[jira] [Reopened] (HBASE-14802) Replaying server crash recovery procedure after a failover causes incorrect handling of deadservers

["stack (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-14802?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack reopened HBASE-14802:
---------------------------

Reverted from master branch. Since this went in, the related TestDeadServers is failing.

See https://builds.apache.org/job/HBase-Trunk_matrix/lastCompletedBuild/jdk=latest1.8,label=Hadoop/testReport/org.apache.hadoop.hbase.master/TestDeadServer/testCrashProcedureReplay/history/

Failed twice in master jdk8 builds (passed once).

Failed once on jdk7 https://builds.apache.org/job/HBase-Trunk_matrix/464/jdk=latest1.7,label=Hadoop/testReport/junit/org.apache.hadoop.hbase.master/TestDeadServer/testCrashProcedureReplay/history/

Seems to be just timing out the start up which is odd.

I have not reverted from branch-1 or branch-1.2 because have had a successful build in the latter. In the former it seemed to be something else.





> Replaying server crash recovery procedure after a failover causes incorrect handling of deadservers
> ---------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-14802
>                 URL: https://issues.apache.org/jira/browse/HBASE-14802
>             Project: HBase
>          Issue Type: Bug
>          Components: master
>    Affects Versions: 2.0.0, 1.2.0, 1.2.1
>            Reporter: Ashu Pachauri
>            Assignee: Ashu Pachauri
>             Fix For: 2.0.0, 1.2.0, 1.3.0
>
>         Attachments: HBASE-14802-1.patch, HBASE-14802-2.patch, HBASE-14802-3.patch, HBASE-14802.patch
>
>
> The way dead servers are processed is that a ServerCrashProcedure is launched for a server after it is added to the dead servers list. 
> Every time a server is added to the dead list, a counter "numProcessing" is incremented and it is decremented when a crash recovery procedure finishes. Since, adding a dead server and recovering it are two separate events, it can cause inconsistencies.
> If a master failover occurs in the middle of the crash recovery, the numProcessing counter resets but the ServerCrashProcedure is replayed by the new master. This causes the counter to go negative and makes the master think that dead servers are still in process of recovery. 
> This has ramifications on the balancer that the balancer ceases to run after such a failover.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)