Unless you have custom named glyphs it's probably okay to wipe out the Unicode table and start anew. Save a backup though. As long as the glyphs are named correctly, you'll be okay. If you have glyphs with custom names you may not be able to do this.

Generate Unicode
standard table
DON'T try to keep unicode indexes

Sort glyphs by Unicode when you're done. Switch to Names mode and make sure the names correspond with the proper glyphs.

. . . or this could be your problem:

I've come across a rare case where I get an error message that multiple glyphs are mapped to the same code but it's not indicated by FontLab in names mode or codepage mode. If you check the supposedly offending glyph, the Unicode is set correctly, the name is set correctly, everything seems fine. If you look around for a duplicate glyph, it can't be found. Mark the glyph with a color then switch to index mode. Now you can find the duplicate glyph. You'll find two identical glyphs wityh the same name and Unicode but two unique index positions. It's only happed to me around five or six times.

Hi Ray,

The first solution did the trick... simply wiping out the existing Unicode table and start anew without checking that little box (!). Thank you very much!

Your description of the 'rare case' is also interesting: I'll keep that in mind (just like all Fontlab users have grown accustomed to all those 'rare cases'... ;-)