Friday, April 22, 2005

Proxy Server Bug Crawling External Sites

In some implementations, the Sharepoint Search Service will fail to crawl content sources with fully-qualified domain names (such as http://www.anyserver.com) but will crawl internal sources using NetBIOS name resoultion (http://anyserver) . Although there has not yet been any confirmation from Microsoft, this appears to be a bug in SPPS. To work around this issue, perform the following steps:

1. Go to Sharepoint Portal Server Central Administration
2. Click on 'Configure account and proxy settings'
3. In the Proxy Server Settings section, click on the 'Use the proxy server specified' radio button.
4. Enter an IP address (it does not have to be a valid address as you will be bypassing it entirely).
5. Enter "80" in the Port field.
6. Check the 'Bypass proxy server for local (intranet) addresses' box.
7. In the 'Do not use proxy server for addresses beginning with:' text box, enter "http://*". This will bypass the proxy address for all HTTP requests. You may enter additional exclusions for HTTPS, FTP, etc. by separating each entry with a semicolon.

Note: Advanced users may recognize the "http://*" entry as the same setting available in the inclusions property of a content source; however, assigning this to a content source or index will not work - it must be assigned in the Proxy settings.