SMTP service failure

Question

I have been quite happy with JES messaging server 2005Q4 until recently. The other night SMTP service failed at around 1:04am. I had to restart the processes using "imsimta restart". Then, last night, the same thing happened again but this time at 1:29am. I had to restart the process again. I got the following alarm message from the server before it stops responding (similar alarm for the other night as well);
Server instance: /opt/SUNWmsgsr
Alarmid: serverresponse
Instance: smtp
Description: server response time in seconds
Current measured value (13/Jun/2006:01:26:21 -0700): 80
Lowest recorded value: 0
Highest recorded value: 80
Monitoring interval: 600 seconds
Alarm condition is when over threshold of 10
Number of times over threshold: 10
I checked my cron list to see if something is running around that time, but nothing. There is also no clue in any log files including /var/adm/messages and /var/opt/SUNWmsgsr/log/* . Nothing has been logged about any failure or error.
I enabled the watcher to restart the failed processes now, but I need to know why it started failing.
I also tried to find a correlation between the number of SMTP connections and that failure. Although it seems to be unneededly high due to spams, there are some other times that it hits much higher values, and it survives. So, it does not seem to be the number of connections handled by the server. It is something else, but what?
Any comments!

Answer 1

Here are the details;
1) lithium# imsimta version
Sun Java(tm) System Messaging Server 6.2-3.04 (built Jul 15 2005)
libimta.so 6.2-3.04 (built 01:43:03, Jul 15 2005)
SunOS lithium 5.10 Generic sun4u sparc SUNW,UltraSPARC-IIi-Engine
2) - SMTP was not responding.
- No core file.
- I do not know how to allow/disallow a core file. But, I guess, the system can produce a core file, because I have seen core files couple of times so far while I was trying to get it running during the installation phase.
3) The followings are taken before the restart:
lithium# imsimta process
USER PID S VSZ RSS STIME TIME COMMAND
mailsrv 349 S 32528 6552 Jun_12 01:08 /opt/SUNWmsgsr/lib/dispatcher
mailsrv 351 S 32360 6624 Jun_12 00:11 /opt/SUNWmsgsr/lib/job_controller
mailsrv 22888 S 33016 11232 01:25:21 00:00 /opt/SUNWmsgsr/lib/tcp_smtp_server
mailsrv 22895 S 33728 13280 01:25:59 00:01 /opt/SUNWmsgsr/lib/tcp_smtp_server
mailsrv 22896 S 33744 13344 01:25:59 00:01 /opt/SUNWmsgsr/lib/tcp_smtp_server
lithium# top
load averages: 0.23, 0.14, 0.13 01:45:12
64 processes: 55 sleeping, 7 running, 1 zombie, 1 on cpu
CPU states: 96.0% idle, 1.6% user, 2.4% kernel, 0.0% iowait, 0.0% swap
Memory: 384M real, 17M free, 618M swap in use, 119M swap free
PID USERNAME LWP PRI NICE SIZE RES STATE TIME CPU COMMAND
499 root 48 59 0 0K 0K run 385:57 2.95% ns-slapd
23666 mailsrv 1 59 0 32M 10M sleep 0:00 2.54% msprobe
23665 root 1 59 0 3720K 1368K cpu 0:00 0.72% top
658 mailsrv 3 59 0 121M 7696K sleep 138:01 0.48% mshttpd
23654 root 1 59 0 8304K 4624K sleep 0:00 0.31% sshd
349 root 14 59 0 32M 6552K sleep 1:08 0.27% dispatcher
597 root 70 59 0 99M 5760K run 69:51 0.24% ns-httpd
747 root 118 59 0 413M 53M run 85:45 0.22% webservd
730 root 31 59 0 138M 18M run 55:02 0.19% webservd
79 daemon 3 59 0 4256K 1544K sleep 0:02 0.09% kcfd
656 mailsrv 2 59 0 103M 2680K run 18:42 0.06% stored
23657 root 1 59 0 8088K 2448K sleep 0:00 0.05% sshd
23659 root 1 59 0 1272K 1000K sleep 0:00 0.05% sh
655 mailsrv 3 59 0 9080K 528K sleep 11:50 0.04% enpd
778 root 1 59 0 2600K 312K run 11:13 0.04% magt
4) No messages in /var/adm/messages, not even a line
5) details in (3)
6) No response to SMTP port connection