Master build acts from master

This MR temporarily builds ACTS from their master branch, rather than from a fixed release. We think this will be more practical whilst we iterate on the ACTS design, but should be reversed once we're in a more stable situation.

I've also made a couple of minor documentation tweaks to the markdown.

I've compiled locally and verified that ACTS libraries are seen in x86*/lib64.

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #466]

Okay I'll fix those @akraszna. Can you see why the CI failed? I can't access the Jenkins log remotely - I think it's just too big and I'm timing out.

added 1 commit

• 4115e848 - Addressing Attila\'s comments on the MR: explicitly state branch and full repo…

Compare with previous version

resolved all discussions

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #470]

Okay - I see the issue: "Build timed out (after 80 minutes). Marking the build as failed."

I'm going to set it to merge if the CI succeeds. I'd really like to have this in before the ACTS meeting tomorrow!

Edit: the gods of CI are not on my side tonight. :-( It failed immediately.

Jenkins please retry a build

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #471]

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #472]

Hi Ed,

It seems that LCG_93a is acting up once again on CVMFS. Until it settles down, all of the CI jobs in the master branch of this repository will keep having issues like:

-- Found 41 package(s)
-- Considering package 1 / 41
-- Package filtering rules read:
--   + External/BAT
--   + External/Davix
--   + External/Eigen
--   + External/FastJet
--   + External/FastJetContrib
--   + External/GPerfTools
--   + External/GoogleTest
--   + External/HDF5
--   + External/KLFitter
--   + External/Lhapdf
--   + External/lwtnn
--   + External/CheckerGccPlugins
--   + External/CLHEP
--   + External/XRootD
--   - .*
-- Configuring the build of package: BAT
CMake Error at /cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase/x86_64/Cmake/3.11.0/Linux-x86_64/share/cmake-3.11/Modules/FindPackageHandleStandardArgs.cmake:137 (message):
  Could NOT find ROOT (missing: ROOT_INCLUDE_DIR ROOT_LIBRARIES
  ROOT_BINARY_PATH ROOT_PYTHON_PATH)
Call Stack (most recent call first):
  /cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase/x86_64/Cmake/3.11.0/Linux-x86_64/share/cmake-3.11/Modules/FindPackageHandleStandardArgs.cmake:378 (_FPHSA_FAILURE_MESSAGE)
  /var/lib/jenkins/workspace/AE-MERGE-REQUEST/Build/AtlasLCG/modules/FindROOT.cmake:66 (find_package_handle_standard_args)
  /var/lib/jenkins/workspace/AE-MERGE-REQUEST/External/BAT/CMakeLists.txt:18 (find_package)


-- Configuring incomplete, errors occurred!
See also "/build2/atlasexternals/master/AthAnalysisExternals/CMakeFiles/CMakeOutput.log".
See also "/build2/atlasexternals/master/AthAnalysisExternals/CMakeFiles/CMakeError.log".

The only way I see out of this is if you also downgrade all projects back to LCG_93 with this MR...

Cheers, Attila

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #473]

added 1 commit

• c1f55f86 - Revert to LCG93

Compare with previous version

Okay - reverting the LCG version does indeed seem to help. sigh BTW I noticed that e.g. AnalysisBaseExternals and TestExternals hadn't been incremented to LCG93a ... was this on purpose.

Jenkins Build FAILURE

Results available at: Jenkins [AE-MERGE-REQUEST #474]

noticed strange failures in the last build. Cleaning aibuild092 and will restart the MR build.

Jenkins Build SUCCESS

Results available at: Jenkins [AE-MERGE-REQUEST #475]

So I can conclude that it was incremental build issue with switching between LCG_93/LCG_93a

Okay, thanks very much Emil!