wkorman@ do you still have the scripts/notes from this?

I don't have anything currently other than what was in the post-mortem doc. I had some hack config files that were the product of me reverse-engineering the environment variables and params passed to build/test commands after reading the build bot stdio/stderr logs but they were only on the couple of VMs I was using at the time, lost in the annals of prod electrons.

I could show someone what sample logs and issues I had to deal with in more detail if helpful but basically it boils down to, how to run the full cycle of git pull, gclient sync, and ninja with goma with the right configs, with a bot pulled out of the normal bot lifecycle. The bots use a local git cache and have other similar precondition setups which are foreign and unexpected to a typical Chromium/Blink eng and so if you're trying to debug a particular build/test situation on a bot it's hard to sync to ToT or a particular version and build and run, which is what this bug was about.

Here is the postmortem doc, Google internal for now.

https://docs.google.com/document/d/1s_RaR9ba2GWgLngJFuo32yRaZGBkJXBEg3xLDnUvoIA/edit

Actually the doc includes the VM host names I was using. Potentially we could log on there and I can find the scripts I was using if the VMs haven't been wiped. But they're like I said just reverse engineered simplified versions of what's in the bot logs, mixed with what an eng would normally run locally in attempt to build and test.

Production environment has surely changed since my work ~1 year ago. I am not actively working in this area. Unassigning self but happy to consult.

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot