[Tilesathome] "done" down and "active" through the roof

[Dirk-Lüder Kreie]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Alex S. schrieb:
> Matthias Julius wrote:
>>> this was also partly due to the API downtime. I would estimate 10-15k
>>> jobs there to be "lost".
>> Actually, I just checked the go_nogo script and it says the queue
>> length is 3.50.  There simply are not more jobs processed because
>> there are no more to process.  This would mean that all of the 16.15k
>> jobs in "active" are actually "lost" and should be re_queued.
> 
> Undoubtedly, some of those have been manually re-queued already.
> 
> Does the server check for and remove entries from the 'active' list even 
> if another client uploads the requested tileset?  If it doesn't, it 
> should, else there will be some unnecessary re-rendering and processor 
> time that could go toward another worthy goal if an automated re-queue 
> is implemented.

The "active" jobs don't record user information so anyone uploading the
data for that tileset clears the entry on a FCFS basis.

Data for the same tileset that arrives later will update the map, but
will not show up "active" or "done", unless a new request has been
generated in the meantime of course.

- --

Dirk-Lüder "Deelkar" Kreie
Bremen - 53.0952°N 8.8652°E

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGk5cwFUbODdpRVDwRAiynAKDBRLEaIhU8imP6Z/xmivL1nWKnOQCfd7/N
kyAKDBDl6aW3Fm8KXRGN/jo=
=fDTF
-----END PGP SIGNATURE-----