Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 75.43%. Comparing base (96f98af) to head (0777a8c).
⚠️ Report is 43 commits behind head on main.

@@            Coverage Diff             @@
##             main    #1993      +/-   ##
==========================================
- Coverage   75.58%   75.43%   -0.15%     
==========================================
  Files         451      454       +3     
  Lines       56369    56348      -21     
  Branches     9299     9299              
==========================================
- Hits        42607    42508      -99     
- Misses      10619    10698      +79     
+ Partials     3143     3142       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

This could also be the right opportunity to drop the deprecated macos-13 runner that you've noted in #1992.

Done!

Now we have an instance of this failure on the "macos-15-intel with Python 3.10" configuration, so it's not just related to Python 3.13:

[gw2] darwin -- Python 3.10.18 /Users/runner/hostedtoolcache/Python/3.10.18/x64/bin/python3
worker 'gw2' crashed while running 'test_reaction.py::TestBlowersMaselRate::test_derivative_ddT'

Now we have an instance of this failure on the "macos-15-intel with Python 3.10" configuration, so it's not just related to Python 3.13:

[gw2] darwin -- Python 3.10.18 /Users/runner/hostedtoolcache/Python/3.10.18/x64/bin/python3
worker 'gw2' crashed while running 'test_reaction.py::TestBlowersMaselRate::test_derivative_ddT'

On the ‘upside’, the post-merge test that was supposed to fail is now running through! Looks like intel macs are now sporadically failing?

@speth ... thus far, we have exactly one of the jobs reliably fail, either 3.10 or 3.13; curiously, never both. Need to take this back ... after writing this, both succeeded, and a different job failed.

This failure of Run Python 3.10 examples on ubuntu-24.04, NumPy ==1.21.6 is different still, and not one I recall seeing before:

samples/python/onedim/catalytic_combustion.py
Traceback (most recent call last):
  File "/home/runner/work/cantera/cantera/samples/python/onedim/catalytic_combustion.py", line 68, in <module>
    surf_phase.advance_coverages(1.0)
  File "build/python/cantera/kinetics.pyx", line 943, in cantera.kinetics.InterfaceKinetics.advance_coverages
cantera._utils.CanteraError: 
*******************************************************************************
CanteraError thrown by CVodesIntegrator::integrate:
CVodes error encountered. Error code: -3
At t = 0.0583727 and h = 4.24999e-08, the error test failed repeatedly or with |h| = hmin.

Components with largest weighted error estimates:
4: -392.219827383851
9: 20.872449014364545
5: -0.0024577841850999837
1: -0.0002620613619660072
0: -0.00011714129027371423
2: -0.00011122842455574104
10: 9.99021834700569e-05
3: -5.6773903789372044e-05
7: -3.898277339646901e-157
8: -9.688141290158232e-158
*******************************************************************************

@speth ... correct (sorry but in all previous instances, the failures on this PR were from the post-merge tests). I had seen this one before, but curiously failing for the ion_flame example(!). For the last CI that was triggered, both previously failing macOS runners succeeded and this one failed. This PR's scope is to test macos-15-intel and address the removal of the macos-13 runners.

@speth ... correct (sorry but in all previous instances, the failures on this PR were from the post-merge tests). I had seen this one before, but curiously failing for the ion_flame example(!). For the last CI that was triggered, both previously failing macOS runners succeeded and this one failed. This PR's scope is to test macos-15-intel and address the removal of the macos-13 runners.

Yes, of course. I commented just to make sure we don't lose information about this new failure mode. I didn't intend to suggest that this CI failure was a blocker for merging this PR.