-
Sub-task
-
Resolution: Done
-
Normal
-
None
-
None
-
None
-
None
-
False
-
-
False
What happened?
Improve the error handler when the cluster was gone.
The CLI is reporting errors when communicating with kube-api correctly, and raising the information to end user[1], but at the end there's a wrong message telling the "pods are ready"[2].
[1] Error on CLI when the cluster was gone
Tue, 14 Jun 2022 20:13:29 -03> Global Status: running JOB_NAME | STATUS | RESULTS | PROGRESS | MESSAGE openshift-conformance-validated | running | | 0/3250 (0 failures) | status=waiting-for=openshift-kube-conformance=(0/-1/0)=[24/100] openshift-kube-conformance | running | | 344/345 (0 failures) | status=running ERRO[2022-06-14T20:14:14-03:00] failed to get namespace openshift-provider-certification: Get "https://api.mrb14.devcluster.openshift.com:6443/api/v1/namespaces/openshift-provider-certification": http2: client connection lost ERRO[2022-06-14T20:14:19-03:00] failed to get namespace openshift-provider-certification: Get "https://api.mrb14.devcluster.openshift.com:6443/api/v1/namespaces/openshift-provider-certification": dial tcp: lookup api.mrb14.devcluster.openshift.com on 10.11.5.19:53: no such host (... x8) ERRO[2022-06-14T20:15:29-03:00] failed to get namespace openshift-provider-certification: Get "https://api.mrb14.devcluster.openshift.com:6443/api/v1/namespaces/openshift-provider-certification": dial tcp: lookup api.mrb14.devcluster.openshift.com on 10.11.5.19:53: no such host
[2] Final message (unexpected) as the cluster was gone
INFO[2022-06-14T20:15:38-03:00] Sonobuoy pods are ready!
Note: the logs above was collected from dev environment which the cluster has a TTL of 10h, the cluster was deleted by pruner while the certification tool was running.
What did you expect to happen?
1) We need to improve the last message when the communication was permanently lost from server. The current message is not correct.
2) We could print an final message after that connection times out (retries exhausted)