Payu archive directory

Hi! I am using the vk83 payu version for the first time, and it is creating this nestled directories in my archive for some unclear reason. I have some restart files I want to start my ACCESS-OM2 run from in /scratch/x77/jn8053/access-om2/archive/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring/, which is being correctly identified when running payu setup.

However, when I start my run, it somehow starts saving output in /scratch/x77/jn8053/access-om2/archive/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring/jn8053/access-om2/archive/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring-93edf978. This is somewhat incovenient.

Hi Julia - can you share your config.yaml ?

I am guessing you have shortpath set. You only need to set this if you want to use a /scratch/ folder in a different project to your default. Shortpath should typically just be /scratch/PROJECT_CODE - payu will append USER/LABORATORY/archive/experiment name automatically

Thanks Anton! I am actually not setting shortpath. This is my config file:

# PBS configuration
# If submitting to a different project to your default uncomment line below 
# and change project code as appropriate
project: x77
queue: normal
walltime: 5:00:00
jobname: 01deg_jra55_ryf_zmin21m
ncpus: 5180
storage:
        gdata:
                - e14
                - ik11
                - x77
                - gv90
                - ua8
                - g40

# Force payu to always find, and save, files in this short (or scratch) project directory
# shortpath: /scratch/x77

# Model configuration
name: common
model: access-om2
input: /g/data/ik11/inputs/access-om2/input_08022019/common_01deg_jra55
submodels:
    - name: atmosphere
      model: yatm
      exe: /g/data/ik11/inputs/access-om2/bin/yatm_1bb8904.exe
      input:
          - /scratch/x77/jn8053/access-om2/input/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring
          - /g/data/ik11/inputs/access-om2/input_08022019/yatm_01deg
          - /g/data/ua8/JRA55-do/RYF/v1-3/
      ncpus: 1

    - name: ocean
      model: mom
      exe: /g/data/ik11/inputs/access-om2/bin/fms_ACCESS-OM_97e3429_libaccessom2_1bb8904.x
      input:
          - /scratch/x77/jn8053/access-om2/input/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring
          - /g/data/ik11/inputs/access-om2/input_08022019/mom_01deg
      ncpus: 4358

    - name: ice
      model: cice5
      exe: /g/data/ik11/inputs/access-om2/bin/cice_auscom_3600x2700_722p_d3e8bdf_libaccessom2_1bb8904.exe
      input: /g/data/ik11/inputs/access-om2/input_08022019/cice_01deg
      ncpus: 799
           
# Collation
collate:
  restart:  true
  mpi:  true
  walltime: 1:00:00
  mem: 64GB
  ncpus: 4
  queue: normal
  exe: /g/data/ik11/inputs/access-om2/bin/mppnccombine-fast

# Misc
runlog: true
stacksize: unlimited
restart_freq: 1  # use tidy_restarts.py instead
###mpirun: -mca coll_hcoll_enable 1  ##--mca pml yalla -x MXM_LOG_FILE=$PBS_JOBFS/mxm.log
qsub_flags: -W umask=027 
#storage:
#      gdata:
#          - ua8
# set number of cores per node (28 for normalbw, 48 for normal on gadi)
platform:
    nodesize: 48
mpi:
    module: openmpi/4.0.1


# DANGER! Do not uncomment this without checking the script is synching 
# to the correct location!
#postscript: sync_output_to_gdata.sh

Odd!

How were you setting up the old restart / warmstart ?

We suggest configuring it in the payu config.yaml file. e.g. add the line:

restart: /scratch/x77/jn8053/access-om2/archive/01deg_jra55_ryf9091_21mbath_sofia_ssp126_control_restoring/restart039/

to the config.yaml, instead of copying/symlinking etc

1 Like

Okk, right, I’ve started from scratch cloning the experiment again and now its good. Another case of me messing up! Thanks Anton.

1 Like

No worries :slight_smile: