Radu Greab wrote:
dormando wrote:
Except the delete job is never getting that message, and the procmanager
code prevents the job monitors subsequent broadcasts from being sent to
the deleter, since the status hasn't changed.
The symptom of this is any deletes destined for those hosts get cycled
through file_to_delete_later and back again every 600 seconds.
I've seen this and sent a patch which was applied as change #1120. Is
that change included into the code you use?
Actually, that's probably why I can't find the bug in the code :P I
missed that patchset a while ago. I've been analyzing the code off of
trunk, but the bug's affected by production (running a previous release).
We need to cut another release of MogileFS. There're enough changes to
trunk that I can't even remember what's been fixed anymore :\
Brad; ping? :)
-Dormando
