SSEC forums

Hello,

I am using geocat software to to convert himawari9 data to netcdf before doing a WRFDA run.
I am getting the himawari 9 data fom NCI (https://opus.nci.org.au/spaces/NDP/page ... DS+Upgrade). Recently they upgrade their services. Since then, I have WRFDA runs ending with segmentation fault at the moment of reading the radiance data. I am wondering if , maybe, geocat software would not be compatible with these new data release? On their website there is this obscure (to me) comment:
"Unidata has changed the THREDDS NetCDF Subset Service (NCSS) output in the new version. This can result in differences in the NCSS dimension ordering. For example, the lat/long values can display in a different order in the output. The issue has been reported to Unidata and remains under development."
Could that be a problem?
I should add that when running geocat, I don't have an error values and the netcdf files are generated apparently without problem. But I was wondering if it could be somehow, corrupted?

Best

Thanks for your question about the NCI Himawari 9 data and Geocat. I just want to confirm, the files you are obtaining from NCI are of the form "HS_H09_20170308_0230_B15_FLDK_R20_S0310.DAT" etc..., and you are having trouble running these files from the new server through geocat, as opposed to from the now-deprecated server?

I will try to obtain some data from NCI and replicate your error, and may have followup questions.

Thanks,
Geoff

Thank you for your answer.
Yes the files are of this type exactly.
And prior to the change of server by NCI (my previous runs are about 1 year ago), I was never having issues.
After running geocat I don't see any problem but strangely WRFDA gives a segmentation fault. Since nothing has changed except the migration/update of nci...

Do you have access to samples of the H9 data before and after the update? If yes, could you try and upload a tarball of the data to ftp.ssec.wisc.edu:/pub/incoming ?

Thanks,
Geoff

Yes I have old data.
I created a tarball but failed to push it:
curl -T RADIANCE_DATA_20240428_0600.tar.bz2 ftp://ftp.ssec.wisc.edu:/pub/incoming
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- 0:00:03 --:--:-- 0
curl: (25) Failed FTP upload: 550

I upload a tarball of old files to the ftp server.
The name of the tarball is: RADIANCE_DATA_20240428_0600.tar.bz2

Thanks for your help.

Best

Can you check the file `DATA_20240428_0600/HS_H09_20240428_0600_B09_FLDK_R20_S0410.DAT` does it have the same size as the others? It may have been corrupted.

Hi,
Maybe a problem with the tarball because it is ok . I pushed it alone on the server.

When I try to unpack the tarball I get an error about the tarball being truncated. I ran to check the integrity of the tarball and I got
Can you run the bunzip2 command I gave above on your local copy of the tarball and see what it says. Also, what is the size of the tarball on your machine?

After having a better look, I am missing the files HS_H09_20240428_0600_B08_FLDK_R20_S1010.DAT and HS_H09_20240428_0600_B09_FLDK_R20_S0810.DAT, could you post those files individually?