hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-15614) Report metrics from JvmPauseMonitor
Date Fri, 08 Apr 2016 04:10:25 GMT

    [ https://issues.apache.org/jira/browse/HBASE-15614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15231598#comment-15231598
] 

stack commented on HBASE-15614:
-------------------------------

No magic to JvmPauseMonitor. If it is seeing pauses that are not in GC, that is interesting.
On article, how you think blocked i/o would manifest in the GC log? Wouldn't the timestamps
be just off? Just trying to avoid double metric on same base phenomenon.

You running an hbase cluster now [~ndimiduk] If so, i defer to your operator self... don't
mind my dev BS.

> Report metrics from JvmPauseMonitor
> -----------------------------------
>
>                 Key: HBASE-15614
>                 URL: https://issues.apache.org/jira/browse/HBASE-15614
>             Project: HBase
>          Issue Type: Improvement
>          Components: metrics, regionserver
>            Reporter: Nick Dimiduk
>
> We have {{JvmPauseMonitor}} for detecting JVM pauses; pauses are logged at WARN. Would
also be good to expose this information on a dashboard via metrics system -- make it easier
to get this info off the host and into a central location for the operator.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message