Hello,
I'm having a strange issue on my courier installation and I couldn't find a way to debug in fix this:
When a bunch of email is coming the server can not keep up with it and I can see that it stops delivering email.
For some reason it just does not deliver any other email. What I found into the logs related to that is:
Apr 27 19:31:58 mail courierd: Waiting. shutdown time=none, wakeup time=none, queuedelivering=1073, inprogress=1
The wakeup time is none.
What is more strange is that the number of messages into the queue does not increase even that it accepts new messages and stores them into the msgs directory, but there is no hardlink for them into the msgq dir.
One simple way to fix this problem is to block the incoming connections on port 25 (iptables -j DROP). After that immediatelly it starts to complete a lot of messages and things are starting to deliver:
Apr 27 19:51:13 mail courierd: completed,id=0032C7CC.408E8F74.0000157B
Apr 27 19:51:13 mail courieresmtp: id=0033FC2B.408E8E49.000070A6,from=<[EMAIL PROTECTED]>,addr=<[EMAIL PROTECTED]>: Connection refused
Apr 27 19:51:13 mail courieresmtp: id=0033FC2B.408E8E49.000070A6,from=<[EMAIL PROTECTED]>,addr=<[EMAIL PROTECTED]>,status: deferred
Apr 27 19:51:13 mail courierd: completed,id=0022233B.408E8F7F.0000169C
Apr 27 19:51:13 mail courierd: completed,id=00305626.408E8F79.000015D2
Apr 27 19:51:13 mail courierd: completed,id=00305EE4.408E8F79.000015CC
Apr 27 19:51:13 mail courierd: completed,id=00309724.408E8F7F.000016A4
Apr 27 19:51:13 mail pop3d: Connection, ip=[194.105.24.92]
Apr 27 19:51:13 mail courierd: completed,id=0031F99C.408E8F79.000015EF
Apr 27 19:51:13 mail courierd: completed,id=00322EAB.408E8F79.000015F5
Apr 27 19:51:13 mail courierd: completed,id=0032ABCF.408E8F7A.00001606
Apr 27 19:51:13 mail courieresmtpd: started,ip=[195.229.116.4]
Apr 27 19:51:14 mail courierd: completed,id=0032C72F.408E8F7A.00001619
Apr 27 19:51:14 mail courierd: completed,id=0032C767.408E8F70.000014B5
And the wakeup messages looks like this:
Apr 27 19:32:53 mail courierd: Waiting. shutdown time=none, wakeup time=Tue Apr 27 19:33:03 2004, queuedelivering=1177, inprogress=477
Then the messages start to show into the mailq and this is increasing very rapidly.
This is a very busy server running almost 1 million messages per day.
The hardware is pretty good:
2 x P4 Xeon 2.4
1G RAM
SCSI disks
showconfig output is attached.
Thank you for any suggestions!
--
![]()
www.astral.ro
| Mircea Damian | |
| Manager Infrastructura Metropolitana - Internet & Data | |
| Astral Telecom | |
| Mihai Bravu 223 | |
| Sector 3, cod 030301 | |
| Tel: | 021 326 5003 |
| Fax: | 021 326 6196 |
| GSM: | 0745 128 110 |
<<image/gif>>
backuprelay: (none)
- relay for undeliverable messages
batchsize: 30
- split messages if there are more recipients
defaultdomain: mx.kappa.ro
- domain to append to addresses by default
dsnfrom: "Postmaster on mail.kappa.ro" <[EMAIL PROTECTED]>
- return address on delivery status notifications
dsnlimit: 32 Kb
- maximum message size returned in DSN in is entirety
esmtpacceptmailfor: mx.kappa.ro
- accept mail for these domains via SMTP
esmtpgreeting: mx.kappa.ro ESMTP
- my ESMTP identification banner
esmtphelo: mail.b.astral.ro
- my parameter to the HELO/EHLO SMTP verbs
esmtptimeout: 10 mins
- Timeout for most outbound ESMTP commands
esmtptimeoutconnect: 1 min
- Timeout for ESMTP connection attempt.
esmtptimeoutdata: 5 mins
- Timeout for ESMTP data transfer.
esmtptimeouthelo: 5 mins
- Timeout for the response to the initial EHLO/HELO command.
esmtptimeoutkeepalive: 1 min
- How long to keep outbound ESMTP connections idle, before closing
esmtptimeoutkeepaliveping:
- How often outbound ESMTP connections are pinged
esmtptimeoutquit: 10 secs
- Timeout for the response to the QUIT command.
faxqueuetime: 8 hours
- how long fax messages stay in the queue
localtimeout: 15 mins
- watchdog timeout for local mail deliveries
msgidhost: mx.kappa.ro
- hostname used in generating Message-ID: headers
me: mx.kappa.ro
- my hostname
queuelo: 902
- Message queue cache low watermark
queuehi: 2000
- Message queue cache high watermark
queuefill: 5 mins
- Message queue refill interval
queuetime: 1 day
- how long messages stay in the queue
respawnlo: 1 hour
- courierd automatic restart low watermark
respawnhi: 1 week
- courierd automatic restart high watermark
retryalpha: 5 mins
- 'alpha' retry interval
retrybeta: 3
- 'beta' retry count
retrygamma: 15 mins
- 'gamma' retry interval
retrymaxdelta: 3
- 'maxdelta' retry exponent
sizelimit: 14.3 Mb
- maximum size of a message
submitdelay:
- delay before initial delivery attempt
warntime: 4 hours
- how long messages stay in the queue before a warning message is sent
(a delayed delivery status notification)