2
  ip address 10.10.2.3 255.255.255.0
  no shut
ip route 10.10.14.1 255.255.255.255 10.10.2.100

Example of IOS SLB with Stateful Backup

This example configuration focuses on the IOS SLB real server connections configured as part of a server farm, with real and virtual servers over Fast Ethernet interfaces configured with stateful backup standby connections.

Figure 14 is an example of a stateful backup configuration, using HSRP on both the client and server sides to handle failover. The real servers route outbound flows to 10.10.3.100, which is the HSRP address on the server side interfaces. The client (or access router), routes to the virtual IP address (10.10.10.12) through 10.10.2.100, HSRP address on the client side.

Notice the loopback interfaces configured on both boxes for the exchange of these messages. Each IOS SLB should also be given duplicate routes to the other switch loopback address. This allows replication messages to flow despite an interface failure.


Note To allow HSRP to function properly, set spantree portfast must be configured on any Layer 2 device between the IOS SLB switches.


Figure 14 IOS SLB Stateful Environment

Switch SLB1 Configuration Statements

ip slb serverfarm SF1
  nat server
  real 10.10.3.1
   inservice
  real 10.10.3.2
   inservice
  real 10.10.3.3
   inservice
!
ip slb vserver VS1
  virtual 10.10.10.12 tcp telnet
  serverfarm SF1
  replicate casa 10.10.99.132 10.10.99.99 1024 password PASS
  inservice standby virt
!
interface Loopback1
  ip address 10.10.99.132 255.255.255.255
!
interface FastEthernet1
  ip address 10.10.3.132 255.255.255.0
  no ip redirects
  no ip mroute-cache
  standby priority 5 preempt delay sync 20
  standby name out
  standby ip 10.10.3.100
  standby track FastEthernet2
!
interface FastEthernet2
  ip address 10.10.2.132 255.255.255.0
  no ip redirects
  standby priority 5 preempt delay sync 20
  standby name virt
  standby ip 10.10.2.100
  standby track FastEthernet1

Switch SLB2 Configuration Statements

ip slb serverfarm SF1
  nat server
  real 10.10.3.1
   inservice
  real 10.10.3.2
   inservice
  real 10.10.3.3
   inservice
!
ip slb vserver VS1
  virtual 10.10.10.12 tcp telnet
  serverfarm SF1
  replicate casa 10.10.99.99 10.10.99.132 1024 password PASS
  inservice standby virt
!
interface Loopback1
  ip address 10.10.99.99 255.255.255.255
!
interface FastEthernet2
  ip address 10.10.2.99 255.255.255.0
  no ip redirects
  no ip route-cache
  no ip mroute-cache
  standby priority 10 preempt delay sync 20
  standby name virt
  standby ip 10.10.2.100
  standby track FastEthernet3
!
interface FastEthernet3
  ip address 10.10.3.99 255.255.255.0
  no ip redirects
  no ip route-cache
  no ip mroute-cache
  standby priority 10 preempt delay sync 20
  standby name out
  standby ip 10.10.3.100
  standby track FastEthernet2

Example of IOS SLB with Active Standby

Figure 15 shows an IOS SLB network configured for active standby, with two IOS SLB devices load-balancing the same virtual IP address while backing up each other. If either device fails, the other takes over its load via normal HSRP failover and IOS SLB stateless redundancy.

Figure 15 IOS SLB Active Standby

The sample network configuration in Figure 15 has the following characteristics:

SLB 1 balances servers 1A and 1B and SLB 2 balances 2A and 2B.

A single virtual IP address (10.10.10.12 for web) is supported across the two IOS SLB devices.

Client traffic is divided in an access router, sending clients with even IP addresses to HSRP1 (10.10.5.100) and clients with odd IP addresses to HSRP2 (10.10.2.100). SLB 1 is configured as primary for clients with odd IP addresses, and SLB 2 is primary for clients with even IP addresses.

The IOS SLB devices balance the traffic to disjoint sets of real servers. (If client NAT was used in this example, this would not be a requirement).

Each set of real servers has a default gateway configured to its IOS SLB device.

The HSRP address on VLAN 105 is 10.10.5.100. The HSRP address on VLAN 102 is 10.10.2.100.

SLB 1 Configuration Statements

ip slb serverfarm EVEN
 nat server
 real 10.10.3.2
  reassign 2
  faildetect numconns 4 numclients 2
  retry 20
  inservice
 real 10.10.3.3
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
!
ip slb serverfarm ODD
 nat server
 real 10.10.3.2
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
 real 10.10.3.3
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
!
ip slb vserver EVEN	; Same EVEN virtual server as in SLB 2
 virtual 10.10.10.12 tcp www
 serverfarm EVEN
 client 0.0.0.0 0.0.0.1
 idle 120
 delay 5
 inservice standby STANDBY_EVEN	; See standby name in Ethernet 3/3 below
!
ip slb vserver ODD	; Same ODD virtual server as in SLB 2
 virtual 10.10.10.12 tcp www
 serverfarm ODD
 client 0.0.0.1 0.0.0.1
 idle 120
 delay 5
 inservice standby STANDBY_ODD	; See standby name in Ethernet 3/2 below
!
interface Ethernet3/2
 ip address 10.10.5.132 255.255.255.0
 standby priority 20 preempt delay sync 20
 standby name STANDBY_ODD	; See standby name in SLB 2, Ethernet 3/5
 standby ip 10.10.5.100
!
interface Ethernet3/3
 ip address 10.10.2.132 255.255.255.0
 standby priority 10
 standby name STANDBY_EVEN	; See standby name in SLB 2, Ethernet 3/1
 standby ip 10.10.2.100

SLB 2 Configuration Statements

ip slb serverfarm EVEN
 nat server
 real 10.10.3.4
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
 real 10.10.3.5
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
!
ip slb serverfarm ODD
 nat server
 real 10.10.3.4
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
 real 10.10.3.5
  reassign 2
  faildetect numconns 4
  retry 20
  inservice
!
ip slb vserver EVEN	; Same EVEN virtual server as in SLB 1
 virtual 10.10.10.12 tcp www
 serverfarm EVEN
 client 0.0.0.0 0.0.0.1
 idle 120
 delay 5
 inservice standby STANDBY_EVEN	; See standby name in Ethernet 3/1 below
!
ip slb vserver ODD	; Same ODD virtual server as in SLB 1
 virtual 10.10.10.12 tcp www
 serverfarm ODD
 client 0.0.0.1 0.0.0.1
 idle 120
 delay 5
 inservice standby STANDBY_ODD	; See standby name in Ethernet 3/5 below
!
interface Ethernet3/1
 ip address 10.10.2.128 255.255.255.0
 standby priority 20 preempt delay sync 20
 standby name STANDBY_EVEN	; See standby name in SLB 1, Ethernet 3/3
 standby ip 10.10.2.100
!
interface Ethernet3/5
 ip address 10.10.5.128 255.255.255.0
 standby priority 10 preempt delay sync 20
 standby name STANDBY_ODD	; See standby name in SLB 1, Ethernet 3/2
 standby ip 10.10.5.100

Access Router Configuration Statements

interface Ethernet0/0
 ip address 10.10.5.183 255.255.255.0
 no ip directed-broadcast
 no ip route-cache
 no ip mroute-cache
!
interface Ethernet0/1
 ip address 10.10.2.183 255.255.255.0
 no ip directed-broadcast
 no ip route-cache
 no ip mroute-cache
!
interface Ethernet0/2
 ip address 10.10.6.183 255.255.255.0
 no ip directed-broadcast
 no ip route-cache
 no ip mroute-cache
 ip policy route-map virts
!
access-list 100 permit ip 0.0.0.1 255.255.255.254 host 10.10.10.12
access-list 101 permit ip 0.0.0.0 255.255.255.254 host 10.10.10.12
route-map virts permit 10
match ip address 100
set ip next-hop 10.10.5.100
!
route-map virts permit 15
match ip address 101
set ip next-hop 10.10.2.100
!

Example of IOS SLB with Redistribution of Static Routes

Figure 16 shows an IOS SLB network configured to distribute static routes to a virtual server's IP address. The route to the address is added to the routing table as static if you advertise the address when you bring the virtual server into service (using the inservice command). See the advertise command for more details about advertising virtual server IP addresses.

Because the routing configuration varies from protocol to protocol, sample configurations for several different routing protocols are given.

Figure 16 IOS SLB Redistribution of Static Routes

Routing Information Protocol (RIP)

Following is the RIP static route redistribution configuration for the IOS SLB switch shown in Figure 16:

router rip
 redistribute static
 network 10.0.0.0
 network 8.0.0.0

Following is the RIP static route redistribution configuration for the access router that is listening for routing updates shown in Figure 16:

router rip
 network 10.0.0.0
 network 8.0.0.0

Open Shortest Path First (OSPF)

Following is the OSPF static route redistribution configuration for the IOS SLB switch shown in Figure 16:

router ospf 1
 redistribute static subnets
 network 10.10.6.217 0.0.0.0 area 0
 network 8.8.8.0 0.0.0.255 area 0

Following is the OSPF static route redistribution configuration for the access router that is listening for routing updates shown in Figure 16:

router ospf 1
 network 10.10.6.2 0.0.0.0 area 0
 network 8.8.8.0 0.0.0.255 area 0

Interior Gateway Routing Protocol (IGRP)

Following is the IGRP static route redistribution configuration for the IOS SLB switch shown in Figure 16:

router igrp 1
 redistribute connected
 redistribute static
 network 8.0.0.0
 network 10.0.0.0

Following is the IGRP static route redistribution configuration for the access router that is listening for routing updates shown in Figure 16:

router igrp 1
 network 8.0.0.0
 network 10.0.0.0

Enhanced Interior Gateway Routing Protocol (Enhanced IGRP)

Following is the Enhanced IGRP static route redistribution configuration for the IOS SLB switch shown in Figure 16:

router eigrp 666
 redistribute static
 network 10.0.0.0
 network 8.0.0.0

Following is the Enhanced IGRP static route redistribution configuration for the access router that is listening for routing updates shown in Figure 16:

router eigrp 666
 network 10.0.0.0
 network 8.0.0.0

Examples of IOS SLB with WAP Load Balancing

Figure 17 shows an IOS SLB network configured to balance WAP flows. In this example:

WAP flows are balanced between WAP gateways 10.10.2.1, 10.10.2.2, and 10.10.2.3.

The clients connect to 10.10.1.1, the IOS SLB virtual server address.

For a given session, load-balancing decisions change if the connection idles longer than the virtual server's idle connection timer (3000 seconds in this example).

Figure 17 IOS SLB with WAP Load Balancing

There are two ways to configure IOS SLB load balancing for WAP:

To load balance sessions running in connection-oriented WSP mode, define a WSP probe and use WAP load balancing. WAP load balancing requires a WAP virtual server configured on one of the WAP ports.

To load balance sessions running in connectionless WSP, connectionless secure WSP, and connection-oriented secure WSP modes, define a ping or WSP probe and use standard UDP load balancing with a low idle timer.

Example with WAP Load Balancing

The following commands configure the IOS SLB device shown in Figure 17 to balance WAP flows on UDP port 9201 (WSP/WTP/UDP):

Router(config)# ip slb probe PROBE3 wsp
Router(config-slb-probe)# url http://localhost/test.txt
!
Router(config)# ip slb serverfarm WAPFARM
Router(config-slb-sfarm)# nat server
Router(config-slb-sfarm)# real 10.10.2.1
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# real 10.10.2.2
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# real 10.10.2.3
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# probe PROBE3
!
Router(config)# ip slb vserver VSERVER
Router(config-slb-vserver)# virtual 10.10.1.1 udp 9201
Router(config-slb-vserver)# serverfarm WAPFARM
Router(config-slb-vserver)# idle 3000
Router(config-slb-vserver)# inservice

Example with UDP Load Balancing

The following commands configure the IOS SLB device shown in Figure 17 to balance WAP flows on UDP port 9203 (WSP/WTP/WTLS/UDP):

Router(config)# ip slb probe PROBE1 ping
!
Router(config)# ip slb serverfarm WAPFARM
Router(config-slb-sfarm)# nat server
Router(config-slb-sfarm)# real 10.10.2.1
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# real 10.10.2.2
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# real 10.10.2.3
Router(config-slb-sfarm)# inservice
Router(config-slb-sfarm)# probe PROBE1
!
Router(config)# ip slb vserver VSERVER
Router(config-slb-vserver)# virtual 10.10.1.1 udp 9203
Router(config-slb-vserver)# serverfarm WAPFARM
Router(config-slb-vserver)# idle 3000
Router(config-slb-vserver)# inservice

Examples of IOS SLB with Route Health Injection

This section contains the following examples, illustrating several different IOS SLB route health injection configurations:

Example with Two Distributed Sites with One Web Server Each

Example with Two Distributed Sites with Two Web Servers Each

Example with Two Distributed Sites with One Web Server and a Backup IOS SLB Switch Each

Example with Two Distributed Sites with One Web Server Each

Figure 18 shows an IOS SLB network configured with route health injection with the following characteristics:

Both IOS SLB devices are configured with the same virtual IP address.

Each IOS SLB device has a server farm containing only the locally attached web server as a real server.

The path to SLB A has the lower weight.

Figure 18 Two Distributed Sites with One Web Server Each

When both web servers in Figure 18 are operational, the client router receives the host route from both IOS SLB devices.

If Web Server A fails, the virtual server for the virtual IP address on SLB A enters FAILED state and stops advertising the host route for the virtual IP address. The client router then begins using the route to SLB B.

When Web Server A is again available, the virtual server again advertises the host route for the virtual IP address, and the client router begins using SLB A.

Example with Two Distributed Sites with Two Web Servers Each

Figure 19 shows an IOS SLB network configured with route health injection with the following characteristics:

Both IOS SLB devices are configured with the same virtual IP address.

Each IOS SLB device has a server farm containing two locally attached web servers as real servers.

The path to SLB A has the lower weight.

Figure 19 Two Distributed Sites with Two Web Servers Each

When all web servers in Figure 19 are operational, the client router receives the host route from both IOS SLB devices.

If one web server in either server farm fails, the route continues to be advertised by the given IOS SLB device.

If both Web Server A1 and Web Server A2 fail, the virtual server for the virtual IP address on SLB A enters FAILED state and stops advertising the host route for the virtual IP address. The client router then begins using the route to SLB B.

When either Web Server A1 or Web Server A2 is again available, the virtual server again advertises the host route for the virtual IP address, and the client router begins using SLB A.

Example with Two Distributed Sites with One Web Server and a Backup IOS SLB Switch Each

Figure 20 shows an IOS SLB network configured with route health injection with the following characteristics:

Both IOS SLB devices are configured with the same virtual IP address.

Each IOS SLB device has a server farm containing only the locally attached web server as a real server.

Each site has a primary IOS SLB device and a backup IOS SLB device.

The path to SLB A has the lower weight.

Figure 20 Two Distributed Sites with One Web Server and a Backup IOS SLB Switch Each

When both web servers in Figure 20 are operational, the client router receives the host route from both SLB A Primary and SLB B Primary.

If SLB A Primary fails, SLB A Backup begins advertising the host route to the virtual IP address. If SLB A Backup also fails, the virtual server for the virtual IP address on SLB A Primary and SLB A Backup enters FAILED state and stops advertising the host route for the virtual IP address. The client router then begins using the route to SLB B Primary (or to SLB B Backup, if SLB B Primary is not available).

When either SLB A Primary or SLB A Backup is again available, the virtual server again advertises the host route for the virtual IP address, and the client router begins using SLB A Primary or SLB A Backup.

Example of IOS SLB with Sticky Connections

The following sample configuration assigns all HTTP connections from a subnet to the same real server in server farm PUBLIC:

Router(config)# ip slb vserver http
Router(config-slb-vserver)# serverfarm PUBLIC
Router(config-slb-vserver)# sticky 30 group 1 netmask 255.255.255.248
Router(config-slb-vserver)# virtual 20.20.20.20 tcp 80
Router(config-slb-vserver)# inservice

The following sample configuration adds HTTPS connections to the above configuration, using the same sticky information but with a different virtual server:

Router(config)# ip slb vserver https
Router(config-slb-vserver)# serverfarm PUBLIC
Router(config-slb-vserver)# sticky 30 group 1 netmask 255.255.255.248
Router(config-slb-vserver)# virtual 20.20.20.20 tcp 443
Router(config-slb-vserver)# inservice

Now, all HTTP and HTTPS connections from the subnet are assigned to the same real server. For example, if a user connects to HTTP, then a second user connects to HTTPS, both connections are assigned to the same real server.

Command Reference

This section documents only new and modified commands.

access

address (HTTP probe)

address (ping probe)

address (WSP probe)

advertise

agent

bindid

clear ip slb

client (virtual server)

credentials

delay (firewall farm TCP protocol)

delay (virtual server)

expect

faildetect (ping probe)

faildetect (real server)

header

idle (firewall farm TCP protocol)

idle (firewall farm UDP protocol)

idle (virtual server)

inservice (DFP agent)

inservice (firewall farm)

inservice (firewall farm real server)

inservice (server farm real server)

inservice (server farm virtual server)

interval (DFP agent)

interval (HTTP probe)

interval (ping probe)

interval (WSP probe)

ip dfp agent

ip slb dfp

ip slb entries

ip slb firewallfarm

ip slb natpool

ip slb probe (HTTP probe)

ip slb probe (ping probe)

ip slb probe (WSP probe)

ip slb serverfarm

ip slb vserver

manager

maxconns (firewall farm TCP protocol)

maxconns (firewall farm UDP protocol)

maxconns (server farm)

mls aging slb normal

mls aging slb process

mls ip slb search wildcard

nat

password

port (DFP agent)

port (HTTP probe)

predictor (server farm)

predictor hash address (firewall farm)

probe (firewall farm real server)

probe (server farm)

real (firewall farm)

real (server farm)

reassign

replicate casa (firewall farm)

replicate casa (virtual server)

request method, request url

retry

serverfarm

show ip dfp

show ip slb conns

show ip slb dfp

show ip slb firewallfarm

show ip slb natpool

show ip slb probe

show ip slb reals

show ip slb replicate

show ip slb serverfarms

show ip slb stats

show ip slb sticky

show ip slb vserver

standby authentication

standby name

standby priority, standby preempt

standby timers

standby track

sticky (firewall farm TCP protocol)

sticky (firewall farm UDP protocol)

sticky (virtual server)

synguard (virtual server)

tcp

udp

url (WSP probe)

virtual (virtual server)

weight (firewall farm real server)

weight (server farm)

access

To route specific flows to a firewall farm, use the access firewall farm configuration command. To restore the default settings, use the no form of this command.

access [source source-ip-address network-mask] [destination destination-ip-address network-mask]

no access [source source-ip-address network-mask] [destination destination-ip-address network-mask]

Syntax Description

source

(Optional) Routes flows based on source IP address.

source-ip-address

(Optional) Specifies the source IP address. The default is 0.0.0.0 (all sources).

network-mask

(Optional) Specifies the source IP network mask. The default is 0.0.0.0 (all source subnets).

destination

(Optional) Routes flows based on destination IP address.

destination-ip-address

(Optional) Specifies the destination IP address. The default is 0.0.0.0 (all destinations).

network-mask

(Optional) Specifies the destination IP network mask. The default is 0.0.0.0 (all destination subnets).


Defaults

The default source IP address is 0.0.0.0 (route flows from all sources to this firewall farm).

The default source IP network mask is 0.0.0.0 (route flows from all source subnets to this firewall farm).

The default destination IP address is 0.0.0.0 (route flows from all destinations to this firewall farm).

The default destination IP network mask is 0.0.0.0 (route flows from all destination subnets to this firewall farm).

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(7)E

This command was introduced.


Usage Guidelines

You can specify more than one source or destination for each firewall farm. To do so, configure multiple access statements, making sure the network masks do not overlap each other.

Examples

The following example routes flows with a destination IP address of 10.1.6.0 to firewall farm FIRE1:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# access destination 10.1.6.0 255.255.255.0

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.


address (HTTP probe)

To configure an IP address to which to send HTTP probes, use the address HTTP probe configuration command. To restore the default settings, use the no form of this command.

address [ip-address]

no address [ip-address]

Syntax Description

ip-address

(Optional) Configures the destination IP address that is to respond to the HTTP probe.


Defaults

If the HTTP probe is associated with a firewall farm, you must specify an ip-address.

If the HTTP probe is associated with a server farm, and you do not specify an ip-address, the address is inherited from the server farm real servers.

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example configures an HTTP probe named PROBE2, changes the CLI to IOS SLB HTTP probe submode, and configures the probe to receive responses from IP address 13.13.13.13:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# address 13.13.13.13

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures an HTTP probe name and changes to HTTP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


address (ping probe)

To configure an IP address to which to send ping probes, use the address ping probe configuration command. To restore the default settings, use the no form of this command.

address [ip-address]

no address [ip-address]

Syntax Description

ip-address

(Optional) Configures the destination IP address that is to respond to the ping probe.


Defaults

If the ping probe is associated with a firewall farm, you must specify an ip-address.

If the ping probe is associated with a server farm, and you do not specify an ip-address, the address is inherited from the server farm real servers.

Command Modes

Ping probe configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example configures a ping probe named PROBE1, changes the CLI to IOS SLB ping probe submode, and configures the probe to receive responses from IP address 13.13.13.13:

Router(config)# ip slb probe PROBE1 ping
Router(config-slb-probe)# address 13.13.13.13

Related Commands

Command
Description

ip slb probe (ping probe)

Configures a ping probe name and changes to ping probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


address (WSP probe)

To configure an IP address to which to send WSP probes, use the address WSP probe configuration command. To restore the default settings, use the no form of this command.

address [ip-address]

no address [ip-address]

Syntax Description

ip-address

(Optional) Configures the destination IP address that is to respond to the WSP probe.


Defaults

If the WSP probe is associated with a firewall farm, you must specify an ip-address.

If the WSP probe is associated with a server farm, and you do not specify an ip-address, the address is inherited from the server farm real servers.

In dispatched mode, ip-address is the same as the virtual server IP address. In directed (NAT) mode, ip-address is unnecessary.

Command Modes

WSP probe configuration

Command History

Release
Modification

12.1(5a)E

This command was introduced.


Examples

The following example configures a ping probe named PROBE3, changes the CLI to IOS SLB WSP probe submode, and configures the probe to receive responses from IP address 13.13.13.13:

Router(config)# ip slb probe PROBE3 wsp
Router(config-slb-probe)# address 13.13.13.13

Related Commands

Command
Description

ip slb probe (WSP probe)

Configures a WSP probe name and changes to WSP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


advertise

To control the installation of a static route to the Null0 interface for a virtual server address, use the advertise virtual server configuration command. To prevent the installation of a static route for the virtual server IP address, use the no form of this command.

advertise [active]

no advertise [active]

Syntax Description

active

(Optional) Indicates that the host route is to be advertised only when the virtual IP address is available (that is, when there is at least one real server in OPERATIONAL, DFP_THROTTLED, or MAXCONNS state).


Defaults

The virtual server IP address is advertised. That is, a static route to the Null0 interface is installed for the virtual server IP addresses and it is added to the routing table.

If you do not specify active, the host route is advertised regardless of whether the virtual IP address is available.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(7)E

The active keyword was added.


Usage Guidelines

Advertisement of a static route using the routing protocol requires that you configure redistribution of static routes for the routing protocol.

The advertise command does not affect virtual servers used for transparent webcache load balancing.

HTTP probes and route health injection require a route to the virtual server. The route is not used, but it must exist in order for HTTP probes and route health injection to function correctly.

For HTTP probes, the route can be either a host route (advertised by the virtual server) or a default route (specified using the ip route 0.0.0.0 0.0.0.0 command, for example). If you specify either no advertise or advertise active, you must specify a default route.

For route health injection, the route must be a default route.

HTTP probes and route health injection can both use the same default route; you do not need to specify two unique default routes.

Examples

The following example prevents advertisement of the virtual server's IP address in routing protocol updates:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# no advertise

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.


agent

To identify a DFP agent with which IOS SLB can initiate connections, use the agent DFP configuration command. To remove a DFP agent definition from the DFP configuration, use the no form of this command.

agent ip-address port-number [timeout [retry_count [retry_interval]]]

no agent ip-address port

Syntax Description

ip-address

Agent IP address

port-number

Agent TCP or UDP port number

timeout

(Optional) Time period, in seconds, during which the DFP manager must receive an update from the DFP agent. The valid range is 0 to 65535 seconds. The default is 0 seconds, which means there is no timeout.

retry_count

(Optional) Number of times the DFP manager attempts to establish the TCP connection to the DFP agent. The valid range is 0 to 65535 times. The default is 0 retries, which means there are infinite retries.

retry_interval

(Optional) Interval, in seconds, between retries. The valid range is 1 to 65535 seconds. The default is 180 seconds.


Defaults

Timeout default: 0 seconds (no timeout)

Retry count default: 0 (infinite retries)

Retry interval default: 180 seconds

Command Modes

DFP configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

The password specified on the ip slb dfp command in the DFP manager must match the password specified on the password command in the DFP agent.

You can identify up to 1024 DFP agents.

Examples

The following example sets the DFP password to Cookies (to match the DFP agent's password), sets the timeout to 360 seconds, changes the CLI to DFP configuration mode, enables IOS SLB to connect to the DFP agent with IP address 10.1.1.1 and port number 2221, and sets a timeout of 30 seconds, an infinite number of retries, and a retry interval of 10 seconds:

Router(config)# ip slb dfp password Cookies 360
Router(config-slb-dfp)# agent 10.1.1.1 2221 30 0 10

Related Commands

Command
Description

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.

manager

Defines the port number to be used by the DFP manager to connect to the IOS SLB DFP agent.


bindid

To configure a bind ID, use the bindid server farm configuration command. To remove a bind ID from the server farm configuration, use the no form of this command.

bindid [bind_id]

no bindid [bind_id]

Syntax Description

bind_id

(Optional) Bind ID number. The default bind ID is 0.


Defaults

The default bind ID is 0.

Command Modes

Server farm configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

You can configure one bind ID on each bindid command.

The bind ID allows a single physical server to be bound to multiple virtual servers and report a different weight for each one. Thus, the single real server is represented as multiple instances of itself, each having a different bind ID. DFP uses the bind ID to identify for which instance of the real server a given weight is specified.

Examples

The following example configures bind ID 309:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# bindid 309

Related Commands

Command
Description

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.

show ip slb serverfarms

Displays information about the IOS SLB server farms.


clear ip slb

To clear IP IOS SLB connections or counters, use the clear ip slb command.

clear ip slb {connections [serverfarm farm_name | vserver server_name] | counters}

Syntax Description

connections

Clears the IOS SLB connection database.

serverfarm

(Optional) Clears the connection database for the serverfarm named.

farm_name

(Optional) Character string used to identify the serverfarm.

vserver

(Optional) Clears the connection database for the virtual server named.

server_name

(Optional) Character string used to identify the virtual server.

counters

Clears the IOS SLB counters.


Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.1(1)E

This command was introduced.


Examples

The following example clears the connection database of server farm FARM1:

Router# clear ip slb connections serverfarm FARM1

The following example clears the connection database of virtual server VSERVER1:

Router# clear ip slb connections vserver VSERVER1

The following example clears the IOS SLB counters:

Router# clear ip slb counters

Related Commands

Command
Description

show ip slb conns

Displays information about active IOS SLB connections.

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb serverfarms

Displays information about the IOS SLB server farms.

show ip slb stats

Displays IOS SLB statistics.

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.


client (virtual server)

To define which clients are allowed to use the virtual server, use the client virtual server configuration command. You can use more than one client command to define more than one client. To remove a client definition from the IOS SLB configuration, use the no form of this command.

client ip-address network-mask

no client ip-address network-mask

Syntax Description

ip-address

Client IP address. The default is 0.0.0.0 (all clients).

network-mask

Client IP network mask. The default is 0.0.0.0 (all subnets).


Defaults

The default client IP address is 0.0.0.0 (all clients).

The default client IP network mask is 0.0.0.0 (all subnets).

Taken together, the default is client 0.0.0.0 0.0.0.0 (allows all clients on all subnets to use the virtual server).

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

The network-mask value is applied to the source IP address of incoming connections. The result must match the ip-address value for the client to be allowed to use the virtual server.

Examples

The following example allows only clients from 10.4.4.x access to the virtual server:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# client 10.4.4.0 255.255.255.0

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


credentials

To configure basic authentication values for the HTTP IOS SLB probe, use the credentials HTTP probe configuration command. To remove a credentials configuration, use the no form of this command.

credentials username [password]

no credentials username [password]

Syntax Description

username

Configures the authentication username of the HTTP probe header. The character string is limited to 15 characters.

password

(Optional) Configures the authentication password of the HTTP probe header. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Examples

The following example configures an HTTP probe named PROBE2, changes the CLI to IOS SLB HTTP probe submode, sets the HTTP authentication to username lauren, and sets the password to develop:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# credentials lauren develop

Related Commands

Command
Description

show ip slb probe

Displays information about an IOS SLB probe.


delay (firewall farm TCP protocol)

To change the amount of time IOS SLB maintains TCP connection context after a connection has terminated, use the delay firewall farm TCP protocol configuration command. To restore the default delay timer, use the no form of this command.

delay duration

no delay

Syntax Description

duration

Delay timer duration in seconds. The valid range is 1 to 600 seconds. The default value is 10 seconds.


Defaults

Duration default: 10 seconds

Command Modes

Firewall farm TCP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

The delay timer allows out-of-sequence packets and final acknowledgments (ACKs) to be delivered after a TCP connection ends.

Do not set this value to zero (0).

If you are configuring a delay timer for HTTP flows, choose a low number such as 5 seconds as a starting point.

Examples

The following example specifies that IOS SLB maintains TCP connection context for 30 seconds after a connection has terminated:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# tcp
Router(config-slb-fw-tcp)# delay 30

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

tcp

Initiates TCP protocol configuration mode.


delay (virtual server)

To change the amount of time IOS SLB maintains TCP connection context after a connection has terminated, use the delay virtual server configuration command. To restore the default delay timer, use the no form of this command.

delay duration

no delay

Syntax Description

duration

Delay timer duration in seconds. The valid range is 1 to 600 seconds. The default value is 10 seconds.


Defaults

Duration default: 10 seconds

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

The delay timer allows out-of-sequence packets and final acknowledgments (ACKs) to be delivered after a TCP connection ends.

Do not set this value to zero (0).

If you are configuring a delay timer for HTTP flows, choose a low number such as 5 seconds as a starting point.

Examples

The following example specifies that IOS SLB maintains TCP connection context for 30 seconds after a connection has terminated:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# delay 30

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


expect

To configure a status code or regular expression to expect from the HTTP probe, use the expect HTTP probe configuration command. To restore the default settings, use the no form of this command.

expect [status status-code] [regex regular-expression]

no expect [status status-code] [regex regular-expression]

Syntax Description

status

(Optional) Specifies that a status code is to be expected.

status-code

(Optional) Configures the expected HTTP status code. The valid range is 100 to 599. The default expected status code is 200.

regex

(Optional) Specifies that a regular expression is to be expected.

regular-expression

(Optional) Configures the regular expression expected in the HTTP response.


Defaults

The default expected status code is 200.

There is no default expected regular expression.

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.

12.1(3a)E

The regex keyword and regular-expression variable were added.


Usage Guidelines

The expect command configures the expected status code or regular expression to be received from the servers. A real server is considered to have failed and is taken out of service if any of the following events occurs:

A status number other than the expected one is received.

The expected regular expression is not received in the first 2920 bytes of probe output. (IOS SLB searches only the first 2920 bytes for the expected status code or regular expression.)

The server fails to respond.

For IOS SLB firewall load balancing, configure the HTTP probe to expect status code 40l.

Examples

The following example configures an HTTP probe named PROBE2 changes the CLI to HTTP submode, and configures the HTTP probe to expect the status code 40l and the regular expression Copyright:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# expect status 401 regex Copyright

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures an HTTP probe name and changes to HTTP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


faildetect (ping probe)

To specify the conditions that indicate a server failure, use the faildetect ping probe configuration command. To restore the default values that indicate a server failure, use the no form of this command.

faildetect number-of-pings

no faildetect

Syntax Description

number-of-pings

Number of consecutive unacknowledged pings allowed before a real server is considered to have failed. Valid range is 1 to 255. The default is 10 unacknowledged pings.


Defaults

The default value is 10 unacknowledged pings.

Command Modes

Ping probe configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

In the following example the unacknowledged ping threshold is set to 16:

Router(config)# ip slb probe PROBE1 ping
Router(config-slb-probe)# faildetect 16

Related Commands

Command
Description

real (firewall farm)

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode.

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.


faildetect (real server)

To specify the conditions that indicate a server failure, use the faildetect real server configuration command. To restore the default values that indicate a server failure, use the no form of this command.

faildetect numconns number-conns [numclients number-clients]

no faildetect

Syntax Description

numconns

Number of consecutive TCP connection reassignments allowed before a real server is considered to have failed.

number-conns

Connection reassignment threshold value in the range from 1 to 255. The default is 8 connection failures.

numclients

(Optional) Number of unique client connection failures allowed before a real server is considered to have failed.

number-clients

(Optional) Client connection reassignment threshold value in the range from 1 to 8. The default is 2 client connection failures.


Defaults

If you do not specify the faildetect command, the default value of the connection reassignment threshold is 8.

If you do not specify the numclients keyword, the default value of the unique client failure threshold is 2.

Command Modes

Real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

In the following example the connection reassignment threshold is set to 16 and, because the number-clients keyword is not configured, the threshold for unique client connection failure is set to the default value 8. The real server is considered to have failed when 8 unique clients have had connection failures and there have been 16 connection reassignments.

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1
Router(config-slb-real)# faildetect numconns 16

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


header

To configure the basic authentication values for the HTTP probe, use the header HTTP probe configuration command. To remove a header HTTP probe configuration, use the no form of this command.

header field-name [field-value]

no header field-name [field-value]

Syntax Description

field-name

Configures the name of the HTTP probe header. The character string is limited to 15 characters.

field-value

(Optional) Configures the value of the HTTP probe header.


Defaults

No default behavior or values, although the following headers are inserted in the request by default:

Accept: */*
Connection: close
User-Agent: cisco-slb-probe/1.0
Host: virtual IP address

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

The header HTTP probe configuration command configures the name and value parameters of the header.


Note The colon ( : ) separating the field-name and field-value is automatically inserted if not provided. Multiple headers with the same name are not supported.


Examples

The following example configures an HTTP probe named PROBE2, changes the CLI to HTTP submode, and configures the HTTP probe header name as Cookie and value as Monster:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# header Cookie Monster

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures an HTTP probe name and changes to HTTP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


idle (firewall farm TCP protocol)

To specify the minimum amount of time IOS SLB maintains connection information in the absence of packet activity, use the idle firewall farm TCP protocol configuration command. To restore the default idle duration value, use the no form of this command.

idle duration

no idle

Syntax Description

duration

Idle connection timer duration in seconds. Valid values range from 10 to 65535. The default is 3600 seconds (1 hour).


Defaults

Duration default: 3600 seconds

Command Modes

Firewall farm TCP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

If a client sends a TCP packet that is not a sequence number (SYN) or reset (RST) packet, and IOS SLB does not have a TCP connection object in its table (possibly due to expiration of the idle timer), IOS SLB sends a TCP RST to the client.

If you are configuring an idle timer for HTTP flows, choose a low number such as 120 seconds as a starting point. A low number ensures that the IOS SLB connection database maintains a manageable size if problems at the server, client, or network result in a large number of connections. However, do not choose a value under 60 seconds; such a low value can reduce the efficiency of IOS SLB.

Examples

The following example instructs IOS SLB to maintain connection information for an idle connection for 120 seconds.

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# tcp
Router(config-slb-fw-tcp)# idle 120

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

tcp

Initiates TCP protocol configuration mode.


idle (firewall farm UDP protocol)

To specify the minimum amount of time IOS SLB maintains connection information in the absence of packet activity, use the idle firewall farm UDP protocol configuration command. To restore the default idle duration value, use the no form of this command.

idle duration

no idle

Syntax Description

duration

Idle connection timer duration in seconds. Valid values range from 10 to 65535. The default is 3600 seconds (1 hour).


Defaults

Duration default: 3600 seconds

Command Modes

Firewall farm UDP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example instructs IOS SLB to maintain connection information for an idle connection for 120 seconds.

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# udp
Router(config-slb-fw-udp)# idle 120

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

udp

Initiates UDP protocol configuration mode.


idle (virtual server)

To specify the minimum amount of time IOS SLB maintains connection information in the absence of packet activity, use the idle virtual server configuration command. To restore the default idle duration value, use the no form of this command.

idle duration

no idle

Syntax Description

duration

Idle connection timer duration in seconds. Valid values range from 10 to 65535. The default is 3600 seconds (1 hour).


Defaults

The default idle duration is 3600 seconds.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

If a client sends a TCP packet that is not a sequence number (SYN) or reset (RST) packet, and IOS SLB does not have a TCP connection object in its table (possibly due to expiration of the idle timer), IOS SLB sends a TCP RST to the client.

If you are configuring an idle timer for HTTP flows, choose a low number such as 120 seconds as a starting point. A low number ensures that the IOS SLB connection database maintains a manageable size if problems at the server, client, or network result in a large number of connections. However, do not choose a value under 60 seconds; such a low value can reduce the efficiency of IOS SLB.

Examples

The following example instructs IOS SLB to maintain connection information for an idle connection for 120 seconds.

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# idle 120

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


inservice (DFP agent)

To enable the DFP agent for communication with a DFP manager, use the inservice DFP agent configuration command. To remove the DFP agent from service, use the no form of this command.

inservice

no inservice

Syntax Description

This command has no arguments or keywords.

Defaults

If the inservice command is not specified, the DFP agent is inactive.

Command Modes

DFP agent configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

A DFP agent is inactive until both of the following conditions are met:

The DFP agent has been enabled using the inservice (DFP agent) command.

The client subsystem has changed the DFP agent's state to ACTIVE.

When you use the no form of this command to remove a DFP agent from service, the DFP agent closes all open connections, and no new connections are assigned.

Examples

In the following example, the DFP agent is enabled for communication with a DFP manager:

Router(config)# ip dfp agent slb
Router(config-dfp)# inservice

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.


inservice (firewall farm)

To enable the firewall farm for use by IOS SLB, use the inservice firewall farm configuration command. To remove the firewall farm from service, use the no form of this command.

inservice [standby group-name]

no inservice [standby group-name]

Syntax Description

standby

(Optional) Configures the HSRP standby firewall farm for use with stateless and stateful backup.

group-name

(Optional) Specifies the HSRP group name with which the IOS SLB firewall farm is associated.


Defaults

If the inservice command is not specified, the firewall farm is defined to IOS SLB but is not used.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

When you use the no form of this command to remove a firewall farm from service, the firewall farm quiesces gracefully. No new connections are assigned, and existing connections are allowed to complete.

Examples

In the following example, the firewall farm is enabled for use by the IOS SLB feature:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# inservice

Related Commands

Command
Description

ip slb firewallfarm

Identifies a firewall farm and initiates firewall farm configuration mode.

show ip slb firewallfarm

Displays information about the firewall farm configuration.


inservice (firewall farm real server)

To enable the firewall for use by IOS SLB, use the inservice firewall farm real server configuration command. To remove the firewall from service, use the no form of this command.

inservice

no inservice

Syntax Description

This command has no arguments or keywords.

Defaults

If the inservice command is not specified, the firewall is defined to IOS SLB but is not used.

Command Modes

Firewall farm real server configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

IOS SLB firewall load balancing uses probes to detect failures. Therefore, if you have not configured a probe, the firewall is not placed in service.

When you use the no form of this command to remove a firewall from service, the firewall quiesces gracefully. No new connections are assigned, and existing connections are allowed to complete.

Examples

In the following example, the firewall is enabled for use by the IOS SLB feature:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# real 10.10.1.1
Router(config-slb-fw-real)# inservice

Related Commands

Command
Description

real (firewall farm)

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode.

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.


inservice (server farm real server)

To enable the real server for use by IOS SLB, use the inservice server farm real server configuration command. To remove the real server from service, use the no form of this command.

inservice

no inservice

Syntax Description

This command has no arguments or keywords.

Defaults

If the inservice command is not specified, the real server is defined to IOS SLB but is not used.

Command Modes

Server farm real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

When you use the no form of this command to remove a real server from service, the real server quiesces gracefully. No new connections are assigned, and existing connections are allowed to complete.

Examples

In the following example, the real server is enabled for use by the IOS SLB feature:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1
Router(config-slb-sfarm-real)# inservice

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


inservice (server farm virtual server)

To enable the virtual server for use by IOS SLB, use the inservice server farm virtual server configuration command. To remove the virtual server from service, use the no form of this command.

inservice [standby group-name]

no inservice [standby group-name]

Syntax Description

standby

(Optional) Configures the HSRP standby virtual server for use with stateless and stateful backup.

group-name

(Optional) Specifies the HSRP group name with which the IOS SLB virtual server is associated.


Defaults

If the inservice command is not specified, the virtual server is defined to IOS SLB but is not used.

Command Modes

Server farm virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(1)E

The standby keyword and group-name variable were added.


Usage Guidelines

When you use the no form of this command to remove a virtual server from service, the virtual server quiesces gracefully. No new connections are assigned, and existing connections are allowed to complete.

Examples

In the following example, the virtual server is enabled for use by the IOS SLB feature:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# inservice

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


interval (DFP agent)

To configure a DFP agent weight recalculation interval, use the interval DFP agent configuration command. To restore the default setting, use the no form of this command.

interval seconds

no interval seconds

Syntax Description

seconds

Specifies the number of seconds to wait before recalculating weights for the DFP manager. Valid values range from 5 to 65535 seconds. The default interval is 10 seconds.


Defaults

The default interval value is 10 seconds.

Command Modes

DFP agent configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

The DFP agent sends the new weight to the DFP manager only if the new weight is different from the old weight. If the new weight is the same as the old weight, it is not sent to the DFP manager.

Examples

The following example configures the DFP agent to recalculate weights every 11 seconds:

Router(config)# ip dfp agent slb
Router(config-dfp)# interval 11

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.


interval (HTTP probe)

To configure an HTTP probe interval, use the interval HTTP probe configuration command. To remove an HTTP probe interval configuration, use the no form of this command.

interval seconds

no interval seconds

Syntax Description

seconds

Designates the number of seconds to wait before reattempting the probe. Valid values range from 1-65535 seconds. The default interval is 8 seconds.


Defaults

The default interval value is 8 seconds.

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Examples

The following example configures an HTTP probe named PROBE2, changes the CLI to HTTP submode, and configures the HTTP probe timer interval to transmit every 11 seconds:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# interval 11

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures an HTTP probe name and changes to HTTP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


interval (ping probe)

To configure a ping probe interval, use the interval ping probe configuration command. To remove a ping probe interval configuration, use the no form of this command.

interval seconds

no interval seconds

Syntax Description

seconds

Designates the number of seconds to wait before reattempting the probe. Valid values range from 1-65535 seconds. The default interval is 1 second.


Defaults

The default interval value is 1 second.

Command Modes

Ping probe configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example configures a ping probe named PROBE1, changes the CLI to ping submode, and configures the ping probe timer interval to transmit every 11 seconds:

Router(config)# ip slb probe PROBE1 ping
Router(config-slb-probe)# interval 11

Related Commands

Command
Description

ip slb probe (ping probe)

Configures a ping probe name and changes to ping probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


interval (WSP probe)

To configure a WSP probe interval, use the interval WSP probe configuration command. To remove a WSP probe interval configuration, use the no form of this command.

interval seconds

no interval seconds

Syntax Description

seconds

Designates the number of seconds to wait before reattempting the probe. Valid values range from 1-65535 seconds. The default interval is 8 seconds.


Defaults

The default interval value is 8 seconds.

Command Modes

WSP probe configuration

Command History

Release
Modification

12.1(5a)E

This command was introduced.


Examples

The following example configures a ping probe named PROBE3, changes the CLI to WSP probe submode, and configures the WSP probe timer interval to transmit every 11 seconds:

Router(config)# ip slb probe PROBE3 wsp
Router(config-slb-probe)# interval 11

Related Commands

Command
Description

ip slb probe (WSP probe)

Configures a WSP probe name and changes to WSP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


ip dfp agent

To identify a DFP agent subsystem and initiate DFP agent configuration mode, use the ip dfp agent global configuration command. To remove the DFP agent identification, use the no form of this command.

ip dfp agent subsystem-name

no ip dfp agent subsystem-name

Syntax Description

subsystem-name

Character string used to identify the DFP agent subsystem, such as slb for IOS SLB. The subsystem name enables the subsystem to send weights to a DFP manager. The subsystem name is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

To discover the subsystem names that are available in your environment, enter the following command:

ip dfp agent ?

Examples

The following example identifies a DFP agent subsystem named slb:

Router(config)# ip dfp agent slb
Router(config-dfp)#?

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.


ip slb dfp

To configure DFP, supply an optional password, and initiate DFP configuration mode, use the ip slb dfp global configuration command. To remove the DFP configuration, use the no form of this command.

ip slb dfp [password [0 | 7] password [timeout]]

no ip slb dfp

Syntax Description

password

(Optional) Specifies a password for MD5 authentication.

0

(Optional) Specifies that the password is unencrypted. This is the default setting.

7

(Optional) Specifies that the password is encrypted.

password

(Optional) Password value for MD5 authentication. This password must match the password configured on the host agent.

timeout

(Optional) Delay period, in seconds, during which both the old password and the new password are accepted. The valid range is 0 to 65535 seconds. The default value is 180 seconds, if a password is specified.


Defaults

The password encryption default is 0 (unencrypted).

The password timeout default is 180 seconds, if a password is specified.

Command Modes

Global configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(3a)E

The 0 and 7 keywords were added.


Usage Guidelines

The password specified on the ip slb dfp command in the DFP manager must match the password specified on the password command in the DFP agent.

The timeout option allows you to change the password without stopping messages between the DFP agent and its manager. The default value is 180 seconds.

During the timeout, the agent sends packets with the old password (or null, if there is no old password), and receives packets with either the old or new password. After the timeout expires, the agent sends and receives packets only with the new password; received packets that use the old password are discarded.

If you are changing the password for an entire load-balanced environment, set a longer timeout. This allows enough time for you to update the password on all agents and servers before the timeout expires. It also prevents mismatches between agents and servers that have begun running the new password and agents, and servers on which you have not yet changed the old password.

Examples

The following example configures DFP, sets the DFP password to Cookies and the timeout to 360 seconds, and changes the CLI to DFP configuration mode:

Router(config)# ip slb dfp password Cookies 360
Router(config-slb-dfp)#

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.


ip slb entries

To configure an initial allocation and a maximum value for IOS SLB database entries, use the ip slb entries global configuration command. To restore the default values, use the no form of this command.

ip slb entries [conn [init-conn [max-conn]] | frag [init-frag [max-frag]] | sticky [init-sticky [max-sticky]]]

no ip slb entries [conn | frag | sticky]

Syntax Description

conn

(Optional) Configures an initial allocation and a maximum value for IOS SLB connection database entries.

init-conn

(Optional) Initial allocation of connection database entries. The number of connection database entries can grow dynamically: When the number of available connection database entries is less than half of init-conn, IOS SLB allocates additional connection database entries.

Valid range is 1 to 1000000 connection database entries. The default is 8000 connection database entries.

max-conn

(Optional) Maximum number of connection database entries that can be allocated. Valid range is 1 to 8000000 connection database entries. The default is 8000000 connection database entries.

frag

(Optional) Configures an initial allocation and a maximum value for IOS SLB fragment database entries.

init-frag

(Optional) Initial allocation of fragment database entries. The number of fragment database entries can grow dynamically: When the number of available fragment database entries is less than half of init-frag, IOS SLB allocates additional fragment database entries.

Valid range is 1 to 1000000 fragment database entries. The default is 2000 fragment database entries.

max-frag

(Optional) Maximum number of fragment database entries that can be allocated. Valid range is 1 to 8000000 fragment database entries. The default is 32000 fragment database entries.

sticky

(Optional) Configures an initial allocation and a maximum value for IOS SLB sticky connection database entries.

init-sticky

(Optional) Initial allocation of sticky database entries. The number of sticky database entries can grow dynamically: When the number of available sticky database entries is less than half of init-sticky, IOS SLB allocates additional sticky database entries.

Valid range is 1 to 1000000 sticky database entries. The default is 4000 sticky database entries.

max-sticky

(Optional) Maximum number of sticky database entries that can be allocated. Valid range is 1 to 8000000 sticky database entries. The default is 8000000 sticky database entries.


Defaults

For connections, the default initial allocation is 8000 connections, and the default maximum is 8000000 connections.

For fragments, the default initial allocation is 4000 fragments, and the default maximum is 8000000 fragments.

For sticky connections, the default initial allocation is 2000 sticky connections, and the default maximum is 3200 sticky connections.

Command Modes

Global configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

If you configure an initial allocation value that exceeds the amount of available memory, memory might not be available for other features. In extreme cases, the router or switch might not boot properly. Therefore, be careful when you configure initial allocation values.

Examples

The following example configures an initial allocation of 128,000 connections, which can grow dynamically to a limit of 512,000 connections:

Router(config)# ip slb entries conn 128000 512000

Related Commands

Command
Description

show ip slb conns

Displays all connections handled by IOS SLB, or, optionally, only those connections associated with a particular virtual server or client.


ip slb firewallfarm

To identify a firewall farm and initiate firewall farm configuration mode, use the ip slb firewallfarm global configuration command. To remove the firewall farm from the IOS SLB configuration, use the no form of this command.

ip slb firewallfarm firewallfarm-name

no ip slb firewallfarm firewallfarm-name

Syntax Description

firewallfarm-name

Character string used to identify the firewall farm. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

Grouping real servers into firewall farms is an essential part of IOS SLB firewall load balancing. Using firewall farms enables IOS SLB to assign new connections to the real servers based on their weighted capacities, and on the load-balancing algorithms used.

Examples

The following example identifies a firewall farm named FIRE1:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)#?

Related Commands

Command
Description

real (firewall farm)

Identifies a firewall as a member of a firewall farm and initiates real server configuration mode.


ip slb natpool

To configure an IOS SLB NAT, use the ip slb natpool configuration command to create at least one client address pool. To remove an ip slb natpool configuration, use the no form of this command.

ip slb natpool pool-name start-ip end-ip [netmask netmask | prefix-length leading_1_bits] [entries init-addr [max-addr]]

no ip slb natpool pool-name

Syntax Description

pool-name

Configures a character string used to identify this client address pool. The character string is limited to 15 characters.

start-ip

Configures a starting IP address that defines the range of addresses in the address pool.

end-ip

Configures an ending IP address that defines the range of addresses in the address pool.

netmask

(Optional) Keyword used to configure the subnet mask.

netmask

(Optional) Mask for the associated IP subnet.

prefix-length

(Optional) Keyword used to configure the subnet mask.

leading_1_bits

(Optional) Mask for the associated IP subnet.

entries

(Optional) Configures an initial allocation and a maximum value for IOS SLB client NAT address entries for pool-name.

init-addr

(Optional) Initial allocation of client NAT address entries. The number of client NAT address entries can grow dynamically: When the number of available client NAT address entries is less than half of init-addr, IOS SLB allocates additional client NAT address entries.

Valid range is 1 to 1000000 client NAT address entries. The default is 8000 client NAT address entries.

max-addr

(Optional) Maximum number of client NAT address entries that can be allocated. Valid range is 1 to 8000000 client NAT address entries.

The default is the maximum number of ports that can be allocated within the IP address range specified for pool-name. For example, the following command:

ip slb natpool 3.3.3.1 3.3.3.5 prefix-length 24 entries 8000

has a default max-addr of (3.3.3.1-3.3.3.5)*54535, or 4*54535, or 218140.


Defaults

The default initial allocation is 8000 client NAT address entries.

The default maximum number of client NAT address entries that can be allocated is the maximum number of ports that can be allocated within the IP address range.

Command Modes

Global configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

If you want to use client NAT, you must create at least one client address pool.

Examples

The following example configures an IOS SLB NAT server farm pool of addresses with the name web-clients, the IP address range from 128.3.0.1 through 128.3.0.254, and a subnet mask of 255.255.0.0:

Router(config)# ip slb natpool web-clients 128.3.0.1 128.3.0.254 netmask 
255.255.0.0

Related Commands

Command
Description

show ip slb natpool

Displays information about the IOS SLB NAT configuration.

show ip slb serverfarms

Displays information about the server farm configuration.


ip slb probe (HTTP probe)

To configure an HTTP probe name and to change to HTTP probe configuration submode, use the ip slb probe (HTTP probe) configuration command. To remove an ip slb probe configuration, use the no form of this command.

ip slb probe name http

no ip slb probe name

Syntax Description

name

Configures a name for the HTTP probe. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

This command configures the HTTP probe name and application protocol, and changes the user interface to HTTP submode.

The HTTP probe cannot be unconfigured while it is being used by the server farm or firewall farm.

You can configure more than one probe, in any combination of types (HTTP, ping, or WSP), for each server farm, or for each firewall in a firewall farm.


Note HTTP probes require a route to the virtual server. The route is not used, but it must exist in order for HTTP probes to function correctly. The route can be either a host route (advertised by the virtual server) or a default route (specified using the ip route 0.0.0.0 0.0.0.0 command, for example).


Examples

The following example configures an IOS SLB probe named PROBE2, then changes to HTTP probe configuration submode:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)#

Related Commands

Command
Description

show ip slb probe

Displays information about an IOS SLB probe.


ip slb probe (ping probe)

To configure a ping probe name and to change to ping probe configuration submode, use the ip slb probe (ping probe) configuration command. To remove an ip slb probe configuration, use the no form of this command.

ip slb probe name ping

no ip slb probe name

Syntax Description

name

Configures a name for the ping probe. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

This command configures the ping probe name and application protocol, and changes the user interface to ping submode.

The ping probe cannot be unconfigured while it is being used by the server farm or firewall farm.

You can configure more than one probe, in any combination of types (HTTP, ping, or WSP), for each server farm, or for each firewall in a firewall farm.

Examples

The following example configures an IOS SLB probe named PROBE1, then changes to ping probe configuration submode:

Router(config)# ip slb probe PROBE1 ping
Router(config-slb-probe)#

Related Commands

Command
Description

show ip slb probe

Displays information about an IOS SLB probe.


ip slb probe (WSP probe)

To configure a WSP probe name and to change to WSP probe configuration submode, use the ip slb probe (WSP probe) configuration command. To remove an ip slb probe configuration, use the no form of this command.

ip slb probe name wsp

no ip slb probe name

Syntax Description

name

Configures a name for the WSP probe. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.1(5a)E

This command was introduced.


Usage Guidelines

This command configures the WSP probe name and application protocol, and changes the user interface to WSP probe configuration submode.

The WSP probe cannot be unconfigured while it is being used by the server farm or firewall farm.

You can configure more than one probe, in any combination of types (HTTP, ping, or WSP), for each server farm, or for each firewall in a firewall farm.

Examples

The following example configures an IOS SLB probe named PROBE3, then changes to WSP probe configuration submode:

Router(config)# ip slb probe PROBE3 wsp
Router(config-slb-probe)#

Related Commands

Command
Description

show ip slb probe

Displays information about an IOS SLB probe.


ip slb serverfarm

To identify a server farm and initiate server farm configuration mode, use the ip slb serverfarm global configuration command. To remove the server farm from the IOS SLB configuration, use the no form of this command.

ip slb serverfarm serverfarm-name

no ip slb serverfarm serverfarm-name

Syntax Description

serverfarm-name

Character string used to identify the server farm. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

Grouping real servers into server farms is an essential part of IOS SLB. Using server farms enables IOS SLB to assign new connections to the real servers based on their weighted capacities, and on the load-balancing algorithms used.

Examples

The following example identifies a server farm named PUBLIC:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)#?

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.


ip slb vserver

To identify a virtual server and initiate virtual server configuration mode, use the ip slb vserver global configuration command. To remove a virtual server from the IOS SLB configuration, use the no form of this command.

ip slb vserver virtual_server-name

no ip slb vserver virtual_server-name

Syntax Description

virtual_server-name

Character string used to identify the virtual server. The character string is limited to 15 characters.


Defaults

No default behavior or values.

Command Modes

Global configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example identifies a virtual server named PUBLIC_HTTP:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)#

Related Commands

Command
Description

serverfarm

Associates a real server farm with a virtual server.

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.


manager

This command has been removed. Its function is now performed by the ip dfp agent global configuration command, and by the following DFP agent configuration commands:

inservice (DFP agent)

interval (DFP agent)

password

port (DFP agent)

maxconns (firewall farm TCP protocol)

To limit the number of active connections to the real server, use the maxconns firewall farm TCP protocol configuration command. To restore the default of no limit, use the no form of this command.

maxconns maximum-number

no maxconns

Syntax Description

maximum-number

Maximum number of simultaneous active TCP connections using the firewall farm. Valid values range from 1 to 4294967295. The default is 4294967295.


Defaults

Maximum_number default: 4294967295

Command Modes

Firewall farm TCP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example limits the real server to a maximum of 1000 simultaneous active connections:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# tcp
Router(config-slb-fw-tcp)# maxconns 1000

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.

tcp

Initiates TCP protocol configuration mode.


maxconns (firewall farm UDP protocol)

To limit the number of active connections to the real server, use the maxconns firewall farm UDP protocol configuration command. To restore the default of no limit, use the no form of this command.

maxconns maximum-number

no maxconns

Syntax Description

maximum-number

Maximum number of simultaneous active UDP connections using the firewall farm. Valid values range from 1 to 4294967295. The default is 4294967295.


Defaults

Maximum_number default: 4294967295

Command Modes

Firewall farm UDP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example limits the real server to a maximum of 1000 simultaneous active connections:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# udp
Router(config-slb-fw-udp)# maxconns 1000

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.

udp

Initiates UDP protocol configuration mode.


maxconns (server farm)

To limit the number of active connections to the real server, use the maxconns real server configuration command. To restore the default of no limit, use the no form of this command.

maxconns maximum-number

no maxconns

Syntax Description

maximum-number

Maximum number of simultaneous active connections on the real server. Valid values range from 1 to 4294967295. The default is 4294967295.


Defaults

The default maximum number of simultaneous active connections on the real server is 4294967295.

Command Modes

Real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example limits the real server to a maximum of 1000 simultaneous active connections:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1
Router(config-slb-real)# maxconns 1000

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


mls aging slb normal

To configure the aging time for flows, use the mls aging slb normal global configuration command. To restore the default setting, use the no form of this command.

mls aging slb normal time

no mls aging slb normal time

Syntax Description

time

Idle time, in milliseconds, before a flow is aged. The valid range is 1 milliseconds to 10000 milliseconds. The default setting is 2000 milliseconds.

Note Heavier-than-normal loads can age flows more aggressively than this time.


Defaults

The default aging idle time is 2000 milliseconds.

Command Modes

Global configuration

Command History

Release
Modification

12.1(8)E

This command was introduced.


Usage Guidelines

This command is supported for Catalyst 6000 Family Switches only.

Examples

The following example sets the idle time to 4000 milliseconds:

Router(config)# mls aging slb normal 4000

Related Commands

Command
Description

ip slb firewallfarm

Identifies a firewall farm and initiates firewall farm configuration mode.

ip slb serverfarm

Associates a real server farm with a virtual server.

ip slb vserver

Identifies a virtual server.

mls aging slb process

Controls how often the aging process runs.


mls aging slb process

To control how often the aging process runs, use the mls aging slb process global configuration command. To restore the default setting, use the no form of this command.

mls aging slb process time

no mls aging slb process time

Syntax Description

time

Aging process interval, in milliseconds. The valid range is 1 millisecond to 10000 milliseconds. The default setting is 2000 seconds.


Defaults

The default aging process interval is 2000 milliseconds.

Command Modes

Global configuration

Command History

Release
Modification

12.1(8)E

This command was introduced.


Usage Guidelines

This command is supported for Catalyst 6000 Family Switches only.

Examples

The following example sets the aging process interval to 4000 milliseconds:

Router(config)# mls aging slb process 4000

Related Commands

Command
Description

ip slb firewallfarm

Identifies a firewall farm and initiates firewall farm configuration mode.

ip slb serverfarm

Associates a real server farm with a virtual server.

ip slb vserver

Identifies a virtual server.

mls aging slb normal

Configures the aging time for flows.


mls ip slb search wildcard

To specify the behavior of IOS SLB wildcard searches, use the mls ip slb search wildcard global configuration command. To restore the default setting, use the no form of this command.

mls ip slb search wildcard [pfc | rp]

no mls ip slb search wildcard [pfc | rp]

Syntax Description

pfc

Specifies that IOS SLB wildcard searches are to be performed by the Policy Feature Card (PFC). This is the default setting.

rp

Specifies that IOS SLB wildcard searches are to be performed by the route processor.


Defaults

The default setting is for the PFC to perform IOS SLB wildcard searches.

Command Modes

Global configuration

Command History

Release
Modification

12.1(7)E

This command was introduced.


Usage Guidelines

This command is supported for Catalyst 6000 Family Switches only.

If you configure IOS SLB and either input ACLs or firewall load balancing on the same Catalyst 6000 Family Switch, you can exceed the capacity of the TCAM on the PFC. To correct the problem, use the mls ip slb search wildcard rp command to reduce the amount of TCAM space used by IOS SLB, but be aware that this command can result in a slight increase in route processor utilization.

Examples

The following example limits wildcard searches to the route processor:

Router(config)# mls ip slb search wildcard rp

Related Commands

Command
Description

ip slb firewallfarm

Identifies a firewall farm and initiates firewall farm configuration mode.

ip slb serverfarm

Associates a real server farm with a virtual server.

ip slb vserver

Identifies a virtual server.


nat

To configure IOS SLB NAT and specify a NAT mode, use the nat server farm configuration command. To remove a NAT configuration, use the no form of this command.

nat {server | client pool-name}

no nat {server | client}

Syntax Description

server

Configures the destination address in load-balanced packets sent to the real server as the address of the real server chosen by the server farm load-balancing algorithm.

client

Configures the client address in load-balanced packets using addresses from the client address pool.

pool-name

Configures the pool name and must match the pool-name parameter from a previous ip slb natpool command.


Defaults

No default behavior or values.

Command Modes

Server farm configuration

Command History

Release
Modification

12.1(1)E

This command was introduced.

12.1(2)E

The client keyword and pool-name variable were added.


Usage Guidelines

The no nat command is allowed only if the virtual server was removed from service with the no inservice command.

Examples

The following example changes to IOS SLB server farm configuration mode and configures NAT mode as server address translation on server farm FARM2:

Router# ip slb serverfarm FARM2
Router(config-slb-sfarm)# nat server

The following example configures the NAT mode on server farm FARM2 to client translation mode and, using the real (server farm) command, configures the real server IP address as 10.3.1.1:

Router(config-slb-sfarm)# nat client web-clients
Router(config-slb-sfarm)# real 10.3.1.1

Related Commands

Command
Description

ip slb serverfarm

Associates a real server farm with a virtual server.

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.


password

To configure a DFP agent password for MD5 authentication, use the password DFP agent configuration command. To remove the DFP agent password, use the no form of this command.

password [0 | 7] password [timeout]

no password

Syntax Description

0

(Optional) Specifies that the password is unencrypted. This is the default setting.

7

(Optional) Specifies that the password is encrypted.

password

(Optional) Password value for MD5 authentication. This password must match the password configured on the host agent.

timeout

(Optional) Delay period, in seconds, during which both the old password and the new password are accepted. The valid range is 0 to 65535 seconds. The default value is 180 seconds.


Defaults

The password encryption default is 0 (unencrypted).

The password timeout default is 180 seconds.

Command Modes

DFP agent configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

The password specified on this command must match the password specified on the DFP manager.

The timeout option allows you to change the password without stopping messages between the DFP agent and its manager. The default value is 180 seconds.

During the timeout, the agent sends packets with the old password (or null, if there is no old password), and receives packets with either the old or new password. After the timeout expires, the agent sends and receives packets only with the new password; received packets that use the old password are discarded.

If you are changing the password for an entire load-balanced environment, set a longer timeout. This allows enough time for you to update the password on all agents and servers before the timeout expires. It also prevents mismatches between agents and servers that have begun running the new password and agents, and servers on which you have not yet changed the old password.

Examples

The following example sets the DFP agent password (unencrypted by default) to Cookies and the timeout to 360 seconds:

Router(config)# ip dfp agent slb
Router(config-dfp)# password Cookies 360

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.

replicate casa (firewall farm)

Configures a stateful backup of IOS SLB decision tables to a backup switch.

replicate casa (virtual server)

Configures a stateful backup of IOS SLB decision tables to a backup switch.


port (DFP agent)

To define the port number to be used by the DFP manager to connect to the DFP agent, use the port (DFP agent) DFP agent configuration command. To disable the port number definition and remove existing connections, use the no form of this command.

port port-number

no port port-number

Syntax Description

port-number

Port number used by the DFP manager to connect to the DFP agent. The valid range is 1 to 65535.


Defaults

No default behavior or values.

Command Modes

DFP agent configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Examples

In the following example, the DFP manager is enabled to connect to the DFP agent using port number 2221:

Router(config)# ip dfp agent slb
Router(config-dfp)# port 2221

Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.


port (HTTP probe)

To specify the port to which an HTTP probe is to connect, use the port (HTTP probe) HTTP probe configuration command. To restore the default settings, use the no form of this command.

port port-number

no port port-number

Syntax Description

port-number

Configures the TCP or UDP port number to which the HTTP probe is to connect.


Defaults

In dispatched mode, the port number is inherited from the virtual server.

If port translation is configured for the real server, that port number is used. See the real (server farm) command for more details.

Command Modes

HTTP probe configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example configures an HTTP probe named PROBE2, changes the CLI to IOS SLB HTTP probe submode, and configures the probe to connect to port number 8:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# port 8

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures an HTTP probe name and changes to HTTP probe configuration submode.

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb probe

Displays information about an IOS SLB probe.


predictor (server farm)

To specify the load-balancing algorithm for selecting a real server in the server farm, use the predictor server farm configuration command. To restore the default load-balancing algorithm of weighted round robin, use the no form of this command.

predictor [roundrobin | leastconns]

no predictor

Syntax Description

roundrobin

(Optional) Uses the weighted round robin algorithm for selecting the real server to handle the next new connection for the server farm. See the "Weighted Round Robin" section for a detailed description of this algorithm. This is the default value.

leastconns

(Optional) Uses the weighted least connections algorithm for selecting the real server to handle the next new connection for this server farm. See the "Weighted Least Connections" section for a detailed description of this algorithm.


Defaults

If you do not specify a load-balancing algorithm, the weighted round robin algorithm is used.

Command Modes

Server farm configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example specifies the weighted least connections algorithm:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# predictor leastconns

Related Commands

Command
Description

show ip slb serverfarms

Displays information about the server farm configuration.

weight (server farm)

Specifies the real server's capacity, relative to other real servers in the server farm.


predictor hash address (firewall farm)

To specify the load-balancing algorithm for selecting a firewall in the firewall farm, use the predictor hash address firewall farm configuration command. To restore the default load-balancing algorithm, use the no form of this command.

predictor hash address [port]

no predictor

Syntax Description

port

(Optional) Uses the source and destination TCP or UDP port numbers, in addition to the source and destination IP addresses, when selecting a firewall.


Defaults

Uses the source and destination IP addresses when selecting a firewall.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example specifies that source and destination IP addresses are to be used when selecting a firewall:

Router(config)# ip slb firewall FIRE1
Router(config-slb-fw)# predictor hash address

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

weight (firewall farm real server)

Specifies the firewall's capacity, relative to other firewalls in the firewall farm.


probe (firewall farm real server)

To associate a probe with a firewall farm, use the probe firewall farm real server configuration command. To remove the association, use the no form of this command.

probe name

no probe name

Syntax Description

name

Name of the probe to associate with this firewall farm.


Defaults

No default behavior or values.

Command Modes

Firewall farm real server configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

You can configure more than one probe for each firewall in a firewall farm.

Examples

The following example associates probe DAWN with server farm FIRE1:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw-real)# probe DAWN

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the server farm configuration.


probe (server farm)

To associate a probe with a server farm, use the probe server farm configuration command. To remove the association, use the no form of this command.

probe name

no probe name

Syntax Description

name

Name of the probe to associate with this server farm.


Defaults

No default behavior or values.

Command Modes

Server farm configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

You can configure more than one probe for each server farm.

Examples

The following example associates probe PROBE1 with server farm PUBLIC:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# probe PROBE1

Related Commands

Command
Description

show ip slb serverfarms

Displays information about the server farm configuration.


real (firewall farm)

To identify a firewall as a member of a firewall farm and initiate real server configuration mode, use the real firewall farm configuration command. To remove the firewall from the IOS SLB configuration, use the no form of this command.

real ip-address

no real ip-address

Syntax Description

ip-address

Real server IP address.


Defaults

No default behavior or values.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

A firewall farm comprises a number of firewalls. The firewalls are the physical devices that provide the firewall load-balanced services.

Examples

The following example identifies a firewall as a member of firewall farm FIRE1:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# real 10.1.1.1
Router(config-slb-real)#

Related Commands

Command
Description

inservice (firewall farm real server)

Enables the firewall for use by IOS SLB.

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.


real (server farm)

To identify a real server as a member of a server farm and initiate real server configuration mode, use the real server farm configuration command. To remove the real server from the IOS SLB configuration, use the no form of this command.

real ip-address [port_number]

no real ip-address [port_number]

Syntax Description

ip-address

Real server IP address.

port_number

(Optional) Port translation for the server. Valid values range from 1 to 65535.


Defaults

No default behavior or values.

Command Modes

Server farm configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(2)E

The port-number variable was added.


Usage Guidelines

A server farm comprises a number of real servers. The real servers are the physical devices that provide the load-balanced services.

Examples

The following example identifies a real server as a member of the server farm:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.1.1.1
Router(config-slb-real)#

Related Commands

Command
Description

inservice (server farm real server)

Enables the real server for use by IOS SLB.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


reassign

To specify the threshold of consecutive unacknowledged synchronizations that, if exceeded, result in an attempted connection to a different real server, use the reassign real server configuration command. To restore the default reassignment threshold, use the no form of this command.

reassign threshold

no reassign

Syntax Description

threshold

Number of unacknowledged TCP SYNs that are directed to a real server before the connection is reassigned to a different real server. An unacknowledged SYN is one for which no SYN or ACK is detected before the next SYN arrives from the client. IOS SLB allows 30 seconds for the connection to be established or for a new SYN to be received. If neither of these occurs within that time, the connection is removed from the IOS SLB database.

The 30-second timer is restarted for each SYN as long as the number of connection reassignments specified on the faildetect (real server) command's numconns keyword is not exceeded. See the faildetect (real server) command for more information.

Valid threshold values range from 1 to 4 SYNs. The default value is 3.


Defaults

Threshold default: 3 SYNs

Command Modes

Real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

IOS SLB does not reassign sticky connections if either of the following conditions is true:

The real server is not OPERATIONAL or MAXCONNS_THROTTLED.

The connection is the first for this sticky.

Examples

The following example sets the threshold of unacknowledged SYNs to 2:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1
Router(config-slb-real)# reassign 2

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


replicate casa (firewall farm)

To configure a stateful backup of IOS SLB decision tables to a backup switch, use the replicate casa firewall farm configuration command. To remove a replicate casa configuration, use the no form of this command.

replicate casa listening-ip remote-ip port-number [interval] [password [0 | 7] password [timeout]]

no replicate casa listening-ip remote-ip port-number

Syntax Description

listening-ip

Specifies the listening IP address for state exchange messages that are advertised.

remote-ip

Specifies the destination IP address for all state exchange signals.

port-number

The ports and the valid name or number for the port-number argument are as follows:

Domain Name System: dns   53

File Transfer Protocol: ftp   21

HTTP over Secure Socket Layer: https   443

Mapping of Airline Traffic over IP, Type A: matip-a   350

Network News Transport Protocol: nntp   119

Post Office Protocol v2: pop2   109

Post Office Protocol v3: pop3   110

Simple Mail Transport Protocol: smtp   25

Telnet: telnet   23

World Wide Web (HTTP): www   80

interval

(Optional) Maximum replication delivery interval from 1 to 300 seconds. The default value is 10 seconds.

password

(Optional) Specifies a password for MD5 authentication.

0

(Optional) Specifies that the password is unencrypted. This is the default setting.

7

(Optional) Specifies that the password is encrypted.

password

(Optional) Password value for MD5 authentication. This password must match the password configured on the host agent.

timeout

(Optional) Delay period, in seconds, during which both the old password and the new password are accepted. The default value is 180 seconds.


Defaults

The interval default is 10 seconds.

The password encryption default is 0 (unencrypted).

The password timeout default is 180 seconds.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

The timeout option allows you to change the password without stopping messages between the backup and primary Layer 3 switches. The default value is 180 seconds.

During the timeout, the backup sends packets with the old password (or null, if there is no old password), and receives packets with either the old or new password. After the timeout expires, the backup sends and receives packets only with the new password.

When setting a new password timeout, keep the following in mind:

If you are configuring a new backup, set the timeout to 0 (send packets with the new password immediately). This prevents password mismatches between the new backup and its primary.

If you are changing the password for an existing backup, set a longer timeout. This allows enough time for you to update the password on the primary before the timeout expires. It also prevents mismatches between the backup and primary.

Examples

The following example configures a stateful backup Layer 3 switch with a listening IP address of 10.10.10.11, a remote IP address of 10.10.11.12, over HTTP port 4231:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# replicate casa 10.10.10.11 10.10.11.12 4231

Related Commands

Command
Description

show ip slb replicate

Displays the configuration of IOS SLB IP replication.

show ip slb firewallfarm

Displays information about the firewall farm configuration.


replicate casa (virtual server)

To configure a stateful backup of IOS SLB decision tables to a backup switch, use the replicate casa virtual server configuration command. To remove a replicate casa configuration, use the no form of this command.

replicate casa listening-ip remote-ip port-number [interval] [password [0 | 7] password [timeout]]

no replicate casa listening-ip remote-ip port-number

Syntax Description

listening-ip

Specifies the listening IP address for state exchange messages that are advertised.

remote-ip

Specifies the destination IP address for all state exchange signals.

port-number

Specifies the TCP or UDP port number or port name for all state exchange signals.

interval

(Optional) Maximum replication delivery interval from 1 to 300 seconds. The default value is 10 seconds.

password

(Optional) Specifies a password for MD5 authentication.

0

(Optional) Specifies that the password is unencrypted. This is the default setting.

7

(Optional) Specifies that the password is encrypted.

password

(Optional) Password value for MD5 authentication. This password must match the password configured on the host agent.

timeout

(Optional) Delay period, in seconds, during which both the old password and the new password are accepted. The default value is 180 seconds.


Defaults

The interval default is 10 seconds.

The password encryption default is 0 (unencrypted).

The password timeout default is 180 seconds.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.

12.1(3a)E

The 0 and 7 keywords were added.


Usage Guidelines

The timeout option allows you to change the password without stopping messages between the backup and primary Layer 3 switches. The default value is 180 seconds.

During the timeout, the backup sends packets with the old password (or null, if there is no old password), and receives packets with either the old or new password. After the timeout expires, the backup sends and receives packets only with the new password.

When setting a new password timeout, keep the following in mind:

If you are configuring a new backup, set the timeout to 0 (send packets with the new password immediately). This prevents password mismatches between the new backup and its primary.

If you are changing the password for an existing backup, set a longer timeout. This allows enough time for you to update the password on the primary before the timeout expires. It also prevents mismatches between the backup and primary.

Examples

The following example configures a stateful backup Layer 3 switch with a listening IP address of 10.10.10.11, a remote IP address of 10.10.11.12, over HTTP port 4231:

Router(config)# ip slb vserver VS1
Router(config-slb-vserver)# replicate casa 10.10.10.11 10.10.11.12 4231

Related Commands

Command
Description

show ip slb replicate

Displays the configuration of IOS SLB IP replication.

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.


request method, request url

To configure an HTTP probe to check the status of the real servers, use the request method or request url configuration command. To remove a request method or request url configuration, use the no form of this command.

request [method {get | post | head | name name}] [url path]

no request [method {get | post | head | name name}] [url path]

Syntax Description

method

Configures the way the data is requested from the server.

get

Configures the Get method to request data from the server.

post

Configures the Post method to request data from the server.

head

Configures the header data type to request data from the server.

name

Name keyword to be followed by the name of the data to request data from the server.

name

Configures the name string of the data to send to the servers. The character string is limited to 15 characters.

url

(Optional) Configures the path from the server.

path

(Optional) Path from the server.


Defaults

If no values are configured following the method keyword, the default is Get.

If no URL path is set to the server, the default is /.

Command Modes

HTTP IOS SLB probe configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Usage Guidelines

The request method/url command configures the IOS SLB HTTP probe method used to receive data from the server. Only one IOS SLB HTTP probe can be configured for each server farm.

Examples

The following example configures an IOS SLB HTTP probe named PROBE2, changes the CLI to IOS SLB probe submode, and configures HTTP requests to use the post method and the URL /probe.cgi?all:

Router(config)# ip slb probe PROBE2 http
Router(config-slb-probe)# request method post url /probe.cgi?all

Related Commands

Command
Description

ip slb probe (HTTP probe)

Configures the IOS SLB IP probe name.

show ip slb probe

Displays information about an IOS SLB probe.


retry

To specify how long to wait before a new connection is attempted to a failed server, use the retry real server configuration command. To restore the default retry value, use the no form of this command.

retry retry-value

no retry

Syntax Description

retry-value

Time, in seconds, to wait after the detection of a server failure before a new connection to the server is attempted.

If the new connection attempt succeeds, the real server is placed in OPERATIONAL state. If the connection attempt fails, the timer is reset, the connection is reassigned, and the process repeats until it is successful or until the server is placed OUTOFSERVICE by the network administrator.

Valid values range from 1 to 3600. The default value is 60 seconds.

A value of 0 means do not attempt a new connection to the server when it fails.


Defaults

The retry-value default is 60 seconds.

Command Modes

Real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example specifies that 120 seconds must elapse after the detection of a server failure before a new connection is attempted:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1
Router(config-slb-real)# retry 120

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


serverfarm

To associate a real server farm with a virtual server, or to configure a backup server farm, use the serverfarm virtual server configuration command. To remove the server farm association from the virtual server configuration, use the no form of this command.

serverfarm primary-serverfarm-name [backup backup-serverfarm-name [sticky]]

no serverfarm primary-serverfarm-name [backup backup-serverfarm-name [sticky]]

Syntax Description

primary-serverfarm-name

Name of a server farm that has already been defined using the ip slb serverfarm command.

backup

(Optional) Specifies that a backup server farm is to be configured.

backup-serverfarm-name

(Optional) Name of a server farm that has already been defined using the ip slb serverfarm command.

sticky

(Optional) Specifies that sticky connections are to be used in the backup server farm.


Defaults

If backup backup-serverfarm-name is not specified, no backup server farm is configured.

If a backup server farm is configured but sticky is not specified, sticky connections are not used in the backup server farm.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(8a)E

The backup and sticky keywords and the backup-serverfarm-name arguments were added.


Examples

The following example shows how the ip slb vserver, virtual, and serverfarm commands are used to associate the real server farm named PUBLIC with the virtual server named PUBLIC_HTTP.

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# virtual 10.0.0.1 tcp www
Router(config-slb-vserver)# serverfarm PUBLIC

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


show ip dfp

To display information about DFP agents, use the show ip dfp privileged EXEC command.

show ip dfp [agent subsystem-name] [detail]

Syntax Description

agent subsystem-name

(Optional) Displays information about the specified DFP agent, such as slb for IOS SLB.

detail

(Optional) Displays detailed DFP agent information.


Defaults

If no options are specified, the command displays output for all DFP agents identified by ip dfp agent commands, regardless of whether those agents are currently in service (Inservice: yes) or active (AppActive: yes).

Command Modes

Privileged EXEC

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

Detailed output for show ip dfp includes information about all DFP agents identified by ip slb agent commands, regardless of whether those agents are currently in service (Inservice: yes) or active (AppActive: yes).

Examples

The following example shows detailed information for DFP agent slb:

Router# show ip dfp agent slb detail

Unexpected errors: 0

DFP Agent for service: SLB
      Port: 666 Interval: 10
      Current passwd: <none> Pending passwd: <none>
      Passwd timeout: 0
      Inservice: yes  AppActive: yes

      Manager IP Address   Timeout
      ------------------   -------
      172.18.45.27         0

Weight Table Report for Agent SLB

      Weights for Port: 80  Protocol: TCP

          IP Address        Bind ID  Weight
          ---------------   -------  -------
          1.1.1.1           0        65535

      Weights for Port: 0 (wildcard)  Protocol: 0 (wildcard)

          IP Address        Bind ID  Weight
          ---------------   -------  -------
          0.0.0.0           65534    0

Bind ID Table Report for Agent SLB

      Bind IDs for Port: 80  Protocol: TCP

          Bind ID   Client IP        Client Mask
          -------   ---------------  ---------------
          0         0.0.0.0          0.0.0.0

Table 1 show ip dfp Field Descriptions 

Field
Description

Port

TCP port number of the agent.

Interval

Number of seconds to wait before recalculating weights.

Current passwd

Current DFP password for MD5 authentication.

Pending passwd

Pending new DFP password for MD5 authentication.

Passwd timeout

Delay period, in seconds, during which both the current password and the new password are accepted.

Inservice

Indicates whether the DFP agent is enabled for communication with a DFP manager.

AppActive

Indicates whether the DFP agent is active.

Manager IP Address

IP address of the manager to which weights are being sent.

Timeout

Time period, in seconds, during which the DFP manager must receive an update from the DFP agent. A value of 0 means there is no timeout.

Weights for Port

Port for which the following weights are reported. 0 indicates a wildcard value.

Protocol

Protocol used for the port. 0 indicates a wildcard value.

IP Address

IP address for which weight is reported.

Bind ID

Bind ID associated with the IP address.

Weight

Weight calculated for the IP address.

Bind IDs for Port

Port for which the following bind IDs are reported.

Protocol

Protocol used for the port.

Bind ID

Bind ID of this instance of the real server.

Client IP

IP address of client using the virtual server.

Client Mask

IP network mask of client using the virtual server.


Related Commands

Command
Description

agent

Identifies a DFP agent to which IOS SLB can connect.

ip dfp agent

Identifies a DFP agent subsystem and initiates DFP agent configuration mode.

ip slb dfp

Configures DFP, supplies an optional password, and initiates DFP configuration mode.


show ip slb conns

To display the active IOS SLB connections, use the show ip slb conns privileged EXEC command.

show ip slb conns [vserver virtual_server-name | client ip-address | firewall firewallfarm-name] [detail]

Syntax Description

vserver

(Optional) Displays only those connections associated with a particular virtual server.

virtual_server-name

(Optional) Name of the virtual server to be monitored.

client

(Optional) Displays only those connections associated with a particular client IP address.

ip-address

(Optional) IP address of the client to be monitored.

firewall

(Optional) Displays only those connections associated with a particular firewall farm.

firewallfarm-name

(Optional) Name of the firewall farm to be monitored.

detail

(Optional) Displays detailed connection information.


Defaults

If no options are specified, the command displays output for all active IOS SLB connections.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(7)E

The firewall keyword and firewallfarm-name variable were added.


Examples

The following example shows IOS SLB active connection data:

Router# show ip slb conns

vserver          prot   client                real                  state
----------------------------------------------------------------------------
TEST             TCP    7.150.72.183:328      80.80.90.25:80        INIT 
TEST             TCP    7.250.167.226:423     80.80.90.26:80        INIT 
TEST             TCP    7.234.60.239:317      80.80.90.26:80        ESTAB 
TEST             TCP    7.110.233.96:747      80.80.90.26:80        ESTAB 
TEST             TCP    7.162.0.201:770       80.80.90.30:80        CLOSING 
TEST             TCP    7.22.225.219:995      80.80.90.26:80        CLOSING 
TEST             TCP    7.2.170.148:169       80.80.90.30:80        ZOMBIE 

Table 2 show ip slb conns Field Descriptions 

Field
Description

vserver

Name of the virtual server whose connections are being monitored and displayed. Information about each connection is displayed on a separate line.

prot

Protocol being used by the connection.

client

Client IP address being used by the connection.

real

Real IP address of the connection.

state

Current state of the connection:

CLOSING—IOS SLB TCP connection deactivated (awaiting a delay timeout before cleaning up the connection)

ESTAB—IOS SLB TCP connection processed a SYN-SYN/ACK exchange between the client and server

FINCLIENT—IOS SLB TCP connection processed a FIN from the client

FINSERVER—IOS SLB TCP connection processed a FIN from the server

INIT—Initial state of the IOS SLB TCP connection

SYNBOTH—IOS SLB TCP connection processed one or more TCP SYNs from both the client and the server

SYNCLIENT—IOS SLB TCP connection processed one or more client TCP SYNs

SYNSERVER—IOS SLB TCP connection processed one or more server 1 TCP SYNs

ZOMBIE—Destruction of the IOS SLB TCP connection failed, possibly because of bound flows. Destruction will proceed when the flows are unbound.


show ip slb dfp

To display DFP manager and agent information, such as passwords, timeouts, retry counts, and weights, use the show ip slb dfp privileged EXEC command.

show ip slb dfp [agent agent_ip_address port-number | manager manager_ip_address | detail | weights]

Syntax Description

agent

(Optional) Displays information about an agent.

agent_ip_address

(Optional) Agent IP address.

port-number

(Optional) Agent TCP or UDP port number.

manager

(Optional) Displays information about a manager.

manager_ip_address

(Optional) Manager IP address.

detail

(Optional) Displays all data available.

weights

(Optional) Displays information about weights assigned to real servers for load balancing.


Defaults

If no options are specified, the command displays summary information.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(5a)E

The manager keyword and manager_ip_address variable were added.


Usage Guidelines

The following example displays high-level information about all DFP agents and managers:

Router# show ip slb dfp
DFP Manager:
      Current passwd:NONE Pending passwd:NONE
      Passwd timeout:0 sec 

Agent IP          Port    Timeout   Retry Count   Interval
---------------------------------------------------------------
161.44.2.34       61936   0         0             180 (Default)

Table 3 show ip slb dfp Field Descriptions 

Field
Description

Agent IP

IP address of the agent about which information is being displayed.

Port

TCP or UDP port number of the agent.

Timeout

Time period, in seconds, during which the DFP manager must receive an update from the DFP agent. A value of 0 means there is no timeout.

Retry Count

Number of times the DFP manager attempts to establish the TCP connection to the DFP agent. A value of 0 means there are infinite retries.

Interval

Interval, in seconds, between retries.

State

Current state of the connection.

CONNING—This state is set after the Agent opens the socket, and is connecting.

FAILED—This state is set when there is an error in the data, or when the maximum number of retries is reached.

NOT_OPEN—This is the initial state when the Agent block is created.

OPEN—The Agent is talking to the Host. This is the functioning state.

SEC_ERR—This state is set when there is a security error in the received message. Either there is no security vector when one was required, or the MD5 string in the message did not match the expected result (password not correct).

TRYING—When the Agent tries to open the socket, it is moved to this state.

Timeout

Time period, in seconds, during which the DFP manager must receive an update from the DFP agent. A value of 0 means there is no timeout.


The following example displays detailed information about DFP agents and managers:

Router# show ip slb dfp detail
DFP Manager
      Current passwd <none> Pending passwd <none>
      Passwd timeout 0 sec 
      Unexpected errors 0
% No DFP Agents configured

The following example displays detailed information about DFP manager 55.55.55.2:

Router# show ip slb dfp manager 55.55.55.2
DFP Manager 55.55.55.2 Connection state Connected
   Timeout = 20   
   Last message sent 033537 UTC 01/02/00

The following example displays detailed information about weights assigned to real servers for load balancing:

Router# show ip slb dfp weights
Real IP Address 17.17.17.17 Protocol TCP Port 22 Bind_ID 111 Weight 111
      Set by Agent 161.44.2.3458490 at 132241 UTC 12/03/99
Real IP Address 17.17.17.17 Protocol TCP Port www Bind_ID 1 Weight 1
      Set by Agent 161.44.2.3458490 at 132241 UTC 12/03/99
Real IP Address 68.68.68.68 Protocol TCP Port www Bind_ID 4 Weight 4
      Set by Agent 161.44.2.3458490 at 132241 UTC 12/03/99
Real IP Address 85.85.85.85 Protocol TCP Port www Bind_ID 5 Weight 5
      Set by Agent 161.44.2.3458490 at 132241 UTC 12/03/99

show ip slb firewallfarm

To display firewall farm information, use the show ip slb firewallfarm configuration command.

show ip slb firewallfarm [detail]

Syntax Description

detail

(Optional) Displays detailed information.


Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example shows firewall farm data:

Router# show ip slb firewallfarm

firewall farm    hash        state         reals
------------------------------------------------
FIRE1            IPADDR      OPERATIONAL     2

Table 4 show ip slb firewallfarm Field Descriptions 

Field
Description

firewall farm

Name of the firewall farm.

hash

Load-balancing algorithm used to select a firewall for the firewall farm:

IPADDR—Uses the source and destination IP addresses in the algorithm.

IPADDRPORT—Uses the source and destination TCP or UDP port numbers, in addition to the source and destination IP addresses, in the algorithm.

See the predictor hash address (firewall farm) command for more details.

state

Current state of the firewall farm.

OPERATIONAL—Functioning properly

OUTOFSERVICE—Removed from the load-balancing predictor lists

STANDBY—Backup firewall farm, ready to become operational if active firewall farm fails

reals

Number of firewalls that are members of the firewall farm.


show ip slb natpool

To display the IP IOS SLB NAT configuration, use the show ip slb natpool command.

show ip slb natpool [name pool-name] [detail]

Syntax Description

name

(Optional) Keyword to display a specific NAT pool.

pool-name

(Optional) NAT pool name string to display.

detail

(Optional) Lists all the interval ranges currently allocated in the client NAT pool.


Defaults

No default behavior or values.

Command Modes

EXEC configuration

Command History

Release
Modification

12.1(2)E

This command was introduced.


Examples

The following example displays the default show ip slb natpool command:

Router# show ip slb natpool
nat client B  1.1.1.6  1.1.1.8  Netmask 255.255.255.0
		nat client A  1.1.1.1  1.1.1.5  Netmask 255.255.255.0

The following example displays the show ip slb natpool command with the additional detail parameter:

Router# show ip slb natpool detail
nat client A  1.1.1.1  1.1.1.5  Netmask 255.255.255.0
     Start NAT        Last NAT         Count     ALLOC/FREE 
     -------------------------------------------------------
     1.1.1.1:11001    1.1.1.1:16333    0005333   ALLOC
     1.1.1.1:16334    1.1.1.1:19000    0002667   ALLOC
     1.1.1.1:19001    1.1.1.5:65535    0264675   FREE

nat client B  1.1.1.6  1.1.1.8  Netmask 255.255.255.0
     Start NAT        Last NAT         Count     ALLOC/FREE 
     -------------------------------------------------------
     1.1.1.6:11001    1.1.1.6:16333    0005333   ALLOC
     1.1.1.6:16334    1.1.1.6:19000    0002667   ALLOC
     1.1.1.6:19001    1.1.1.8:65535    0155605   FREE

show ip slb probe

To display information about an IOS SLB probe, use the show ip slb probe configuration command.

show ip slb probe [name probe_name] [detail]

Syntax Description

name

(Optional) Displays information about the specific probe named.

probe_name

(Optional) Probe name to display.

detail

(Optional) Displays detailed information, including the SA Agent operation ID, which you can use to correlate the output of the show rtr operational-state EXEC command.


Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.1(2)E

This command was introduced.


Examples

The following example shows IOS SLB probe data:

Router# show ip slb probe

Server:Port            State        Outages  Current  Cumulative
----------------------------------------------------------------
10.10.4.1:0            OPERATIONAL        0  never    00:00:00
10.10.5.1:0            FAILED             1  00:00:06 00:00:06

Table 5 show ip slb probe Field Descriptions 

Field
Description

Server:Port

IP address and port of the real server.

State

Operational state of the probe:

FAILED—The probe has succeeded in the past but has currently failed.

OPERATIONAL—The probe is functioning normally.

TESTING—The probe has never succeeded, due to no response. IOS SLB keeps no counters or timers for this state.

Outages

Number of intervals between successful probes.

Current

Time since the last probe success. That is, the duration (so far) of the current outage.

Cumulative

Total time the real server has been under test by the probe and has failed the probe test. This value is the sum of the Current time plus the total time of all previous outages.


show ip slb reals

To display information about the real servers, use the show ip slb reals privileged EXEC command.

show ip slb reals [vserver virtual_server-name] [detail]

Syntax Description

vserver

(Optional) Displays information about only those real servers associated with a particular virtual server.

virtual_server-name

(Optional) Name of the virtual server.

detail

(Optional) Displays detailed information.


Defaults

If no options are specified, the command displays information about all real servers.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example shows IOS SLB real server data:

Router# show ip slb reals

real             farm name        weight   state           conns
--------------------------------------------------------------------
80.80.2.112      FRAG             8        OUTOFSERVICE    0        
80.80.5.232      FRAG             8        OPERATIONAL     0        
80.80.15.124     FRAG             8        OUTOFSERVICE    0        
80.254.2.2       FRAG             8        OUTOFSERVICE    0        
80.80.15.124     LINUX            8        OPERATIONAL     0        
80.80.15.125     LINUX            8        OPERATIONAL     0        
80.80.15.126     LINUX            8        OPERATIONAL     0        
80.80.90.25      SRE              8        OPERATIONAL     220      
80.80.90.26      SRE              8        OPERATIONAL     216      
80.80.90.27      SRE              8        OPERATIONAL     216      
80.80.90.28      SRE              8        TESTING         1        
80.80.90.29      SRE              8        OPERATIONAL     221      
80.80.90.30      SRE              8        OPERATIONAL     224      
80.80.30.3       TEST             100      READY_TO_TEST   0        
80.80.30.4       TEST             100      READY_TO_TEST   0        
80.80.30.5       TEST             100      READY_TO_TEST   0        
80.80.30.6       TEST             100      READY_TO_TEST   0        

Table 6 show ip slb reals Field Descriptions 

Field
Description

real

IP address of the real server about which information is being displayed. Used to identify each real server. Information about each real server is displayed on a separate line.

farm name

Name of the server farm or firewall farm with which the real server is associated.

weight

Weight assigned to the real server. The weight identifies the real server's capacity, relative to other real servers in the server farm.

state

Current state of the real server.

DFP_THROTTLED—The DFP agent sent a weight of 0 for this real server (send no further connections to this real server).

FAILED—The real server has failed due as a result of either no response or RST responses to client traffic. (See the faildetect (real server) command for more information about controlling tolerance for no-responses and RSTs). The real server has been removed from use by the predictor algorithms. The retry timer has started.

MAXCONNS_THROTTLE—The number of connections on the real server exceeds the configured maximum number of simultaneous active connections (maxconns).

OPERATIONAL—The real server is functioning properly and is being used for load-balancing.

OPER_WAIT—The real server is waiting to become operational (waiting for a timeout or some other condition to be met).

OUTOFSERVICE—The real server was configured with no inservice and has been removed from the load-balancing predictor lists.

PROBE_FAILED—The probe has succeeded in the past but has currently failed. This might occur at the same time user connections fail, or it might not.

PROBE_TESTING—The probe has never succeeded, due to no response. The initial probe timed out waiting for a success.

READY_TO_TEST—The real server is queued for testing after being in FAILED state until the retry timer expired.

TESTING—The real server is queued for assignment. When a single user connection is assigned to a real server that is in READY_TO_TEST state, the real server is placed in TESTING state. If the test succeeds, the real server is placed back in OPERATIONAL state.

TEST_WAIT—The real server is waiting to begin testing (waiting for a timeout or some other condition to be met).

conns

Number of connections associated with the real server.


The following is sample output from the show ip slb reals detail command for a real server in a server farm:

Router# show ip slb reals detail

10.10.1.7, S, state = OPERATIONAL, type = server
  conns = 0, dummy_conns = 0, maxconns = 4294967295
  weight = 8, weight(admin) = 8, metric = 0, remainder = 0
  reassign = 3, retry = 60
  failconn threshold = 8, failconn count = 0
  failclient threshold = 2, failclient count = 0
  total conns established = 0, total conn failures = 0
  server failures = 0

The following is sample output from the show ip slb reals detail command for a real server in a firewall farm:

Router# show ip slb reals detail

10.10.3.2, F, state = OPERATIONAL, type = firewall
  conns = 0, dummy_conns = 0, maxconns = 4294967295
  weight = 8, weight(admin) = 8, metric = 0, remainder = 0
  total conns established = 8377, hash count = 0
  server failures = 0
  interface FastEthernet1/0, MAC 0000.0c41.1063

Table 7 describes the fields shown in the above detail displays.

Table 7 show ip slb reals detail Field Descriptions 

Field
Description

IP address

IP address of the real server about which information is being displayed. Used to identify each real server. Information about each real server is displayed on a separate line.

farm name

Name of the server farm or firewall farm with which the real server is associated.

state

Current state of the real server.

DFP_THROTTLED—The Dynamic Feedback Protocol (DFP) agent sent a weight of 0 for this real server (send no further connections to this real server).

FAILED—The real server has failed as a result of either no response or reset (RST) responses to client traffic. (See the faildetect (real server) command for more information about controlling tolerance for no responses and RSTs.) The real server has been removed from use by the predictor algorithms. The retry timer has started.

MAXCONNS_THROTTLE—The number of connections on the real server exceeds the configured maximum number of simultaneous active connections (maxconns).

OPERATIONAL—The real server is functioning properly and is being used for load-balancing.

OPER_WAIT—The real server is waiting to become operational (waiting for a timeout or some other condition to be met).

OUTOFSERVICE—The real server was configured with no inservice and has been removed from the load-balancing predictor lists.

PROBE_FAILED—The probe has succeeded in the past but has currently failed. This failure might occur at the same time user connections fail, or it might not.

PROBE_TESTING—The probe has never succeeded, due to no response. The initial probe timed out waiting for a success.

READY_TO_TEST—The real server is queued for testing after being in FAILED state until the retry timer expired.

TESTING—The real server is queued for assignment. When a single user connection is assigned to a real server that is in READY_TO_TEST state, the real server is placed in TESTING state. If the test succeeds, the real server is placed back in OPERATIONAL state.

TEST_WAIT—The real server is waiting to begin testing (waiting for a timeout or some other condition to be met).

type

Indicates whether the real server is associated with a server farm (server) or firewall farm (firewall).

conns

Number of connections associated with the real server.

In general packet radio service (GPRS) load balancing, number of sessions associated with the real server.

In per-packet server load balancing, number of request packets that have been load balanced to each real server, using the connection count.

dummy_conns

Internal counter used in debugging.

maxconns

Maximum number of active connections allowed on the real server at one time.

weight

Weight assigned to the real server. The weight identifies the real server's capacity, relative to other real servers in the server farm. This value could be changed by DFP.

weight(admin)

Configured (or default) weight assigned to the real server.

metric

Internal counter used in debugging.

remainder

Internal counter used in debugging.

reassign

Total number of consecutive unacknowledged SYNchronize sequence numbers (SYNs) or Create Packet Data Protocol (PDP) requests since the last time the clear ip slb counters command was issued.

retry

Interval, in seconds, to wait between the detection of a failure on the real server and the next attempt to connect to the server.

failconn threshold

Maximum number of consecutive connection failures allowed before the real server is considered to have failed.

failconn count

Total number of consecutive connection failures since the last time the clear ip slb counters command was issued.

failclient threshold

Maximum number of unique client connection failures allowed before the real server is considered to have failed.

failclient count

Total number of unique client connection failures since the last time the clear ip slb counters command was issued.

total conns established

Total number of successful connection assignments since the last time the clear ip slb counters command was issued.

total conn failures

Total number of unsuccessful connection assignments since the last time the clear ip slb counters command was issued.

server failures

Total number of times this real server has been marked failed.

hash count

Total number of times the hash algorithm has been called.

interface

Type of interface.

MAC

MAC address of the firewall.


show ip slb replicate

To display the IOS SLB replication configuration, use the show ip slb replicate privileged EXEC command.

show ip slb replicate

Syntax Description

This command has no arguments or keywords.

Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.1(2)E

This command was introduced.


Examples

The following example displays the IOS SLB replication configuration:

Router# show ip slb replicate 
VS1, local = 10.10.99.132 remote = 10.10.99.99 port = 1024 
  current password = none pending password = none 
  password timeout = 180 sec (Default) 
  unsent conn updates:        0 
  conn updates received:      32 
  conn updates transmitted:   471 
  update packets received:    12 
  update packets transmitted: 34 
  failovers:                  0 
Router#

Table 8 describes the fields shown in the display.

Table 8 show ip slb replicate Field Descriptions 

Field
Description

local

Listening IP address for state exchange messages that are advertised.

remote

Destination IP address for all state exchange signals.

port

TCP or User Datagram Protocol (UDP) port number or port name for all state exchange signals.

current password

Current password for Message Digest Algorithm Version 5 (MD5) authentication, if any.

pending password

Pending password for MD5 authentication, if any.

password timeout

Delay period during which both the current (old) password and the pending (new) password are accepted.

unsent conn updates

Number of connection updates (changes in connection status) waiting to be sent to the remote switch.

conn updates received

Number of connection updates (changes in connection status) received from the remote switch.

conn updates transmitted

Number of connection updates (changes in connection status) sent to the remote switch.

update packets received

Number of packets containing connection updates (changes in connection status) received from the remote switch.

update packets transmitted

Number of packets containing connection updates (changes in connection status) sent to the remote switch.

failovers

Number of failovers detected. That is, the number of times the Hot Standby Router Protocol (HSRP) state has changed from Active to Standby, or from Standby to Active.


show ip slb serverfarms

To display information about the server farms, use the show ip slb serverfarms privileged EXEC command.

show ip slb serverfarms [name serverfarm-name] [detail]

Syntax Description

name

(Optional) Displays information about only a particular server farm.

serverfarm-name

(Optional) Name of the server farm.

detail

(Optional) Displays detailed server farm information.


Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example shows IOS SLB server farm data:

Router# show ip slb serverfarms

server farm      predictor     reals   bind id
-------------------------------------------------
FRAG             ROUNDROBIN    4       0       
LINUX            ROUNDROBIN    3       0       
SRE              ROUNDROBIN    6       0       
TEST             ROUNDROBIN    4       0       

Table 9 show ip slb serverfarms Field Descriptions 

Field
Description

server farm

Name of the server farm about which information is being displayed. Information about each server farm is displayed on a separate line.

predictor

Type of load-balancing algorithm (ROUNDROBIN or LEASTCONNS) used by the server farm.

reals

Number of real servers configured in the server farm.

bind id

Bind ID configured on the server farm.


show ip slb stats

To display IOS SLB statistics, use the show ip slb stats privileged EXEC command.

show ip slb stats

Syntax Description

This command has no arguments or keywords.

Defaults

No default behavior or values.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example shows IOS SLB statistics:

Router# show ip slb stats
 Pkts via normal switching:   0
 Pkts via special switching:  12534087
 Connections Created:         12466960
 Connections Established:     12240100
 Connections Destroyed:       12466960
 Connections Reassigned:      0
 Zombie Count:                0
 Connections Reused:          0
 Connection Flowcache Purges: 0
 Failed Connection Allocs:    0
 Failed Real Assignments:     0
 Total indications:           0
   Unknown conn indications:  0
   SYNACK indications:        0
   RST client indications:    0
   RST server indications:    0
   Two FINs indications:      0
   Flow age indications:      0

Table 10 show ip slb stats Field Descriptions 

Field
Description

Pkts via normal switching

Number of packets handled by IOS SLB via normal switching since the last time counters were cleared. Normal switching is when IOS SLB packets are handled on normal IOS switching paths (CEF, fastswitching, and process level switching).

Pkts via special switching

Number of packets handled by IOS SLB via special switching since the last time counters were cleared. Special switching is when IOS SLB packets are handled on hardware-assisted switching paths.

Connections Created

Number of connections created since the last time counters were cleared.

Connections Established

Number of connections created and have become established since the last time counters were cleared.

Connections Destroyed

Number of connections destroyed since the last time counters were cleared.

Connections Reassigned

Number of connections reassigned to a different real server since the last time counters were cleared.

Zombie Count

Number of connections that are currently pending destruction (awaiting a timeout or some other condition to be met).

Connections Reused

Number of zombie connections reused since the last time counters were cleared. A zombie connection is reused if it receives a TCP SYN or UDP packet and succeeds in connecting to a real server. The zombie connection becomes a real connection and the zombie count is decremented.

Connection Flowcache Purges

Number of times the connection flow cache was purged since the last time counters were cleared.

Failed Connection Allocs

Number of times the allocation of a connection failed since the last time counters were cleared.

Failed Real Assignments

Number of times the assignment of a real server failed since the last time counters were cleared.

Total indications

Total number of indications of all types since the last time counters were cleared.

  Unknown conn indications

Number of times unknown connections were detected since the last time counters were cleared.

  SYNACK indications

Number of TCP SYN/ACK events since the last time counters were cleared.

  RST client indications

Number of TCP RST client events since the last time counters were cleared.

  RST server indications

Number of TCP RST server events since the last time counters were cleared.

  Two FINs indications

Number of times two FINs (one in each direction) were detected on the connection since the last time counters were cleared.

  Flow age indications

Number of times TCP flows aged out since the last time counters were cleared.


show ip slb sticky

To display the IOS SLB sticky database, use the show ip slb sticky privileged EXEC command.

show ip slb sticky [client ip_address]

Syntax Description

client

(Optional) Displays only those sticky database entries associated with a particular client IP address or subnet.

ip-address

(Optional) IP address of the client.


Defaults

If no options are specified, the command displays information about all virtual servers or firewall farms.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example shows the IOS SLB sticky database:

Router# show ip slb sticky

client           netmask          group  real                  conns
-----------------------------------------------------------------------
10.10.2.12       255.255.0.0      4097   10.10.3.2             1 

Table 11 show ip slb sticky Field Descriptions 

Field
Description

client

Client IP address or subnet which is bound to this sticky assignment.

netmask

Subnet mask for this sticky assignment.

group

Group ID for this sticky assignment.

real

Real server used by all clients connecting with the client IP address or subnet detailed on this line.

conns

Number of connections currently sharing this sticky assignment.


show ip slb vserver

To display information about the virtual servers, use the show ip slb vserver privileged EXEC command.

show ip slb vserver [name virtual_server-name] [detail]

Syntax Description

name

(Optional) Displays information about only this virtual server.

virtual_server-name

(Optional) Name of the virtual server.

detail

(Optional) Displays detailed information.


Defaults

If no options are specified, the command displays information about all virtual servers.

Command Modes

Privileged EXEC

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example shows virtual server data:

Router# show ip slb vserver

slb vserver      prot   virtual               state          conns   
---------------------------------------------------------------------
TEST             TCP    80.80.254.3:80        OPERATIONAL    1013    
TEST21           TCP    80.80.254.3:21        OUTOFSERVICE   0       
TEST23           TCP    80.80.254.3:23        OUTOFSERVICE   0       

Table 12 show ip slb vserver Field Descriptions 

Field
Description

slb vserver

Name of the virtual server about which information is being displayed. Information about each virtual server is displayed on a separate line.

prot

Protocol being used by the virtual server.

virtual

Virtual IP address of the virtual server, including the network mask, if configured.

state

Current state of the virtual server:

FAILED—Real server represented by this virtual server has been removed from use by the predictor algorithms; retry timer started

OPERATIONAL—Functioning properly

OUTOFSERVICE—Removed from the load-balancing predictor lists

STANDBY—Backup virtual server, ready to become operational if active virtual server fails

TESTING_REAL—Real server represented by this virtual server is queued for assignment

conns

Number of connections associated with the virtual server.


The following example shows detailed data for a virtual server with route health injection (advertise=TRUE):

Router#show ip slb vserver detail
RH1, state = OPERATIONAL, v_index = 6
  virtual = 5.5.5.5/32:80, TCP, service = NONE, advertise = TRUE
  server farm = RHSF, delay = 10, idle = 3600
  backup server farm = BACKUP, use count = 0, backup sticky = FALSE
  sticky timer = 0, sticky subnet = 255.255.255.255
  sticky group id = 0
  synguard counter = 0, synguard period = 0
  conns = 1, total conns = 31484, syns = 0, syn drops = 0
  standby group = None

Table 13 show ip slb vserver detail Field Descriptions 

Field
Description

RH1

Name of the virtual server about which information is being displayed (in this case, RH1). Information about each virtual server is displayed on a separate line.

state

Current state of the virtual server:

FAILED—Real server represented by this virtual server has been removed from use by the predictor algorithms; retry timer started

OPERATIONAL—Functioning properly

OUTOFSERVICE—Removed from the load-balancing predictor lists

STANDBY—Backup virtual server, ready to become operational if active virtual server fails

TESTING_REAL—Real server represented by this virtual server is queued for assignment

v_index

Virtual index, out of a maximum of 127.

virtual

Virtual IP address of the virtual server, including the network mask, if configured.

TCP

Protocol being used by the virtual server (in this case, TCP).

service

Service, such as HTTP or Telnet, associated with the virtual server.

advertise

Current state of host route advertisement for this virtual server:

TRUE—Host route is being advertised.

FALSE—Host route is not being advertised.

server farm

Name of the server farm associated with the virtual server.

delay

Delay timer duration, in seconds, for this virtual server.

idle

Idle connection timer duration, in seconds, for this virtual server.

backup server farm

Name of the backup server farm associated with the virtual server.

use count

Number of times the backup server farm has taken over for the primary server farm in this period.

backup sticky

Indicates whether sticky connections are used in the backup server farm:

TRUE—Sticky connections are used in the backup server farm.

FALSE—Sticky connections are not used in the backup server farm.

sticky timer

Sticky timer duration, in seconds, for this virtual server.

sticky subnet

Sticky subnet in which this virtual server is placed, for coupling of services.

sticky group id

Sticky group in which this virtual server is placed, for coupling of services.

synguard counter

Number of unacknowledged SYNs that are allowed to be outstanding to this virtual server.

synguard period

Interval, in milliseconds, for SYN threshold monitoring for this virtual server.

conns

Number of active connections currently associated with the virtual server.

total conns

Total number of connections that have been associated with the virtual server since coming INSERVICE.

syns

Number of SYNs handled by the virtual server in this period.

syn drops

Number of SYNs dropped by the virtual server in this period.

standby group

HSRP group name with which the virtual server is associated.


standby authentication

To configure an authentication string for the Hot Standby Router Protocol (HSRP), use the standby authentication interface configuration command. To delete an authentication string, use the no form of this command.

standby [group-number] authentication string

no standby [group-number] authentication string

Syntax Description

group-number

(Optional) Group number on the interface to which this authentication string applies.

string

Authentication string. It can be up to eight characters in length.


Defaults

The group number default is 0.

The string default is cisco.

Command Modes

Interface configuration

Command History

Release
Modification

10.0

This command was introduced.


Usage Guidelines

The authentication string is transmitted unencrypted in all HSRP messages. The same authentication string must be configured on all routers and access servers on a cable to ensure interoperation. Authentication mismatch prevents a device from learning the designated Hot Standby IP address and the Hot Standby timer values from other routers configured with HSRP. Authentication mismatch does not prevent protocol events such as one router taking over as the designated router.

When group number 0 is used, no group number is written to NVRAM, providing backward compatibility.

Examples

In the following example, word is configured as the authentication string required to allow Hot Standby routers in group 1 to interoperate:

Router(config)# interface fastethernet 1
Router(config-if)# standby 1 authentication word

standby name

To specify an HSRP group name with which to associate an IOS SLB interface, use the standby name interface configuration command. To remove the group name association on the interface, use the no form of this command.

standby [group-number] name group-name

no standby [group-number] name group-name

Syntax Description

group-number

(Optional) Group number of the interface to which the timers apply. The default is 0.

group-name

Specifies the HSRP group name with which the IOS SLB virtual server is associated.


Defaults

The default group number is 0.

Command Modes

Interface configuration

Command History

Release
Modification

12.1(1)E

This command was introduced.


Usage Guidelines

The HSRP group name must first be specified on the inservice (server farm virtual server) command.

Examples

In the following example, HSRP is enabled for group number 1, group name Web-Group, on Ethernet port 0 on the EIP that is installed in slot 5:

Router(config)# interface Ethernet5/0
Router(config-if)#   ip address 172.18.48.154 255.255.255.128
Router(config-if)#   standby 1 ip 172.18.48.124
Router(config-if)#   standby 1 priority 2 preempt delay sync 20
Router(config-if)#   standby 1 name Web-Group

Related Commands

Command
Description

inservice (server farm virtual server)

Enables the virtual server for use by IOS SLB.


standby priority, standby preempt

To configure Hot Standby Router Protocol (HSRP) priority, preemption, and preemption delay, use the standby interface configuration command. To restore the default values, use the no form of this command.

standby [group-number] priority priority [preempt [delay [delay] [minimum [min-delay] | sync [sync-period]]]]

standby [group-number] [priority priority] preempt [delay [delay] [minimum [min-delay] | sync [sync-period]]]]

no standby [group-number] priority priority [preempt [delay [delay] [minimum [min-delay] | sync [sync-period]]]]

no standby [group-number] [priority priority] preempt [delay [delay] [minimum [min-delay] | sync [sync-period]]]]

Syntax Description

group-number

(Optional) Group number on the interface to which the other arguments in this command apply.

priority priority

(Optional) Priority value that prioritizes a potential Hot Standby router. The range is 1 to 255.

preempt

(Optional) The router is configured to preempt, which means that when the local router has a Hot Standby priority higher than the current active router, the local router should attempt to assume control as the active router. If preempt is not configured, the local router assumes control as the active router only if it receives information indicating that there is no router currently in the active state (acting as the designated router).

delay delay

(Optional) Time in seconds. The delay argument causes the local router to postpone taking over the active role for delay seconds since that router was last restarted. The range is 0 to 3600 seconds (1 hour). The default value is 0 seconds (if the router wants to preempt, it will do so immediately).

minimum min-delay

(Optional) Specifies the minimum time the local router waits before taking over the active role. The range is 0 to 3600 seconds (1 hour). The default value is 0 seconds (if the router wants to preempt, it will do so immediately).

sync sync-period

(Optional) Specifies the synchronization period for the local router, which is the maximum time the local router waits before attempting to assume control as the active router. The synchronization period gives the local router enough time to synchronize its functions with the active router before assuming control. The range is 0 to 3600 seconds (1 hour). The default value is 0 seconds (if the router wants to preempt, it will do so immediately, without synchronizing).

For IOS SLB, the recommended sync-period is 60 seconds.


Defaults

The group number default is 0.

The priority default is 100.

The delay default is 0 seconds (if the router wants to preempt, it will do so immediately).

The minimum delay default is 0 seconds (if the router wants to preempt, it will do so immediately).

The synchronization period default is 0 seconds (if the router wants to preempt, it will do so immediately, without synchronizing).

Command Modes

Interface configuration

Command History

Release
Modification

11.3

This command was introduced.

12.0(2)T

The syntax was restructured, and the minimum and sync keywords and min-delay and sync-period arguments were added.


Usage Guidelines

When using this command, you must specify at least one keyword (priority or preempt), or you can specify both.

When group number 0 is used, no group number is written to NVRAM, providing backward compatibility.

The assigned priority is used to help select the active and standby routers. Assuming preemption is enabled, the router with the highest priority becomes the designated active router. In case of ties, the primary IP addresses are compared, and the higher IP address has priority.

Note that the device's priority can change dynamically if an interface is configured with the standby track command and another interface on the router goes down.

When a router first comes up, it does not have a complete routing table. If it is configured to preempt, it will become the active router, yet it is unable to provide adequate routing services. This problem is solved by configuring a delay before the preempting router actually preempts the currently active router.

If you use the standby preempt command, you must also set the preempt synchronization delay or critical information, such as mobility bindings or load-balancing information, cannot be retrieved before the home agent preempts to become active. In the case where one router is designated as the active home agent, the priority is set highest in the HSRP group and preempt is set.

If you use the standby preempt delay command on both the backup server and backup client, you must specify the same delay on both.

Use the standby preempt delay sync command to ensure that all critical information, such as mobility bindings or load-balancing information, is downloaded to the local router before it takes the active role. When all critical information is downloaded or when the timer expires, whichever comes first, the router becomes active.

Examples

In the following example, the router has a priority of 120 (higher than the default value) and waits for 20 seconds before attempting to become the active router:

Router(config)# interface fastethernet 1
Router(config-if)# standby ip 172.19.108.254
Router(config-if)# standby priority 120 preempt delay sync 20

Related Commands

Command
Description

standby track

Configures the standby track on an interface so that the Hot Standby priority changes based on the availability of other interfaces.


standby timers

To configure the time between hellos and the time before other routers declare the active Hot Standby or standby router to be down, use the standby timers interface configuration command. To restore the timers to their default values, use the no form of this command.

standby [group-number] timers hellotime holdtime

no standby [group-number] timers hellotime holdtime

Syntax Description

group-number

(Optional) Group number on the interface to which the timers apply.

hellotime

Hello interval in seconds. This is an integer from 1 to 255.

holdtime

Time in seconds before the active or standby router is declared to be down. This is an integer from 1 to 255.


Defaults

The default group number is 0.

The default hellotime is 3 seconds.

The default holdtime is 3 seconds.

Command Modes

Interface configuration

Command History

Release
Modification

10.0

This command was introduced.


Usage Guidelines

The standby timers command configures the time between standby hellos and the time before other routers declare the active or standby router to be down. Routers or access servers on which timer values are not configured can learn timer values from the active or standby router. The timers configured on the active router always override any other timer settings. All routers in a Hot Standby group should use the same timer values. Normally, holdtime is greater than or equal to 3 times hellotime (holdtime > 3 * hellotime).

When group number 0 is used, no group number is written to NVRAM, providing backward compatibility.

Examples

In the following example, for group number 1 on Fast Ethernet interface 1, the time between hello packets is set to 5 seconds, and the time after which a router is considered to be down is set to 15 seconds:

Router(config)# interface fastethernet 1
Router(config-if)# standby 1 ip
Router(config-if)# standby 1 timers 5 15

standby track

To configure an interface so that the Hot Standby priority changes based on the availability of other interfaces, use the standby track interface configuration command. To remove the tracking, use the no form of this command.

standby [group-number] track type number [interface-priority]

no standby [group-number] track type number [interface-priority]

Syntax Description

group-number

(Optional) Group number on the interface to which the tracking applies.

type

Interface type (combined with interface number) that will be tracked.

number

Interface number (combined with interface type) that will be tracked.

interface-priority

(Optional) Amount by which the Hot Standby priority for the router is decremented (or incremented) when the interface goes down (or comes back up).


Defaults

The default group number is 0.

The default interface priority is 10.

Command Modes

Interface configuration

Command History

Release
Modification

10.3

This command was introduced.


Usage Guidelines

This command ties the router's Hot Standby priority to the availability of its interfaces. It is useful for tracking interfaces that are not configured for the HSRP.

When a tracked interface goes down, the Hot Standby priority decreases by 10. If an interface is not tracked, its state changes do not affect the Hot Standby priority. For each interface configured for Hot Standby, you can configure a separate list of interfaces to be tracked.

The optional argument interface-priority specifies how much to decrement the Hot Standby priority by when a tracked interface goes down. When the tracked interface comes back up, the priority is incremented by the same amount.

When multiple tracked interfaces are down and interface-priority values have been configured, these configured priority decrements are cumulative. If tracked interfaces are down, but none was configured with priority decrements, the default decrement is 10 and it is noncumulative.

When group number 0 is used, no group number is written to NVRAM, providing backward compatibility.

Examples

In the following example, Fast Ethernet interface 1 tracks Fast Ethernet interface 10 and Gigabit Ethernet interface 49. If one or both of these two interfaces go down, the Hot Standby priority of the router decreases by 10. Because the default Hot Standby priority is 100, the priority becomes 90 when one or both of the tracked interfaces go down.

Router(config)# interface fastethernet 1
Router(config-if)# ip address 198.92.72.37 255.255.255.240
Router(config-if)# no ip redirects
Router(config-if)# standby track fastethernet 10
Router(config-if)# standby track gigabitethernet 49
Router(config-if)# standby preempt delay sync 20
Router(config-if)# standby ip 198.92.72.46

Related Commands

Command
Description

standby priority, standby preempt

Configures the Hot Standby Router Protocol (HSRP) priority, preemption, and preemption delay.


sticky (firewall farm TCP protocol)

To assign all connections from a client to the same firewall, use the sticky firewall farm TCP protocol configuration command. To remove the client/server coupling use the no form of this command.

sticky duration [netmask netmask]

no sticky

Syntax Description

duration

Sticky timer duration in seconds. Valid values range from 0 to 65535.

netmask

(Optional) Places the virtual server as part of a sticky subnet, for coupling of services.

netmask

(Optional) Sticky subnet mask number.


Defaults

Virtual servers are not associated with any groups.

Command Modes

Firewall farm TCP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example specifies that if a client's subsequent request for a firewall farm is made within 60 seconds of the previous request, then the same firewall is used for the connection:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# tcp
Router(config-slb-fw-tcp)# sticky 60

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb sticky

Displays information about the IOS SLB database.

tcp

Initiates TCP protocol configuration mode.


sticky (firewall farm UDP protocol)

To assign all connections from a client to the same firewall, use the sticky firewall farm UDP protocol configuration command. To remove the client/server coupling use the no form of this command.

sticky duration [netmask netmask]

no sticky

Syntax Description

duration

Sticky timer duration in seconds. Valid values range from 0 to 65535.

netmask

(Optional) Places the virtual server as part of a sticky subnet, for coupling of services.

netmask

(Optional) Sticky subnet mask number.


Defaults

Virtual servers are not associated with any groups.

Command Modes

Firewall farm UDP protocol configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example specifies that if a client's subsequent request for a firewall farm is made within 60 seconds of the previous request, then the same firewall is used for the connection:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# udp
Router(config-slb-fw-udp)# sticky 60

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb sticky

Displays information about the IOS SLB database.

udp

Initiates UDP protocol configuration mode.


sticky (virtual server)

To assign all connections from a client to the same real server, use the sticky virtual server configuration command. To remove the client/server coupling use the no form of this command.

sticky duration [group group-id] [netmask netmask]

no sticky

Syntax Description

duration

Sticky timer duration in seconds. Valid values range from 0 to 65535.

group

(Optional) Places the virtual server in a sticky group, for coupling of services. In essence, the group keyword and group-id argument tie multiple virtual servers together.

group-id

(Optional) Number identifying the sticky group to which the virtual server belongs. Valid values range from 0 to 255.

netmask

(Optional) Places the virtual server as part of a sticky subnet, for coupling of services. Client sessions whose source IP addresses fall within the netmask are directed to the same real server.

netmask

(Optional) Sticky subnet mask number.


Defaults

Sticky connections are not tracked.

Virtual servers are not associated with any groups.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(2)E

The netmask keyword and netmask variable were added.


Usage Guidelines

The last real server that was used for a connection from a client is stored for the set duration seconds. If a new connection from the client to the virtual server is initiated during that time, the same real server that was used for the previous connection is chosen for the new connection. If two virtual servers are placed in the same group, coincident connection requests for those services from the same IP address are handled by the same real server.

Examples

The following example specifies that if a client's subsequent request for a virtual server is made within 60 seconds of the previous request, then the same real server is used for the connection. This example also places the virtual server in group 10.

Router(config)# ip slb vserver VS1
Router(config-slb-vserver)# sticky 60 group 10

Related Commands

Command
Description

show ip slb sticky

Displays information about the IOS SLB database.

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


synguard (virtual server)

To limit the rate of TCP SYNs handled by a virtual server to prevent a SYN flood denial-of-service attack, use the synguard virtual server configuration command. To remove the threshold, use the no form of this command.

synguard syn-count [interval]

no synguard

Syntax Description

syn-count

Number of unacknowledged SYNs that are allowed to be outstanding to a virtual server. Valid values range from 0 (off) to 4294967295. The default is 0.

interval

(Optional) Interval, in milliseconds, for SYN threshold monitoring. Valid values range from 50 to 5000. The default is 100 ms.


Defaults

Syn-count default: 0 (off)

Interval default: 100 ms

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Examples

The following example sets the threshold of unacknowledged SYNs to 50:

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# synguard 50

Related Commands

Command
Description

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.

virtual (virtual server)

Configures the virtual server attributes.


tcp

To initiate TCP protocol configuration mode, use the tcp firewall farm configuration command.

tcp

Syntax Description

This command has no arguments or keywords.

Defaults

No default behavior or values.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

There is not a no form of this command.

Examples

The following example initiates TCP protocol configuration mode:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# tcp
Router(config-slb-fw-tcp)# exit

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.


udp

To initiate UDP protocol configuration mode, use the udp firewall farm configuration command.

udp

Syntax Description

This command has no arguments or keywords.

Defaults

No default behavior or values.

Command Modes

Firewall farm configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Usage Guidelines

There is not a no form of this command.

Examples

The following example initiates UDP protocol configuration mode:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# udp
Router(config-slb-fw-udp)# exit

Related Commands

Command
Description

show ip slb firewallfarm

Displays information about the firewall farm configuration.


url (WSP probe)

To specify the URL path that a WSP probe is to request from the server, use the url WSP probe configuration command. To restore the default settings, use the no form of this command.

url [path]

no url [path]

Syntax Description

path

(Optional) Path from the server. This argument is case-sensitive.


Defaults

If no URL path is specified, the default is /.

Command Modes

WSP probe configuration

Command History

Release
Modification

12.1(5a)E

This command was introduced.


Examples

The following example configures a ping probe named PROBE3, changes the CLI to IOS SLB WSP probe submode, and configures the probe to request URL path http://localhost/test.txt:

Router(config)# ip slb probe PROBE3 wsp
Router(config-slb-probe)# url http://localhost/test.txt

Related Commands

Command
Description

ip slb probe (WSP probe)

Configures a WSP probe name and changes to WSP probe configuration submode.

show ip slb probe

Displays information about an IOS SLB probe.


virtual (virtual server)

To configure virtual server attributes, use the virtual virtual server configuration command. To remove the attributes, use the no form of this command.

virtual ip-address [network-mask] {tcp | udp} [port-number | wsp | wsp-wtp | wsp-wtls | wsp-wtp-wtls] [service service-name]

no virtual

Syntax Description

ip-address

IP address for this virtual server instance, used by clients to connect to the server farm.

network-mask

(Optional) Specifies the IP network mask for transparent webcache load balancing. The default is 0.0.0.0 (all subnets).

tcp

Performs load balancing for only TCP connections.

udp

Performs load balancing for only UDP connections.

port-number

(Optional) IOS SLB virtual port (the TCP or UDP port number or port name). If specified, only the connections for the specified port on the server are load-balanced. The ports and the valid name or number for the port-number argument are as follows:

Domain Name System: dns   53

File Transfer Protocol: ftp   21

HTTP over Secure Socket Layer: https   443

Mapping of Airline Traffic over IP, Type A: matip-a   350

Network News Transport Protocol: nntp   119

Post Office Protocol v2: pop2   109

Post Office Protocol v3: pop3   110

Simple Mail Transport Protocol: smtp   25

Telnet: telnet   23

World Wide Web (HTTP): www   80

Specify a port number of 0 to configure an all-port virtual server (that is, a virtual server that accepts flows destined for all ports).

wsp

(Optional) Performs load balancing in connectionless WSP mode on port 9200.

wsp-wtp

(Optional) Performs load balancing in connection-oriented WSP mode on port 9201, and activates the WAP FSM, which monitors WSP and drives the session FSM accordingly.

wsp-wtls

(Optional) Performs load balancing in connectionless secure WSP mode on port 9202.

wsp-wtp-wtls

(Optional) Performs load balancing in connection-oriented secure WSP mode on port 9203.

service

(Optional) Couples connections associated with a given service, such as HTTP or Telnet, so all related connections from the same client use the same real server.

service-name

(Optional) Type of connection coupling:

ftp—Couples FTP data connections with the control session that created them.

wsp-wtp—Couples WSP data connections with the control session that created them.


Defaults

No default behavior or values.

Command Modes

Virtual server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(5a)E

The wsp, wsp-wtp, wsp-wtls, and wsp-wtp-wtls keywords were added.


Usage Guidelines

The no virtual command is allowed only if the virtual server was removed from service by the no inservice command.

For some applications, it is not feasible to configure all the virtual server TCP or UDP port numbers for IOS SLB. To support such applications, you can configure IOS SLB virtual servers to accept flows destined for all ports. To configure an all-port virtual server, specify a port number of 0.

On Catalyst 6000 Family Switches, if you use FTP sessions with MNLB you must configure a port-bound virtual server bound to port 21 on the MNLB Services Manager (the LocalDirector).

In general, you should use port-bound virtual servers instead of all-port virtual servers. When you use all-port virtual servers, flows can be passed to servers for which no application port exists. When the servers reject these flows, IOS SLB might fail the servers and remove them from load balancing. To prevent this problem, use one of the following procedures:

Configure one or more port-bound virtual servers.

Configure access control lists on the IOS SLB ingress interface to permit only ports for supported applications on the real server.

Configure faildetect numconns 255 numclients 8 on the real server.

Examples

The following example specifies that the virtual server with the IP address 10.0.0.1 performs load balancing for TCP connections for the port named www. The virtual server processes HTTP requests.

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# virtual 10.0.0.1 tcp www

The following example specifies that the virtual server with the IP address 10.0.0.13 performs load balancing for UDP connections for all ports. The virtual server processes HTTP requests.

Router(config)# ip slb vserver PUBLIC_HTTP
Router(config-slb-vserver)# virtual 10.0.0.13 udp 0

Related Commands

Command
Description

ip slb vserver

Identifies a virtual server.

show ip slb vserver

Displays information about the virtual servers defined to IOS SLB.


weight (firewall farm real server)

To specify a real server's capacity, relative to other real servers in the firewall farm, use the weight firewall farm real server configuration command. To restore the default weight value, use the no form of this command.

weight weighting-value

no weight

Syntax Description

weighting-value

Weighting value to use for real server predictor algorithm. Valid values range from 1 to 255. The default weighting value is 8.


Defaults

Weighting-value default: 8

Command Modes

Firewall farm real server configuration

Command History

Release
Modification

12.1(3a)E

This command was introduced.


Examples

The following example specifies the relative weighting values of three real servers as 16, 8 (by default), and 24, respectively:

Router(config)# ip slb firewallfarm FIRE1
Router(config-slb-fw)# real 10.10.1.1		First real server
Router(config-slb-fw-real)# weight 16		Assigned weight of 16
Router(config-slb-fw-real)# inservice		Enabled
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# real 10.10.1.2		Second real server
Router(config-slb-fw-real)# inservice		Enabled; default weight
Router(config-slb-fw-real)# exit
Router(config-slb-fw)# real 10.10.1.3		Third real server
Router(config-slb-fw-real)# weight 24		Assigned weight of 24; 	 not enabled

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb firewallfarm

Displays information about the firewall farm configuration.

show ip slb reals

Displays information about the real servers.


weight (server farm)

To specify a real server's capacity, relative to other real servers in the server farm, use the weight real server configuration command. To restore the default weight value, use the no form of this command.

weight weighting-value

no weight

Syntax Description

weighting-value

Weighting value to use for real server predictor algorithm. Valid values range from 1 to 255. The default weighting value is 8.


Defaults

Weighting-value default: 8

Command Modes

Real server configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.


Usage Guidelines

The static weights you define using this command are overridden by the weights calculated by DFP. If DFP is removed from the network, IOS SLB reverts to these static weights.

Examples

The following example specifies the relative weighting values of three real servers as 16, 8 (by default), and 24, respectively:

Router(config)# ip slb serverfarm PUBLIC
Router(config-slb-sfarm)# real 10.10.1.1		First real server
Router(config-slb-real)# weight 16		Assigned weight of 16
Router(config-slb-real)# inservice		Enabled
Router(config-slb-real)# exit
Router(config-slb-sfarm)# real 10.10.1.2		Second real server
Router(config-slb-real)# inservice		Enabled; default weight
Router(config-slb-real)# exit
Router(config-slb-sfarm)# real 10.10.1.3		Third real server
Router(config-slb-real)# weight 24		Assigned weight of 24; 	 not enabled

Related Commands

Command
Description

real (server farm)

Identifies a real server as a member of a server farm and initiates real server configuration mode.

show ip slb serverfarms

Displays information about the server farm configuration.

show ip slb reals

Displays information about the real servers.


Debug Commands

This section documents the following new debug command related to the IOS SLB feature:

debug ip dfp agent

debug ip slb

debug ip dfp agent

To display debug messages for the DFP agent subsystem, use the debug ip dfp EXEC command. To stop debug output, use the no form of this command.

debug ip dfp agent

no debug ip dfp agent

Syntax Description

This command has no arguments or keywords.

Defaults

No default behavior or values.

Command Modes

EXEC configuration

Command History

Release
Modification

12.1(8a)E

This command was introduced.


Usage Guidelines

See the following caution before using debug commands:


Caution Because debugging output is assigned high priority in the CPU process, it can render the system unusable. For this reason, use debug commands only to troubleshoot specific problems or during troubleshooting sessions with Cisco technical support staff. Moreover, it is best to use debug commands during periods of lower network flows and fewer users. Debugging during these periods reduces the effect these commands have on other users on the system.

This command displays debug messages for the DFP agent subsystem.

Examples

The following example configures a DFP agent debug session:

Router# debug ip dfp agent
DFP debugging is on
Router#

The following example stops all debugging:

Router# no debug all
All possible debugging has been turned off
Router#

debug ip slb

To display debug messages for IOS SLB, use the debug ip slb EXEC command. To stop debug output, use the no form of this command.

debug ip slb {conns | dfp | firewallfarm | icmp | natpool | probe | reals | replication | sessions | vservers | all}

no debug ip slb {conns | dfp | firewallfarm | icmp | natpool | probe | reals | replication | sessions | vservers | all}

Syntax Description

all

Displays all debug messages for IOS SLB.

conns

Displays debug messages for all connections being handled by IOS SLB, including WSP events and states.

dfp

Displays debug messages for DFP. To display debug messages for the DFP agent subsystem, use debug ip dfp agent.

firewallfarm

Displays debug messages related to firewall load balancing.

icmp

Displays all Internet Control Message Protocol debug messages for IOS SLB.

natpool

Displays debug messages related to the IOS SLB client NAT pool.

probe

Displays debug messages related to probes.

reals

Displays debug messages for all real servers defined to IOS SLB.

replication

Displays debug messages related to IOS SLB stateful backup virtual server.

sessions

Displays debug messages for all sessions being handled by IOS SLB.

vservers

Displays debug messages for all virtual servers defined to IOS SLB.


Defaults

No default behavior or values.

Command Modes

EXEC configuration

Command History

Release
Modification

12.0(7)XE

This command was introduced.

12.1(2)E

The natpool and replication keywords were added.

12.1(3a)E

The firewallfarm keyword was added.

12.1(7)E

The vservers keyword was added.

12.1(8a)E

The sessions keyword was added.


Usage Guidelines

See the following caution before using debug commands:


Caution Because debugging output is assigned high priority in the CPU process, it can render the system unusable. For this reason, use debug commands only to troubleshoot specific problems or during troubleshooting sessions with Cisco technical support staff. Moreover, it is best to use debug commands during periods of lower network flows and fewer users. Debugging during these periods reduces the effect these commands have on other users on the system.

Examples

The following example configures a debug session to check all IP IOS SLB parameters:

Router# debug ip slb all
SLB All debugging is on
Router# 

The following example stops all debugging:

Router# no debug all
All possible debugging has been turned off
Router#

The following example configures debugging to check IP IOS SLB replication used with stateful backup and displays the output from the send or transmit virtual server:

Router# debug ip slb replication
*Mar  2 08:02:38.019:  SLB Replicate: (send) update vs: VS1 update_count 42

FAQ (Frequently Asked Questions)

The following questions and answers can help you troubleshoot IOS SLB, if you have problems.

Question
Answer

Can I use IOS SLB to load balance clients and real servers that are on the same LAN or VLAN?

NO!

IOS SLB does not support load balancing of flows between clients and real servers that are on the same LAN or VLAN. The packets being load balanced cannot enter and leave the load-balancing device on the same interface.

Why is IOS SLB not marking my connections as ESTABLISHED even though I'm transferring data?

If you are using dispatched mode, make sure there are no alternate paths that allow outbound flows to bypass IOS SLB. Also, make sure the clients and real servers are not on the same IP subnet.

Why am I able to connect to real servers directly, but unable to connect to the virtual server?

Make sure that the virtual IP address is configured as a loopback in each of the real servers (if you are running in dispatched mode).

Why is IOS SLB not marking my real server as failed when I disconnect it from the network?

Tune the values for the numclients, numconns, and delay keywords.

If you have a very small client population (for example, in a test environment), the numclients keyword could be causing the problem. This parameter prevents IOS SLB from mistaking the failure of a small number of clients for the failure of a real server.

Why does IOS SLB show my real server as INSERVICE even though I have taken it down or physically disconnected it?

The INSERVICE and OUTOFSERVICE states indicate whether the network administrator intends for that real server to be used when it is operational. A real server that was INSERVICE but was removed from the selection list dynamically by IOS SLB as a result of automatic failure detection, is marked as FAILED. Use the show ip slb reals detail command to display these real server states.

Beginning with release 12.1(1)E, INSERVICE is changed to OPERATIONAL, to better reflect what is actually occurring.

Why is IOS SLB not balancing correctly? I am using dispatched mode, the servers are leaving sockets open, and I am seeing RSTs in response to a number of SYNs. Curiously, sometimes things work fine.

Enter the show mls flow command:

Router# sh mls flow
current ip flowmask for unicast: full flow
current ipx flowmask for unicast: destination only

The current IP flowmask must be full flow. If it is not, correct the problem using the mls flow ip full command:

Router# config t
Enter configuration commands, one per line.
End with CNTL/Z.
Router(config)# mls flow ip full
Router(config)# 

How can I verify that IOS SLB sticky connections are working properly?

Use the following procedure:

1. Configure the sticky connections.

2. Start a client connection.

3. Enter the show ip slb reals detail and show ip slb conns commands.

4. Examine the real server connection counts. The real server whose count increased is the one to which the client connection is assigned.

5. Enter the show ip slb sticky command to display the sticky relationships that IOS SLB stored.

6. End the connection.

7. Ensure that the real server's connection count decreased.

8. Restart the connection, after waiting no longer than the sticky timeout value.

9. Enter the show ip slb conns command again.

10. Examine the real server connection counts again, and verify that the sticky connection is assigned to the same real server as before.

How can I verify that server failures are being detected correctly?

Use the following procedure:

1. Use a large client population. If the number of clients is very small, tune the numclients keyword on the faildetect (real server) command so that the servers are not displayed as FAILED.

2. Enter the show ip slb reals detail command to show the status of the real servers.

3. Examine the status and connection counts of the real servers.

Servers that failed show a status of FAILED, TESTING, or READY_T0_TEST, based on whether IOS SLB is checking that the server came back up when the command was sent.

When a real server fails, connections that are assigned but not established (no SYN or ACK is received) are reassigned to another real server on the first inbound SYN after the reassign threshold is met. However, any connections that were already established are forwarded to the same real server because, while it might not be accepting new connections, it might be servicing existing ones.

For weighted least connections, a real server that has just been placed in service starts slowly so that it is not overloaded with new connections. (See the "Slow Start" section for more information.) Therefore, the connection counts displayed for a new real server show connections going to other real servers (despite the new real server's lower count). The connection counts also show "dummy connections" to the new real server, which IOS SLB uses to artificially inflate the connection counts for the real server during the slow start period.

Does the no inservice command take a resource out of service immediately?

When you use the no form of the inservice command to remove a firewall, firewall farm, real server, or virtual server from service, the resource quiesces gracefully. No new connections are assigned, and existing connections are allowed to complete.

To stop all existing connections for an entire firewall farm or virtual server immediately, use the clear ip slb command.

I configured both IOS SLB and input ACLs on the same Catalyst 6000 Family Switch, and now I see TCAM Capacity Exceeded messages. Why?

If you configure IOS SLB and either input ACLs or firewall load balancing on the same Catalyst 6000 Family Switch, you can exceed the capacity of the TCAM on the Policy Feature Card (PFC). To correct the problem, use the mls ip slb search wildcard rp command to reduce the amount of TCAM space used by IOS SLB, but be aware that this command can result in a slight increase in route processor utilization.


Glossary

active standby—Redundancy scheme in which two IOS SLB devices can load-balance the same virtual IP address while at the same time acting as backups for each other. See also active standby, stateful backup.

bearer network—Network that carries messages of a transport-layer protocol, and ultimately also of the session-layer protocols, between physical devices. A single session can use more than one bearer network.

CASA—Content Aware Services Architecture. CASA is a protocol designed to allow network appliances to selectively control the flow of IP packets through a router, switch, or other network device.

client NAT—Translation scheme in which the client IP address is replaced with an IP address associated with one of a group of load-balancing devices, resulting in outbound flows being routed to the correct device. See also directed mode, server NAT.

client subsystem—Users, such as IOS SLB, of the DFP agent function.

cluster—Set of computer systems that are connected through multisystem hardware or software to supply services traditionally provided by a single system. This arrangement provides higher availability and better scalability of the system.

content-aware networking—Networking strategy that enables content to be dynamically distributed. Because content can be dynamically cached, it can be located at any given place at any given time and distributed between the servers and the location of the webcache. Cisco has developed the ContentFlow architecture and the DFP to enable networks to provide content-aware networking services.

ContentFlow architecture—Cisco's content-aware networking architecture that describes message flows and actions in a distributed environment.

DFP—Dynamic Feedback Protocol. Allows host agents to dynamically report the change in status of the host systems providing a virtual service. The status reported is a relative weight that specifies a host server's capacity to perform work.

DFP agent—Host agent in a load-balanced environment that dynamically reports changes in status of the host systems that provide a virtual service. The status reported is a relative weight that specifies a host server's capacity to perform work. See also DFP manager.

DFP manager—Host manager in a load-balanced environment that collects status reports from DFP agents. See also DFP agent.

directed mode—Session redirection mode in which the virtual server can be assigned an IP address that is not known to any of the real servers. IOS SLB translates packets exchanged between a client and real server, translating the virtual server IP address to a real server IP address through NAT. See also dispatched mode, NAT.

dispatched mode—Session redirection mode in which the virtual server address is known to the real servers. The virtual server IP address must be configured as a loopback address, or secondary IP address, on each of the real servers. IOS SLB redirects packets to the real servers at the media access control (MAC) layer. Since the virtual server IP address is not modified in dispatched mode, the real servers must be Layer 2-adjacent to IOS SLB, or intervening routers might not be able to route to the chosen real server. See also directed mode.

Dynamic Feedback Protocol—See DFP.

firewall—Router or access server, or several routers or access servers, designated as a buffer between any connected public networks and a private network. A firewall router uses access lists and other methods to ensure the security of the private network.

firewall farm—Group of firewalls.

firewall load balancing—Load-balancing scheme in which the network administrator configures a group of firewalls into a firewall farm. When a client initiates a connection, IOS SLB chooses a firewall for the connection based on a hash algorithm.

HSRP—Hot Standby Router Protocol. Provides high network availability and transparent network topology changes. HSRP creates a Hot Standby router group with a lead router that services all packets sent to the Hot Standby address. The lead router is monitored by other routers in the group, and if it fails, one of these standby routers inherits the lead position and the Hot Standby group address. See also redundancy, stateful backup, stateless backup.

HTTP redirect load balancing—Load-balancing scheme in which all HTTP requests that belong to the same transaction are directed to the same real server.

IOS SLB—IOS Server Load Balancing. Load-balancing scheme in which the network administrator defines a virtual server that represents a group of real servers in a cluster of network servers known as a server farm. When a client initiates a connection to the virtual server, IOS SLB chooses a real server for the connection based on a configured load-balancing algorithm.

load balancing—Spreading user requests among available servers within a cluster of servers, based on a variety of algorithms.

MD5—Message Digest Algorithm Version 5. Neighbor router authentication scheme used to ensure reliability and security when routing updates are exchanged between neighbor routers.

Message Digest Algorithm Version 5—See MD5.

NAT—Network Address Translation. Modification of one or more of the following fields in an IP packet: source IP address, destination IP address, source TCP/UDP port, destination TCP/UDP port. See also client NAT, server NAT.

NetFlow switching—High-performance network-layer switching path that captures as part of its switching function a rich set of traffic statistics including user, protocol, port, and type of service information.

Network Address Translation—See NAT.

port-bound—Server configuration scheme in which a virtual server IP address represents one set of real servers for one service, such as Hypertext Transfer Protocol (HTTP), and a different set of real servers for another service, such as Telnet.

real server—The specification of a physical server associated with a virtual server. The specification includes the real server's IP address and an optional weight to be used by the virtual server predictor.

redundancy—The duplication of devices, services, or connections so that, in the event of a failure, the redundant devices, services, or connections can perform the work of those that failed. See also stateful backup, stateless backup.

round robin—See weighted round robin.

Secure Socket Layer—See SSL.

server cluster—See server farm.

server farm—Also called a server cluster. Group of real servers that provide various applications and services.

Server Load Balancing—See IOS SLB.

server NAT—Translation scheme in which the virtual server IP address is replaced with the real server IP address (and vice versa), allowing servers to be many hops away from the load-balancing device, and enabling intervening routers to route to them without requiring tunnelling. See also client NAT, directed mode, server port translation.

server port translation—Translation scheme in which the virtual server port number is replaced with the real server port number (and vice versa), allowing servers to be many hops away from the load-balancing device, and enabling intervening routers to route to them without requiring tunnelling. See also client NAT, directed mode, server NAT.

services manager—Functionality built into IOS SLB that makes load-balancing decisions based on application availability, server capacity, and load distribution algorithms such as weighted round robin or weighted least connections, or the DFP. The services manager determines a real server for the packet flow using load balancing and server/application feedback.

SLB—See IOS SLB.

SSL—Secure Socket Layer. Encryption technology for the web used to provide secure transactions such as the transmission of credit card numbers for e-commerce.

stateful backup—Redundancy scheme that enables IOS SLB to incrementally backup its load-balancing decisions, or "keep state," between primary and backup switches. See also active standby, stateless backup.

stateless backup—Redundancy scheme that provides high network availability by routing IP flows from hosts on Ethernet networks without relying on the availability of a single Layer 3 switch. See also active standby, stateful backup.

sticky connections—Load-balancing scheme in which new connections from a client IP address or subnet are assigned to the same real server (for server load balancing) or firewall (for firewall load balancing) as were previous connections from that address or subnet.

virtual server—Presents a single address that represents an application server farm to clients.

WAP—Wireless Application Protocol. Suite of protocols used to deliver services to wireless devices.

weighted least connection—Load-balancing algorithm in which the next real server chosen for a new connection to the virtual server is the server with the fewest active connections. Each real server is assigned a weight, n, that represents its capacity to handle connections, as compared to the other real servers associated with the virtual server. The server with the fewest connections is based on the number of active connections on each server, and on the relative capacity of each server. The capacity of a given real server is calculated as the assigned weight of that server divided by the sum of the assigned weights of all of the real servers associated with that virtual server, or n1/(n1+n2+n3...).

weighted round robin—Load-balancing algorithm in which the real server used for a new connection to the virtual server is chosen in a circular fashion. Each real server is assigned a weight, n, that represents its capacity to handle connections, as compared to the other real servers associated with the virtual server. New connections are assigned to a given real server n times before the next real server in the list is chosen.

Wireless Application Protocol—See WAP.

Wireless Session Protocol—See WSP.

Wireless Transaction Protocol—See WTP.

Wireless Transport Security Layer—See WTLS.

workload agents—Value-added software components developed for specific platforms by third-party developers. Workload agents run on server platforms or on platforms that manage server farms. Workload agents deliver server and application information to the services manager. This information enables the services manager to make optimum server selection.

WSP—Wireless Session Protocol. Session-layer protocol of the WAP suite.

WTLS—Wireless Transport Security Layer. Layer that provides security between WAP clients and WAP gateways.

WTP—Wireless Transaction Protocol. Transaction-layer protocol of the WAP suite.