Update release notes draft to version v1.34.0-alpha.2

[Urvashi0109]

What type of PR is this?
/kind documentation

What this PR does / why we need it:
This PR updates the Release Notes Draft to k/k v1.34.0-alpha.2

Which issue(s) this PR fixes:

Special notes for your reviewer:
This is an automated PR generated from krel The Kubernetes Release Toolbox

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: Urvashi0109
Once this PR has been reviewed and has the lgtm label, please assign fsmunoz for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• releases/release-1.34/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[Vyom-Yadav]

@michellengnx @Urvashi0109 @ArvindParekh I'm confused about the procedure the docs team is following for generating/editing release notes, can y'all please clarify/correct me on that.

In this PR, the following release note has been edited using a map file: https://github.com/kubernetes/sig-release/pull/2809/files#diff-8b2e03503a6361c4c7c386629b92cbf09c0ab36eb95b21c2d50162d124d870e6

i.e.

pr: 132305
releasenote:
  text: Changed Job controller to use the controller UID index for Pod lookups to improve performance.
pr_body: ""

However, looking at both json and md file in this PR:

  "132305": {
    "commit": "8f1f17a04f62ab64ebe4f0b9d7f5f799bf56a0d9",
    "text": "Job controller used the controller UID index for Pod lookups.",
    "markdown": "Job controller used the controller UID index for Pod lookups. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))",
    "author": "xigang",
    "author_url": "https://github.com/xigang",
    "pr_url": "https://github.com/kubernetes/kubernetes/pull/132305",
    "pr_number": 132305,
    "kinds": [
      "cleanup"
    ],
    "sigs": [
      "apps"
    ],
    "is_mapped": true
  },

- Job controller used the controller UID index for Pod lookups. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))

I think the actual markdown/json file should have these changes as well. I know when you do alpha 3, these changes would be reflected then. The tool would pickup the notes file and update the md and json, but I don't think that is correct as that makes alpha2 notes not up to date.

I'm also not sure if we should be directly updating the md and json file (without a corresponding map). I'm unsure if that has some side affects (technically I don't think there will be, but still).

I was looking at a few past v1.33 PRs as well, and I've noticed this pattern there as well i.e. a maps file was added but the json and md weren't updated again and then in the next subsequent release cuts when krel was run again, it updated json and md to reflect those changed files in the last release notes. Which I don't think is correct.

I think the correct procedure to follow is: https://github.com/kubernetes/sig-release/blob/master/release-team/role-handbooks/docs/editing-flow.md

(I fully understand that editing files using a CLI might not be what everyone is used to and this should be clearer. I see maps as patches that are applied to edit the source release notes file)

[Vyom-Yadav]

In the krel logic I see the function fixReleaseNotes: (which is triggered by the --fix flag)

https://github.com/kubernetes/release/blob/3c696bde3c0cf41421ba18269641fc819b3b2501/cmd/krel/cmd/release_notes.go#L1214-L1222

		if noteMaps != nil {
			fmt.Println("✨ Note contents was previously modified with a map")

			for _, noteMap := range noteMaps {
				if err := note.ApplyMap(noteMap, true); err != nil {
					return fmt.Errorf("applying notemap for PR #%d: %w", pr, err)
				}
			}
		}

But I think this should be something in the linter rather than something for the next release notes PR. This is also an item on how to improve the experience with krel (or just even log that somewhere) so we can be more accurate.

[rashansmith]

https://github.com/kubernetes/sig-release/blob/master/release-team/role-handbooks/docs/editing-flow.md

Hey @Vyom-Yadav! These are great questions around the krel editing process. We currently use the cli to create the PRs, and also follow the editing flow guide. We should bring this topic up in the Release Notes channel for more insight into why this current flow exists, and maybe what we can do to improve it. I'll also tag @npolshakova @puerco @ramrodo @fsmunoz as well here for input, and remember to add this as a release retro item for continued discussion.

In the meantime, to keep the release notes drafts on schedule, I would recommend to continue with the PR approval process, unless this is viewed as something to be resolved immediately. Let me know if you have any questions!

[Vyom-Yadav]

Hey @rashansmith

Map != Release Note md/json. That is a blocking thing in my opinion. Alpha 2 release notes should be reflect the changes made for alpha 2 which is not the case right now as I mentioned in the above comment.

To move ahead with this PR, I'd recommend just running krel once more with --fix so that the current map changes are propagated to the markdown and json.

I'm ok with discussing the matter of directly editing md/json later.

[wendy-ha18]

This PR is LGTM except what @Vyom-Yadav pointed out about the updates in individual map PRs didn't reflect in release-notes-draft.json and release-notes-draft.md.

If we have a look in release notes draft to version v1.34.0-alpha.1, we can see how the next PR (which is the one) updates the Json and Mardown file.

[rashansmith]

Hey @rashansmith

Map != Release Note md/json. That is a blocking thing in my opinion. Alpha 2 release notes should be reflect the changes made for alpha 2 which is not the case right now as I mentioned in the above comment.

To move ahead with this PR, I'd recommend just running krel once more with --fix so that the current map changes are propagated to the markdown and json.

I'm ok with discussing the matter of directly editing md/json later.

@Vyom-Yadav here is the command that was used to create a PR:

krel release-notes --create-draft-pr --fork=Urvashi0109 --tag v1.34.0-alpha.2 --list-v2 --log-level trace --fix

To confirm, the --fix tag was used to create this PR. In this case, running the command again may not resolve your concern, and I suggest we continue with the flow we have now for this release.

[Vyom-Yadav]

Hmm, going through the source code and experimenting locally with krel, it is not designed for editing release notes after PR has been made kind of workflow. I was able to hack through and generate md and json files again. It wasn't pretty.

The problem with that was:

diff --git a/releases/release-1.34/release-notes/release-notes-draft.md b/releases/release-1.34/release-notes/release-notes-draft.md
index 9d4cb9f9b67..fcc2b692663 100644
--- a/releases/release-1.34/release-notes/release-notes-draft.md
+++ b/releases/release-1.34/release-notes/release-notes-draft.md
@@ -22,19 +22,19 @@
 
 ### API Change
 
+- Added `omitempty` and `opt` tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct. ([#132338](https://github.com/kubernetes/kubernetes/pull/132338), [@PatrickLaabs](https://github.com/PatrickLaabs))
 - Added a `runtime.ApplyConfiguration` interface implemented by all generated apply configuration types. ([#132194](https://github.com/kubernetes/kubernetes/pull/132194), [@alvaroaleman](https://github.com/alvaroaleman)) [SIG API Machinery and Instrumentation]
-- Added omitempty and opt tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct. ([#132338](https://github.com/kubernetes/kubernetes/pull/132338), [@PatrickLaabs](https://github.com/PatrickLaabs))
-- Corrected the documentation to clarify that `podSelector` is optional and to describe its default behavior. ([#131354](https://github.com/kubernetes/kubernetes/pull/131354), [@tomoish](https://github.com/tomoish))
+- Corrected the documentation to clarify that `podSelector` is optional and described its default behavior. ([#131354](https://github.com/kubernetes/kubernetes/pull/131354), [@tomoish](https://github.com/tomoish))
 - DRA API: resource.k8s.io/v1alpha3 now only contains DeviceTaintRule. All other types got removed because they became obsolete when introducing the v1beta1 API in 1.32.
   before updating a cluster where resourceclaims, resourceclaimtemplates, deviceclasses, or resourceslices might have been stored using Kubernetes < 1.32, delete all of those resources before updating and recreate them as needed while running Kubernetes >= 1.32. ([#132000](https://github.com/kubernetes/kuber
netes/pull/132000), [@pohly](https://github.com/pohly)) [SIG Etcd, Node, Scheduling and Testing]
 - DRA: Starting with Kubernetes 1.34, the alpha-level `resource.k8s.io/admin-access` label has been updated to `resource.kubernetes.io/admin-access`. Admins using the alpha feature and updating from 1.33 can set both labels, upgrade, then remove `resource.k8s.io/admin-access` when no downgrade is going to ha
ppen anymore. ([#131996](https://github.com/kubernetes/kubernetes/pull/131996), [@ritazh](https://github.com/ritazh)) [SIG Node and Testing]
 - Extended the NodePorts scheduling plugin to consider hostPorts used by restartable init containers. ([#132040](https://github.com/kubernetes/kubernetes/pull/132040), [@avrittrohwer](https://github.com/avrittrohwer)) [SIG Scheduling and Testing]
-- Introduced OpenAPI format support for `k8s-short-name` and `k8s-long-name` in `CustomResourceDefinition` schemas. ([#132504](https://github.com/kubernetes/kubernetes/pull/132504), [@jpbetz](https://github.com/jpbetz)) [SIG API Machinery, Architecture, Auth, CLI, Cloud Provider, Cluster Lifecycle, Instrumen
tation, Network, Node, Scheduling and Storage]
+- Introduced OpenAPI format support for `k8s-short-name` and `k8s-long-name` in CustomResourceDefinition schemas. ([#132504](https://github.com/kubernetes/kubernetes/pull/132504), [@jpbetz](https://github.com/jpbetz)) [SIG API Machinery, Architecture, Auth, CLI, Cloud Provider, Cluster Lifecycle, Instrumenta
tion, Network, Node, Scheduling and Storage]
 - Kube-apiserver: Added support for disabling caching of authorization webhook decisions in the `--authorization-config` file. The new fields `cacheAuthorizedRequests` and `cacheUnauthorizedRequests` could be set to `false` to prevent caching for authorized or unauthorized requests. See the https://kubernete
s.io/docs/reference/access-authn-authz/authorization/#using-configuration-file-for-authorization for more details. ([#129237](https://github.com/kubernetes/kubernetes/pull/129237), [@rfranzke](https://github.com/rfranzke)) [SIG API Machinery and Auth]
 - Kube-apiserver: Promoted the `StructuredAuthenticationConfiguration` feature gate to GA. ([#131916](https://github.com/kubernetes/kubernetes/pull/131916), [@aramase](https://github.com/aramase)) [SIG API Machinery, Auth and Testing]
 - Kube-apiserver: the AuthenticationConfiguration type accepted in `--authentication-config` files has been promoted to `apiserver.config.k8s.io/v1`. ([#131752](https://github.com/kubernetes/kubernetes/pull/131752), [@aramase](https://github.com/aramase)) [SIG API Machinery, Auth and Testing]
 - Kube-log-runner: Added the `-log-file-size` parameter to rotate log output into a new file once it reached a certain size. Introduced `-log-file-age` to enable automatic removal of old output files, and `-flush-interval` to support periodic flushing. ([#127667](https://github.com/kubernetes/kubernetes/pull
/127667), [@zylxjtu](https://github.com/zylxjtu)) [SIG API Machinery, Apps, Architecture, Auth, Autoscaling, CLI, Cloud Provider, Cluster Lifecycle, Etcd, Instrumentation, Network, Node, Release, Scheduling, Storage, Testing and Windows]
-- Kubectl: Graduated `kuberc` support to beta. A `kuberc` configuration file provided a mechanism for customizing `kubectl` behavior (distinct from kubeconfig, which configures cluster access across different clients). ([#131818](https://github.com/kubernetes/kubernetes/pull/131818), [@soltysh](https://githu
b.com/soltysh)) [SIG CLI and Testing]
+- Kubectl: Graduated kuberc support to beta. A `kuberc` configuration file provided a mechanism for customizing `kubectl` behavior (distinct from kubeconfig, which configures cluster access across different clients). ([#131818](https://github.com/kubernetes/kubernetes/pull/131818), [@soltysh](https://github.
com/soltysh)) [SIG CLI and Testing]
 - Promoted Job Pod Replacement Policy to general availability. The `JobPodReplacementPolicy` feature gate is now locked to true, and will be removed in a future release of Kubernetes. ([#132173](https://github.com/kubernetes/kubernetes/pull/132173), [@dejanzele](https://github.com/dejanzele)) [SIG Apps and T
esting]
 - Promoted the `RelaxedEnvironmentVariableValidation` feature gate to GA and locked it in the enabled state by default. ([#132054](https://github.com/kubernetes/kubernetes/pull/132054), [@HirazawaUi](https://github.com/HirazawaUi)) [SIG Apps, Architecture, Node and Testing]
 - Removed an inaccurate statement about requiring ports when the Pod spec `hostNetwork` field was set. ([#130994](https://github.com/kubernetes/kubernetes/pull/130994), [@BenTheElder](https://github.com/BenTheElder)) [SIG Network and Node]
@@ -65,7 +65,7 @@
   now allows an underscore (`_`) where a dash (`-`) would be allowed, and it allows search strings be a single dot `.`. ([#132036](https://github.com/kubernetes/kubernetes/pull/132036), [@adrianmoisey](https://github.com/adrianmoisey)) [SIG Network and Testing]
 - Graduated scheduler `QueueingHint` support to GA (general availability) ([#131973](https://github.com/kubernetes/kubernetes/pull/131973), [@sanposhiho](https://github.com/sanposhiho)) [SIG Scheduling and Testing]
 - Included namespace in the output of the kubectl delete for clearer identification of resources. ([#126619](https://github.com/kubernetes/kubernetes/pull/126619), [@totegamma](https://github.com/totegamma)) [SIG CLI]
-- Kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` now surfaced health checks named `etcd-override-<index>` and `etcd-override-readiness-<index>`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-readiness` directives. ([#129438](htt
ps://github.com/kubernetes/kubernetes/pull/129438), [@pacoxu](https://github.com/pacoxu)) [SIG API Machinery and Testing]
+- Kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` surfaced health checks named `etcd-override-<index>` and `etcd-override-readiness-<index>`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-readiness` directives. ([#129438](https:/
/github.com/kubernetes/kubernetes/pull/129438), [@pacoxu](https://github.com/pacoxu)) [SIG API Machinery and Testing]
 - Kube-apiserver: Promoted the `ExternalServiceAccountTokenSigner` feature to beta, which enabled external signing of service account tokens and fetching of public verifying keys. This was accomplished by enabling the beta `ExternalServiceAccountTokenSigner` feature gate and specifying the `--service-account
-signing-endpoint` flag. The flag value could either be the path to a Unix domain socket on the filesystem, or be prefixed with @ to indicate a Unix domain socket in the abstract namespace. ([#131300](https://github.com/kubernetes/kubernetes/pull/131300), [@HarshalNeelkamal](https://github.com/HarshalNeelkama
l)) [SIG API Machinery, Auth and Testing]
 - Kube-proxy: Checked whether IPv6 was available on Linux before using it. ([#131265](https://github.com/kubernetes/kubernetes/pull/131265), [@rikatz](https://github.com/rikatz))
 - Kubeadm: Added support for ECDSA-P384 as an encryption algorithm type in v1beta4. ([#131677](https://github.com/kubernetes/kubernetes/pull/131677), [@lalitc375](https://github.com/lalitc375))
@@ -96,7 +96,7 @@
 - Checked for newer resize fields when deciding the recovery feature status in the kubelet. ([#131418](https://github.com/kubernetes/kubernetes/pull/131418), [@gnufied](https://github.com/gnufied))
 - DRA: Ensured that ResourceClaims requesting a fixed number of devices with `adminAccess` were no longer allocated the same device multiple times. ([#131299](https://github.com/kubernetes/kubernetes/pull/131299), [@nojnhuh](https://github.com/nojnhuh))
 - Disabled reading of disk geometry before calling expansion for ext and xfs filesystems. ([#131568](https://github.com/kubernetes/kubernetes/pull/131568), [@gnufied](https://github.com/gnufied))
-- Fixed API response for `StorageClassList` queries and returns a graceful error message, if the provided `ResourceVersion` is too large. ([#132374](https://github.com/kubernetes/kubernetes/pull/132374), [@PatrickLaabs](https://github.com/PatrickLaabs)) [SIG API Machinery and Etcd]
+- Fixed API response for StorageClassList queries and returns a graceful error message, if the provided ResourceVersion is too large. ([#132374](https://github.com/kubernetes/kubernetes/pull/132374), [@PatrickLaabs](https://github.com/PatrickLaabs)) [SIG API Machinery and Etcd]
 - Fixed ReplicationController reconciliation when the `DeploymentReplicaSetTerminatingReplicas` feature gate was enabled. ([#131822](https://github.com/kubernetes/kubernetes/pull/131822), [@atiratree](https://github.com/atiratree))
 - Fixed a bug in CEL's common.UnstructuredToVal where `==` evaluates to false for identical objects when a field is present but the value is null.  This bug does not impact the Kubernetes API. ([#131559](https://github.com/kubernetes/kubernetes/pull/131559), [@jpbetz](https://github.com/jpbetz)) [SIG API Mac
hinery]
 - Fixed a bug in the Job controller that could result in creating unnecessary Pods for Jobs already marked as finished (either successful or failed). ([#130333](https://github.com/kubernetes/kubernetes/pull/130333), [@kmala](https://github.com/kmala)) [SIG Apps and Testing]
@@ -106,9 +106,9 @@
 - Fixed a possible deadlock in the watch client that could happen if the watch was not stopped. ([#131266](https://github.com/kubernetes/kubernetes/pull/131266), [@karlkfi](https://github.com/karlkfi)) [SIG API Machinery]
 - Fixed a regression introduced in 1.33 where some paginated LIST calls fell back to `etcd` instead of being served from cache. ([#132244](https://github.com/kubernetes/kubernetes/pull/132244), [@hakuna-matatah](https://github.com/hakuna-matatah))
 - Fixed an incorrect reference to `JoinConfigurationKind` in the error message when no ResetConfiguration is found during `kubeadm reset` with the `--config` flag. ([#132258](https://github.com/kubernetes/kubernetes/pull/132258), [@J3m3](https://github.com/J3m3)) [SIG Cluster Lifecycle]
-- Fixed an issue that allowed Custom Resources to be created using Server-Side, apply even when their `CustomResourceDefinition` was terminating. ([#132467](https://github.com/kubernetes/kubernetes/pull/132467), [@sdowell](https://github.com/sdowell))
 - Fixed an issue where Windows kube-proxy’s `ModifyLoadBalancer` API updates did not match the HNS state in version 15.4. Support for `ModifyLoadBalancer` policy began with Kubernetes 1.31+. ([#131506](https://github.com/kubernetes/kubernetes/pull/131506), [@princepereira](https://github.com/princepereira))
 - Fixed an issue where `insufficientResources` was logged as a pointer during pod preemption, making logs more readable. ([#132183](https://github.com/kubernetes/kubernetes/pull/132183), [@chrisy-x](https://github.com/chrisy-x)) [SIG Node]
+- Fixed an issue which allowed Custom Resources to be created with Server-Side Apply even when its CustomResourceDefinition was terminating. ([#132467](https://github.com/kubernetes/kubernetes/pull/132467), [@sdowell](https://github.com/sdowell)) [SIG API Machinery]
 - Fixed incorrect behavior for AllocationMode: All in ResourceClaim when used in subrequests. ([#131660](https://github.com/kubernetes/kubernetes/pull/131660), [@mortent](https://github.com/mortent)) [SIG Node]
 - Fixed misleading response codes in admission control metrics. ([#132165](https://github.com/kubernetes/kubernetes/pull/132165), [@gavinkflam](https://github.com/gavinkflam)) [SIG API Machinery, Architecture and Instrumentation]
 - Fixed the `allocatedResourceStatuses` field name mismatch in PVC status validation. ([#131213](https://github.com/kubernetes/kubernetes/pull/131213), [@carlory](https://github.com/carlory))
@@ -136,12 +136,12 @@
 - Bumped kube-dns to v1.26.4. ([#132012](https://github.com/kubernetes/kubernetes/pull/132012), [@pacoxu](https://github.com/pacoxu))
 - Bumped the cel-go dependency to v0.25.0. The changeset is available at: https://github.com/google/cel-go/compare/v0.23.2...v0.25.0. ([#131444](https://github.com/kubernetes/kubernetes/pull/131444), [@erdii](https://github.com/erdii)) [SIG API Machinery, Auth, Cloud Provider and Node]
 - By default, binaries like kube-apiserver were built with the `grpcnotrace` tag enabled. Used the `DBG` flag to enable Golang tracing if needed. ([#132210](https://github.com/kubernetes/kubernetes/pull/132210), [@dims](https://github.com/dims))
+- Changed Job controller to use the controller UID index for Pod lookups to improve performance. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))
 - Changed apiserver to treat failures decoding a mutating webhook patch as failures to call the webhook so they trigger the webhook failurePolicy and count against metrics like `webhook_fail_open_count` ([#131627](https://github.com/kubernetes/kubernetes/pull/131627), [@dims](https://github.com/dims)) [SIG A
PI Machinery]
 - DRA kubelet- Updated logging to use `driverName` instead of `pluginName`, aligning with the rest of the Kubernetes components. ([#132096](https://github.com/kubernetes/kubernetes/pull/132096), [@pohly](https://github.com/pohly)) [SIG Node and Testing]
 - DRA kubelet: Simplified recovery from mistakes like scheduling a Pod onto a node where the required driver was not running, as the kubelet no longer unnecessarily blocked Pod deletion. ([#131968](https://github.com/kubernetes/kubernetes/pull/131968), [@pohly](https://github.com/pohly)) [SIG Node and Testin
g]
 - Fixed some missing white spaces in the flag descriptions and logs. ([#131562](https://github.com/kubernetes/kubernetes/pull/131562), [@logica0419](https://github.com/logica0419)) [SIG Network]
 - Hack/update-codegen.sh now automatically ensured goimports and protoc. ([#131459](https://github.com/kubernetes/kubernetes/pull/131459), [@BenTheElder](https://github.com/BenTheElder))
-- Job controller used the controller UID index for Pod lookups. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))
 - Kube-apiserver: removed the deprecated `apiserver_encryption_config_controller_automatic_reload_success_total` and `apiserver_encryption_config_controller_automatic_reload_failure_total` metrics in favor of `apiserver_encryption_config_controller_automatic_reloads_total`. ([#132238](https://github.com/kube
rnetes/kubernetes/pull/132238), [@aramase](https://github.com/aramase)) [SIG API Machinery, Auth and Testing]
 - Kube-scheduler: removed the deprecated scheduler_scheduler_cache_size metric  in favor of scheduler_cache_size ([#131425](https://github.com/kubernetes/kubernetes/pull/131425), [@carlory](https://github.com/carlory)) [SIG Scheduling]
 - Kubeadm: fixed missing space when printing the warning about pause image mismatch. ([#131563](https://github.com/kubernetes/kubernetes/pull/131563), [@logica0419](https://github.com/logica0419)) [SIG Cluster Lifecycle]
@@ -158,9 +158,9 @@
 - Removed support for API streaming from the metadata client’s `List() method`. ([#132149](https://github.com/kubernetes/kubernetes/pull/132149), [@p0lyn0mial](https://github.com/p0lyn0mial)) [SIG API Machinery and Testing]
 - Removed the `kubernetes.io/initial-events-list-blueprint` annotation from the synthetic "Bookmark" event in watch stream requests. ([#132326](https://github.com/kubernetes/kubernetes/pull/132326), [@p0lyn0mial](https://github.com/p0lyn0mial))
 - Removed the deprecated `--register-schedulable` command line argument from the kubelet. ([#122384](https://github.com/kubernetes/kubernetes/pull/122384), [@carlory](https://github.com/carlory)) [SIG Cloud Provider, Node and Scalability]
-- The conntrack reconciler now considered a Service’s target port during cleanup of stale flow entries. ([#130542](https://github.com/kubernetes/kubernetes/pull/130542), [@aroradaman](https://github.com/aroradaman))
 - Types: CycleState, StateData, StateKey and ErrNotFound moved from pkg/scheduler/framework to k8s.io/kube-scheduler/framework.
   Type CycleState that is passed to each plugin in scheduler framework is changed to the new interface CycleState (in k8s.io/kube-scheduler/framework) ([#131887](https://github.com/kubernetes/kubernetes/pull/131887), [@ania-borowiec](https://github.com/ania-borowiec)) [SIG Node, Scheduling, Storage and Testi
ng]
 - Updated CNI plugins to v1.7.1 ([#131602](https://github.com/kubernetes/kubernetes/pull/131602), [@adrianmoisey](https://github.com/adrianmoisey)) [SIG Cloud Provider, Node and Testing]
+- Updated `conntrack` reconciler to consider a Service’s target port during cleanup of stale flow entries. ([#130542](https://github.com/kubernetes/kubernetes/pull/130542), [@aroradaman](https://github.com/aroradaman))
 - Updated cri-tools to v1.33.0. ([#131406](https://github.com/kubernetes/kubernetes/pull/131406), [@saschagrunert](https://github.com/saschagrunert)) [SIG Cloud Provider]
 - Upgraded CoreDNS to v1.12.1. ([#131151](https://github.com/kubernetes/kubernetes/pull/131151), [@yashsingh74](https://github.com/yashsingh74)) [SIG Cloud Provider and Cluster Lifecycle]
\ No newline at end of file

Because the md file was directly edited without a map, running krel on it again reverts those changes (as they weren't recorded in a map). But it also picked the changes made to the map in the PR to update the md and json files.

Going with the current workflow can be problematic when the changes that were made without a map get reverted and need human intervention to manually correct those - the docs team would have to make sure they manually revert changes that weren't made through a map.

The session files are also weird to me. Only upstream session files are fetched and not the ones in the current PR, which is weird (and if I read code correctly, wrong session mechanism - you lose everything you reviewed ). The session files are taken from the clone that krel creates. (I haven't checked what's the complete interaction with --maps-from)

Overall, krel needs modifications to make the release notes creation experience better. If you'd like to manually update the json and md for this PR, here is the updated json (with changes that need to reverted as md/json have been directly modified). It's also fine to do these updates in alpha.3.

diff --git a/releases/release-1.34/release-notes/release-notes-draft.json b/releases/release-1.34/release-notes/release-notes-draft.json
index 70350ebeb5a..610e3efcc85 100644
--- a/releases/release-1.34/release-notes/release-notes-draft.json
+++ b/releases/release-1.34/release-notes/release-notes-draft.json
@@ -243,8 +243,8 @@
   },
   "129438": {
     "commit": "b569406b792fef24bb5613f1263ea354755bbae0",
-    "text": "kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` now surfaced health checks named `etcd-override-\u003cindex\u003e` and `etcd-override-readiness-\u003cindex\u003e`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-readin
ess` directives.",
-    "markdown": "Kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` now surfaced health checks named `etcd-override-\u003cindex\u003e` and `etcd-override-readiness-\u003cindex\u003e`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-re
adiness` directives. ([#129438](https://github.com/kubernetes/kubernetes/pull/129438), [@pacoxu](https://github.com/pacoxu)) [SIG API Machinery and Testing]",
+    "text": "kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` surfaced health checks named `etcd-override-\u003cindex\u003e` and `etcd-override-readiness-\u003cindex\u003e`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-readiness`
 directives.",
+    "markdown": "Kube-apiserver: Each unique set of etcd server overrides specified with `--etcd-servers-overrides` surfaced health checks named `etcd-override-\u003cindex\u003e` and `etcd-override-readiness-\u003cindex\u003e`. These checks were still excluded by the `?exclude=etcd` and `?exclude=etcd-readin
ess` directives. ([#129438](https://github.com/kubernetes/kubernetes/pull/129438), [@pacoxu](https://github.com/pacoxu)) [SIG API Machinery and Testing]",
     "author": "pacoxu",
     "author_url": "https://github.com/pacoxu",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/129438",
@@ -409,8 +409,8 @@
   },
   "130542": {
     "commit": "e43ab77aedfa9257022e90642fed0034d937056e",
-    "text": "The conntrack reconciler now considered a Service’s target port during cleanup of stale flow entries.",
-    "markdown": "The conntrack reconciler now considered a Service’s target port during cleanup of stale flow entries. ([#130542](https://github.com/kubernetes/kubernetes/pull/130542), [@aroradaman](https://github.com/aroradaman))",
+    "text": "Updated `conntrack` reconciler to consider a Service’s target port during cleanup of stale flow entries.",
+    "markdown": "Updated `conntrack` reconciler to consider a Service’s target port during cleanup of stale flow entries. ([#130542](https://github.com/kubernetes/kubernetes/pull/130542), [@aroradaman](https://github.com/aroradaman))",
     "author": "aroradaman",
     "author_url": "https://github.com/aroradaman",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/130542",
@@ -845,8 +845,8 @@
   },
   "131354": {
     "commit": "354520cd9b8060a33f77bd62357e78f89358807a",
-    "text": "Corrected the documentation to clarify that `podSelector` is optional and to describe its default behavior.",
-    "markdown": "Corrected the documentation to clarify that `podSelector` is optional and to describe its default behavior. ([#131354](https://github.com/kubernetes/kubernetes/pull/131354), [@tomoish](https://github.com/tomoish))",
+    "text": "Corrected the documentation to clarify that `podSelector` is optional and described its default behavior.",
+    "markdown": "Corrected the documentation to clarify that `podSelector` is optional and described its default behavior. ([#131354](https://github.com/kubernetes/kubernetes/pull/131354), [@tomoish](https://github.com/tomoish))",
     "author": "tomoish",
     "author_url": "https://github.com/tomoish",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/131354",
@@ -2715,8 +2715,8 @@
   },
   "132305": {
     "commit": "8f1f17a04f62ab64ebe4f0b9d7f5f799bf56a0d9",
-    "text": "Job controller used the controller UID index for Pod lookups.",
-    "markdown": "Job controller used the controller UID index for Pod lookups. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))",
+    "text": "Changed Job controller to use the controller UID index for Pod lookups to improve performance.",
+    "markdown": "Changed Job controller to use the controller UID index for Pod lookups to improve performance. ([#132305](https://github.com/kubernetes/kubernetes/pull/132305), [@xigang](https://github.com/xigang))",
     "author": "xigang",
     "author_url": "https://github.com/xigang",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/132305",
@@ -2771,8 +2771,8 @@
   },
   "132338": {
     "commit": "70c25cbe0c01bb49dd09b8c5497b8d2299a82abe",
-    "text": "Added omitempty and opt tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct.",
-    "markdown": "Added omitempty and opt tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct. ([#132338](https://github.com/kubernetes/kubernetes/pull/132338), [@PatrickLaabs](https://github.com/PatrickLaabs))",
+    "text": "Added `omitempty` and `opt` tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct.",
+    "markdown": "Added `omitempty` and `opt` tag to the API `v1beta2` AdminAccess type in the `DeviceRequestAllocationResult` struct. ([#132338](https://github.com/kubernetes/kubernetes/pull/132338), [@PatrickLaabs](https://github.com/PatrickLaabs))",
     "author": "PatrickLaabs",
     "author_url": "https://github.com/PatrickLaabs",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/132338",
@@ -2804,8 +2804,8 @@
   },
   "132374": {
     "commit": "49bff137a7211ff300028bfdba3ce1929a5dfb2b",
-    "text": "Fixed API response for `StorageClassList` queries and returns a graceful error message, if the provided `ResourceVersion` is too large.",
-    "markdown": "Fixed API response for `StorageClassList` queries and returns a graceful error message, if the provided `ResourceVersion` is too large. ([#132374](https://github.com/kubernetes/kubernetes/pull/132374), [@PatrickLaabs](https://github.com/PatrickLaabs)) [SIG API Machinery and Etcd]",
+    "text": "Fixed API response for StorageClassList queries and returns a graceful error message, if the provided ResourceVersion is too large.",
+    "markdown": "Fixed API response for StorageClassList queries and returns a graceful error message, if the provided ResourceVersion is too large. ([#132374](https://github.com/kubernetes/kubernetes/pull/132374), [@PatrickLaabs](https://github.com/PatrickLaabs)) [SIG API Machinery and Etcd]",
     "author": "PatrickLaabs",
     "author_url": "https://github.com/PatrickLaabs",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/132374",
@@ -2820,8 +2820,7 @@
       "api-machinery",
       "etcd"
     ],
-    "duplicate": true,
-    "is_mapped": true
+    "duplicate": true
   },
   "132376": {
     "commit": "885db534cabb9c1975553f43e32c90f0be256a51",
@@ -2841,8 +2840,8 @@
   },
   "132467": {
     "commit": "ec80dcaf3892d0cf89e4fd7b1ea60300fb417d8a",
-    "text": "Fixed an issue that allowed Custom Resources to be created using Server-Side, apply even when their `CustomResourceDefinition` was terminating.",
-    "markdown": "Fixed an issue that allowed Custom Resources to be created using Server-Side, apply even when their `CustomResourceDefinition` was terminating. ([#132467](https://github.com/kubernetes/kubernetes/pull/132467), [@sdowell](https://github.com/sdowell))",
+    "text": "Fixed an issue which allowed Custom Resources to be created with Server-Side Apply even when its CustomResourceDefinition was terminating.",
+    "markdown": "Fixed an issue which allowed Custom Resources to be created with Server-Side Apply even when its CustomResourceDefinition was terminating. ([#132467](https://github.com/kubernetes/kubernetes/pull/132467), [@sdowell](https://github.com/sdowell)) [SIG API Machinery]",
     "author": "sdowell",
     "author_url": "https://github.com/sdowell",
     "pr_url": "https://github.com/kubernetes/kubernetes/pull/132467",
@@ -2852,8 +2851,7 @@
     ],
     "sigs": [
       "api-machinery"
-    ],
-    "is_mapped": true
+    ]
   },
   "132504": {
     "commit": "1d932bd6cc951b9182d07d701946aebaf667df94",

[rashansmith]

@Vyom-Yadav yes, if I remember correctly during my time with release notes we had set aside time to manually correct these inconsistencies.

It can definitely be an improvement to the krel tool, and in the meantime @Urvashi0109 can make the changes manually.

[Urvashi0109]

Hey team, instead of doing manual edit wherein there may be chances of errors, I'll be creating the PR from scratch again with relevant suggestions incorporated within the map file itself and then having krel do the required changes to md and json file itself

CC @michellengnx, @rashansmith, @Vyom-Yadav