Fix branch runs with run_acme #1594
Conversation
There was a problem hiding this comment.
This looks fine as far as I can tell. The real proof would be to actually use the code to perform a branch run - which I don't have time to do right now because I'm on my way to a conference. Have you done this Michael? I'd say once you've confirmed you can branch, this PR is ready for master. Thanks for working on this.
Another comment is that the machine Hopper doesn't exist any more, so we should delete all mention of it.
|
I believe so; I've set the run type to branch to verified that it's able to run without crashing and with the correct slurm options on Edison. |
Update version number Remove hopper from consideration when updating the case.run file
|
Were branch runs broken in general or just inside the run_acme script? |
|
AFAIK, just inside the run_acme script - some filenames changed |
|
@golaz or @cameronsmith1 - would you mind trying a branch run using the run_acme script? I don't think I can manage it from the CESM meeting... |
|
Note: I have a PR to fix a different problem in run_acme (run_acme stops when running a compset with a data atmosphere): PR #1586. The only conflict I can see is that both PRs are designated v3.0.9. |
|
Whenever I have to do a branch run, I usually end up needing to edit the branch setup section manually to achieve what I need. However, it is useful to have a simple case in that section that serves as a template (so I support this update). Hence, I am happy to let this go forward without additional testing. |
There was an independent PR (#1594) which became version 3.0.9, so changing the version for this PR to 3.0.10.
Add some maps to config_grids.xml Added the f19_g17 maps, as wall as the ww3 <-> tx1v1 maps; all files are in inputdata as well. Test suite: none, just some stand-alone checks Test baseline: N/A Test namelist changes: N/A Test status: bit for bit, plus resolutions that didn't run previously now run Fixes #1539 User interface changes?: N/A Code review: @dabail10 can you please look at this? (And maybe see about getting added to ESMCI so I can list you as an official reviewer?)
This PR enables branch runs with the run_acme script
It also removes some sed commands
BFB