Philipp Schmid's questions

I am using the following appcmd to add an autoStartProvider to the list of autoStartProviders:

appcmd.exe set config -section:system.applicationHost/serviceAutoStartProviders /+"[name='MyAutoStartup',type='PreWarmCache, MyAssembly, version=1.0.0.0, Culture=neutral, PublicKeyToken=null']" /commit:apphost

Running the script for the first time, everything works correctly. But when running the script for the second time it complains that "MyAutoStartup" already exists. I know I can clear out the collection by using:

appcmd.exe clear config -section:system.applicationHost/serviceAutoStartProviders -commit:apphost

but that would remove all keys, not just MyAutoStart.

Is there a way to only remove MyAutoStart if it is present?

I got a second hand Zonet KVM3322 (USB switcher). There's supposed to be a keyboard hotkey combination that can be used to switch between the two attached computers (instead of manually click on the appropriate button on the device). I cannot find the manual online that would tell me what the hotkey combination is.

We are using Office 365 as our Exchange server. Our O365 account is a P1 plan and so our email accounts are not in the same domain as our local TFS accounts.

I have configured TFS 2010 to use our O365 Exchange server (same server address that Outlook is using), but of course there is no UI to enter the account credentials (only server address and email account to use). Therefore, the following error message in the event log is not surprising:

Detailed Message: TF271001: An error occurred while attempting to send an e-mail notification to the following address: [email protected]. Further e-mail notification errors that occur within the next five minutes might not be logged. Verify that the e-mail notification settings are correct in the Team Foundation Administration Console. Exception Message: The SMTP server requires a secure connection or the client was not authenticated. The server response was: 5.7.1 Client was not authenticated (type SmtpException)

Is there a way to configure TFS2010 to use an email server that needs authentication credentials (such as O365 or GMail)?

Edit: It appears that the information can be added/edited in TfsJobAgent.exe.config. So far I have not been able to successfully use a GMail account for the email relay.

I have successfully migrated a TFS 2005 Workgroup server to a new Team Project Collection in TFS 2010.

I am trying to use the TfsConfig Identity /change /fromdomain:WORKGROUPNAME /todomain:DOMAINNAME command, but nothings is being changed. Is this even possible or will I have to fat-finger the SQL Database directly to get this mapping working?

Or is there another way to fix-up those non-domain account identities in the new TFS 2010 project?

Using my NETGEAR WPN824v2, I would like to forward all ports to a single machine. However, when I try to add a single forwarding entry to an otherwise empty forwarding list, it complains that there is a conflict, but then seems to end up with a corrupted forwarding table internally (at least I am observing some of the mapped ports not being forwarded at all).

What is the correct way to forward all ports? Is it even possible? I realize it might not be advisable to do so, and instead list all ports explicitly.

Does a machine (e.g., mobile laptop) have to be domain joined in order to use Direct Access (DA)? Or does DA allow the user to provide credentials just like VPN does when setting up DA?

After an unexpected server shutdown due to a power failure, I can no longer connect to the internal windows database MICROSOFT##SSEE which is hosting Central Admin for my SBS 2008 server.

The log shows:

  Error: 18456, Severity: 14, State: 16.
  Login failed for user 'NT AUTHORITY\NETWORK SERVICE'. [CLIENT: <named pipe>]

I've tried to connect using the SQL Management studio (connecting to .pipemssql$microsoft##sseesqlquery) but no luck.

The SQL Server Configuration Manager doesn't show a entry for 'Protocols for MICROSOFT##SSEE' (but shows it for 2 other database hosted on the same SQL server 2005 Express edition.

I have tried to restore the master.ldf and mastlog.log files from a backup, but the issue persists.

We would like to host multiple customers in a single WSS 3.0 web application (so we can use https instead of custom SSL ports). Because they are different customers, we don't want them to 'see' each other when doing directory lookups. We have not been able to configure WSS 3.0 to map different OUs to different site collections within the same web application. Is that even possible? Or is there another mechanism to achieve our desired behavior?

We believe we can create the desired behavior by using separate web applications for each customer, but then we loose the ability to use a common SSL port.

We can use either an AD with separate OUs for the different customers, or a custom membership provider.

My server machine has 2 disks: a primary disk containing SBS 2008 and a second disk with an copy of Windows Server 2008 that I am not currently using (but don't want to delete at this time).

Windows Backup is automatically selecting both disks for backup because they both contain OS installations. I would like to exclude the second disk from backup. I have tried to hide the second disk before configuring my nightly backup, but alas it contains the boot loader (another issue to fix I guess) and therefore needs to be present.

So I am looking for a way to temporarily configure my second drive so it won't be considered an OS drive when setting up my nightly Windows backup. Can I rename some files/folders, or maybe modify the boot information using bcdedit?

For some unknown reason, my SBS 2008 port mappings are incorrect!

My SBS 2008 server uses the (static) IP address 192.168.0.10, but important ports (80, 443, 4125) are pointing to 192.168.0.20. My router (NETGEAR WPN824v2) is managed by SBS 2008 and there doesn't seem to be a way to change them via the browser windows (launched from the SBS console, some ports show, but not the one mentioned above), nor the Windows SBS Console application.

How can I either take manual control of the router, or change the incorrect port mappings using the SBS tools?

Update (clarification really): I can access the router directly from the SBS machine (other machines will show me a message saying that the router is managed by 192.168.0.10), but the 3 ports are not listed in the list of forwarded ports. Trying to set them using the browser interface will result in an error message stating that the port is already being forwarded!

Maybe there is a way to set the router to no longer be managed by 192.168.0.10?

Solution: Turn the UPnP functionality off on the Netgear router and voila, I can now control the port mapping on the router again for the 3 ports.

I know that adding the Hyper-V role to a Small Business Server 2008 installation is not supported as stated by Microsoft Support (KB 958829). I'm aware that I can use Virtual Server 2005 to host my testing virtual machines on SBS 2008 if I have to.

Nevertheless I was wondering if anyone has been able to fix the resulting networking issue (DHCP binding to the virtual adapter instead of the real adaptor) to allow for this very appealing scenario to work?

Does anyone know if a Windows Server 2008 R2-based version of SBS 2008 will not have this Hyper-V role limitation?

I followed these instructions on www.hottonetworking.com to set up VPN on a clean install of Windows Server 2008 SP2 (not R2 yet).

When I then establish a VPN connection to that machine from a client machine (running Windows 7 RC), everything succeeds (it seems since I get a 'Connected' state in the network sharing center window), but I end up with a subnet mask (according to ipconfig /all) of 255.255.255.255 instead of 255.255.255.0. The net effect is that I don't have local network or internet capability.

What additional configuration steps do I have to do to get VPN with the proper subnet mask working?

Update: Using the steps outlined in the Technet article mentioned by Mr. Nimble, I was able to get internet connection. Apparently the subnet mask is not an issue as my coworker was able to connect using his VPN connection and ping the server machine by name as well.

I have backed up Windows Server 2008 running virtualized on Hyper-V to a Windows Home Server 2008 SP1 (I know I should have backed up the VHD instead).

Now I need to restore the contents of the VM from WHS. I have created a restore CD ISO and used it to create a new VM.

It all works as advertised up to the point where the restore process wants to load the network drivers (it only finds 4 disk drivers on the restore CD. but no network drivers). So I created a virtual floppy and copied the contents of 'Home Server Drivers for Restore onto it. But no luck! I have tried moving the 4 subdirectories into the root of the floppy, but that didn't work either. Finally, I started another instance of the WS 2008 to identify the network driver that the virtualized instance is using (%WINDOWS%\system32\drivers\netvsc60.sys) and copied that file onto the virtual floppy, without success.

Does anyone have any suggestions on how to get networking working on a Hyper-V instance running off the Windows Home Server Restore CD?

UPDATE: As suggested by delenda, I have added a legacy network adapter to my VM, and indeed I now get a network driver listed! However, the WHS it still not found, even after entering the home server name manually.

PHS

I am able to connect Outlook 2007 to a WSS 3.0 calendar on our intranet site (same domain). Even though the dialog specifies 'Read' permissions, I am able to create or change an appointment in either Outlook or SharePoint and have the other party synchronize within a minute or so.

I am trying to connect to a second WSS 3.0 calendar which I can access via the internet (hosted by Small Business Server 2008 using SSL). I get the same setup experience (Permissions: Read) but any attempts to create or change an appointment from Outlook fail with a error message stating that I only have read permissions.

Why does it work in one scenario, and not the other? Is there a specific configuration setting that I need to apply to the SBS 2008 site? Or is a domain membership required for this feature to work (this seems to be the only difference between the 2 cases)?

My current configuration of WSS 3.0 is suddenly missing the 'Connect to Outlook' actions menu item!

Update: I have verified using the same browser instance loading a calendar from a different SharePoint site (hosted on a different machine) shows me the 'Connect to Outlook' action menu. This rules out any client-side issues.

It used to be there but I don't recall making any changes at all (I know, famous last words!). I'd like to avoid having to re-install SBS 2008 because of it.

Is there a log file that I could inspect for any errors?