Crawling BCS fails RRS feed

  • Question

  • After having deployed an external content type that connects to MS SQL I created a new search content source using my external content type. After a FAST full crawl all the item from the BCS connection is shown on my search page – so all works fine. When I look at the item link on search result page all URL starts with “bdc3://…….”. This is because I have not created a profile page for my external content type. Adding a profile page also works fine except that a full crawl cannot crawl the BCS content source anymore. The crawl log shows no errors. I have found out that if I remove the <actions> section of the BCS model the full crawl works fine again. I have tested the scenario on a FAST installation with and without SP1 but in both cases I have the same problem.  

    Any suggestions what can be wrong?


    Best regards,

    Steen Westergaard

    Steen Westergaard
    Friday, August 19, 2011 4:07 PM

All replies

  • Hi Steen,


    I had some questions regarding the behavior you observed:

    1.       You indicated that the crawl log shows no error, which sounds like all successes were logged.  Can you confirm that the log was not actually empty?  An empty log would indicate a problem with SharePoint and not specifically with FastSearch.

    2.       Is the profile page of the BCS model created? This is needed for the crawler to index each BCS document separately

    3.       Is the user a member of BCS metadata store?  (I can email you a  screenshot if that will help)

    4.       Do you have full read rights to the object?

    5.       Has the farm config wizard been run after installing SharePoint 2010 SP1?


    Let us know your findings.



    Rob Vazzana | Sr Support Escalation Engineer | US Customer Service & Support

    Customer Service & Support                          Microsoft | Services

    Wednesday, August 24, 2011 8:17 PM
  • Hi Rob


    Thanks for your suggestions for solving my problem. See my comment to your point below:


    1. Before crawling I cleared the crawl log as well as the content collection. After importing my BCS model (which connect to AdventureWorks sample DB for testing purposes) I started a full crawl which resulted in 2 successes and 0 errors. The two successes were actual not entries from my database but two “stange” results:


    • Title: AdventureWorksLT2008R2&s_ce=0408c4g11gi000204080g010231408c4
    • URL:  bdc3://adventureworkslt2008r2_adventureworkslt2008r2/Default/00000000%2D0000%2D0000%2D0000%2D000000000000/2150/AdventureWorksLT2008R2/2152&s_ce=0408c4g11gi000204080g001023100


    • Title: 2152&s_ce=0408c4g11gi000204080g001023100
    • URL: bdc3://adventureworkslt2008r2_adventureworkslt2008r2/Default/00000000%2D0000%2D0000%2D0000%2D000000000000/AdventureWorksLT2008R2/AdventureWorksLT2008R2&s_ce=0408c4g11gi000204080g010231408c4


    I then went to CA and selected “View External Content Type” and deleted the default action (which is the link to the profile page). After a new full crawl I get 409 successes and 0 errors. From the crawl logs I can see that the URL for all the crawled is in the form:


    • bdc3://adventureworkslt2008r2_adventureworkslt2008r2/Default/00000000%2D0000%2D0000%2D0000%2D000000000000/2150/AdventureWorksLT2008R2/2152?s_id=izHMAAA==&s_ce=0408c4g11gi000204080g0010231001ue


    I can reproduce the above scenario on both my own development environment as well on another environment.


    2. Yes the profile page is created and works fines.


    3. Both the crawler and administrator account is member of the BCS metadata store with full rights. The same permissions are set on the BDC model and External Content Type.


    4. Both the crawler and administrator account is set to DB owner on the SQL database that are crawled.


    5. Yes, the farm config wizard has been run after installing SharePoint 2010 SP1.



    Steen Westergaard

    Steen Westergaard
    Friday, August 26, 2011 12:08 PM