Re: [exim-dev] [Bug 1443] New: Mailq corruption

Top Page
Delete this message
Reply to this message
Author: Jasen Betts
Date:  
To: exim-dev
Subject: Re: [exim-dev] [Bug 1443] New: Mailq corruption
On 2014-02-12, Bernhard Eringa <mberinga@???> wrote:
> ------- You are receiving this mail because: -------
> You are on the CC list for the bug.
>
> http://bugs.exim.org/show_bug.cgi?id=1443
>            Summary: Mailq corruption
>            Product: Exim
>            Version: 4.72
>           Platform: x86
>         OS/Version: Linux
>             Status: NEW
>           Severity: bug
>           Priority: medium
>          Component: Queues
>         AssignedTo: nigel@???
>         ReportedBy: mberinga@???
>                 CC: exim-dev@???

>
>
> I've had exim go beserk twice, I'm not sure why, but the queue ends up with
> 1000+ bounce/retry emails, after I've run an email campaign to customers.


easiest cure may be to use a valid return address

> The only fix is killing the mailq, which is fine for me, but hard/impossible
> for any provider using Exim.
> It flogs our Bband connection so much we can't surf. I can't even get DNS
> lookups when it's in full flight.


That doesn't parse to anything I can grok - need more detail.

> I've changed the default retry string to only retry once after 0.5 hr, but
> haven't run any campaigns on this new config as yet, so will let you know if it
> happens again.
>
> I wonder if it isn't some kind of anti spam backlash from the recipients SMTP
> servers (Probably Australian SMTP or Exchange engines).


Anti-spam: I don't think so.

Once after half an hour may hurt if you hut greylisting.

> PS: Is there a way to force Debian to use a later version?, I'm currently using
> 4.72.


4.72 is standard in "Squeeze" (old-stable) upgrade to debian "wheezy"
(stable) for 4.80

--
Neither the pheasant plucker, nor the pheasant plucker's son.