hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicolas Liochon (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-11799) AsyncProcess can get locked when running PE
Date Mon, 25 Aug 2014 10:13:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-11799?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14108983#comment-14108983

Nicolas Liochon commented on HBASE-11799:

using exactly what Elliott wrote, all values (heap & so on) being the default.

On master, it does not work because the ZK thread embedded in the server stops (CancelledKey
exception not managed in ZK code)

On 0.98 top, for my first run, there are some nasty exceptions in the server, but the test
completed. I will run it a few more more times.
2014-08-25 11:57:37,967 WARN  [defaultRpcServer.handler=19,queue=1,port=40554] regionserver.MemStoreFlusher:
Memstore is above high water mark and block 71ms
2014-08-25 11:57:37,968 ERROR [MemStoreFlusher.1] regionserver.MemStoreFlusher: Cache flusher
failed for entry org.apache.hadoop.hbase.regionserver.MemStoreFlusher$WakeupFlushThread@291daa39
        at com.google.common.base.Preconditions.checkState(Preconditions.java:133)
        at org.apache.hadoop.hbase.regionserver.MemStoreFlusher.flushOneForGlobalPressure(MemStoreFlusher.java:208)
        at org.apache.hadoop.hbase.regionserver.MemStoreFlusher.access$500(MemStoreFlusher.java:66)
        at org.apache.hadoop.hbase.regionserver.MemStoreFlusher$FlushHandler.run(MemStoreFlusher.java:238)
        at java.lang.Thread.run(Thread.java:745)
2014-08-25 11:57:38,051 INFO  [MemStoreFlusher.0] regionserver.DefaultStoreFlusher: Flushed,
sequenceid=144, memsize=138.5 M, hasBloomFilter=true, into tmp file file:/tmp/hbase-liochon/hbase/data/default/TestTable/0434b24c6f20a4fbc259a3a54eafbdb0/.tmp/2ee7375469d644e4b905c745f2374053

> AsyncProcess can get locked when running PE
> -------------------------------------------
>                 Key: HBASE-11799
>                 URL: https://issues.apache.org/jira/browse/HBASE-11799
>             Project: HBase
>          Issue Type: Bug
>          Components: Client
>    Affects Versions: 0.98.7
>            Reporter: Elliott Clark
>         Attachments: jstack
> On a clean checkout of 0.98.
> {code}
> mvn clean package -DskipTests
> bin/start-hbase.sh
> bin/hbase pe --nomapred randomWrite  10 1000{code}
> Everything runs fine for a while then the client hangs.
> The region server thinks that all requests have completed successfully. The PE process
is all stuck on the final flush.

This message was sent by Atlassian JIRA

View raw message