TestTraffic/sniffing/auto-tcp_a->headless_from_cluster-0 flakes frequently

[hzxuzhonghu]

https://prow.istio.io/view/gs/istio-prow/pr-logs/pull/istio_istio/26638/integ-pilot-k8s-tests_istio/18950

@howardjohn has some fix recently

[istio-policy-bot]

🤔 ❄️ Hey there's been no update for this test flakes for 3 days.

Courtesy of your friendly test flake nag.

[stevenctl]

I think this is fixed for single cluster, will wait a couple days now that it's re-enabled before closing this.

We still get some failures on http/grpc in multicluster mode. I think we're hitting limits on our test infra, can't replicate in GKE clusters.

=== RUN   TestTraffic/sniffing/auto-http_a->vm_from_cluster-1
2020-09-09T17:17:59.568511Z	info	tf	=== BEGIN: Test: 'pilot[TestTraffic/sniffing/auto-http_a->vm_from_cluster-1]' ===
    traffic.go:69: call for auto-http a->vm from cluster-1/0 failed, retrying: failed calling a->'http://vm.echo-1-70707.svc.cluster.local:81/': rpc error: code = Unknown desc = 20/20 requests had errors; first error: Get "http://vm.echo-1-70707.svc.cluster.local:81": read tcp 10.20.0.10:59864->10.255.20.200:81: read: connection reset by peer

[stevenctl]

We've disabled TCP sniffing tests so this passes but it isn't fixed yet. In the multicluster pilot job I see a lot of connection resets for http/grpc.

[stevenctl]

https://prow.k8s.io/view/gcs/istio-prow/pr-logs/directory/integ-pilot-multicluster-tests_istio/999

[howardjohn]

Not stale

[stevenctl]

Not stale

[istio-policy-bot]

🚧 This issue or pull request has been closed due to not having had activity from an Istio team member since 2021-06-18. If you feel this issue or pull request deserves attention, please reopen the issue. Please see this wiki page for more information. Thank you for your contributions.

Created by the issue and PR lifecycle manager.