Recieved this message "This cluster is in a failed state. If you didn't do an operation, AKS may resolve the provisioning status automatically ..." but monitored for a few hours and pods are good and otherwise solution is good.

Question

CONTEXT:
-While using our current AKS solution we discover that some of the pods where down and the age of the nodes correlated with an update initiated by Azure Container Services. Or at least that was the only log that correlated with the age on the nodes. (Solution is a month or to old, but the nodes where 40 hours)

-Rebuild the feature on the front end and the pods came back up.

-Disabled the automatic update feature for AKS through settings>cluster configuration.

-The disablement of the update feature produced a message (This cluster is in a failed state. If you didn't do an operation, AKS may resolve the provisioning status automatically if your cluster applications continue to run.)

-However, after further testing and monitoring the solution via K9s for a couple of hours there is no further direct evidence of service disruption on the pods or the solution itself.

QUESTION:
What would be the cleanest method of getting rid of the message without service disruption.

Answer 1

Hello EDWARD REYESARROYO,

I believe you are looking for high availability when using workloads running on your AKS cluster.

For that please visit the following links:

1-https://learn.microsoft.com/en-us/azure/architecture/guide/aks/aks-high-availability

2-https://learn.microsoft.com/en-us/azure/aks/best-practices-app-cluster-reliability

3-https://learn.microsoft.com/en-us/azure/aks/ha-dr-overview

If it was helpful, please click "Upvote" Accept the answer on this post to let us know.

Thank You.

Answer 2

Hello EDWARD REYESARROYO,

I wanted to check if you had the opportunity to review the information which was provided in my previous comment.

If it was helpful, please click "Upvote" on this post to let us know.

Thank You.