Re: FATAL: hudson.remoting.RequestAbortedException: java.io.IOException: Unexpected termination of the channel

To me, the issue is on hadoop6, while hadoop1, hadoop2 seem to be ok? On Thu, Sep 6, 2012 at 7:27 AM, Stack wrote: > On Wed, Sep 5, 2012 at 5:28 PM, Jonathan Hsieh wrote: > > I'm pretty convinced this was an environmental issue -- things seem well > > enough on some our internal builds of apach

Jenkins build is back to normal : HBase-0.92 #559

See

Re: Hbase Assignments in trunk.

On the Async vs. sync: there are 3 different ways to write multiple znodes in ZK, and huge differences in the performances between them: 1) for loop sync 2) for loop async 3) multi Async will be 20 to 100 times faster than sync. multi will be 2 to 4 times faster than async (that is, 80 to 400 tim

Re: Hbase Assignments in trunk.

On Wed, Sep 5, 2012 at 4:08 PM, Stack wrote: > On Wed, Sep 5, 2012 at 12:38 PM, Jonathan Hsieh wrote: > ... > > We've also talked about defining design and code invariants -- here's the > > one that I've gotten so far: (We can pull up more from discussion) > > > > * ZK state should transient (

Re: Hbase Assignments in trunk.

IMO, moving to new ZK seems to makes sense for HBase trunk. Jon. On Thu, Sep 6, 2012 at 12:53 AM, n keywal wrote: > On the Async vs. sync: there are 3 different ways to write multiple znodes > in ZK, and huge differences in the performances between them: > > 1) for loop sync > 2) for loop asyn

Re: Thoughts about large feature dev branches

On Wed, Sep 5, 2012 at 3:49 PM, Stack wrote: > On Wed, Sep 5, 2012 at 11:38 AM, Jonathan Hsieh > wrote:> Where the folks working on those features committers at the time? > What do > > we do for contributions from folks who aren't committers yet? > > > > Yes. > > For folks not yet committers, l

Re: Thoughts about large feature dev branches

On Wed, Sep 5, 2012 at 4:43 PM, Jesse Yates wrote: > On Wed, Sep 5, 2012 at 3:58 PM, Elliott Clark >wrote: > > > - I would suggest that since hbase's code base moves so rapidly, a > > rebased branch should probably be a requirement before merging. > > Otherwise the merge will get pretty interest

Re: Thoughts about large feature dev branches

On Wed, Sep 5, 2012 at 11:38 AM, Jonathan Hsieh wrote: > Though I wasn't around yet, it seems like this is what we did for > coprocs/security, probably for the 0.90 master. > http://search-hadoop.com/m/byzZYZMktx1/hbase+windows&subj=Re+Proposed+feature+branch+for+HBase+security Though we had mult

Re: Hbase Assignments in trunk.

There's a discussion on the ZK mailing list about releasing ZK 3.4.4, which will have multi and some other fixes. Once that is out, we can move to that on trunk. That will also help with one of the replication patches that Himanshu currently has pending, which relies on multi. - Dave On Thu, Se

Porting policy from 0.94 to 0.92

Hi, As release manager of 0.92.x, I want to poll your opinion on porting policy from 0.94 to 0.92 Earlier there was email thread 'Porting policy from 0.96+ to 0.94'. >From that thread, I think there was green light for porting non invasive, small new features if some committer/party shows interest

Re: Porting policy from 0.94 to 0.92

I think there's a distinction from porting to the newest release (0.94.x, which is not yet really widely deployed, though starting to get there) compared to porting to a one-old release (0.92.x). I think we should be especially conservative about adding even non-invasive features to "stable" branch

Re: Porting policy from 0.94 to 0.92

Thanks for your comment, Todd. I have observed some flexibility in this regard from, say cdh3u4a: http://archive.cloudera.com/ebay/cdh3u4a-rc1/cdh/3/hbase-0.92.1+82.releasenotes.html Cheers On Thu, Sep 6, 2012 at 10:50 AM, Todd Lipcon wrote: > I think there's a distinction from porting to the

Build failed in Jenkins: HBase-TRUNK #3310

See Changes: [tedyu] HBASE-6659 Port HBASE-6508 Filter out edits at log split time [jxiang] HBASE-6715 TestFromClientSide.testCacheOnWriteEvictOnClose is flaky [stack] HBASE-6352 Add copy method in Bytes -

Re: Porting policy from 0.94 to 0.92

Hi Ted, With my Apache hat on: the policies used by distribution vendors may differ from the policies in the project itself. I think the Apache project should be concerned with pushing the ball forward on new releases and not on backporting features to older branches. Vendors have internal QA team

Build failed in Jenkins: HBase-0.94 #451

See Changes: [jxiang] HBASE-6715 TestFromClientSide.testCacheOnWriteEvictOnClose is flaky -- [...truncated 581 lines...] Running org.apache.hadoop.hbase.coprocessor.TestClassLoading Tests run: 7, Failu

Re: Porting policy from 0.94 to 0.92

On Thu, Sep 6, 2012 at 11:03 AM, Ted Yu wrote: > Thanks for your comment, Todd. > > I have observed some flexibility in this regard from, say cdh3u4a: > > http://archive.cloudera.com/ebay/cdh3u4a-rc1/cdh/3/hbase-0.92.1+82.releasenotes.html > Irrelevant. Off-topic. Probably info. that doesn't be

Re: Porting policy from 0.94 to 0.92

The same principal should guide 0.94.x, right ? On Thu, Sep 6, 2012 at 12:31 PM, Stack wrote: > On Thu, Sep 6, 2012 at 11:03 AM, Ted Yu wrote: > > Thanks for your comment, Todd. > > > > I have observed some flexibility in this regard from, say cdh3u4a: > > > > > http://archive.cloudera.com/ebay

Re: Porting policy from 0.94 to 0.92

On Thu, Sep 6, 2012 at 12:32 PM, Ted Yu wrote: > The same principal should guide 0.94.x, right ? > No. Reread Todd's first note on difference between 0.92 and 0.94 (one is 'old', other is 'current') St.Ack

Re: Porting policy from 0.94 to 0.92

On Thu, Sep 6, 2012 at 12:32 PM, Ted Yu wrote: > The same principal should guide 0.94.x, right ? Since 0.96.0 hasn't been released yet, I'm more OK with putting minor features in 0.94.x, considering it is the "bleeding edge" release. It would be nice to formalize some of these rules/guidelines/w

Re: Thoughts about large feature dev branches

On Thu, Sep 6, 2012 at 3:33 AM, Jonathan Hsieh wrote: >> For folks not yet committers, lets look at them. If they are working >> on big features for HBase, they probably should be committers? >> >> This is probably case by case. > Agree St.Ack

Re: Hbase Assignments in trunk.

On Thu, Sep 6, 2012 at 3:16 AM, Jonathan Hsieh wrote: > On Wed, Sep 5, 2012 at 4:08 PM, Stack wrote: > >> On Wed, Sep 5, 2012 at 12:38 PM, Jonathan Hsieh wrote: ... >> We should post these invariants somewhere? In dev section of refguide? >> >> We should definitely put this in the javadoc. May

Build failed in Jenkins: HBase-TRUNK #3311

See Changes: [stack] Add an invariants section to the developer docs; added the one about no permanent data in zk -- [...truncated 2640 lines...] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Ti

Build failed in Jenkins: HBase-TRUNK-on-Hadoop-2.0.0 #163

See Changes: [stack] Add an invariants section to the developer docs; added the one about no permanent data in zk [tedyu] HBASE-6659 Port HBASE-6508 Filter out edits at log split time [jxiang] HBASE-6715 TestFromClientSide

Unnecessary file copying during the bulkload: should we backport the fix in 0.96?

In HBase 0.94, bulkload would always copy the files: // Move the file if it's on another filesystem FileSystem srcFs = srcPath.getFileSystem(conf); if (!srcFs.equals(fs)) { LOG.info("File " + srcPath + " on different filesystem than " + "destination store - moving to this filesystem.");

Re: Unnecessary file copying during the bulkload: should we backport the fix in 0.96?

On Thu, Sep 6, 2012 at 5:30 PM, pig user wrote: > In HBase 0.94, bulkload would always copy the files: > > // Move the file if it's on another filesystem > FileSystem srcFs = srcPath.getFileSystem(conf); > if (!srcFs.equals(fs)) { >LOG.info("File " + srcPath + " on different filesystem than "

Re: Unnecessary file copying during the bulkload: should we backport the fix in 0.96?

The fix, HBASE-6529, would be in 0.94.2 Expect 0.94.2 RC0 next week. Thanks On Thu, Sep 6, 2012 at 5:30 PM, pig user wrote: > In HBase 0.94, bulkload would always copy the files: > > // Move the file if it's on another filesystem > FileSystem srcFs = srcPath.getFileSystem(conf); > if (!srcFs.e

Build failed in Jenkins: HBase-TRUNK #3312

See Changes: [stack] HBASE-6286 Upgrade maven-compiler-plugin to 2.5.1 [stack] HBASE-6707 TEST org.apache.hadoop.hbase.backup.example.TestZooKeeperTableArchiveClient.testMultipleTables flaps -- [.

Build failed in Jenkins: HBase-0.94 #452

See Changes: [stack] HBASE-6286 Upgrade maven-compiler-plugin to 2.5.1 -- [...truncated 588 lines...] Running org.apache.hadoop.hbase.io.encoding.TestChangingEncoding Tests run: 4, Failures: 0, Errors:

Build failed in Jenkins: HBase-TRUNK #3313

See Changes: [stack] HBASE-6706 Remove TotalOrderPartitioner [stack] HBASE-6707 TEST org.apache.hadoop.hbase.backup.example.TestZooKeeperTableArchiveClient.testMultipleTables flaps; REVERT -- [...