On Sun, 22 May 2005, J. Kendzorra wrote:
>On Sunday 22 May 2005 05:34, Peter Stuge wrote:
>> Unfortunately this strace output isn't useful. This is the strace of
>> either tcpserver or bincimap-up and we need the strace from
>> bincimapd.
>Sorry; here's one from bincimapd (snippets):

So this could simply be a race condition where one client is deleting and 
recreating a mailbox, while another is polling its contents. If the 
mailbox is present (new/, cur/ and tmp/ exist) the very moment after, then 
that would be the problem.

I'm not quite sure how to deal with it. Binc fails to stat the folder 
because it's in an intermediary state. Binc can't try to "clean up" the 
mailbox when this happens, and it can't really "just ignore it" either. 
But perhaps it could wait a little and try again, treating it as a 
temporary failure.

Could you check if the mailbox is "complete" after the NOOP failed?

Andy :-)

--
Andreas Aardal Hanssen   | http://www.andreas.hanssen.name/gpg
Author of Binc IMAP      |  "It is better not to do something
http://www.bincimap.org/ |        than to do it poorly."

Reply via email to