-
Cisco ACNS Software Configuration Guide for Centrally Managed Deployments, Release 5.4
-
Index
-
Preface
-
Chapter 1: Understanding the ACNS 5.4 Network
-
Chapter 2: Getting Started
-
Chapter 3: Using the Content Distribution Manager GUI
-
Chapter 4: Setting up Content Request Routing in the ACNS Network
-
Chapter 5: Configuring the ACNS Network for Content Distribution
-
Chapter 6: Configuring the ACNS Network for Content Acquisition
-
Chapter 7: Creating and Managing Programs
-
Chapter 8: Configuring Caching Services
-
Chapter 9: Configuring Streaming Media Services
-
Chapter 10: Managing Caching Performance and Differentiated Services
-
Chapter 11: Viewing Content Replication Status
-
Chapter 12: Configuring Login Authentication, Configuration Authorization, and Accounting
-
Chapter 13: Working with Device Configurations
-
Chapter 14: Configuring Network Interfaces
-
Chapter 15: Configuring Request Authentication and Authorization
-
Chapter 16: Configuring Request Processing Services
-
Chapter 17: Creating and Managing IP Access Control Lists
-
Chapter 18: Configuring Group Authorization Using an Access Control List
-
Chapter 19: Using the Transaction Logs
-
Chapter 20: Configuring Platform and System Settings
-
Chapter 21: Monitoring and Troubleshooting the ACNS Network
-
Chapter 22: Monitoring with SNMP
-
Appendix A: Creating Manifest Files
-
Appendix B: IP Multicast Addressing
-
Appendix C: Advanced Routing Configurations
-
Appendix D: Mapping ACNS Software CLI Commands to the Content Distribution Manager GUI
-
Table Of Contents
Configuring the ACNS Network for Content Distribution
Preparing for Channel Distribution of Content
Creating and Modifying Locations
Creating and Modifying Content Providers
Creating and Modifying Websites
Configuring Channels for Channel Distribution
Creating and Modifying Channels
Designating the Root Content Engine
Adding and Removing Content Engines from Channels
Adding and Removing Device Groups from Channels
Distributing Content Through Unicast Replication
Configuring Location Leader Preference and Forwarder Selection Probability
Configuring the Forwarder Lookup Level
Configuring the Content Metadata Poll Interval Multiplier
Configuring Unicast Content Distribution Through a NAT Firewall
Commands Related to Distribution Settings
Distributing Content Through Multicast Replication
About Configuring Carousel Passes
About Multicast Logging Enhancements
About the NACK Interval Multiplier
About the File Resend or Stop Feature
About Configuring Multicast Forward Error Correction
About Configuring the Multicast Sender Delay Interval
Configuring PGM and File Transmission Parameters Using Multicast Expert Mode
About Configuring Channel Options for Content Replication
Enabling Content Engines for Multicasting
Configuring Multicast Cloud Properties
Assigning Content Engines to a Multicast Cloud
Assigning and Removing Multicast Clouds from Channels
Verifying the Multicast Configuration
Commands Related to Multicasting
Troubleshooting the Distribution Hierarchy
Configuring the ACNS Network for Content Distribution
Once the hardware devices that make up your ACNS network are properly installed and configured, you are ready to use the information contained in this chapter to organize those devices into an ACNS network that is capable of serving pre-positioned content or live media to end users.
In the ACNS network, content is replicated through a channel distribution architecture. Content in channels can be transmitted by unicast pull or, if multicasting is enabled, by multicast push. This chapter provides information about configuring the network elements for channel distribution and multicasting. It contains the following sections:
•
Preparing for Channel Distribution of Content
•
•
•
•
Figure 5-1 shows the task flow for configuring channels for unicast and multicast content distribution.
Figure 5-1 Distributing Content in the ACNS Network
Preparing for Channel Distribution of Content
Channel distribution describes the manner in which content is distributed from the origin web server to all the Content Engines that are subscribed to the content or, more specifically, assigned to the channel that represents that group of content objects. (For an overview of the channel distribution architecture, see the "Content Acquisition and Distribution Architecture" section on page 1-10.)
In preparing for channel distribution of content, you should configure the following network elements in this order:
1.
2.
3.
4.
Creating and Modifying Locations
Locations are set up in the Content Distribution Manager GUI, and they organize and group Content Engines into virtual networks for distribution of content through channels.
Locations need to be configured before you can activate Content Engines and Content Routers and bring them online in the ACNS network. A default location is created automatically when you use the Content Distribution Manager Quick Start tool to activate your devices. If you wish to create a new location or modify the default location, use the procedure in this section.
Note
To create a new location, follow these steps:
Step 1
Figure 5-2 Locations Window
Step 2
Figure 5-3 Creating New Location Window
Step 3
Step 4
A location with no parent location assigned is a level 1 location. A location with a level 1 parent becomes a level 2 location, and so forth. The location level is displayed after you choose a parent location (or choose None) and click Submit to save the configuration.
Step 5
Step 6
To modify an existing location, follow these steps:
Step 1
Step 2
Step 3
Step 4
Configuring Channel Elements
This section describes how to create and modify content providers and websites. If you wish to associate a company name with its website, you must create a content provider directory listing. Websites must be set up in your ACNS network before you can create channels.
Creating and Modifying Content Providers
When you are creating a channel, you must choose the name of a content provider from the Content Provider drop-down list, or you can choose None. To populate this list, you must create a directory listing with the name, address, and contact information of a content provider. Content provider listings are created in the Content Distribution Manager GUI.
To create a content provider listing, follow these steps:
Step 1
Step 2
Step 3
Step 4
To modify a content provider listing, follow these steps:
Step 1
Figure 5-4 Modifying Content Provider Window
Step 2
Note
Step 3
Deleting Content Providers
You can delete content providers as needed.
Note
To delete a content provider, follow these steps:
Step 1
Step 2
Step 3
Creating and Modifying Websites
Websites must be defined in your ACNS network before you can create channels. Channels then map the website content to the Content Engines in your ACNS network. The Content Distribution Manager GUI allows you to define websites in your ACNS network. A website listing must provide the URL for the origin web server that stores the website content.
To create a website, follow these steps:
Step 1
Step 2
Figure 5-5 Creating New Web Site Window
Step 3
Table 5-1 Website Properties
Name*1
Unique name for the website.
Content Provider*
Name of the company that is providing the content. The default is None.
Note
Origin Server*
Fully qualified domain name of the origin server where the content resides.
Request Routed FQDN
Fully qualified domain name to route requests when using a Content Router.
WMT Authentication Type
Basic authentication or NTLM authentication.
Note
Pure DNS Routing
When checked, uses only the DNS server specified in the incoming packets for DNS requests. This causes the DNS caching name service to determine the original destination IP address.
When unchecked (the default), allows the DNS caching name service to use the servers configured on the Content Engine and allows WCCP interception of DNS requests.
Ignore case of URLs during playback
When checked, allows case-insensitive URL matching during the playback of pre-positioned content.
1 * = required field. All other fields are optional.
Step 4
Note
Step 5
Step 6
Step 7
To modify a website, follow these steps:
Step 1
Step 2
Step 3
Step 4
Deleting Websites
Note
To delete a website, follow these steps:
Step 1
Step 2
Step 3
Configuring Channels for Channel Distribution
Channel distribution describes the manner in which content is distributed from the origin web server to all the Content Engines that are subscribed to the content, or more specifically, assigned to the channel that represents that group of content objects. (For an overview of the channel distribution architecture, see the "Content Acquisition and Distribution Architecture" section on page 1-10.)
Creating and Modifying Channels
Channels map the content from a website to the devices in your ACNS network. When you create a channel, you first create a directory of your content providers and then provide URLs to their websites. (See the "Creating and Modifying Content Providers" section and the "Creating and Modifying Websites" section.)
Using the Content Distribution Manager, you can create, modify, or remove channels from the ACNS network. Keep in mind that modifying channels can affect the availability of content on the ACNS network.
You create a channel by defining the following items:
•
•
•
•
•
Creating a Channel
To create a channel, follow these steps:
Step 1
Step 2
Figure 5-6 Creating New Channel—Channel Properties
Step 3
Step 4
Table 5-2 Channel Properties
Channel information
Name*1
Unique name for the channel.
Request Routed FQDN
Fully qualified domain name to route requests. Read only.
This field is configured from the Creating New Web Site window (see the "Creating and Modifying Websites" section) and is used if you have a Content Router in your network.
Origin Server
Fully qualified domain name of the origin server. Read only.
This field is configured from the Creating New Web Site window. (See the "Creating and Modifying Websites" section.)
Content Provider*
Name of the content provider associated with this channel. The default is None.
To add a new content provider to this drop-down list, see the "Creating and Modifying Content Providers" section.
WebSite*
Name of the website belonging to the content provider that is associated with this channel.
Note
Live Channel
When checked, creates a live channel to distribute live or scheduled programs to the Content Engines associated with this channel. This channel does not have a related manifest file and cannot be used to distribute file-based content as regular channels do. The live channel learns about a live program through a program file that describes the attributes of the program.
Checking this check box disables the Channel Quota field and fields in the Acquisition and Distribution Properties section.
Channel Quota*
Maximum content storage size in megabytes for pre-positioning content for this channel.
See the "Channel Quota" section on page 6-26.
Acquisition and Distribution Properties
Distribution Priority*
Content distribution priority setting. Options are High, Normal, and Low. The default is Normal.
Multicast only, Multicast/Unicast, or Unicast only*
Enables multicast-only, multicast transmission with failover to unicast, or unicast-only transmission.
See the "About IP Multicasting" section.
Use null cipher for Distribution
When checked, disables encryption for distribution.
Root CE failover/fallback grace period*
Number of minutes before a root Content Engine failover or a temporary root Content Engine fallback occurs. The range is 20 to120 minutes. (See the "About the Root Content Engine Failover and Fallback Grace Period Setting" section.)
Never
When checked, Content Engine failover or fallback will never occur.
Use system-wide settings for QoS for unicast data
When checked, applies the system-wide QoS settings for unicast data to the channel. (See the "Configuring System-Wide Distribution QoS Settings Based on Channel Priority" section on page 20-16.)
To override the system-wide QoS settings with channel-specific QoS values, leave this check box unchecked, and configure the channel-specific QoS values in the QoS value for unicast data field.
QoS value for unicast data*
Configures a Differentiated Services Code Point (DSCP) value for the QoS. (See Table 5-3.)
If you choose Other, enter a decimal value in the corresponding field.
Note
1 * = required field. All other fields are optional.
Step 5
Step 6
Table 5-3 lists and describes the DSCP values for unicast QoS.
About the Root Content Engine Failover and Fallback Grace Period Setting
Cisco ACNS software supports failover of the root Content Engine to another Content Engine in your network when the root Content Engine is down or not functioning.
The secondary Content Engine polls the root Content Engine at preset intervals to make sure that the root Content Engine is functioning properly. If the polling of the root Content Engine does not return a positive response, the secondary Content Engine waits for the grace period and then takes over as the root Content Engine until the end of the grace period. The time interval before a Content Engine takes over as the root Content Engine is set by using the Root CE failover/fallback grace period option in the Create a New Channel window. You can choose a time interval between 20 and 120 minutes using increments of 10.
The acting root Content Engine keeps polling the original root Content Engine to see if the root Content Engine is back online. When the acting root Content Engine obtains a positive response from the original root Content Engine, the acting root Content Engine waits for the fallback grace period and then goes back to its original mode if the positive responses are consistent. When the acting root Content Engine goes back to its original mode, the original root Content Engine takes over the duties of the root Content Engine. Both failover and fallback grace period values are configured in the same field, and use the same value.
Note
Modifying a Channel
To modify a channel, follow these steps:
Step 1
Step 2
Figure 5-7 Modifying Channel Window
Step 3
To modify the Content Provider or WebSite fields, click the Edit Content Provider or Edit Website button. (See the "Configuring Channel Elements" section.)
Step 4
After you create a new channel, you need to designate a root Content Engine for the channel and then assign Content Engines or device groups to the channel. Although designating a root Content Engine and assigning devices to the channel can be done from the same GUI window, the following sections break out these steps into separate procedures.
Designating the Root Content Engine
The root Content Engine is the one Content Engine that is authorized to go directly to the origin web server for content. The root Content Engine then publishes the content to other Content Engines in the channel. You must designate a root Content Engine for content distribution to take place.
To designate one Content Engine to be the root Content Engine for a channel, follow these steps:
Step 1
Step 2
Step 3
•
•
Step 4
Note
Step 5
Adding and Removing Content Engines from Channels
To add or remove Content Engines from a channel, follow these steps:
Step 1
Step 2
Step 3
This window displays a list of the Content Engines in all locations to which they have been assigned, as well as the unreserved and total disk space in the cdnfs partition for each Content Engine. Content Engines can be added to the channel from more than one location.
Step 4
Step 5
Step 6
Step 7
If the Content Engine that you removed was the root Content Engine, and if there is at least one Content Engine still assigned to the channel, you must designate a new root Content Engine. (See the "Designating the Root Content Engine" section.)
Adding and Removing Device Groups from Channels
Device groups are assigned to channels through the Content Distribution Manager GUI. Whenever a channel is created and additional device groups are added, or when a channel assignment to the device group changes, devices in the group are notified of their assignment to the associated channel.
The relationship between device groups and channels is many-to-many. A channel can have multiple device groups and device groups, in turn, can belong to multiple channels.
To add a device group from a channel, follow these steps.
Step 1
Step 2
Step 3
Step 4
Step 5
Note
To remove a device group from a channel, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
Deleting Channels
To delete a channel, follow these steps:
Step 1
Step 2
Step 3
Step 4
Viewing Channels
To view all channels across your ACNS network, follow these steps:
Step 1
The Channels window appears, listing all the channels in your ACNS network.
The channel type and the website that is associated with each channel are displayed in the row for each channel.
Step 2
Step 3
Viewing All Content Engines Assigned to a Channel
In the Content Distribution Manager GUI, a distinction is made between devices that are assigned directly to a channel or devices that are assigned through a device group. The assignments are tracked separately. For tracking purposes, the View all Content Engines assigned to Channel window displays all Content Engines assigned to a channel, regardless of whether they are assigned directly or through device groups.
To view the list of all Content Engines assigned to a channel, follow these steps:
Step 1
Step 2
Step 3
.
Distributing Content Through Unicast Replication
Once content is acquired from the origin server by the root Content Engine of a channel, it can be replicated through the channel either by unicast or multicast transmission. The basic channel distribution architecture provides for unicast content replication using a hop-by-hop, store-and-forward methodology with the forwarder Content Engines systematically selected on the basis of the manually configured location hierarchy.
To distribute content through unicast, the ACNS network automatically creates a unidirectional distribution tree for each channel. The root node of the tree is the root Content Engine of the channel, and each Content Engine subscribed to the channel is a node on the tree.
For each node, its parent node is also called its forwarder Content Engine. The algorithm for automatically designating the forwarder Content Engine is called the channel routing algorithm.
Three general rules in the current channel routing algorithm are as follows:
1.
The location leader is computed automatically by the channel routing algorithm. ACNS 5.4 software enables you to influence which Content Engines are chosen as location leaders by setting a priority level for individual Content Engines in the location. Content Engines are all set to the lowest priority by default. The Content Engine with the highest priority is more likely to be chosen as the location leader. (See the next section, "Configuring Location Leader Preference and Forwarder Selection Probability".)
Use the show distribution channel EXEC command to see which Content Engine is the current forwarder for a channel. The reason/status field in the command output shows why a Content Engine is unable find a forwarder. Use the show distribution forwarder-list EXEC command to see the forwarder selection order of a Content Engine for a channel.
2.
ACNS 5.4 software allows you to set the number of levels that a location leader can search to find a forwarder. (See the "Configuring the Forwarder Lookup Level" section.)
You can also influence the selection of a forwarder by configuring the Content Engine with a distribution weight, by setting it to be used only as a last resort, or by taking it out of the forwarder list completely. The Content Engine configured with the heaviest weight has the highest probability of becoming the forwarder. (See the next section, "Configuring Location Leader Preference and Forwarder Selection Probability".)
3.
When multicast transmission is desirable, channels and Content Engines can be enabled for multicasting.
(See also the "Configuring Weighted Load Balancing for Live Stream Splitting" section on page 9-35.)
Configuring Location Leader Preference and Forwarder Selection Probability
To configure the location leader preference and forwarder probability settings, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
If you choose Weighted, click and slide the control button on the scale to the weight (1-20) that you want. The Content Engine with the greatest weight is most likely to be chosen as the forwarder.
Step 6
Configuring the Forwarder Lookup Level
To configure the forwarder lookup level, follow these steps:
Step 1
Step 2
Step 3
Figure 5-8 Distribution Settings
Step 4
Step 5
Configuring the Content Metadata Poll Interval Multiplier
Sender and receiver Content Engines exchange information before they transmit the actual content. This information is called metadata. Included in the metadata is the wait time between content updates (or the metadata polling interval), which the sender Content Engine sends to the receiver. The sender calculates this wait time based on a number of factors, such as the number of devices in the ACNS network or the number of receivers in the forwarder's queue. You can lengthen or shorten the wait time by configuring the content metadata poll interval multiplier on receiver Content Engines. This multiplier enables you to control how often receiver Content Engines notice changes in content. The polling interval that is calculated by the sender Content Engine will be either shorter or longer by a factor of the content metadata poll interval multiplier that you configure.
When the receiver Content Engine obtains the wait time from the sender, it multiplies the wait time by the value you specify in the multiplier. You can scale the unicast receiver Content Engine's next metadata poll interval down to 10 percent of normal or up to 10 times normal. For example, if you set the multiplier to 10 percent of normal, and the wait time between update polls is 120 seconds, the effective polling interval is shortened to 12 seconds (10 percent of normal or 0.1 x 120 is 12 seconds).
The higher the multiplier, the longer it takes for the receiver to be notified of changes in content and the less the network traffic that results from polling. On the other hand, the lower the multiplier, the less time it takes for the receiver to be notified of content changes, and the greater the network traffic that results from polling.
Note
To configure the content metadata poll interval, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
Configuring Unicast Content Distribution Through a NAT Firewall
ACNS 5.4 software supports communication between two Content Engines that are located behind different NAT devices. Content Engines can be configured for unicast content distribution with a NAT address and port, as well as with a primary IP address and port.
For example, if a forwarder Content Engine is behind a NAT device, you can set its NAT address on the Content Distribution Manager GUI, and the Content Distribution Manager will propagate the forwarder Content Engine NAT address to all the devices in the ACNS network so that the receiver Content Engine can set up a connection with the forwarder using the NAT address.
NAT address configuration also supports the following types of communications between Content Engines and other devices in a centrally managed network:
•
•
•
•
Note
NAT configuration logic works in the following manner: A receiver Content Engine first attempts to contact a forwarder Content Engine using the Content Engine primary IP address. If the receiver cannot contact the forwarder Content Engine by using the primary IP address, the receiver attempts to contact the forwarder Content Engine using the Content Engine NAT IP address, if a NAT IP address has been configured. If the receiver successfully initiates communication with the forwarder using the NAT IP address, the receiver will retry the primary IP address periodically and fall back to using the primary IP address, if it becomes operational.
To configure a NAT IP address for a Content Engine, follow these steps:
Step 1
Step 2
In the Contents pane, chose Device Activation. The Device Activation window appears with fields for editing the properties of the selected Content Engine.
Step 3
•
•
•
Note
Step 4
Commands Related to Distribution Settings
Table 5-5 lists and describes CLI commands that are related to distribution settings. For complete descriptions of these commands, including usage guidelines and examples, refer to the Cisco ACNS Software Command Reference, Release 5.4 publication.
Distributing Content Through Multicast Replication
Multicasting allows efficient distribution of content to multiple Content Engines and is useful when many end users are interested in the same content. ACNS software supports Pragmatic General Multicast (PGM)-based multicast replication using either satellite or multicast-enabled terrestrial infrastructures.
To prepare for multicast transmission of content, you should perform these tasks in the order presented in this section:
1.
2.
3.
4.
About IP Multicasting
Multicast delivery enables the distribution of streaming media by allowing different receiving devices on the IP multicast to receive a single stream of media content from the Content Engine simultaneously. This type of delivery can save significant network bandwidth consumption because a single stream is sent to many devices, rather than sending a single stream to a single device every time that this stream is requested.
This multicast delivery feature is enabled by setting up a multicast address on the Content Engine to which different devices, configured to receive content from the same channel, can subscribe. The delivering device sends content to the multicast address set up at the Content Engine, from which it becomes available to all subscribed receiving devices.
To take advantage of multicasting, all devices, including Content Engines and routers, as well as clients, must be multicast-enabled. For this reason, multicasting is mostly used in local networks where routers can be configured for multicasting. Multicast delivery over the Internet can only be accomplished when all devices that participate in the multicast have been enabled for multicasting.
About Multicast Clouds
For multicast content replication, Content Engines are grouped into multicast clouds. A multicast cloud consists of one sender Content Engine, an optional backup sender Content Engine, and at least one receiver Content Engine in a hub and spoke topology. All Content Engines in one multicast cloud share a unique advertisement address that allows them to communicate multicast session information. The multicast cloud is then associated with one or more multicast-enabled channels.
In pull-based unicast content distribution, a unicast receiver pulls file data out of the proper forwarder (or root Content Engine) when a client requests the content. By contrast, in multicast content distribution, the sender Content Engine in a multicast cloud proactively pushes content into the cloud according to a preconfigured schedule.
The receiver Content Engines listen on the advertisement IP address for information on content to be replicated from the sender, and then the receiver Content Engines decide whether or not to accept an advertisement and receive the corresponding content.
Content metadata (machine-readable information that describes the characteristics of the content) must be distributed to a receiver first before the content itself can be replicated. Content metadata helps to define what content to retrieve, how content will be retrieved, how recently content has been updated, how content is to be pre-positioned (for example, expiration time), and so forth. Metadata is always distributed using unicast. Content, however, can be replicated using either multicast or unicast. A multicast receiver rejects the multicast sender's advertisement of a file if the proper content metadata has not yet arrived.
The Content Distribution Manager GUI allows you to configure multicast cloud parameters in the Creating New Multicast Cloud window. These parameters are discussed in the following sections.
Figure 5-9 shows the task flow for configuring a multicast cloud.
Figure 5-9 Multicast Cloud Configuration Task Flow
About Configuring Carousel Passes
The carousel pass setting defines the maximum number of times that the sender Content Engine can retransmit the multicast for missing files. The primary sender sends the first carousel pass automatically. After the first round, multicast receiver Content Engines request missing content by sending a negative acknowledgment (NACK) to the sender that identifies the missing content. The multicast sender sends out the requested content when it receives the NACK from the receiver. After all receiver Content Engines have received all the multicast content or the sender has reached the maximum number of carousel passes, whichever comes first, the sender stops transmitting content.
If the primary sender fails and the backup sender becomes active, the backup sender takes charge of NACK processing. The backup sender's carousel passes are always triggered by a NACK. When the maximum number of carousel passes is reached for a file on the current active sender, you can configure file distribution to fall back to unicast, if this is desired. (See the "About Configuring Channel Options for Content Replication" section for more information.)
To adjust the pacing of the multicast transmission, you can specify how much time must elapse before missing files are resent (the Delay between passes field) in the Content Distribution Manager GUI Modifying Multicast Cloud window (Devices > Multicast Clouds > Modifying Multicast Cloud).
About Multicast Logging Enhancements
The ACNS 5.3.5 software release includes enhanced multicast logging to identify the receiver Content Engine that is sending the retransmission requests (NACKs) and to identify why a file is scheduled for multicasting. These logging enhancements provide the following details:
•
•
•
About the NACK Interval Multiplier
To identify missing content and trigger a resend of a file, receiver Content Engines send a negative acknowledgement (NACK) message to the sender Content Engine. NACK messages generated by many receiver Content Engines could generate more traffic than the sender can handle. ACNS 5.x software allows you to adjust the average interval between NACKs by configuring a NACK interval multiplier for an individual receiver Content Engine. This value (an integer between 0.1 and 10) adjusts the default average NACK interval (the default is 20 minutes) by the value configured as the interval multiplier. For example, if you set the NACK interval multiplier to 3, the interval between NACKs becomes 20 minutes x 3, or 60 minutes.
This adjustment can be made as needed by choosing Devices > Devices > Prepositioning > Distribution in the Content Distribution Manager GUI.
To send an immediate NACK request rather than wait for the scheduled interval, you can issue the distribution multicast send-nack-now EXEC command on a multicast receiver Content Engine.
To configure the NACK interval multiplier, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
About the File Resend or Stop Feature
To trigger resending of a particular file, or the content in a particular channel or all channels, you can issue the distribution multicast resend EXEC CLI command on a multicast sender. This command internally resets the number of carousel passes completed to zero for the desired target. Because the first resend of content always happens automatically without being triggered by a NACK, setting the carousel pass count to zero triggers a resend of the content. However, if the on-demand-only option is additionally specified, the first carousel pass made after the command has been issued can only be triggered by a NACK for the content involved.
To stop sending a particular file, a particular channel, or all channels, you can issue the distribution multicast stop EXEC command on a multicast sender. This command internally increases the number of carousels completed to the maximum integer, preventing further carousel passes from being triggered.
About Configuring Multicast Forward Error Correction
Forward error correction (FEC) is a type of data encoding that protects transmissions against errors, without requiring retransmission. The FEC number denotes the number of packets that will be encoded into one FEC transmission group. When the FEC number goes up, the transmission group becomes larger, so the multicast may be more error-resistant. However, there will be more computational overhead on the multicast sender and receivers. Note that there is no bandwidth overhead related to FEC.
In ACNS 5.0.1 software, the FEC default value is 8, whereas in ACNS 5.0.3 software and later, the FEC default is 16. If the multicast sender device is a high-end Content Engine model such as a CE-7325, you can set this number higher in order to improve multicast reliability, especially when your network connectivity has a high uniform loss rate. However, we do not recommend that you set this number beyond 64 because it may place too much of a load on all the receiver Content Engines.
Separately, you can also set proactive FEC using the PGM configuration file (a text file accessible from the Content Engine CLI in the /local1/multicast-expert-config/ directory). (See the "Configuring PGM and File Transmission Parameters Using Multicast Expert Mode" section.) Proactive FEC is the number of extra packets that the multicast sender proactively sends out for every FEC number of data packets. The proactive FEC default value is 0. You can set it higher for better multicast reliability, for example, 2 proactive packets for every 16 FEC packets, at the expense of 12.5 percent traffic overhead (2 divided by 16).
Proactive FEC is an additional reliability measure above and beyond that of normal FEC. Although normal FEC does not incur bandwidth overhead, proactive FEC does. Proactive FEC primarily protects the multicast from uniform losses. For example, if the network has a uniform loss rate of 15 percent, then a proactive FEC of 2 extra packets for every 16 FEC packets (a 12.5 percent bandwidth overhead) cuts the effective loss rate down to 2.5 percent. Most network losses are not completely uniform. Still, during bursts, proactive FEC similarly undercuts the effective burst loss rate. For example, if the burst loss rate is 20 percent while the average loss rate is 2 percent, with proactive FEC at 12.5 percent, the receiver Content Engines experience a burst loss rate of 7.5 percent and an average loss rate near 0 percent.
About Configuring the Multicast Sender Delay Interval
The multicast sender delay interval is the amount of time before each multicast transmission begins. A period of delay before the actual multicast transmission begins is required to allow content metadata time to propagate to the receiver Content Engine. Metadata contains the content file and configuration information that is necessary for the successful transmission of content files. The default sender delay interval is 16 minutes. The sender delay parameter is used to configure an extra delay before a multicast transmission can begin.
When configuring the sender delay interval, you must take into account that the content metadata must first be propagated to the receiver before the multicast transmission can commence. During a multicast session, a receiver Content Engine sends out periodic requests for files that it has not yet received. The sender retransmits files only as requested by the receiver Content Engine. A multicast receiver will reject a multicast sender's advertisement of a file if the associated content metadata has not arrived yet. The sender delay option allows you to configure enough time for the metadata to propagate to the receiver and avoid having the receiver reject the multicast sender's advertisement of a file.
To configure the sender delay interval, use the multicast sender-delay global configuration command on a sender Content Engine. You can configure the duration of the delay based on your expectation or best guess of the amount of time required for metadata to be propagated. The system takes this user-configured sender delay value and delays the multicast transmission by such a period.
Note
Configuring PGM and File Transmission Parameters Using Multicast Expert Mode
PGM is a reliable multicast protocol defined in IETF RFC 3208. It is designed for applications that require ordered or unordered, duplicate-free, multicast data delivery from multiple sources to multiple receivers. Support for reliable multicasting and file transmission is provided by the TIBCOSmartPGM FX tool set, which is integrated with Cisco ACNS software, beginning with the ACNS 5.2 relelase. You can configure some PGM and file transmission (FX) parameters through the Content Distribution Manager GUI, such as the advertisement IP address, multicast-out bandwidth, TTL, FEC transmission group, and so forth.
In some cases, expert users might want to change other PGM and FX parameters to make the multicast file transfer more robust and efficient for their multicast environment. ACNS software allows you to change the configuration parameters of the TIBCOSmartPGM FX configuration file manually by using multicast expert mode on the Content Engine.
Caution
ACNS software contains default TIBCOSmartPGM FX configuration files for multicast sender and receiver Content Engines. The multicast sender and receiver Content Engines determine the medium being used for the multicast by checking the configuration of the multicast cloud, and they then read the configuration parameter values from the PGM configuration file that corresponds to the medium (terrestrial or satellite) being used for the multicast.
ACNS 5.4 software uses the following default TIBCOSmartPGM FX configuration files:
•
•
The Content Engine stores sample copies of the default TIBCOSmartPGM FX configuration files in the /local1/multicast-expert-config/ directory for reference. You can modify one of these sample configuration files, rename it, and save it back to the multicast sender or receiver Content Engine. When you create customized configuration files and copy them to the /local1/multicast-expert-config/ directory using the default configuration filenames, the Content Engine uses the customized configuration file instead of the default configuration file. The customized configuration file becomes effective after the Content Engine is restarted.
Content Engines running ACNS 5.4 software contain the following sample configuration files:
•
•
•
•
To change the configuration parameters of a default TIBCOSmartPGM FX configuration file, follow these steps:
Step 1
a.
b.
c.
d.
Step 2
ftp> cd multicast-expert-config250 CWD command is successful.ftp>Step 3
ftp> ls200 PORT command successful.150 Opening ASCII mode data connection for file list.fxdSatellite.conf.rcv.samplefxdSatellite.conf.src.samplefxdTerra.conf.rcv.samplefxdTerra.conf.src.sample226 Transfer complete.ftp: 297 bytes received in 0.01Seconds 29.70Kbytes/sec.Determine which sample file you want to retrieve, based on whether you are configuring the multicast sender or the receiver and whether your network is using terrestrial or satellite media transmission.
Step 4
ftp> bin200 Type set to I.Step 5
ftp> get fxdSatellite.conf.rcv.sample200 PORT command successful.Opening BINARY mode data connection for fxdSatellite.conf.rcv.sample (5607 bytes).226 Transfer complete.ftp: 5607 bytes received in 0.00Seconds 5607000.00Kbytes/sec.ftp>Step 6
ftp> quitStep 7
Step 8
Step 9
Step 10
a.
b.
c.
d.
C:\>ftp 128.19.220.79Connected to 128.19.220.79.220 CONTENTENGING FTP server (Version wu-2.7.0(2) Tue Sep 7 17:20:20 PDT 2004) ready.User (128.19.220.79:(none)):admin331 Password required for admin.Password:230 User admin logged in. Access restrictions apply.ftp> cd multicast-expert-config250 CWD command successful.ftp> bin200 Type set to I.ftp> put fxd.conf.rcv200 PORT command successful.150 Opening BINARY mode data connection for fxd.conf.src.226 Transfer complete.ftp:5607 bytes sent in 0.01Seconds 560.70Kbytes/sec.ftp> quitStep 11
About Configuring Channel Options for Content Replication
Once a multicast cloud is configured, it must be associated with one or more multicast-enabled channels for multicast replication of content to take place within the channel. The channel configuration offers various transmission options for replicating content; a channel can be configured for multicast and unicast (multicast with failover to unicast), multicast-only, or unicast-only transmission.
Multicast and Unicast
When a channel is configured for multicast and unicast, the receiver Content Engine uses unicast to download content only after all carousel passes have been exhausted and after the preconfigured multicast transmission fails. In a multicast cloud configuration that uses a backup sender, when the channel is enabled for multicast and unicast, the failover to unicast occurs when the current active multicast sender has exhausted all the carousel passes for the file.
If the administrator wants the Content Engines to fall back to unicast (for example, with a multitier unicast deployment using a terrestrial multicast medium), the multicast cloud should be configured for a low number of carousel passes (such as 1, 2, or 3).
Multicast Only
However, if only multicast replication is desired (for example, with a hub and spoke or star topology deployment using a satellite multicast medium), the channels should be configured as multicast-only, with a high number of carousel passes configured in the multicast cloud (such as 10 or more).
When a channel is configured to be multicast only (that is, when the channels are associated with a multicast cloud and the subscribing receiver Content Engine has multicast service successfully licensed and enabled), content replication takes place only through multicasting. No retransmission takes place in unicast at all. This prevents background unicast polls from happening and taking up bandwidth. However, if a Content Engine in the multicast-only channel is not licensed and enabled for multicasting, it can continue to request all the content from a multicast-only channel through unicasting.
Enabling Content Engines for Multicasting
Before you can create a multicast cloud, you must have a multicast distribution license key (purchased from Cisco Systems) and Content Engines that are enabled for multicasting. These multicast-enabled Content Engines can then be assigned as sender and receiver Content Engines when you configure the multicast cloud.
Content Engines for multicasting must be assigned to the multicast cloud, which is assigned to multicast-enabled channels. Also, you need to assign individual Content Engine senders and receivers of the cloud to the particular multicast-enabled channel. You must do this additional step even though the multicast cloud is associated with the channel.
Note
To enable Content Engines for multicasting, follow these steps:
Step 1
Step 2
Step 3
The Multicast Distribution Agreement for Content Engine window appears only if you have not read and accepted the multicast distribution license for the device. To accept the license, follow these steps:
a.
b.
c.
The window refreshes, and the Multicast Distribution License Settings for Content Engine window appears.
Step 4
Alternatively, check the Evaluate check box to enable an evaluation license key for a set period.
Step 5
Step 6
Step 7
•
•
•
When a device is associated with one or many device groups that have been configured with the settings displayed in this window, you can choose the device group name from the drop-down list whose settings you want to apply to the device.
Step 8
Configuring Multicast Cloud Properties
A multicast cloud is configured by specifying an IP multicast address for advertising the data being transferred, an IP multicast address range for transferring the data, a primary multicast sender Content Engine and an optional backup sender, a set of receiver Content Engines, and a maximum rate at which to send the data.
Note
To configure a multicast cloud, follow these steps:
Step 1
Step 2
Figure 5-10 Creating New Multicast Cloud Window
Step 3
a.
b.
When creating the multicast cloud, you must assign a unique advertisement address. Assigning a unique advertisement address provides all the Content Engines in one cloud with the same advertisement address and allows them to communicate multicast session information.
c.
Step 4
a.
Once the multicast cloud is configured and you associate the multicast cloud with one or more multicast-enabled channels, this multicast address range is used to provide each channel associated with it a unique data channel multicast address. When you assign a channel to the multicast cloud, an unused IP address is automatically selected from this range to ensure that the address is used by only one channel and by only one multicast cloud.
b.
Tip
c.
Step 5
a.
b.
The TTL field dictates how many hops a packet travels before it is discarded, regardless of whether or not the packet has reached its destination.
c.
This number of carousel passes is the maximum number of times that a multicast sender can send the content of the channel to which it is assigned. The multicast sender can be set to fall back to unicast after the maximum number of carousel passes has been reached.
If you want the Content Engines to fall back to unicast (for example, with a multitier unicast deployment using a terrestrial multicast medium), the multicast cloud should be configured for a low number of carousel passes (such as 1, 2, or 3).
However, if multicast replication is preferred (for example, with a hub and spoke or star topology deployment using a satellite multicast medium), use a high number of carousel passes, such as 10 or more.
Note
d.
FEC data encoding protects transmissions against errors, without requiring retransmission. The FEC number denotes the number of packets that will be encoded into one FEC transmission group. The higher the number, the bigger the transmission group size, and hence, the more error-resistant the multicast may be, but also the more computational overhead there will be on the multicast sender and receivers. Note that there is no bandwidth overhead related to FEC.
e.
f.
This option specifies whether IP routers are to be used to assist in distribution of content.
g.
This is the amount of time allotted for the backup sender to detect whether the primary sender is active.
h.
This is the amount of time allotted for the primary sender to detect whether the backup is active.
Step 6
Step 7
Your multicast cloud configuration is saved, and you can proceed to Step 3 in the "Assigning Content Engines to a Multicast Cloud" section.
Note
Table 5-6 describes the parameters that you need to define in the Content Distribution Manager GUI when configuring a multicast cloud. Required fields are marked with an asterisk in the GUI and in the table.
Table 5-6 Multicast Cloud Parameters
Multicast Cloud Information
Name*1
Identifier for the multicast cloud. The name must be unique across the system.
Advertisement IP address*
Multicast address collectively known and used by a multicast cloud to implement a communication channel.
The advertisement IP address must conform to these guidelines:
•
•
•
•
•
•
•
See Appendix B, "IP Multicast Addressing," for information on multicast address restrictions.
Port*
Port used for file addresses. The default port is 7777.
Multicast Address Settings
Start IP address*
Start of the IP address range, which must be within the range 224.0.0.0 to 239.255.255.255.
The IP address range must conform to the following:
•
•
Note
End IP address*
End of the IP address range, which must be higher than the start IP address. The end IP address must be within the range 224.0.0.0 to 239.255.255.255.
See Appendix B, "IP Multicast Addressing," for information on multicast address restrictions.
Default Multicast-out Bandwidth*
Maximum multicast rate in kilobits per second (kbps). The minimum rate is 10 kbps.
This is the default multicast-out bandwidth, which is applied 24 hours a day, 7 days a week. The multicast-out bandwidth is associated with the sender Content Engine only.
To customize bandwidth rates for different days, use the A&D Bandwidth Settings window. These settings override the default for the period that you specify. (See the "Configuring Acquisition and Distribution Default Bandwidth Settings" section on page 6-56.)
Synchronize Primary and Backup CE Multicast-out Bandwidth
When checked, default multicast-out bandwidth settings are used by both primary and backup senders, if a backup sender is configured.
Backup CE Default Multicast-out Bandwidth
Maximum multicast rate in kilobits per second (kbps) for the backup sender. The minimum rate is 10 kbps.
Advanced Settings
Multicast medium
Means of transmitting the multicast, either satellite or terrestrial. The default is satellite.
TTL
Time To Live expressed as the number of hops (1-255). The default is 255.
FEC transmission group
Size of an FEC (forward error correction) block in packets. (Refer to RFC 3208 PGM Reliable Transport Protocol Specification for more information.) The allowable inputs are 2, 4, 8, 16, 32, 64, and 128. The default is 16.
# of Carousel passes
Maximum number of times (1-1000000000) that a multicast sender will attempt to send missing content for on-demand carousels. The default is 5.
Note
Delay between passes
Delay between file transmissions, in minutes (0-10080) (7 days * 24 hours * 60 minutes). The default is 0 minutes.
PGM Router-assist
Specifies whether IP routers are to be used to assist in distribution of content. Checking this check box enables the IP router alert option for PGM packets.
Note
Primary-to-backup failover grace period
Amount of time allotted for the backup sender to detect whether the primary sender is active. If the backup sender does not hear a heartbeat from the primary sender within this grace period, the backup sender assumes the active role.
The range is 5-7200 minutes. The default is 30 minutes.
Backup-to-primary fallback grace period
Amount of time allotted for the primary sender to detect whether the backup is active. If the backup sender does not respond within this grace period, the primary sender assumes the active role.
The range is 5-7200 minutes. The default is 30 minutes.
Comments
Comments about the multicast cloud.
1 * = required field. All other fields are optional.
Modifying a Multicast Cloud
To modify a multicast cloud, follow these steps:
Step 1
Step 2
Step 3
Step 4
Deleting Multicast Clouds
To delete a multicast cloud, follow these steps:
Step 1
Step 2
Step 3
Step 4
Viewing Multicast Clouds
To view a list of all the multicast clouds configured in your network, follow these steps:
Step 1
The table also displays the sender Content Engine for each multicast cloud.
Step 2
Assigning Content Engines to a Multicast Cloud
Note
To add Content Engines to a multicast cloud, follow these steps:
Step 1
Step 2
Step 3
Step 4
a.
Alternatively, click the Assign all Content Engines icon in the taskbar to perform a bulk addition of all Content Engines in various locations to the multicast cloud and click Submit.
b.
c.
Note
To remove a Content Engine from the multicast cloud, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
Assigning and Removing Multicast Clouds from Channels
You can assign or remove multicast clouds only from channels that are multicast-enabled. A channel is multicast-enabled when you click the Multicast/Unicast radio button or the Multicast-only radio button when you create or modify the channel. (To create a new channel, see the "Creating and Modifying Channels" section.)
Note
To assign a multicast cloud from a multicast-enabled channel, follow these steps:
Step 1
Step 2
Step 3
a.
b.
Step 4
Figure 5-11 Assigning Multicast Clouds to Channel Window
Step 5
Step 6
In Figure 5-12 a multicast cloud has already been chosen and the window has refreshed, showing the IP address range for the cloud.
Figure 5-12 Assign Multicast Cloud to Channel Assignment Window
Step 7
To remove a multicast cloud from a channel, follow these steps:
Step 1
Step 2
Step 3
Step 4
Step 5
Verifying the Multicast Configuration
Use the following checklist to verify your multicast configuration:
1.
2.
3.
4.
5.
6.
Commands Related to Multicasting
Table 5-8 lists and describes CLI commands that are related to multicasting. For complete descriptions of these commands, including usage guidelines and examples, refer to the Cisco ACNS Software Command Reference, Release 5.4 publication.
Troubleshooting the Distribution Hierarchy
Because distribution-related problems are design-dependant, your initial strategy is to discover whether or not the correct Content Engine is sending content in the correct distribution path.
•
ce1# show distribution remote traceroute ?forwarder-next-hop next forwarder along the pathunicast-sender check status for unicast senderce1# show distribution remote traceroute forwarder-next-hop ?channel-id Channel-id of a Channelce1# show distribution remote traceroute forwarder-next-hop channel-id 133 ?max-hop Trace route till specified number of hops is reachedtrace-till-good traceroute till probe is good or the object is foundtrace-till-root traceroute till the root cece1# show distribution remote traceroute forwarder-next-hop channel-id 133 trace-till-rootHop NextHop_CEId NextHop_CEName NextHop_CEIp GenID Status/Reason--- ------------ -------------- ------------ ----- -------------1 1100 ce3 10.255.0.43 1 LOC-LEAD1 1100 ce3 128.107.193.183 1 LOC-LEAD (Reached RootCE)•
ce1# show distribution remote traceroute unicast-sender channel-id 133 ?cdn-url check the object on remote CE using cdn-urlprobe probe the remote unicast senderrelative-cdn-url check the object on remote CE using relative-cdn-urlce1# show distribution remote traceroute unicast-sender channel-id 133 probe ?max-hop Max-hop to traceroute totrace-till-good traceroute till probe is good or the object is foundtrace-till-root traceroute till the root cece1# show distribution remote traceroute unicast-sender channel-id 133 probe trace-till-rootPolling .... ce3 [10.255.0.43] Fwdr_Id:1100Polling .... ce3 [128.107.193.183] Fwdr_Id:1100(Reached RootCE)Where to Go Next
In this chapter you have accomplished the following tasks:
•
•
•
•
•
•
Now you are ready to configure additional device settings for acquiring pre-positioned content and to define the content that you want to acquire. Proceed to the next chapter, Chapter 6, "Configuring the ACNS Network for Content Acquisition."
