new lastmin

CHANGELOG.md

@@ -16,19 +16,17 @@ Monitor the release status by regions at [AKS-Release-Tracker](https://releases.
 
 * Preview Features:
   * AKS version [1.30](https://kubernetes.io/blog/2024/04/17/kubernetes-v1-30-release/) is available in preview. 
-
-* Behavioral Changes:
-  * Azure Service Mesh-1-19 is no longer supported. If this revision is in use, please upgrade for continued support. More information about mesh upgrades and version support can be found [here](https://aka.ms/asm-aks-upgrade-docs).
-
-* Bug Fixes:
+  * Istio service mesh addon revision asm-1-19 is no longer supported. If you are still using this revision on your cluster, please upgrade for continued support. More information about mesh upgrades and version support can be found [here](https://aka.ms/asm-aks-upgrade-docs).
+
+ * Bug Fixes:
   * CoreDNS has been updated to use image [v1.9.4-hotfix.20240520](https://github.com/aks-lts/coredns/tree/release-1.9) on all AKS clusters above version 1.24. This updated image addresses [CVE vulnerabilities](https://github.com/aks-lts/coredns/commit/ba698d28c2ab8d9db0951592be631885e4134e5).
   * Updated cilium to version 1.14.10 for K8s version 1.29+, to fix the issue [where the host network is broken and remains broken even if the underlying interface goes up again](https://github.com/cilium/cilium/issues/18706).
   * Removes the post-upgrade annotation on [hubble-generate-cert](https://github.com/cilium/cilium/blob/aa10df3a4c6a9e7bd947a4a32613cedf22b3731d/Documentation/gettingstarted/hubble-configuration.rst#L81) Job. On each aks cluster reconcile, the helm chart revision is incremented which counts as an upgrade. Each time the helm chart is upgraded or installed this job will restart. This change fixes that to not restart on helm chart upgrades and successfully clean up.
    * Windows containerd has been upgraded from v1.7.14 to v1.7.17 in K8s v1.28+. This upgrade fixes two bugs resulting in a wrong [default path](https://github.com/Azure/AKS/issues/4196) and a [deadlock issue](https://github.com/containerd/ttrpc/issues/72#issuecomment-2105545516).
-   * Fixed the following issues for [AKS Edge zone support](https://learn.microsoft.com/en-us/azure/aks/edge-zones?tabs=azure-resource-manager) -
-      * Fixed bug where clusters with ExtendedLocation set would accept create AgentPool with AvailabilityZones even though AvailabilityZones aren't supported in ExtendedLocation mode. 
-      * Fixed bug where we accepted edgezone case, when it should have been `EdgeZone`.
-      * Fixed a related bug where even when the user has specified `EdgeZone`, we were mistakenly resuming edgezone.
+   * Fixed the following issues for [AKS Edge zone support](https://learn.microsoft.com/azure/aks/edge-zones?tabs=azure-resource-manager) -
+      * Fixed bug where clusters with ExtendedLocation set would accept create node pool with availability zones even though availability zones aren't supported in ExtendedLocation mode. 
+      * Fixed bug where `edgezone` was previously being wrongly accepted in small case. Only `EdgeZone` is accepted.
+
 
 * Component Updates:
   * Changing [cilium operator](https://docs.cilium.io/en/stable/internals/cilium_operator/) tolerations to match cilium-agent. Adding tolerations for NoExecute and NoSchedule. This should fix a race condition in upgrades, where cilium-operator cannot schedule due to node taint.