[
https://issues.apache.org/jira/browse/HBASE-10058?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13835206#comment-13835206
]
Lars Hofhansl commented on HBASE-10058:
---------------------------------------
In 0.94 the new test fails without HBASE-9915 (touches more blocks than
necessary) and passes with HBASE-9915 in place. Had we had a test like this in
the beginning this HBASE-9915 would not have gone unnoticed for a year.
I have no idea what TestBlocksScanned in trunk is supposed to test, but it sure
as hell does not test "blocks scanned". It doesn't count any block metrics in
trunk (maybe that got removed during the metric switch).
I'll make this 0.94 only for now.
Simple test-only change, I assume there are no objections.
> Test for HBASE-9915 (avoid reading index blocks)
> ------------------------------------------------
>
> Key: HBASE-10058
> URL: https://issues.apache.org/jira/browse/HBASE-10058
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Fix For: 0.94.15
>
> Attachments: 10058-0.94.txt
>
>
> Was just trying to write a unit test to make sure we won't regress on
> HBASE-9915 in the future and found the dataBlockReadCnt metric is not
> correctly updated for encoded blocks (type is ENCODED_DATA as opposed to
> DATA).
--
This message was sent by Atlassian JIRA
(v6.1#6144)
