Anomaly forcing run adds extra streams at runtime, causing an error
- Thread starter James King
- Start date
-
- Tags
- clm5 data models namelist
ncdump -k ~/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP5-8.5/af.allvars.CESM.SSP5-8.5.2015-2100_c20220628.nc
ncdump: /data4/home/dpeng/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP5-8.5/af.allvars.CESM.SSP5-8.5.2015-2100_c20220628.nc: NetCDF: Unknown file format
I tried to download the file with FileZilla ftp, but the problem is still.
I tried to use Matlab to open this file but failed.
Matlab log
>> ncdisp('C:\Users\Pe-Do\af.allvars.CESM.SSP5-8.5.2015-2100_c20220628.nc')
Error using matlab.internal.imagesci.netcdflib
'5' is not a recognized netcdf format.
Error in netcdf.inqFormat (line 26)
fmt = matlab.internal.imagesci.netcdflib('inqFormat',ncid);
Error in internal.matlab.imagesci.nc/openToRead (line 1293)
this.Format = lower(netcdf.inqFormat(this.ncRootid));
Error in internal.matlab.imagesci.nc (line 124)
this.openToRead();
Error in ncdisp (line 69)
ncObj = internal.matlab.imagesci.nc(ncFile);
>> h5read('C:\Users\Pe-Do\af.allvars.CESM.SSP5-8.5.2015-2100_c20220628.nc', '/time')
Error using h5readc
Unable to open the file because of HDF5 Library error. Reason:Unknown
Error in h5read (line 93)
[data,var_class] = h5readc(Filename,Dataset,start,count,stride);
ncdump cannot recognize which format of this file. I tried browser download, wget, FTP(fileZilla). Tried to use different software to open this file, like ncview, panoply(NASA), Matlab, and ncdump. No solution yet.
I do see now that there is a strange problem with that file. An ncdump -k works fine on the file in its original location. However, if I scp it to our ftp site and do an ncdump -k on it I get a Unknown file format error.
So I made a new file that is "Classic" file format and put it on our ftp site:
ftp://ftp.cgd.ucar.edu/pub/oleson/af.allvars.CESM.SSP5-8.5.2015-2100_c20220628_classic.nc
So, download that and see if you get a valid file format when you do ncdump -k.
So I made a new file that is "Classic" file format and put it on our ftp site:
ftp://ftp.cgd.ucar.edu/pub/oleson/af.allvars.CESM.SSP5-8.5.2015-2100_c20220628_classic.nc
So, download that and see if you get a valid file format when you do ncdump -k.
Hi, Oleson. I don't have access to get into this path. I cannot see any folders within ftp://ftp.cgd.ucar.edu/pub/.
Can you try accessing the ftp site using anonymous login, e.g.,
username: anonymous
password: <your-email-here@email.com>
username: anonymous
password: <your-email-here@email.com>
If you're still unable to connect, we can likely confirm any firewall issues you might be having too. However, it may require us to know the IP address that you're reaching out from (or at least the somewhat precise time you tried to connect to our system from).
Now, I can access the path with the password <your-email-here@email.com>. Before using this password, I used "user@example.edu" as the password, it can't let me get into the pub/oleson. Thanks for the help, I will try the nc file soon.
Oleson, it works. I now can use ncview and matlab to open this nc file. Some of the files in the anomaly path have similar issues, could you please help replace them with new files?
By the way, I want to know where I can find the citations about the anomaly forcing data. So then I can share the information with readers and reviewers, and acknowledge the helps when I try to publish my findings. Appreciate the help from you and the community.
This was related to a failing mount on the system. The /pub directory should be remounted and available once again, so feel free to use your own email address to connect to the ftp system (although you can continue using either option; using your own email address will allow us to better troubleshoot any issues you might run into)
Can you be specific about what files have issues?
I don't know of any formal citations. Perhaps you could cite the User's Guide (1.5.12. Running with anomaly forcing — ctsm CTSM master documentation). There is some additional information in the global attributes of each file.
I don't know of any formal citations. Perhaps you could cite the User's Guide (1.5.12. Running with anomaly forcing — ctsm CTSM master documentation). There is some additional information in the global attributes of each file.
Oleson, both of the files in the folders have the same issue. Can you share the _classic files in the pub/oleson?
Thanks, Oleson. All of those files work well.