|
| 1 | +## v2.4.3 - 2025-07-08 |
| 2 | +### Added |
| 3 | +* Added `noGenerate` to `Password`. When `true`, it prevents the operator from generating non-existent Secrets and non-existent Secret keys that are specified via `valueFrom` in User custom resource |
| 4 | +### Changed |
| 5 | +* It is not the case that the OperatorQuiescent condition for the v1 operator cannot be True unless the ClusterConfigured condition is also True. |
| 6 | + |
| 7 | + The status.observedGeneration will only update when the cluster reaches the OperatorQuiescent state. |
| 8 | +* The operator will try stripping off a layer of quotation from configuration values when interpreting numeric and boolean values. These may be accidentally introduced upstream of the CR, but where the intent is obvious we don't need to be strict about it. |
| 9 | +### Fixed |
| 10 | +* The operator now unconditionally produces statefulsets that have environment variables available to the initContainer that are used for CEL-based config patching. |
| 11 | + |
| 12 | + Previously it attempted to leave existing sts resources unpatched if it seemed like they had already been bootstrapped. With the adoption of CEL patching for node configuration, that left sts pods unable to restart. |
| 13 | +* Improved support for multi-STSes (e.g., multiple NodePools) in the ghost broker decommissioning logic. |
| 14 | + |
| 15 | + - Desired replicas were previously fetched from a single STS, leading to incorrect broker count decisions when multiple STSes were present. Now, the logic accounts for all STSes. |
| 16 | + - Fixed incorrect broker map keying: previously used pod ordinal, which is not unique across STSes (e.g., `blue-0` and `green-0` both mapped to `0`). Switched to using the pod name as the key to correctly distinguish brokers. |
| 17 | + - Disabled ordinal-based broker deletion logic in Operator v1 mode, as it doesn't work reliably in a multi-STS setup. |
| 18 | + |
| 19 | +* Setting `serviceAccount.create` to `false` no longer prevents the Kubernetes ServiceAccountToken volume from being mounted to the operator Pod. |
| 20 | +* updated operator v1 to ignore "cluster.redpanda.com/node-pool-spec" annotation for pod rolls. previously, under certain conditions, the operator started rolling pods if this annotation changed - but there is no need to do so. |
| 21 | +* Added the missing `https` port to the operator Pod that was referenced by the [`ServiceMonitor`](https://github.com/redpanda-data/redpanda-operator/blob/4e34c5ea79b00fa0caeda64955e3291666194274/operator/chart/servicemonitor.go#L42) |
| 22 | +* Empty Redpanda specs (i.e. `clusterSpec: null` or otherwise unspecified) no longer causes the operator to panic |
| 23 | +* `get` permissions on `Node` resources is now correctly configured by default. |
| 24 | + |
| 25 | + `--set rbac.createAdditionalControllerCRs=true` is no longer required for rack awareness to work. |
| 26 | +* Crashes from the Sidecar container (e.g. due to a temporary API Server outage) no longer forcefully restart the redpanda container. |
0 commit comments