Monday, January 3, 2011

ASCII v UNICODE Windows 2008 Shares

Recently I was troubleshooting a problem with a Xerox printer/scanner. It has a feature that allows you to scan to a network share. This function stopped working with the file server in question was upgraded from Windows 2003 to Windows 2008. It took a network sniff to realize what was going on. The printer was presenting the share name to the server in ASCII and Windows 2008 required double byte UNICODE. Pretty ironic given the international nature of Xerox. The double byte UNICODE is required to support Asian alphabets. Anyhow, since Xerox were unlikely to fix that for we turned to Microsoft. There is a hot fix for Windows 2008 that restores the ability to use ASCII. KB975512 explains.

This issue surfaced again with an IBM AIX Unix system trying to mount a share on a Windows 2008 R2 clustered file share. Again, the sniff revealed ASCII and not UNICODE. Here is the line from the sniff, if the share name was correct you would see the 'zeroes' before each letter e.g. 'f i l e s e r v e r...' etc. The bytes will the value zero would appear as spaces.

SMB: C; Tree Connect Andx, Path = \\\APPS, Service = A:

Here is the response from Windows:

SMB: R; Tree Connect Andx - DOS OS Error, (67) BAD_NET_NAME


No comments:

Post a Comment