Hmm, I looked at that and for some reason didn't think it would do what I wanted...
Anyway, my combo test looks like: IVR.procs = (node1.procs + node2.procs + node3.procs + node4.procs) == 1 Although I'm not sure how to make it only test on the 'ivr' processes....
but I'm getting the same error as this thread (http://www.xymon.com/archive/2009/09/msg00052.html) reports. My Xymon server is a RHEL5.5 virtual host running under Xen.
On an aside, this environment 'reuses' hostnames only differentiating systems by the domain suffix (no, not my idea and no I don't like it). I have node[1-4].prod and node[1-3].dev (no not their real hostnames or domain names), all are in Xymon. Can the combo test handle being provided the FQDN? i.e. IVR.procs = (node1\.prod.procs + node2\.prod.procs + node3\.prod.procs
- node4\.prod.procs) == 1
Thanks
CC
On Tue, Oct 19, 2010 at 9:59 AM, Vernon Everett <everett.vernon at gmail.com> wrote:
Hi Colin
We did something similar at previous client we know too well. :-) It's called a combo-test. Look for "Combination Tests" under the man pages.
What you need to do is define a new test. Call it "Cluster" for lack of a better name. You then define servername.Cluster=(insert logical arguments here) into bbcombotest.cfg This will generate a new test which is dependant on the logic you define. The server name doesn't even need to exist. So you can define your combo test as cluster.app=(logic) This will send the results of a new test "app" as if it came from a new server called "cluster" You then add cluster to bb-hosts as follows 0.0.0.0 Cluster # noconn Very powerful, and the man page does explain it fairly well.
Let me know how you go.
Cheers Vernon
On Tue, Oct 19, 2010 at 9:42 AM, Colin Coe <colin.coe at gmail.com> wrote:
Hi all
I'm trying to do a depends-like test for processes. What I'd like is ensure that process foo is running on one of four hosts, if the process is running on more than one host or on none, a red flag should be raised.
Has anyone done this previously?
Thanks
CC
-- RHCE#805007969328369
To unsubscribe from the xymon list, send an e-mail to xymon-unsubscribe at xymon.com
-- RHCE#805007969328369