Durga Prasad Malyala:
> Thank you for your reply Wietse,
> You are my Yoda Jedi Master and I have greatest regard for you - Sincerely.
>
> Ive seen the above link earlier too and am at my wits end on this. I
> was wondering if there is any issue in LMTP handing mail over to
> dovecot.
I suppose that Postfix is overwhelming Dovecot, or that Dovecot is
overwhelming the file system, or both.
Have you looked at Dovecot logs? If not, then you may be missing
important information.
Have you looked at file system performance indicators? If not, then
you may be missing important information.
No-one can help you you aren't willing to do some of the work.
Although Postfix and Dovecot are common, you are unlikely to fix
this with some cut and paste.
You can configure Postfix to make fewer Dovecot connections, for
example by reducing in master.cf the number of processes that can
talk to Dovecot (this change requires "postfix reload"). This will
slow things down a bit and move the bottleneck to the Postfix queue.
You really need to make sure that Dovecot is not logging errors,
and that your file system is performing properly.
Wietse
> I've mailed this to the Dovecot group as well but no-one is
> volunteering to get involved.
> This mail is to get responses if anyone has faced similar issues
> earlier. Since Dovecot is very common and maybe someone can give
> pointers.
>
> Rgds/DP
>
> On 4 May 2018 at 16:54, Wietse Venema <wie...@porcupine.org> wrote:
> > Durga Prasad Malyala:
> >> Hello all,
> >> I am seeing consistent delays in writing to disk (my System redhat 7.2
> >> using GFS2 file system cluster)
> >>
> >> May 4 10:03:34 mail1 postfix/lmtp[11662]: E4EB75048C19:
> >> to=<x...@xyz.com>, relay=mail.xyz.com[private/dovecot-lmtp], delay=50,
> >> delays=0.02/0/0/50, dsn=2.0.0, status=sent (250 2.0.0 <x...@xyz.com>
> >> IIt4Ejji61o3LgAAuUaIWw Saved)
> >
> > The 0.02/0/0/50 means that Postfix spends 50s waiting for DOVECOT
> > to accept the mail.
> >
> > Look in your DOVECOT logs for warnings or errors.
> >
> > The format of the "delays=a/b/c/d" logging is as follows:
> >
> > a = time from message arrival to last active queue entry
> > b = time from last active queue entry to connection setup
> > c = time in connection setup, including DNS, EHLO and STARTTLS
> > d = time in message transmission
> >
> > http://www.postfix.org/postconf.5.html#delay_logging_resolution_limit
> >
> > Wietse
>
