Stop migration unless there are some metric of CI stability

[malfet]

Number of builds jobs failed today with various non-descriptive which in the past was exceeding rare for Linux builds.
Few of them are

• https://github.com/pytorch/pytorch/actions/runs/8739408133/job/23981738189
• https://github.com/pytorch/pytorch/actions/runs/8748004920/job/24007340735
• https://github.com/pytorch/pytorch/actions/runs/8748004920/job/24007341497
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976682807
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976683347
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976681715
• https://github.com/pytorch/pytorch/actions/runs/8738940413/job/23979092050
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994213718
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994215119
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994213896
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994214290
• https://github.com/pytorch/pytorch/actions/runs/8738177261/job/23991638097

Please do not migrate any new workflows until there is an explanation on what is causing those failures and how to address them

[zxiiro]

• https://github.com/pytorch/pytorch/actions/runs/8739408133/job/23981738189

This one both build and test do not appear to be running in ARC runners. So I think this is unrelated to ARC migration.

• https://github.com/pytorch/pytorch/actions/runs/8748004920/job/24007340735
• https://github.com/pytorch/pytorch/actions/runs/8748004920/job/24007341497
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976682807
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976683347
• https://github.com/pytorch/pytorch/actions/runs/8738176604/job/23976681715
• https://github.com/pytorch/pytorch/actions/runs/8738940413/job/23979092050

This one looks like #134 that @seemethere raised about the docker daemon issue.

• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994213718
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994215119
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994213896
• https://github.com/pytorch/pytorch/actions/runs/8743512800/job/23994214290
• https://github.com/pytorch/pytorch/actions/runs/8738177261/job/23991638097

This one looks like symptoms of #118 and #94 since the logs abruptly stop.

[jeanschmidt]

pytorch/pytorch/actions/runs/8743512800/job/23994213718
pytorch/pytorch/actions/runs/8743512800/job/23994215119
pytorch/pytorch/actions/runs/8743512800/job/23994213896
pytorch/pytorch/actions/runs/8743512800/job/23994214290
pytorch/pytorch/actions/runs/8738177261/job/23991638097
This one looks like symptoms of #118 and #94 since the logs abruptly stop.

For those failures, It looks to me that they are still victims of the issue that got fixed on pytorch/test-infra#5102 still I cannot guarantee that there might be other issues that can cause similar problems.

Thanks for looking into it and reporting problems you are seeing. But I believe we can benefit more by having a single open issue tracking this.

[jeanschmidt]

In agreement with the team, I'll close this issue now, and we will follow up in the issues there.

[malfet]

Well, it still acting up: https://github.com/pytorch/pytorch/actions/runs/8758173278/job/24038390608?pr=124511