Re-indexing causes time out

[Daisuke Niwa]

Hi there,

Our customer has found out following error is recorded in the log.

Regarding this,

1) What does this error stand for?

2) What is "Threshold" parameter about? It seems this has something to do with re-indexing.

If you have an idea, please let me know.

atlassian-jira.log(78878): 2012-12-05 16:45:59,796 http-8080-13 INFO 0000106263

1005x251702x1 j5r05q <masked IP address> /secure/EditIssue.jspa [jira.issue.
index.DefaultIndexManager] Optimizing the index because 4000 issues have been re-
indexed since last optimize. Threshold is 4000.

atlassian-jira.log(78879): 2012-12-05 16:46:35,141 http-8080-40 ERROR 0000103356
1006x251776x2 1ee5bwn <masked IP address> /secure/EditIssue.jspa [jira.issue.
index.DefaultIndexManager] Wait attempt timed out - waited 30000 milliseconds
atlassian-jira.log(78880): com.atlassian.jira.issue.index.IndexException: Wait attempt
timed out - waited 30000 milliseconds

atlassian-jira.log(79018): 2012-12-05 16:46:35,161 http-8080-40 INFO 0000103356
1006x251776x2 1ee5bwn <masked IP address> /secure/EditIssue.jspa [jira.issue.
index.DefaultIndexManager] Optimizing the index because 4001 issues have been re-
indexed since last optimize. Threshold is 4000.

atlassian-jira.log(79019): 2012-12-05 16:46:37,024 http-8080-50 ERROR 0000112279
1006x251780x4 px3dv6 <masked IP address> /secure/AssignIssue.jspa [jira.issue.
index.DefaultIndexManager] Wait attempt timed out - waited 30000 milliseconds

Regards,

Daisuke Niwa

[Naren · Accepted solution]

The error -

Optimizing the index because 4000 issues have been re-
indexed since last optimize. Threshold is 4000.

Basically the optimzing has caused the thread to time out, because of large JIRA instances.

The second error -

Wait attempt
timed out - waited 30000 milliseconds

This is caused because of JIRA acquiring the lock on Lucene index. Please find the detailed documentation here.

You can solve this problem by increasing the waittime, following the steps provided in the documentation.

[win bolton]

I have this same problem. Whenever I see this message in the log, com.atlassian.jira.issue.index.IndexException: Wait attempt timed out - waited 30000 milliseconds, I can guarantee that Jira is locked up for everyone and I need to restart. When this happens there is always 1 CPU at 100% utilization and doesnt go away. I believe I have ample specs on our Jira server with 4 CPU and 10 GB Ram (5 alocated to JVM) so I would doubt its a resource issue. This seems to happen to us about once per month. Whenever I take a thread dump I see several threads blocked on the NextPreviousPager. It looks like there is a bug on this. JRA-21557. Any insight into what is causing the hang?

[Erik Saline [BlackPearl PDM]]

This article talks about possible disk speed issues, ant-virus or garbage collection. https://confluence.atlassian.com/display/JIRAKB/Lock+timout+-+Waited+30000+Milliseconds+Error+Message