Only filter HTML documents with Index Server 2.0

[|Jan Rygh]

Hi,

I have a document root for my webserver that contains both word and html
documents.
I only want the Index Server to filter the html documents and skip the word
documents.

I have managed to set the Persistent Handlers for Office documents to point
to the query.dll filter
in the registry. This should cause the Word documents to be filtered by the
null filter, or actually not filtered at all.
So I thought everything was OK. But when I start the Index Server the
Persistent Handlers pops back to the
original settings.

Suggestions on what to do?

jan