[Netarchivesuite-users] Upload of files in oldjobs

[aponb at gmx.at]

Thanks, for that information.
Yes I would need it for getting a job completed, but I didn't know that 
I can also use an Harvester Instance which have a differnt priority. And 
you are right. I have of course a free harvester with a high priority 
running on my machines.

Thanks for you help
a.

> If your job has been stopped "in the right way" - that is, the job finished (could be failed) but reported its state back to the GUI (database) as well as statistics per domain you don't need to move the job-directory from oldjobs back to a harvester instance. That workaround is only needed to get a job completed.
>
> If your problem is only a missing upload you should definately use the UploadTool - much easier and don't require any workarounds.
>
> For the complete workaround (jobs that have not reported back to the system) you need a free HarvesterApplication instance. This works in Netarchive.dk because we have both HIGH_PRIORITY and LOW_PRIORITY harvesters running on all machines - so there is almost always a free HIGH_PRIORITY harvester to use (even for jobs originally run by a LOW_PRIORITY harvester - this does not matter in this case).
>
> So if you don't have a free harvester you could maybe setup a HIGH_PRIORITY harvester on the machine and use that ?
>
> best
> Bjarne Andersen
> ________________________________________
> Fra: netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users> [netarchivesuite-users-bounces at lists.gforge.statsbiblioteket.dk <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users>] P&#229; vegne af aponb at gmx.at <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users> [aponb at gmx.at <https://lists.gforge.statsbiblioteket.dk/mailman/listinfo/netarchivesuite-users>]
> Sendt: 16. september 2009 09:02
> Til: NetarchiveSuite-users
> Emne: [Netarchivesuite-users] Upload of files in oldjobs
>
> After my Problems with my JMS-Queue there is still one not uploaded
> arc-File in the oldjobs directory of one crawler. The procedure for
> uploading that is to move the whole directory of the jobs (i.e.
> 2860_1253011869233) to the harvester instance of that machine - and then
> to stop and start that instance. And now my question. How can I do that
> when there are jobs for that harvester in the queue. How can I say that
> the system should give that special harvester instance another job, so
> that I can move my job and restart instance. After that processing I
> would like to have that the system is scheduling jobs again for that
> instance. Could you tell me how I can do that? Is using the Upload-Tool
> the right way?
>
> Regards
> a.