Description of problem (please be detailed as possible and provide log
snippests):
- Customer needs to have ODF and infrastructure (logging/monitoring/router/etc) stacks workloads running on the same nodes.
Version of all relevant components (if applicable):
- OpenShift 4.10
- ODF 4.10
Does this issue impact your ability to continue to work with the product
(please explain in detail what is the user impact)?
- Customer is impacted since the procedure cannot be followed until there's a confirmation from eng/pm that the procedure is correct, tested and supported.
Is there any workaround available to the best of your knowledge?
- We have provided one, although we need to know if the provided solution is tested/supported and that changes to the different object (to add tolerations) won't be overwriten by an ODF upgrade.
Additional info:
To summarize, the process to be followed would be:
1. https://access.redhat.com/solutions/5034771 to set nodes with infra taints
2. https://docs.openshift.com/container-platform/4.10/monitoring/configuring-the-monitoring-stack.html#moving-monitoring-components-to-different-nodes_configuring-the-monitoring-stack to move logging stack to infra/storage nodes
3. https://docs.openshift.com/container-platform/4.10/logging/config/cluster-logging-moving-nodes.html to move logging stack to infra/storage nodes
4. https://access.redhat.com/solutions/6983629 to configure elasticseach-im cronjobs with tolerations for the infra/storage taints
5. https://access.redhat.com/articles/6408481 to configure ODF deployment to tolerate infra taints.
There is the concern in point #5 that after placing all configurations, an ODF upgrade will overwrite this and Pods won't be able to get scheduled because tolerations would be gone.
This concern comes mainly with the need to edit subscriptions objects like stated in the article, but it also applies to the rest of objects where the changes need to be done, storagecluster and rook-ceph-operator-config configMap.