cannot upgrade to 9.5.1

[zeroday]

got this message today

Installing the package sw-cp-server-1.0-6.201004011431.centos5.i386
Stoppping SWsoft control panels server... stale pidfile. [ OK ] Starting SWsoft control panels server... stale pidfile. [FAILED]
error: %post(sw-cp-server-1.0-6.201004011431.centos5.i386) scriptlet failed, exit status 255
Error: An error occurred during installation of packages.
Attention! Your software might be inoperable.
Please, contact product technical support.

i thought http://kb.odin.com/en/8338 was the issue but my log tells following error:

all children busy, launch additional (total 4, limit 30)
2010-04-23 21:58:21: (log.c.75) server started
2010-04-23 21:58:21: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-04-23 21:58:21: (log.c.75) server started
2010-04-23 21:58:21: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use

 

[DariuszR]

Had the same problem. CentOS 5.4 with Plesk 9.3 running. What I did was stopping Plesk before running autoinstaller update (/etc/rc.d/init.d/psa stop). Updated successfully after that.

 

[zeroday]

how can you run the autostaller when stopping plesk?

 

[DanijelD]

Plesk Web Server down

Hi guys,
I have a very similar problem. The error is the same, I just didn't do any updates, it just happend.

First I got two messages "The Plesk Web Server service on host www....com is down."
Then after 30 minutes I got the message "The Plesk Web Server service on host www....com has been released from monitoring."

Then I logged in to Plesk to check it and found out that Plesk Web Server is down. I tried to start it with
"/etc/init.d/sw-cp-server start"
but that gave me an error:
"Starting SWsoft control panels server... stale pidfile. [FAILED]"

Then I went to see the log file at
/var/log/sw-cp-server/
and it says
010-05-13 10:04:07: (log.c.75) server started
2010-05-13 10:04:07: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use

I also thought this http://kb.odin.com/en/8338 was the problem, but not, I have the latest fix for that one.

So, my Plesk Web Server is down. What to do?

 

[IgorG]

It is other problem, not as described in mentioned KB article. You have opened port 10001 already and therefore sw-cp-server can't start.
Check it with something like

netstat -vl|grep 10001

It looks like your upgrade was not successfully completed.

 

[DanijelD]

It is other problem, not as described in mentioned KB article.

That is exactly what I said. It's not that article.

You have opened port 10001 already and therefore sw-cp-server can't start.
Check it with something like

netstat -vl|grep 10001

That command gives me the following result:
[root /]# netstat -vl|grep 10001
netstat: no support for `AF INET (sctp)' on this system.
netstat: no support for `AF IPX' on this system.
netstat: no support for `AF AX25' on this system.
netstat: no support for `AF X25' on this system.
netstat: no support for `AF NETROM' on this system.

It looks like your upgrade was not successfully completed.

Which upgrade? I did not do any upgrades in past few days.

Please help...

 

[IgorG]

Try stop sw-cp-server and check port 10001 with other command:

# lsof -i tcp:10001

 

[DanijelD]

This is what I get:

[root /]# lsof -i tcp:10001
COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME
sw-cp-ser 415 sw-cp-server 5u IPv4 57176896 TCP localhost.localdomain:scp-config (LISTEN)

 

[DanijelD]

Any idea? :-(

 

[IgorG]

It means that you have already started sw-cp-server.
What about

rpm -qa | grep sw-cp-server

Maybe you have two rpms installed?

 

[DanijelD]

Here it is, only one:

[root /]# rpm -qa | grep sw-cp-server

sw-cp-server-1.0-6.201004011431.centos5

[root /]#

:-(

 

[IgorG]

Stop sw-cp-server and check that you haven't any sw-cp-server processes with

ps ax | grep sw-cp-server

Kill these processes if they are exists, and check that port 10001 is not opened with

lsof -i tcp:10001

Then try to start sw-cp-server.

 

[DanijelD]

Wuuhuuu... that helped, thanks!

Here is what I did:


[root /]# /etc/init.d/sw-cp-server stop

Stoppping SWsoft control panels server... stale pidfile. [ OK ]

[root /]# ps ax | grep sw-cp-server

404 pts/1 S+ 0:00 grep sw-cp-server
415 ? S 0:28 /usr/sbin/sw-cp-serverd -f /etc/sw-cp-server/config

[root /]# kill 415

[root /]# ps ax | grep sw-cp-server

577 ? S 0:00 /usr/sbin/sw-cp-serverd -f /etc/sw-cp-server/config
604 pts/1 S+ 0:00 grep sw-cp-server

[root /]# /etc/init.d/sw-cp-server start

Starting SWsoft control panels server... already running. [ OK ]

[root /]#


So, it means that it was running. Also, as soon as I killed it - it startet automaticaly again (under new ID 577).

All ok now, thanks a lot.

 

[ThomasR]

Hi,

We got the same problems on a brand new virtual server with (an up-to-date version of) Plesk 9.5.2. The solution suggested by IgorG is a working, but seems to be a temporary workaround. In the few past days, Watchdog indicated dozens of times a "Plesk Web Server is down" message. After to much tries, it stops with the "released from monitoring" message. the only way to start monitoring again, is kill the proces "sw-cp-serverd" and start sw-cp-server again. This VPS has no resources problems and, as it is brand new, is hardly doing anything.

The VPS is running CentOS 5.4 with x86_64 architecture and has following packages:

# rpm -qa | grep sw-cp-server
sw-cp-server-1.0-6.201004011130.centos5

# rpm -qa | grep "psa"
psa-mod_fcgid-1.10-3
psa-mail-driver-common-9.5.2-cos5.build95100504.10
psa-api-9.5.2-cos5.build95100504.10
psa-libpam-plesk-9.5.2-cos5.build95100504.10
psa-migration-manager-9.5.2-cos5.build95100504.10
psa-watchdog-2.0.3-cos5.build95100504.10
psa-horde-3.3.6-cos5.build95100504.10
psa-mimp-1.1.3-cos5.build95100504.10
psa-mod-fcgid-configurator-1.0-14
psa-logrotate-3.7-cos5.build95100504.10
psa-pylibplesk-9.5.2-cos5.build95100504.10
psa-locale-base-en-US-9.5.2-2010041911
psa-api-rpc-9.5.2-cos5.build95100504.10
psa-updates-9.5.2-cos5.build95100504.10
psa-spamassassin-9.5.2-cos5.build95100504.10
psa-mail-pc-driver-1.0.0-cos5.build95100504.10
psa-imp-4.3.6-cos5.build95100504.10
psa-mnemo-2.2.3-cos5.build95100504.10
psa-php5-configurator-1.5.3-cos5.build95100504.10
psa-proftpd-1.3.2e-cos5.build95100504.10
psa-backup-manager-9.5.2-cos5.build95100504.10
psa-awstats-configurator-1.0.0-cos5.build95100504.10
psa-mailman-configurator-9.5.2-cos5.build95100504.10
psa-ingo-1.2.3-cos5.build95100504.10
psa-passwd-3.1.2-cos5.build95100504.10
psa-fcgi-2.4.0-2
psa-ftputil-2.1.1-0.98221
psa-pear-1.4.9-20060517.1
psa-autoinstaller-3.6.0-100413.11
psa-selinux-9.5.2-cos5.build95100504.10
psa-proftpd-xinetd-1.3.2e-cos5.build95100504.10
psa-9.5.2-cos5.build95100504.10
psa-backup-manager-vz-9.5.2-cos5.build95100504.10
psa-migration-agents-9.5.2-cos5.build95100504.10
psa-drweb-configurator-9.5.2-cos5.build95100504.10
psa-kronolith-2.3.3-cos5.build95100504.10
psa-turba-2.3.3-cos5.build95100504.10

In the error_log you can find:

# cat /var/log/sw-cp-server/error_log | grep -B 1 "10001"
2010-05-20 05:46:36: (log.c.75) server started
2010-05-20 05:46:36: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 05:46:37: (log.c.75) server started
2010-05-20 05:46:37: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 05:51:39: (log.c.75) server started
2010-05-20 05:51:39: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 05:51:39: (log.c.75) server started
2010-05-20 05:51:39: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 05:56:41: (log.c.75) server started
2010-05-20 05:56:41: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 05:56:41: (log.c.75) server started
2010-05-20 05:56:41: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 06:01:44: (log.c.75) server started
2010-05-20 06:01:44: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 06:01:44: (log.c.75) server started
2010-05-20 06:01:44: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 09:05:40: (log.c.75) server started
2010-05-20 09:05:40: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 09:05:40: (log.c.75) server started
2010-05-20 09:05:40: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 09:06:35: (log.c.75) server started
2010-05-20 09:06:35: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use
2010-05-20 09:06:35: (log.c.75) server started
2010-05-20 09:06:35: (network.c.300) can't bind to port: 127.0.0.1 10001 Address already in use

Is this a bug in the Watchdog module?

 

[DanijelD]

Hi,
I can also confirm that this happens repeatedly. We need a final solution.

 

[DanijelD]

Maybe this should be split into a separate topic? I am not having problems with upgrade like the subject says, but with plesk server suddenly stopping.

 

[ThomasR]

Problem still exists. As you can see in the next "Watchdog weekly report May 16, 2010 - May 22, 2010" there is hardly any CPU nor RAM usage and the problem only exists with "Plesk Web Server":

Watchdog is running since May 24, 2010 01:00 AM.
Watchdog is monitoring services:
Plesk Web Server
Web Server (Apache)
SMTP Server(Postfix)
IMAP/POP3 Server (Courier-IMAP)
DNS Server (BIND)
MySQL
Watchdog is monitoring:
[normal] /dev/vzfs (mount point /)


Average CPU load: 0.0691%.
Average RAM usage: 137937.1416.
Security scans number: 1.

Events:
[May 22, 2010 06:26 AM] 'Plesk Web Server' service started.
[May 22, 2010 06:21 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 06:11 AM] 'Plesk Web Server' service started.
[May 22, 2010 06:01 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 05:56 AM] 'Plesk Web Server' service started.
[May 22, 2010 05:51 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 05:46 AM] 'Plesk Web Server' service started.
[May 22, 2010 05:40 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 05:00 AM] 'Plesk Web Server' service started.
[May 22, 2010 04:55 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 04:50 AM] 'Plesk Web Server' service started.
[May 22, 2010 04:40 AM] 'Plesk Web Server' service stopped.
[May 22, 2010 04:34 AM] 'Plesk Web Server' service started.
[May 22, 2010 04:29 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 10:00 AM] 'Web Server (Apache)' service started.
[May 21, 2010 09:55 AM] 'Web Server (Apache)' service stopped.
[May 21, 2010 06:23 AM] 'Plesk Web Server' service started.
[May 21, 2010 06:13 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 05:52 AM] 'Plesk Web Server' service started.
[May 21, 2010 05:42 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 04:11 AM] 'Plesk Web Server' service started.
[May 21, 2010 04:01 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 03:55 AM] 'Plesk Web Server' service started.
[May 21, 2010 03:45 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 03:20 AM] 'Plesk Web Server' service started.
[May 21, 2010 03:15 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 03:10 AM] 'Plesk Web Server' service started.
[May 21, 2010 03:04 AM] 'Plesk Web Server' service stopped.
[May 21, 2010 02:49 AM] 'Plesk Web Server' service started.
[May 21, 2010 02:44 AM] 'Plesk Web Server' service stopped.
[May 20, 2010 09:10 AM] 'Plesk Web Server' service started.
[May 20, 2010 09:05 AM] 'Plesk Web Server' service stopped.
[May 20, 2010 06:06 AM] The 'Plesk Web Server' service was released from monitoring because it failed frequently.
[May 20, 2010 05:46 AM] 'Plesk Web Server' service stopped.
[May 20, 2010 05:41 AM] 'Plesk Web Server' service stopped.
[May 20, 2010 04:05 AM] 'Plesk Web Server' service started.
[May 20, 2010 03:45 AM] 'Plesk Web Server' service stopped.
[May 20, 2010 03:39 AM] 'Plesk Web Server' service started.
[May 20, 2010 03:34 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 11:16 AM] 'Plesk Web Server' service started.
[May 19, 2010 10:55 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 10:50 AM] 'Plesk Web Server' service started.
[May 19, 2010 10:40 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 06:27 AM] 'Plesk Web Server' service started.
[May 19, 2010 06:22 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 06:06 AM] 'Plesk Web Server' service started.
[May 19, 2010 06:01 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 05:46 AM] 'Plesk Web Server' service started.
[May 19, 2010 05:41 AM] 'Plesk Web Server' service stopped.
[May 19, 2010 03:35 AM] 'Plesk Web Server' service started.
[May 19, 2010 03:30 AM] 'Plesk Web Server' service stopped.
[May 18, 2010 07:35 AM] 'Plesk Web Server' service started.
[May 18, 2010 07:30 AM] 'Plesk Web Server' service stopped.
[May 18, 2010 05:53 AM] 'Plesk Web Server' service started.
[May 18, 2010 05:43 AM] 'Plesk Web Server' service stopped.
[May 18, 2010 04:17 AM] 'Plesk Web Server' service started.
[May 18, 2010 04:12 AM] 'Plesk Web Server' service stopped.
[May 18, 2010 02:46 AM] 'Plesk Web Server' service started.
[May 18, 2010 02:40 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 02:09 PM] 'Plesk Web Server' service stopped.
[May 17, 2010 02:07 PM] 'Plesk Web Server' service stopped.
[May 17, 2010 09:37 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 09:31 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 04:08 AM] The 'Plesk Web Server' service was released from monitoring because it failed frequently.
[May 17, 2010 03:48 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 03:43 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 02:41 AM] 'Plesk Web Server' service started.
[May 17, 2010 02:36 AM] 'Plesk Web Server' service stopped.
[May 17, 2010 01:00 AM] Security scanning report is in the file /usr/local/psa/var/modules/watchdog/report/security_2010-05-17.
[May 16, 2010 06:06 AM] 'Plesk Web Server' service started.
[May 16, 2010 06:01 AM] 'Plesk Web Server' service stopped.
[May 16, 2010 05:46 AM] 'Plesk Web Server' service started.
[May 16, 2010 05:40 AM] 'Plesk Web Server' service stopped.
[May 16, 2010 03:34 AM] 'Plesk Web Server' service started.
[May 16, 2010 03:29 AM] 'Plesk Web Server' service stopped.
[May 16, 2010 03:14 AM] 'Plesk Web Server' service started.
[May 16, 2010 03:09 AM] 'Plesk Web Server' service stopped.
[May 16, 2010 02:43 AM] 'Plesk Web Server' service started.
[May 16, 2010 02:38 AM] 'Plesk Web Server' service stopped.

Might be important: this virtual server is running on a Virtuozzo Hardware Node with "Offline Management" enabled, so the Service Container is used for running Plesk at https://www.myserver.com:8443/vz/cp/panel/plesk/frameset.

Someone who can confirm this problem is related to the Service Container?

 

[2danis2do]

I am having the same problem.

http://kb.odin.com/en/8119 does not seem to be relevant as I have a firewall anyway.