Crawl of Large Secure Sites Fails with Site Account

• Microsoft Search, included with:
  • Microsoft Site Server version 3.0

SYMPTOMS

When you specify a site-specific account to crawl a large secure site, the crawl begins to fail to authenticate at some point. Access to the URLs is denied to the Search Gatherer.

CAUSE

The site-specific account information is not persisted during the crawl. If an error is encountered or if Search runs out of memory and reloads the information for the crawl, the information for the site-specific account is lost.

The site-specific account information also is lost if the site has a large number of secure pages that are being indexed by the crawl.

RESOLUTION

To resolve this problem, apply the latest Site Server 3.0 service pack.

To work around this problem, change the Search Default Content Access Account to the account/password needed to access the URLs for the site in question. If crawling many secure sites, a separate catalog is necessary for each site. This may not be a feasible workaround for crawling many secure sites as part of a single catalog.

To change the Default Content Access Account, perform the following steps:

1. In the Microsoft Management Console (MMC), open the Search snap-in and open the server in question.



2. Right-click the Catalog Build Server and select the Accounts tab.



3. Fill in the information for the Default Content Access Account.

STATUS

Microsoft has confirmed this to be a problem in Site Server version 3.0. This problem has been corrected in the latest U.S. service pack for Microsoft Site Server version 3.0. For information on obtaining the service pack, query on the following word in the Microsoft Knowledge Base (without the spaces):

S E R V P A C K

MORE INFORMATION

The site specific account information is now persisted, so the crawl will successfully recover from an error.

Additional query words: sp

Keywords : prodsitesrv3 prodsrch SS3Sp1Public Ss3Sp1fix
Version : WINNT:3.0
Platform : winnt
Issue type : kbbug