Bug 248725 – EWS pulls the wrong revision when checking out the tree for testing

• Status: RESOLVED
• Resolution: FIXED
• Component: Tools / Tests

[Simon Fraser (smfr)]

We've had many occurrences of test failed in EWS that seem unexplained. After looking at some buildbot logs, I think I see why.

https://ews-build.webkit.org/#/builders/70/builds/22723 is an EWS run for this PR: https://github.com/WebKit/WebKit/pull/6939. It's testing against 257312@main.

The archive is built in https://ews-build.webkit.org/#/builders/43/builds/31687, and appears to behave correctly. I verified locally that the cherry-pick logic does the right thing. The result is uploaded as https://s3-us-west-2.amazonaws.com/ews-archives.webkit.org/mac-bigsur-x86_64-release/88c31a9a.zip and downloaded OK by the tester.

Things go wrong when the tester checks out (trying to check out to 257312@main):
https://ews-build.webkit.org/#/builders/70/builds/22723/steps/5/logs/stdio

This does:
git fetch -f -t https://github.com/WebKit/WebKit.git main --progress
From https://github.com/WebKit/WebKit
 * branch                      main       -> FETCH_HEAD
Auto packing the repository in background for optimum performance.
See "git help gc" for manual housekeeping.
warning: The last gc run reported the following. Please correct the root cause
and remove .git/gc.log.
Automatic cleanup will not be performed until the file is removed.
warning: There are too many unreachable loose objects; run 'git prune' to remove them.
git reset --hard FETCH_HEAD --
HEAD is now at fc72928b66d9 [Live Range Selection] editing/pasteboard/cleanup-on-move.html fails https://bugs.webkit.org/show_bug.cgi?id=248673
git checkout -B main
Reset branch 'main'
git rev-parse HEAD
fc72928b66d943febf0a19941ced1d2b2e8f3496

which is checking out the current TOT, not 257312@main. Maybe FETCH_HEAD is unset?

[Alexey Proskuryakov]

I suspect that this may be an indirect regression from 257193@main.

There is CheckOutSpecificRevision class in steps.py, but it's entirely unused. We probably need to start using it (again?), as the checkout revision matters. Plus, ShowIdentifier is clearly printing out incorrect information.

[Radar WebKit Bug Importer]

<rdar://problem/102964437>

[Jonathan Bedard]

*** Bug 248760 has been marked as a duplicate of this bug. ***

[Jonathan Bedard]

Pull request: https://github.com/WebKit/WebKit/pull/7150

[EWS]

Committed 257373@main (171728f99da7): <https://commits.webkit.org/257373@main>

Reviewed commits have been landed. Closing PR #7150 and removing active labels.