Aviv did NOT land his changes around this, so he's not to blame.

An example failure (I hope this isn't normal):

https://uberchromegw.corp.google.com/i/chromeos_release/builders/peach_pit-release%20release-R52-8350.B/builds/2/steps/SyncChrome/logs/stdio

[0:02:42] Fetching +refs/tags/*:refs/tags/*
[0:02:42] running "git fetch -v --progress origin +refs/tags/*:refs/tags/*" in "/b/git-cache/chromium.googlesource.com-external-github.com-catapult--project-catapult"
[0:02:42] Fetching +refs/branch-heads/*:refs/branch-heads/*
[0:02:42] running "git fetch -v --progress origin +refs/branch-heads/*:refs/branch-heads/*" in "/b/git-cache/chromium.googlesource.com-external-github.com-catapult--project-catapult"

________ running 'git -c core.deltaBaseCacheLimit=2g fetch origin --verbose' in '/b/cbuild/shared_internal/.cache/distfiles/target/chrome-src-internal/src/third_party/catapult'
[0:02:42] error: refs/remotes/origin/mac_symbols_fix does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mappersv22 does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mac_symbols_fix does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mappersv22 does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mac_symbols_fix does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mappersv22 does not point to a valid object!
[0:02:42] From /b/git-cache/chromium.googlesource.com-external-github.com-catapult--project-catapult
[0:02:42]  = [up to date]      infra/config -> origin/infra/config
[0:02:42]  = [up to date]      mappersv2  -> origin/mappersv2
[0:02:42]  = [up to date]      mappersv2_merge -> origin/mappersv2_merge
[0:02:42] error: refs/remotes/origin/mac_symbols_fix does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mappersv22 does not point to a valid object!
[0:02:42]    0728abf..d01ff3b  master     -> origin/master
[0:02:42]  = [up to date]      md-pages   -> origin/md-pages
[0:02:42]  = [up to date]      polymer10  -> origin/polymer10
[0:02:42]  = [up to date]      polymer10-dashboard -> origin/polymer10-dashboard
[0:02:42]    a0dc5f8..b948445  polymer10-migration -> origin/polymer10-migration
[0:02:42] error: refs/remotes/origin/mac_symbols_fix does not point to a valid object!
[0:02:42] error: refs/remotes/origin/mappersv22 does not point to a valid object!

Steven, do you have any feedback here?

None of that makes any sense to me.

We are investigating issue 591814, but have not landed anything yet.

We can probably fix this by clobbering the affected builders, but I don't want to do that until after we finish getting GCE builders to sync successfully.

I don't think this have to do with issue 591814.

Comment#2 - looks like some stuck remote branches, that didn't exit in release branches. Maybe it's ok.

This one is more scaring:

Error: Command 'git merge-base HEAD origin' returned non-zero exit status 1 in /b/cbuild/shared_internal/.cache/distfiles/target/chrome-src-internal/src/third_party/libjingle/source/talk

Looks like local git corruption -- per discussion with dgarrett@ can we just clobber the builder and proceed?

(If you're waiting on me, go for it).

Git corruption isn't unheard of during milestone transitions. The next "beta" inherits the previous "beta"'s checkouts, so not only do they jump pretty far ahead abruptly, but any change in how Chromite handles checkouts may hit a case suddenly dealing with a (from its POV) really old checkout from the previous milestone's Chromite.

Starting the clean up, semi-manually.

PS: Not every builder was affected, only most of them.

All of the problematic builders have been clobbers. 

There are some in-progress builds that should be killed, preferably after they finish the SyncChrome stages.

They should all have been clobbered, and re-sync'd.

Closing... please feel free to reopen if its not fixed.