I've been scratching my head on this one for a few days now. I'm getting spurious file writing errors (as in, randomly it FAILS to write the output file) with a routine sandwiched in a much larger piece of code, that resembles:
open INFILE, "$inputfn" || die "Error reading input file: $!\n";
open OUTFILE, ">$outputfn" || die "Error opening output file: $!\n";
while (<INFILE>) {
##...a bunch of stuff munged with regexes omitted...
print OUTFILE "$_\n";
}
close OUTFILE;
close INFILE;
This has only been happening on Linux, but on machines running two different distros, two different (albeit both 2.2.x) kernels, and both on nfs-mounted filesystems and not. It's as if the open call is dying (ie. there isn't even a zero-length file lying around) without the die() being called.
Anyone have any thoughts? Known idiosyncracies that are similar? etc.?
Spud Zeppelin * spud@spudzeppelin.com