1

I am trying to run the UniData admin tools "guide" and "fixfile" on a data file with with a corrupted record. However, both programs result in this error message "Locale C.UTF-8 does not belong to current language group 255/192/129". I spent a little bit of time reading through the internationalization manual for UniData and researching localization for Linux. I ran "locale -a" and I noticed that I have a en_US.utf8 locale, but no C.UTF-8. I found a post online where some people were talking about adding C.UTF-8 support to centos, but in the end decided not to because concerns over the disk space and a few other things. They proposed that anyone that needs support should just handle it in "the usual way".

I'm not really sure what that means though. At this point, I'd settle for anything that works even if it is just a temporary workaround.

UniData is version 7.3 and centos is version 6.

MrGoodfix
  • 317
  • 2
  • 14

2 Answers2

1

Sorry for the inconvenience everyone. I'm working from a Windows 10 workstation and I'm using Ubuntu on the Windows Subsystem for Linux (WSL). I've been working on the server over ssh. The default configuration for ssh on WSL Ubuntu is to forward the locale settings. I fixed the issue by editing the ssh config on my Ubuntu.

I edited the config with ...

sudo vi /etc/ssh/ssh_config

I commented out the line with ...

SendEnv LANG LC_*

so it turned into ...

# SendEnv LANG LC_*

Then I ssh'ed back into the server and my programs worked fine. Or maybe that remains to be seen, but at least they didn't give the language group error.

MrGoodfix
  • 317
  • 2
  • 14
1

Try setting LANG=C before running the tool.

On Red Hat Enterprise Linux Server 7.9, we resolved the Locale C.UTF-8 does not belong to current language group 255/192/129 error when running udt by adding export LANG=C to the appropriate shell initialization file, like .bashrc.

Luke Fritz
  • 108
  • 1
  • 7