-
Cisco IOS Quality of Service Solutions Command Reference, Release 12.1
-
About the Cisco IOS Software Documentation
-
Using Cisco IOS Software
-
Commands: access-list rate-limit -- frame-relay ip rtp priority
-
Commands: ip rsvp atm-peak-rate-limit -- ip rtp priority
-
Commands: match access-group -- pvc-bundle
-
Commands: queue-limit -- set ip precedence
-
Commands: show access-lists rate-limit -- traffic-shape rate
-
Index
-
Table Of Contents
random-detect exponential-weighting-constant
random-detect flow average-depth-factor
queue-limit
To specify or modify the maximum number of packets the queue can hold for a class policy configured in a policy map, use the queue-limit policy-map class configuration command. To remove the queue packet limit from a class, use the no form of this command.
queue-limit number-of-packets
no queue-limit number-of-packets
Syntax Description
number-of-packets
A number in the range 1 to 64 specifying the maximum number of packets that the queue for this class can accumulate.
Defaults
64
Command Modes
Policy-map class configuration
Command History
Usage Guidelines
Weighted fair queueing (WFQ) creates a queue for every class for which a class map is defined. Packets satisfying the match criteria for a class accumulate in the queue reserved for the class until they are sent, which occurs when the queue is serviced by the fair queueing process. When the maximum packet threshold you defined for the class is reached, enqueueing of any further packets to the class queue causes tail drop or, if Weighted Random Early Detection (WRED) is configured for the class policy, packet drop to take effect.
Examples
The following example configures a policy map called policy11 to contain policy for a class called acl203. Policy for this class is set so that the queue reserved for it has a maximum packet limit of 40.
policy-map policy11class acl203bandwidth 2000queue-limit 40Related Commands
queue-list default
To assign a priority queue for those packets that do not match any other rule in the queue list, use the queue-list default global configuration command. To restore the default value, use the no form of this command.
queue-list list-number default queue-number
no queue-list list-number default queue-number
Syntax Description
Defaults
This command is disabled by default.
Command Modes
Global configuration
Command History
Usage Guidelines
When you use multiple rules, remember that the system reads the queue-list commands in order of appearance. When classifying a packet, the system searches the list of rules specified by queue-list commands for a matching protocol or interface type. When a match is found, the system assigns the packet to the appropriate queue. The system searches the list in the order specified, and the first matching rule terminates the search.
Queue number 0 is a system queue. It is emptied before any of the other queues are processed. The system enqueues high-priority packets, such as keepalives, to this queue.
Use the show interfaces command to display the current status of the output queues.
Examples
In the following example, the default queue for list 10 is set to queue number 2:
queue-list 10 default 2Related Commands
queue-list interface
To establish queueing priorities on packets entering on an interface, use the queue-list interface global configuration command. To remove an entry from the list, use the no form of this command.
queue-list list-number interface interface-type interface-number queue-number
no queue-list list-number interface interface-type interface-number queue-number
Syntax Description
Defaults
No queueing priorities are established.
Command Modes
Global configuration
Command History
Usage Guidelines
When you use multiple rules, remember that the system reads the queue-list commands in order of appearance. When classifying a packet, the system searches the list of rules specified by queue-list commands for a matching protocol or interface type. When a match is found, the system assigns the packet to the appropriate queue. The list is searched in the order specified, and the first matching rule terminates the search.
Examples
In the following example, queue list 4 establishes queueing priorities for packets entering on interface tunnel 3. The queue number assigned is 10.
queue-list 4 interface tunnel 3 10Related Commands
queue-list lowest-custom
To set the lowest number for a queue to be treated as a custom queue, use the queue-list lowest-custom command in global configuration mode. To restore the default value, use the no form of this command.
queue-list list-number lowest-custom queue-number
no queue-list list-number lowest-custom queue-number
Syntax Description
list-number
Number of the queue list. Any number from 1 to 16 that identifies the queue list.
queue-number
Number of the queue. Any number from 1 to 16.
Defaults
The default number of the lowest custom queue is 1.
Command Modes
Global configuration
Command History
Usage Guidelines
All queues from queue 0 to the queue prior to the one specified in the queue-list lowest-custom command use the priority queue. (Queue 0 has the highest priority.)
All queues from the one specified in the queue-list lowest-custom command to queue 16 use a round-robin scheduler.
Use the show queueing custom command to display the current custom queue configuration.
Examples
In the following example, the lowest custom value is set to 2 for queue list 4:
queue-list 4 lowest-custom 2Related Commands
queue-list protocol
To establish queueing priority based upon the protocol type, use the queue-list protocol global configuration command. To remove an entry from the list, use the no form of this command with the appropriate arguments.
queue-list list-number protocol protocol-name queue-number queue-keyword keyword-value
no queue-list list-number protocol protocol-name queue-number queue-keyword keyword-value
Syntax Description
list-number
Number of the queue list. Any number from 1 to 16.
protocol-name
Protocol type: aarp, apollo, appletalk, arp, bridge (transparent), clns, clns_es, clns_is, cmns, compressedtcp, decnet, decnet_node, decnet_routerl1, decnet_routerl2, dlsw, ip, ipx, pad, rsrb, stun, vines, xns, and x25.
queue-number
Number of the queue. Any number from 1 to 16.
queue-keyword keyword-value
Possible keywords are fragments, gt, list, lt, tcp, and udp. See Table 8 from the priority-list protocol command.
Defaults
No queueing priorities are established.
Command Modes
Global configuration
Command History
Usage Guidelines
When you use multiple rules for a single protocol, remember that the system reads the queue-list commands in order of appearance. When classifying a packet, the system searches the list of rules specified by queue-list commands for a matching protocol. When a match is found, the system assigns the packet to the appropriate queue. The system searches the list in the order specified, and the first matching rule terminates the search.
The decnet_router-l1 keyword refers to the multicast address for all level 1 routers, which are intra-area routers, and the decnet_router-l2 keyword refers to all level 2 routers, which are interarea routers.
The dlsw, rsrb, and stun keywords refer only to direct encapsulation.
Use Table 8, Table 9, and Table 10 from the priority-list protocol command to configure the queueing priorities for your system.
Examples
The following example assigns 1 as the custom queue list, specifies DECnet as the protocol type, and assigns 3 as a queue number to the packets sent on this interface:
queue-list 1 protocol decnet 3The following example assigns DECnet packets with a size greater than 200 bytes to queue number 2:
queue-list 2 protocol decnet 2 gt 200The following example assigns DECnet packets with a size less than 200 bytes to queue number 2:
queue-list 4 protocol decnet 2 lt 200The following example assigns traffic that matches IP access list 10 to queue number 1:
queue-list 1 protocol ip 1 list 10The following example assigns Telnet packets to queue number 2:
queue-list 4 protocol ip 2 tcp 23The following example assigns UDP Domain Name Service packets to queue number 2:
queue-list 4 protocol ip 2 udp 53The following example assigns traffic that matches Ethernet type code access list 201 to queue number 1:
queue-list 1 protocol bridge 1 list 201Related Commands
queue-list queue byte-count
To specify how many bytes the system allows to be delivered from a given queue during a particular cycle, use the queue-list queue byte-count global configuration command. To return the byte count to the default value, use the no form of this command.
queue-list list-number queue queue-number byte-count byte-count-number
no queue-list list-number queue queue-number byte-count byte-count-number
Syntax Description
Defaults
This command is not enabled by default.
The default byte count is 1500 bytes.
Command Modes
Global configuration
Command History
Examples
In the following example, queue list 9 establishes the byte count as 1400 for queue number 10:
queue-list 9 queue 10 byte-count 1400Related Commands
queue-list queue limit
To designate the queue length limit for a queue, use the queue-list queue limit global configuration command. To return the queue length to the default value, use the no form of this command.
queue-list list-number queue queue-number limit limit-number
no queue-list list-number queue queue-number limit limit-number
Syntax Description
Defaults
The default queue length limit for a queue is 20.
Command Modes
Global configuration
Command History
Examples
In the following example, the queue length of queue 10 is increased to 40:
queue-list 5 queue 10 limit 40Related Commands
random-detect (interface)
To enable Weighted Random Early Detection (WRED) or VIP-Distributed WRED (DWRED), use the random-detect interface configuration command. To configure WRED as class policy in a policy map, use the random-detect policy-map class configuration command. To disable WRED or DWRED, use the no form of this command.
random-detect
no random-detect
Syntax Description
This command has no arguments or keywords.
Defaults
WRED and DWRED are disabled by default.
Command Modes
Interface configuration when used on an interface.
Policy-map class configuration when used to specify class policy in a policy map.
Command History
Usage Guidelines
WRED is a congestion avoidance mechanism that slows traffic by randomly dropping packets when congestion exists. DWRED is similar to WRED but uses the Versatile Interface Processor (VIP) instead of the Route Switch Processor (RSP). WRED and DWRED are most useful with protocols like TCP that respond to dropped packets by decreasing the transmission rate.
The router automatically determines parameters to use in the WRED calculations. To change these parameters, use the random-detect precedence command.
The DWRED feature is only supported on Cisco 7000 series routers with an RSP7000 card and Cisco 7500 series routers with a VIP2-40 or greater interface processor. A VIP2-50 interface processor is strongly recommended when the aggregate line rate of the port adapters on the VIP is greater than DS3. A VIP2-50 interface processor is required for OC-3 rates.
To use DWRED, distributed Cisco Express Forwarding (dCEF) switching must first be enabled on the interface. For more information on dCEF, refer to the Cisco IOS Switching Services Configuration Guide and the Cisco IOS Switching Services Command Reference.
WRED in a Policy Map
You can configure WRED as part of the policy for a standard class or the default class. The WRED random-detect command and the weighted fair queueing (WFQ) queue-limit command are mutually exclusive for class policy. If you configure WRED, its packet drop capability is used to manage the queue when packets exceeding the configured maximum count are enqueued. If you configure the WFQ queue-limit command for class policy, tail drop is used.
To configure a policy map and create class policies, use the policy-map and class (policy-map) commands. When specifying class policy within a policy map, you can use the random-detect command with either of the following commands:
•
bandwidth (policy-map class)
•
Note that if you use WRED packet drop instead of tail drop for one or more classes composing a policy map, you must ensure that WRED is not configured for the interface to which you attach that service policy.
The DWRED feature is not supported for class policy.
Examples
The following example configures WRED on the High-Speed Serial Interface (HSSI) 0/0/0 interface:
interface Hssi0/0/0random-detectThe following example configures the policy map called policy1 to contain policy specification for the class called class1. During times of congestion, WRED packet drop is used instead of tail drop.
! The following commands create the class map called class1:class-map class1match input-interface FE0/1! The following commands define policy1 to contain policy specification for class1:policy-map policy1class class1bandwidth 1000random-detectRelated Commands
IP Precedence
random-detect (per VC)
To enable per-virtual circuit (VC) Weighted Random Early Detection (WRED) or per-VC VIP-Distributed WRED (DWRED), use the random-detect VC submode command. To disable per-VC WRED and per-VC DWRED, use the no form of this command.
random-detect [attach group-name]
no random-detect [attach group-name]
Syntax Description
Defaults
WRED and DWRED are disabled by default.
Command Modes
VC submode
Command History
Usage Guidelines
WRED is a congestion avoidance mechanism that slows traffic by randomly dropping packets when congestion exists. DWRED is similar to WRED but uses the Versatile Interface Processor (VIP) instead of the Route Switch Processor (RSP). WRED and DWRED are most useful with protocols like TCP that respond to dropped packets by decreasing the transmission rate.
WRED and DWRED are configurable at the interface and per-VC levels. The VC-level WRED/DWRED configuration will override the interface-level configuration if WRED/DWRED is also configured at the interface level.
Use this command to configure a single ATM VC or a VC that is a member of a bundle.
Remember the following guidelines when using the random-detect (per VC) command:
•
•
When this command is used to configure an interface-level WRED/DWRED group to include per-VC WRED/DWRED as a drop policy, the configured WRED/DWRED group parameters are inherited under the following conditions:
•
•
When an interface-level WRED/DWRED group configuration is removed, per-VC WRED/DWRED parameters are removed from any VC that inherited them from the configured interface-level WRED/DWRED group.
When an interface-level WRED/DWRED group configuration is modified, per-VC WRED/DWRED parameters are modified accordingly if the WRED/DWRED parameters were inherited from the configured interface-level WRED/DWRED group configuration.
This command is only supported on interfaces that are capable of VC-level queueing. The only currently supported interface is the Enhanced ATM port adapter (PA-A3).
The DWRED feature is only supported on Cisco 7000 series routers with an RSP7000 card and Cisco 7500 series routers with a VIP2-40 or greater interface processor. A VIP2-50 interface processor is strongly recommended when the aggregate line rate of the port adapters on the VIP is greater than DS3. A VIP2-50 interface processor is required for OC-3 rates.
To use DWRED, distributed Cisco Express Forwarding (dCEF) switching must first be enabled on the interface. For more information on dCEF, refer to the Cisco IOS Switching Services Configuration Guide and the Cisco IOS Switching Services Command Reference.
Examples
The following example configures per-VC WRED for the permanent virtual circuit (PVC) called cisco. Because the attach keyword was not used, WRED uses default parameters.
pvc cisco 46random-detectThe following example creates a DWRED group called Rome and then applies the parameter group to an ATM PVC:
! The following commands create the DWRED parameter group Rome:random-detect-group Romeprecedence rsvp 46 50 10precedence 1 32 50 10precedence 2 34 50 10precedence 3 36 50 10precedence 4 38 50 10precedence 5 40 50 10precedence 6 42 50 10precedence 7 44 50 10exitexit! The following commands create a PVC on an ATM interface and then apply the! DWRED group Rome to that PVC:interface ATM2/0.23 point-to-pointip address 10.9.23.10 255.255.255.0no ip mroute-cachepvc vc1 201/201random-detect attach Romevbr-nrt 2000 1000 200encapsulation aal5snapThe following show queueing command displays the current settings for each of the IP Precedences following configuration of per-VC DWRED:
Router# show queueing random-detect interface atm2/0.23 vc 201/201random-detect group Rome:exponential weight 9class min-threshold max-threshold mark-probability----------------------------------------------------------0 30 50 1/101 32 50 1/102 34 50 1/103 36 50 1/104 38 50 1/105 40 50 1/106 42 50 1/107 44 50 1/10rsvp 46 50 1/10Related Commands
random-detect exponential-weighting-constant
To configure the Weighted Random Early Detection (WRED) and VIP-Distributed WRED (DWRED) exponential weight factor for the average queue size calculation for the queue, use the random-detect exponential-weighting-constant interface configuration command. To configure the exponential weight factor for the average queue size calculation for the queue reserved for a class, use the random-detect exponential-weighting-constant policy-map class configuration command. To return the value to the default, use the no form of this command.
random-detect exponential-weighting-constant exponent
no random-detect exponential-weighting-constant
Syntax Description
Defaults
The default exponential weight factor is 9.
Command Modes
Interface configuration when used on an interface.
Policy-map class configuration when used to specify class policy in a policy map.
Command History
11.1 CC
This command was introduced.
12.0(5)T
This command was made available as a policy-map class configuration command.
Usage Guidelines
WRED is a congestion avoidance mechanism that slows traffic by randomly dropping packets when congestion exists. DWRED is similar to WRED but uses the Versatile Interface Processor (VIP) instead of the Route Switch Processor (RSP). WRED and DWRED are most useful with protocols like TCP that respond to dropped packets by decreasing the transmission rate.
Use this command to change the exponent used in the average queue size calculation for the WRED and DWRED services. You can also use this command to configure the exponential weight factor for the average queue size calculation for the queue reserved for a class.
Note
The DWRED feature is not supported for class policy.
The DWRED feature is only supported on Cisco 7000 series routers with an RSP7000 card and Cisco 7500 series routers with a VIP2-40 or greater interface processor. A VIP2-50 interface processor is strongly recommended when the aggregate line rate of the port adapters on the VIP is greater than DS3. A VIP2-50 interface processor is required for OC-3 rates.
To use DWRED, distributed Cisco Express Forwarding (dCEF) switching must first be enabled on the interface. For more information on dCEF, refer to the Cisco IOS Switching Services Configuration Guide and the Cisco IOS Switching Services Command Reference.
Examples
The following example configures WRED on an interface with a weight factor of 10:
interface Hssi0/0/0description 45Mbps to R1ip address 10.200.14.250 255.255.255.252random-detectrandom-detect exponential-weighting-constant 10The following example configures the policy map policy1 to contain policy specification for the class called class1. During times of congestion, WRED packet drop is used instead of tail drop. The weight factor used for the average queue size calculation for the queue for class1 is 12.
! The following commands create the class map called class1:class-map class1match input-interface FE0/1! The following commands define policy1 to contain policy specification for class1:policy-map policy1class class1bandwidth 1000random-detectrandom-detect exponential-weighting-constant 12Related Commands
random-detect flow
To enable flow-based Weighted Random Early Detection (WRED), use the random-detect flow interface configuration command. To disable flow-based WRED, use the no form of this command.
random-detect flow
no random-detect flow
Syntax Description
This command has no arguments or keywords.
Defaults
Flow-based WRED is disabled by default.
Command Modes
Interface configuration
Command History
Usage Guidelines
You must use this command to enable flow-based WRED before you can use the random-detect flow average-depth-factor and random-detect flow count commands to further configure the parameters of flow-based WRED.
Before you can enable flow-based WRED, you must enable and configure WRED. For complete information, refer to the Cisco IOS Quality of Service Solutions Configuration Guide.
Examples
The following example enables flow-based WRED on the Serial1 interface:
interface Serial1random-detectrandom-detect flowRelated Commands
random-detect flow average-depth-factor
To set the multiplier to be used in determining the average depth factor for a flow when flow-based Weighted Random Early Detection (WRED) is enabled, use the random-detect flow average-depth-factor interface configuration command. To remove the current flow average depth factor value, use the no form of this command.
random-detect flow average-depth-factor scaling-factor
no random-detect flow average-depth-factor scaling-factor
Syntax Description
Defaults
The default average depth factor is 4.
Command Modes
Interface configuration
Command History
Usage Guidelines
Use this command to specify the scaling factor that flow-based WRED should use in scaling the number of buffers available per flow and in determining the number of packets allowed in the output queue for each active flow. This scaling factor is common to all flows. The outcome of the scaled number of buffers becomes the per-flow limit.
If this command is not used and flow-based WRED is enabled, the average depth scaling factor defaults to 4.
A flow is considered nonadaptive—that is, it takes up too much of the resources—when the average flow depth times the specified multiplier (scaling factor) is less than the depth for the flow, for example:
average-flow-depth * (scaling factor) < flow-depthBefore you use this command, you must use the random-detect flow command to enable flow-based WRED for the interface. To configure flow-based WRED, you may also use the random-detect flow count command.
Examples
The following example enables flow-based WRED on the Serial1 interface and sets the scaling factor for the average flow depth to 8:
interface Serial1random-detectrandom-detect flowrandom-detect flow average-depth-factor 8Related Commands
random-detect flow count
To set the flow count for flow-based Weighted Random Early Detection (WRED), use the random-detect flow count interface configuration command. To remove the current flow count value, use the no form of this command.
random-detect flow count number
no random-detect flow count number
Syntax Description
Defaults
256
Command Modes
Interface configuration
Command History
Usage Guidelines
Before you use this command, you must use the random-detect flow command to enable flow-based WRED for the interface.
Examples
The following example enables flow-based WRED on the Serial1 interface and sets the flow threshold constant to 16:
interface Serial1random-detectrandom-detect flowrandom-detect flow count 16Related Commands
random-detect-group
To define the Weighted Random Early Detection (WRED) or VIP-Distributed WRED (DWRED) parameter group, use the random-detect group global configuration command. To delete the WRED/DWRED parameter group, use the no form of this command.
random-detect-group group-name
no random-detect-group group-name
Syntax Description
Defaults
No WRED/DWRED parameter group exists.
Command Modes
Global configuration
Command History
Usage Guidelines
WRED is a congestion avoidance mechanism that slows traffic by randomly dropping packets when there is congestion. DWRED is similar to WRED but uses the Versatile Interface Processor (VIP) instead of the Route Switch Processor (RSP). WRED and DWRED are most useful when the traffic uses protocols such as TCP, that respond to dropped packets by decreasing the transmission rate.
The router automatically determines parameters to use in the WRED calculations. If you want to change these parameters for a group, use the exponential-weighting-constant or precedence command.
Examples
The following example defines the WRED parameter group called sanjose:
random-detect-group sanjoseprecedence 0 32 256 100precedence 1 64 256 100precedence 2 96 256 100precedence 3 128 256 100precedence 4 160 256 100precedence 5 192 256 100precedence 6 224 256 100precedence 7 256 256 100Related Commands
random-detect precedence
To configure Weighted Random Early Detection (WRED) and VIP-Distributed WRED (DWRED) parameters for a particular IP Precedence, use the random-detect precedence interface configuration command. To configure WRED parameters for a particular IP Precedence for a class policy in a policy map, use the random-detect precedence policy-map class configuration command. To return the values to the default for the precedence, use the no form of this command.
random-detect precedence precedence min-threshold max-threshold mark-prob-denominator
no random-detect precedence precedence min-threshold max-threshold mark-prob-denominator
Syntax Description
precedence
IP Precedence number. The value range is 0 to 7 and RSVP. For Cisco 7000 series routers with an RSP7000 interface processor and Cisco 7500 series routers with a VIP2-40 interface processor (VIP2-50 interface processor strongly recommended), the precedence value range is 0 to 7 only; see Table 12.
min-threshold
Minimum threshold in number of packets. The value range of this argument is 1 to 4096. When the average queue length reaches the minimum threshold, WRED randomly drops some packets with the specified IP Precedence.
max-threshold
Maximum threshold in number of packets. The value range of this argument is the value of the min-threshold argument to 4096. When the average queue length exceeds the maximum threshold, WRED drops all packets with the specified IP Precedence.
mark-prob-denominator
Denominator for the fraction of packets dropped when the average queue depth is at the maximum threshold. For example, if the denominator is 512, one out of every 512 packets is dropped when the average queue is at the maximum threshold. The value range is 1 to 65536. The default is 10; one out of every ten packets is dropped at the maximum threshold.
Defaults
For all precedences, the mark-prob-denominator default is 10, and the max-threshold is based on the output buffering capacity and the transmission speed for the interface.
The default min-threshold depends on the precedence. The min-threshold for IP Precedence 0 corresponds to half of the max-threshold. The values for the remaining precedences fall between half the max-threshold and the max-threshold at evenly spaced intervals.
Table 12 lists the default minimum threshold value for each IP Precedence.
