Forum Discussion
Altocumulusopenshift multi cluster CIS HA
I think I found the reason. The check probe of bigip-ctlr was configured as an Openshift route managed by cis itself. So when starting the primary CIS it had to configure this first on the F5 to get the health check to work. While the secondary did not had this route and was still active, so blocking the creation of this route.
Creating a static route for the health check, using the default ingress controller on port 80 solved the issue
MVPStrange that you have no "probeInterval:" or "retryInterval: " for the multi cluster. Other than that where you got the " autoMonitor: none" option as from openshift-4-9/next-gen-routes/single-vip/ocp-route/route-tea.yaml at main · mdditt2000/openshift-4-9 · GitHub of Mark_Dittmer I don't see this for next gen routes as annotations are used under the route object ?
Ratio and Active-Active mode in Multi-Cluster
That is what came to my mind and as I already mentioned Mark's videos and github are great sources of info.
PaulVogtI got the possible values for autoMonitor from the source code, https://github.com/F5Networks/k8s-bigip-ctlr. Setting it to none in the extended configmap made the warning disappear. CIS 2.19.1 keeps looping with [INFO] Starting Namespace Informer for cluster openshift-engineering-02, flooding the log.
The secondary runs fine. I can try to make the secondary the primary and see what happens then.
