[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15627571#comment-15627571 ] Yu Li commented on HBASE-16931: --- Yes, let me update, thanks for the note sir. > Setting cell's seqId to

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15627566#comment-15627566 ] Mikhail Antonov commented on HBASE-16931: - affectVersion should also be all 1.1+ branches? >

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15627559#comment-15627559 ] Yu Li commented on HBASE-16931: --- Ok, just pushed into 1.3, thanks for the confirmation [~mantonov] and

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15627067#comment-15627067 ] Mikhail Antonov commented on HBASE-16931: - Yeah I don't think we've seen that in the workloads we

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15627052#comment-15627052 ] Gary Helmling commented on HBASE-16931: --- +1 for 1.3. Let's pull it in. > Setting cell's seqId to

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15625068#comment-15625068 ] Anoop Sam John commented on HBASE-16931: bq.We dont clone the seqId and all cloned cells will

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15621806#comment-15621806 ] Yu Li commented on HBASE-16931: --- No problem, will wait for your feedback then. And yes [~aoxiang] has added

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15621761#comment-15621761 ] Mikhail Antonov commented on HBASE-16931: - Sorry for late reply here [~carp84] and great

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15621730#comment-15621730 ] Mikhail Antonov commented on HBASE-16931: - [~anoopsamjohn] for that other possible bug in the

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15610615#comment-15610615 ] Yu Li commented on HBASE-16931: --- Ping [~mantonov], pending on your decision for branch-1.3 to close this

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15607278#comment-15607278 ] Yu Li commented on HBASE-16931: --- Skimmed log of failed post-commit check for trunk/branch-1/branch-1.1 and

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15607269#comment-15607269 ] Yu Li commented on HBASE-16931: --- bq. The choice is between this going into 1.3.0 or waiting for 1.3.1,

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15606042#comment-15606042 ] Hudson commented on HBASE-16931: SUCCESS: Integrated in Jenkins build HBase-1.2-JDK7 #55 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605931#comment-15605931 ] Hudson commented on HBASE-16931: SUCCESS: Integrated in Jenkins build HBase-1.2-JDK8 #50 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605663#comment-15605663 ] Hudson commented on HBASE-16931: FAILURE: Integrated in Jenkins build HBase-1.1-JDK7 #1804 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605578#comment-15605578 ] Hudson commented on HBASE-16931: FAILURE: Integrated in Jenkins build HBase-1.1-JDK8 #1888 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605384#comment-15605384 ] Sean Busbey commented on HBASE-16931: - The choice is between this going into 1.3.0 or waiting for

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605312#comment-15605312 ] Yu Li commented on HBASE-16931: --- Pushed into branch-1.1 and branch-1.2. Now waiting for the decision on

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15604418#comment-15604418 ] Yu Li commented on HBASE-16931: --- Thanks [~busbey] [~stack] and [~anoop.hbase] for the confirmation, will

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602755#comment-15602755 ] Hudson commented on HBASE-16931: FAILURE: Integrated in Jenkins build HBase-1.4 #493 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602741#comment-15602741 ] Hudson commented on HBASE-16931: FAILURE: Integrated in Jenkins build HBase-Trunk_matrix #1846 (See

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602719#comment-15602719 ] Anoop Sam John commented on HBASE-16931: bq.Does this patch cause some change in behavior or

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602715#comment-15602715 ] Anoop Sam John commented on HBASE-16931: bq.It looks like we get the sequenceid twice from a Cell

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602536#comment-15602536 ] Sean Busbey commented on HBASE-16931: - If the problem also exists in the earlier branch-1 release

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602481#comment-15602481 ] stack commented on HBASE-16931: --- Should we do another version of the patch after the hot fix has gone in?

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602459#comment-15602459 ] stack commented on HBASE-16931: --- [~aoxiang] Excellent debugging sir. > Setting cell's seqId to zero in

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602358#comment-15602358 ] Yu Li commented on HBASE-16931: --- Pushed commit to master branch and branch-1. AFAICS all branch-1.1+ have

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602322#comment-15602322 ] Yu Li commented on HBASE-16931: --- All timed out cases failed because of OOME (why so frequent OOME?...)

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602227#comment-15602227 ] Hadoop QA commented on HBASE-16931: --- | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602118#comment-15602118 ] Yu Li commented on HBASE-16931: --- Checking the UT result and all 3 cases failed because of OOME and

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15602098#comment-15602098 ] Yu Li commented on HBASE-16931: --- Ok, let's wait for the HadoopQA report. > Setting cell's seqId to zero in

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601913#comment-15601913 ] binlijin commented on HBASE-16931: -- done > Setting cell's seqId to zero in compaction flow might cause

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601630#comment-15601630 ] Hadoop QA commented on HBASE-16931: --- | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601619#comment-15601619 ] Hadoop QA commented on HBASE-16931: --- | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601550#comment-15601550 ] Yu Li commented on HBASE-16931: --- [~aoxiang] Please update the branch-1 patch to merge UT case into

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601542#comment-15601542 ] Yu Li commented on HBASE-16931: --- Oops, at a second glance, the previous HadoopQA report was about patch v3,

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601524#comment-15601524 ] Yu Li commented on HBASE-16931: --- The failed cases are known as always timeout and irrelative to the patch

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601478#comment-15601478 ] Hadoop QA commented on HBASE-16931: --- | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601457#comment-15601457 ] Anoop Sam John commented on HBASE-16931: +1. > Setting cell's seqId to zero in compaction flow

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601447#comment-15601447 ] Yu Li commented on HBASE-16931: --- +1 on patch v5, good job [~aoxiang] > Setting cell's seqId to zero in

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601403#comment-15601403 ] Ted Yu commented on HBASE-16931: Patch v5 looks good to me. > Setting cell's seqId to zero in compaction

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601370#comment-15601370 ] Yu Li commented on HBASE-16931: --- bq. I agree that avoid set and reset cell seqId is the best way. And clone

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601346#comment-15601346 ] binlijin commented on HBASE-16931: -- upload HBASE-16931_master_v4.patch this patch reset the local

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601331#comment-15601331 ] Anoop Sam John commented on HBASE-16931: I agree that avoid set and reset cell seqId is the best

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601315#comment-15601315 ] binlijin commented on HBASE-16931: -- I do not think clone is a good idea. > Setting cell's seqId to zero

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601307#comment-15601307 ] Yu Li commented on HBASE-16931: --- IMO changing the seqid of a cell which is referred to is dangerous, clone

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601303#comment-15601303 ] binlijin commented on HBASE-16931: -- What name do you think is better for this variable? > Setting

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601244#comment-15601244 ] Anoop Sam John commented on HBASE-16931: There is a small issue no? Within if we set

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601188#comment-15601188 ] Anoop Sam John commented on HBASE-16931: Sure.. Pls post a new patch for affected version. Once

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601166#comment-15601166 ] binlijin commented on HBASE-16931: -- Yes，i do not check the writer, thanks very much for the confirm. >

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601157#comment-15601157 ] binlijin commented on HBASE-16931: -- Yes, it is hard to debug... > Setting cell's seqId to zero in

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601153#comment-15601153 ] binlijin commented on HBASE-16931: -- branch-1 also have this problem. > Setting cell's seqId to zero in

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601100#comment-15601100 ] Anoop Sam John commented on HBASE-16931: U mean call that before calling beforeShipped *also*..

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601097#comment-15601097 ] ramkrishna.s.vasudevan commented on HBASE-16931: I think it will. Instead of doing every

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601092#comment-15601092 ] binlijin commented on HBASE-16931: -- And if we set back sequence id before ((ShipperListener)

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601080#comment-15601080 ] ramkrishna.s.vasudevan commented on HBASE-16931: Ok. So just saw that in Writer {code}

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601077#comment-15601077 ] ramkrishna.s.vasudevan commented on HBASE-16931: Its very difficult to debug this issue.

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601075#comment-15601075 ] ramkrishna.s.vasudevan commented on HBASE-16931: Checking this patch. Fix looks ok but

[jira] [Commented] (HBASE-16931) Setting cell's seqId to zero in compaction flow might cause RS down.

[ https://issues.apache.org/jira/browse/HBASE-16931?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15601057#comment-15601057 ] Anoop Sam John commented on HBASE-16931: I just changed the subject of the jira and upped the