[
https://issues.apache.org/jira/browse/HBASE-10392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13881329#comment-13881329
]
Anoop Sam John commented on HBASE-10392:
----------------------------------------
{code}
- float globalMemstoreLimit =
conf.getFloat("hbase.regionserver.global.memstore.upperLimit", 0.4f);
- int gml = (int)(globalMemstoreLimit * CONVERT_TO_PERCENTAGE);
+ float globalMemstoreSize =
conf.getFloat("hbase.regionserver.global.memstore.size", 0.4f);
+ int gml = (int)(globalMemstoreSize * CONVERT_TO_PERCENTAGE);
{code}
I was/am worried abt this change where we look at the new config alone. What
if a user configure hbase.regionserver.global.memstore.upperLimit (old) to 0.5
and HConstants.HFILE_BLOCK_CACHE_SIZE_KEY to 0.4 ? This check will not find
out that right?
> Correct references to hbase.regionserver.global.memstore.upperLimit
> -------------------------------------------------------------------
>
> Key: HBASE-10392
> URL: https://issues.apache.org/jira/browse/HBASE-10392
> Project: HBase
> Issue Type: Bug
> Reporter: Nick Dimiduk
> Assignee: Nick Dimiduk
> Fix For: 0.99.0
>
> Attachments: HBASE-10392.0.patch, HBASE-10392.1.patch,
> HBASE-10392.2.patch
>
>
> As part of the awesome new HBASE-5349, a couple references to
> {{hbase.regionserver.global.memstore.upperLimit}} was missed. Clean those up
> to use the new {{hbase.regionserver.global.memstore.size}} instead.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
