datamattsson
commented
3 months ago How do I increase the volume limit per node? What is the maximum supported or already tested limit?
The limit we've tested with is up to 100. We don't know what happens beyond that. The factor we take into account is "What if a node needs to attach 100 all at once?", which can happen during outages. I've found in my own testing that attachment rate degradation starts around 40-50.
We've had a few requests to control the limit from an end-user perspective but it hasn't materialized. If you're an HPE customer you can work with your account team to file something called a DER to have the issue considered by product management.
If you want to tinker yourself, you can set MAX_VOLUMES_PER_NODE as an environment variable to the CSI node driver.
Hello Reading the documentation for using the CSI driver at: Here
The CSI driver support a fixed number of volumes per node. Inspect the current limitation by running kubectl get csinodes -o yaml and inspect .spec.drivers.allocatable for "csi.hpe.com". The "count" element contains how many volumes the node can attach from the HPE CSI Driver (default is 100).According to the Knowledgebase of Red Hat:
For other drivers or platforms, if any clarification is needed in this regard, the entity which supports them has to be contacted.I searched the documentation (HPE Storage Container Orchestrator Documentation) on this subject but couldn't find it.
How do I increase the volume limit per node? What is the maximum supported or already tested limit?