On 03/01/2010 10:23 AM, Dave Hallowell wrote:
Helmut,
I recently ran into this problem with imap timing out. I don't use
squirrelmail much, but I was getting timeouts using thunderbird when
trying to send and receive from imap email accounts. I read a post on
this list about timeouts when the number of imap processes reached a
certain number. In my case, when I had the smtp timeouts, I would run
from the command line
# ps -ef | grep imapd | wc -l
and it would always return 43 processes when I was having problems. 42
processes, everything ran fine. When I had 43 imapd processes running
and experiencing issues, I would run this from the command line.
# /usr/bin/killall -9 imapd
this would kill all the current imapd processes and sending and
receiving email resumed properly.
I am running Centos4.8 and I haven't done an update for several
months. Still running qmail-toaster-1.03-1.3.18.
I also analyzed, checked, and repaired my vpopmail database but this
did not solve this issue.
So until I can figure out what is really causing this, I wrote a
little script that runs from cron every 5 minutes and if there are
more than 42 imapd processes running, it kills them. Here is the
script if you wish to use it.
------ snip -----------
DT=`date`
KILLIT="/usr/bin/killall -9 imapd"
PID=`ps -ef | grep imapd | wc -l`
if [ $PID -gt 42 ]
then
#echo "too high"
echo $DT >> /oper/kill.log
echo $PID >> /oper/kill.log
$KILLIT > /dev/null 2>&1
fi
----- snip ---------
Have you looked at the imapd config and ensured the MAXDAEMONS is set
high enough?
