Pushing queued emails

Question

While configuring (and perhaps later in the "real life") every now and then I see messages stuck in a working or held queue (as seen with imsimta qm dir; imsimta qm held).
Quite often this is for a reason, i.e. the relay-out daemon was down (restarting) at the moment a message was trying to relay. The problem is, I can't always easily push the mails to go on after the cause is fixed. It can sit in the queue for hours, sometimes even after a full restart of the messaging services.
While "imsimta run tcp_local" sometimes helps, it takes surprisingly long to run (i.e. 2-3 minutes for a queue with a single stuck message).
What I would like to achieve is something like sendmail's "-qI" flag (perhaps coupled with a fake hoststatus directory path) which tries to immediately send all messages from the queue and ignores previous failures and SMTP retry timeouts which they may have caused.
I did not yet find a correct non-disruptive procedure in the docs/blogs/forums/wikis, perhaps I need some pointing help from the gurus? ;)
Another inconvenience concerns the HELD messages. After I fixed the reason that got several dozen messages into the HELD status, I had to "release" them one-by-one from the interactive mode. In fact, then I hacked up a small script to release a hardcoded amount of messages at once, but there certainly must be a more correct and elegant method:
( echo "dir"; A=1; while [ $A -le 35 ]; do echo release $A; `echo "$A+1" | bc`; done ) | imsimta qm

Answer 1

JimKlimov wrote:
Quite often this is for a reason, i.e. the relay-out daemon was down (restarting) at the moment a message was trying to relay. The problem is, I can't always easily push the mails to go on after the cause is fixed. It can sit in the queue for hours, sometimes even after a full restart of the messaging services.You can use the ETRN command to push emails destined for a specific host in this situation e.g.
bash-2.05$ telnet mta.aus.sun.com 25
Trying 1.2.3.4...
Connected to mta.aus.sun.com.
Escape character is '^]'.
220 mta.aus.sun.com -- Server ESMTP (Sun Java(tm) System Messaging Server 6.3-7.01 (built May 14 2008; 32bit))
etrn relay.aus.sun.com
250 2.0.0 Delivery started on channel.
quit
221 2.3.0 Bye received. Goodbye.This will only re-try those emails which are destined for the specified host rather then re-attempting delivery for all emails in the channel.
While "imsimta run tcp_local" sometimes helps, it takes surprisingly long to run (i.e. 2-3 minutes for a queue with a single stuck message).This does indeed sound like a long time. How many emails are in the tcp_local queue when you ran the command?
Another inconvenience concerns the HELD messages. After I fixed the reason that got several dozen messages into the HELD status, I had to "release" them one-by-one from the interactive mode. You shouldn't need to do this. Try the following steps to select/process multiple held messages in a channel:
./imsimta qm
dir -held <channel name e.g. tcp_local>
release 1-<last message number>So if you had 10 held messages in tcp_local it would look like this:
./imsimta qm
dir -held tcp_local
release 1-10Regards,
Shane.