[workflow] log is not pushed to driver if run multiple times
See original GitHub issueimport ray
from ray import workflow
ray.init(address='auto')
workflow.init()
@workflow.step
def s():
return 10
print(s.step().run())
Run the above script multiple times and only the first one will print the log correctly. Need to always push the log to the driver.
Issue Analytics
- State:
- Created 2 years ago
- Comments:9 (8 by maintainers)
Top Results From Across the Web
7 Github Actions Tricks I Wish I Knew Before I Started
Here are 7 tricks with github actions that changed my life (or at least my CI/CD pipeline). These tricks helped me create a...
Read more >Workflow Best Practices - Support and Troubleshooting
Workflow Best Practices | Tips and Gotchas Things to Avoid Here's list of things to avoid when building a workflow: NEVER delete a...
Read more >Monitor and manage real-time workflow processes
In this article. Monitoring real-time workflows and actions; Status of real-time workflow processes; Deleting process log rows; Next steps.
Read more >Troubleshoot Dataflow errors - Google Cloud
When running in batch mode, bundles including a failing item are retried four times. The pipeline fails completely when a single bundle fails...
Read more >Using workflow run logs - GitHub Docs
For more information about the Checks API, see "Checks." Note: Ensure that you only commit valid workflow files to your repository. If .github/workflows...
Read more >
Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free
Top Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
@lchu-ibm we don’t need to fix it in ray core layer since this is the behavior we’d like.
The thing here is that we need to redirect the log to the right driver by updating the logging library.
We don’t need to do this for most jobs. For workflow, we need to do this because it damages UX.
Even for non-detached actor, the same issue will be there. E.g. in the same example above, if the actor is not detached but script 1 is not exited, script 2 has the same issue of lost logs in the driver.