Hello,
I have a mail server (sendmail) with SUNOS 5.5.1. Just recently it began to respond very slowly.
I used vmstat to check the performance data. Only interupt, system call and CPU context swiching are relatively high. Other statistics are normal, especially CPU utilization are very low.
How I can improve those high statistics if they cause the problems.
Thanks,
Cao Ai
Is this server only running Sendmail? If so, then check to see if you have an increased volume of email that it is handling.
If you save the messages files or have stats on how much mail it handles, please check to see how many messages per week. If spammers have started hitting your system, this could be the only problem you are having.
You can check that you are not an open relay for spammers to use (www.abuse.net)
Next, do you have sar or any monitoring running? What is the version of Sendmail? Is it set up to use DNS? Has DNS recently changed (if sendmail is waiting on DNS, it can slow down your system as messages are not flowing as quickly waiting on timeouts from a DNS request).
Post output from % /usr/lib/sendmail -d0.1 -bt < /dev/null
(change company info as appropriate).
Hi,
Thank you very much.
The server only runs sendmail, and the volume of mail is relatively small. Other situations(spam and DNS) are all seemingly OK.
Here I post information you want and output of vmstat:
> /usr/lib/sendmail -d0.1 -bt </dev/null
Version 8.9.2
Compiled with: MAP_REGEX LOG MATCHGECOS MIME7TO8 MIME8TO7 NAMED_BIND
NDBM NETINET NETUNIX NEWDB NIS NISPLUS QUEUE SCANF SMTP USERDB
XDEBUG
ADDRESS TEST MODE (ruleset 3 NOT automatically invoked)
Enter <ruleset> <address>
>vmstat 5
procs memory page disk faults cpu
r b w swap free re mf pi po fr de sr f0 s0 -- -- in sy cs us sy id
0 0 0 150264 75248 0 0 0 0 0 0 0 0 10 0 0 271 204 217 0 1 99
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 218 205 215 0 0 99
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 216 203 214 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 215 204 212 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 216 205 214 0 0 99
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 216 203 215 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 218 204 213 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 213 203 214 0 0 99
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 217 204 215 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 217 204 215 0 0 100
0 0 0 150264 75248 0 0 0 0 0 0 0 0 0 0 0 211 203 215 0 0 100
0 0 0 150096 75080 0 438 0 0 0 0 0 0 0 0 0 217 931 249 8 8 84
0 0 0 149424 74400 0 4025 0 0 0 0 0 0 0 0 0 219 4438 450 39 61 0
0 0 0 149016 74384 0 4026 0 0 0 0 0 0 0 0 0 214 4446 454 41 59 0
0 0 0 149992 74960 0 1406 0 0 0 0 0 0 7 0 0 267 1768 323 18 23 59
0 0 0 150712 75456 0 0 0 0 0 0 0 0 0 0 0 215 205 215 0 0 100
0 0 0 150712 75456 0 0 0 0 0 0 0 0 0 0 0 213 203 216 0 0 100
0 0 0 150712 75456 0 0 0 0 0 0 0 0 0 0 0 216 203 215 0 0 100
0 0 0 150008 74488 0 0 0 0 0 0 0 0 0 0 0 220 204 214 0 0 99
0 0 0 149728 74248 0 21 0 0 0 0 0 0 0 0 0 214 257 220 1 0 99
0 0 0 149632 74200 0 13 0 0 0 0 0 0 0 0 0 219 219 217 0 1 99
0 0 0 149280 74040 0 0 0 0 0 0 0 0 0 0 0 215 203 214 0 0 100
I have no sar and other monitoring tools. The server supports telnet and ftp services(which are all got very slow response now). I have minimize the number of process, including stopping Openwindows. Following is the process status:
# ps -ef
UID PID PPID C STIME TTY TIME CMD
root 0 0 0 Mar 05 ? 0:00 sched
root 1 0 0 Mar 05 ? 0:00 /etc/init -
root 2 0 0 Mar 05 ? 0:00 pageout
root 3 0 0 Mar 05 ? 0:27 fsflush
root 458 1 0 Mar 05 ? 0:00 /usr/lib/saf/sac -t 300
root 4575 190 0 14:42:15 ? 0:00 /usr/lib/sendmail -bd -q1h
root 111 1 0 Mar 05 ? 0:00 /usr/sbin/keyserv
root 157 1 0 Mar 05 ? 0:00 /usr/sbin/syslogd
root 125 1 0 Mar 05 ? 0:00 /usr/sbin/in.named
root 45 1 0 Mar 05 ? 0:00 /opt/netra/SUNWi150/sysctld -I -h 600
root 129 1 0 Mar 05 ? 0:02 /usr/sbin/inetd -s
root 117 1 0 Mar 05 ? 0:00 /usr/sbin/kerbd
root 132 1 0 Mar 05 ? 0:00 /usr/lib/nfs/statd
root 134 1 0 Mar 05 ? 0:00 /usr/lib/nfs/lockd
root 153 1 0 Mar 05 ? 0:00 /usr/lib/autofs/automountd
root 205 1 0 Mar 05 ? 0:00 /usr/sbin/vold
root 170 1 0 Mar 05 ? 0:02 /usr/sbin/nscd
root 4576 190 0 14:42:20 ? 0:00 /usr/lib/sendmail -bd -q1h
root 190 1 0 Mar 05 ? 0:01 /usr/lib/sendmail -bd -q1h
root 200 1 0 Mar 05 ? 0:00 /usr/lib/utmpd
root 210 1 0 Mar 05 ? 0:00 /opt/netra/SUNWnetra/bin/httpd -f /etc/opt/netraf
root 275 273 0 Mar 05 ? 0:00 /usr/local/netscape/nse-home/admserv/ns-admin -da
root 459 1 0 Mar 05 console 0:00 /usr/lib/saf/ttymon -g -h -p mail console login:c
root 4569 129 0 14:41:03 ? 0:00 in.telnetd
root 272 1 0 Mar 05 ? 0:00 /usr/local/netscape/nse-home/admserv/ns-admin -da
root 261 1 0 Mar 05 ? 0:00 /opt/SUNWipx/bin/NWnvtd
root 273 1 0 Mar 05 ? 0:00 /usr/local/netscape/nse-home/admserv/ns-admin -da
root 465 458 0 Mar 05 ? 0:00 /usr/lib/saf/ttymon
nobody 780 1 0 16:02:32 ? 0:00 ./ns-httpd -d /usr/local/netscape/nse-home/httpsg
root 4582 4572 0 14:42:39 pts/0 0:00 sh
root 4566 190 0 14:38:24 ? 0:00 /usr/lib/sendmail -bd -q1h
caoai 4572 4569 0 14:41:21 pts/0 0:00 -csh
root 4583 190 0 14:42:47 ? 0:00 /usr/lib/sendmail -bd -q1h
root 4567 4566 8 14:39:58 ? 0:16 /usr/lib/sendmail -bd -q1h
root 4584 4582 1 14:42:59 pts/0 0:00 ps -ef
I wonder if there is any strange processe in the system.
Best Regards
Cao Ai
You can still use the "top" which shows
which process is using the cpu time.
--Jagadeesh
Thanks.
I don't have 'top' in my system.
Cao Ai
