You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Prior to this commit all CHANGELOG.md files contained an `### unreleased`
section list all pending changes in changie fragments. The idea at the time was
to make it easy to understand what changes would be present in an upcoming
release.
In practice these sections were never referred to as most developers would rely
on `git log`. Worse yet they were the primary cause of backports requiring
manual intervention.
As the number of release branches has grown, so has the frequency of
backporting and the frustration of needing to manually re-run `task generate`.
This commit removes the unreleased section with the primary goal of reducing
the toil of backporting commits.
(cherry picked from commit 1d28958)
Copy file name to clipboardExpand all lines: charts/redpanda/CHANGELOG.md
-131Lines changed: 0 additions & 131 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,137 +5,6 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
5
5
and is generated by [Changie](https://github.com/miniscruff/changie).
6
6
7
7
8
-
## Unreleased
9
-
### Added
10
-
* Added a chart wide `podTemplate` field which may be used to control Pod attributes chart wide.
11
-
12
-
This field has a lower precedence than `statefulset.podTemplate` and
13
-
`post_install_job.podTemplate` but will still be merged with them.
14
-
*`podTemplate`, `statefulset.podTemplate`, and `post_install_job.podTemplate` may now contain template expressions **with in string fields**
15
-
16
-
To compensate for some of the functionality that was lost with the removal of
17
-
fields like `extraVolumes`, we've upgraded `podTemplate` fields to support
18
-
templating. Rather than allowing full control over a structured value, we've
19
-
reduced the scope to only string fields. This is significantly more
20
-
maintainable and less error prone.
21
-
22
-
As an example, the below snippet will apply the release name as an annotation
23
-
to all Pods created by the chart.
24
-
25
-
```yaml
26
-
podTemplate:
27
-
annotations:
28
-
"keys-cannot-be-templated": '{{ .Release.Name }}'# But values can!
29
-
```
30
-
31
-
See `values.yaml` for additional examples.
32
-
* statefulset.initContainers.configurator.additionalCLIArgs is added to the helm values, to enable the init container to construct an external secret evaluator.
33
-
34
-
In the v2 operator, this value is defaulted from the operator's settings.
35
-
### Changed
36
-
* Promoted the config-watcher sidecar into a real go binary that handles user management and simplifies cluster health checks so they no longer fail when the sole issue is that other nodes in the cluster are unavailable. Additionally the new sidecar subsumes the behavior of the `statefulset.sideCars.controllers` stanza which should now be specified via their own `enabled` flags.
37
-
* `clusterDomain` now defaults to `cluster.local.` (A trialing `.` has been added) and the chart no longer adds trailing `.`'s to internal domains.
38
-
39
-
For users not experiencing issues with trailing `.`'s this change has no
40
-
effect. For users that have had issues with trailing `.`'s, it's now possible
41
-
to opt-out of this behavior by explicitly setting `clusterDomain` to `cluster.local`.
42
-
43
-
For users that override `clusterDomain`, copied a previous releases
44
-
values.yaml, or use the `--reuse-values` flag, trailing `.`'s will be tripped
45
-
from domains upon updating. This behavior may be opted into by appending a
46
-
`.`to `clusterDomain` prior to upgrading.
47
-
* Bump AppVersion to v24.3.6 Redpanda release
48
-
* Bump Redpanda operator side car container tag to `v2.3.7-24.3.6`.
49
-
* `values.schema.json` is now "closed" (`additionalProperties: false`)
50
-
51
-
Any unexpected values will result in a validation error,previously they would
52
-
have been ignored.
53
-
* Update Console depedency to latest version with breaking change. Please visit Console change-log.
54
-
* The name of the container running redpanda is now always set to `redpanda`.
55
-
* bumped `appVersion` to [v25.1.1](https://github.com/redpanda-data/redpanda/releases/tag/v25.1.1).
56
-
* `serviceAccount.create` now defaults to `true`.
57
-
58
-
The previous behavior resulted in using the `default` service account and
59
-
extending it with all bindings generated from the chart. Such behavior is
60
-
unlikely to be desired.
61
-
* `rpk debug bundle --namespace $NAMESPACE` now works by default.
62
-
63
-
The chart now creates a set of `Roles` and `RoleBindings` that satisfy the
64
-
requirements of running `rpk debug bundle` from any redpanda Pod. These
65
-
permissions may be disabled by specifying `rbac.rpkDebugBundle=false`.
66
-
67
-
Additionally, the redpanda container now always has a Kubernetes
68
-
ServiceAccount token mounted to it to ensure `rpk debug bundle` can be
69
-
executed successfully.
70
-
* Update Console depedency to latest version `v3.1.0`. Please visit Console change-log.
71
-
### Deprecated
72
-
* `.statefulset.sidecars.controllers.image` is now deprecated. It may be specified but will not be respected. Use `.statefulset.sidecars.image` instead.
73
-
### Removed
74
-
* Connectors sub-chart integration.
75
-
76
-
The connectors chart may still be deployed separately, though it is not
77
-
officially support. If possible, it is recommended to migrate to redpanda
78
-
connect.
79
-
* Removed the deprecated fields `license_key` and `license_secret_ref` in favor
80
-
of `enterprise.license` and `enterprise.licenseSecretRef`, respectively.
81
-
* `statefulset.securityContext`, `statefulset.sideCars.configWatcher.securityContext` have been removed.
82
-
83
-
These fields previously served as both PodSecurityContext and SecurityContext
84
-
across the entire chart which led to confusing semantics that couldn't be
85
-
fixed without breaking backwards compatiblity.
86
-
87
-
The top level `podTemplate` field may be used to control
88
-
PodSecurityContexts and SecurityContexts across the chart.
89
-
* Fields that would be better served through `podTemplate` have been removed in favor of using `podTemplate`.
* The unrespected`kafkaEndpoint` listener parameter has been removed from `values.yaml`
121
-
### Fixed
122
-
* Reverse order of applying resources to first create ClusterRole and then ClusterRoleBinding.
123
-
When Redpanda custom resource has enabled RBAC the reconciliation was blocked due
124
-
ClusterRoleBinding referencing not yet created ClusterRole.
125
-
126
-
* Fixed an issue where not explicitly specifying a SASL auth mechanism when SASL is enabled caused Console to fail to start up.
127
-
* Prevent broker nodes from restarting when solely the cluster replica amount changes
128
-
* `authentication_method` is no longer set on `http_api` as redpanda itself does not support authentication on the http API.
129
-
* Fixed rack awareness by mounting a service account token to the initcontainer when rack awareness is enabled.
130
-
* Broken `Issuer`s and `Certificate`s are no longer needlessly generated when `tls.<cert>.issuerRef` is provided.
131
-
* Fixed the security contexts' of `set-datadir-ownership` and `set-tiered-storage-cache-dir-ownership`.
132
-
* The `schema_registry_client` and `pandaproxy_client` stanzas of `redpanda.yaml`
133
-
now respect `listeners.kafka.tls.trustStore`, when provided.
134
-
See also [helm-chart 1573 issue](https://github.com/redpanda-data/helm-charts/issues/1573).
135
-
136
-
* Corrected naming of `Role`s to use Fullname instead of Name, which ensures they are unique within their namespace.
137
-
* Updated naming of `ClusterRole`s to include the release namespace. This ensures that they are unique per release and permits installing the chart with the same name across different namespaces.
Copy file name to clipboardExpand all lines: operator/CHANGELOG.md
-95Lines changed: 0 additions & 95 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,101 +5,6 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
5
5
and is generated by [Changie](https://github.com/miniscruff/changie).
6
6
7
7
8
-
## Unreleased
9
-
### Added
10
-
* Added scheduled sync of ghost broker decommissioner to ensure it's running, even if no watches trigger the reconciler.
11
-
* v1 operator: ExternalSecretRefSelector is now provided for referring to external secrets in `clusterConfiguration`. This has an `optional` flag which is honoured if present - it turns errors into warnings if the secret can't be looked up.
12
-
* Added full lifecycle management support for v1alpha2/Redpanda resources by changing all StatefulSets to leverage OnDelete. Additionally, introduce NodePool statuses to the Redpanda structure and deprecate all old Flux-based status fields.
13
-
* statefulset.initContainers.configurator.additionalCLIArgs is added to the helm values, to enable the init container to construct an external secret evaluator.
14
-
15
-
In the v2 operator, this value is defaulted from the operator's settings.
16
-
* Add option to the operator installation to install CRDs as part of the helm install process.
17
-
* The annotation `operator.redpanda.com/config-sync-mode: Declarative` may be added to a Redpanda resource to control the mode in which cluster configuration is updated. `Declarative` mode will *remove* any configuration that's not specified on the resource. The default behaviour is `Additive`, wich preserves the current overwrite-only mode.
18
-
### Changed
19
-
*[Chart] Moved all template rendering into `entry-point.yaml` to match the redpanda and console charts.
20
-
*`values.schema.json` is now "closed" (`additionalProperties: false`)
21
-
22
-
Any unexpected values will result in a validation error,previously they would
23
-
have been ignored.
24
-
* The redpanda operator's helm chart has been merged into the operator itself.
25
-
26
-
Going forward the chart's `version` and `appVersion` will always be equal.
27
-
*`rbac.createRPKBundleCRs` now defaults to `true`.
28
-
* The operator will now populate `.Statefulset.SideCars.Image`, if unspecified, with it's own image.
29
-
30
-
The image and tag may be controlled with pre-existing
31
-
`--configurator-base-image` and `--configurator-tag` flags, respectively.
32
-
33
-
The previous behavior was to defer to the default of the redpanda chart which
34
-
could result in out of sync RBAC requirements or regressions of
35
-
sidecar/initcontainer behavior, if using an older redpanda chart.
36
-
* Bootstrap expansion in the v2 operator's initContainer now uses CEL-patching for its expansion, much like the v1 operator.
37
-
38
-
* It is not the case that the OperatorQuiescent condition for the v1 operator cannot be True unless the ClusterConfigured condition is also True.
39
-
40
-
The status.observedGeneration will only update when the cluster reaches the OperatorQuiescent state.
41
-
* 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.
42
-
### Deprecated
43
-
* v1 operator: the `clusterConfiguration` field `ExternalSecretRef` is deprecated in favour of `ExternalSecretRefSelector`. Since this field was extremely new, it will be removed in the very near future.
44
-
### Removed
45
-
* Removed bundled FluxCD controllers, bundled FluxCD CRDs, and support for delegating control to FluxCD.
and `upgradeFailures` are no longer set on `RedpandaStatus`, similar to their
61
-
behavior when `useFlux: false` was set.
62
-
*`gcr.io/kubebuilder/kube-rbac-proxy` container is deprecated and has been removed from the Redpanda
63
-
operator helm chart. The same ports will continue to serve metrics using kubebuilder's built in RBAC.
64
-
65
-
Any existing prometheus rules don't need to be adjusted.
66
-
67
-
For more details see: https://github.com/kubernetes-sigs/kubebuilder/discussions/3907
68
-
69
-
* The V1 operator now requires a minimum Redpanda version of 23.2; all feature-gated behaviour that supported older versions is now enabled unconditionally.
70
-
* The [`kube-prometheus-stack`](https://prometheus-community.github.io/helm-charts) subchart has been removed.
71
-
72
-
This integration was not being up kept and most use cases will be better served by deploying this chart themselves.
73
-
### Fixed
74
-
* Certificate reloading for webhook and metrics endpoints should now behave correctly.
75
-
* The operator will restart the redpanda cluster on any change to the cluster configuration
76
-
* Expanded the set of rules in both Roles and ClusterRoles to be appropriately in sync with the redpanda helm chart.
77
-
* DeprecatedFullNameOverride was interpreted differently between rendering resources and creating
78
-
kafka, admin and schema registry client. Now deprecated fullNameOverride will be used only
79
-
if correct FullNameOverride is not provided and handled the same way for both
80
-
client creation and render function.
81
-
* The Redpanda license was not set by operator. Now it will be set in the first reconciliation. After initial setup the consequent license re-set will be reconciled after client-go cache resync timeout (default 10h).
82
-
* The operator now unconditionally produces statefulsets that have environment variables available to the initContainer that are used for CEL-based config patching.
83
-
84
-
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.
85
-
* The operator now unconditionally produces an environment for the initContainer that supports CEL-based patching.
86
-
87
-
This is required to ensure that a pre-existing sts can roll over to new configuration correctly.
88
-
* Improved support for multi-STSes (e.g., multiple NodePools) in the ghost broker decommissioning logic.
89
-
90
-
- 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.
91
-
- 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.
92
-
- Disabled ordinal-based broker deletion logic in Operator v1 mode, as it doesn't work reliably in a multi-STS setup.
93
-
94
-
* Setting `serviceAccount.create` to `false` no longer prevents the Kubernetes ServiceAccountToken volume from being mounted to the operator Pod.
95
-
* 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.
96
-
* 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)
97
-
* Empty Redpanda specs (i.e. `clusterSpec: null` or otherwise unspecified) no longer causes the operator to panic
98
-
*`get` permissions on `Node` resources is now correctly configured by default.
99
-
100
-
`--set rbac.createAdditionalControllerCRs=true` is no longer required for rackawareness to work.
101
-
* Fixed a bug where pods would be restarted indefinitely in the case of config version changes when syncing cluster configuration.
0 commit comments