 Feedback
|
Table Of Contents
Release Notes for Cisco Internet Streamer
CDS 3.1.2Enhancements in Release 3.1.2-b34
Enhancements in Release 3.1.2-b26
Enhancements in Release 3.1.2-b17
Enhancements in Release 3.1.2-b11
New Features in Releaes 3.1.2-b11
Open Caveats in Release 3.1.2-b34
Open Caveats in Release 3.1.2-b26
Open Caveats in Release 3.1.2-b17
Open Caveats in Release 3.1.2-b11
Resolved Caveats in Release 3.1.2-b34
Resolved Caveats in Release 3.1.2-b26
Resolved Caveats in Release 3.1.2-b17
Resolved Caveats in Release 3.1.2-b11
Content Abstraction Layer (CAL)
Downgrading from Release 3.1.2
Obtaining Documentation and Submitting a Service Request
Release Notes for Cisco Internet Streamer
CDS 3.1.2
These release notes cover Cisco Internet Streamer CDS Release 3.1.2-b34.
Revised: March 2013, OL-28262-03Contents
The following information is included in these release notes:
•
Enhancements in Release 3.1.2-b34
•
•
•
•
•
Note
Enhancements in Release 3.1.2-b34
There are no new enhancements in Release 3.1.2-b34.
Enhancements in Release 3.1.2-b26
There are no new enhancements in Release 3.1.2-b26.
Enhancements in Release 3.1.2-b17
There are no new enhancements in Release 3.1.2-b17.
Enhancements in Release 3.1.2-b11
The following enhancements have been added in Release 3.1.2:
•
•
•
•
New Features in Releaes 3.1.2-b11
The Web Engine supports Adobe Zeri content in cache mode. In cache mode, the SE gets the content fragments from the Origin server and streams them; similar to any other web objects that the Web Engine caches and delivers. The Web Engine does not support acting as the Origin server for Adobe Zeri files, nor is the Web Engine capable of generating content fragments from a large Zeri file.
Note
System Requirements
The Internet Streamer CDS runs on the CDE205, CDE220, and the CDE250 hardware models, as well as two UCS models.
Table 1 lists the different device modes for the Cisco Internet Streamer CDS software, and which CDEs support them.
The new CDE250 models (CDE250-2S8, CDE250-2S9, and CDE250-2S10) have four interfaces at 10 gigabit Ethernet speeds and four interfaces at gigabit Ethernet speeds (plus two additional gigabit ethernet interfaces for management).
The new CDE250 models only support the SE device mode and have the following storage capacities:
•
•
•
The Cisco UCS models (UCS C200 and UCS C210) and the Cisco Internet Streamer Release 3.1 software are sold separately and ship independently of each other.
CDE250-2S6 and CDE250-2M0 platforms have four interfaces at 10 gigabit Ethernet speeds and four interfaces at gigabit Ethernet speeds (plus two additional gigabit ethernet interfaces for management).
The CDE220-2S3i platform has a total of 14 gigabit Ethernet ports in this CDE. The first two ports (1/0 and 2/0) are management ports. The remaining 12 gigabit Ethernet ports can be configured as two port channels. See the Cisco Content Delivery Engine CDE205/220/250/420 Hardware Installation Guide for set up and installation procedures for the CDE220-2S3i and the Cisco Internet Streamer CDS 2.6 Software Configuration Guide for information on configuring the Multi Port Support feature.
The CDE220-2G2 platform has a total of ten gigabit Ethernet ports. The first two ports (1/0 and 2/0) are management ports. The remaining eight gigabit Ethernet ports can be configured as one port channel. See the Cisco Content Delivery Engine CDE205/220/250/420 Hardware Installation Guide for set-up and installation procedures for the CDE220-2G2.
The CDE205 can run as the CDSM, SR or SE. See the Cisco Content Delivery Engine CDE205/220/250/420 Hardware Installation Guide for set-up and installation procedures for the CDE205.
Note
Limitations and Restrictions
This release contains the following limitations and restrictions:
•
•
•
•
Note
The CDS does support clients that are behind a NAT device or firewall that have shared external IP addresses. In other words, there could be a firewall between the CDS network and the client device. However, the NAT device or firewall must support RTP/RTSP.
System Limits and Thresholds
This release has the following limits and thresholds:
•
•
•
•
Service Router Limits and Thresholds
The Service Router has memory-related limits and thresholds. Memory usage of the Service Router depends on the number of coverage zone entries, the number of Content Origin servers, the distribution of subnets in the Coverage Zone file, and the number of Service Engines in the CDS. From our tests using a sample Coverage Zone file, we have observed that we can support 20,000 Coverage Zone entries with 26 SEs, and 40 Content Origins servers.
Note
We recommend keeping the memory usage (both virtual and resident) below 1.5 GB.
Frequent configuration updates could cause memory fragmentation, which raises the memory usage.
Service Monitor Limits and Thresholds
When the Service Monitor thresholds are exceeded, an alarm is raised on the respective device and an SNMP trap is sent to the CDSM. The parameters monitored and thresholds for each component or protocol engine can be modified. The default thresholds are as outlined below.
Following are the parameters that are monitored on each device (SE, SR, and CDSM) and the default threshold setting of each parameter:
•
•
•
•
•
•
Following are the parameters that are monitored only on the SE, along with default threshold setting of each parameter:
•
•
•
•
•
•
•
Web Engine Limits and Thresholds
The Web Engine has the following limits and thresholds:
Memory Usage
In Release 2.5.9, the memory threshold on each SE is 3.2 GB. If the threshold is exceeded, the memory_exceeded alarm is raised and trickle mode is enabled. In Release 2.5.9, the admission control is based on 30,000 session and 3.2 GB of memory.
In Release 2.6.1, the memory threshold on each SE is 3.2 GB. If the threshold is exceeded, the memory_exceeded alarm is raised. In cases where the memory reaches 3.7 GB, trickle mode is enabled and eventually the Web Engine is restarted. The above memory values, and the 20,000-60,000 sessions and 100,000 open file/socket descriptor (FD) limit are used for admission control in Release 2.6.1.
Session Limits
Web Engine supports the following session-threshold limits:
•
•
The max_session_exceeded alarm is raised if the session-threshold limit is reached. If further requests are sent to the SE even when the session threshold is reached, the Web Engine attempts to process the requests but does not accept any more requests when the request count reaches 60,000 on a CDE250, and 20,000 on all other CDEs.
CAL Limits
Outstanding CAL Lookup threshold is 25,000 on the CDE250 and 15,000 on all other CDEs. The WebCalLookupThreshold alarm is raised on reaching this threshold limit.
Outstanding CAL disk Write threshold is 3,000 CAL requests (create, update, delete, popularity update) on the CDE250, and 1,500 on all other CDEs. The WebCalDiskWriteThreshold alarm is raised on reaching this threshold.
Other CAL thresholds are as follows:
•
•
•
Note
Web Engine thresholds are also applicable to adaptive bit rate (ABR) streaming.
CDSM Limits and Thresholds
The CDSM has the following limits and thresholds:
•
•
RPC Connections
A maximum of 40 RPC connections are supported among the managed devices (SE, SR, standby CDSM, and primary CDSM). The RPC connection maximum is defined in the httpd.conf.rpc configuration file located in the /state directory.
File Synchronization
The primary CDSM checks for file updates and synchronization with the managed devices (SE, SR, and standby CDSM) every ten minutes.
CDSM Availability (primary and standby)
The SE and SR check for the availability of the primary and standby CDSM on a regular interval; however, if the CDSM does not respond, the SE and SR use an exponential-backoff call for retrying the connection.
The exponential backoff call means that if the CDSM does respond to the first attempt, the SE or SR sleep for ten seconds before trying again. If the second attempt does not succeed, the wait time doubles (20 seconds), if that attempt does not succeed, the wait time doubles again (40 seconds). The wait time doubles every attempt (10, 20, 40, 80, and so on) until the maxWaitingTime of 320 seconds.
SE Configuration Change Synchronization
The period of time before the local configuration manager (LCM) on an SE sends a configuration change to the primary CDSM is a maximum of 2.25 times the polling rate. The polling rate is configurable through the CDSM GUI (System > Configuration > System Properties, System.datafeed.pollRate).
RTSP Gateway and Movie Streamer
The default RTSP Gateway transactions per second (tps) is 40. There are no other limits to the RTSP Gateway.
The Movie Streamer default maximum concurrent session is 200 and the default maximum bandwidth is 200 Mbps.
Windows Media Streaming
Windows Media Streaming has the following limits and thresholds:
•
Note
•
•
•
Flash Media Streaming
With the basic license, Flash Media Streaming the default maximum concurrent sessions is 200 and the default maximum bandwidth is 200 Mbps.
Buying more licenses can increase the concurrent sessions and maximum bandwidth as follows:
•
•
We recommend that the Flash Media Streaming process memory usage not exceed 3 GB resident set size (RSS). If the memory usage for Flash Media Streaming exceeds 3 GB RSS, a threshold exceeded alarm is raised.
Note
Important Notes
To maximize the content delivery performance of a CDE205, CDE220, or CDE250, we recommend you do the following:
1.
Configure interfaces on the quad-port gigabit Ethernet cards into a single port-bonding interface. Use this bonding channel, which provides instantaneous failover between ports, for all client-facing traffic. Use interfaces number 1 and 2 (the two on-board Ethernet ports) for intra-CDS traffic, such as management traffic, and configure these two interfaces either as standby or port-channel mode. Refer to the Cisco Internet Streamer CDS 2.6 Software Configuration Guide for detailed instruction.
2.
Assuming ether-channel (also known as port-channel) is used between the upstream router/switch and the SE for streaming real-time data, the ether-channel load balance algorithms on the upstream switch/router and the SE should be configured as "Src-ip" and "Destination IP" respectively. Using this configuration ensures session stickiness and general balanced load distribution based on clients' IP addresses. Also, distribute your client IP address space across multiple subnets so that the load balancing algorithm is effective in spreading the traffic among multiple ports.
Note
3.
The CDE205, or CDE220 performance has been optimized for HTTP and WMT bulk traffic, individually. While it is entirely workable to have mixed HTTP and WMT traffic flowing through a single server simultaneously, the aggregate performance may not be as optimal as the case where the two traffic types are separate, especially when the traffic volume is high. So, if you have enough client WMT traffic to saturate the full capacity of a server, we recommend that you provision a dedicated server to handle WMT; and likewise for HTTP. In such cases, we do not recommended that you mix the two traffic types on all CDE servers which could result in suboptimal aggregate performance and require more servers than usual.
4.
If your deployment must have Streamers handle HTTP and WMT traffic simultaneously, it is best that you configure the Streamer to limit each of its HTTP sessions below a certain bitrate (for example, 1Mbps, 5Mbps, or the typical speed of your client population). This prevents HTTP sessions from running at higher throughput than necessary, and disrupting the concurrent WMT streaming sessions on that Streamer. To turn on this pacing feature, use the HTTP bitrate field in the CDSM Delivery Service GUI page.
Please be aware of the side effects of using the following commands for Movie Streamer:
Config# movie-streamer advanced client idle-timeout <30-1800>Config# movie-streamer advanced client rtp-timeout <30-1800>These commands are only intended for performance testing when using certain testing tools that do not have full support of the RTCP receiver report. Setting these timeouts to high values causes inefficient tear down of client connections when the streaming sessions have ended.
For typical deployments, it is preferable to leave these parameters set to their defaults.
5.
Open Caveats
The open caveat section has the following subsections:
•
•
•
•
Open Caveats in Release 3.1.2-b34
There are no new open caveats for 3.1.2-b34.
Open Caveats in Release 3.1.2-b26
Release 3.1.2-b26 contains the following open caveats:
CDSM
•
Symptom:
When configuring a URL signature in GUI: Devices (choose a Service Engine) -> Service Control -> URL Signing, if the "Public Key URL" or "Private Key URL" is given an URL, which is invalid, the GUI accepts the configuration, but on the change, it doesn't reflect in the running-config in CLI on the SE side.
Conditions:
The "Public Key URL" or "Private Key URL" is given an URL which is invalid, either it doesn't exist or it's not valid key.
Workaround:
Specify valid "Public Key URL" and "Private Key URL".
•
Symptom:
In page Devices -> Application Control -> Windows Media Streaming -> General Settings ->Windows Media Cache Settings, change Maximum TTL and Minimum TTL. The new value of Maximum TTL may not take effect and "Cli command failed" appears in cms log.
Conditions:
This only happens when all following conditions are met:
1. Both Maximum TTL and Minimum TTL are modified at the same time.
2. Maximum TTL is changed to be lower than or equal to 60 minutes.
Note: 60 minutes is the default value of Minimum TTL.
Workaround:
Avoid modifying Maximum TTL and Minimum TTL at the same time.
•
Symptom:
Sometimes editing, Devices->Application Control ->Bandwidth Schedules, may result in an error message like "bandwith.obj.get failed 39838".
Conditions:
This happens only when all following conditions are met:
1. Modification is made at the time of SE syncing data to GUI.
2. The schedule rule of bandwidth is "specify day range: " and start/end day are configured with same value.
Workaround:
Open bandwidth edit page and do the modification again.
or
Avoid configuring start/end day of "Specify day range: " with same value. If user wants to set that kind of schedule, the right way is to use " Use specific days: " rule instead.
For example, instead of configuring both start and end day of "Specify day range" with Sunday, user should select "Use specific days:" rule and check "Sunday".
Platform
•
Symptom:
During upgrade, the system may go into KDB.
Conditions:
This occurs very rarely and it happened when there is streaming traffic running during upgrade.
Workaround:
1. Wait until traffic is completely off before upgrade.
If the box does go into KDB:
2. If the box has KDB enabled at CLI, reboot it from Console, or power cycle.
3. If the box doesn't have KDB enabled at CLI, the box would reboot by itself, and this failure will be no harm to customer.
Service Routing
•
Symptom:
When delivery service session exceeds quota and sends a rtsp request directly to SE. The content is not playing as expected and we receive 499 response in SE translog. But session is established with OS and wmt hdr file is cached in SE. The transaction log is as follows:
5.50.0.11 2013-01-23 05:43:11 - rtsp://V9-2S3-Fiber.se.spcdn.com/vod/test.wmv 0 0 0 499 {3300AD50-2C39-46C0-AE0A-AB5E3960BA83} - - "WMPlayer/11.0.5721.5251" - - - - - - 0 0 - rtsp - - - - 0 - 0 - - - - - - - - - - 4.25.1.14 - 1 -
Conditions:
When the delivery service session exceeds quota, and the request for a cache miss content.
Workaround:
None.
Open Caveats in Release 3.1.2-b17
Release 3.1.2-b17 contains the following open caveats:
Multicast Distribution
•
Symptom:
Sometimes the files are transferred by way of unicast when the delivery service is of the type Multicast Unicast.
Conditions:
When the multicast receiver has an alternate forwarder path to receive the content via unicast distribution.
Workaround:
Please note that
- the content is pre-positioned successfully
- the issue is observed rarely.
- and failover from multicast to unicast is expected in case of an error in multicast distribution
so there is no impact of this issue on the feature as far as the customer is concerned. No workaround is needed.
Web Engine
•
Symptom:
Web Engine core dumps and hence all the active sessions will be lost, leading to service disruption.
Conditions:
This occurs in scenarios when the following conditions are met for HLS playout-
- HLS Session Tracking feature is enabled
- Cache Bypass (no-store option in the response header)
- The .m3u8 file requested is an empty file (0 bytes)
Workaround:
Avoid zero byte files OR Disable Session Tracking.
•
Symptom:
The CLI "web-engine range-cache-fill enable" was hidden in Release 3.0 since KCF was introduced and this CLI is not used anymore. However, HTTPS does not use the KCF code and still needs this CLI to be executed to do a range-cache-fill.
Conditions:
When HTTPS is used.
Workaround:
This CLI still works as expected and can be executed. It's just hidden. This will not affect regular HTTP (KCF enabled) functioning.
•
Symptom:
If the primary and streaming interface are configured to be the same physical interface and then separated out by creating a new primary interface (no change to streaming), the web-engine process does not run on port 443. This disables HTTPS functionality.
eg: Initial Config -
primary-interface PortChannel 1
streaming-interface PortChannel 1
Now, Web-Engine process runs on port 80 and rpc_httpd process runs on port 443 on interface PortChannel 1
Change to -
no primary-interface PortChannel 1
primary-interface PortChannel 2
Now, the rpc_httpd process getsrestarted and is running on port 443 on PortChannel 2.
However, web-engine process is notrestarted and thus runs only on port 80. There is no trigger for it to run on port 443 on the Streaming interface - PortChannel 1
Conditions:
See Symptoms for details.
Workaround:
In the above scenario, execute the following command -
no streaming-interface PortChannel 1
streaming-interface PortChannel 1
This will restart the web-engine process and will now run on port 443 on the streaming interface - PortChannel 1 and HTTPS should work fine.
WMT
•
Symptom:
SE admin shell responds with latency (around 50 seconds).
Conditions:
The common_edm process hangs during unregistering from the data server.
Workaround:
None.
Adobe-Fms
•
Symptom:
fmscore process coredumps seen under very high stress.
Conditions:
Under too much of stress (alarms generated before), fmscore process coredumps randomly.
Workaround:
Nothing found at this point. But if SR is there in the deployment, then this issue should not be seen as there were alarms generated before this coredump.
SPCDN
•
Symptom:
M3U playlists are returned empty when requested through CDS-IS.
Conditions:
Response from Origin Server doesn't consist "Content-Length" field.
Workaround:
Configure Origin Server to include "Content-Length" field if possible.
•
Symptom:
CDS-CLIENT-INFO not printed in custom transaction log for cache hit case.
Conditions:
When enabling "content flow trace" + "filter trace flow to client" with custom transaction log including "%{CDS-CLIENT-INFO}i" field, it's not printed correctly in cache hit scenario.
Workaround:
None.
•
Symptom:
Webengine session limit alarm is raised when there are more than 20000 webengine sessions, even though user can set maximum session to 60000 via CLI.
Conditions:
When there are more than 20000 webengine sessions.
Workaround:
None.
Open Caveats in Release 3.1.2-b11
Release 3.1.2-b11 contains the following open caveats:
Multicast Distribution
•
Symptom:
Sometimes the files are transferred by way of unicast when the delivery service is of the type Multicast Unicast.
Conditions:
When the backup Sender is in the same location as that of primary Sender. When an SE that is not assigned to the multicast cloud resides in any of the location as the Content Acquirer, primary Sender, or a middle-tier Receiver.
Workaround:
Assign the backup Sender to a new location. Assign the SE that is not part of the multicast cloud to a location that is not in the direct upstream path of the edge location. Content is distributed properly and no issues are observed.
Web Engine
•
Symptom:
On very rare instances, the Web Engine crashes and restarts.
Conditions:
During Web Engine startup if there are any configuration changes occurring simultaneously, corruption to the Web Engine data structures may occur and cause a crash of the Web Engine.
Workaround:
After the SE or Web-Engine process restarts, wait for 30 seconds before making any configuration changes using the CLI or CDSM GUI.
Data Server
•
Symptom:
SE admin shell responds with latency (around 50 seconds).
Conditions:
The common_edm process hangs during unregistering from the data server.
Workaround:
None.
CDSM
•
Symptom:
On the CDSM GUI device activation page, when the streaming interface IP address is set as the management IP address (no https-enabled delivery service exists), rpc_httpd fails to start up due to port 443 binding issue
Conditions:
In Release 3.0.0 and later releases, when the streaming interface IP address is set as the management IP address.
Upgrading the SE and CDSM from Release 2.x.x to Release 3.x.x with this GUI setting can also cause this issue.
Workaround:
Before upgrading the software, either check the Use SE's primary IP address for management communication checkbox on the CDSM GUI device activation page or use another IP address that is different than any streaming interface IP address (and assigned the IP address) if this device has more NICs.
•
Symptom:
CDSM may run slow when there are a lot of alarms.
Conditions:
This happens when large number of alarms are generated back to back. This problem was hit in 3.1.1 due to other bugs such as CSCuc92136 and CSCuc03325. Because CSCuc92136 and CSCuc03325 are both fixed this release, this problem is unlikely to cause any visible impact.
Workaround:
None.
Platform
•
Symptom:
The no shutdown command does not work with standby interfaces.
Conditions:
Configure two interfaces for standby interface; the no shutdown command does not work on the non-active interface.
Workaround:
Reboot the device.
Service Routing
•
Symptom:
SR does not redirect .m3u8 requests to SE unless the Movie Streamer service is enabled on the SE.
Conditions:
URL contains string ".sdp."
Workaround:
Enable Movie Streamer service on SEs.
•
Symptom:
The Service Router should limit simultaneous RPC calls to 20, but this limit is not applied.
Conditions:
The Service Router should limit simultaneous RPC calls to 20, but this limit is not applied.
Workaround:
None.
Resolved Caveats
The following caveats have been resolved since Cisco Internet Streamer CDS Release 3.1.2. Not all the resolved issues are mentioned here. The following list highlights the resolved caveats associated with customer deployment scenarios.
The resolved caveats section has the following subsecions:
•
•
•
•
Resolved Caveats in Release 3.1.2-b34
Table 2 lists the issues resolved in the Cisco Internet Streamer CDS 3.1.2-b34 release.
Click on the bug ID to view the bug details. This information is displayed in the Bug Toolkit.
Resolved Caveats in Release 3.1.2-b26
The following caveats are resolved in 3.1.2-b26:
Web Engine
•
Symptom:
Web Engine core dumps and hence all the active sessions will be lost, leading to service disruption.
Conditions:
This is seen in error scenarios when the following conditions are met for HLS playout-
- HLS Session Tracking feature is enabled
- Cache Bypass (no-store option in the response header)
- The .m3u8 file requested is an empty file (0 bytes)
•
Symptom:
WebEngine crashes after 3.1.2-b11 upgrade.
Conditions:
Player sends HTTP range request to SE and Windows Media Server is configured as the HTTP Origin Server.
Update to the problem scenarios, this problem happens in two cases
1) When a range request is sent (either WMT or non-WMT request) and OS replies with a 200 (ignoring range request) instead of 206.
2) When a small file request is sent and OS responds with HTTP header and complete body in one packet, with "Cache-Control: no-store" HTTP header option.
•
Symptom:
An incorrect webengine error log message indicating that Internal Server Error (HTTP code 500) has occurred, although the actual error is different from 500.
Conditions:
Certain error scenarios - 504 gateway timeout, Zero byte file request, Server not found (403).
•
Symptom:
Webengine session limit alarm is raised when there are more than 20000 webengine sessions, even though user can set max session to 60000 via CLI.
Conditions:
When there are more than 20000 webengine sessions.
CDSM
•
Symptom:
1) System -> Configuration -> HTTPS -> Root CA File Registration
a) Upload CA.crt - accepted
b) Upload test.txt - not accepted
c) Upload test.xlsx - not accepted
2) System -> Configuration -> HTTPS -> HTTPS Certification Files Registration
a) Upload Certificate File CA.crt - accepted
b) Upload Certificate File test.txt - accepted
c) Upload Certificate File test.xlsx - accepted
d) Upload Key File CA.crt - accepted
e) Upload Key File test.txt - accepted
f) Upload Key File test.xlsx - accepted
Conditions:
3.0 or later releases.
Transaction Logs
•
Symptom:
CDS-CLIENT-INFO not printed in custom transaction log for cache hit case.
Conditions:
When enabling "content flow trace" + "filter trace flow to client" with custom transaction log including "%{CDS-CLIENT-INFO}i" field, it is not printed correctly in cache hit scenario.
WMT
•
Symptom:
In syslog.txt, there are lines as
Nov 5 14:12:29 edgecache501-cardiff wmt_be: %SE-SYSLOGD-4-581002: pid-2426: HTEndLoop=0, should set timeout!
Conditions:
In wmt http vod playback, there is possiblity that full-cached content refreshes incorrectly.
Mp3-Live
•
Symptom:
M3U playlists are returned as empty when requested through CDS-IS.
Conditions:
Response from Origin Server does not consist "Content-Length" field.
Multicast Distribution
•
Symptom:
When the replication delivery type is set as Multicast or Unicast, we expect the distribution to all multicast receivers to occur over Multicast. But in some rare occasions the distribution occurs over Unicast.
Conditions:
The issue specifically occurs, if the multicast receiver has an alternate forwarder path to receive the content via unicast distribution. In cases where the notification to pre-position the content reaches ahead of the notification from the CDSM to the SE, saying that it is now a part of a multi-cast cloud for that DeliveryService this issue occurs.
Workaround:
Please note
- The content is pre-positioned successfully.
- The issue is observed rarely.
- And failover from multicast to unicast is expected in case of an error in multicast distribution.
So there is no impact of this issue on the feature as far as the customer is concerned. No workaround is needed.
Resolved Caveats in Release 3.1.2-b17
The following caveats are resolved in 3.1.2-b17:
Web Engine
•
Symptom:
If ACQ path is down (tier-1), the edge SE returns 500 (internal error) to client instead of 504(gateway-timeout) or 503(DNS failure).
Conditions:
The SE returns 500 to the client when its tier-1 nodes are down.
•
Symptom:
When client plays HDS stream, intermittently it gets 504 Gateway Timeout.
Conditions:
Race condition while OS timeout one TCP connection and SE reuse this connection.
•
Symptom:
WebEngine crashed after 3.1.2-b11 upgrade.
Conditions:
Player sends HTTP range request to SE and Windows Media Server is configured as the HTTP Origin Server.
CDSM
•
Symptom:
Flash Player(HDS) receives 504 error when admin adds new dummy Origin Server or Delivery Service into system.
Conditions:
Operator add new OS or DS from GUI.
•
Symptom:
Adobe HDS client received 504 errors while Admin creates a non-related delivery service on CDSM.
Conditions:
Creation of delivery service will impact other delivery service on 3.1.2.
Authorization Server
•
Symptom:
Bad usage of CdsRegex regular expression leads to memory leak, it doesn't free the momory pointed to by "real_preg".
Conditions:
Valgrind checks if Auther server releases rule file in memory when deliver Service changes rule file from one file to antoher file(none).
Service Routing
•
Symptom:
SR does not redirect .m3u8 requests to SE unless the MovieStreamer service is enabled on the SE.
Conditions:
URL contains string ".sdp"
•
Symptom:
The service router should limit simultaneous RPC calls to 20, but this limit is not applied.
Conditions:
The limit is always not working.
•
Symptom:
Service router core dump.
Conditions:
When client sends http/rtsp request with no '.' in its url.
Platform
•
Symptom:
Service engine, service router does not reply with an ACK on a TCP syn.
Conditions:
Execute TAC supplied script to disable/enable tcp_tw_recycle or tcp_tw_reuse --or-- execute script that ignores the TCP timestamp. Use persistent connections on the clients.
•
Symptom:
Add a show command to know the tcp settings after addition of these CLIs.
Conditions:
•
Symptom:
The below alarm condition may be seen:
-Rootfs partition out of free inodes
In addition, the following error messages may be seen in the syslog:
... hil-cds-se1 kernel: %SE-SYS-3-900000: IPMI Watchdog: response: Error ff on cmd 22
Conditions:
This condition only occurs on CDE250 platforms.
•
Symptom:
Upgrade from 2.5.X based releases to 2.X and/or 3.X releases results in intermittent premature aborts.Subsequently, rebooting / power-cycling the system results in the original 2.5.X release coming up.
Conditions:
Under very rare circumstances, pending access(es) to CDNFS volumes (even after offloading the device),may cause this upgrade issue.
•
Symptom:
After upgrading all SEs from 2.6.3 to 3.1.2.7, it was found that that 15 SEs out of 50 SEs have their CPU threshold exceeded. The processor list displays 10 to 20 WMT Backend processes indicating abnormally high CPU usage, some at 100%. The aggregate of these WMT processes uses up all available processor availability and causes the service monitor CPU threshold exceeded to be raised, which in turn causes the service router to stop sending new service request to these SEs.
Conditions:
upgrading all SE's from 2.6.3 to 3.1.2.7.
DSS
•
Symptom:
Signed URLs in movie-streamer do not log correctly in translog.
Conditions:
When URL signing is applied for movie-streamer request;
and S * 2 + 10 + R + E > 255 (S: length of query string excluding question mark in a signed URL, R: length of request URL, E: length of the string "sename.se." in RFQDN).
Resolved Caveats in Release 3.1.2-b11
The following caveats are resolved in 3.1.2-b11:
Web Engine
•
Symptom:
During mixed profile (cache miss and cache hit requests) HTTPS stress test, some sessions were left hanging after a 15 hour run. There were associated Data sources also leftover. These hanging sessions look to be a cache hit scenario as we saw Managed Cache Data Source being left over.
Conditions:
This happens only under stress. Not easily reproducible. Some corner case is resulting in this behavior. The exact scenario is yet to be identified. This happens when there is a disk file create failure.
•
Symptom:
The Web Engine goes into a core dump on long URLs.
Conditions:
The request is a POST message that is posted by the client for the WMT translog.
•
SE goes offline during upgrade from 2.6.3 to 3.1.0/3.1.1.
Conditions:
Upgrade from 2.6.3 to 3.1.x when there are more than 25 delivery services.
•
Symptom:
Additional requests for revalidation of content despite 304 response within cache timeout.
Conditions:
Additional requests for revalidation of content despite 304 response within cache timeout.
•
Symptom:
1.
2.
3.
4.
Conditions:
SRs/SEs/backup CDSM send http[s] messages to the primary CDSM which is handled by the rpc_httpd process on the CDSM. These requests are the http[s] messages that report the health of the various nodes to CDSM.
Apache(rpc_httpd) uses ssl_scache.pag file to speed up parallel request processing by avoiding unnecessary session handshakes. At every SSLSessionCacheTimeout interval the global/inter-process SSL Session Cache information is timed out, with the httpd process acquiring a lock and traversing the records. Due to the size of the file (44gb) this operation is taking excessively long time thereby blocking other processes from reading the file for session information.
Since the customer enabled "Fast SE Offline Detection mode", the SEs health is communicated to the CDSM using UDP messages (and not the http[s] mechanism). This corresponds to what was observed, where only the backup CDSM and the SRs were offline, while the SEs were reported to be online.
Happens once every 6 months (with 50+ SEs, 4SRs and 1 backup CDSM communicating with the primary CDSM using SSL.
•
Symptom:
System is not functioning well, such as wget to SE may hang, SE may become very slow, SR may see SE offline, and so on.
Conditions:
Installing the software by using the installation CD or rescue CD may result in this problem.
•
Symptom:
The files delivered through cache bypass flow may have a corrupted body with contents beginning at offset 0 repeats later in the body.
Conditions:
This problem happens in following scenarios:
–
–
•
Symptom:
HLS streaming stops when HLS session tracking is enabled.
Conditions:
The Origin server configures "private" or "no-store" directive in HTTP response Cache-Control header.
•
Symptom:
WebEngine in edge SE bypasses Content Acquirer and reaches Origin server directly.
Conditions:
1. Edge SE, which means serving client directly
2. HTTP head request
3. The requesting content is cache-hit and expired and need revalidation
4. There is an existing Datasource for the requesting url but *no* existing DataSourceFinder for the request url.
•
Symptom:
Users may get HTTP 404 errors when the new "Follow Origin Server Redirect" feature is enabled on CDS delivery service. This occurs because the hostname is not being re-written to what is listed in the 302 redirect location.
Conditions:
Enable the new "Follow Origin Server Redirect" feature on a delivery service. Redirected web server needs proper hostname to route traffic to proper application pool.
•
Symptom:
Custom access logs (transaction logs) shows large value in the column which corresponds to the "Bytes-Transferred-Excluding-Header" when no data is sent out to the downstream SE, for example:
x.x.x.x y.y.y.y 18446744073709551288 ....
(ClientIP) (SEIP) (BytesTransferred Excluding Header) ...
Conditions:
Only during overloaded conditions when the SE or Content Acquirer takes too long to send the data out to the downstream SE. But the downstream SE (after waiting for a certain time; default= 5secs), disconnects from the upstream SE and sends a 504-Gateway timeout to the client.
Windows Media Streaming
•
Symptom:
WMT process generate coredumps frequently.
Conditions:
WMT received a specific transaction log POST message from client.
•
Symptom:
Whenever the end user changes the resolution of the video, a garbled video image is seen and the CDS operator has to restart the live program through CDSM GUI.
Conditions:
End user changes video resolution.
Content Manager
•
Symptom:
When the clear-cache-all command is entered, sometimes the content is not completely deleted. This is because the Content Manager is not aware of all the content cached by the Web Engine.
Conditions:
When the Web Engine creates more than three million objects and the slow scan (slowscan) process has just finished running, some content is not known by the Content Manager.
CLI
•
Symptom:
NTP server is accepting an invalid IP address range.
Conditions:
While configuring the NTP server IP, the IP address should validate.
•
Symptom:
In CLI prompt, RTSP max-request-rate shows a range of <1-2147483647>.
Conditions:
When the rtsp advanced max-request-rate ? command is entered.
•
Symptom:
Upgrade failure from Release 2.6.3 to 3.1.
Conditions:
System has just been upgraded from older release to Release 2.6.3, Disk RAID is still sync pending.
•
Symptom:
Cannot access server via console. Keep getting "Username" prompt and "System Initializing. Please wait....."
Conditions:
Upgrade "failure" from Release 2.6.3 to Release 3.1.
Movie Streamer
•
Symptom:
Movie Streamer does not return error code for invalid URL signature.
Conditions:
Service rule is enabled and a request with signed URL is sent to the Movie Streamer engine.
•
Symptom:
When show programs, a DBAccess error appears
Conditions:
200 programs are created.
CDSM
•
Symptom:
After downgrading from 3.1.2 to 2.6.3, can not log in to the new primary CDSM, and communication is lost between primary CDSM and standby CDSM.
Conditions:
This may happen when database was manually edited and had some old entries in there. This should rarely happen in production network.
•
Symptom:
Device states turn yellow and red when lowering the quota for content services
Conditions:
1.
2.
3.
4.
•
Symptom:
The rpc_httpd process (the Apache process) goes in to a core dump when reloading.
Conditions:
When the device is reloading, the rpc_httpd process is exiting.
•
Symptom:
Unable to enter in Management Communication Address in Device Activation Screen when it's register the first time and inactive.
The reason is that customer need to do more steps to activate and deactivate to enter the Management Communication Address in Device Activation Screen. As production have around 90 plus devices to upgrade & activate, they looking for the quickest way possible to do this to reduce the workload.
Conditions:
Unable to enter in Management Communication Address in Device Activation Screen when it's register the first time and inactive.
•
Symptom:
When CDSM configuration for NTP apply to CDSM. The previous configuration still persist.
Conditions:
Applying the CDSM configuration for NTP applied to CDSM.
Steps To Reproduce the issue:
1.
2.
1.1.1.1 2.2.2.2 3.3.3.3 4.4.4.43.
4.
5.
cde-205-b-cdsm1# show running-config | in ntpntp server 1.1.1.1ntp server 2.2.2.2ntp server 3.3.3.3ntp server 4.4.4.46.
2.2.2.2 3.3.3.3 4.4.4.47.
cde-205-b-cdsm1# show running-config | in ntpntp server 1.1.1.1ntp server 2.2.2.2ntp server 3.3.3.3ntp server 4.4.4.48.
•
Symptom:
After reload, rules do not apply immediately.
Conditions:
It happens only for a short period after reloading the device.
•
Symptom:
CDSM GUI may become slow or unresponsive.
Conditions:
When devices send alarms to CDSM.
•
Symptom:
SE cannot be unassigned from the delivery service after downgraded the CDSM from Release 3.1 to Release 3.0. When unassigning an SE from a delivery service, the operation is rejected, with an error message saying this SE cannot be unassigned, because it is the Sender of a Multicast Cloud assigned to the delivery service.
Conditions:
Before downgrading the CDSM from Release 3.1, Multicast Clouds are configured and the SE is the Sender in the Multicast Cloud, and this Multicast Cloud has been assigned to the delivery service. And, the downgrade script "cms database downgrade script Downgrade3_1_to_3_0"(documented in the Release Notes for 3.1 was not run before running "cms database downgrade" on the CDSM.
•
Symptom:
When multiple delivery services are created and a new delivery service is added, the keep alive message (heartbeat) may not sent out in a timely manner. Use the tcpdump udp port 2000 command to monitor the packets sent out from the SE; some delay can be seen.
Conditions:
Add or delete a delivery service, the keep live message is missed or delayed.
•
Symptom:
A delivery service is showing a replication status of yellow and "No status reported."
Conditions:
When the content type of a delivery service has zero pre-position storage quota.
CSCuc38057
Symptom:
Cannot configure single word domain name in CDSM. Following error is seen:
------------------------------------------------Transaction not completed.Domain Names must match following rules:1. Each label between dots must be no less than 1 character and no more than 63 characters.2. Each label between dots must start with letters and end with letters or numbers.3. Only letters, numbers and "-" are valid for labels between dots.------------------------------------------------Conditions:
Configure DNS in CDSM under General Settings > Network > DNS
•
Symptom:
After reload, rules do not apply immediately.
Conditions:
It happens only for a short period after reloading the device.
Authorization Server
•
Symptom:
The Service Rule does not include a left parenthesis "("when the NOT ("!") operator is used.
Conditions:
When the Internet Streamer CDS is configured with service rules by using the Service Rule XML file and if the rule has !(<some regex>), then the left parenthesis after the NOT operator is not considered.
•
Symptom:
The Authorization Server returns a 302 message in release 2.6.3 compared to a 403 message in release 2.5.11. This happens when the Geo Server fails during a URL signing pass.
Conditions:
URL signing is applied and the error-redirect-url is configured.
Content Abstraction Layer (CAL)
•
Symptom:
HIgh disk usage.
Conditions:
Slow scan in progress.
Web Services
•
Symptom:
Based on the investigation over the error logs and the web-engine code, the root cause is:
1. Some existing data sources in the memory ended up with a bad status due to content downloading being terminated by OS HTTP response body read timeout 504 (due to OS being overloaded)
2. At the same time, many new requests kept hitting the existing data sources
3. Before using those data sources, there is not status check. Therefore, 500 was returned to client
4. Since new requests kept hitting the data sources, the data sources could not be evicted from the memory
In other words,
1. OS 504 causes data source in bad status;
2. Heavy traffic causes data sources hanging in the memory
3. Bad status causes 500 returned to client
Conditions:
Origin server is under heavy load.
Syslog
•
Symptom:
In syslog.txt, the following lines appear:
Nov 5 14:12:29 edgecache501-cardiff wmt_be: %SE-SYSLOGD-4-581002: pid-2426:HTEndLoop=0, should set timeout!Conditions:
WMT HTTP and acquirer.
Service Monitor
•
Symptom:
When service_monitor process is hung, SE state inside SR is flapping. The interval of UDP packet for keepalive is increased from 2 seconds to 8 seconds. It is because there are multiple SR devices in production network. Each query will add extra delay.
Conditions:
SE service_monitor process is hung.
Service Routing
•
Symptom:
SR is using streaming interface's ip address for its PRC call back to SE. And RPC call is not successful because SE's streaming interface is not reachable from SR.
Note
Conditions:
1. SE is using primary interface to send KAL message to SR
2. The streaming interface's ip address is not reachable from SR
•
Symptom:
There is no external visible symptom. This is an internal code enhancement for reliability.
Conditions:
This is related to whole-sale feature.
•
Symptom:
Lots of established TCP connections do not timeout.
Conditions:
Client connects to SR (TCP port 80/554) and does not send any requests.
•
Symptom:
The Service Router has a core dump while checking the SR statistics for the API request and 302 redirect.
Conditions:
1.
http://ServicerouterIP/routeURL?CDNURL=<RequestedURL>&ClientIP=<IPOfTheClient>;2.
3.
•
Symptom:
The ability of the content based routing feature is limited when URL signing is used. Content based routing generates a hash on the whole URL which includes the signed part and therefore is always unique. So all traffic go to origin server.
Conditions:
When URL signing feature is used.
Acquisition and Distribution
•
Symptom:
The correct play length is not displayed for .wmv files.
Conditions:
When WMV files are prefetched, the play length is displayed as 00:00:00.
Platform
•
Symptom:
False Power Supply Temperature alarm(s) seen from the show alarms command.
Major Alarms:-------------Alarm ID Module/Submodule Instance-------------------- -------------------- -------------------------1 temp sysmon System TemperatureMar 26 20:26:07.828 UTC, Equipment Alarm, #000002, 1000:445009System temperature threshold(s) exceeded.Please run 'show tech-support';check the System Health section for details.Conditions:
Enter the show tech command, and look for the "-- system health --"section.
---------------------- system health -----------------------*****************************************************************************Supero Doctor II - Linux version 2.79(110504)Copyright(c) 1993-2010 by Super Micro Computer, Inc. http://supermicro.com/*****************************************************************************Monitored Item High Limit Low Limit Status----------------------------------------------------------------------Fan2 Fan Speed 712 6221Fan3 Fan Speed 712 6459Fan8/CPU2 Fan Speed 712 6279Power1 Fan 1 Speed 4007 5267Power1 Fan 2 Speed 4007 7099CPU1 Vcore Voltage 1.48 0.92 1.09CPU2 Vcore Voltage 1.48 0.92 1.08-12V Voltage -10.58 -13.40 -12.21+1.5V Voltage 1.98 1.34 1.52+3.3V Voltage 3.63 2.96 3.33+12V Voltage 13.25 10.75 11.81+5V Voltage 5.50 4.49 4.92+5Vsb Voltage 5.50 4.49 4.87VBAT Voltage 3.66 2.99 3.17PECI Agent1 Temperature 93/199 LowPECI Agent2 Temperature 93/199 LowSystem Temperature 75/167 25/77Power1 Temperature 0/32 27/80 Warning! <-- Invalid High Limit (0/32)Power2 Temperature 0/32 110/230 Warning! <-- Invalid High Limit (0/32)Chassis Intrusion Bad Warning!Power Supply Failure Bad Warning!Power1 Supply Failure GoodPower2 Supply Failure Bad Warning!--------------------------------------------- Mon Apr 9 09:44:50 2012If you see "0/32" for either Power1 / Power2 Temperature High Limits,
then the Power Supply Temperature alarm(s) is / are false and can be ignored.
•
Symptom:
PSU alarm is raised when CDE205 has only a single PSU.
Conditions:
If CDE205 has only a single PSU, an alarm will be raised.
•
Symptom:
Write Memory Errors displayed; "No space left on device" /state partition not mounted to the appropriate internal SSD partition.
Conditions:
Trying to save running config to startup flash, this is seen on new CDE250 boxes.
Transaction Logs
•
Symptom:
Alarms from ftp_export does not get cleared.
Conditions:
The instance of the alarm (address of ftp-server) is stuck. The configuration of the address has been removed.
•
Symptom:
The ftp_export process was stuck.
Conditions:
The sftp-server (SSH) process does not respond to the ftp_export and does not break the connection.
RTSP Gateway
•
Symptom:
Alarm is not clearing despite rtspg logs showing that there is no traffic hitting the device, at the same time the service_monitor_errorlog is filled up with threshold exceeded alarm.
Major Alarms:-------------Alarm ID Module/Submodule Instance-------------------- -------------------- -------------------------1 tpsquotaexceed streaming/rtspg rtspg_tps_40Conditions:
These were generated during a change where 11 SEs were removed from a custom Device Group and assigned to "BASELINE" Device Group.
MP3 Live
•
Symptom:
When the CDS serves an AAC-encoded shoutcast stream (Content-type: audio/aacp), it also adds the HTTP header "Content-length: 0". This header is normally used to indicate there is no data to be transferred.
Some clients like VLC ignore this header and play the stream properly, but, some other clients such as Iphone/Ipad honor the header and immediately close the connection.
Conditions:
When the CDS serves an AAC-encoded shoutcast stream.
Cache Router
•
Symptom:
Cache router hash level calculation goes wrong if the URL used for calculating starts with "xxx.se."
Conditions:
If the OFQDN has the format "xxx.se.yyy," it is stripped by the Web Engine, and makes the hash level use the default value 0.
Data Server
•
Symptom:
Service Router does not serve new added delivery service\
Conditions:
There are some alarm need to raise simultaneously on the SR, a race condition could happen.
Accessing Bug Tool kit
This section explains how to use the Bug Toolkit to search for a specific bug or to search for all bugs in a release.
Step 1
Step 2
Note
Step 3
Step 4
•
•
•
•
•
–
–
Choose Terminated to view terminated bugs. To filter terminated bugs, uncheck the Terminated check box and select the appropriate suboption (Closed, Junked, or Unreproducible) that appears below the Terminated check box. Select multiple options as required.
Choose Open to view all open bugs. To filter the open bugs, uncheck the Open check box and select the appropriate suboptions that appear below the Open check box. For example, if you want to view only new bugs in Prime Optical 9.5, choose only New.
Choose Fixed to view fixed bugs. To filter fixed bugs, uncheck the Fixed check box and select the appropriate suboption (Resolved or Verified) that appears below the Fixed check box.
–
–
–
Step 5
Step 6
a.
b.
c.
If you cannot export the spreadsheet, log into the Technical Support website at http://www.cisco.com/cisco/web/support/index.html or contact the Cisco Technical Assistance Center (TAC).
Upgrading to Release 3.1.2
Release 3.1.2 supports upgrades from Release 2.5.9, Release 2.5.11, Release 2.6.x, Release 3.0.x, and Release 3.1.x.
Note
Note
If your CDS is running an older release than Release 2.5.9, you need to upgrade to Release 2.5.9 or 2.5.11 before upgrading to Release 3.1.2.
When upgrading from Release 2.5.9 or 2.5.11, all content is erased. For Service Engines, this means that prefetched metadata and content need to be redistributed from upstream SEs after the upgrade, and that cached content is not preserved. Additionally, Flash Media Streaming service rules must be converted from device-based service rules to the Service Rule XML file. For more information on upgrading from Release 2.5.9 and 2.5.11, see the Cisco Internet Streamer CDS 2.6 Software Upgrade Guide (http://www.cisco.com/en/US/docs/video/cds/cda/is/2_6/upgrade_guide/upgrade.html).
We strongly recommend that you upgrade your CDS network devices in the following order:
1.
2.
3.
4.
5.
6.
7.
8.
Note
Alternatively, if you do not switch roles of the standby CDSM and primary CDSM, you can use the CDSM GUI to upgrade the primary CDSM. The primary CDSM loses connectivity with the CDS devices for a short time during the upgrade, but this is not service affecting.
When using the CDSM GUI to upgrade from Release 2.6.3 and later releases to Release 3.11, after you upgrade the standby CDSM, if you switch roles of the standby CDSM and primary CDSM to maintain an active CDSM at all times, the new primary CDSM GUI can be used to upgrade the new standby CDSM.
For more information on the upgrade procedure, see the Cisco Internet Streamer CDS 3.1 Software Configuration Guide.
After the upgrade procedure starts, do not make any configuration changes until all the devices have been upgraded.
Downgrading from Release 3.1.2
Note
For software downgrades from Release 3.1.2 on systems with primary and standby CDSMs, you need to do the following:
Step 1
If you are using the CLI, downgrade the primary CDSM first, followed by the standby CDSM.
Step 2
•
cms database downgrade script downgrade/Downgrade3_1_1_to_3_1cms database downgrade script downgrade/Downgrade3_1_to_3_0cms database downgrade script downgrade/Downgrade3_0_to_2_6cms database downgradecms enableThen, consult the "Downgrading the Internet Streamer CDS Software" chapter in the Cisco Internet Streamer CDS 2.6 Software Upgrade Guide for downgrading from Release 2.6.x to Release 2.5.9 or 2.5.11.
•
cms database downgrade script downgrade/Downgrade3_1_1_to_3_1cms database downgrade script downgrade/Downgrade3_1_to_3_0cms database downgrade script downgrade/Downgrade3_0_to_2_6cms database downgradecms enable•
cms database downgrade script downgrade/Downgrade3_1_1_to_3_1cms database downgrade script downgrade/Downgrade3_1_to_3_0cms database downgradecms enable•
cms database downgrade script downgrade/Downgrade3_1_1_to_3_1cms database downgradecms enableStep 3
Note
Documentation Updates
The following document has been added for this release:
•
Related Documentation
Refer to the following documents for additional information about Cisco Internet Streamer CDS 3.1:
•
http://www.cisco.com/en/US/docs/video/cds/cda/is/3_1/configuration_guide/icds31confg.html
•
http://www.cisco.com/en/US/docs/video/cds/cda/is/3_0/quick_guide/ISCDSQuickStart.html
•
http://www.cisco.com/en/US/docs/video/cds/cda/is/3_1/developer_guide/iscds31APIGuide.html
•
http://www.cisco.com/en/US/docs/video/cds/cda/is/3_1/command_reference/Command_Ref.html
•
http://www.cisco.com/en/US/docs/video/cds/cda/is/3_1/message_guide/message_guide.html
•
•
http://www.cisco.com/en/US/docs/video/cds/overview/CDS_Roadmap3.x.html
•
•
•
http://www.cisco.com/en/US/docs/video/cds/cde/regulatory/compliance/CDE_RCSI.html
•
http://www.cisco.com/en/US/docs/unified_computing/ucs/c/hw/C200M1/install/c200M1.html
•
http://www.cisco.com/en/US/docs/unified_computing/ucs/c/hw/C210M1/install/C210M1.html
The entire CDS software documentation suite is available on Cisco.com at:
http://www.cisco.com/en/US/products/ps7127/tsd_products_support_series_home.html
The entire CDS hardware documentation suite is available on Cisco.com at:
http://www.cisco.com/en/US/products/ps7126/tsd_products_support_series_home.html
The Cisco UCS C-Series Rack Servers documentation is available on Cisco.com at:
http://www.cisco.com/en/US/products/ps10493/prod_installation_guides_list.html
Obtaining Documentation and Submitting a Service Request
For information on obtaining documentation, submitting a service request, and gathering additional information, see the monthly What's New in Cisco Product Documentation, which also lists all new and revised Cisco technical documentation, at:
http://www.cisco.com/en/US/docs/general/whatsnew/whatsnew.html
Subscribe to the What's New in Cisco Product Documentation as a Really Simple Syndication (RSS) feed and set content to be delivered directly to your desktop using a reader application. The RSS feeds are a free service and Cisco currently supports RSS version 2.0.
This document is to be used in conjunction with the documents listed in the "Related Documentation" section.
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (1110R)
Any Internet Protocol (IP) addresses used in this document are not intended to be actual addresses. Any examples, command display output, and figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses in illustrative content is unintentional and coincidental.
© 2013 Cisco Systems, Inc. All rights reserved.
