Closed bartlettroscoe closed 6 years ago
CC: @trilinos/stokhos
Since this is a library build failure, this just be backed out immediately. This will take down practically everyone's build (including all of the auto PR builds) that builds Stokhos or any packages that depend on Stokhos upstream or downstream.
I will back out this commit immediately and push with the checkin-test-sems.sh script.
I just pushed the commit 5ee27ba3b0e6e043d4979157f80fb08d877cc325:
commit 5ee27ba3b0e6e043d4979157f80fb08d877cc325
Author: Roscoe A. Bartlett <rabartl@sandia.gov>
Date: Thu Mar 1 14:33:23 2018 -0700
Revert "inject null space now." (#2315)
This reverts commit 5585e8cdbc10ee28cae85c240138fabf27289778.
This commit breaks the Stokhos library build.
Build/Test Cases Summary
Enabled Packages: MueLu
Disabled Packages: PyTrilinos,Claps,TriKota
0) MPI_RELEASE_DEBUG_SHARED_PT => passed: passed=80,notpassed=0 (13.32 min)
M packages/muelu/src/Transfers/SemiCoarsen/MueLu_SemiCoarsenPFactory_decl.hpp
M packages/muelu/src/Transfers/SemiCoarsen/MueLu_SemiCoarsenPFactory_def.hpp
Therefore, the next CI iteration should not show this failure.
Can someone on the @trilinos/muelu team work with @rstumin to help set up a topic branch and create a PR for this commit? That will test it and then it can be safely merged to 'develop' once the automated builds and tests pass?
@bartlettroscoe thanks for posting the issue and reverting that commit :D
Sadly, it looks like that commit caused my PR #2313 to fail the auto-tester though but I can trigger a re-test I think ;)
Sadly, it looks like that commit caused my PR #2313 to fail the auto-tester though but I can trigger a re-test I think ;)
@william76, just rebase your branch on top of develop
and force push. That will trigger another round of testing.
The current CI build shows clean. See:
See the -3
by the 0
in the "Build | Error" column.
Now this can be fixed in a calm and relaxed way offline.
@bartlettroscoe Not sure if Ray gets these notifications; he might need a separate e-mail or phone call.
@mhoemmen @bartlettroscoe plus Ray is going to parallel processing so he probably won't be working on this for some time.
Ray is going to parallel processing so he probably won't be working on this for some time.
No hurry. The commit has been reverted so fixing it can be done anytime someone gets around to it and test and push it.
I sent Ray an e-mail just in case. Thanks Ross!
?Yup. I saw that I screwed up. Sorry about that. I'm getting a little slow witted in my waning years.
From: Mark Hoemmen notifications@github.com Sent: Friday, March 2, 2018 1:04 PM To: trilinos/Trilinos Cc: Tuminaro, Raymond S; Mention Subject: [EXTERNAL] Re: [trilinos/Trilinos] Broken stokhos library build with untested push to Muelu on 3/1/2018 (#2315)
I sent Ray an e-mail just in case. Thanks Ross!
- You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/trilinos/Trilinos/issues/2315#issuecomment-370052024, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ALR0uY6ZFEktaGl49d0KcgOZhEIRs_p9ks5tabP5gaJpZM4SZBeZ.
@rstumin Awww Ray, I think what happened is that Stokhos wasn't getting enabled for you. It's an honest mistake. Trilinos doesn't exactly make things easy sometimes ;-) .
@rstumin PR #3422 has been merged, let us wait until tomorrow's nightly tests are run and if it all looks good we should be able to close this issue!
So I did not notice any particular breakage on the dashboard related to PR #3422 so I am closing this.
CC: @trilinos/framework, @trilinos/stokhos, @trilinos/muelu, @rstumin
Next Action Status:
Commit 5585e8c was reverted in pushed commit 5ee27ba3b. Next: Someone help @rstumin to put commit 5585e8c on a rebased topic branch and create PR once errors are resolved?
Description
The current CI iteration running right now shown at:
is showing build failures for the Stokhos libraires (not tests or examples).
The only changes pulled in this CI iteration are shown at:
and are: