| Q |
I can't search PDF files ! |
| A |
By default, Index Server can only search HTML,
Text and Microsoft Office files. To allow Index
Server to search PDF documents you need to install
a special filter (Adobe PDF IFilter) from Adobe.
This can be downloaded from the following location:
http://www.adobe.com/support/downloads/product.jsp?product=1&platform=Windows
|
| Q |
None of my links work ! |
| A |
To solve this problem, you need to associate your
catalog with the appropriate WWW server. This is
explained in point 5 of my Index Server tutorial:
http://www.simongibson.com/intranet/indexserv/
Alternatively, if you are using an earlier version
of the script, you should upgrade to v2.0 which
now includes automatic link generation. Existing
users can use their original download link and password
to update.
|
| Q |
My abstracts don't appear ! |
| A |
The Windows 2000, 2000 Pro, XP and Windows Server
2003 implementations of Index Server (Indexing Services)
don't generate abstracts by default. So, to set
this option you need to ensure that "Generate
Abstracts" is checked in your catalog properties
(see point 5 of my Index Server tutorial at the
link below):
http://www.simongibson.com/intranet/indexserv/
|
| Q |
Wildcard searches (eg. car* for cars,
carton, carbide etc) don't return the expected results
! |
| A |
This is because the default query language for
Microsoft Indexing Services 3.0 is "free-text".
That means that the * in the query is taken literally
and hence a search for "car*" will return
less (or no) results when compared to a search for
"carbide".
This problem has now been fixed in version 1.1 of
the SearchSimon script (12/06/2003). Existing users
should download the updated version using their
original link and password.
Special thanks to Ken Otto for his help in solving
this problem.
This problem is explained in more detail in the
Microsoft article link below:
http://support.microsoft.com/?id=320942
|
| Q |
Links for some documents don't appear
! |
| A |
Where the title of a document consists only of
spaces, the document title won't appear and hence
the link won't be available.
This is due to a bug in the script and has now been
fixed (14/06/2003). Existing users should download
the updated version using their original link and
password.
Special thanks to Mason Cooper for his help in solving
this problem.
|
| Q |
My results don't contain the expected
documents!
Some documents are missing, why? |
| A |
The Indexing Service is quite intelligent when
searches are performed on documents contained on
the local machine. This means that it will only
display results that the user has the permissions
to view. People often find that documents are not
returned in the list of results when they should
be.
If your web site uses Anonymous Access, the account
that is used to query the Indexing Service is
IUSR_<machinename>. This account MUST have
access to all documents you are indexing.
If your web site uses Integrated Windows Authentication,
you can apply read permissions to all indexed
documents for the DOMAIN USERS group. Or, leave
the permissions as they are if you want to take
advantage of this Indexing Service feature.
When the Indexing Service is indexing documents
contained on another machine’s network share
it uses a single user account to access the documents.
In this case, the service will not check permissions.
|
| Q |
Links point to the full physical path of the document and not the UNC share |
| A |
To force the indexing service to use the appropriate UNC path instead of the physical path for local indexed documents, follow the steps in this tutorial:
http://www.simongibson.com/intranet/forcepathalias/ |
