rezaebrahimzadeh commented on issue #10422: URL: https://github.com/apache/apisix/issues/10422#issuecomment-2953605304
Hi team, I'm currently using APISIX in a Kubernetes cluster, deployed via Helm. I have the following components enabled: - etcd - apisix - apisix-ingress-controller - dashboard Everything works fine in general, but I've encountered an issue where APISIX continues to send requests to old pod IPs when a backend pod restarts or its IP changes. This causes request failures until APISIX updates the upstream IP list. I expected that using Kubernetes Services as upstreams would solve this, but I’m still seeing requests being routed to stale IPs. I saw related issues like: - https://github.com/apache/apisix-ingress-controller/issues/2341 My questions are: 1. What is the **recommended and reliable way** to ensure APISIX always routes to the **current pod IPs** in a Kubernetes environment? 2. Is there any **specific configuration** I should apply to upstream definitions or the ingress controller to ensure proper endpoint syncing? 3. Would enabling **active health checks** or adjusting `dns_validity` help in this case, even when using Kubernetes Services? I'd appreciate any guidance or best practices on how to configure APISIX in Kubernetes to handle pod IP changes gracefully. Thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: notifications-unsubscr...@apisix.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
