hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Gray (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HBASE-3180) Review periodic master logging, especially ServerManager once a minute
Date Tue, 02 Nov 2010 03:47:24 GMT

    [ https://issues.apache.org/jira/browse/HBASE-3180?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12927288#action_12927288
] 

Jonathan Gray commented on HBASE-3180:
--------------------------------------

It'd be a complete hack.  Would have to add a new argument into LB method which wouldn't make
much sense.  We should be seeing dead server processing regardless?

> Review periodic master logging, especially ServerManager once a minute
> ----------------------------------------------------------------------
>
>                 Key: HBASE-3180
>                 URL: https://issues.apache.org/jira/browse/HBASE-3180
>             Project: HBase
>          Issue Type: Improvement
>          Components: master
>    Affects Versions: 0.90.0
>            Reporter: Jonathan Gray
>            Assignee: Jonathan Gray
>             Fix For: 0.90.0
>
>         Attachments: HBASE-3180-v1.patch, HBASE-3180-v2.patch
>
>
> Not including HBASE-3179, the master logs of a steady-state cluster are something like:
> {noformat}
> 2010-10-30 10:03:59,876 INFO org.apache.hadoop.hbase.master.ServerManager: regionservers=86,
averageload=28.05
> 2010-10-30 10:04:59,886 INFO org.apache.hadoop.hbase.master.ServerManager: regionservers=86,
averageload=28.05
> 2010-10-30 10:05:59,894 INFO org.apache.hadoop.hbase.master.ServerManager: regionservers=86,
averageload=28.05
> 2010-10-30 10:06:59,904 INFO org.apache.hadoop.hbase.master.ServerManager: regionservers=86,
averageload=28.05
> 2010-10-30 10:07:59,914 INFO org.apache.hadoop.hbase.master.ServerManager: regionservers=86,
averageload=28.05
> 2010-10-30 10:08:05,941 DEBUG org.apache.hadoop.hbase.master.LoadBalancer: Skipping load
balancing.  servers=86 regions=2401 average=27.918604 mostloaded=28 leastloaded=28
> 2010-10-30 10:08:07,128 INFO org.apache.hadoop.hbase.master.CatalogJanitor: Scanned 2399
catalog row(s) and gc'd 0 unreferenced parent region(s)
> {noformat}
> Let's review these and see if there's anything we want to take out or add.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message