Clean up after process crash

[huard]

Description

The database entries (pywps_stored_requests, pywp_requests) for requests are not decremented when processes crash. After some time the server reaches the maximum number of parallel processes allowed and rejects all new requests, even though no process is active.

Is there a way to catch the crashes and remove those processes from the count to avoid this issue ?

[omerlevko]

hi,
@cehbrecht - do you have any idea about how to solve this? i will implement it

[cehbrecht]

@huard Probably this will be solved by the watchdog daemon in PR @497.