hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: hadoop 2.0 dependency upgraded to 2.0.3-alpha
Date Fri, 22 Feb 2013 19:49:20 GMT
Trying out #2, I got:

testExcludeAllFromMinorCompaction(org.apache.hadoop.hbase.mapreduce.TestHFileOutputFormat)
 Time elapsed: 35.746 sec  <<< FAILURE!
java.lang.AssertionError
        at org.junit.Assert.fail(Assert.java:86)
        at org.junit.Assert.assertTrue(Assert.java:41)
        at org.junit.Assert.assertTrue(Assert.java:52)
        at
org.apache.hadoop.hbase.mapreduce.TestHFileOutputFormat.runIncrementalPELoad(TestHFileOutputFormat.java:468)
        at
org.apache.hadoop.hbase.mapreduce.TestHFileOutputFormat.testExcludeAllFromMinorCompaction(TestHFileOutputFormat.java:713)

Arun Murthy got to know this defect. Another 2.0 release is coming soon
(next week, hopefully).

I plan to wait for the new release.

Thanks

On Fri, Feb 22, 2013 at 11:39 AM, Ted Yu <yuzhihong@gmail.com> wrote:

> The root cause was the removal of capacity-scheduler.xml<http://svn.apache.org/viewvc/hadoop/common/branches/branch-2/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/conf/capacity-scheduler.xml>
from
> 2.0.3-alpha maven artifacts.
> This config file is needed to start CapacityScheduler which is the default.
>
> After talking to Sid, I got two suggestions:
> 1. embed minimal capacity-scheduler.xml<http://svn.apache.org/viewvc/hadoop/common/branches/branch-2/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/conf/capacity-scheduler.xml>
in
> our src/test/resources
> 2. use FiFo scheduler instead which doesn't require presence of config file
>
> I think #2, though temporary fix, is easy to change once a future 2.0
> release addresses this issue.
>
> Once consensus is reached, I will attach another patch to HBASE-7904
>
> Cheers
>
>
> On Fri, Feb 22, 2013 at 10:31 AM, Andrew Purtell <apurtell@apache.org>wrote:
>
>> For minicluster unit tests? In my opinion, no.
>>
>>
>> On Fri, Feb 22, 2013 at 10:10 AM, Ted Yu <yuzhihong@gmail.com> wrote:
>>
>> > I can try to find the proper usage.
>> >
>> > The question is: do we have to maintain compatibility with 2.0.2-alpha ?
>> >
>> > Thanks
>> >
>> > On Fri, Feb 22, 2013 at 10:07 AM, Andrew Purtell <apurtell@apache.org
>> > >wrote:
>> >
>> > > Sigh. Every release seems to break our uses of MiniMRCluster all over
>> > > again.
>> > >
>> > >
>> > > On Fri, Feb 22, 2013 at 10:04 AM, Ted Yu <yuzhihong@gmail.com> wrote:
>> > >
>> > > > I temporarily reverted this change due to:
>> > > >
>> > > > org.apache.hadoop.yarn.YarnException: Failed to Start
>> > > > org.apache.hadoop.mapred.MiniMRCluster
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.service.CompositeService.start(CompositeService.java:78)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.mapred.MiniMRClientClusterFactory.create(MiniMRClientClusterFactory.java:75)
>> > > >         at
>> > > >
>> org.apache.hadoop.mapred.MiniMRCluster.<init>(MiniMRCluster.java:180)
>> > > >         at
>> > > >
>> org.apache.hadoop.mapred.MiniMRCluster.<init>(MiniMRCluster.java:170)
>> > > >         at
>> > > >
>> org.apache.hadoop.mapred.MiniMRCluster.<init>(MiniMRCluster.java:162)
>> > > >         at
>> > > >
>> org.apache.hadoop.mapred.MiniMRCluster.<init>(MiniMRCluster.java:123)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.hbase.HBaseTestingUtility.startMiniMapReduceCluster(HBaseTestingUtility.java:1574)
>> > > > ...
>> > > > Caused by: org.apache.hadoop.yarn.YarnException:
>> > > > java.lang.IllegalStateException: Queue configuration missing child
>> > queue
>> > > > names for root
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.MiniYARNCluster$ResourceManagerWrapper.start(MiniYARNCluster.java:181)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.service.CompositeService.start(CompositeService.java:68)
>> > > >         ... 38 more
>> > > > Caused by: java.lang.IllegalStateException: Queue configuration
>> missing
>> > > > child queue names for root
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.parseQueue(CapacityScheduler.java:328)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.initializeQueues(CapacityScheduler.java:255)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.resourcemanager.scheduler.capacity.CapacityScheduler.reinitialize(CapacityScheduler.java:220)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.resourcemanager.ResourceManager.init(ResourceManager.java:226)
>> > > >         at
>> > > >
>> > > >
>> > >
>> >
>> org.apache.hadoop.yarn.server.MiniYARNCluster$ResourceManagerWrapper.start(MiniYARNCluster.java:161)
>> > > >
>> > > > Cheers
>> > > >
>> > > > On Fri, Feb 22, 2013 at 9:55 AM, Andrew Purtell <
>> apurtell@apache.org>
>> > > > wrote:
>> > > >
>> > > > > Have you tried running the mapreduce unit tests with this version
>> > Ted?
>> > > > Just
>> > > > > curious. They have given us grief in the past.
>> > > > >
>> > > > >
>> > > > > On Fri, Feb 22, 2013 at 9:35 AM, Ted Yu <yuzhihong@gmail.com>
>> wrote:
>> > > > >
>> > > > > > Hi,
>> > > > > > I have integrated the following JIRA to trunk:
>> > > > > > HBASE-7904 Upgrade hadoop 2.0 dependency to 2.0.3-alpha
>> > > > > >
>> > > > > > Earlier this week, Ramkrishna encountered some problem where
>> HBase
>> > > was
>> > > > > > built against 2.0.2-alpha but hadoop cluster deployment
was
>> using
>> > > > > > 2.0.3-alpha
>> > > > > >
>> > > > > > In case you use HBase against hadoop 2.0, please note this
>> change.
>> > > > > >
>> > > > > > Cheers
>> > > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > > --
>> > > > > Best regards,
>> > > > >
>> > > > >    - Andy
>> > > > >
>> > > > > Problems worthy of attack prove their worth by hitting back.
-
>> Piet
>> > > Hein
>> > > > > (via Tom White)
>> > > > >
>> > > >
>> > >
>> > >
>> > >
>> > > --
>> > > Best regards,
>> > >
>> > >    - Andy
>> > >
>> > > Problems worthy of attack prove their worth by hitting back. - Piet
>> Hein
>> > > (via Tom White)
>> > >
>> >
>>
>>
>>
>> --
>> Best regards,
>>
>>    - Andy
>>
>> Problems worthy of attack prove their worth by hitting back. - Piet Hein
>> (via Tom White)
>>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message