Comment 3 for bug 148607

Thank you for taking the time to report this issue and help to improve Ubuntu.

Breaking down your suggested changes to smb.conf:

 unix charset = UTF8
 display charset = UTF8

These should be equivalent to the defaults, there is no need to set them.

 character set = UTF8
 client code page = 866

These are not valid options with current samba.

 default case = upper
 preserve case = yes
 short preserve case = yes

These don't seem to have anything to do with filename encoding, why do you need to set them? (The last two are also the same as the built-in defaults.)

 dos charset = cp866

This is the last remaining change, and the one that directly affects the character set used for displaying filenames to the client.

It's true that the default 'dos charset' value is only correct for Western European languages. On the other hand, the dos charset setting is only used when communicating with non-Unicode clients; clients from NT4 and above (which includes Windows 2000, Windows XP, and Windows Vista) are handled natively using Unicode. So I don't think the impact of this issue is very large; on the other hand, mapping all of our available installation locales to the (unrelated) set of DOS codepages would be a lot of work. While I'm not marking this as "wontfix" now, I think that's probably what will happen with it.