-
Bug
-
Resolution: Done-Errata
-
Major
-
4.12
-
-
-
Important
-
No
-
Rejected
-
False
-
-
cinder-csi-driver got updated to fix issues with long volume attaches when number of Cinder volume in the project exceeded 1000.
-
Bug Fix
-
Customer Escalated
Description:
Large cluster (~190 nodes) with a big number of volumes provisioned trough cinder.csi.openstack.org is reporting slowness while attaching volumes.
Investigation:
After upgrading the cluster, csi-attacher is displaying the following messages:
VA csi-c1d4bf03ce41693ad5e735575da97411972f8c999797ac1e8f880d44f771e62e for volume fae03218-8940-4064-a94e-656a121bfebf has attached status true but actual state false. Adding back to VA queue for forced reprocessing $ oc logs openstack-cinder-csi-driver-controller-bd46d48f9-lb8hk -c csi-attacher |grep "14:43" |grep VA |wc -l 2043
This matches with the following Upstream issues
https://github.com/kubernetes/cloud-provider-openstack/issues/2137
https://github.com/kubernetes/cloud-provider-openstack/issues/2295
The pull request https://github.com/kubernetes/cloud-provider-openstack/pull/2296 were the issue was fixed was not merged with Cinder CSI driver.
- depends on
-
OCPBUGS-20133 Pagination issue causes delays on attaching cinder volumes
- Closed
- is cloned by
-
OCPBUGS-20130 Pagination issue causes delays on attaching cinder volumes
- Closed
-
OCPBUGS-20133 Pagination issue causes delays on attaching cinder volumes
- Closed
- links to
-
RHBA-2023:5677 OpenShift Container Platform 4.12.38 bug fix update