maik's questions

I'm having (yet another) strange problem with IIS. When viewing an ASPX page I've designed on my local machine by browsing to http://localhost/page.aspx the page looks as expected (and looks the same in IE, Firefox and Chrome. If I change localhost to my_hostname the page is rendered with a disabled vertical scroll bar.

The behavior was first noticed when I published my site to our live server and saw the same discrepancy. After beating my head against the wall I tried what I described above and was able to duplicate my "problem". So with that, I turn to you guys.

This wouldn't really be an issue (save for the cross-browser inconsistency) except that this screws up an "absolute"ly positioned <div> moving it partway off the screen instead of being centered like it should be (and is when viewed any other way except in IE when the address is anything but localhost).

As another test I added a new aspx page to my project and didn't add or change any of the default code. If I browse to the page using localhost there is no scrollbar. If I browse to the page using my_hostname the scrollbar is there. Whatever the difference is it's making IE's processing of CSS get screwed up, to the point where at first everything works the same in all browsers I'm testing in, and afterwards IE just makes up its own rules. This is incredibly frustrating and I'm really hoping I'm just doing something wrong and it's not an inherent problem.

I've developed a couple of internal sites for my organization that use integrated authentication. Ultimately we want these sites to be accessible externally to users with domain-joined computers. The sites work as expected on domain computers while on the internal network. The problem comes when I take my laptop home and try to access those sites.

IIS only has integrated authentication enabled for the two sites. When I browse to the site using IE8 I get a username/password prompt asking for domain credentials. I can put those in and it will work, but the goal is to use the cached token for integrated authentication.

Next I reasoned that IE wouldn't response to an integrated auth request (is NTLM the right term for this?) unless the site was trusted. I tried adding the site to Trusted Sites but I get the same behavior as the before. I then added the site to Local Intranet sites and that is where things get weird. I get a generic error page from IE, no error code or anything.

Just for funsies I loaded up Firefox (which I had previously set up to use integrated authentication) and I added this new site to network.automatic-ntlm-auth.trusted-uris. Much to my surprise I was able to load the pages up with no problem at all and saw exactly what I was expecting (including verification that the integrated authentication worked).

My mind is a bit boggled at the moment as I'm not really sure where to go from here. I was hoping some of you may be able to provide some insight.

Update 05/11/10 I got to messing with Fiddler as recommended below and saw some discrepancies between how Firefox and IE handle the page request. When I tested Internet Explorer it makes one request, hits a 401.2 (server configuration error) and dies. There are no security log entries when this happens. Firefox encounters this same error initially, but subsequently sends another request and authentication is successful.

Keeping in mind the initial testing was all from outside the network, I looked at a successful auth by IE8 when done inside and it appears to behave the same way as Firefox both internally and externally. The initial request is met with 401.2, then another one is sent and is met with 401.1 (the challenge?) followed by a 200 (success).

So the problem seems to be that when accessing the "intranet" site from outside the network, IE8 experiences a 401.2 and then dies when it should (by measure of other browsers) keep going with the authentication process.

Can someone verify the Fiddler activity that I described?

I'm not sure if I'm just not asking Google properly or what, but I can't come up with a good answer to this problem.

We have MDT 2010 setup and have a Task Sequence for refreshing Windows XP machines. It doesn't seem to happen all the time, but a lot of the time when we start a refresh it goes through the normal motions and when it gets into the first part of Windows XP setup (the blue screen) it stops, telling me a Windows installation already exists at that location and I can press L to continue, erasing everything and using that folder.

I've poured over the unattend file and can't find an option that will just delete the old files and keep going, so I'm at a loss. Any ideas?

I'm trying to sort out what's expected from DNS in a Windows domain environment when a non-Windows, non-domain member tries to resolve a computer's hostname without it being a FQDN.

Windows clients (for what I agree to be good reason) will take their domain name received through DHCP and use it as the DNS suffix when resolving hostnames so that if I do nslookup server-1 it requests the record for server-1.example.com. If I try the same thing from my Mac it only does the lookup for server-1 and fails. The domain name received from DHCP is in /etc/resolv.conf as domain example.com, but because it's not listed as search example.com the lookup fails without appending the domain. The latter also happens for my linux-based mobile phone.

I don't really care about the why because I'm sure both platforms have their reasons, what I'm trying to figure out is how to rectify the situation with no modification to the clients. IANA says DHCP/BOOTP option 119 is the option for DNS domain search list, but it appears most platforms don't support option 119 out of the box. Windows doesn't seem to at all anymore, though that isn't a big concern, and *nix platforms only support it when ISC DHCP4 or greater is used. Not sure about Mac but I've read here and elsewhere that it doesn't support option 119 either.

Any ideas?

I asked this question on SuperUser a few days ago but haven't gotten any responses yet. I figured I would try my luck here since a) there isn't much of a difference between disk manager in Vista and Server 08 and b) it's backup related and I suppose could fall into a server problem should anyone run into this when performing backups on a server with the same (albeit bad) configuration.

Also, I haven't been around these communities too long, so if this is a no-no please let me know and I'll delete it (or rephrase it :)).

I have an odd situation. I have 2 drives in my computer running Vista, the OS drive (C:, E:) and a storage drive (D:). In Disk Management the C: volume is listed as the boot partition and has all of the system files. The D: volume has the "System" flag, but (right now) there's absolutely nothing on it. I'm hesitant about pulling the drive or even repartitioning using a 3rd party tool (since Windows won't let me) as I've read that if a system volume isn't detected Windows can have a conniption. Any ideas?

The main reason this is a problem is because Windows won't let me exclude this volume from backups, but it's where I am already storing preliminary backups of other things.

I'm a bit of an Exchange admin noob so I don't know if this is normal, but on our Exchange 2007 server under C:\Program Files\Microsoft\Exchange Server\Mailbox\First Storage Group\ there are 30,000 1MB .log files. I don't know if those should be there or not, but they keep multiplying like little jackrabbits. I've so far resisted deleting them as I'm sure something would break, but it's come to the point now where our 80GB disk is completely full up and Exchange has stopped receiving messages.

My best guess is that these are transaction logs and would be useful in rebuilding the database should something go awry, but 30GB of transaction logging seems a bit excessive. Is there a way to limit it or is that even a good idea?

And of course did I completely miss the point?

I'm trying to get TS gateway going on a 2008 Server. I'm running into trouble with authentication. I found that I needed to enable Windows Authentication in the RPC site. While Windows Authentication is enabled the RD gateway works flawlessly. After about 5 minutes though it appears to reset to disabled. I've looked high and low and can't find a setting to make it the default or to keep it from inheriting some parent object's settings. Any ideas? I'm assuming it's more of an overall IIS7 thing than just the RPC site not playing nice.