[
https://issues.apache.org/jira/browse/STORM-3772?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Shiromani Bharti updated STORM-3772:
------------------------------------
Description:
With Pacemaker storage strategy,
When topology is submitted, there is entry created inside zNode
/strom/workerbeats. But same is not deleted when topology is killed. Because of
this we are ending up having all old topology entries in this zNode.
Since pacemaker is taking care of worker heartbeat, shouldn't we stop adding
entries in zNode /strom/workerbeats when topology submitted in case of
pacemaker storage strategy?
was:
With Pacemaker storage strategy,
When topology is submitted, there is entry created inside zNode
/strom/workerbeats. But same is not deleted when topology is killed. Because of
this we are ending up having all old topology entries in this zNode.
Since pacemaker is taking care of worker heartbeat, should be drop
> With pacemaker, zNode /strom/workerbeats entries are not removed once
> topology is killed
> ----------------------------------------------------------------------------------------
>
> Key: STORM-3772
> URL: https://issues.apache.org/jira/browse/STORM-3772
> Project: Apache Storm
> Issue Type: Bug
> Components: storm-server
> Affects Versions: 2.2.0
> Environment: Software platform
> Reporter: Shiromani Bharti
> Priority: Major
>
> With Pacemaker storage strategy,
> When topology is submitted, there is entry created inside zNode
> /strom/workerbeats. But same is not deleted when topology is killed. Because
> of this we are ending up having all old topology entries in this zNode.
> Since pacemaker is taking care of worker heartbeat, shouldn't we stop adding
> entries in zNode /strom/workerbeats when topology submitted in case of
> pacemaker storage strategy?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
