Watchdog 2.0.7 on Centos 6 - even more problems
Centos 6 does not use
/etc/inittab, the scripts are in
/etc/init/. But they do not work for me. Neither stop, nor start.
I cannot shutdown my machine - last message in console is
init: psa-monit main process ended, respawning and it comes every 5 minutes or so.
The service does not start correctly too, or at least Panel does not see it enabled although I can see it running:
/usr/local/psa/admin/bin/modules/watchdog/monit -Ic /usr/local/psa/etc/modules/watchdog/monitrc
Actually only the monit, not the collecting part - graphs do not show anything.
Not only does not it start correctly, it even does something wrong with the nginx and httpd processes. It looks like it starts some wrong instance of apache, which does not see all the virtual hosts, and than Nginx fails to start with the following message:
Starting sshd: [ OK ]
Starting xinetd: [ OK ]
Starting ntpd: [ OK ]
Starting spamd:
Starting nginx: nginx: [emerg] bind() to [2001:8d8:8ae:e500::ae:4abd]:80 failed (99: Cannot assign requested address)
[FAILED]
Starting psa... Starting sw-engine-fpm: [ OK ]
Please see, that
httpd is not among started services (I can see it there when I uninstall WD).
I have workarounds to solve shutdown and startup problems, but they make 11.5.3 unfit for a production use:
Restart:
/usr/local/psa/admin/bin/modules/watchdog/wdservice monit stop /usr/local/psa/etc/modules/watchdog/monitrc
shutdown -r now
After start:
/etc/init.d/nginx restart
/etc/init.d/httpd restart
/usr/local/psa/admin/bin/modules/watchdog/wdservice monit start /usr/local/psa/etc/modules/watchdog/monitrc
/usr/local/psa/admin/bin/modules/watchdog/wdservice wdcollect start /usr/local/psa/etc/modules/watchdog/monitrc
My research and workaround:
http://brablc.com/2013/09/21/plesk-watchdog-nginx-shutdown-problem/