I'm trying to read large data files (> 4 GB) into fluent and get an error message.

These files are older and were created 4 yrs ago. The person who ran them had similar problems but does not remember the fix.

If I gunzip (using /bin/gunzip) on the command line and read that file in it works since Fluent is not doing the gunzip.

So it seems that Fluent is pointing at a different out of date gunzip.

How can I check? Can I change this setting?

Thanks

Scott

You are probably just running out of memory. Given the size of the data files, I assume the cas file is fairly large, and is already loaded into fluent, so your memory usage is already high. When you read a gzipped data file, fluent creates a new process via fork/exec to run gzip and read the output. If you don't have enough swap space (virtual memory), the fork will fail.