severity 420175 whishlist
merge 420175 511060
thanks

* 2009-01-07 06:23, Jim Studt wrote:
> Dovecot imapd terminates itself deliberately if the system clock moves
> backwards.  This is the same underlying issue as bug 420175, but that was
> a laptop after a suspend and was deemed unimportant because of that.

I prefer to merge the two bug reports, because the problem is exactly the
same.

> This occurs on a XEN based virtual private server at a commercial hosting 
> facility.
> Jan  7 04:54:49 xxx dovecot: Time just moved backwards by 81 seconds.
> This might cause a lot of problems, so I'll just kill myself now.
> http://wiki.dovecot.org/TimeMovedBackwards

Are you running ntp on your server? You should, it would prevent your clock
to be moved backwards by 81 seconds all together, and would probably fix
this behaviour.

> A timebomb like this renders dovecot unsuitable for use on servers as all
> email clients will be unable to connect to the mail server until an
> administrator discovers the problem and restarts the mail server.

To be honest, I run my mail servers under Xen, I run dovecot and I never
had such an issue. Again, I think that having the clock going backwards by
81 seconds is a problem and should be fixed, instead of considering dovecot
a time bomb.

I'm pretty sure that upstream won't fix this (as stated in the wiki page),
and I'm reluctant to add a wrapper script to simply restart dovecot in
these situations: you would never discover the real problem.

All in all, such a wrapper script is pretty easy to write if your server is
affected by clock moving too fast; we can probably provide one in the
package, as an example.

>   - warn administrators that they are installing a fragile solution during 
> install.

I don't think that dovecot should be considered a fragile solution.

Thanks,

-- 
Fabio Tranchitella                         http://www.kobold.it
Free Software Developer and Consultant     http://www.tranchitella.it
_____________________________________________________________________
1024D/7F961564, fpr 5465 6E69 E559 6466 BF3D 9F01 2BF8 EE2B 7F96 1564

Attachment: signature.asc
Description: Digital signature

Reply via email to