[
http://opencast.jira.com/browse/MH-8081?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=28339#comment-28339
]
Adam McKenzie commented on MH-8081:
-----------------------------------
Hi Ruben,
Overloaded services are handled at service call time i.e. if I want to process
a segmentation and I ask a service to process one and it replies that it is
overloaded no big deal I move to the next service provider.
The service registry heartbeat is for cases when the server has crashed, the
network has crashed etc. that would make these calls impossible.
The way the code read before my patch was that an overloaded service returns a
value that isn't OK so the service is marked offline. Which is not correct.
The future work for this problem is to also make the heartbeat a little bit
more kind in turning servers to offline mode so that they have a chance to
recover if it was just a high load for example on the machine that caused it to
not return in the time that the service registry wanted.
Adam
> Distributed servers go offline in service registry after heatbeat check
> -----------------------------------------------------------------------
>
> Key: MH-8081
> URL: http://opencast.jira.com/browse/MH-8081
> Project: Matterhorn Project
> Issue Type: Bug
> Components: Administrative Tools
> Affects Versions: 1.2
> Reporter: Adam McKenzie
> Assignee: Greg Logan
> Priority: Release Blocker
> Fix For: 1.2.1
>
> Attachments: opencast.log, serviceRegistryPatch1.2
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
http://opencast.jira.com/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
_______________________________________________
Matterhorn mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn
To unsubscribe please email
[email protected]
_______________________________________________
