5.1SP1 VCS Agents hung up

book

Article ID: 100006860

calendar_today

Updated On:

Description

Error Message

--In Oracle_A.log:
2011/05/03 10:12:32 VCS WARNING V-16-2-13139 Thread(4156308368) Canceling thread (4153408400)   <<<< oracle agent hung?????
2011/05/03 10:12:32 VCS ERROR V-16-2-13027 Thread(4149095312) Resource(Ora_Oracle) - monitor procedure did not complete within the expected time.
2011/05/03 10:14:33 VCS WARNING V-16-2-13139 Thread(4156308368) Canceling thread (4149095312)
2011/05/23 11:18:54 VCS ERROR V-16-2-13120 Thread(4158568144) Error receiving from the engine. Agent(Oracle) is exiting.
 

Cause

--That is a known issue in the VCS Release Notes for 5.1SP1.
The agent framework does not detect if service threads hang inside an entry point
In rare cases, the agent framework does not detect if all service threads hang inside a C entry point. In this case it may not cancel them successfully. [1511211]
Workaround: If the service threads of the agent are hung, send a kill signal to restart the agent. Use the following command: kill -9 hung agent's pid. The haagent -stop command does not work in this situation.
 

Resolution

--Workaround: If the service threads of the agent are hung, send a kill signal to restart the agent. Use the following command: kill -9 hung agent's pid.The haagent -stop command does not work in this situation.
--Stop and start VCS.


Applies To

--VCS 5.1SP1 on Solaris

Issue/Introduction

The customer reports during testing of the QA VCS cluster, an attempt was made to switch the db10 database from node db12 to node db11The had daemon did not send the command to the oracle database to offline the database resource, nor to bring down any other of the resources in the OraGrp. He needs to stop and restart VCS.

Additional Information

ETrack: 1442255 ETrack: 1511211