Skip to content

RQ "leaks" worker registrations #55

Closed
nvie opened this Issue Apr 9, 2012 · 8 comments

5 participants

@nvie
Owner
nvie commented Apr 9, 2012

Workers cannot be unregistered upon cold shut down, since there is no hook for that, UNIX-wise. Therefore, worker keys in Redis won't be cleaned up.

Affected keys:

  • rq:worker:<uuid>
  • rq:workers (set members)

This bug is even more annoying due to #48.

@nvie
Owner
nvie commented Apr 9, 2012

The most reliable approach to solving this problem is to put Redis timeouts on the worker keys.

@passy
passy commented Apr 12, 2012

+1 on timeouts. I was confused when I took a look at rqinfo after a few restarts and saw a couple of dozen workers hanging around where I expected only one to be active.

@armonge
armonge commented Jun 8, 2012

This also affects when running under supervisor, with a configuration like the following rq fails to restart without manually deleting some keys

[program:rq]
command=rqworker -v --name=myrq
process_name=%(program_name)s
directory=/my/path
autostart=true
autorestart=true
stdout-logfile=/my/path/logs/rq.log
stderr_logfile=/my/path/logs/rq.log
@nvie
Owner
nvie commented Aug 5, 2012

Fixed in 0.3.0.

@selwin
Collaborator
selwin commented Aug 16, 2012

@nvie since this is fixed in 0.3.0, this issue should probably be closed

@nvie
Owner
nvie commented Aug 17, 2012

Fixed in 0.3.0.

@nvie nvie closed this Aug 17, 2012
@drewyeaton

I realize this issue is fixed currently, but what's the best way to clean up these leaked worker registrations?

@nvie
Owner
nvie commented Dec 18, 2012
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Something went wrong with that request. Please try again.