Custom Query (3366 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (4 - 6 of 3366)

1 2 3 4 5 6 7 8 9 10 11 12
Ticket Resolution Summary Owner Reporter
#2451 fixed adding new diagnostics in STASHmaster haochi
Description

Hi

I’ve changed my STASHmaster_A file to output some variables(CCN and kappa). The model finished the run with success, but I didn’t see any expected output.

Should the STASHmaster_A file in the cylc-run directory be same with that in my branch? I checked that file in cylc-run directory, and it doesn’t include any changes I have made.

The file is in here on Monsoon /home/d02/haohi/cylc-run/u-ax382/work/20150901T0000Z/atmos_main/STASHmaster

However, I have already committed my branch, and added this two lines in my rose-suite.conf

[file:app/um/file/STASHmaster] source=fcm:um.xm_br/dev/haochiche/vn10.6.1_ukca10.6.1/rose-meta/um-atmos/HEAD/etc/stash/STASHmaster@HEAD

And also added this line inside ENV of rose-app.conf.

STASHMSTR=STASHmaster

My suite id is u-ax382

Many thanks

Haochi

#3087 answered Advice on using mule-cumf on MONSOON for summary comparison between dumps gmann
Description

Dear NCAS-CMS Helpdesk,

Wuhu Feng and I have been discussing the output files from mule-cumf on MONSOON to compare two dump files from two different UKESM runs.

There seems to be a problem because the cumf utility doesn't seem to be able to recognise/compare the STASH fields in two different dumps files from the reference UKESM PI-control (aw310) and Wuhu's copy of the release job bc694 (bp061) — it's a direct copy of the release job bc694, except for start date and start dump files.

Basically, mule-cumf doesn't seem to be able to recognise that the same fields are contained in both dump files.

In contrast, when we compare dump files from the same job it does recognise that there are differences between the fields and then lists the differences accordingly.

Please can you advise about this problem.

It looks like cumf is not able to compare fields from the v10.9 job and the v11.2 job.

Or we need to add a flag to the command we're running mule-cumf on xcs via for example:

/projects/um1/bin/mule-cumf aw310a.da30090101_00 bp061a.da20001001_00

The two .txt files are off Wuhu's homepages.see for reference.

http://homepages.see.leeds.ac.uk/~earfw/UM/d.txt

and

http://homepages.see.leeds.ac.uk/~earfw/UM/diff_jan.txt

See the job where we compare between the aw310 and bp061 runs the cumf is saying there are no differences within the "additional parameters", see it says

0 differences in additional_parameters (with 0 ignored indices)

Whereas when we compare dumps from the same aw310 job it does recognise the fields simply based on their order I guess.

Are we missing a flag or so to pick up the STASH items to do the comparison properly?

See below for more info.

Thanks for your help with this.

Cheers Graham

Below is the cumf output from comparing the aw310 and bp061 dump files.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

  • (CUMF-II) Module Information *

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% mule : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/mule/init.py (version 2019.01.1) um_utils : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/um_utils/init.py (version 2019.01.1) um_packing : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/um_packing/init.py (version 2019.01.1) (packing lib from SHUMlib: 2018061)

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

  • CUMF-II Comparison Report *

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

File 1: aw310a.da30090101_00 File 2: bp061a.da20000101_00 Files DO NOT compare

  • 8 differences in fixed_length_header (with 7 ignored indices)
  • 2 differences in integer_constants (with 0 ignored indices)
  • 3 differences in real_constants (with 0 ignored indices)
  • 0 differences in additional_parameters (with 0 ignored indices)

Compared 0/111019 fields, with 0 matches 55470 fields found in file 1 were not in file 2 55549 fields found in file 2 were not in file 1

And for the run from the same aw310 v10.9 job but different years we get 7142 differences:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

  • (CUMF-II) Module Information *

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% mule : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/mule/init.py (version 2019.01.1) um_utils : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/um_utils/init.py (version 2019.01.1) um_packing : /projects/um1/mule/mule-2019.01.1/system/python2.7/lib/um_packing/init.py (version 2019.01.1) (packing lib from SHUMlib: 2018061)

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

  • CUMF-II Comparison Report *

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

File 1: bp061a.da20000101_00 File 2: bp061a.da20100101_00 Files DO NOT compare

  • 1 differences in fixed_length_header (with 7 ignored indices)
  • 1 differences in integer_constants (with 0 ignored indices)
  • 3 differences in real_constants (with 0 ignored indices)
  • 7142 differences in additional_parameters (with 0 ignored indices)

Compared 0/111098 fields, with 0 matches 55549 fields found in file 1 were not in file 2 55549 fields found in file 2 were not in file 1

1) aw310 — the main UKESM pre-industrial control

and a copy of the release job bc694) and we have

http://homepages.see.leeds.ac.uk/~earfw/UM/

#3112 answered Increase storage luciad
Description

Hello,

I would like to increase the storage for my account on 'n02-NEL013746'. I need to test a suite for one month in a Gregorian day calendar, that includes many 3D 3hourly diagnostics for the A-CURE PPE project (collaboration U. Oxford - U. Leeds). For the moment I have 250 GB, of which only 120 GB are available for testing, but this storage is enough only for a 10 day cycle.

If this is not possible, then could you advice what to do?

Thank you

Lucia

1 2 3 4 5 6 7 8 9 10 11 12
Note: See TracQuery for help on using queries.