SSH command and cron job. 'killall -9 sendmail'

I got sendmail shut off, yet I'm still geting this...

10:32:06  up 23:36,  1 user,  load average: 27.08, 24.69, 18.72

PID   USER      PRI  NI  SIZE  RSS SHARE STAT [b][color=red]%CPU[/color][/b] %MEM   TIME CPU COMMAND
27911 root      19   0  2460 2452  1772  D    22.1  0.1  25:38   0 sendmail
14904 root      19   0  2460 2452  1772  D    19.9  0.1   6:58   0 sendmail
18063 root      18   0  2456 2448  1772  D    19.3  0.1  37:51   0 sendmail
3611  root      18   0  2464 2456  1772  D    18.3  0.1  12:53   0 sendmail

Code (markup):

that requires a

killall -9 sendmail

to shut it off.

Is it possible to make a cron job run the SSH command

killall -9 sendmail

at say 5 and 35 minutes after the hour?

 

Create a .sh file for eg name it sendmail.sh put that command in that file and put the cron to that file

 

Enter crontab -e
Insert
5,35 * * * * killall -9 sendmail
Press Esc, :, x, enter.
Now you're fine of sendmail.

If you installed webmin, you can change the Max load average for sending setting, or try editting /etc/mail/sendmail.cf

 

I'd say there's a bigger problem here. First of all, the only way sendmail would be consuming that many resources is if it were sending (or receiving) a large amount of email. You might want to watch your mail log file and see exactly what it's doing. Typically this would be in /var/log/maillog and you can watch it in realtime by doing 'tail -f /var/log/maillog'.

As to fully shutting it off, if this is a RedHat based system (Redhat/CentOS/etc) you can do 'chkconfig sendmail off; service sendmail stop'

 

Thanks. The

04 * * * * root killall -9 sendmail
34 * * * * root killall -9 sendmail

stoped it. I got a vBulletin forum on the server, so as long as this works, I don't want to completly shut it off.

For checking the logs for errors, one error is

Jul 25 13:55:32 ensim sendmail[24849]: grew WorkList for /home/virtual/FILESYSTEMTEMPLATE/services/sendmail/mqueue.scanned to 45000

looks like there's a ton of messages there. The directory is way too big for me to look at it through ftp to see if it's safe to do a

rm -f /home/virtual/FILESYSTEMTEMPLATE/services/sendmail/mqueue.scanned/*

 

im sorry, but ive never got the -9 option working for me. shouldnt you be killing the pid?