Alex KeySmith's questions

When running a Test-NetConnection, I can see port 5985 connects ok from server A to B.

And doing a Test-WsMan from desktop C to server B, it works.

However a Test-WsMan from server A to B, does not work.

All machines are on different subnets, both server firewalls indicate a domain connection. And it appears the windows firewall is not on.

The WsMan error indicates it's a networking issue (I've replaced the domain name with an example name).

Test-WSMan : <f:WSManFault xmlns:f="http://schemas.microsoft.com/wbem/wsman/1/wsmanfault" Code="2150859046"
Machine="1.example.org"><f:Message>WinRM cannot complete the operation. Verify that the specified
computer name is valid, that the computer is accessible over the network, and that a firewall exception for the WinRM
service is enabled and allows access from this computer. By default, the WinRM firewall exception for public profiles
limits access to remote computers within the same local subnet. </f:Message></f:WSManFault>
At line:1 char:1

Most articles help diagnose firewall issues, but this does not appear to be one if the tcp connection works. I'm not sure what diagnostic techniques to try next to understand what the error could be.

I've created an Azure Active directory under my MSDN Azure subscription for testing.

The subscription is tied to our main company Azure AD.

I turned Azure Active Directory Identity Protection a while ago to experiment on my own AD.

However I suspect it's interfering (not the best word) with apps logging in, as I often see warnings that my login was blocked during an oauth flow, but after closing the dialog, it works fine.

I suspect when doing oauth flows that may not be scoped to a particular directory, the flow attempts to connect to both directories?

Is there a way to delete the azure active directory, if I select the directory in the top right of the azure portal, it instantly says my login is blocked.

I can't see a way of managing the directory without switching to it.

EDIT:

An example repro in powershell:

"Before Login"
Get-AzureRmContext

Login-AzureRmAccount

"After login"
Get-AzureRmContext

Here's the screen I get during login, but the "after login" shows it has logged in ok.

I'm running a .net 4.0 app in IIS6 in a separate app pool.

Is there a way to see a count of active sessions for that website?

I can see the active sessions in the performance counters, but only for all .net 4.0 apps, is there a way to break this down further? It's using in process session state.

Perhaps using powershell? Thanks for the help.

Can the recenttly opened sourced engine used on Azure for git deployment be installed on IIS 6?

https://github.com/projectkudu/kudu

The documentation mentions IIS7, but doesn't explicitly say IIS6 is incompatible.

I'm considering enabling SQLCLR on our SQL2005 production server, however technet states:

Computers configured with large amounts of memory and a large number of processors may fail to load the CLR integration feature of SQL Server when starting the server. To address this issue, start the server by using the -g memory_to_reserve SQL Server service startup option, and specify a memory value large enough. For more information, see Using the SQL Server Service Startup Options.

How much memory is "large amounts" in this case? As I realise that this was probably written prior to 2005, so large amounts back then could be the norm now? (same for the large number of processors).

Is there a way of deleting locked files in mailroot\queue in IIS6 Smtp?

We've been having problems with a huge bulk of e-mail being sent, we have now resolved this but there are 6 e-mails stuck in the queue which I wish to delete.

I've tried stopping the service, but it is now stuck in a stopping state.

E-mails were previously being sent even when in stopping. But these last few seem to be blocking other e-mails in the pickup directory.

I have read about a tool called aqadmcli.exe, but have been unable to find a working download link. I'm hesitant to use anything other than from an authoritative site.

To give a bit of context, we are looking to host our Microsoft dynamics e-mail router to connect to Google apps e-amil.

We are looking at many options from varying angles, but one possible option is looking at a cloud VM, however it occurred to me is the cost of bandwidth.... Ignoring the other options at the moment, this is just for arguments sake:

Is there a way in which to setup a cloud environment to have near-free bandwidth to gmail? I'm thinking perhaps a tunnel to a specific IP or perhaps hosting in Google Cloud Compute engine (I release GCCE probably isn't an options as I think they only support Linux).

My thoughts behind this are similar to how DB providers operate (e.g. mongolab.com) at the moment by having their database platform installed in your cloud provider to get it closer to the app server.

What happens if a stored procedure is run in sql server management studio (SSMS 2012 to 2005 server) and it crashes?

I'm guessing the stored procedure continues to run?

On the other hand I have read somewhere that if a connection is lost and a transaction is used, then the transaction automatically rolls back?

Is it possible to have a single site in IIS6, with multiple domains (not subdomains) some domains using SSL? The domains with SSL using a unique IP.

I've read similar situations, but nothing exactly the same as this. But close enough to think that it might cause problems.

I've possibly seen that wildcard ssl's might help due to them not being as mapped to IP address?

We are thinking of using sql users / roles for row level security.

I have read that the max is approx 32,000+/- users, because the sys.users has a smallint id.

However I've not been able to find anything absolutely conclusive.

n.b. I'm not after the number of concurrent users, but the max number of registered users.

thanks for the help

I wish to do some testing on my local machine which requires me to put the date forward a year. So I can test some stuff in SQL Server (i've found it not possible to override the sql server time).

However because I'm connected to a domain, I believe this is causing the time to be reset to normal a few moments after I try to override it.

I understand that it's probably important for the local and domain time to be insync, but is it possible to temporarily block the clock from being reset?

Disconnecting from the network isn't an option, as I require an internet connection.

If there is a manual way to block the time reset what are the likely reprocutions? I don't want to screw anything up on the network :-)

Is it possible to set a folder permission so that you (in this case a .net app), can see if a file exists - but not read any data.

I want to be able to set permissions, so:

1. A user can tell if a file exists
2. A user can write to the folder.
3. No other permissions (can't read / delete etc).

-thanks Alex.

If I use a UNC share for a virtual directory, and use "Connect As" to save the username and password.

I'm presuming that these saved credentials are secure? Is that correct? Are the credentials saved encrypted?

-thanks for the help.

If classic asp got into a redirect loop, would it be enough to kill the IIS service. Or does the act of redirecting not consume that many resources?

Is it possible to setup internet explorer favourites for everyone (or perhaps a group) on a domain?

We have a number of favourites that would be useful to share with all users.

I'm moving to a new machine and i'm hoping to transfer my repositories as easily as possible. I have had a quick look around and have seen people using SVN dump.

As i'm planning on copying all repositories can I simply copy the c:/repositories folder ?

I'm using visual SVN on windows XP.

What is the difference between adding MIME types in these 2 different situations?

When in IIS 6 manager...

1. Right click on the computer name (local computer) > properties > mime types
2. Right click on the "Web sites" folder > properties > http headers > mime types

I'm guessing that perhaps option 1 adds MIME types for FTP also? However if that were true i'd expect to be able to add MIME types specifically in the properties of FTP (and not just websites).

thanks for your help.