Justin James's questions

I have the strangest IIS error that I've ever seen in my life.

I have an application/directory on an IIS server, that throws an error 500 when accessing ANY of the content in it, including HTML documents, when accessed via HTTPS AND the machines FQDN. When I access it with "localhost" it works fine. When I added a bogus entry for the NIC's IP in the hosts file, it worked fine. When I access it with the machines name and HTTP it works fine.

Here's a chart (the machine's name is "lofn.titaniumcrowbar.com"):

http - lofn.titaniumcrowbar.com: works https - lofn.titaniumcrowbar.com: broken https - localhost: works https - temp.titaniumcrowbar.com (put into hosts file): works

I set up tracing, and I got some useless information:

"The I/O operation has been aborted because of either a thread exit or an application request. (0x800703e3)"

This would make sense, except this happens when pulling up static content. While the directory may be an "application", the content is all static in it.

Any/all suggestions, no matter how strange, are VERY appreciated.

Thanks!

J.Ja

I've been struggling for a while with a problem with ActiveSync, it is giving 401.2 errors when trying to access the mailboxes. Oddly enough, clients can get to the calendars just fine. I've tried from an Android device as well as the Exchange remote connectivity troubleshooter, same error both times. Here is the configuration:

• Connectivity through ISA 2006
• Exchange 2007 on Windows 2008
• Basic authentication is on
• ISA is passing through all authentication directly to the Exchange server
• The user IS allowed to get to ActiveSync in their Exchange mailbox settings

I have tried the following, without success:

• Making SSL not required
• Enabling Windows Integrated and Digest authentication

Here is an example of the errors in the IIS log:

OPTIONS /Microsoft-Server-ActiveSync/default.eas - 443 - 192.168.32.18 TestActiveSyncConnectivity 401 2 5 0

Here is what I am seeing in my trace log:

ModuleName IIS Web Core 
Notification 2 
HttpStatus 401 
HttpReason Unauthorized 
HttpSubStatus 2 
ErrorCode 2147942405 
ConfigExceptionInfo  
Notification AUTHENTICATE_REQUEST 
ErrorCode Access is denied. (0x80070005) 

I'm 100% stumped. Any clues?

I have an existing OCS 2007 R2 install (single server internal, single Edge Server). It's been running for a while, and now I am trying to change its configuration. We have been using a NIC directly connected to the public network for A/V Edge, as R1 required, and I am changing it so that the A/V Edge is connected to the public Internet via NAT, behind our ISA server.

As required, we have our internal DNS server configured so that the A/V Edge machine pulls up the public IP address when it queries the A/V Edge FQDN. In addition, I went into the A/V Edge configuration, and told it to enable NAT.

Internal clients can use the A/V fine. However, external clients are doing something very strange. A packet capture on the external client machine show that it sets up the conversation with the usual chatter to the access and conferencing edge servers. Suddenly, instead of trying to access the A/V Edge's public IP, it tries accessing our base domain name's IP address via HTTPS. In other words, instead of looking up av.domain.com, it looks up domain.com. Then, after that, it does actually try the correct public IP address of the A/V Edge Server via port 3478 (TURN).

But here's where it gets weird. After a little more back-and-forth with the access and conferencing edge servers, it then tries the STUN conversation (over ports in the 50,000+ range) with the internal private IP address of our OCS server. Needless to say, these queries go nowhere, and the client is unable to get the audio/video.

Any suggestions?

Thanks!

J.Ja

I set up System Center Virtual Machine Manager some time ago. We barely use it, except to do conversions from VMWare to Hyper-V. When I initially set it up, it worked fine for this purpose. Now, when I try to do a V2V conversion, the job fails during the transfer of data over BITS; it says that the connection timed out. The source VM, destination, and SCVMM machine are all the same machine. I've looked at the event logs, and I see the URL that is trying to be accessed by BITS. I've popped that URL into my browser, and it doesn't work. I've tried packet captures, but the traffic never appears on any of the adapters since it is all loopback traffic. When I try the hostname of the URL in a manual telnet session to the SSL port, I get a connection, so I know something is listening. I checked the SSL cert bindings, and I beleive that a cert is bound. I don't get any errors in my browser regarding certs when I try. I did try trimming the URL up a bit, and I am getting 404 errors when I try the top level directory. However, SCVMM uses BITS for the transfer, not pure HTTP, so I may not be doing the right kind of test.

UPDATE: I tried it with a VMWare VM copied from a different server, it magically worked. So, somehow, SCVMM's ability to dish out the file is based on the file itself. I also tried to start the same said VM on a different machine, under VMWare Workstation, and it failed too. There is something very, very wrong here. It could have occured on the transfer to the SCVMM machine (I doubt it, I tried twice), or it could be something not right about VMWare Server.

Any help would be much appreciated! I've been banging my head on this for half of a week now.

Thanks!

J.Ja