ekarlsso@purdue_edu
New Member
I've been working on porting CLM5.0 to our cluster for some time now. I'm slowly making progress, but have a few questions. 1. How and where do I access the baseline data that are used in e.g. class O_TestTestScheduler? I currently don't have an entry for BASELINE_ROOT in my config_machines.xml so I'm guessing that 's the issue, but I'm not sure where to access the baslines and if I need to download and store those locally?2. My system does not require a PROJECT, so I have set PROJECT_REQUIRED to FALSE. The test still gives me en error "project not found". Is there a way to override this?3. I currently have access to one node on my system. After some digging I found that cime (in config_tests.xml) forces the addition of three spare nodes. Does the system require at least 4 nodes? Or will this not be an issue when running cases? 4. Lastly, in T_TestRunRestart I get this error: ---------------------------------------------------2019-03-22 15:56:38: ERROR: RUN FAIL: Command 'mpirun /scratch/brown/ekarlsso/scripts_regression_test.20190322_154929/NODEFAIL_P1.f09_g16.X.brown_intel.fake_testing_only_20190322_154929/bld/fake.sh >> cesm.log.$LID 2>&1 ' failedSee log file for details: /scratch/brown/ekarlsso/scripts_regression_test.20190322_154929/NODEFAIL_P1.f09_g16.X.brown_intel.fake_testing_only_20190322_154929/run/cesm.log.3786153.brown-adm.rcac.purdue.edu.190322-155636 The cesm.log file gives me the following error message, which I'm unable to decipher: JGF FAKE NODE FAIL
Any help on what is going on here is greatly appreciated.
Any help on what is going on here is greatly appreciated.