Skip to content

Commit

Permalink
Merge pull request #4361 from kaarthis/rel-20240609
Browse files Browse the repository at this point in the history
Minor edit
  • Loading branch information
kaarthis authored Jul 2, 2024
2 parents 993cc4f + b7d057c commit 4ef5aac
Showing 1 changed file with 5 additions and 2 deletions.
7 changes: 5 additions & 2 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,9 @@ Monitor the release status by regions at [AKS-Release-Tracker](https://releases.
### Announcements
* Starting 1.30 Kubernetes version and 1.27 LTS versions, beta apis will be disabled by default, when you upgrade to [them](https://learn.microsoft.com/azure/aks/upgrade-aks-cluster?tabs=azure-cli#before-you-begin). There will be an option provided to explicitly enable beta apis closer to the 1.30 release.
* 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).
* 1.30 is the next LTS version after 1.27. Customers will have at least 6 months of migration time available between 1.27 LTS to 1.30 LTS i.e before the 1.27 LTS goes End of Life by July 2025. Refer [long term support](https://learn.microsoft.com//azure/aks/long-term-support) for more information.
* Container Insights has automatically migrated from legacy authentication to managed authentication on AKS clusters where the Container Insights addon was enabled with legacy authentication. This migration occurs when any feature, such as the cost-analysis addon or authorized IP ranges, is enabled using the preview API version 2023-07-02-preview or later. This unintended migration has caused monitoring to break, this issue has been fixed for new clusters. To mitigate this issue on existing clusters, re-onboarding or re-configuring of Container Insights is required.
* Released improvements to [planned maintenance schedule](https://learn.microsoft.com/azure/aks/planned-maintenance), so that there are no missed or delayed windows. Also you no longer have to wait for 24 hours between the creation or update of a maintenance configuration and the scheduled start time, this will now only take few minutes going forward.

### Release Notes

Expand All @@ -18,15 +21,15 @@ Monitor the release status by regions at [AKS-Release-Tracker](https://releases.
* AKS version [1.30](https://kubernetes.io/blog/2024/04/17/kubernetes-v1-30-release/) is available in preview.

* 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 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/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:
* Windows containerd has been upgraded from v1.7.14 to v1.7.17 in K8s v1.28+.
* 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.
* Retina Enterprise and Operator image update [v0.0.8](https://github.com/azure-networking/retina-enterprise/releases/tag/v0.0.8).
* Updated linux cni versions to [v1.4.54](https://github.com/Azure/azure-container-networking/releases/tag/v1.4.54) and [v1.5.28](https://github.com/Azure/azure-container-networking/releases/tag/v1.5.28).
Expand Down

0 comments on commit 4ef5aac

Please sign in to comment.