a quick workaround for increasing the mpi task numbers on orion for ctest :: rrfs_3denvar_rdasens #788
Conversation
…d the program hanging as reported in Issue 766
|
@TingLei-daprediction , please reach out and find two peer reviewers for this PR. |
There was a problem hiding this comment.
The ctest rrfs_3denvar_rdasens completed successfully for me using this PR. It is still slower than Hercules but at least the test is working.
(base) orion-login-3[130] sdegelia$ ctest -R rrfs_3denvar_rdasens
Test project /work/noaa/wrfruc/sdegelia/GSI/build
Start 3: rrfs_3denvar_rdasens
1/1 Test #3: rrfs_3denvar_rdasens ............. Passed 1146.50 sec
100% tests passed, 0 tests failed out of 1
Total Test time (real) = 1146.51 sec
|
Thank you @SamuelDegelia-NOAA for running the As the developer of this change and the author of this PR, @TingLei-daprediction should run and document ctest results. @TingLei-daprediction , please review GSI: How to Make Changes |
|
@SamuelDegelia-NOAA Thanks for your reviewing and ctests on your side. That will always be helpful for ctests by a "third-party". |
|
@TingLei-daprediction , you are right. As this PR only alters one test on one machine, running |
|
@RussTreadon-NOAA Thanks. I had updated the description for this PR. Please let me know if more details are needed. |
|
@TingLei-daprediction , what you added is sufficient for this PR. In the future please follow @SamuelDegelia-NOAA 's example and include the ctest output.
|
|
@RussTreadon-NOAA Thanks! |
JingCheng-NOAA
left a comment
JingCheng-NOAA
left a comment
There was a problem hiding this comment.
Since there are already two successful ctests (from Ting and Samuel) for this change on Orion, I have no issue to approve it.
* origin/develop: Move to contrib spack-stack on Jet (#787) a quick workaround for increasing the mpi task numbers on orion for ctest :: rrfs_3denvar_rdasens (#788) Recover the capability of handling model fields from operation gfs.v16.3 (#785) fix a bug in deter_sfc_gmi (#781) add safeguard to thompson_reff (#779) Fix incorrect usage of real(i_kind) in mg_input.f90 (#760) Transition to Thompson Microphysics for Microwave All-sky Assimilation (#743) Format changes for EUMETSAT metop-sg and CADS debug fix (#773) Update global_4denvar and global_enkf ctests to reflect GFS v17 (#774) fix for cris-fsr memory corruption (#767) Gnssrwnd1.0 (#747)
…test :: rrfs_3denvar_rdasens (NOAA-EMC#788) This PR would provide a quick fix/workaround for rrfs_3denvar_rdasens running on orion by increasing the mpi task numbers running the test.
6 participants
It has been under investigation why the GSI would become idle in rrfs_3denvar_rdasens issue 766. Following discussions with @RussTreadon-NOAA discussion , this PR would provide a quick fix/workaround for rrfs_3denvar_rdasens running on orion by increasing the mpi task numbers running the test.
Update: it should be noted. the corresponding local branch had passed regression tests on orion before this PR was submitted.