Checkout of um at a particular revision

Hi CMS team,

I am debugging an input file that is failing to open. I have done a fcm co of relevant um code for local testing, but I would like a branch on my development on trac, as I think this will be helpful in the future (right now I just have a local copy of code I can’t check back onto trac).

I could like to have a branch of the cmip6 branch here

I am copying the branch using:

fcm branch-create --bob -k no_ticket_needed pacemaker_runs_code_bob fcm:um.x-br/pkg/Share/vn10.7_CMIP6_production_mods@43043

however, when I do a diff of the code branch I just created and the original branch I tried to copy, I can see they are different.

Any idea why the branch I created is not a copy of the URL I gave it?

Thank you in advance,

Penny

Penny

Input files should not fail to open – what is the offending suite/file

Grenville

I was able to identify an input filename had the extension ‘.n’ instead of ‘.nc’. So it was just a typo. I was able to fix that aspect.

Hi Penny,

I just tried this out and the branch I created is the same as the original branch. I ran this:

ros@puma2$ fcm branch-create -t test --bob ros_test fcm:um.x-br/pkg/Share/vn10.7_CMIP6_production_mods@43043
ros@puma2$ fcm co fcm:um.x-br/test/rosalynhatcher/r43043_ros_test
ros@puma2$ fcm co fcm:um.x-br/pkg/Share/vn10.7_CMIP6_production_mods@43043
ros@puma2$ diff -r vn10.7_CMIP6_production_mods r43043_ros_test
Binary files vn10.7_CMIP6_production_mods/.svn/wc.db and r43043_ros_test/.svn/wc.db differ

Regards,
Ros.

Hi Penny,

I’ve just checked out your branch and a diff shows it’s the same as the code your branched from…

ros@puma2$ diff -r r43043_pacemaker_runs_code_bob vn10.7_CMIP6_production_mods
Binary files r43043_pacemaker_runs_code_bob/.svn/wc.db and vn10.7_CMIP6_production_mods/.svn/wc.db differ

Cheers,
Ros.

Thanks Ros.

I just took a new copy of vn10.7_CMIP6_production_mods. So the diff you did was against the files I just downloaded.

I was suspicious that my folder vn10.7_CMIP6_production_mods_local_copy was the problem. Which it was in the end. On reflection I must not have had the code at the revision I expected and was actually at the HEAD. Thanks for helping me debug it.

Hi Penny,

Glad you resolved it. You may already be aware of it but in case not, fcm info is a useful command that will tell you what revision of the code you are looking at in a working copy.

Regards,
Ros.