[
https://issues.apache.org/jira/browse/NIFI-1183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joseph Witt updated NIFI-1183:
------------------------------
Description:
The issue manifested itself with some failure in GetKafka. The thread dump
appears to be pointing to deadlock originated with some Zookeeper failure
The general background to this is that in various cases there is a very key
lock involved in ensuring the consistency of the nifi configuration. This is
evident in the pathological case of a processor which for example has an
OnScheduled method that never finishes. If the the user starts such a
processor it will never actually start and further if they then try to stop
that processor it will block indefinitely. The framework has to handle such
cases in a deterministic manner and without resorting to blocking because no
single bad processor should be able to bring down the house in such a case.
This is not only present in this suggested path. It would apply to others so
this needs holistic review and approach. Again, whenever this occurs it is a
buggy and broken processor. But the framework must be able to handle these
chaos monkey type processors in a deterministic manner.
was:The issue manifested itself with some failure in GetKafka. The thread
dump appears to be pointing to deadlock originated with some Zookeeper failure
> Failed to stop failing processor
> ---------------------------------
>
> Key: NIFI-1183
> URL: https://issues.apache.org/jira/browse/NIFI-1183
> Project: Apache NiFi
> Issue Type: Bug
> Components: Core Framework
> Affects Versions: 0.3.0
> Reporter: Oleg Zhurakousky
> Assignee: Oleg Zhurakousky
> Fix For: 0.5.0
>
> Attachments: thread-dump.txt
>
>
> The issue manifested itself with some failure in GetKafka. The thread dump
> appears to be pointing to deadlock originated with some Zookeeper failure
> The general background to this is that in various cases there is a very key
> lock involved in ensuring the consistency of the nifi configuration. This is
> evident in the pathological case of a processor which for example has an
> OnScheduled method that never finishes. If the the user starts such a
> processor it will never actually start and further if they then try to stop
> that processor it will block indefinitely. The framework has to handle such
> cases in a deterministic manner and without resorting to blocking because no
> single bad processor should be able to bring down the house in such a case.
> This is not only present in this suggested path. It would apply to others so
> this needs holistic review and approach. Again, whenever this occurs it is a
> buggy and broken processor. But the framework must be able to handle these
> chaos monkey type processors in a deterministic manner.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
