[
https://issues.apache.org/jira/browse/HBASE-2388?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12851246#action_12851246
]
stack commented on HBASE-2388:
------------------------------
+1 on commit. The troubleshooting should probably become an xdoc that is
checked in and rides along w/ the hbase version, but thats another project. I
thought the link in the message an auto-generated link but it looks like we
have an explicit anchor in the wiki named '9' so should be fine as long as
folks leave anchors as they are. Good stuff J-D.
> Give a very explicit message when we figure a big GC pause
> ----------------------------------------------------------
>
> Key: HBASE-2388
> URL: https://issues.apache.org/jira/browse/HBASE-2388
> Project: Hadoop HBase
> Issue Type: Improvement
> Affects Versions: 0.20.3
> Reporter: Jean-Daniel Cryans
> Assignee: Jean-Daniel Cryans
> Fix For: 0.20.4, 0.21.0
>
> Attachments: HBASE-2388.patch
>
>
> Our new users get confused when they see their region servers disappearing.
> If they are proactive enough, they even show us with some messages like:
> {code}
> 2010-03-28 20:24:27,439 WARN org.apache.hadoop.hbase.util.Sleeper: We slept
> 79410ms, ten times longer than scheduled: 5000
> 2010-03-28 20:24:27,439 WARN org.apache.hadoop.hbase.util.Sleeper: We slept
> 78781ms, ten times longer than scheduled: 3000
> {code}
> For us it's very obvious what just happened, for our users not at all. I
> think we could easily just create a very comprehensive entry about this issue
> in the troubleshooting page and put a link directly into the error message.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
