Monitoring Microsoft Clusters

[dlukinski]

Please advice what is the proper way to monitor Microsoft clusters? (Windows 2008 R2 specifically)

- tried monitoring nodes separately, by configuring their exact FQDN (fails)
- about to try monitoring "as a single" host (no sure if this is correct)
- found some dated plugin in Nagios Exchange (from year 2010, not sure if that one is a solution either)

Thank you

[jdalrymple]

Favorable reviews:
https://exchange.nagios.org/directory/P ... pt/details

I've personally worked with people that have gotten what they needed out of this plugin.

[dlukinski]

Favorable reviews:
https://exchange.nagios.org/directory/P ... pt/details

I've personally worked with people that have gotten what they needed out of this plugin.

Thank you,

Unsure what this means (while configuring): "Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")".

[dlukinski]

Favorable reviews:
https://exchange.nagios.org/directory/P ... pt/details

I've personally worked with people that have gotten what they needed out of this plugin.

Thank you,

Unsure what this means (while configuring): "Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")".

Received (Return code of 255 is out of bounds) error (client currently enable on one of two nodes).

[dlukinski]

Favorable reviews:
https://exchange.nagios.org/directory/P ... pt/details

I've personally worked with people that have gotten what they needed out of this plugin.

Thank you,

Unsure what this means (while configuring): "Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")".

Received (Return code of 255 is out of bounds) error (client currently enable on one of two nodes).
Unsure how SERVICE is configured via GUI:

use generic-service - do nothing in GUI (?)
host_name MSCSSVC01 - assign cluster host/ip (not the nodes) ?
service_description MSCS Resources - my own description
process_perf_data 0 - not sure where this one goes
check_command check_mscs_nrpe!CLRESP - select check_mscs_nrpe commant and add CLRESP as the ARG#1 ?

[dlukinski]

Favorable reviews:
https://exchange.nagios.org/directory/P ... pt/details

I've personally worked with people that have gotten what they needed out of this plugin.

[/quote]

Thank you,

Unsure what this means (while configuring): "Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")".

Received (Return code of 255 is out of bounds) error (client currently enable on one of two nodes).
Unsure how SERVICE is configured via GUI:

use generic-service - do nothing in GUI (?)
host_name MSCSSVC01 - assign cluster host/ip (not the nodes) ?
service_description MSCS Resources - my own description
process_perf_data 0 - not sure where this one goes
check_command check_mscs_nrpe!CLRESP - select check_mscs_nrpe commant and add CLRESP as the ARG#1 ?

Now down to this - "Could not construct return packet in NRPE handler check client side (nsclient.log) logs..."

[ssax]

"Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")"

What this means is point it at the cluster IP/cluster Name (the VIP of the cluster that will point to the active node).

Also, please try running it from the command line and show us the command and the full output of the command.

Thank you

[dlukinski]

"Please remember to point this check to the corresponding clustered ressource (type "Cluster Name" & "Cluster IP Address"),
not to a single node (please look at section "testing & debugging")"

What this means is point it at the cluster IP/cluster Name (the VIP of the cluster that will point to the active node).

Also, please try running it from the command line and show us the command and the full output of the command.

Thank you

This way simply did not work for me at all. I mean the entire check. As the matter of fact NRPE does not seem to work for me at all.

So instead let's discuss check_cluster in here and how I could monitor CLUSTER nodes resources for active/passive cluster so that I do not troubles when cluster switches over to another node (therefore Windows services and drives change making me re-configure services every single time)

[rkennedy]

I saw your other posts come up this morning regarding check_nrpe not working properly, I believe that once this is working for you than the Microsoft clustering plugin will work for you.

Depending on the checks you're looking to run, NRPE may be needed for the kinds of checks you're looking to run. Do you want hold off on this until we're able to help you with your NRPE agents on windows / linux?

[dlukinski]

I saw your other posts come up this morning regarding check_nrpe not working properly, I believe that once this is working for you than the Microsoft clustering plugin will work for you.

Depending on the checks you're looking to run, NRPE may be needed for the kinds of checks you're looking to run. Do you want hold off on this until we're able to help you with your NRPE agents on windows / linux?

Yes please. Let's fix NRPE first