Update_buffer_pos .NE. unpack_buffer_pos

ANCOMS Regional MOM6
1

Hi Charles

I think it was my fault for using some data the wrong resolution!

But I am now getting an error once I actually try and run the model with initial conditions and boundary forcing generated using regional mom6.

FATAL from PE 9: set_group_update: update_buffer_pos .NE. unpack_buffer_pos

I don’t know if it is related to the warning message

WARNING from PE 9: categorize_axes: Failed to identify x- and y- axes in the axis list (nx_segment_003, ny_segment_003, time) of a variable being read from INPUT/forcing_obc_segment_003.nc

I used using regional-mom6 version 1.0.1 from xp65 analysis3-25.05
And pulled the main MOM code from dev-MC_100km_jra_iaf GitHub - ACCESS-NRI/access-om3-configs: ACCESS-OM3 MOM6-CICE6 configurations with optional WW3 and Wombat. All the configurations use the Payu and pre-built executables available on NCI.

2

Hi Lizzie,

This doesn’t look like it’s due to the xp65 environment to my eyes: this

looks like it originates here. I’ll see if I can find someone who has a better understanding of this & if so move this off to a separate thread.

3

Hi Lizzie,

@helen has just let me know that she thinks this is probably her domain so she should be able to help you. Unfortunately she’s ill today, so it won’t be until next week.

In the meantime, would you be able to upload your config files to github? It should help Helen pinpoint the problem.

Cheers!

1 Like
4

Ok thank you
My files are uploaded here

6

Hi @Lizzie – apologies for the delay and thanks for uploading your files! There have been a few moving parts in terms of updates in the analysis 3 environment and the regional mom6 environments which have created some issues. I think we are past these now but the first thing we need to do is to make sure we are all on the same page with what version we are using. Can I please get you to recreate the input files using analysis3-25.05 and the notebook created by @mmr0 here (also, see Madi’s post here). These are relatively up-to date and what I plan to use. No guarantees this fixes the issue – but I think it is a good place to start - please report back if you get the same or a different bug.

7

This is a normal warning that occurs all the time so I think is unrelated to the current issue.

8

Yes, using 25.05 and the notebook Madi uploaded has worked for me again this week :slight_smile:

1 Like
9

Not sure if this issue was solved, but when we get it in the CESM, it’s because we have too many MPI tasks (i.e. requested too many resources for a small domain, like 10 points over 10 nodes). Reducing them gets rid of the issue.

2 Likes
10

Thanks @manishvenu! We are stating to think about the number of tasks requested as we optimise some of the domains - so it is good to know that this is the error message when we go overboard!

1 Like