recover from a stuck queue.

[kanderson]

I have a user with something that must have been stuck in a queue somewhere. My bigger concern is that right now, the server has both a SR and LD sitting with about 30 thousand mesages in each. They're worthless messages, but is there a command similar to omtidyu that I can run against the different services? Otherwise it'll take days before this server catches back up.

Thanks in advance

[dkelly]

There isn't an omtidyu equivalent as such.

What we've done in the past is to do this:

Code: Select all

omstat -q router | cut -d" " -f 1 | while read msgid
do
    omstat -q router -j -R -m $msgid
done

Unfortunately, with 30K messages, this will take a long time. You can be more specific by grepping for subjects if you want to.

Alternatively, using omqdump to delete the messages is an option.

Cheers

Dave

[kanderson]

It'll be a while yet, I can take them out in the SR, so LD never needs to proces them. That'll mean it happens far faster than letting them actually get processed.

Thanks Dave.

This is awesome.

[mhanisch]

...but how did you find out?

The "-j -R " options to omstat are not documented in the manual page, but are very
useful to me (my local queue has > 160000 messages at the moment... don't ask).

Are there any other "secret" options?

[KiwiStag]

After an error in sendmail it stored all mail to /var/spool/mqueue
How can I get it to reprocess the queue to deliver the messages that never got processed by mailer=scalix_mime as they were only going mailer=relay
?

[Valerion]

sendmail -q -v should do the trick for you.