← sources

Need advice: Kubernetes 1.34 → 1.35 upgrade with Elasticsearch running in the cluster

source: reddit (reddit:kubernetes)published 2026-10-04 17:19ingested 2026-10-05 02:40

retained text

Hi everyone,

I’m planning a Kubernetes cluster upgrade from v1.34.8 to v1.35, as the current version is approaching retirement/EOL.

The cluster currently has Elasticsearch running on it, so I’m concerned about potential compatibility or stability issues after upgrading Kubernetes.

I’d like some guidance on:  
\- How can I verify Elasticsearch compatibility with Kubernetes 1.35 before upgrading?  
\- Are there any known issues I should be aware of when upgrading Kubernetes while Elasticsearch is running?  
\- What should I check across Elasticsearch, ECK/operator, StatefulSets, storage classes/CSI drivers, ingress, CRDs, APIs, and other dependencies?  
\- What is the recommended pre-upgrade checklist?  
\- Should I upgrade Elasticsearch/ECK first, or Kubernetes first?  
\- What kind of backup/rollback strategy should I have?  
\- Is it better to upgrade the control plane and worker nodes separately?  
\- What would be a safe upgrade plan for a production cluster to minimize downtime/risk?  
\- Are there any tools or commands you recommend for detecting deprecated APIs or incompatible workloads before the upgrade?

If anyone has gone through a similar Kubernetes 1.34 → 1.35 upgrade with Elasticsearch, I’d really appreciate hearing about your experience and any issues you encountered.

Thanks! 

document fetched 2026-10-05 00:33 · canonical https://reddit.com/r/kubernetes/comments/1wxl3ma/need_advice_kubernetes_134_135_upgrade_with