Help Needed for Health Check

If a tcp port check to the syslog port is not sufficient to tell whether syslogd is actually functioning correctly then I agree with cmbhatt. Just write a shell script which ssh's to the box, checks for syslogd (ps or whatever), and returns "UP" if it succeeds. Then just create an external monitor type that calls the script. I would back the default monitor interval off from 5 seconds to something a little less aggressive though.

 

 

A tip: the regular shell variables $1, $2 correspond to the pool member and port, respectively, and $3 is any argument called in the monitor definition in the GUI. But LTM is totally IPv6 internally, so $1 will be rendered with an extra "ffff" which may cause you headache wondering why the ssh to the pool member IP fails. I would take care of this by assigning the IP to a variable, stripping off the IPv6 stuff, and then performing ssh to the variable name:

 

 

MEMBER=`echo $1 | sed 's/::ffff://'` 2>/dev/null

 

 

has worked for me before.

 

 

Hope that helps!

 

Denny