Hello,
I have the next problem :
...
Loading optical class data from /usr/local/seadas/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf
-E- /data5/swdev/seadas/build/src/l2gen/owt.c: Error reading SDS class_means from /usr/local/seadas/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf.
...
I don't know why the program has the problem read the file:
-rw-r--r-- 1 derpro pysta 4558 Mar 27 17:26 /usr/local/seadas/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf
The file is the same on other computer:
-rw-rw-r-- 1 pypro pypro 4558 Nov 21 15:06 /home/pypro/seadas-7.4/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf
Could someone give me suggestions?
Thanks,
Yuri
/l2gen/owt.c: Error reading SDS class_means from owt16_modis_stats_101111.hdf
/l2gen/owt.c: Error reading SDS class_means from owt16_modis_stats_101111.hdf
When reporting a problem, you should provide enough detail so someone else can easily reproduce the problem (see How to Report Bugs Effectively). Remember that NASA people are very busy working to give us SeaDAS 7.5, so as users we need to do everyting we can to let them focus on that task.
There have been permission problems for users when the OCSSW software was installed to a "system" directory such as
There is also a possibility that the "
There have been permission problems for users when the OCSSW software was installed to a "system" directory such as
/usr/local by running the installed as the "root" user (e.g., using the "sudo" command). Allowing random users to make changes to files in "system" directories that may be used by others is a security risk, so recent OS versions are using additional layers with names like "Security Enhanced Linux" to protect users from potential malicous changes to shared files. There is also a possibility that the "
owt16_modis_stats_101111.hdf" file is corrupt. Compare the output of "md5sum /usr/local/seadas/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf" with "md5sum /home/pypro/seadas-7.4/ocssw/run/data/modis/class/owt16_modis_stats_101111.hdf" on your other system. If they are the same, the file is unlikely to be corrupt. If they are different, they could still be OK, as some software opens hdf4 libraries with "write" enabled to update some sort of access log without changing the data. On the other (working) system, you have group write permission, but the failing system only allows the owner to write to the file.