Hi,
We are currently running a Windows web server with approximately 250 sites and are recieving an intermittent 500.19 error, which affects a random selection of sites, and it affects ASP, ASP.NET and sites with naked HTML.
We see the following error when we view an effected site (apologies for munging):
Server Error ============ Internet Information Services 7.0 Error
Summary HTTP Error 500.19 - Internal Server Error
----------------------------------------- The requested page cannot be
accessed because the related configuration data for the page is
invalid. Detailed Error Information Module IIS Web Core Notification
BeginRequest Handler Not yet determined Error Code 0x80070038 Config
Error Cannot read configuration file Config File
\\?\UNC\10.0.3.25\home\domain.co.uk\public_html\web.config
Requested URL http://domain.co.uk:80/ Physical Path
\\10.0.3.25\home\domain.co.uk\public_html Logon Method Not yet
determined Logon User Not yet determined Config Source -1: 0: Links and
More Information This error occurs when there is a problem reading the
configuration file for the Web server or Web application. In some
cases, the event logs may contain more information about what caused
this error. [1]View more information » 1.
http://go.microsoft.com/fwlink/?LinkID=62293&IIS70Error=500,19,0x80070038,6001
This is often accompanied in the event log with an event with the message :
Could not load all ISAPI filters for site 'DOMAIN.CO.UK'. Therefore site startup aborted
I've previously seen a similar error reported at the link below, but this does not seem to help us as changing the MaxCmds registry entry seemingly has no effect.
http://forums.iis.net/t/1157405.aspx
The web server is Windows Server 2008 and the UNC share is a RHEL box with samba, site authentication is set to a specific domain user, asp impersonation is enabled, and all sites have individual application pools with their identity configured to the specific domain user.
When we run Network Monitor, we seem to find packets indicating that Windows has failed to send packets it's requested a response from for an affected site, however the network is running at 30-50% of available bandwidth and very occasional TCP sacks seem to be unrelated to this problem.
I would appreciate if any one could point me in the right direction.