July 18, 2013 at 7:02 am
After a peak load test carried out for an application for two consecutive days, the DB server got automatically bounced the next day with following event logged.
Does a deadlock situation/lot of concurrent connections bring down the server? What should be taken care of in order to address this issue?
The following is the error message from the logs:
[sqsrvres] printODBCError: sqlstate = 08S01; native error = 0; message = [Microsoft][SQL Server Native Client 10.0]The connection is no longer usable because the server failed to respond to a command cancellation for a previously executed statement in a timely manner. Possible causes include application deadlocks or the server being overloaded. Open a new connection and re-try the operation.
Cluster resource 'SQL Server' in clustered service or application 'SQL Server (MSSQLSERVER)' failed.
Cluster Agent: The cluster resource SQL Server has failed.
[SNMP TRAP: 15006 in CPQCLUS.MIB]
Cluster Agent: The cluster resource SQL Server Agent has failed.
[SNMP TRAP: 15006 in CPQCLUS.MIB]
The SQL Server Agent (MSSQLSERVER) service terminated unexpectedly. It has done this 2 time(s).
Cluster Agent: The cluster resource SQL Server Agent has become degraded.
[SNMP TRAP: 15005 in CPQCLUS.MIB]
[sqsrvres] CheckQueryProcessorAlive: sqlexecdirect failed
[sqsrvres] printODBCError: sqlstate = HYT00; native error = 0; message = [Microsoft][SQL Server Native Client 10.0]Query timeout expired
[sqsrvres] OnlineThread: QP is not online.
July 18, 2013 at 7:19 am
The SQL instance failed on the cluster. You refer (with deadlock, concurrency) to the error in the ODBC connection. This error is not the rootcause. This error just occurs because the instance on the cluster has failed and can't be reached.
Can you post the errors in the SQL logging?
Also take a look in the Windows Event log, allthough I suspect they won't be very descriptive. But maybe you can relate the failure of SQL to other errors in the eventlog, just prior to the failure of SQL. It could for example be that a disk error occurred or the clusternode is rebooted due to a heartbeat failure.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply