Managed machines not registering with Task Server post-upgrade to ITMS 8.6 RU2

Products

• Client Management Suite

Issue/Introduction

After upgrading to ITMS 8.6 RU2 none of the managed machines are able to register with their Task Server.  The Sym Agents & plug-ins have all upgraded to the latest version as have the Task Management plug-ins (Altiris Client Task Agent & Altiris Client Task Server Agent) on the Site Servers.

The SMA agent.logs include errors like the following:

Agent Service will be terminated due to an excess of plugin stopping time over the limit or unexpected stopping failure!
Agent Service exit time limit of 15 seconds expired
Agent Service will be terminated due to an excess of stopping time over the limit!
Failed to send SMP frame '{XXXX131A-XXXX-4XXX-XXXX-XXXXXXXXXXXX}', ID: {XXXXE414-XXXX-XXXX-XXXX-XXXXXXXXXXXX}, timeout: 120 seconds, error: This operation returned because the timeout period expired (0x000005B4)
Task Server Connection: Failed to receive persistent properties from NS, error: This operation returned because the timeout period expired (0x800705B4)
Task Server Connection: Failed to get server list from NS, error: This operation returned because the timeout period expired (0x800705B4)
Task Server Connection: Failed to refresh server list or register, error: This operation returned because the timeout period expired (0x800705B4)

Environment

ITMS 8.6 RU2

Cause

The upgrade didn't complete correctly.

When you see the error "Failed to send SMP frame," it suggests that the Symantec Management Agent on a given client machine has encountered a problem when trying to communicate or send data back to the SMP server. This could be due to a variety of reasons including network connectivity issues, server unavailability, configuration issues, or potential issues with the agent itself.

Resolution

Typically, you can resolve this issue by going into Symantec Installation Manager (SIM) and running a full repair & reconfigure.

The following is an explanation of the above errors:

• The Agent Service time limit error can be seen when agents cannot shutdown or restart in 15 seconds, so it kills itself and restarts. That happens when some plugins cannot stop in time. In this case this is the CTA agent that wasn't quick enough during shutdown and the cause is those 120 seconds request timeouts. While there is certainly something happening in the CTA stopping routine, the time limit error is completely recoverable and not important now.
• About the 120 second timeout - this issue is on the NS side, where the NS does not answer for Task Server requests coming over persistent connections. Task Management on the NS uses the same persistent connection as the NS, and the policy requests and the events are working over persistent connections just fine but TS requests like "persistent properties" are not working. Do not confuse "persistent connection" with "persistent properties requests", the last one is just one of the CTA requests, it can go over persistent connections or HTTP/S.
• See if Task Management has been upgraded successfully on the NS and look for websocket related errors in the NS a.log files.
• The client logs provided indicate that TS requests are not being processed over persistent connections at all for some reason.