Spurious panel restart from the watchdog in a Plesk 9.5.4 version

[Miguel_Tellería]

PRODUCT, VERSION, MICROUPDATE, OPERATING SYSTEM, ARCHITECTURE

Plesk Panel 9.5.4, MU 29, Centos 5, amd64

PROBLEM DESCRIPTION

We observe some panel restarts from the watchdog monit daemon in our Plesk 9.5.4 panel. The system recovers itself fully but this provides some downtime for our users.

STEPS TO REPRODUCE

Unfortunately we are still unable to trigger this error manually.

ANY ADDITIONAL INFORMATION

• Alert email sent to root by wdcollect saying (translated from Spanish):

X-Mailer: wdcollect 1.0.0

Plesk Web Server service on host <FQDN> is down.
Problem has been discovered on jun 28, 2015 08:37 .

20 minutes later an email saying that the panel has been restarted is received.

• wdcollect.log

7200 Jun 28 08:52:24 [notice] Connection to server has been established.
7200 Jun 28 08:53:30 [info] Connection to SMTP server has been closed.
7200 Jun 28 08:57:29 [notice] Connection to server has been established.
7200 Jun 28 08:58:30 [info] Connection to SMTP server has been closed.

• monit.log

[CEST Jun 28 08:37:34] HTTP: error receiving data -- Resource temporarily unavailable
[CEST Jun 28 08:37:52] 'plesk_apache' failed protocol test [HTTP] at INET[localhost:8443].
[CEST Jun 28 08:37:58] Collector: error receiving data from http://127.0.0.1:2912/ -- Resource temporarily unavailable
[CEST Jun 28 08:37:58] Collector: communication failure -- aborting message
[CEST Jun 28 08:37:58] 'plesk_apache' trying to restart
[CEST Jun 28 08:37:58] 'plesk_apache' stop: /etc/init.d/sw-cp-server
[CEST Jun 28 08:38:42] 'plesk_apache' start: /etc/init.d/sw-cp-server
[CEST Jun 28 08:38:50] monit: pidfile `/var/run/sw-cp-server.pid' does not contain a valid pidnumber
[CEST Jun 28 08:39:17] Collector: error receiving data from http://127.0.0.1:2912/ -- Resource temporarily unavailable
[CEST Jun 28 08:39:17] Collector: communication failure -- aborting message
[CEST Jun 28 08:45:15] monit: embed_ssl_socket (): SSL service timeout!
[CEST Jun 28 08:45:30] 'plesk_apache' failed, cannot open a connection to INET[localhost:8443]
[CEST Jun 28 08:45:43] 'plesk_apache' trying to restart
[CEST Jun 28 08:45:43] 'plesk_apache' stop: /etc/init.d/sw-cp-server
[CEST Jun 28 08:49:50] 'plesk_apache' start: /etc/init.d/sw-cp-server
[CEST Jun 28 08:50:46] monit: pidfile `/var/run/sw-cp-server.pid' does not contain a valid pidnumber
[CEST Jun 28 08:50:48] monit: pidfile `/var/run/sw-cp-server.pid' does not contain a valid pidnumber
[CEST Jun 28 08:50:49] monit: pidfile `/var/run/sw-cp-server.pid' does not contain a valid pidnumber
[CEST Jun 28 08:50:50] monit: pidfile `/var/run/sw-cp-server.pid' does not contain a valid pidnumber
[CEST Jun 28 08:50:56] GENERIC: error receiving data -- Resource temporarily unavailable
[CEST Jun 28 08:51:00] 'apache' failed protocol test [generic] at INET[178.33.116.67:80].
[CEST Jun 28 08:51:06] Collector: error receiving data from http://127.0.0.1:2912/ -- Resource temporarily unavailable
[CEST Jun 28 08:51:06] Collector: communication failure -- aborting message
[CEST Jun 28 08:51:06] 'apache' trying to restart
[CEST Jun 28 08:51:06] 'apache' stop: /usr/local/psa/admin/bin/websrvmng
[CEST Jun 28 08:52:20] 'apache' start: /usr/local/psa/admin/bin/websrvmng
[CEST Jun 28 08:57:28] 'plesk_apache' connection passed
[CEST Jun 28 08:57:28] 'apache' connection passed

• sw-cp-server/error_log

2015-06-28 08:38:20: (log.c.135) server stopped
2015-06-28 08:38:48: (log.c.75) server started
2015-06-28 08:38:48: (mod_fastcgi.c.1337) Dynamic spawning with max_procs > 1 is not supported; setting min_procs = max_procs
2015-06-28 08:38:48: (mod_fastcgi.c.1337) Dynamic spawning with max_procs > 1 is not supported; setting min_procs = max_procs
2015-06-28 08:45:15: (connections.c.299) SSL: 1 error:140780E5:SSL routines:SSL23_READ:ssl handshake failure
2015-06-28 08:47:37: (log.c.135) server stopped
2015-06-28 08:50:20: (log.c.75) server started
2015-06-28 08:50:20: (mod_fastcgi.c.1337) Dynamic spawning with max_procs > 1 is not supported; setting min_procs = max_procs
2015-06-28 08:50:20: (mod_fastcgi.c.1337) Dynamic spawning with max_procs > 1 is not supported; setting min_procs = max_procs

Any ideals or further details will be appreciated.

 

[Miguel_Tellería]

I forgot to say that I have checked http://kb.odin.com/en/112824 and verify that at least now pidfiles in /var/run match pid number of threads.