|
For the dependencies I’ve always
setup the dep_behavior switch as well to tell mon what should and how a dependency
is understood. Here’s a clip from the man page
describing how it works. dep_behavior {a|m} The evaluation of dependency graphs can control the
suppression of either alert or monitor invocations. Alert suppression.
If this option is set to "a", then the dependency _expression_ will be
evaluated after the monitor for the service exits or after a trap is received.
An alert will only be sent if the evaluation succeeds, meaning that none of the
nodes in the dependency graph indicate failure. Monitor suppression.
If it is set to "m", then the dependency _expression_ will be evaulated
before the monitor for the service is about to run. If the evaulation succeeds,
then the monitor will be run. Otherwise, the monitor will not be run and the
status of the service will remain the same. For the double dependency I’m not
100% sure if that works, I’ve never tried using it that way.
Typically the root depend is either DNS resolving or ping if it is just an ip
address. Here’s a typical host: Host A Service ping Service
SNMP Service
SSH Service
etc.. And all of the services would rely on ping
on the host before they would report failures.
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On
Behalf Of Aaron Segura Hi all, sorry if this has already been covered, but I
coulnd't find anything in the list archive indicating that it has... I have two systems that *pretty much* run the same services,
except one. If this service is running on A, it should not be running on
B, and if running on B, it should not be running on A. So I set up dependencies as follows: dep_behavior = a Watch A service ping interval 60s monitor ping.monitor period... alert... service snmp interval 60s monitor snmp.monitor depend SELF:ping period... alert... service IMGService interval 60s monitor .... depend SELF:ping
&& SELF:snmp && ! B:IMGService Watch B service ping interval 60s monitor ping.monitor period... alert... service snmp interval 60s monitor snmp.monitor depend SELF:ping period... alert... service IMGService interval 60s monitor .... depend SELF:ping
&& SELF:snmp && ! A:IMGService ...which seemed like the right thing to do at the time, but
I'm getting "dep recursion too deep" message in the logs whenever mon
attempts to resolve the dependencies. Looking at the logs leads me to
believe the following is happening: A:ping (OK) A:snmp (OK) B:IMGService -> Depends on... B:ping (OK) B:snmp (OK) A:IMGService -> Depends on... A:ping (OK) A:snmp (OK) B:IMGService
-> Depends on...
B:ping (OK)
B:snmp (OK)
A:IMGService -> Depends on...
A:ping(OK)
A:snmp(OK)
B:IMGService -> Depends on...
B:... etc...etc... So, my question is: How does one set up dependencies
for such a thing? I'm sure this is going to be one of those "Did you
even READ the manual page?" things...it always does when I finally break
down and ask for help...but yeah, I did...honest! :) ams |
_______________________________________________ mon mailing list mon@linux.kernel.org http://linux.kernel.org/mailman/listinfo/mon
