Oliver Gould

While work on this specific feature hasn't moved forward, we do have some work in flight that will be foundational to implementing a feature like this. The [new policy CRDs](https://github.com/linkerd/linkerd2/pull/8007)...

I'm unable to reproduce this on the latest edge: ```console :; k delete po -n feh feh-server-10000b10x-65c59865dd-w8plw pod "feh-server-10000b10x-65c59865dd-w8plw" deleted :; linkerd diagnostics proxy-metrics -n feh deploy/feh-client-10000b10x| grep "feh-server-10000b10x-65c59865dd-w8plw" |wc...

@ra-grover Thank you for the additional details. I am also traveling at the moment, so my responses may be delayed.

Ah! This is exactly what the 'ingress' mode is for :) In the default mode, Linkerd uses the destination IP as configured by traefik. That is, traefik does its own...

Are you able to provide the output of `linkerd diagnostics proxy-metrics` and `kubectl logs` against a client in this state? This should help shine light on the nature of the...

@1ovsss It sounds to me that Linkerd is not doing load balancing... this would explain why the cilium cluster has no metrics associated with the Service -- the proxy is...

``` 2024-06-06T14:03:06.497366974Z [ 61799.863570s] INFO ThreadId(01) inbound:server{port=80}:rescue{client.addr=10.109.192.187:37222}: linkerd_app_core::errors::respond: HTTP/1.1 request failed error=error trying to connect: Connection refused (os error 111) error.sources=[Connection refused (os error 111)] ``` This indicates that the...

Linkerd doesn't have any default request timeouts. The log messages you include do not indicate any sort of timeout behavior. ``` [ 1081.455684s] TRACE ThreadId(01) outbound:accept{client.addr=10.3.109.171:33118 server.addr=172.20.117.227:7233}:proxy{addr=172.20.117.227:7233}:http:service{ns= name=service port=0}:endpoint{addr=10.3.140.90:7233}: linkerd_stack_tracing:...

I'm unable to reproduce this with an example application that sleeps for 60s. Clients successfully observe responses. And in the screenshot, you include, we see other endpoints with response times...

@prajithp13 Unfortunately this doesn't really get us any closer to a repro, though it does indicate that your application is seeing an error that originates in linkerd: > ``` >...