GKE cluster upgrade guide [Linux]
(On-premise) Upgrading Kubernetes for Google Kubernetes Engine (GKE) is essential for maintaining optimal performance and security. Regular updates ensure access to the latest features, enhancements, and critical security patches. Each Sisense version supports the exact list of Kubernetes versions so it’s crucial to be on a supported Kubernetes version before the Sisense upgrade or installation.
Step-by-Step Guide:
We expect that you have already installed and initialized gcloud according to the general requirements for the Sisense installation on GKE.
1. Get the available Kubernetes version in your region/zone, e.g.
gcloud container get-server-config --zone us-east1-b or
gcloud container get-server-config --region us-east1
2. Check the Kubernetes versions supported by your current Sisense version here:
https://docs.sisense.com/main/SisenseLinux/linux-minimum-requirements.htm#CompatibilityMatrix
3. Upgrade the control plane of the cluster with
gcloud container clusters upgrade --zone us-east1-b CLUSTER_NAME --master --cluster-version VERSION
If the cluster is regional use --region instead of --zone
In case you don’t remember the exact cluster name, feel free to use the following command to list available clusters in a zone/region:
gcloud container clusters list --zone us-east1-b
(If the cluster is regional, use --region instead of --zone)
4. To update all nodes to the same version as the control plane, use the next command:
gcloud container clusters upgrade --zone us-east1-b CLUSTER_NAME
If your cluster contains more than one nodepool you’ll need to update the nodepools one by one.
Updating the particular node-pool would be like:
gcloud container clusters upgrade --zone us-east1-b --node-pool=YOUR_NODEPOOL_NAME CLUSTER_NAME
Make sure to update other node pools using the same scenario.
If a node-pool upgrade step takes too much time (e.g., more than 30 minutes), it’s worth checking the nodes:
and describe the node which is being processed (SchedulingDisabled) in our case
kubectl describe node NODE_NAME
There can be 1-2 Sisense pods that we should delete manually, e.g.
kubectl -n sisense delete pod exporting-578f8bccff-mdn4f
Conclusion:
This guide describes the regular flow of GKE cluster updates that you may need while running the Sisense application on Google Cloud. If you face difficulties with the process, we would suggest you check Google Cloud documentation or reach the Google Cloud support team.
References/Related Content
- https://cloud.google.com/kubernetes-engine/docs/how-to/upgrading-a-cluster
- https://docs.sisense.com/main/SisenseLinux/deploying-sisense-on-google-gke.htm
Disclaimer: This post outlines a potential custom workaround for a specific use case or provides instructions regarding a specific task. The solution may not work in all scenarios or Sisense versions, so we strongly recommend testing it in your environment before deployment. If you need further assistance with this, please let us know.
