[jira] [Commented] (HBASE-16359) NullPointerException in RSRpcServices.openRegion()

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

    [ https://issues.apache.org/jira/browse/HBASE-16359?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15602205#comment-15602205 ] 

Andrew Purtell commented on HBASE-16359:
----------------------------------------

See also HBASE-16830, where this was re-identified as a problem in 1.3, with discussion at tail suggesting it should go back further. When committing to master and branch-1, please remember that we have other active code lines for releases:

branch-1.1: for 1.1.x
branch-1.2: for 1.2.x
branch-1.3: for 1.3.x
0.98: for 0.98.x

So should this change be committed back further? Let's do so. [~ted_yu] [~syuanjiang] [~mantonov] [~busbey] [~ndimiduk]

> NullPointerException in RSRpcServices.openRegion()
> --------------------------------------------------
>
>                 Key: HBASE-16359
>                 URL: https://issues.apache.org/jira/browse/HBASE-16359
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 1.1.2
>            Reporter: Ted Yu
>            Assignee: Ted Yu
>             Fix For: 2.0.0, 1.4.0
>
>         Attachments: 16359.addendum, 16359.v2.txt
>
>
> I was investigating why some region failed to move out of transition within timeout 120000ms and found the following in region server log:
> {code}
> 2016-08-04 09:19:52,616 INFO  [B.priority.fifo.QRpcServer.handler=12,queue=0,port=16020] regionserver.RSRpcServices: Open hbck_table_772674,,1470302211047.                                da859880bb51bc0fd25979798a96c444.
> 2016-08-04 09:19:52,620 ERROR [B.priority.fifo.QRpcServer.handler=12,queue=0,port=16020] ipc.RpcServer: Unexpected throwable object
> java.lang.NullPointerException
>   at org.apache.hadoop.hbase.regionserver.RSRpcServices.openRegion(RSRpcServices.java:1530)
>   at org.apache.hadoop.hbase.protobuf.generated.AdminProtos$AdminService$2.callBlockingMethod(AdminProtos.java:22737)
>   at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:2127)
>   at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:107)
>   at org.apache.hadoop.hbase.ipc.RpcExecutor.consumerLoop(RpcExecutor.java:133)
> {code}
> Here is related code - NPE was thrown from the last line:
> {code}
>         htd = htds.get(region.getTable());
>         if (htd == null) {
>           htd = regionServer.tableDescriptors.get(region.getTable());
>           htds.put(region.getTable(), htd);
>         }
> ...
>           if (region.isMetaRegion()) {
>             regionServer.service.submit(new OpenMetaHandler(
>               regionServer, regionServer, region, htd, masterSystemTime, coordination, ord));
>           } else {
>             regionServer.updateRegionFavoredNodesMapping(region.getEncodedName(),
>               regionOpenInfo.getFavoredNodesList());
>             if (htd.getPriority() >= HConstants.ADMIN_QOS || region.getTable().isSystemTable()) {
> {code}
> region.getTable() shouldn't be null since it is called via htds.get(region.getTable()) unconditionally.
> It seems htd was null.



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