Repository navigation
DATM CPLHIST test needs to be fixed and then restored #408
Description
I prematurely merged #405 without realizing that the DATM CPLHIST test (SMS_Ld5.f10_f10_mt232.1850_DATM%CPLHIST_SLND_SICE_SOCN_SROF_SGLC_SWAV_SESP.derecho_intel.datm-cplhist) is currently failing.
The problem is that a dtlimit is exceeded due to too big of a discrepancy between dtmin and dtmax:
(shr_strdata_readstrm) reading file ub: /glade/campaign/cesm/cesmdata/inputdata/cplhist/cesm3_0/b.e30_alpha08o.B1850C_MTso.ne30_t232_wgx3.330/atm/b.e30_alpha08o.B1850C_MTso.ne30_t232_wgx3.330.cpl.hx.atm.1h.inst.0101-01-01-03600.nc
(shr_strdata_advance) Stream: 1 reading new data with LB ymd,tod = 1010101 2700 UB ymd,tod = 1010101 6300
(shr_strdata_advance) ERROR: for stream 1
(shr_strdata_advance) ERROR: calendar = NO_LEAP
(shr_strdata_advance) ERROR: dday = 0
(shr_strdata_advance) ERROR: dtime, dtmax, dtmin, dtlimit = 4.166667E-02 3.340417E+02 4.166667E-02 3.000000E+00
I'm going to temporarily remove this test from the test list until this problem can be resolved. @ekluzek whenever you have time to resolve it (no rush!!!), it can be reinstated.
Activity
billsacks commentedon May 21, 2026
This will mean backing out the changes in #409 .
Temporarily remove datm cplhist test
Merge pull request #409 from billsacks/remove_datm_cplhist_test
wwieder commentedon May 21, 2026
Does this mean we should make a cesm alpha09 tag that uses #409, or should we wait until this issue is resolved before making the alpha tag?
Second, Is this something that @slevis-lmwg could help with, or is it lower priority and can wait for @ekluzek to return?
billsacks commentedon May 21, 2026
I could be wrong, but my impression is that the data set pointed to here was just for software testing purposes, so this shouldn't hold anything up... but I haven't been in the loop with the plan for the land model spinup so I could be wrong on that.
slevis-lmwg commentedon May 21, 2026
@wwieder requested that investigate this:
@billsacks @ekluzek @wwieder I interpret this to mean that Erik's commit that looks for Cecile's cplhist data in /inputdata fails to find it there. I see two options, and will wait for feedback:
wwieder commentedon May 21, 2026
Thanks for looking into this, @slevis-lmwg.
It seems like a simple enough issue to address, but I'll let more SE-minded collaborators decide how to maintain this long-term.
Let's shelve this until @ekluzek and @billsacks are back from leave to address this.
slevis-lmwg commentedon May 21, 2026
Browsing the list of missing files suggests that the model does not find 12 months of data (present in Cecile's /scratch) and finds only 1 month in /inputdata. For the full list of missing files, see
/glade/derecho/scratch/slevis/SMS_Ld5.f10_f10_mt232.1850_DATM%CPLHIST_SLND_SICE_SOCN_SROF_SGLC_SWAV_SESP.derecho_intel.datm-cplhist.20260521_154000_oiao8p/TestStatus.logslevis-lmwg commentedon May 22, 2026
I will mark "done" in my Sprint accounting. Can always change later.
7 remaining items
Increase dtlimit for the CPLHIST options so that the CPLHIST test wil…