This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Youcheng, > I still get "Segmentation fault" when try "make all". > Attached below are related files. It turns out we need a copy of the output from running the configure script, even if it appears there are no errors. So please remove the config.cache file, if there is one, and run configure with exactly the same arguments you used before, and please send us the invocation line, the CUSTOMIZE file (if you have changed it), and the output from running the configure script. Also, instead of just running "make all", try running "make -n all" and see if there is any output before you get a sementation fault. This should tell what program is being invoked that causes the segmentation fault (unless it is "make"). Finally, see if you can change your configuration so that a core file is produced for errors like this. Often there is some user-defined configuration parameter for the shell that determines whether programs can produce core files. A core file would help in diagnosing this problem. Other Linux users have reported success in building netcdf 2.4 and 2.4.1 on Linux systems, so we know it can be done. --Russ