Wait for prediction endpoint hangs for ever in the case there is no external loadbalancer
Here: https://github.com/kubeflow/fairing/blob/cd3b0581c2a70c352ac2f66432bc2e023e3fbc7c/fairing/kubernetes/manager.py#L77
We wait for the loadbalancer to be ready; but there will never be a loadbalancer if we configured it to only use cluster ip
Issue-Label Bot is automatically applying the label kind/bug to this issue, with a confidence of 0.88. Please mark this comment with :thumbsup: or :thumbsdown: to give our bot feedback!
Links: app homepage, dashboard and code for this bot.
Seems we can waiting for same long seconds (such as 300s), and raise error message if the loadbalancer is not ready? comments? @karthikv2k
/area engprod