You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for this. The statement is run only when the container is about to be killed. There seems to be a regression that took place here when curl was accidentially removed.
During deployment of test models, Kubernetes dashboard produces an error:
Liveness probe failed: Get http://172.17.0.24:8082/ready: dial tcp 172.17.0.24:8082: getsockopt: connection refused
Exec lifecycle hook ([/bin/sh -c curl 127.0.0.1:8000/pause && /bin/sleep 5]) for Container "seldon-container-engine" in Pod "deep-mnist-single-model-svc-orch-6b5c55bb97-7bjng_seldon(a3704d8b-e0f3-11e8-8dac-00155d09a30d)" failed - error: command '/bin/sh -c curl 127.0.0.1:8000/pause && /bin/sleep 5' exited with 127: /bin/sh: curl: not found , message: "/bin/sh: curl: not found\n"
Does it mean that curl is not installed in created pod? I see numerous errors like this, while looking at the pods list of the model deployment
The text was updated successfully, but these errors were encountered: