Changes since v21.10.0
- IMPORTANT: The v21.10.0 release has an issue that can put the Trident controller into a CrashLoopBackOff state when a node is removed and then added back to the Kubernetes cluster. This issue is fixed in v21.10.1.
Fixes:
- Fixed potential race condition when importing a volume on a GCP CVS backend resulting in failure to import
- Fixed an issue that can put the Trident controller into a CrashLoopBackOff state when a node is removed and then added back to the Kubernetes cluster (Issue #669).
- Fixed issue where SVM's were no longer discovered if no SVM name was specified (Issue #612).