supervise: fatal: unable to start log/run: file does not exist

[Diana Calder]

I'm running Qmailtoaster utilities on CentOS6 VM on a XenServer. Data
is stored on NFS. Over the weekend, the VM went into Read-Only file
system due to a lag in communication. I had to manually run FSCK
(there were errors and I told it to fix them) and reboot after
restoring it to Read-Write mode. After doing so, I had errors from
supervise. I managed to find information in the archives to let me fix
the clamd/run error but the other one still persists and I have no
idea what else I can do to fix it. I get a continuous stream of
"supervise: fatal: unable to start log/run: file does not exist".

Running qmailctl stat results in the following:

clamd: up (pid 12127) 6 seconds
log: up (pid 12138) 1 seconds
send: up (pid 8239) 1037 seconds
smtp: up (pid 8245) 1037 seconds
spamd: up (pid 11877) 96 seconds
submission: up (pid 8244) 1037 seconds
clamd/log: up (pid 8240) 1037 seconds
send/log: up (pid 8277) 1032 seconds
smtp/log: up (pid 8246) 1037 seconds
spamd/log: up (pid 12139) 1 seconds
submission/log: up (pid 8273) 1032 seconds

Mail appears to be working okay inbound and outbound but the stream of
error messages continues. I'm stumped. I really don't want to have to
backup the database, rollback to an old snapshot, and restore the
database but I might have to. I've tried starting up a copy of an old
snapshot on another VM to check for differences but can't find
anything obvious.

Suggestions?
--
Best regards,
Diana Calder mailto:***@essexcountylibrary.ca
Automation Technician [phone] x.131
Essex County Library
Essex, ON


---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Eric Shubert]

Post by Diana Calder
I'm running Qmailtoaster utilities on CentOS6 VM on a XenServer. Data
is stored on NFS. Over the weekend, the VM went into Read-Only file
system due to a lag in communication. I had to manually run FSCK
(there were errors and I told it to fix them) and reboot after
restoring it to Read-Write mode. After doing so, I had errors from
supervise. I managed to find information in the archives to let me fix
the clamd/run error but the other one still persists and I have no
idea what else I can do to fix it. I get a continuous stream of
"supervise: fatal: unable to start log/run: file does not exist".
clamd: up (pid 12127) 6 seconds
log: up (pid 12138) 1 seconds
send: up (pid 8239) 1037 seconds
smtp: up (pid 8245) 1037 seconds
spamd: up (pid 11877) 96 seconds
submission: up (pid 8244) 1037 seconds
clamd/log: up (pid 8240) 1037 seconds
send/log: up (pid 8277) 1032 seconds
smtp/log: up (pid 8246) 1037 seconds
spamd/log: up (pid 12139) 1 seconds
submission/log: up (pid 8273) 1032 seconds
Mail appears to be working okay inbound and outbound but the stream of
error messages continues. I'm stumped. I really don't want to have to
backup the database, rollback to an old snapshot, and restore the
database but I might have to. I've tried starting up a copy of an old
snapshot on another VM to check for differences but can't find
anything obvious.
Suggestions?

You shouldn't have a /var/qmail/supervise/log directory. I think if you
simply remove it (or move it aside to /root/ perhaps) then you'll be fine.

Here's what I have in mine (COS6 with new packages):
[***@tacs-mail ~]# ls -l /var/qmail/supervise/
total 20
drwx-----T 4 qmaill qmail 4096 Apr 7 10:24 clamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 send
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 smtp
drwx-----T 4 qmaill qmail 4096 Apr 7 14:58 spamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 submission


Note for posterity, clamav and spamd are very soon to be changed to use
syslogd and initd instead of supervise and multilog. This is in
preparation for COS7, which will be using systemd, at which point
daemontools and ucscp-tcp will no longer be used by QMT.

[Diana Calder]

Post by Eric Shubert
You shouldn't have a /var/qmail/supervise/log directory. I think if you
simply remove it (or move it aside to /root/ perhaps) then you'll be fine.

I DIDN'T have one when the errors started. I also checked an old
snapshot of the Qmail server and it didn't have one either. Despite
that, I tried creating one to see if it would resolve the problem. No
such luck, so I removed it again.

Post by Eric Shubert
total 20
drwx-----T 4 qmaill qmail 4096 Apr 7 10:24 clamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 send
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 smtp
drwx-----T 4 qmaill qmail 4096 Apr 7 14:58 spamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 submission

Yes, I have the same directories actually present:

[***@qmail ~]# ls -l /var/qmail/supervise
total 20
drwx-----T 4 qmaill qmail 4096 Jun 30 10:00 clamd
drwx-----T 4 qmaill qmail 4096 Mar 4 10:15 send
drwx-----T 4 qmaill qmail 4096 Apr 11 12:13 smtp
drwx-----T 4 qmaill qmail 4096 Jun 23 10:35 spamd
drwx-----T 4 qmaill qmail 4096 Apr 11 15:39 submission

Still getting the same error:

[***@qmail ~]# qmailctl start
Starting qmail-toaster: svscan.
[***@qmail ~]# supervise: fatal: unable to start log/run: file does not exist
supervise: fatal: unable to start log/run: file does not exist

*sigh* Even though mail seems to be working, all is obviously not well
with the server. Memory stays maxed out at 8Gb and the CPU
at 100%.

Here's my qmail startup script:

[***@qmail ~]# cat /etc/rc.d/init.d/qmail
#!/bin/sh
#
# qmailStartup script for QmailToaster MTA
#
# chkconfig: 345 55 45
# description: Qmailtoaster MTA
#
# Udated May 19, 2006
# Nick Hemmesch <***@ndhsoft.com>
#

# Source function library.
if [ -f /etc/init.d/functions ]; then
. /etc/init.d/functions
elif [ -f /etc/rc.d/init.d/functions ]; then
. /etc/rc.d/init.d/functions
fi

PATH=/var/qmail/bin:/usr/local/bin:/usr/bin:/bin
export PATH

case "$1" in
start)
# have a silent kill in case someone tries to start the service when it
# is already running
$0 stop >/dev/null 2>&1
echo -n "Starting qmail-toaster: svscan"
cd /var/qmail/supervise
env - PATH="$PATH" svscan &
echo $! > /var/run/qmail-svscan.pid
touch /var/lock/subsys/qmail
echo "."
;;
stop)
echo -n "Stopping qmail-toaster: svscan"
kill `cat /var/run/qmail-svscan.pid`
echo -n " qmail"
svc -dx /var/qmail/supervise/*
echo -n " logging"
svc -dx /var/qmail/supervise/*/log
echo "."
rm -rf /var/run/qmail-svscan.pid
rm -rf /var/lock/subsys/qmail
;;
stat)
cd /var/qmail/supervise
svstat * */log
;;
doqueue|alrm)
echo "Sending ALRM signal to qmail-send."
svc -a /var/qmail/supervise/send
;;
queue)
/var/qmail/bin/qmail-qstat
/var/qmail/bin/qmail-qread
;;
reload|hup)
echo "Sending HUP signal to qmail-send."
svc -h /var/qmail/supervise/send
;;
pause)
echo "Pausing qmail-send"
svc -p /var/qmail/supervise/send
echo "Pausing qmail-smtpd"
svc -p /var/qmail/supervise/smtp
;;
cont)
echo "Continuing qmail-send"
svc -c /var/qmail/supervise/send
echo "Continuing qmail-smtpd"
svc -c /var/qmail/supervise/smtp
;;
restart)
echo "Restarting qmail:"
echo "* Stopping qmail-smtpd."
svc -d /var/qmail/supervise/smtp
echo "* Sending qmail-send SIGTERM and restarting."
svc -t /var/qmail/supervise/send
echo "* Restarting qmail-smtpd."
svc -u /var/qmail/supervise/smtp
;;
cdb)
tcprules /etc/tcprules.d/tcp.smtp.cdb /etc/tcprules.d/tcp.smtp.tmp < /etc/tcprules.d/tcp.smtp
chmod 644 /etc/tcprules.d/tcp.smtp*
echo "Reloaded /etc/tcprules.d/tcp.smtp"
if [ -f /var/qmail/bin/qmail-badmimetypes ] ; then
/var/qmail/bin/qmail-badmimetypes;
echo "Reloaded /var/qmail/control/badmimetypes.cdb";
fi
if [ -f /var/qmail/bin/qmail-badloadertypes ] ; then
/var/qmail/bin/qmail-badloadertypes;
echo "Reloaded /var/qmail/control/badloadertypes.cdb";
fi
if [ -f /var/qmail/bin/simscanmk ] ; then
/var/qmail/bin/simscanmk -g >/dev/null 2>&1;
echo "Reloaded /var/qmail/control/simversions.cdb";
if [ -f /var/qmail/control/simcontrol ] ; then
/var/qmail/bin/simscanmk >/dev/null 2>&1;
echo "Reloaded /var/qmail/control/simcontrol.cdb";
fi
fi
;;
help)
cat <<HELP
stop -- stops mail service (smtp connections refused, nothing goes out)
start -- starts mail service (smtp connection accepted, mail can go out)
pause -- temporarily stops mail service (connections accepted, nothing leaves)
cont -- continues paused mail service
stat -- displays status of mail service
cdb -- rebuild the tcpserver cdb file for smtp
restart -- stops and restarts smtp, sends qmail-send a TERM & restarts it
doqueue -- sends qmail-send ALRM, scheduling queued messages for delivery
reload -- sends qmail-send HUP, rereading locals and virtualdomains
queue -- shows status of queue
alrm -- same as doqueue
hup -- same as reload
HELP
;;
*)
echo "Usage: $0 {start|stop|restart|doqueue|reload|stat|pause|cont|cdb|queue|help}"
exit 1
;;
esac

exit 0


I must be missing something but I can't figure out what. I don't see
anything here that should be causing the problem.

I've never paid very close attention to qmailctl stat results other
than up or not, but it looks to me like maybe clamd and spamd are
repeatedly stopping and restarting? They never seem to go above a few
(less than 10) seconds.

[***@qmail ~]# qmailctl stat
clamd: up (pid 12991) 1 seconds
send: up (pid 12874) 46 seconds
smtp: up (pid 12877) 46 seconds
spamd: up (pid 12878) 46 seconds
submission: up (pid 12879) 46 seconds
clamd/log: up (pid 12873) 46 seconds
send/log: up (pid 12876) 46 seconds
smtp/log: up (pid 12872) 46 seconds
spamd/log: up (pid 12992) 0 seconds
submission/log: up (pid 12880) 46 seconds
[***@qmail ~]# supervise: fatal: unable to start log/run: file does not exist
supervise: fatal: unable to start log/run: file does not exist

So, I have a server throwing a hissy fit and a holiday tomorrow.
Hopefully, it doesn't melt down before I find a fix.

Post by Eric Shubert
Note for posterity, clamav and spamd are very soon to be changed to use
syslogd and initd instead of supervise and multilog. This is in
preparation for COS7, which will be using systemd, at which point
daemontools and ucscp-tcp will no longer be used by QMT.

The less custom stuff that has to be added to a system for QMT, the
better, as far as I'm concerned.
--
Best regards,
Diana Calder mailto:***@essexcountylibrary.ca
Automation Technician [phone] x.131
Essex County Library
Essex, ON


---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Eric Shubert]

Unfortunately, the error message doesn't appear to say which log file
it's having a problem with.

I'm going to guess it's clamav, since that's being moved. I probably
need to have the clamav package remove some leftover stuff.

First, verify that clamav is logging to /var/log/maillog. If it's not,
then I wonder which clamav package you have installed.

This should fix you up:
# cd /var/qmail/supervise/clamd
# mv log log.old

I'm guessing that'll take care of things. Sorry I haven't tested the
latest package better. I hope to get to that this week.

Thanks.
--
-Eric 'shubes'

Post by Diana Calder

Post by Eric Shubert
You shouldn't have a /var/qmail/supervise/log directory. I think if you
simply remove it (or move it aside to /root/ perhaps) then you'll be fine.

I DIDN'T have one when the errors started. I also checked an old
snapshot of the Qmail server and it didn't have one either. Despite
that, I tried creating one to see if it would resolve the problem. No
such luck, so I removed it again.

Post by Eric Shubert
total 20
drwx-----T 4 qmaill qmail 4096 Apr 7 10:24 clamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 send
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 smtp
drwx-----T 4 qmaill qmail 4096 Apr 7 14:58 spamd
drwx-----T 4 qmaill qmail 4096 Mar 24 11:29 submission

total 20
drwx-----T 4 qmaill qmail 4096 Jun 30 10:00 clamd
drwx-----T 4 qmaill qmail 4096 Mar 4 10:15 send
drwx-----T 4 qmaill qmail 4096 Apr 11 12:13 smtp
drwx-----T 4 qmaill qmail 4096 Jun 23 10:35 spamd
drwx-----T 4 qmaill qmail 4096 Apr 11 15:39 submission
Starting qmail-toaster: svscan.
supervise: fatal: unable to start log/run: file does not exist
*sigh* Even though mail seems to be working, all is obviously not well
with the server. Memory stays maxed out at 8Gb and the CPU
at 100%.
#!/bin/sh
#
# qmailStartup script for QmailToaster MTA
#
# chkconfig: 345 55 45
# description: Qmailtoaster MTA
#
# Udated May 19, 2006
#
# Source function library.
if [ -f /etc/init.d/functions ]; then
. /etc/init.d/functions
elif [ -f /etc/rc.d/init.d/functions ]; then
. /etc/rc.d/init.d/functions
fi
PATH=/var/qmail/bin:/usr/local/bin:/usr/bin:/bin
export PATH
case "$1" in
start)
# have a silent kill in case someone tries to start the service when it
# is already running
$0 stop >/dev/null 2>&1
echo -n "Starting qmail-toaster: svscan"
cd /var/qmail/supervise
env - PATH="$PATH" svscan &
echo $! > /var/run/qmail-svscan.pid
touch /var/lock/subsys/qmail
echo "."
;;
stop)
echo -n "Stopping qmail-toaster: svscan"
kill `cat /var/run/qmail-svscan.pid`
echo -n " qmail"
svc -dx /var/qmail/supervise/*
echo -n " logging"
svc -dx /var/qmail/supervise/*/log
echo "."
rm -rf /var/run/qmail-svscan.pid
rm -rf /var/lock/subsys/qmail
;;
stat)
cd /var/qmail/supervise
svstat * */log
;;
doqueue|alrm)
echo "Sending ALRM signal to qmail-send."
svc -a /var/qmail/supervise/send
;;
queue)
/var/qmail/bin/qmail-qstat
/var/qmail/bin/qmail-qread
;;
reload|hup)
echo "Sending HUP signal to qmail-send."
svc -h /var/qmail/supervise/send
;;
pause)
echo "Pausing qmail-send"
svc -p /var/qmail/supervise/send
echo "Pausing qmail-smtpd"
svc -p /var/qmail/supervise/smtp
;;
cont)
echo "Continuing qmail-send"
svc -c /var/qmail/supervise/send
echo "Continuing qmail-smtpd"
svc -c /var/qmail/supervise/smtp
;;
restart)
echo "Restarting qmail:"
echo "* Stopping qmail-smtpd."
svc -d /var/qmail/supervise/smtp
echo "* Sending qmail-send SIGTERM and restarting."
svc -t /var/qmail/supervise/send
echo "* Restarting qmail-smtpd."
svc -u /var/qmail/supervise/smtp
;;
cdb)
tcprules /etc/tcprules.d/tcp.smtp.cdb /etc/tcprules.d/tcp.smtp.tmp < /etc/tcprules.d/tcp.smtp
chmod 644 /etc/tcprules.d/tcp.smtp*
echo "Reloaded /etc/tcprules.d/tcp.smtp"
if [ -f /var/qmail/bin/qmail-badmimetypes ] ; then
/var/qmail/bin/qmail-badmimetypes;
echo "Reloaded /var/qmail/control/badmimetypes.cdb";
fi
if [ -f /var/qmail/bin/qmail-badloadertypes ] ; then
/var/qmail/bin/qmail-badloadertypes;
echo "Reloaded /var/qmail/control/badloadertypes.cdb";
fi
if [ -f /var/qmail/bin/simscanmk ] ; then
/var/qmail/bin/simscanmk -g >/dev/null 2>&1;
echo "Reloaded /var/qmail/control/simversions.cdb";
if [ -f /var/qmail/control/simcontrol ] ; then
/var/qmail/bin/simscanmk >/dev/null 2>&1;
echo "Reloaded /var/qmail/control/simcontrol.cdb";
fi
fi
;;
help)
cat <<HELP
stop -- stops mail service (smtp connections refused, nothing goes out)
start -- starts mail service (smtp connection accepted, mail can go out)
pause -- temporarily stops mail service (connections accepted, nothing leaves)
cont -- continues paused mail service
stat -- displays status of mail service
cdb -- rebuild the tcpserver cdb file for smtp
restart -- stops and restarts smtp, sends qmail-send a TERM & restarts it
doqueue -- sends qmail-send ALRM, scheduling queued messages for delivery
reload -- sends qmail-send HUP, rereading locals and virtualdomains
queue -- shows status of queue
alrm -- same as doqueue
hup -- same as reload
HELP
;;
*)
echo "Usage: $0 {start|stop|restart|doqueue|reload|stat|pause|cont|cdb|queue|help}"
exit 1
;;
esac
exit 0
I must be missing something but I can't figure out what. I don't see
anything here that should be causing the problem.
I've never paid very close attention to qmailctl stat results other
than up or not, but it looks to me like maybe clamd and spamd are
repeatedly stopping and restarting? They never seem to go above a few
(less than 10) seconds.
clamd: up (pid 12991) 1 seconds
send: up (pid 12874) 46 seconds
smtp: up (pid 12877) 46 seconds
spamd: up (pid 12878) 46 seconds
submission: up (pid 12879) 46 seconds
clamd/log: up (pid 12873) 46 seconds
send/log: up (pid 12876) 46 seconds
smtp/log: up (pid 12872) 46 seconds
spamd/log: up (pid 12992) 0 seconds
submission/log: up (pid 12880) 46 seconds
supervise: fatal: unable to start log/run: file does not exist
So, I have a server throwing a hissy fit and a holiday tomorrow.
Hopefully, it doesn't melt down before I find a fix.

Post by Eric Shubert
Note for posterity, clamav and spamd are very soon to be changed to use
syslogd and initd instead of supervise and multilog. This is in
preparation for COS7, which will be using systemd, at which point
daemontools and ucscp-tcp will no longer be used by QMT.

The less custom stuff that has to be added to a system for QMT, the
better, as far as I'm concerned.

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Diana Calder]

Post by Eric Shubert
Unfortunately, the error message doesn't appear to say which log file
it's having a problem with.
I'm going to guess it's clamav, since that's being moved. I probably
need to have the clamav package remove some leftover stuff.
First, verify that clamav is logging to /var/log/maillog. If it's not,
then I wonder which clamav package you have installed.

It was installed by the qmailtoaster install script when the one for
COS6 was still running from the testing repos. It is logging to
/var/log/maillog.

Post by Eric Shubert
# cd /var/qmail/supervise/clamd
# mv log log.old

Done. But...

Post by Eric Shubert
I'm guessing that'll take care of things.

Nope, no such luck. Same error.

However, once I removed log.old it cleared up the error when I did a
qmailctl start - so, that's one problem solved.

Unfortunately, now it's throwing an error:

ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.

Tried rebooting the VM and I spotted this in the bootup messages:

Starting qmail-toaster: svscan.
Starting sshd: [ OK ]
Starting xinetd: [ OK ]
Starting ntpd: [ OK ]
Starting Clam AntiVirus Daemon: ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
[ OK ]
Starting freshclam: [ OK ]
Starting mysqld: [ OK ]
Starting Dovecot Imap: [ OK ]
Starting vusaged: started
Starting httpd: [ OK ]
Starting crond: [ OK ]

Renamed /tmp/clamd.socket and started qmailctl - looked okay. Rebooted
to check and the socket file in use error is back.

So, is clamd trying to start twice? Once via svscan and once on its
own? I did run a yum update when I first saw the errors, on the
offchance that it would fix the problem - maybe something odd happened
there?

More testing...

[***@qmail ~]# /etc/init.d/clamd stop
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
[***@qmail ~]# /etc/init.d/clamd stop
Stopping Clam AntiVirus Daemon: [ OK ]
[***@qmail ~]# qmailctl stop
Stopping qmail-toaster: svscan qmail logging.
[***@qmail ~]# qmailctl start
Starting qmail-toaster: svscan.
[***@qmail ~]# ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
qmailctl stop
Stopping qmail-toaster: svscan qmail logging.
[***@qmail ~]# /etc/init.d/clamd stop
Stopping Clam AntiVirus Daemon: [ OK ]
[***@qmail ~]# qmailctl start
Starting qmail-toaster: svscan.
[***@qmail ~]# ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
qmailctl stat
clamd: up (pid 2678) 8 seconds
send: up (pid 2562) 82 seconds
smtp: up (pid 2569) 82 seconds
spamd: up (pid 2571) 82 seconds
submission: up (pid 2566) 82 seconds
send/log: up (pid 2568) 82 seconds
smtp/log: up (pid 2570) 82 seconds
submission/log: up (pid 2567) 82 seconds

So, clamd keeps starting and stopping and restarting. *sigh* Now what?
--
Best regards,
Diana Calder mailto:***@essexcountylibrary.ca
Automation Technician [phone] x.131
Essex County Library
Essex, ON


---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Finn Buhelt]

Hi Diana.

I've also had this thing happening to my system - check if You have
clamd starting in the init.d directory - my issue was that I happended
to get clamd installed by another repo file which put the clamd startup
script under /etc/init.d (and linked to which ever rcX.d You are using -
so check that out (clamd will be handled by QMT /var/qmail/supervise...)


Hope this can be helpfull?

Cheers
Finn

Post by Diana Calder

Post by Eric Shubert
Unfortunately, the error message doesn't appear to say which log file
it's having a problem with.
I'm going to guess it's clamav, since that's being moved. I probably
need to have the clamav package remove some leftover stuff.
First, verify that clamav is logging to /var/log/maillog. If it's not,
then I wonder which clamav package you have installed.

It was installed by the qmailtoaster install script when the one for
COS6 was still running from the testing repos. It is logging to
/var/log/maillog.

Post by Eric Shubert
# cd /var/qmail/supervise/clamd
# mv log log.old

Done. But...

Post by Eric Shubert
I'm guessing that'll take care of things.

Nope, no such luck. Same error.
However, once I removed log.old it cleared up the error when I did a
qmailctl start - so, that's one problem solved.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
Starting qmail-toaster: svscan.
Starting sshd: [ OK ]
Starting xinetd: [ OK ]
Starting ntpd: [ OK ]
Starting Clam AntiVirus Daemon: ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
[ OK ]
Starting freshclam: [ OK ]
Starting mysqld: [ OK ]
Starting Dovecot Imap: [ OK ]
Starting vusaged: started
Starting httpd: [ OK ]
Starting crond: [ OK ]
Renamed /tmp/clamd.socket and started qmailctl - looked okay. Rebooted
to check and the socket file in use error is back.
So, is clamd trying to start twice? Once via svscan and once on its
own? I did run a yum update when I first saw the errors, on the
offchance that it would fix the problem - maybe something odd happened
there?
More testing...
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
Stopping Clam AntiVirus Daemon: [ OK ]
Stopping qmail-toaster: svscan qmail logging.
Starting qmail-toaster: svscan.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
qmailctl stop
Stopping qmail-toaster: svscan qmail logging.
Stopping Clam AntiVirus Daemon: [ OK ]
Starting qmail-toaster: svscan.
Closing the main socket.
qmailctl stat
clamd: up (pid 2678) 8 seconds
send: up (pid 2562) 82 seconds
smtp: up (pid 2569) 82 seconds
spamd: up (pid 2571) 82 seconds
submission: up (pid 2566) 82 seconds
send/log: up (pid 2568) 82 seconds
smtp/log: up (pid 2570) 82 seconds
submission/log: up (pid 2567) 82 seconds
So, clamd keeps starting and stopping and restarting. *sigh* Now what?

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Eric Shubert]

Yeah, it's trying to run from both supervise and init.d.

Which clamd version is installed?
--
-Eric 'shubes'

Post by Finn Buhelt
Hi Diana.
I've also had this thing happening to my system - check if You have
clamd starting in the init.d directory - my issue was that I happended
to get clamd installed by another repo file which put the clamd startup
script under /etc/init.d (and linked to which ever rcX.d You are using -
so check that out (clamd will be handled by QMT /var/qmail/supervise...)
Hope this can be helpfull?
Cheers
Finn

Post by Diana Calder

Post by Eric Shubert
Unfortunately, the error message doesn't appear to say which log file
it's having a problem with.
I'm going to guess it's clamav, since that's being moved. I probably
need to have the clamav package remove some leftover stuff.
First, verify that clamav is logging to /var/log/maillog. If it's not,
then I wonder which clamav package you have installed.

It was installed by the qmailtoaster install script when the one for
COS6 was still running from the testing repos. It is logging to
/var/log/maillog.

Post by Eric Shubert
# cd /var/qmail/supervise/clamd
# mv log log.old

Done. But...

Post by Eric Shubert
I'm guessing that'll take care of things.

Nope, no such luck. Same error.
However, once I removed log.old it cleared up the error when I did a
qmailctl start - so, that's one problem solved.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
Starting qmail-toaster: svscan.
Starting sshd: [ OK ]
Starting xinetd: [ OK ]
Starting ntpd: [ OK ]
Starting Clam AntiVirus Daemon: ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
[ OK ]
Starting freshclam: [ OK ]
Starting mysqld: [ OK ]
Starting Dovecot Imap: [ OK ]
Starting vusaged: started
Starting httpd: [ OK ]
Starting crond: [ OK ]
Renamed /tmp/clamd.socket and started qmailctl - looked okay. Rebooted
to check and the socket file in use error is back.
So, is clamd trying to start twice? Once via svscan and once on its
own? I did run a yum update when I first saw the errors, on the
offchance that it would fix the problem - maybe something odd happened
there?
More testing...
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
Stopping Clam AntiVirus Daemon: [ OK ]
Stopping qmail-toaster: svscan qmail logging.
Starting qmail-toaster: svscan.
Closing the main socket.
ERROR: LOCAL: Socket file /tmp/clamd.socket is in use by another process.
Closing the main socket.
qmailctl stop
Stopping qmail-toaster: svscan qmail logging.
Stopping Clam AntiVirus Daemon: [ OK ]
Starting qmail-toaster: svscan.
Closing the main socket.
qmailctl stat
clamd: up (pid 2678) 8 seconds
send: up (pid 2562) 82 seconds
smtp: up (pid 2569) 82 seconds
spamd: up (pid 2571) 82 seconds
submission: up (pid 2566) 82 seconds
send/log: up (pid 2568) 82 seconds
smtp/log: up (pid 2570) 82 seconds
submission/log: up (pid 2567) 82 seconds
So, clamd keeps starting and stopping and restarting. *sigh* Now what?

---------------------------------------------------------------------

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Diana Calder]

Post by Eric Shubert
Yeah, it's trying to run from both supervise and init.d.

That's what I thought - the question is, how do I fix this without
accidentally making things worse? (Well, the *other* question is, what
caused the problem in the first place, but at the moment I'm more
interested in solving it.)

Post by Eric Shubert
Which clamd version is installed?

[***@qmail ~]# yum list clamd
Loaded plugins: fastestmirror, priorities, refresh-packagekit
Loading mirror speeds from cached hostfile
* base: less.cogeco.net
* extras: less.cogeco.net
* qmailtoaster-current: mirror2.qmailtoaster.com
* qmailtoaster-current-nodist: mirror2.qmailtoaster.com
* remi: rpms.famillecollet.com
* rpmforge: ftp.heanet.ie
* updates: mirror.science.uottawa.ca
162 packages excluded due to repository priority protections
Available Packages
clamd.x86_64 0.98.4-1.el6.rf rpmforge

Okay, that was odd. One clamd package available but none installed. So, try a slight variation...

[***@qmail ~]# yum list clam*
Loaded plugins: fastestmirror, priorities, refresh-packagekit
Loading mirror speeds from cached hostfile
* base: less.cogeco.net
* extras: less.cogeco.net
* qmailtoaster-current: mirror2.qmailtoaster.com
* qmailtoaster-current-nodist: mirror2.qmailtoaster.com
* remi: rpms.famillecollet.com
* rpmforge: ftp.heanet.ie
* updates: mirror.science.uottawa.ca
162 packages excluded due to repository priority protections
Installed Packages
clamav.x86_64 0.98.4-0.qt.el6 @qmailtoaster-testing
Available Packages
clamav-db.x86_64 0.98.4-1.el6.rf rpmforge
clamav-devel.x86_64 0.98.4-1.el6.rf rpmforge
clamav-milter.x86_64 0.98.4-1.el6.rf rpmforge
clamd.x86_64 0.98.4-1.el6.rf rpmforge
clamtk.noarch 4.25-1.el6.rf rpmforge

Okay, looks like the package is actually clamav not clamd and I have
version 0.98.4-0 installed.

As for my qmailtoaster itself:

Installed Packages
qmailtoaster-release.noarch 2.0-1.qt.nodist installed

Installed Packages
qmailtoaster-util.noarch 2.0-1.qt.nodist @qmailtoaster-current-nodist

qmailtoaster-testing is disabled in my repos and qmailtoaster-current
is enabled (and set to a slightly higher priority than the testing
repo back when I changed over from testing to current, just in case
that made any difference).
--
Best regards,
Diana Calder mailto:***@essexcountylibrary.ca
Automation Technician [phone] x.131
Essex County Library
Essex, ON


---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Eric Shubert]

Sorry about the clamd/clamav confusion. The package name is clamav.

You have the latest version from .qt., but I'm curious as to how you got
the testing version, if that repo is disabled (as it should be now in
most setups).

That's where the the problem lies though. The testing version is the
first version to include the init.d file for clamd instead of the
supervise setup. I thought that rpm would remove the old supervise run
file since it's not part of the package any more, but it appears I'll
need to put some code in the rpm spec file to do that 'manually'.

I think the best fix for you at this point will be to simply get rid of
the /var/qmail/supervise/clamd/ directory. You might move it to /root/
as a backup rather than simply deleting it. It'd be safest to do this
with qmail/supervise stopped, so that supervise doesn't get confused.
You might then need to "service clamd start" to get clamd running before
starting qmail back up.

That should fix it for you.

Thanks for testing this out, even though it was apparently inadvertent.
We all appreciate your help.
--
-Eric 'shubes'

Post by Diana Calder

Post by Eric Shubert
Yeah, it's trying to run from both supervise and init.d.

That's what I thought - the question is, how do I fix this without
accidentally making things worse? (Well, the *other* question is, what
caused the problem in the first place, but at the moment I'm more
interested in solving it.)

Post by Eric Shubert
Which clamd version is installed?

Loaded plugins: fastestmirror, priorities, refresh-packagekit
Loading mirror speeds from cached hostfile
* base: less.cogeco.net
* extras: less.cogeco.net
* qmailtoaster-current: mirror2.qmailtoaster.com
* qmailtoaster-current-nodist: mirror2.qmailtoaster.com
* remi: rpms.famillecollet.com
* rpmforge: ftp.heanet.ie
* updates: mirror.science.uottawa.ca
162 packages excluded due to repository priority protections
Available Packages
clamd.x86_64 0.98.4-1.el6.rf rpmforge
Okay, that was odd. One clamd package available but none installed. So, try a slight variation...
Loaded plugins: fastestmirror, priorities, refresh-packagekit
Loading mirror speeds from cached hostfile
* base: less.cogeco.net
* extras: less.cogeco.net
* qmailtoaster-current: mirror2.qmailtoaster.com
* qmailtoaster-current-nodist: mirror2.qmailtoaster.com
* remi: rpms.famillecollet.com
* rpmforge: ftp.heanet.ie
* updates: mirror.science.uottawa.ca
162 packages excluded due to repository priority protections
Installed Packages
Available Packages
clamav-db.x86_64 0.98.4-1.el6.rf rpmforge
clamav-devel.x86_64 0.98.4-1.el6.rf rpmforge
clamav-milter.x86_64 0.98.4-1.el6.rf rpmforge
clamd.x86_64 0.98.4-1.el6.rf rpmforge
clamtk.noarch 4.25-1.el6.rf rpmforge
Okay, looks like the package is actually clamav not clamd and I have
version 0.98.4-0 installed.
Installed Packages
qmailtoaster-release.noarch 2.0-1.qt.nodist installed
Installed Packages
qmailtoaster-testing is disabled in my repos and qmailtoaster-current
is enabled (and set to a slightly higher priority than the testing
repo back when I changed over from testing to current, just in case
that made any difference).

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Finn Buhelt]

Hi Eric.

Has this 'testing' something to do with my spamd is logging in the
maillog file rather than the /spamd/current file ? (since last update
beginning of April) or has my age caught up with me making me forgetting
things I have read/done ;-)

(repo = 1 in current)

Cheers,
Finn

BTW when will the latest clamav be available for COS6 ?

Post by Eric Shubert
Sorry about the clamd/clamav confusion. The package name is clamav.
You have the latest version from .qt., but I'm curious as to how you got
the testing version, if that repo is disabled (as it should be now in
most setups).
That's where the the problem lies though. The testing version is the
first version to include the init.d file for clamd instead of the
supervise setup. I thought that rpm would remove the old supervise run
file since it's not part of the package any more, but it appears I'll
need to put some code in the rpm spec file to do that 'manually'.
I think the best fix for you at this point will be to simply get rid of
the /var/qmail/supervise/clamd/ directory. You might move it to /root/
as a backup rather than simply deleting it. It'd be safest to do this
with qmail/supervise stopped, so that supervise doesn't get confused.
You might then need to "service clamd start" to get clamd running before
starting qmail back up.
That should fix it for you.
Thanks for testing this out, even though it was apparently inadvertent.
We all appreciate your help.

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Eric Shubert]

Post by Finn Buhelt
Hi Eric.
Has this 'testing' something to do with my spamd is logging in the
maillog file rather than the /spamd/current file ? (since last update
beginning of April) or has my age caught up with me making me forgetting
things I have read/done;-)

That probably depends on the version you're running. IIRC, I made the
logging changes to spamassassin-3.3.2-1.qt, which is in the testing
repo. I'll probably also add the init script to this version before
releasing it to current. Guess I should make that 3.3.2-2 then. At that
point, both clamd and spamd will be severed from daemontools dependence.
All log messages will (eventually) be going to /var/log/maillog (via the
standard syslogd) for the interim, until I get the ELK (Elasticsearch,
Logstash, Kibana) logging stack put together. That's going to be a
really neat feature.

Post by Finn Buhelt
(repo = 1 in current)
Cheers,
Finn
BTW when will the latest clamav be available for COS6 ?

It's available in the testing repo, but as Diana found, it has just a
small issue to be fixed. I simply haven't had time to test it myself,
but hope to get to that this week. Should be next week at the latest.

For those of you running (legacy) clamav-toaster, the latest version
(0.98.4) is available now (and has been for a while, to be honest).

For those running the new packages on COS6, you should change your yum
configuration to disable the testing repo now. The testing repo should
not be used on mission critical hosts. All stable packages will be
available in the 'current' repo.

Thanks Finn.

[Diana Calder]

Hello Eric,

Post by Eric Shubert
Sorry about the clamd/clamav confusion. The package name is clamav.
You have the latest version from .qt., but I'm curious as to how you got
the testing version, if that repo is disabled (as it should be now in
most setups).

I double-checked and it was disabled BUT I don't remember whether I
ran a yum clean all after I changed it (nor do I remember exactly when
I changed it), so I did a yum clean all and ran another yum update
(there weren't any).

Post by Eric Shubert
That's where the the problem lies though. The testing version is the
first version to include the init.d file for clamd instead of the
supervise setup. I thought that rpm would remove the old supervise run
file since it's not part of the package any more, but it appears I'll
need to put some code in the rpm spec file to do that 'manually'.

Ah. Okay, that makes sense.

Post by Eric Shubert
I think the best fix for you at this point will be to simply get rid of
the /var/qmail/supervise/clamd/ directory. You might move it to /root/
as a backup rather than simply deleting it. It'd be safest to do this
with qmail/supervise stopped, so that supervise doesn't get confused.
You might then need to "service clamd start" to get clamd running before
starting qmail back up.

I did a reboot after moving the directory yesterday afternoon, just to
make sure that everything works cleanly.

Post by Eric Shubert
That should fix it for you.

Yes, the toaster ran happily overnight and was still fine this
morning, so it looks like all is well.

Post by Eric Shubert
Thanks for testing this out, even though it was apparently inadvertent.
We all appreciate your help.

Well, at least we found the problem and resolved it relatively
quickly. Thanks for the assistance! When you're a one-gal IT
department, outside help is greatly appreciated when running into
these sorts of situations. There's too many things on the server side
of things that only come up once in a blue moon and I just don't
remember exactly how to resolve them without a little assistance.
Sometimes my Google-fu suffices but, for the times when it doesn't,
lists like this one are essential!
--
Best regards,
Diana Calder mailto:***@essexcountylibrary.ca
Automation Technician [phone] x.131
Essex County Library
Essex, ON


---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com

[Finn Buhelt]

Hi Diana.

Sounds like an error I have seen a couple of times in my QMT running in
KVM : a run files is missing

(/var/qmail/supervise/'smtp|clamd|spamd|send|submission' /run - and
/var/qmail/supervise/ 'smtp|clamd|spamd|send|submission' /log/run )

Check to make sure the run files are present

This is my /var/qmail/supervise/spamd/log/run file

#!/bin/sh
LOGSIZE=`cat /var/qmail/control/logsize`
LOGCOUNT=`cat /var/qmail/control/logcount`
exec /usr/bin/setuidgid qmaill /usr/bin/multilog t s$LOGSIZE n$LOGCOUNT
/var/log/qmail/spamd


Hope it helps

Regards,
Finn

Post by Diana Calder
I'm running Qmailtoaster utilities on CentOS6 VM on a XenServer. Data
is stored on NFS. Over the weekend, the VM went into Read-Only file
system due to a lag in communication. I had to manually run FSCK
(there were errors and I told it to fix them) and reboot after
restoring it to Read-Write mode. After doing so, I had errors from
supervise. I managed to find information in the archives to let me fix
the clamd/run error but the other one still persists and I have no
idea what else I can do to fix it. I get a continuous stream of
"supervise: fatal: unable to start log/run: file does not exist".
clamd: up (pid 12127) 6 seconds
log: up (pid 12138) 1 seconds
send: up (pid 8239) 1037 seconds
smtp: up (pid 8245) 1037 seconds
spamd: up (pid 11877) 96 seconds
submission: up (pid 8244) 1037 seconds
clamd/log: up (pid 8240) 1037 seconds
send/log: up (pid 8277) 1032 seconds
smtp/log: up (pid 8246) 1037 seconds
spamd/log: up (pid 12139) 1 seconds
submission/log: up (pid 8273) 1032 seconds
Mail appears to be working okay inbound and outbound but the stream of
error messages continues. I'm stumped. I really don't want to have to
backup the database, rollback to an old snapshot, and restore the
database but I might have to. I've tried starting up a copy of an old
snapshot on another VM to check for differences but can't find
anything obvious.
Suggestions?

---------------------------------------------------------------------
To unsubscribe, e-mail: qmailtoaster-list-***@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-***@qmailtoaster.com