Internet Mail gateway not processing messages - take 3

[easysoft]

Hi All,

We have a problem with the UNIX queue (Internet Mail gateway) on out Scalix 10.0.5 server. This problem keeps reappearing every so often so I'm inclined to think there is some other underlying cause.

Basically, every so often all outbound mail gets stuck in the UNIX queue (which is displayed as Internet Mail Gateway when you run a 'service scalix status').

Previously, 'mikethebike' suggested the following procedure to restart the queue :
1) omoff -d 0 -s unix
2) ps -ef | grep unix ; and then kill off all unix.out processes.
3) Use omqdump to move all the messages to the error queue
4) omon -s unix
5) omscan -A -q -f
6) Now use omqdump to move messages back in batches to the unix queue

This procedure has worked for me the last 3 or 4 times this happened, but now I just can't get the UNIX queue to start processing mail.

Does any one have any idea how to solve this problem - and prevent it reoccurring ?

Just as an indication, there are currently only 450 messages in the outbound queue so it can't be a case of the machine being overloaded.

Many thanks in advance,
Regards
Arif

[easysoft]

One difference between this time and previous times is that now when I move messages from the unix queue to the error queue using omqdump there are always 3 messages left in the unix that won't move. If I try to Zap them I get a Request TIME_OUT error. However, the omscan on the UNIX queue completes fine ... strange.

Its been nearly 2 days now and no mail is going out ... help !

Regards
Arif

[mikethebike]

Arif,

those three email are being processed and cannot be moved. I would assume you have two aux processes for unix.
You would need to stop the gateway and move them (omoff -d 0 -s unix, if that takes too long use "omreset -o off unix")

Mick

[easysoft]

Thanks for your reply Mike.

I ran an omscan -A -a -f and that seemed to clear up my problems. Mail started going out and then I started moving back emails from error in batches of 100. After the second batch it jammed up again. I ran the omscan again, and it started again. So now I'm adding the old mails back again in batches of 50 - so far so good. Just 120 left ...

There must be something that is causing this to keep happening ... I am now considering a fresh install of Scalix 11 and then importing all my user data into that ...

Many thanks all the same,
Regards
Arif

[chrisburatto]

Hi

I am kind of new to this, but we are having a similar problem. We have Scalix 11 installed for almost a year without problems. Last month, our outbound queue started building up. We kept receiving emails normally, but no email was being sent to the internet.

I tried several things and noticed that when you restart the whole server (shutdown -r now), when it comes back it empties the queue. It works fine for hours (I can't precise, but from 5 to 10 hours), then it starts holding outbound emails again. --- again, the queue is really small, with a few hundred messages (sometimes even 60 messages are held up)

This is really weird. Is it possible that there is some kind of message in loop causing this? We've been receiving lots of Spam (no SpamAssassin here yet), and thought maybe a wrongly addressed message could be causing a loop after infinite bouncings (just a really wild guess here). There could be also a 'zoombie' machine inside our network sending these emails? I am just trying to imagine external problems, not Scalix problems.

Please help! I am about to reinstall everything too, but if this is a bug or external issue, it might come back.

Thanks!

[JasonWarren]

Hi,
did this get resolved?

best regards,

Jason