Repoint branches

[mcoduoza]

No description provided.

[mcoduoza]

Hi @steveri, could you please look into this failure? https://buildkite.com/stanford-aha/aha-flow/builds/10889

This seems to have passed all the checks on garnet.
Thanks

[steveri]

?? You merged this? Did the problem get fixed? I think I know how to remove the error, even though it apparently was not complaining earlier...?

[steveri]

I am attempting to repeat the earlier failure here:
https://buildkite.com/stanford-aha/aha-flow/builds/10893
...this should fail, same as your build 10889

I started a new build with a potential hotfix, it should run here when/if it gets a free agent
https://buildkite.com/stanford-aha/aha-flow/builds/10894

[steveri]

Looks like a VCS versioning problem. Below, two buildkite runs, each using the same garnet repo version, one fails and one succeeds:

https://buildkite.com/stanford-aha/aha-flow/builds/10889
FAILS using aha df2e802, garnet 4479910f, vcs T-2022.06
"Error-[XMREMX] Cross-module reference resolution error"
Chronologic VCS (TM)
Version T-2022.06-SP2_Full64 -- Thu Dec 5 14:00:19 2024

https://buildkite.com/stanford-aha/garnet/builds/5369
SUCCEEDS using garnet 4479910f, vcs Q-2020.033
Chronologic VCS (TM)
Version Q-2020.03-SP2_Full64 -- Wed Dec 4 16:12:38 2024

[mcoduoza]

I merged it then reverted when I realized it failed. Hmm, so the garnet CI is using a different VCS version than aha?

[steveri]

I merged it then reverted when I realized it failed. Hmm, so the garnet CI is using a different VCS version than aha?

Looks like that may be the case. I am continuing my investigation, I should be able to come up with code that works with both versions of vcs I think.

[steveri]

I have code that works with

• vcs version Q-2020 running on either ubuntu or centos
• vcs version T-2022 running on centos but BREAKS if run on ubuntu

I am hoping to get this resolved by tomorrow sometime... :(

[mcoduoza]

oh wow :( ok thanks for your help!