Hi,

we got an application which fail to start just because the monitor
operation before "start" exit with return code different than 3.
Actually the rc is 127 which means "file not found" and this happens
because the master script in /etc/init.d is llnked to other scripts on
the shared storage which, of course, is not available on all cluster
nodes at the same time (file system is mounted only on the active node)
Using the LSB script here.

I thinking of writing a OCF script which can somehow "fool" the master
script, for instance, where the monitor operation does not really use
the master script but makes other checks.

Make sense? Any hints?

Is already there an OCF script of example which works in a similar way?

Thanks,
Jan
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to