i am still working on integrating a customer’s kubernetes cluster into a checkmk 2.0p6 enterprise installation. We achieved to have everything running fine with piggyback data and dynamic host management.
Now as we have the basics, the customer would like to monitor functionality of the pods themselves. For example we do have some pods running MySQL instances - we would like to have mysql specific checks for these pods.
My first idea was to just add the “pod_ip” from the kubernetes special agent’s piggyback data to the pod hosts created via dynamic host management. That way the customer could install the check_mk_agent inside the mysql pods and we then can monitor these agents via ssh connection automatically? Sadly i couldn’t find a way to get piggyback-data into host attributes via the dynamic host management. Is it possible?
If not, are there any other ways how the above could be achived?
I would love to have some suggestions or hints how that could work.
Hallo,
perhaps another idea:
Use robotmk to check the webinterfaces of the programs running etc.
So you can check ig everything is working. If there are problems you can analyse the containers one by one.
I ve some similar ideas but kubernets is supporting http/ https out of but I ve not found a solution how to forward ports like 22 or 3306.
Our old docker / rancher setup uses a dns server that syncs all containers with a syntax nodename.stack.container. So it is possible to have access to containers pines node node.
Ralf
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.