Unicode-like encoding for filenames using Windows10-NFS-client and Linux-NFS-server

Question

Problem

The NFS-Client of Windows10 Pro mount an NFS-share with lang=ansi by default. This causing problems using filenames with umlauts or asian characters (e.g. Kanji, Hiragana, Katakana, etc).

Details about my setup

Server is Debian GNU/Linux 10.6 (buster). Support NFS-Version by that server from 3 to 4.2 (2 is disabled):

$ sudo cat /proc/fs/nfsd/versions -2 +3 +4 +4.1 +4.2 

The client is Windows10-Pro build 19041. It is unknown for me which NFS-protocol version is used by the client. This are the options for the mount point.

UID=-2, GID=-2 rsize=131072, wsize=131072 mount=soft, timeout=0.8 retry=1, locking=yes fileaccess=755, lang=ANSI casesensitive=no s=sys 

I know there are other options for lang=.

euc-jp: Japanese euc-tw: Chinese euc-kr: Korean shift-jis: Japanese Big5: Chinese Ksc5601: Korean Gb2312-80: Simplified Chinese Ansi: ANSI-encoded 

But none of them looks usefull for me.

Is there any possible solution?

Answer 1

AFAIK, windows supports NFSv3, which is uses XDR string as a type for the file names., which is limited to ASCII:

NFSv3 spec https://www.rfc-editor.org/rfc/rfc1813#section-2.5

typedef string filename3<>;

https://www.rfc-editor.org/rfc/rfc1014#section-3.10

The standard defines a string of n (numbered 0 through n-1) ASCII bytes to be the number n encoded as an unsigned integer (as described above), and followed by the n bytes of the string.

IOW, NFS v3 doesn't supports unicode

https://www.rfc-editor.org/rfc/rfc1813#section-3.2