[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-02-04 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13570961#comment-13570961 ] Hudson commented on HBASE-7551: --- Integrated in HBase-0.94-security-on-Hadoop-23 #11 (See

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-24 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13561714#comment-13561714 ] Hudson commented on HBASE-7551: --- Integrated in HBase-0.94-security #96 (See

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553932#comment-13553932 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- +1 Lars. Actually seeing

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553980#comment-13553980 ] stack commented on HBASE-7551: -- Patch looks good to me. Not sure how useful the version

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553998#comment-13553998 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- bq.Not sure how useful the

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554004#comment-13554004 ] stack commented on HBASE-7551: -- bq. If the nodeChildrenEvent happens just after RS creates

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554020#comment-13554020 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- Yes. On creating node we

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Ted Yu (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554057#comment-13554057 ] Ted Yu commented on HBASE-7551: --- minor comment: {code} + for (String child :

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554120#comment-13554120 ] Lars Hofhansl commented on HBASE-7551: -- Thanks for the trunk patch Ted. The Stat

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554180#comment-13554180 ] Hadoop QA commented on HBASE-7551: -- {color:red}-1 overall{color}. Here are the results

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Ted Yu (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554245#comment-13554245 ] Ted Yu commented on HBASE-7551: --- Looped TestSplitTransactionOnCluster successfully 29 times

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554262#comment-13554262 ] Lars Hofhansl commented on HBASE-7551: -- This looks good. I'll commit this soon unless

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554267#comment-13554267 ] Hadoop QA commented on HBASE-7551: -- {color:red}-1 overall{color}. Here are the results

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554307#comment-13554307 ] Lars Hofhansl commented on HBASE-7551: -- TestLocalHBaseCluster passes locally for me.

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Ted Yu (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554335#comment-13554335 ] Ted Yu commented on HBASE-7551: --- 2 more rounds of test run with 29 iterations each passed.

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554415#comment-13554415 ] Lars Hofhansl commented on HBASE-7551: -- Anybody opposed to committing this? After

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554447#comment-13554447 ] Lars Hofhansl commented on HBASE-7551: -- Thanks Ram and Ted!

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Ted Yu (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554453#comment-13554453 ] Ted Yu commented on HBASE-7551: --- @Lars: Do you want to assign this issue to yourself ?

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554472#comment-13554472 ] Lars Hofhansl commented on HBASE-7551: -- Either way :) Ram did the work of tracking

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554513#comment-13554513 ] Hudson commented on HBASE-7551: --- Integrated in HBase-0.94 #736 (See

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554630#comment-13554630 ] Hudson commented on HBASE-7551: --- Integrated in HBase-TRUNK #3752 (See

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-15 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13554648#comment-13554648 ] Hudson commented on HBASE-7551: --- Integrated in HBase-TRUNK-on-Hadoop-2.0.0 #349 (See

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552547#comment-13552547 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- Will also check the impact

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread nkeywal (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552552#comment-13552552 ] nkeywal commented on HBASE-7551: Nice catch. bq. We should ensure that the transition

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552561#comment-13552561 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- Yes N. Something like the

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread nkeywal (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552574#comment-13552574 ] nkeywal commented on HBASE-7551: bq. May be as in AM assigning regions can we start with

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552892#comment-13552892 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- Some concerns here is, The

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552900#comment-13552900 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- @N Instead of starting

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread nkeywal (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552927#comment-13552927 ] nkeywal commented on HBASE-7551: If I not wrong, we would still have to manage that the

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13552980#comment-13552980 ] Lars Hofhansl commented on HBASE-7551: -- I have to study the code a bit more. But

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553035#comment-13553035 ] Lars Hofhansl commented on HBASE-7551: -- In fact looking at the code... Could somebody

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553331#comment-13553331 ] Lars Hofhansl commented on HBASE-7551: -- Is AssignmentManager.handleRegion idempotent?

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553363#comment-13553363 ] stack commented on HBASE-7551: -- bq. Could somebody explain to me why we create the node first

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553369#comment-13553369 ] Lars Hofhansl commented on HBASE-7551: -- I see. Hmm... I have a test patch, which just

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553472#comment-13553472 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- @N bq.While with the

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553501#comment-13553501 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- @Lars Change seems fine and

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread Lars Hofhansl (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553508#comment-13553508 ] Lars Hofhansl commented on HBASE-7551: -- Don't we have to set the watch in case we see

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread ramkrishna.s.vasudevan (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553531#comment-13553531 ] ramkrishna.s.vasudevan commented on HBASE-7551: --- Ok Lars. I will prepare a

[jira] [Commented] (HBASE-7551) nodeChildrenChange event may happen after the transition to RS_ZK_REGION_SPLITTING in SplitTransaction causing the SPLIT event to be missed in the master side.

2013-01-14 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-7551?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13553549#comment-13553549 ] stack commented on HBASE-7551: -- Lars patch looks pretty good. Looking forward to Ram patch.