I have a bash script that has been running fine for months that scans a bunch of files and gives me a log file output, it has suddenly started putting 1.5M of a repeating sequence of ^@^@^@^@^@^@^@^@^@^@ on the first line of the logfile, is this a unicode problem I should be setting something in my bash script? I already had PATH issues after awhile of running (that I solved with your help, thanks!). Any suggestions?

um, thanks?

Where do you see those characters? When you look at the file using vi/vim? If so, '^@' (one character, not two) is the representation of the NULL-byte. Example of this happening is when you truncate a file while there's still an open file-handle.

It's an extremely long sequence of those 2 characters with no spaces between the 2 character pattern, only on the first line of the file when viewing with either vi or cat, it's almost like I'm not cleaning up from the last time I ran the script or something? I can't understand why it ran once a day for months and just now started doing this. The script takes >20 hours to run once.

This is usually the case in VI when the line is "too long", or? (seen it on AIX, Solaris - but tried it on Linux, and that doesn't happen somehow!)