You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geode.apache.org by "Anthony Baker (JIRA)" <ji...@apache.org> on 2017/02/16 16:11:42 UTC
[jira] [Updated] (GEODE-2463) Review Netty shutdown in
GeodeRedisServiceImpl
[ https://issues.apache.org/jira/browse/GEODE-2463?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Anthony Baker updated GEODE-2463:
---------------------------------
Fix Version/s: (was: 1.2.0)
> Review Netty shutdown in GeodeRedisServiceImpl
> ----------------------------------------------
>
> Key: GEODE-2463
> URL: https://issues.apache.org/jira/browse/GEODE-2463
> Project: Geode
> Issue Type: Sub-task
> Components: redis
> Reporter: Galen O'Sullivan
>
> Do this after [GEODE-2449].
> in {{GeodeRedisServiceImpl.stop()}}:
> The old code used to call {{syncUninterruptibly}} on the workerGroup, bossGroup, and serverChannel. However, the close method is likely being called as the result of a shutdown message on the channel future, and we can't call netty sync / await on an I/O thread. We saw deadlocks in testing. Look into this and make sure we're shutting down right.
> http://netty.io/wiki/index.html
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)