Hello
I am getting a stupid error. I have two Web Applications, one of them is the MySites and one is my corporate Web App.
So, I setup my Search Service, and crawling on the corporate Web App is working fine. But MySites throwing and error, and does not get crawled.
The errors are:
- sps3://my.corp.domain.com –> the object was not found
- http://my.corp.domain.com –> the object was not found
My apps are all with a host-name header, and all three, the http://corp.domain.com, http://my.corp.domain.com and sps3://my.server.domain.com are in the “Start adresses area”.
I checked that all my used accounts have the “retrieve people data for search crawlers” permission.
One of these accounts is my Default content access account.
So what could the issue be?
/emo
You still should have been able to have my.domain.com on port 80 too. You just need to have something unique between IP/Host Name/Port. So the host name of my would have made it different to say corp on port 80. You would have had to remove the listen to anything on port 80 binding if it was there to get it started in IIS as that would have conflicted with the corp web app. I am by no means an IIS expert though so I may be getting myself confused.
So if you can I would strongly recommend you get it set that way as I am sure that is how you wanted it to be?
Out of interest what are the binding settings of the two web apps in IIS?
I think a lot of the issue comes down to my.corp.domain.com as I am thinking there will be no zone in dns to cover this namespace.
I know i should change the account for search, but I am just testing so I thought I could go with that one. I do have a crawl account which has full read on the Web-app.
So I should try to add that one as my default content access account, and give that one retrieve people data for search crawlers” permission.
The reason why my Web-apps are called that is, that i have a corporate webapp with a host-name “corp” on port 80. I then created mysites web-app also on port 80, with a host-name too, so thats why it is my.corp.domain.com.
Is this a strongly wrong way to do it?
I will look at the post from technet.
I would strongly object to that configuration. You do not want the crawl account to have any greater permissions than Read as it can kill the security trimming of search results.
So it is my.corp.domain.com and not my.domain.com, was there a reason for this? I cannot help but think this is a major difference compared to the working web app of corp.domain.com
Are you able to resolve my.corp.domain.com on the server that is performing the crawls?
Might be also worth checking out a post such as:Â http://social.technet.microsoft.com/Forums/sharepoint/en-US/70e15ad6-0387-4a13-9d5e-58f56ccd8481/object-not-found-sharepoint-search-top-level-error
Sorry, i wrote the adresses wrong in the first time. They are corp.domain.com and my.corp.domain.com.
The user running the crawl is also my “setup” user, so it is the one I logon to CA with and create web-apps, so by default he is assigned the full permissions on my Web-app.
Just as a double check, does your content access account have read access at the web application level? So if you go into CA > Web Applications > Select my.server.domain.com web application > User Policy > Is the account that is being used to crawl the content specified in here with Full Read access?
Also on the server that is performing the crawl can you resolve the start addresses that you have specified? You could hope to check this by performing a ping from CMD to the required start addresses.
I also have never done the start address with FQDN address, cannot see this as being an issue though. I see that the format is name.server and then domain. Where I would normally expect just name of the web app and then domain. So I would definitely you can resolve those start addresses on the server performing the crawls.
