pyrra-dev
feat(examples): add OpenShift examples and readme
Signed-off-by: Rick Rackow [email protected]
Since OpenShift has some different behavior especially around connecting to built-in services, it makes sense to have a separate Readme for it with examples adjusted to work on OpenShift. This PR adds exactly that: A readme and all files to deploy a running Pyrra on Openshift
cc @vrutkovs
The push error in the workflow isn't related to my change @metalmatze, but probably needs a bit of extra attention
The rest should be good, maybe @vrutkovs or @nautilux could run through the Readme on a cluster-bot cluster and see if it's working
Hi! Any updates here? This would be a fantastic addition to Pyrra. Especially on use cases for migration from OpenShift (on-prem) to AWS. Having Pyrra SLOs established on-prem to monitor and compare to when moving an application between on-prem and AWS would be really helpful.
I think this one is good to go, just waiting for someone with openshift knowledge or a cluster to cross check
@RiRa12621 We've successfully deployed this to our OpenShift cluster, but I can't confirm if the prometheus-operator logic works (the part where rules should be automatically picked up py prometheus-operator). We use standalone instances outside of the cluster looking in. We did however get the SLO definition deployed to Pyrra and using this https://github.com/pyrra-dev/pyrra/issues/314 we managed to generate the alert/recording rules outside of the cluster and then deploy them to the server with Ansible. The only changes we made was adding a patch to both the pyrra-api and pyrra-kubernetes to bump up the memory as the pod got OOM-killed. Also we changed the namespace during testing so we could just purge the namespace if needed, this might be something that should be on the same namespace as prometheus-operator to be able to pick up the recording/alert-rules? Also added service and route-entry, but that might not be necessary to include in the examples.