Closed kvudata closed 6 years ago
Have the same issue with same environment
@kvudata @kirgene Is it duplication of https://github.com/istio/istio/issues/5806 ?
@ymesika, perhaps. That issue is specifically mentioning timeouts on the Gateway (which I did see as well) while this issue is in particular about hangs/timeouts between 2 proxy's.
@ymesika Seems to be. Upgrading to 1.10.4-gke.0 (as mentioned in https://github.com/istio/istio/issues/5806) fixes this issue.
@kirgene Thanks for verifying.
Is this a BUG or FEATURE REQUEST?: BUG
Did you review https://istio.io/help/ and existing issues to identify if this is already solved or being worked on?: Yes, #152 seems potentially related but there were no error logs I could see in the proxy like that issue.
Bug: Y
What Version of Istio and Kubernetes are you using, where did you get Istio from, Installation details
Installed Istio using istio.io Kubernetes Quick Start instructions and following the directions to Render Kubernetes manifest with Helm and deploy with kubectl. Also tried installing via
helm install
.Is Istio Auth enabled or not ?
What happened: Everything seemed to have installed fine i.e. all the deployments in the istio-system namespace were available, but pods with the proxy side car in the cluster were unreachable. Tried following the Basic Authentication Policy Task but the first curl just hangs. I verified that the proxied service was working within the pod by exec'ing into the proxy container and curl'ing localhost. I also checked that the proxy-config's of the two proxy's had inbound/outbound clusters as expected.
Everything seems to work though on a cluster running k8s 1.9.7-gke.1
What you expected to happen: Be able to connect to the proxied containers.
How to reproduce it: Create a v1.10.2-gke.3 GKE cluster, install Istio, add some proxied containers.
Feature Request: N
Describe the feature: N/A