vllm - 💡(How to fix) Fix [CI Failure]: mi355_2: Distributed Tests (2 GPUs)(H100-MI355) [3 comments, 2 participants]

TL;DR

Investigate the mi355_2: Distributed Tests (2 GPUs)(H100-MI355) test group failure on the AMD nightly build.

Guidance

• Review the test logs and hardware status for mi355_2 to identify the cause of the failure.
• Check the sync_ready_tickets.py script for any issues or errors that may be contributing to the failure.
• Verify that the test environment and hardware configuration are correct and consistent with previous successful runs.
• Investigate the build history to see if any recent changes may have introduced the failure.

Notes

The issue lacks detailed information about the failure, such as error messages or stack traces, which would be necessary to provide a more specific fix or workaround.

Recommendation

Apply workaround: Investigate and resolve the issue with the mi355_2 test group to prevent further failures, as there is no clear indication of a fixed version available.