> Shoult it not be 10000000 instead of 9855805? Yes that is very interesting. Could you please post the logs if you have them.
Enis On Fri, Mar 22, 2013 at 7:41 PM, Jean-Marc Spaggiari < jean-m...@spaggiari.org> wrote: > Hi Enis, > > What the expected result of the IntegrationTestLoadAndVerify? > > I got that on my side: > 13/03/22 22:36:34 INFO mapred.JobClient: > org.apache.hadoop.hbase.test.IntegrationTestLoadAndVerify$Counters > 13/03/22 22:36:34 INFO mapred.JobClient: ROWS_WRITTEN=0 > 13/03/22 22:36:34 INFO mapred.JobClient: REFERENCES_CHECKED=9855805 > > Shoult it not be 10000000 instead of 9855805? > > > 2013/3/22 Enis Söztutar <e...@hortonworks.com>: > > +1 > > > > - verified md5, asc > > - Build with hadoop 2 and hadoop 1 > > - Run local mode cluster, and these tests: > > - bin/hbase org.apache.hadoop.hbase.util.LoadTestTool -write 10:10:100 > > -num_keys 1000000 -read 100:30 > > - bin/hbase "org.apache.hadoop.hbase.test.IntegrationTestLoadAndVerify > > -Dloadmapper.backrefs=10 -Dloadmapper.map.tasks=10 > > -Dloadmapper.num_to_write=100000 -Dverify.reduce.tasks=1 > > -Dverify.scannercaching=10000 loadAndVerify" > > bin/hbase "org.apache.hadoop.hbase.test.IntegrationTestBigLinkedList > Loop > > 2 1 3000000 /tmp/biglinkedlist 1" > > > > Also based on Jeff's tests run on multi-node cluster. > > > > Enis > > > > > > > > On Fri, Mar 22, 2013 at 3:15 PM, Jeffrey Zhong <jzh...@hortonworks.com > >wrote: > > > >> +1 based on the test suit I ran, which includes some large tests such as > >> BigtopLoadAndVerify, LoatTestTool, GoraCI and TestAcidGuarantee. I've > >> updated the release spreadsheet. > >> > >> > >> Thanks, > >> -Jeffreyz > >> > >> On 3/22/13 7:33 AM, "Jean-Marc Spaggiari" <jean-m...@spaggiari.org> > wrote: > >> > >> >Hi, > >> > > >> >I deployed 0.94.6 on my 8 nodes "production" cluster. > >> > > >> >Ran some MR jobs (Row count and own MRs), everything went well. > >> >Ran the test suite (mvn test -P runAllTests), everything went well. > >> >(Tests run: 1325, Failures: 0, Errors: 0, Skipped: 13) > >> >Done many operatoins (c.f. google spreadsheet), everything went well. > >> > > >> >Only thing is we lost 1% in scan performances but this should not sink > >> >the release. > >> > > >> >So overall, I'm +1. > >> > > >> >JM > >> > > >> >2013/3/22 Jean-Marc Spaggiari <jean-m...@spaggiari.org>: > >> >> Sorry for the delay. > >> >> > >> >> I will test it deeply today and you will have my feedback before EOD. > >> >> > >> >> JM > >> >> > >> >> 2013/3/22 Stack <st...@duboce.net>: > >> >>> Hmm. Ignore my +1 from earlier this evening on RC1. > >> >>> > >> >>> Here is a +1 for RC2. > >> >>> > >> >>> Checked md5. > >> >>> > >> >>> Checked signing. > >> >>> > >> >>> Again, multipage book is missing but single page is here so thats > fine > >> >>>I'd > >> >>> say. The javadocs AND the src xref are in this version. > >> >>> > >> >>> Loaded ten million into standalone and scanned verified all loaded. > >> >>> > >> >>> Logs look good. > >> >>> > >> >>> St.Ack > >> >>> > >> >>> > >> >>> On Thu, Mar 21, 2013 at 9:49 PM, lars hofhansl <la...@apache.org> > >> >>>wrote: > >> >>> > >> >>>> Friendly reminder to please have a look (voting period ended > >> >>>>yesterday). > >> >>>> > >> >>>> I stress loaded 1bn 1k rows into a test cluster with 15 RS using > 150 > >> >>>> mappers, watched key metrics. Besides maxing out our 10ge TOR > switch > >> >>>>all > >> >>>> was good. > >> >>>> > >> >>>> -- Lars > >> >>>> > >> >>>> > >> >>>> > >> >>>> ________________________________ > >> >>>> From: lars hofhansl <la...@apache.org> > >> >>>> To: hbase-dev <dev@hbase.apache.org> > >> >>>> Sent: Thursday, March 14, 2013 6:24 PM > >> >>>> Subject: [ANNOUNCE] The 3rd hbase 0.94.6 release candidate is > >> >>>>available > >> >>>> for download > >> >>>> > >> >>>> The 3rd and hopefully last 0.94.6 RC is available for download at > >> >>>> http://people.apache.org/~larsh/hbase-0.94.6-rc2/ > >> >>>> > >> >>>> Signed with my code signing key: C7CFE328 > >> >>>> > >> >>>> 0.94.6 is a primarily a bug fix release. This RC has 55 issues > >> >>>>resolved > >> >>>> against it, contributed by 28 individuals. > >> >>>> These are mostly bug fixes. Don't be fooled by the comparatively > small > >> >>>> number of issues; one of them is a giant back port of the HBase > >> >>>>snapshots > >> >>>> feature. > >> >>>> RC2 also fixes a nasty replication/restart issue and also pulls in > all > >> >>>> fixed committed since the last RC. > >> >>>> > >> >>>> The full list of changes is available here: > >> >>>> https://issues.apache.org/jira/browse/HBASE/fixforversion/12323907 > >> >>>> > >> >>>> Please try out the RC, check out the doc, take it for a spin, etc, > and > >> >>>> vote +1/-1 by March 20th on whether we should release this as > 0.94.6. > >> >>>> > >> >>>> The release testing spreadsheet is available here: > >> >>>> > >> >>>> > >> >>>> > >> https://docs.google.com/spreadsheet/ccc?key=0AvLqcVIqNtlTdGkwN3NIeWw1bG > >> >>>>JyQkZJMHJVVTB3aHc#gid=0 > >> >>>> (if you test this release, please add your test to this > spreadsheet, > >> >>>>so > >> >>>> that we can gauge the coverage) > >> >>>> > >> >>>> Thanks. > >> >>>> > >> >>>> -- Lars > >> >>>> > >> > >> > >> >
