Paul Lee

Sharepoint bumps and bruises

  Home  |   Contact  |   Syndication    |   Login
  11 Posts | 0 Stories | 7 Comments | 0 Trackbacks

News

Archives

Post Categories

Wednesday, April 22, 2009 #

Environment: SIngle server farm - Windows 2003 - MOSS SP1

I found another server with the same "access denied" error in the crawl logs.  So I applied the Loopback fix but then another error:

The system cannot find the path specified. (Exception from HRESULT: 0x80070003)

After searching the internet, I found the solution.  You have to make sure the "gthrsvc" folder exists under your search account's Temp directory.  Then restart the search service. 

CMD> net stop osearch

CMD> net start osearch

Then run full crawl.  Worked for me. 

I think this is another side effect of one of the recent server patches.  This particular server has been around for a while and the crawls never had a problem until now.  Going to check the rest of my MOSS servers now...

  • Share This Post:
  • Share on Twitter
  • Share on Facebook
  • Share on Technorati

Got woken up this morning by frantic people in Amsterdam.  Their document libraries stopped receiving emails! 

First thing I checked was to make sure emails were still making it to the sharepoint server.  To the IIS drop box at least...  So I sent a test email there.

Yep, they were definitely getting there.  And they were disappearing too so that means the timer service was working and picking them up.  So where the hell were they going? 

So I throttled the email events:

After sending another test email, here's what showed up in the event logs:

Event Type: Warning
Event Source: Windows SharePoint Services 3
Event Category: E-Mail
Event ID: 6873
Date:  4/20/2009
Time:  3:15:26 PM
User:  N/A
Computer: XXXXXXXXXXX
Description:
An error occurred while processing the incoming e-mail file C:\Inetpub\mailroot\Drop\f83e047f01c9c1b9000000bb.eml. The error was: Unknown alias..

The event type was Warning.  This is just a Warning?!?!?

Here's what I found out before calling premier and before doing any more research on this.  If you go to the Incoming email settings of the document library:

You can just click OK on the settings page without changing anything and emails will magically start working again.  Wow.  What a fix!  The best thing is that the end users can do this themselves.  :)  So I called them and asked them to go to each of the document libraries and do this. 

Here's the real cause and fix for this after doing some research.  Occasionally, the email addresses in the config db can get deleted (corrupt) without warning. 

The quick fix is:

stsadm -o refreshdms -url <url of web application>

The permanent fix is to install the hotfix http://support.microsoft.com/kb/961750 or even better, install the Feb CU and get it over with.

Apparently, if sharepoint doesn't know what to do with an email or there is some kind of error (or warning) it will just be launched into orbit.  Never to be seen again.  Shouldn't it go into the Badmail folder or something?

  • Share This Post:
  • Share on Twitter
  • Share on Facebook
  • Share on Technorati

Environment: MOSS SP1 with Feb CU - WIndows Server 2003

I noticed this issue today after installing the Feb CU last week.  The sharepoint crawler was not crawling any sites that had a DNS alias to itself.  For example, my servername was myserver01 and I have a DNS alias called myportal.  The crawler would not crawl http://myportal and anything under it.  This sucks. 

My error from the event viewer:

Access is denied. Check that the Default Content Access Account has access to this content, or add a crawl rule to crawl this content. (0x80041205)

My error from the crawl logs:

Deleted by the gatherer (This item was deleted because its parent was deleted.)  Huh?

The fix

Apparently, now you have to Disable the Loopback Check. 

- Open regedit

- Go to HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa

- Under Lsa, create a DWORD value called DisableLoopbackCheck

- In Value, type 1

- Close regedit

Crawls work as normal.

Update:

I wrongly blamed the Feb CU for this.  It was just a coincidence as I see the problem with one of my servers without the patch.  I'm now thinking it was one of the recent (conflicker?) patches put on by the server boys...

  • Share This Post:
  • Share on Twitter
  • Share on Facebook
  • Share on Technorati