If you are reporting any crash or any potential security issue, do not
open an issue in this repo. Please report the issue via emailing
envoy-security@googlegroups.com where the issue will be triaged appropriately.
Title: Envoy pod stuck with dns time outs in kuberenetes when kubedns service restarts and point to a new IP
Description:
We deployed Envoy proxy as a forward http proxy container in kubernetes with "STRICT DNS" resolution. We observed that after kubernetes kubedns service restartes with new IP, Envoy pod unable to detect new kubedns ip during dns resolution. It fails all requests with dns timeout error. Is there any configuration in Envoy to forcefully reset dns socket after a no of timeouts so that it can check the iptables and pick new up new DNS server IP for future DNS resolutions?
Now our only option to recover envou is to restart the container.
ggreenway
commented
1 day ago
If you are reporting any crash or any potential security issue, do not open an issue in this repo. Please report the issue via emailing envoy-security@googlegroups.com where the issue will be triaged appropriately.
Title: Envoy pod stuck with dns time outs in kuberenetes when kubedns service restarts and point to a new IP
Description: We deployed Envoy proxy as a forward http proxy container in kubernetes with "STRICT DNS" resolution. We observed that after kubernetes kubedns service restartes with new IP, Envoy pod unable to detect new kubedns ip during dns resolution. It fails all requests with dns timeout error. Is there any configuration in Envoy to forcefully reset dns socket after a no of timeouts so that it can check the iptables and pick new up new DNS server IP for future DNS resolutions?
Now our only option to recover envou is to restart the container.
[optional Relevant Links:] Same issue as https://github.com/envoyproxy/envoy/issues/7965