Re: [exim] Exim crashed after reboot

Philip Hazel wrote:
> On Wed, 10 Jan 2007, Adam Stephens wrote:
>
>
>> We have a mailstore which accepts mail via SMTP and delivers it to
>> Cyrus IMAP using the LMTP driver. The Exim version is 4.63 and it's
>> started like this:
>> exim -bd &
>> exim -q5m &
>>
>
> As a matter of interest, why not
>
> exim -bd -q15m
>
> ? That is, why two daemons instead of one? (And you don't need the '&'
> when starting a daemon; it disconnects itself from the terminal unless
> debugging is turned on.)
>
>
We usually do start a single daemon; the start script was changed on
this particular box because one of my colleagues wanted to be able to
stop the listener and queue runner processes separately.


>> We rebooted this server last week and exim failed to start up. It logged
>> this:
>>
>> 2007-01-02 17:52:57 exim 4.63 daemon started: pid=524, no queue at
>> startupruns, listening for SMTP on port 25 (IPv4)
>> 2007-01-02 17:52:57 exim 4.63 daemon started: pid=525, -q5m, not
>> listening for SMTP
>> 2007-01-02 17:52:58 1H1gZk-000154-44 ==
>> rtdjf@??? R=local_cyrus_user
>> T=local_cyrus_deliver defer (-1): Failed to connect to socket
>> /var/cyrus/imap/socket/lmtp for local_cyrus_deliver transport:
>> Connection refused
>> 2007-01-02 17:53:04 queue run: process 563 crashed with signal 15 while
>> delivering 1H1ga7-000187-7q
>>
>> Both exim processes die, and nothing further is logged until the server
>> is restarted.
>>
>
> If you can reproduce the effect, try adding -d to the exim commands to
> get debugging output (written to stderr). Without more information,
> there's not much more to say.
>
OK, thanks. I'll try to reproduce the problem on one of our test systems.

--
--------------------------------
Adam Stephens
Network Specialist - Email & DNS
adam.stephens@???