On 20.9.2013, at 9.07, /dev/rob0 <rob0@gmx.co.uk> wrote:
This issue still occurs. It varies which of the four director instances gets it, but it seems that once one of them does, the only fix is to restart all four.
Do you see any other errors or warnings besides this? Are any of the directors restarted or do they for any reason get disconnected from each others before this happens? Are the clocks synchronized in all the directors? I'm guessing the directors did get restarted at some point and some of the other directors didn't notice this because of a bug:
http://hg.dovecot.org/dovecot-2.2/rev/b78c705bbb8d
I think with 3 directors this error wouldn't happen, because all directors have direct connections to each others and this bug doesn't affect them.
I'm fixing bugs for the rest of this week and I'll make new Dovecot release hopefully next week. And -ee release with this fix probably sooner. Maybe I'll even have time to go through this mailing list at some point. :)