Project

General

Profile

Bug #4050

Factory should handle os.fork() failing

Added by Burt Holzman over 7 years ago. Updated about 1 year ago.

Status:
Assigned
Priority:
Low
Assignee:
Burt Holzman
Category:
Factory
Target version:
v3_7_x
Start date:
06/11/2013
Due date:
% Done:

0%

Estimated time:
First Occurred:
v2_7_1
Occurs In:
Stakeholders:
Duration:

Description

The GOC saw os.fork() fail because no more memory could be allocated.
If this happens, we should catch the OSError and do something.
(It would be good to figure out why it happened, too.)

History

#1 Updated by Igor Sfiligoi over 7 years ago

os.fork() failure may not necessarily be fatal.

We should be able to recover, especially if it is a group fork, since we can have an arbitrary number of them.

Reducing the max number automatically on failure may also be a good idea.

#2 Updated by Burt Holzman over 7 years ago

Parag had the same comment, but I think fork failing should be fatal -- we really can't do useful work at that point and the machine
is nearly out of memory.

Remember that the factory will restart the entryGroup when it exits, so the factory will keep chugging along. And restarting
the entryGroup should defragment the memory too (if that's what causes the PSS to increase over time).

#3 Updated by Parag Mhashilkar over 7 years ago

  • Target version changed from v2_7_x to v2_7_2

#4 Updated by Parag Mhashilkar about 7 years ago

  • Target version changed from v2_7_2 to v3_2_x

#5 Updated by Burt Holzman about 7 years ago

  • Priority changed from Normal to Low

#6 Updated by Marco Mambelli over 2 years ago

  • Target version changed from v3_2_x to v3_4_x

#7 Updated by Marco Mambelli about 2 years ago

  • Target version changed from v3_4_x to v3_5_x

#8 Updated by Marco Mambelli about 1 year ago

  • Target version changed from v3_5_x to v3_7_x

Also available in: Atom PDF