I'm not sure exactly how to articulate this request, and maybe one of
the system architects can formulate it more precisely,
but it seems there needs to be some kind of backup plan or timer or
fuse to blow up jobs that have failed processing (for
whatever happens to be the problem of the day) to clear the job queue
and get these things to a final state.

I had a retract operation fail on me and now I'm looking at the
recordings listing and these jobs are listed as "Processing: Archived"
when clearly they are not processing, not going anywhere, and will
stay in that state forever until I do something (and I'm not sure what
that something is at the moment).

Whoever spawned these things ought to look at them and see if they
complete in some reasonable time, and if they go belly up ought to
nuke them
and let the system get on with its life.

$**+ happens for many unforseen reasons (bad code, bad data, acts of
God, etc.) and a robust system can deal with it at more than one
level.

Hank
_______________________________________________
Matterhorn-users mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn-users

Reply via email to