cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Russ Hatch (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-6943) very contentious writes result in performance degradation
Date Thu, 27 Mar 2014 18:54:14 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-6943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13949786#comment-13949786
] 

Russ Hatch commented on CASSANDRA-6943:
---------------------------------------

I did a couple of runs with no reproducing, and then got it again earlier than usual (about
800k operations in).

the output from stress looks like this when it's happening:
{noformat}
TimedOutException(acknowledged_by:0)
TimedOutException(acknowledged_by:0)
TimedOutException(acknowledged_by:0)
TimedOutException(acknowledged_by:0)
TimedOutException(acknowledged_by:0)
TimedOutException(acknowledged_by:0)
java.lang.RuntimeException: Timed out waiting for a timer thread - seems one got stuck
        at org.apache.cassandra.stress.util.Timing.snapInterval(Timing.java:58)
        at org.apache.cassandra.stress.util.Timing.snapInterval(Timing.java:83)
        at org.apache.cassandra.stress.StressMetrics.update(StressMetrics.java:121)
        at org.apache.cassandra.stress.StressMetrics.access$200(StressMetrics.java:35)
        at org.apache.cassandra.stress.StressMetrics$1.run(StressMetrics.java:70)
        at java.lang.Thread.run(Thread.java:744)
{noformat}

> very contentious writes result in performance degradation
> ---------------------------------------------------------
>
>                 Key: CASSANDRA-6943
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6943
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Russ Hatch
>         Attachments: screenshot.png, screenshot2.png
>
>
> Running performance scenarios I have seen this characteristic drop in performance happen
several times. A similar effect was reproduced in another test cluster. Operations eventually
come to a standstill.
> !screenshot.png!
> !screenshot2.png!



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message