[
https://issues.apache.org/jira/browse/ZOOKEEPER-785?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Henry Robinson updated ZOOKEEPER-785:
-------------------------------------
+1, this looks good (although I'd remove the 'out of place in this class'
comment now that you've moved it).
> Zookeeper 3.3.1 shouldn't infinite loop if someone creates a server.0 line
> ---------------------------------------------------------------------------
>
> Key: ZOOKEEPER-785
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-785
> Project: Zookeeper
> Issue Type: Bug
> Components: server
> Affects Versions: 3.3.1
> Environment: Tested in linux with a new jvm
> Reporter: Alex Newman
> Assignee: Patrick Hunt
> Fix For: 3.3.2, 3.4.0
>
> Attachments: ZOOKEEPER-785.patch, ZOOKEEPER-785.patch,
> ZOOKEEPER-785_2.patch, ZOOKEEPER-785_2.patch, ZOOKEEPER-785_2_br33.patch,
> ZOOKEEPER-785_2_br33.patch
>
>
> The following config causes an infinite loop
> [zoo.cfg]
> tickTime=2000
> dataDir=/var/zookeeper/
> clientPort=2181
> initLimit=10
> syncLimit=5
> server.0=localhost:2888:3888
> Output:
> 2010-06-01 16:20:32,471 - INFO [main:quorumpeerm...@119] - Starting quorum
> peer
> 2010-06-01 16:20:32,489 - INFO [main:nioservercnxn$fact...@143] - binding to
> port 0.0.0.0/0.0.0.0:2181
> 2010-06-01 16:20:32,504 - INFO [main:quorump...@818] - tickTime set to 2000
> 2010-06-01 16:20:32,504 - INFO [main:quorump...@829] - minSessionTimeout set
> to -1
> 2010-06-01 16:20:32,505 - INFO [main:quorump...@840] - maxSessionTimeout set
> to -1
> 2010-06-01 16:20:32,505 - INFO [main:quorump...@855] - initLimit set to 10
> 2010-06-01 16:20:32,526 - INFO [main:files...@82] - Reading snapshot
> /var/zookeeper/version-2/snapshot.c
> 2010-06-01 16:20:32,547 - INFO [Thread-1:quorumcnxmanager$liste...@436] - My
> election bind port: 3888
> 2010-06-01 16:20:32,554 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@620] - LOOKING
> 2010-06-01 16:20:32,556 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@649] - New election. My
> id = 0, Proposed zxid = 12
> 2010-06-01 16:20:32,558 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@689] - Notification: 0,
> 12, 1, 0, LOOKING, LOOKING, 0
> 2010-06-01 16:20:32,560 - WARN
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@623] - Unexpected exception
> java.lang.NullPointerException
> at
> org.apache.zookeeper.server.quorum.FastLeaderElection.totalOrderPredicate(FastLeaderElection.java:496)
> at
> org.apache.zookeeper.server.quorum.FastLeaderElection.lookForLeader(FastLeaderElection.java:709)
> at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:621)
> 2010-06-01 16:20:32,560 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@620] - LOOKING
> 2010-06-01 16:20:32,560 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@649] - New election. My
> id = 0, Proposed zxid = 12
> 2010-06-01 16:20:32,561 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@689] - Notification: 0,
> 12, 2, 0, LOOKING, LOOKING, 0
> 2010-06-01 16:20:32,561 - WARN
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@623] - Unexpected exception
> java.lang.NullPointerException
> at
> org.apache.zookeeper.server.quorum.FastLeaderElection.totalOrderPredicate(FastLeaderElection.java:496)
> at
> org.apache.zookeeper.server.quorum.FastLeaderElection.lookForLeader(FastLeaderElection.java:709)
> at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:621)
> 2010-06-01 16:20:32,561 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@620] - LOOKING
> 2010-06-01 16:20:32,562 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@649] - New election. My
> id = 0, Proposed zxid = 12
> 2010-06-01 16:20:32,562 - INFO
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:fastleaderelect...@689] - Notification: 0,
> 12, 3, 0, LOOKING, LOOKING, 0
> 2010-06-01 16:20:32,562 - WARN
> [QuorumPeer:/0:0:0:0:0:0:0:0:2181:quorump...@623] - Unexpected exception
> java.lang.NullPointerException
> Things like HBase require that the zookeeper servers be listed in the
> zoo.cfg. This is a bug on their part, but zookeeper shouldn't null pointer in
> a loop though.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
