Hello,

  One of our small servers/sites used to (and very occasionally still
does) get this error, which was simply due to a "slow" database. 
Especially during backups (mysqldump) it would do that, but even in
normal use it would occasionally.  We updated hardware on the machine
(and maybe mysql versions, too?  Davel would know, as he did it...) and
it's almost completely gone.  But to help you may also be able to
optimize things better .. eg. check that your indexes are appropriate
and used, and maybe look into what specifically is slowing it down and
see if that can be improved (eg. things like the recent discussion in
dbmail-dev about a slow query in imap, and there was another similar
improvement for an inefficient UPDATE query in pop3 which I think still
needs to be put into svn (hint, hint, Paul  :) )).  Also if you can use
lmtp, even though the actual problem will still exist for you, the way
it handles errors/retries will be different and you won't get the
bounced error message (at least not the same message .. I think it'll
probably retry things silently until they work, but haven't tried
offhand).




> We've got a dbmail server with around the 200 domains configured on it.
> And some of our customers will receive the following error:
>
> <e-mail address>: Command time limit exceeded:
> "/usr/local/sbin/dbmail-smtp"
>
> After a reset of the server it will work again, but it wil return after
> a couple of weeks.

-- 
Jesse Norell
Kentec Communications, Inc.

Reply via email to