Acquiring Content to Pre-Position

Now that you've learned how to set up a location tree, you can pre-position content to the CEs located within the tree, so that the tree can provide content to the clients requesting content from the network. The content acquisition model of content networking begins at the origin server containing the channel's content. You should choose the root CE for the channel as a CE in close proximity to the channel's origin server because, as you learned previously, the root CE acquires the content for the channel by unicast-pulling the content from the origin server.

The type of content determines the protocols that you can use to fetch the content, which are as follows:

• HTTP for acquiring content from a website

• FTP for files from an FTP server

• Windows Common Internet File System (CIFS)

• UNIX Network File System (NFS) for retrieving content from Windows- and UNIX-based file servers

• Microsoft Media Servics (MMS) or RTSP for acquiring streaming content for live or ondemand viewing.

Keep in mind that the protocols used to acquire the content into the ACNS network need not be the same as those used by clients for viewing or downloading the content. For example, the acquirer can use FTP to pull Apple QuickTime streaming files from an FTP server, whereas clients can use their Apple media player to view the content using RTSP and RTP.

You can configure content acquisition using either the CDM GUI or with external manifest files.

Configuring Acquisition Using Manifest Files

You should use manifest files if you have content developers in your organization who would like to specify what content to preposition to the ACNS network from their websites. These content developers are resources who do not have administrative access to the CDM GUI. They simply need to write the manifest file and let you know of the URL where they published it for you to specify when you configure the channel for their content.

The manifest file instructs the root CE for the channel to pull either individual items of content or to crawl through entire directories of content. You write manifest files in XML using tags specific to ACNS acquisition. Example 14-1 gives a sample manifest file for acquiring content in an ACNS network.

Example 14-1. Sample Manifest File

 <CdnManifest>   <server name="cisco">     <host name="http://www.cisco.com" />   </server>   <crawler     server="cisco"     start-url="tac/docs.html"     depth="10"     prefix="tac"   </crawler>   <item src="/books/2/203/1/html/2/sales/docs.pdf" />   </CdnManifest> 

You can use the following tags in your manifest file:

• <CdnManifest>...</CdnManifest> Indicates the beginning and end of the manifest file.

• <server>...</server> Enables you to identify a server for applying crawler tags to. This way, you do not need to write the entire URL for the server in every crawler job.

• <crawler.../> Crawls a directory starting at the location specified in start-url for a specific depth, within the given prefix. In Example 14-2, the acquirer will crawl the URL http://www.cisco.com/tac for a depth of 10, starting with http://www.cisco.com/tac/docs.html.

• <item>.../> Specifies the URL source of an individual item to pull.

Note

Refer to your ACNS documentation for the XML schema that outlines the entire manifest XML file specification.

Figure 14-10 shows how to assign the manifest file to a channel.

Figure 14-10. Assigning a Manifest File to a Channel

Configuring Acquisition Using the CDM GUI

The CDM GUI enables you, as an ACNS administrator, to configure content acquisition directly by creating Quick Crawl Filters, as Figure 14-11 illustrates.

Figure 14-11. Configuring Content Acquisition from the CDM GUI