#general
- b
bored-art-40741
08/24/2015, 5:47 PMI'll check - c
cuddly-architect-27013
08/24/2015, 5:47 PMok, iirc i saw some differences between what ./rbt diff was giving me and what git diff master was showing...not sure what I screwed up here, but I'll keep at it till the diff looks right - c
cuddly-architect-27013
08/24/2015, 5:47 PMthx! - w
witty-crayon-22786
08/24/2015, 5:47 PMmight have accidentally committed to master - c
cuddly-architect-27013
08/24/2015, 5:48 PMok, i'll check the log - w
witty-crayon-22786
08/24/2015, 5:48 PMmake sure
worksgit pull --ff-only origin/master - c
cuddly-architect-27013
08/24/2015, 5:48 PMok - c
cuddly-architect-27013
08/24/2015, 5:49 PM[tw-mbp-jtrobec pants-build (jtrobec/coverage/fix-cobertura-open)]$ git pull --ff-only origin/master fatal: 'origin/master' does not appear to be a git repository fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists. - c
cuddly-architect-27013
08/24/2015, 5:49 PMyeah, i've broken something - b
bored-art-40741
08/24/2015, 5:50 PMgit remote -v - w
witty-crayon-22786
08/24/2015, 5:50 PMthe contributor guide might have suggested naming it "upstream" or something... but github.com/pantsbuild/pants is ground truth - c
cuddly-architect-27013
08/24/2015, 5:50 PM[tw-mbp-jtrobec pants-build (jtrobec/coverage/fix-cobertura-open)]$ git remote -v origin https://github.com/jtrobec/pants (fetch) origin https://github.com/jtrobec/pants (push) pantsbuild https://github.com/pantsbuild/pants.git (fetch) pantsbuild https://github.com/pantsbuild/pants.git (push) upstream https://github.com/pantsbuild/pants (fetch) upstream https://github.com/pantsbuild/pants (push) - c
cuddly-architect-27013
08/24/2015, 5:50 PMyeah, ok - w
witty-crayon-22786
08/24/2015, 5:50 PMhaving
as your own master is hecka confusing, heh.origin - w
witty-crayon-22786
08/24/2015, 5:51 PMhave been meaning to edit that - c
cuddly-architect-27013
08/24/2015, 5:51 PMyeah, i think i did things out of order or something, didn't fork till after i started the change - a
ancient-stone-50795
08/24/2015, 5:55 PM@witty-crayon-22786: I think your proposed scheme would work for production pants, but I’m pretty sure that in tests
isn’t always relative topants_workdir
You may find some ‘gotchas’ there.buildroot - w
witty-crayon-22786
08/24/2015, 5:56 PM@ancient-stone-50795: but it should be, in theory. - w
witty-crayon-22786
08/24/2015, 5:56 PMor, it need not not be =P - a
ancient-stone-50795
08/24/2015, 5:56 PMI guess it can always be expressed as a relative path, but it might not always be ${BUILDROOT}/.pants.d - w
witty-crayon-22786
08/24/2015, 5:56 PMthat's fine. - a
ancient-stone-50795
08/24/2015, 6:02 PM@witty-crayon-22786: Thanks for your feedback on https://rbcommons.com/s/twitter/r/2688/. Do you care to take another look before I push it? We are planning to another internal release today. - w
witty-crayon-22786
08/24/2015, 6:03 PMlooking - w
witty-crayon-22786
08/24/2015, 6:07 PM@ancient-stone-50795: #shipit - w
witty-crayon-22786
08/24/2015, 6:07 PM(with comment) - c
cuddly-architect-27013
08/24/2015, 6:13 PMok, diff is fixed: https://rbcommons.com/s/twitter/r/2670/ thanks for the help! - b
bored-art-40741
08/24/2015, 6:15 PM@rough-minister-58256: I'm having trouble getting another repro, this might be getting into heisenbug territory - b
bored-art-40741
08/24/2015, 6:15 PMCan you enumerate the major changes you made to nailgun/process management? - b
bored-art-40741
08/24/2015, 6:15 PMDid anywhere you have a tight loop trying to read from a socket followed by acquiring a lock? - b
bored-art-40741
08/24/2015, 6:16 PMThough it does a lot more socket reading than lock acquiring