[
http://opencast.jira.com/browse/MH-6944?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=26044#comment-26044
]
Christopher Brooks commented on MH-6944:
----------------------------------------
<cab938_> failed ingest
<adhocbot> stuff in ingest temp
<cab938_> network drop halfway
<jholtzman> no, there's no new workflow for a failed zip upload, just
(possibly) an old workflow for the scheduled recording, if there was a
scheduled event for this recording
<adhocbot> good question though chris :)
<cab938_> so if I upload a recording and it dies halfway through, it is
archived as failed, yes?
<cab938_> so a zip file is created
<jholtzman> the UI does something in these cases...
<adhocbot> ?
<jholtzman> no, nothing happens, the files sit in the ingest temp dir forever
<cab938_> ah, so it's just UI and the underlying workflow doesn't get failed?
<cab938_> ok, so it maybe is just partial ingest files taking up the space
forever
<jholtzman> right, because we don't know if, for instance, this is an offline
CA that's going to upload later
<cab938_> which means when a connection is broken on ingest it should clean
everything up right?
<cab938_> And by ingest I only mean addZippedMediaPackage
<cab938_> not the other forms of ingestion
<jholtzman> I'm not sure whether the code is informed of an IOException if the
connection is lost
<cab938_> since its an atomic operation
<cab938_> ok
<jholtzman> I'd have to watch it in a debugger
<cab938_> finally{} to clean files up?
<jholtzman> there's jetty, the osgi http service, a servlet, and a bunch of
filters between the JAX-RS endpoint and the wire
<jholtzman> lots of opportunities to hide this
<cab938_> either way adhocbot I'm not clear on why we are solving this, it's
not in the iter, I don't know if this is required for this release?
<cab938_> can I just add this chat to the ticket and move on?
<adhocbot> it's not
<jholtzman> this chat will be archived for all time? quick, think of something
witty
<cab938_> selective archiving :)
<cab938_> undoubtedly I'll edit in _micah describing his deep love for all
things flash
<adhocbot> yup. sorry about the distraction. I was just curious because in
the comment from long ago I had asked Micah if he had fixed it
> Capture Agent should respect retry config properties
> ----------------------------------------------------
>
> Key: MH-6944
> URL: http://opencast.jira.com/browse/MH-6944
> Project: Matterhorn Project
> Issue Type: Bug
> Components: Capture (Devices and Software)
> Affects Versions: 1.1
> Reporter: Ruediger Rolf
> Assignee: Micah Sutton
> Priority: Critical
> Fix For: 1.1 , 1.2
>
>
--
This message is automatically generated by JIRA.
-
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]
_______________________________________________
