Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.
And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.
Long prediction using KFServing API 504 Gateway Timeout
See original GitHub issue/kind bug
What steps did you take and what happened: [A clear and concise description of what the bug is.] Ran a custom container, using short/small data that will respond within 10-second is working fine.
However, we ran a custom container with a long input, that will take more than 30 seconds, the curl will get 504 Gateway Timeout. Queue-proxy has 127.0.0.1:8080 connection refused error.
What did you expect to happen: Should wait for a response that takes longer. Should be able to read probe.
Environment:
- Knative Version: 0.11.1
- KFServing Version: 0.2.2
- Kubeflow version: 1.0.0
- Microk8s: 1.18.3
- Kubernetes version: (use
kubectl version): 1.18.3 - OS (e.g. from
/etc/os-release): RHEL
Issue Analytics
- State:
- Created 3 years ago
- Comments:13 (5 by maintainers)
Top Results From Across the Web
Top Related Medium Post
Top Related StackOverflow Question
Troubleshoot Live Code
Top Related Reddit Thread
Top Related Hackernoon Post
Top Related Tweet
Top Related Dev.to Post
Top Related Hashnode Post
@yuzisun How can we change the default timeout value using V1aplha2 specs? I understand the “timeout” option present in the V1beta1PredictorSpec allows to change the default timeout value . We are currently using the KFServing 0.4.1 release and leveraging the kfserving pipeline component to deploy services.
Issue-Label Bot is automatically applying the labels:
Please mark this comment with 👍 or 👎 to give our bot feedback! Links: app homepage, dashboard and code for this bot.