[
https://issues.apache.org/jira/browse/HBASE-24177?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17085611#comment-17085611
]
Gaurav Kanade commented on HBASE-24177:
---------------------------------------
Without this fix we were outputting the same value for data size of flush and
time taken. This fixes it. See log below and before and after screenshots
attached
2020-04-17 15:08:28,832 INFO
[rs(ramvasu-vm,16020,1587116254987)-flush-proc-pool9-t1] regionserver.HRegion:
Finished flush of *dataSize ~27 B*/27, heapSize ~344 B/344, currentSize=0 B/0
for b0745021e7535fe7a984eddded4347c9 in *400m*s, sequenceid=12, compaction
requested=false
> MetricsTable#updateFlushTime is wrong
> -------------------------------------
>
> Key: HBASE-24177
> URL: https://issues.apache.org/jira/browse/HBASE-24177
> Project: HBase
> Issue Type: Bug
> Components: metrics
> Affects Versions: 1.1.2
> Reporter: ramkrishna.s.vasudevan
> Assignee: Gaurav Kanade
> Priority: Minor
> Fix For: 3.0.0, 2.3.0, 2.1.10, 2.2.5
>
>
> MetricsRegionServer does an update on the MetricsRegionServerSource,
> MetricsTable etc.
> While doing updateFlushTime, the time taken for flush is rightly updated in
> the RegionServerSource but at the MetricsTable level we update the
> memstoresize instead of the time.
> This applies from 1.1 version onwards.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
