[
https://issues.apache.org/jira/browse/STORM-1376?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15046952#comment-15046952
]
Robert Joseph Evans commented on STORM-1376:
--------------------------------------------
[~knusbaum] and [~redsanket] can you two jump on this and take a look. This
should be simple to reproduce so we can see where the issues are comming from.
I suspect that it is probably in the blob store code as that is a big rewrite
to work with nimbus HA, where as the heartbeat code we have been running with
for a long time.
> ZK Becoming deadlocked with zookeeper_state_factory
> ---------------------------------------------------
>
> Key: STORM-1376
> URL: https://issues.apache.org/jira/browse/STORM-1376
> Project: Apache Storm
> Issue Type: Bug
> Components: storm-core
> Affects Versions: 0.11.0
> Reporter: Daniel Schonfeld
>
> Since the introduction of blobstore and pacemaker we've noticed that when
> using nimbus with the new zookeeper_state_factory backing cluster state
> module, some of our ZK nodes become unresponsive and show and increasing
> amounts of outstanding requests (STAT 4-letter command).
> Terminating storm supervisors and nimbus usually gets zookeeper to realize
> after a few minutes those connections are dead and to become responsive
> again. In some extreme cases we have to kill that ZK nodes and bring it back
> up.
> Our topologies ran across ~10 supervisor nodes with each having about
> ~400-500 executors.
> I mention the amount of executors cause I am not sure if someone made each
> executor by mistake start sending heartbeats instead of each worker and that
> might possibly be the reason for this slow down.
> Final note. If someone can jot a few ideas of why this might be happening
> i'd be more than happy to dig further in the storm code and submit a PR
> myself. But I need some hint or direction of where to go with this...
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
