Thanks Stack! I have created HBASE-18832 to track LTT exception. We can include the fix in next release. Everything else looks good.
Considering alpha release I vote +1. Thanks, Umesh On Fri, Sep 15, 2017 at 5:37 PM, Stack <st...@duboce.net> wrote: > On Fri, Sep 15, 2017 at 12:50 PM, Umesh Agashe <uaga...@cloudera.com> > wrote: > > > -1 > > > > > Thanks for taking a look Umesh. > > May I suggest that one of our many tools failing perhaps is not sufficient > reason to sink an alpha release. The barrier for alpha releases is allowed > to be lower given these releases will not see production. We have a > workaround in that the PerformanceEvaluation loading tool works. Folks > could use this instead? (It worked for me). > > Thanks Umesh. > > St.Ack > > > > > - Downloaded the src tarball > > - Verified checksums and signature (8ACC93D2). Passed. > > - Extracted it. Passed. > > - Buil from source. openjdk version "1.8.0_131". Passed. > > - Running unit tests. Ongoing > > - mvn apache-rat:check Passed. > > - Downloaded the bin tarball > > - Verified checksums and signature (8ACC93D2). Passed. > > - Extracted it. Passed. > > - Started local instance. Ran the "hbase shell" with some basic > table > > commands. Passed. > > - Nothing unusual in the logs. > > - Ran ltt with 10 million rows and 2 columns/ row. Failed with > > Exception: > > > > 2017-09-15 12:38:38,158 ERROR [main] util.AbstractHBaseTool: Error > running > > command-line tool > > java.lang.ClassCastException: > > org.apache.hadoop.hbase.client.ColumnFamilyDescriptorBuilder$ > > ModifyableColumnFamilyDescriptor > > cannot be cast to org.apache.hadoop.hbase.HColumnDescriptor > > at > > org.apache.hadoop.hbase.util.LoadTestTool.applyColumnFamilyOptions( > > LoadTestTool.java:265) > > at > > org.apache.hadoop.hbase.util.LoadTestTool.initTestTable( > > LoadTestTool.java:540) > > at > > org.apache.hadoop.hbase.util.LoadTestTool.loadTable( > LoadTestTool.java:567) > > at > > org.apache.hadoop.hbase.util.LoadTestTool.doWork(LoadTestTool.java:548) > > at > > org.apache.hadoop.hbase.util.AbstractHBaseTool.run( > > AbstractHBaseTool.java:154) > > at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:70) > > at > > org.apache.hadoop.hbase.util.AbstractHBaseTool.doStaticMain( > > AbstractHBaseTool.java:262) > > at > > org.apache.hadoop.hbase.util.LoadTestTool.main(LoadTestTool.java:793) > > > > On Fri, Sep 15, 2017 at 11:03 AM, Stack <st...@duboce.net> wrote: > > > > > The first release candidate for HBase 2.0.0-alpha-3 is up at: > > > > > > https://dist.apache.org/repos/dist/dev/hbase/hbase-2.0.0-alpha-3RC0/ > > > > > > Maven artifacts are available from a staging directory here: > > > > > > https://repository.apache.org/content/repositories/ > orgapachehbase-1175 > > > > > > All was signed with my key at y 8ACC93D2 > > > <http://pgp.mit.edu/pks/lookup?op=get&search=0x9816C7FC8ACC93D2> > > > > > > I tagged the RC as > > > 2.0.0-alpha-3RC0.2 (a5c8461ca87d6324f16ffd126b765146fdd5315a) > > > > > > hbase-2.0.0-alpha-3 is our third alpha release on our march toward an > > > hbase-2.0.0. It includes all that was in previous alphas (new > assignment > > > manager, offheap read/write path, in-memory compactions, etc.), but > had a > > > focus on polishing our public API: old API that had been deprecated > since > > > hbase-1.0.0 or before was purged and new API was added with sympathetic > > > deprecation of the previous. Along the Admin plane, incompatible > changes > > > were unavoidable; you will not be able to administer a hbase2 cluster > > using > > > an hbase1 client (see adjacent "[DISCUSS] hbase-2.0.0 compatibility > > > expectations" thread for discussion and see [1] for current list of > > > Incompatibles). > > > > > > What is here will be our public API for 2.0.0 unless we get pushback > from > > > our gracious downstreamers. Please try it out. Shout now if you find a > > > problem so we can fix it before we get to beta. > > > > > > Alpha-3 does not have the final version of our Coprocessor API. > Finishing > > > the Coprocessor API for hbase-2.0.0 is the topic of our last planned > > alpha, > > > 2.0.0-alpha-4. The Coprocessor API changes pretty radically in > > hbase-2.0.0 > > > (though Coprocessor Endpoints will continue to work across an upgrade). > > See > > > [2] for why and why it was unavoidable. Input now from Coprocessor API > > > users before alpha-4 would be especially effective (I've cc'd our > Phoenix > > > brothers and sisters toward this end). > > > > > > hbase-2.0.0-alpha-3RC0 is a rough cut ('alpha'), not-for-production > > preview > > > of what hbase-2.0.0 will look like. It is meant for devs and > > downstreamers > > > to test drive and flag us early if we messed up anything ahead of our > > > rolling GAs. > > > > > > The list of features addressed in 2.0.0 so far can be found here [3]. > > There > > > are about 2700+. The list of ~500 fixes in 2.0.0 exclusively can be > found > > > here [3]. > > > > > > I've updated our overview doc. on the state of 2.0.0 [6] but JIRA 2.0.0 > > > label [5] has undergone extensive weeding and presents a fairly good > > > picture on what is yet to do before we 2.0.0. Check it out. > > > > > > Please take it for a spin and vote on whether it ok to put out as our > > first > > > alpha (bar is low for an 'alpha'). Let the VOTE be open for 72 hours > > > (Sunday) > > > > > > Thanks, > > > St.Ack > > > > > > 1. Current list of Incompatibles: > > > https://docs.google.com/document/d/1WCsVlnHjJeKUcl7wHwqb4z9iEu_ > > > ktczrlKHK8N4SZzs/edit#heading=h.723jjn18p2jr > > > 2. Why CPs are Incompatible: > > > https://docs.google.com/document/d/1WCsVlnHjJeKUcl7wHwqb4z9iEu_ > > > ktczrlKHK8N4SZzs/edit#heading=h.9k7mjbauv0wj > > > 3. goo.gl/Gcrp4f > > > 4. goo.gl/6dPqzG > > > 5. https://issues.apache.org/jira/projects/HBASE/versions/12327188 > > > 6. > > > https://docs.google.com/document/d/1WCsVlnHjJeKUcl7wHwqb4z9iEu_ > > > ktczrlKHK8N4SZzs/ > > > > > >
