[
https://issues.apache.org/jira/browse/HBASE-854?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12627096#action_12627096
]
Jim Kellerman commented on HBASE-854:
-------------------------------------
This is very strange. The failures only occur on the second test case in a
Junit test in which we spin up multiple HBase and Hadoop clusters. But it
doesn't always fail - that's the baffling part. The following exception causes
the region server to abort, and because the client application cannot connect,
the test times out.
{code}
[junit] 2008-08-26 21:24:50,536 FATAL [RegionServer:0]
regionserver.HRegionServer(424): Unhandled exception. Aborting...
[junit] java.io.IOException: Call failed on local exception
[junit] at org.apache.hadoop.ipc.Client.call(Client.java:718)
[junit] at
org.apache.hadoop.hbase.ipc.HbaseRPC$Invoker.invoke(HbaseRPC.java:230)
[junit] at $Proxy8.getProtocolVersion(Unknown Source)
[junit] at
org.apache.hadoop.hbase.ipc.HbaseRPC.getProxy(HbaseRPC.java:340)
[junit] at
org.apache.hadoop.hbase.ipc.HbaseRPC.getProxy(HbaseRPC.java:327)
[junit] at
org.apache.hadoop.hbase.ipc.HbaseRPC.getProxy(HbaseRPC.java:364)
[junit] at
org.apache.hadoop.hbase.ipc.HbaseRPC.waitForProxy(HbaseRPC.java:302)
[junit] at
org.apache.hadoop.hbase.regionserver.HRegionServer.reportForDuty(HRegionServer.java:706)
[junit] at
org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:291)
[junit] at java.lang.Thread.run(Thread.java:619)
[junit] Caused by: java.io.EOFException
[junit] at java.io.DataInputStream.readInt(DataInputStream.java:375)
[junit] at
org.apache.hadoop.ipc.Client$Connection.receiveResponse(Client.java:499)
[junit] at org.apache.hadoop.ipc.Client$Connection.run(Client.java:441)
[junit] 2008-08-26 21:24:50,538 DEBUG [RegionServer:0]
hbase.RegionHistorian(316): Offlined
[junit] Exception in thread "RegionServer:0" java.lang.NullPointerException
[junit] at
org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:450)
[junit] at java.lang.Thread.run(Thread.java:619)
{code}
Now it could be related to the fact that the client cache is static in HbaseRPC:
{code}
private static ClientCache CLIENTS=new ClientCache();
{code}
But why don't all tests that have multiple test cases fail all the time on the
second cluster spin up?
> hbase-841 broke build on hudson?
> --------------------------------
>
> Key: HBASE-854
> URL: https://issues.apache.org/jira/browse/HBASE-854
> Project: Hadoop HBase
> Issue Type: Bug
> Reporter: stack
> Assignee: Jim Kellerman
> Priority: Blocker
> Fix For: 0.18.0
>
> Attachments: hbase-854.patch
>
>
> Jim, you want to take a look at it?
> 841 changed interfaces. Changed interfaces can make for odd issues like the
> hangs exhibited up on hudson (stuff is failing for me on my laptop since
> about the commit 841... timeouts. I don't have same issue on branch).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
