hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yu Li (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-15160) Put back HFile's HDFS op latency sampling code and add metrics for monitoring
Date Tue, 26 Apr 2016 03:36:13 GMT

    [ https://issues.apache.org/jira/browse/HBASE-15160?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15257501#comment-15257501
] 

Yu Li commented on HBASE-15160:
-------------------------------

Sorry for the late response [~enis], busy resolving some online issues recently...

bq. are you also seeing these counters get reset?
No, never observed counters reset. Also checked source of {{MutableHistogram}} (both master
and branch-1) and there should be no reset call to {{count}}. Could you tell more details
about "the histograms are reset"? Thanks.

bq. Did you try with block cache disabled
Yes, the test was against a table with BLOCKCACHE=>false. Actually, this is part of the
reason for the regression observed in HBASE-15619, you could find more details about the test
there.

bq. since even if argument pread=false, we might end up doing a pread if we cannot get the
lock
Oh yes, you are right, didn't quite notice this part... Then it's hard to decide whether to
sacrifice such accuracy for performance... Personally, maybe I'd still choose performance.

> Put back HFile's HDFS op latency sampling code and add metrics for monitoring
> -----------------------------------------------------------------------------
>
>                 Key: HBASE-15160
>                 URL: https://issues.apache.org/jira/browse/HBASE-15160
>             Project: HBase
>          Issue Type: Sub-task
>    Affects Versions: 2.0.0, 1.1.2
>            Reporter: Yu Li
>            Assignee: Yu Li
>         Attachments: HBASE-15160.patch, HBASE-15160_v2.patch, HBASE-15160_v3.patch, hbase-15160_v4.patch,
hbase-15160_v5.patch
>
>
> In HBASE-11586 all HDFS op latency sampling code, including fsReadLatency, fsPreadLatency
and fsWriteLatency, have been removed. There was some discussion about putting them back in
a new JIRA but never happened. According to our experience, these metrics are useful to judge
whether issue lies on HDFS when slow request occurs, so we propose to put them back in this
JIRA, and add the metrics for monitoring as well.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message