CoreTrouble
A transferred file is unreadable
Text that looks fine on one platform is gibberish on the other.
Almost always character encoding
Traditional z/OS data is EBCDIC; files from other platforms are usually ASCII or UTF-8. Whether a tool converts depends on the file's tag and the transfer mode used.
Diagnose
- ls -T on the Unix file — is it tagged, and as what?
- Was the transfer done in text or binary mode? Text converts; binary does not.
- View the file with HEX ON. EBCDIC spaces are x'40'; ASCII spaces are x'20'. Digits are F0–F9 in EBCDIC, 30–39 in ASCII.
- Check for stray carriage returns (x'0D') from a Windows source.
Fix
- Tag the file correctly with chtag so tools handle it consistently.
- Convert deliberately with iconv rather than relying on a transfer mode.
- Decide once, per interface, whether files arrive as ASCII or EBCDIC, and document it.
Stuck on something else?
Ask the community or search the full course.