[
https://issues.apache.org/jira/browse/PHOENIX-5051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16816669#comment-16816669
]
William Shen commented on PHOENIX-5051:
---------------------------------------
[~chengfan], I read through the code, do you think it is accurate to say that
the effect of the original bug is that number of row filtered is incorrectly
inflated by number of row scanned?
Every time we call org.apache.phoenix.monitoring.Metric#change, it changes the
metric by that amount, so in the original bug, the number of row filtered was
set incorrectly to _(number of row filtered + number of row scanned)_ and
number of row scanned was not set.
> ScanningResultIterator metric "RowsScanned" not set
> ---------------------------------------------------
>
> Key: PHOENIX-5051
> URL: https://issues.apache.org/jira/browse/PHOENIX-5051
> Project: Phoenix
> Issue Type: Bug
> Affects Versions: 4.14.1
> Reporter: Chen Feng
> Assignee: Cheng Fan
> Priority: Major
> Fix For: 4.15.0, 5.1.0
>
> Attachments: PHOENIX-5051-4.x-HBase-1.2-v1.patch,
> PHOENIX-5051-v1.patch, PHOENIX-5051-v2.patch
>
>
> in ScanningResultIterator, the metric "RowsScanned" was not set, while
> "RowsFiltered" was set twice
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
