man processes of "executable"

[slemke]

many processes of "executable"

Hi @all,

I have a problem with an Opensuse 10.x and Plesk 9.2.1 Server. After a short time a have dozens of "executeable" processes.
Greylisting is enabled, SMTP is enabled, SMTP-after-POP is _disabled_:

15404 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15408 ? S 0:00 | \_ bin/qmail-queue
15409 ? R 1:03 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15412 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15416 ? S 0:00 | \_ bin/qmail-queue
15417 ? R 1:02 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15432 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15436 ? S 0:00 | \_ bin/qmail-queue
15437 ? R 1:01 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15456 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15461 ? S 0:00 | \_ bin/qmail-queue
15462 ? R 1:00 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15465 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15469 ? S 0:00 | \_ bin/qmail-queue
15482 ? R 0:56 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15485 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15489 ? S 0:00 | \_ bin/qmail-queue
15492 ? R 0:56 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15611 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15615 ? S 0:00 | \_ bin/qmail-queue
15616 ? R 0:53 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15676 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15722 ? S 0:00 | \_ bin/qmail-queue
15723 ? R 0:46 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15750 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15794 ? S 0:00 | \_ bin/qmail-queue
15797 ? R 0:46 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
15899 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
15904 ? S 0:00 | \_ bin/qmail-queue
15905 ? R 0:49 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
16177 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
16181 ? S 0:00 | \_ bin/qmail-queue
16182 ? R 0:48 | \_ /usr/local/psa/handlers/info/05-grey-GOODM5/executable context [email protected] [email protected]
16283 ? Ss 0:00 \_ /var/qmail/bin/qmail-smtpd /var/qmail/bin/smtp_auth /var/qmail/bin/true /var/qmail/bin/cmd5checkpw /var/qmail/bin/true
16337 ? S 0:00 | \_ bin/qmail-queue


Any Ideas ?

Regards,
Sebastian

 

[bluecon]

Hello Sebastian,

check this: http://kb.odin.com/en/6345

regards,
Adrian

 

[slemke]

Hello,

"As a workaround for the moment it is recommended to disable POP3 before SMTP authorization"

as I wrote:

"SMTP-after-POP is _disabled_"

Is there anywhere else another field to switch off ?

Regards,
Sebastian

 

[Luis MariaS]

same problem

Hi all!

same problem here, the only workaround we have found is to kill processes or deactivate greylisting.

Using Ubuntu 8.04 VPS inside Virtuozzo Containers 4.0 with CentOS 5.2.

Any update?

Thanks in advanced.

 

[A.M.]

Same problem here.
Ubuntu 8.04 LTS, Plesk 9.2.2, MTA qmail, pop3 before smtp deactivated.
Is there any news about this topic?

 

[A.M.]

Same problem here.
Ubuntu 8.04 LTS, Plesk 9.2.2., MTA qmail, pop3 before smtp deactivated?
Is there any news about this topic?

 

[A.M.]

Same problem here...
Ubuntu 8.04 LTS, Plesk 9.2.2, MTA qmail, SMTP after POP3 disabled
Any news about this topic?

 

[GustavoO]

Same here... looks like we need an update!

 

[FundacioC]

hi, same problem, with FC4 and plesk 9.22
server gets unreachable, althought ping works well.
cpu is at 50%
I'm recording a series of data every minute:
the open files (with cat /proc/sys/fs/file-nr) shows:
from 3000 to 56000 in 30 minutes.!!!
cpu idle from 98% to 11% the same period.

 

[GustavoO]

Do you notice if it happens at a particular time everyday?

Cause we've discovered that it can be a cron problem. Try disabling all cron jobs and see if it still happens.

Also, be sure to do this:

Go into plesk:

1) Mail server settings

* At Relay Options:
2) Uncheck the authentication required: "POP3 lock time"

* Disable greylisting:
3) /usr/local/psa/bin/grey_listing –update-server -status off

 

[FundacioC]

of Cron jobs

Thank you for reply

Yes, it happens between 5:00 and 5:30
I've seen number of open files increasing at this time.

with greylisting disabled, server went ok.
We re-enabled greylisting 24hours after.
The server failed 36 hours after enablig greylist.

Last server error, begins at 4:54... server unreachable ant 05:31

Someone knows which processes in the cron can be safely disabled?
(I mean Plesk cron jobs....)

Next is a shell script I'm using to track the error. It writes a series of values
to a file, ready to paste in excel or calc.

I put it in crontab every minute.
and mails an alert if number of open files is greater than 5000, with the
list of running processes. May help to others.....

DATE=$(date +"%d/%m/%Y %H:%M:%S")
DATE_F=$(date +"%Y%m%d_%H_%M_%S")
RES_FILE=/root/scripts/results.txt
ALERT_FILE=/root/scripts/$DATE_F_alert.txt
TIMEW=`netstat -a --inet -np |grep "TIME_WAIT"|wc -l`
ESTABLISHED=`netstat -a --inet -np |grep "ESTABLISHED"|wc -l`
VMEM=`vmstat |tail -1|sed 's/^ *//g'|sed 's/ */\t/g'`
FILES=`cat /proc/sys/fs/file-nr`
echo -e "$DATE\t$ESTABLISHED\t$TIMEW\t$VMEM\t$FILES">>$RES_FILE
OPEN_F=`echo $FILES|awk '{print $1}'`
if [ $OPEN_F -gt 5000 ]; then
echo $DATE>$ALERT_FILE
echo "Open files: $OPEN_F ">>$ALERT_FILE
echo "LIST OF PROCESSES">>$ALERT_FILE
ps waxvf >>$ALERT_FILE
unix2dos -q $F_ALERTA
mutt -s "Open files ALERT" -a $ALERT_FILE [email protected] <$ALERT_FILE
fi

 

[GustavoO]

Hello!

Try to use "system-snapshot" utility, it can track whats happening in your server when it freezes.


Good luck!

 

[FundacioC]

hi GustavoO

Thanks for the answer.

with greylisting disabled, server is ok, for 4 days.

But I'm not sure if greylisting hang is the cause or a consequence of other things (like a cron job..).

Don't find the utility system-snapshot, which is the rpm ?

 

[IgorG]

Look at new KB article regarding problem with greylists http://kb.odin.com/en/6875
Try apply this hotfix.

 

[FundacioC]

Thanks IgorG.

I've applied the patch. Server is OK. (last 24 h).