[CMIP6 CMOR-ization & ESGF-publication] NorESM2-LM - esm-hist about noresm2cmor HOT 9 CLOSED

OK, I can ask Sigma2 to see if they will have some clue. Otherwise, I will dig further into it.

from noresm2cmor.

There are still some unclear error for some of these realisations, to cite this one as an example:

For this one, 1850-1869 data seems OK, but then there are some errors:

cmorized number of files

variant: r6i1p1f1
Ofx, fx, etc	10
yyyy1	yyyy2	nf
1850	1859	486
1860	1869	486
1870	1879	46
1880	1889	46
1890	1899	46
1900	1909	46
...
2010	2014	46
Total:		1672
Total r6i1p1f1:	1672

So from 1870, only 46 files are able to be produced.

During Cmorization, there is error like:

--------------------------------------------------------------------------
Open MPI failed an OFI Libfabric library call (fi_endpoint).  This is highly
unusual; your job may behave unpredictably (and/or abort) after this.

  Local host: ipcc
  Location: mtl_ofi_component.c:513
  Error: Invalid argument (22)
--------------------------------------------------------------------------
ipcc:rank0: PSM3 can't open nic unit: 0 (err=23)
ipcc:rank0: PSM3 can't open nic unit: 0 (err=23)
ipcc:rank0: PSM3 can't open nic unit: 0 (err=23)
…
ipcc:rank7: PSM3 can't open nic unit: 0 (err=23)
ipcc:rank6: PSM3 can't open nic unit: 0 (err=23)
[ipcc:2062563] 7 more processes have sent help message help-mtl-ofi.txt / OFI call fail
[ipcc:2062563] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages

The cmorization tool is identical and goes generally well with some other realisations. I suspect there are some errors in the file / file system.

Do you have any ideas? I would suspect this is due to some file/file system error?

I can ask Sigma2 if they can quickly spot some error and get some clue?

from noresm2cmor.

I'll have a look and see if I find something

from noresm2cmor.

Hm, I can't find any problems with the files. I can open all, display them and I'm not sure what else to check... maybe you can ask sigma2?

I'm not sure if this is relevant, but a similar problem was discussed here: ofiwg/libfabric#6710
sorry if it is something completely different...

from noresm2cmor.

@YanchunHe Good if you can prioritize the SSP simulations. Are those also problematic?

from noresm2cmor.

Yes, OK. most of the esm-hist simulations are done, then I will soon go with the ssp simulations.

from noresm2cmor.

Supert. Thanks!

from noresm2cmor.

CMORized, and ready to be published to ESGF.

486 datasets for every 10 years.

data path

• /projects/NS9034K/CMIP6/.cmorout/NorESM2-LM/esm-hist
• /projects/NS9034K/CMIP6/CMIP/NCC/NorESM2-LM/esm-hist

version

• v20230616

sha256sum
/projects/NS9034K/CMIP6/CMIP/NCC/NorESM2-LM/esm-hist

• .r6i1p1f1.sha256sum_v20230616

from noresm2cmor.

@YanchunHe published

from noresm2cmor.