You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jackrabbit.apache.org by "licheng (JIRA)" <ji...@apache.org> on 2012/06/15 10:05:46 UTC
[jira] [Created] (JCR-3344) Multhireads hang issue occurs because
of oracle database socket read blocks forever
licheng created JCR-3344:
----------------------------
Summary: Multhireads hang issue occurs because of oracle database socket read blocks forever
Key: JCR-3344
URL: https://issues.apache.org/jira/browse/JCR-3344
Project: Jackrabbit Content Repository
Issue Type: Bug
Components: jackrabbit-core
Affects Versions: 2.2
Environment: Winxp+oracle+JBoss
Reporter: licheng
Fix For: 2.2
Attachments: dead lock_mr1.log, deak lock_mr2.log
During our test, the network between JBoss and Oracle has some problem for a while. At that time, one JR thread which holding Global ISM lock is doing socket read with Oracle. Unfortunately, this blocking socket read doesn't return anymore. Meanwhile, some other threads are warting for the global ISM lock. As a result, many threads are blocked.
Although we only hit it once, but it is really a serious problem. To workarount it, we have to restart our application server. To our custom, it is not acceptable.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (JCR-3344) Multhireads hang issue occurs because
of oracle database socket read blocks forever
Posted by "licheng (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/JCR-3344?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
licheng updated JCR-3344:
-------------------------
Attachment: dead lock_mr1.log
deak lock_mr2.log
The thread dump is attached.
> Multhireads hang issue occurs because of oracle database socket read blocks forever
> -----------------------------------------------------------------------------------
>
> Key: JCR-3344
> URL: https://issues.apache.org/jira/browse/JCR-3344
> Project: Jackrabbit Content Repository
> Issue Type: Bug
> Components: jackrabbit-core
> Affects Versions: 2.2
> Environment: Winxp+oracle+JBoss
> Reporter: licheng
> Fix For: 2.2
>
> Attachments: dead lock_mr1.log, deak lock_mr2.log
>
>
> During our test, the network between JBoss and Oracle has some problem for a while. At that time, one JR thread which holding Global ISM lock is doing socket read with Oracle. Unfortunately, this blocking socket read doesn't return anymore. Meanwhile, some other threads are warting for the global ISM lock. As a result, many threads are blocked.
> Although we only hit it once, but it is really a serious problem. To workarount it, we have to restart our application server. To our custom, it is not acceptable.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira