Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Migrate from envinf1 to frontend1 #2440

Merged
merged 16 commits into from May 17, 2019

Conversation

@bilke
Copy link
Member

commented Mar 25, 2019

Use frontend1 instead of envinf1.

  1. Feature description was added to the changelog
  2. Any new feature or behavior change was documented?

@bilke bilke added the WIP 🏗 label Mar 25, 2019

@bilke

This comment has been minimized.

Copy link
Member Author

commented Mar 25, 2019

@endJunction

This comment has been minimized.

Copy link
Member

commented Mar 25, 2019

Note to myself:
These needs a small tolerance upgrade only:
ogs-PhaseField_2D_StaticHydraulicFracture-time-vtkdiff
ogs-HydroMechanics_HML_square_1e2_unconfined_compression_early-time-vtkdiff
ogs-GroundWaterFlowProcess_Neumann_nonuniform_cosY-vtkdiff
ogs-GroundWaterFlowProcess_wedge_1x1x1_1e3_prism_surfaceflux-vtkdiff
ogs-Mechanics/EhlersDamageNonLocal/bar/p3/bar

The failing ExtractSurface benchmarks need an upgrade in vtkdiff, because they are compared with diff only, but the coordinate values need to be compared. Same for the Bode test. The newer vtk version seems to have better conversion of doubles to ascii.

@bilke

This comment has been minimized.

Copy link
Member Author

commented Apr 8, 2019

@TomFischer RasterToMeshTest.convertRasterToTriMeshElevation now segfaults. The other convertRasterToXXX-tests successfully run. Any idea?

@endJunction endJunction force-pushed the bilke:migrate-frontend1 branch from 3465d7e to dda6be5 May 8, 2019

@bilke bilke force-pushed the bilke:migrate-frontend1 branch 3 times, most recently from 27a8d4f to bff70fb May 9, 2019

@bilke

This comment has been minimized.

Copy link
Member Author

commented May 10, 2019

@endJunction New progess:

frontend-serial:

Fails once in 3 test runs:

unknown file: Failure
C++ exception with description "std::bad_alloc" thrown in the test body
[  FAILED  ] NumLibLocalToGlobalIndexMapTest.MultipleVariablesMultipleComponentsHeterogeneousElements

frontend-parallel:

66 - ogs-Mechanics/Linear/disc_with_hole (Failed)

Reason:

Calling /gpfs1/data/ogs/jenkins-frontend1/workspace/ufz_ogs_PR-2440@2/build/bin/vtkdiff from the VTKDIFF_EXE environment variable didn't work as expected. Return value was 32512.
...
vtkdiff not found. at TestDefinition.cpp, line 132

vtkdiff is there and works (as long as the env script loading the modules is sourced) but somehow its test call (vtkdiff --version) returns 127. When I comment out this check (line 108 in TestDefinition.cpp) it works.

When running vtkdiff without loading the module files it also return 127 because it cannot load the VTK shared libraries. Could it be that std::system does not pass the environment? But still I do not know why it works when commenting out the mentioned check.

@bilke bilke force-pushed the bilke:migrate-frontend1 branch from bff70fb to f6efdd6 May 13, 2019

@bilke

This comment has been minimized.

Copy link
Member Author

commented May 14, 2019

@endJunction Running on frontend2 reduces failing ctests from 44 to 14.

@endJunction endJunction force-pushed the bilke:migrate-frontend1 branch from 5cb5de8 to da007a5 May 16, 2019

@bilke bilke force-pushed the bilke:migrate-frontend1 branch 2 times, most recently from 415a6e2 to f967993 May 17, 2019

@bilke bilke force-pushed the bilke:migrate-frontend1 branch from f967993 to d45ad6b May 17, 2019

@bilke bilke force-pushed the bilke:migrate-frontend1 branch from d45ad6b to ea9addf May 17, 2019

@bilke bilke removed the WIP 🏗 label May 17, 2019

@bilke

This comment has been minimized.

Copy link
Member Author

commented May 17, 2019

@endJunction I think I found the reason for the internal compiler errors: They appeared when two jobs ran singularity1 at the same time. I limited singularity1 to just 1 job executor slot to hopefully prevent that.

@codecov

This comment has been minimized.

Copy link

commented May 17, 2019

Codecov Report

Merging #2440 into master will decrease coverage by 0.14%.
The diff coverage is n/a.

Impacted file tree graph

@@            Coverage Diff             @@
##           master    #2440      +/-   ##
==========================================
- Coverage   32.95%   32.81%   -0.15%     
==========================================
  Files         554      554              
  Lines       20699    20696       -3     
  Branches     9749     9749              
==========================================
- Hits         6822     6792      -30     
- Misses      10407    10411       +4     
- Partials     3470     3493      +23
Impacted Files Coverage Δ
MathLib/Vector3.cpp 33.33% <0%> (-33.34%) ⬇️
GeoLib/SimplePolygonTree.cpp 61.9% <0%> (-4.77%) ⬇️
MeshLib/IO/VtkIO/VtuInterface-impl.h 13.63% <0%> (-4.55%) ⬇️
GeoLib/AnalyticalGeometry-impl.h 72.95% <0%> (-3.28%) ⬇️
MeshLib/VtkOGSEnum.cpp 3.75% <0%> (-2.5%) ⬇️
...ng/Algorithms/IterationNumberBasedTimeStepping.cpp 78.57% <0%> (-2.39%) ⬇️
GeoLib/MinimalBoundingSphere.cpp 43.92% <0%> (-1.87%) ⬇️
NumLib/DOF/LocalToGlobalIndexMap.cpp 48.4% <0%> (-1.6%) ⬇️
...eStepping/Algorithms/EvolutionaryPIDcontroller.cpp 68.18% <0%> (-1.52%) ⬇️
MeshLib/MeshSurfaceExtraction.cpp 29.48% <0%> (-1.29%) ⬇️
... and 5 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6e8d9fd...ea9addf. Read the comment docs.

@bilke bilke merged commit 1e5df79 into ufz:master May 17, 2019

4 of 6 checks passed

codecov/project 32.81% (-0.15%) compared to 6e8d9fd
Details
ufz.ogs in progress
Details
Codacy/PR Quality Review Up to standards. A positive pull request.
Details
codecov/patch Coverage not affected when comparing 6e8d9fd...ea9addf
Details
continuous-integration/jenkins/pr-merge This commit looks good
Details
deploy/netlify Deploy preview ready!
Details

@bilke bilke deleted the bilke:migrate-frontend1 branch May 17, 2019

@bilke

This comment has been minimized.

Copy link
Member Author

commented May 17, 2019

@jbathmann The migration has been done. Please check the updated docs on usage of the eve frontends for development.

@jbathmann

This comment has been minimized.

Copy link
Contributor

commented May 17, 2019

@bilke Thanks for the information.
Running ogs on highmem and idiv nodes works for me now. I build ogs on frontend2 using the cli.sh script as described in the documentation (no conan).
However I still have trouble running ogs on the eve nodes (no response and immediate job progression after executing ogs).

Am I still doing things wrong?

@endJunction

This comment has been minimized.

Copy link
Member

commented May 17, 2019

@jbathmann Let's have a look together on Monday. Maybe Tom can be of some help.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
3 participants
You can’t perform that action at this time.