MOSS 2007 crawl broken
From: "alexlomas tta gmail tod com" <alexlomas att gmail dot com>
Newsgroups: microsoft.public.sharepoint.portalserver
Subject: MOSS 2007 crawl broken
Date: 30 Dec 2006 05:05:24 -0800
Hi All,
I've just done an (in place) upgrade from SPS 2003 to MOSS 2007 - everything worked just fine apart from the crawl service. Searching & crawling was fine before the upgrade. We also did a test upgrade with a copy of the SPS 2003 data on another server and that went fine (the only difference is that the test server doesn't use SSL and the production one does).
It will crawl external sites, file shares etc. without a problem but it won't crawl the local sharepoint sites. The Office server search service is running; you can also search from inside Sharepoint although it won't return any matches.
~~~~~~~~~~~~~~~~~~~~~~~~~~
The crawl log gives errors of:
https://sharepoint.mydomain.com
Error in the Site Data Web Service. (*** Client found response content type of 'text/html', but expected 'text/xml'. The request failed with the error message: -- <html> <head> <title>Server Unavailable </title> </head> <body> <h1><font face=Verdana color=#ff3300>Server Application Unavailable </font></h1> <p> <font face=Verdana> The web application you are attempting to access on this web server is currently unavailable. Please hit the "Refresh" button in your web browser to retry your r)
sps3s://sharepoint.mydomain.com
Error in PortalCrawl Web Service. (*** Client found response content type of 'text/html', but expected 'text/xml'. The request failed with the error message: -- <html> <head> <title>Server Unavailable </title> </head> <body> <h1><font face=Verdana color=#ff3300>Server Application Unavailable </font></h1> <p> <font face=Verdana> The web application you are attempting to access on this web server is currently unavailable. Please hit the "Refresh" button in your web browser to retry your r)
The event log has:
Event Type: Warning
Event Source: Office Server Search
Event Category: Gatherer
Event ID: 2436
Date: 30/12/2006
Time: 12:49:51
User: N/A
Computer: xxxxx
Description:
The start address <https://sharepoint.mydomain.com> cannot be crawled.
Context: Application 'Default Web Site', Catalog 'Portal_Content'
Details:
Error in the Site Data Web Service. (0x80042616)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I've tried resetting the index, removing and replacing the indexing functionality, removing SSL from the IIS site, set IIS authentication to NTLM, told the crawler to ignore errors in SSL certs (although there isn't a problem with name mismatches) and extended MOSS onto another IIS website on another port and asked it to crawl that... All to no avail.
The rest of Sharepoint is fine, that is, you can browse to https://sharepoint.mydomain.com without problem and use all the rest of the functionality (oh, except the spell checker, but that's for another day!).
Foolishly I've done the upgrade at a quiet time thinking it would all work, so I'm unable to get hold of MS PSS until Tuesday now. Although the search stuff isn't critical it would be nice if I could get things resolved before people are back to work (and at least I can take the site down and tinker with it at the moment).
I have seen people with similar issues on newsgroups, but I haven't spotted a resolution yet so any help is very welcome! :)
Thanks,
Alex
**************************************
From: "Peter Lawton"
Subject: Re: MOSS 2007 crawl broken
Date: Sat, 30 Dec 2006 21:26:56 -0000
Newsgroups: microsoft.public.sharepoint.portalserver
Hi Alex ;-)
I've just spent the last 2 weeks (over xmas too, I am a sad ....) struggling with MOSS indexing locking up and the crawls never ending, I suspect the MOSS search will get a few patches soon
As it turns out it was a corrupt file in a document folder causing all my problems, it doesn't seem to take much to upset it.
What you could try is creating another shared services provider, moving all the applications over to the new one and then deleting the old one - it might help
Peter Lawton
**************************************
From: "alex lomas"
Newsgroups: microsoft.public.sharepoint.portalserver
Subject: Re: MOSS 2007 crawl broken
Date: 30 Dec 2006 13:57:18 -0800
On Dec 30, 9:26 pm, "Peter Lawton" wrote:
> What you could try is creating another shared services provider, moving all the applications over to the new one and then deleting the old one - it might help.
I had seen that mentioned but I'd shied away from it because I wasn't quite sure on how to do it... any pointers? I'd hate to loose all the content of the server :)
**************************************
From: "Peter Lawton"
Subject: Re: MOSS 2007 crawl broken
Date: Sat, 30 Dec 2006 22:18:38 -0000
Newsgroups: microsoft.public.sharepoint.portalserver
I have to admit I've only done it once on a test server that had very little content anyway and I didn't check everythign to see what got lost, probably best to practise on a test system if you have one first.
Basically you just create a new SSP then move the existing applications onto it.
Peter
**************************************
From: "alex lomas"
Newsgroups: microsoft.public.sharepoint.portalserver
Subject: Re: MOSS 2007 crawl broken
Date: 4 Jan 2007 03:56:43 -0800
Having spent a couple of days working with PSS on this, the fault is apparently a bug in MOSS 2007 that means it can't crawl SSL enabled Sharepoint sites.
The workaround is to extend the existing sharepoint web application onto another IIS web site (e.g. port 81) - the same content will now be exposed on two URLs.
You then configure the default SSP's searcher to point at the new (non-SSL) website, e.g.:
http://mysite.com:81
sps3://mysite.com:81
As both URLs use the same index, search results are exposed on your original site *with the correct URLs* (you can also use name mappings to correct them if you like)
You can then set some IP restrictions on the port 81 site so that only the index server can access it and avoid confusing your end users.
Posted using 