test: Increase service/DNS timeout from 30 to 240 seconds #16820
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jrajahalme
added
area/CI-improvement
tklauser
approved these changes
Jul 8, 2021
aditighag
reviewed
Jul 8, 2021
There was a problem hiding this comment.
I don't quite follow why absence of service backends would fail DNS resolution.
In the above case the service did not have any backends yet, even when
the backend pod was up and running. Give k8s apiserver a bit more time
to converge.
$ k get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
my-service ClusterIP 10.39.240.33 <none> 80/TCP 6s
$ k get ep
NAME ENDPOINTS AGE
my-service <none> 15s
$ k exec -it dnsutils nslookup my-service
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
Server: 10.39.240.10
Address: 10.39.240.10#53
Name: my-service.default.svc.cluster.local
Address: 10.39.240.33
This was a hasty assumption of mine. Not sure but maybe DNS resolution could fail for a headless service if there are no backends, but that should not be the case here. Now I don't know why the DNS resolution for the service failed, but a longer timeout might still avoid the test flake. |
|
test-me-please |
Increase the timeout for service top acquire it's backends in hopes
the 30 seconds was simply too optimistic in some scenarios. This could
avoid test flakes like:
Failed to resolve kafka-service DNS entry in pod empire-outpost-8888-544cdcd9b8-72hgl
Expected
<*errors.errorString | 0xc001e05fd0>: {
s: "unable to resolve DNS for service kafka-service in pod empire-outpost-8888-544cdcd9b8-72hgl: 30s timeout expired",
}
to be nil
/home/jenkins/workspace/Cilium-PR-K8s-1.21-kernel-4.9/src/github.com/cilium/cilium/test/k8sT/KafkaPolicies.go:157
In the above case the service did not have any backends yet, even when
the backend pod was up and running. Give k8s apiserver a bit more time
to converge.
"status": {
"loadBalancer": {}
}
Signed-off-by: Jarno Rajahalme <jarno@isovalent.com>
jrajahalme
force-pushed
the
pr/jrajahalme/test-kafka-service-timeout
branch
from
b8b824e
to
1da0082
Compare
|
rebased due to completely broken CI |
|
test-me-please |
|
test-1.21-4.9 |
|
test-1.21-4.9 |
jrajahalme
added
the
ready-to-merge
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Increase the timeout for service to acquire it's backends in hopes
the 30 seconds was simply too optimistic in some scenarios. This could
avoid test flakes like:
In the above case the service did not have any backends yet, even when
the backend pod was up and running. Give k8s apiserver a bit more time
to converge.
(from
svc.jsonin https://jenkins.cilium.io/job/Cilium-PR-K8s-1.21-kernel-4.9/939/testReport/junit/Suite-k8s-1/21/K8sKafkaPolicyTest_Kafka_Policy_Tests_KafkaPolicies/)Signed-off-by: Jarno Rajahalme jarno@isovalent.com