Outgoing Mail Delays

[lswindale]

Hi. I'm have been running scalix (ver 11.2.0-GA) for a couple of months now without any major issues. Within the last week however; outgoing mail seems to be delayed in being sent. The delay can be anywhere between 12 and 24 hours. The mail does eventually find it's destination, but it seems to be taking a long time to send. I'm relaying my outgoing mail through my ISP's SMTP server, however this doesn't appear to be the issue (a friend is also relaying through the same ISP's server without delay, and after contacting the ISP they tell me they're not experiencing any probs).

Can you suggest where I should look in the logs to see when the email is sent (from my client) and when the email is relayed? Or, if this sounds like a familiar issue could you please point me in the direction of a post in the forum.

Also, the outgoing delays are experienced whether I send from Outlook or webmail. The issue appears to be further down the line than the client.

Cheers.

[techsharp]

lswindale -

Make sure to check the maillog in /var/log. This will tell you when the mail is sent and will be timestamped.

Also when you send a mail see if it is in the mailq.

Also Scalix logs are found in /var/opt/scalix/??/s/logs - check to see if you can find anything that might give you some clues as to why mail is being delayed.

[charon]

i also realized this issue,

added a script that restarts all smtp and ldap stuff. now it seems to be fixed.

[techsharp]

Great!

Log files are a excellent way to tell what is going on ;p

I should of mentioned doing a omstat -s && omstat -a - that will show all Scalix services and if they are running.

[lswindale]

Thanks for the reply guys. I've tried restarting everything but to no avail. Also, I had a look through the logs. I sent a test email to a hotmail address at 9:05am on 29/12. I then had a look in /var/logs/maillog a few minutes later. I couldn't find any sign of any outgoing mail sent from my address. Looking further back through maillog I could find previous emails that I had sent, but they were timestamped as being sent during the middle of the night (obviously delayed). I did manage to find a record of the email I sent at 9:05 however, in /var/opt/scalix/s1/s/logs/audit. I don't know if this helps, but it contained the following:

Code: Select all

routing
time 1198883117 Sat Dec 29 09:05:17 2007 +600
type 0 message
priority 0 normal
sensitivity 0 normal
importance 0 normal
created-locally 1
hop-count 1
originator Luke Swindale / mymailnode/CN=Swindale, Luke
subject test 0913:2912
ua-message-id H000006900010195.1198883117.serv01.myserver.com.au
mta-message-id H000006900010195.1198883117.serv01.myserver.com.au
part-size 306
part-type 1166 DISTRIBUTION LIST
part-size 8345
part-type 2130 Microsoft RTF
recipient-to external.address / internet/CN=external.address DDT1=RFC-822; DDV1=external_address@hotmail.com;
ack-req 0 none
queue UNIX:MIME
max-nest-depth 0
message-size 13349
part-count 2
delivered-count 1


[lswindale]

Hey guys,

Just following up from the last log message that I posted. As mentioned in my previous post, the test email was sent out at 9:05am on 29/12. I didn't see any sign of the message in /var/log/maillog until 8:34am on 30/12 (see below). Does anything abnormal stand out to you guys? The only thing I've noticed is that the timestamp on the initial routing log says "2007 + 600". I'm guessing this should be "2007 +1000" (assuming +600 is meant to be the offset from GMT).

Code: Select all

Dec 30 08:34:24 serv01 sendmail[19599]: lBTMYMV3019599: from=<luke@mydomain.com>, size=3268, class=0, nrcpts=1, msgid=<H000006900010195.1198883117.serv01.myserver.com.au@MHS>, proto=ESMTP, daemon=MTA, relay=localhost.localdomain [127.0.0.1]
Dec 30 08:34:24 serv01 spamd[15485]: spamd: connection from localhost.localdomain [127.0.0.1] at port 51347
Dec 30 08:34:24 serv01 spamd[15485]: spamd: setuid to sa-milt succeeded
Dec 30 08:34:24 serv01 spamd[15485]: spamd: processing message <H000006900010195.1198883117.serv01.myserver.com.au@MHS> for sa-milt:496
Dec 30 08:34:24 serv01 spamd[15485]: spamd: clean message (-1.4/5.0) for sa-milt:496 in 0.5 seconds, 3740 bytes.
Dec 30 08:34:24 serv01 spamd[15485]: spamd: result: . -1 - ALL_TRUSTED,AWL,BAYES_40,DATE_IN_PAST_12_24,HTML_MESSAGE scantime=0.5,size=3740,user=sa-milt,uid=496,required_score=5.0,rhost=localhost.localdomain,raddr=127.0.0.1,rport=51347,mid=<H000006900010195.1198883117.serv01.myserver.com.au@MHS>,bayes=0.222806,autolearn=no
Dec 30 08:34:24 serv01 sendmail[19599]: lBTMYMV3019599: Milter add: header: X-Spam-Status: No, score=-1.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_40,\n\tDATE_IN_PAST_12_24,HTML_MESSAGE autolearn=no version=3.2.3
Dec 30 08:34:24 serv01 sendmail[19599]: lBTMYMV3019599: Milter add: header: X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on\n\tserv01.myserver.com.au
Dec 30 08:34:24 serv01 spamd[2452]: prefork: child states: II
Dec 30 08:34:25 serv01 sendmail[19632]: lBTMYMV3019599: to=<external_address@hotmail.com>, delay=00:00:01, xdelay=00:00:01, mailer=relay, pri=123268, relay=mail.iinet.net.au. [203.0.178.192], dsn=2.0.0, stat=Sent (ok:  Message 157829764 accepted)


[lswindale]

Ok, I'm wondering whether my mail server is getting flooded with cr@p. My mail delays appear to be getting even longer. After looking at the Internet Mail Gateway queue SAC show'd the following stats. It shows the queue size as being 9047 with an average of 9046 processed in the last minute. This appears odd to me. Is it possible that the queue is so large that it is taking a long time to get around to processing my outgoing emails? Also, how would I go about clearing this queue? Cheers.

Code: Select all

Internet Mail Gateway
messages being processed 1
messages in the queue 9046
total messages processed 20112
averaged 9043.31 messages during last 1 min.
averaged 9042.34 messages during last 5 min.
averaged 9039.56 messages during last 15 min.


[Valerion]

It is quite possible. Messages are processed in the sequence they were received in. The easiest way is to use omqdump to clear the queues (search the forum for usage tips). Be warned it is a powerful diagnostics tool, though, and not to be used lightly.

But I suggest you first find out where the messages are coming from (can be internal or external) and squash that problem.

[lswindale]

Yep, looks like thats why I'm getting the delays. I've cleared the queue and emails are getting delivered in a timely manner now.... so I guess, my next task is to see why that queue keeps filling up. CHeers.

[mikethebike]

I always find it handy to use "omqstat -t" to see the throughput of queues and sizes, also, like Valerion says, check the smtp queue for queue length and age of messages (if using sendmail, check the date/time of the d* in the queue directory).