• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 3472
  • Last Modified:

Terminal Services on Windows Server 2008 freezing

I have an interesting problem for you all!

We have a slightly odd setup at our office for performing remote maintenence on client systems. In the past we used to have a bench of P.Cs on our DMZ subnet, and these were loaded with VPN clients (Cisco, Checkpoint etc..) Our support staff could use them to connect to remote networks, outside of our corporate Lan.

We recently moved offices, and due to the configuration of the building, there were staff spread over 2 floors all requiring access to a DMZ machine.

After tossing a few ideas around, we decided on a solution involving virtual machines.

We purchased the following:

1) Dell server with Win 2K8, 15K SAS drives, Quad-Xeon 3GHz CPU and 8GB of RAM
2) 10x Terminal Services Licenses
3) VMware Workstation

To connect to a remote system, a user can connect to a TS session on this server (which is on the DMZ) and launch an instance of VMware workstation. Once on VMWare workstation, a Virtual Machine is loaded with its own virtual NIC and DMZ ipaddress, and can be used to connect to a remote client as if it were a regular DMZ PC. The advantage is that the only data units being exchanged with the corporate network are RDC packets.

It all works surprisingly well, however, recently, the TS Sessions have been freezing every 10-20 minutes or so, for around 60-120 seconds. Its causing riots in customer services!

Interestingly, it seems to be the TS session that freezes, as when it un-freezes again, the virtual machine jumps forward and remains connected to a VPN, if one has been established.

The freezing happens whether there are many users or few.

The CPU usage never goes above 30%

The Memory usage never goes above 85% (is this high?)

When everything is running, performance is always excellent.

When there is a freeze, it affects everybody

I have monitored the services with a utility to try and catch an offending service when the freezing happens, but nothing jumps out at me.

There are no relevant errors in the event log.

Does anybody have any clues?

BTW, now ESXi and the VMWare infrastructure client are free, we are thinking of going down this route. However, it would be great to solve this as the outlay for the TS CALs and the OS was not insignificant!

Many thanks,

Olly  
0
ollygd
Asked:
ollygd
  • 4
  • 3
1 Solution
 
ACollyerCommented:
Does the server console (i.e. hard wired keyboard/mouse) respond at all when a freeze occurs? It'd be good to know so we can narrow it down to either an RDP/service/resource issue or a more general server issue. I've had servers behave similarly when a hard drive is failing.

Maybe check any monitoring tools supplied with your storage controller (e.g. Intel Matrix Storage Manager comes with their ICH controllers) to check the SMART status and/or health of your drives (although to be honest, I'd expect something to appear in the event log if something was wrong.) While you're at it, actually, try installing the latest storage driver too.
0
 
ollygdAuthor Commented:
Thanks for getting back to me!

When a freeze occurs, the entire server consol locks up. After the wait of about a minute,  the RDC turns grayscale and flashes it's disconnect warning, then it reconnects again and resumes. The vast majority of the time, the session remains intact on the server.

Interestingly, Dell are pushing an urgent upgrade to the SCSI controller. Its a Dell PERC 6i Integrated. I'll run that tonight and see what happens. I'll upgrade the SAS drive firmware aswell.

Thanks so far! Ill keep you posted.
0
 
ollygdAuthor Commented:
Well, all the storage and drives look fine. I've updated all the firmware but there seems to be no let-up to the freezing issues.

There is still nothing of any relevence in the event logs :-(

I updated the network card firmware aswell.....although I forgot to update the drivers. I'll try that tonight. Do you think it could be a network card issue? This server (a PE1950) has duel Broadcom BCM5708C NetXtreme II GigE cards.

Just one card is in use, and is connected to a trunk port on a Cisco Cat 1950 switch.

Thanks so far!
0
Concerto Cloud for Software Providers & ISVs

Can Concerto Cloud Services help you focus on evolving your application offerings, while delivering the best cloud experience to your customers? From DevOps to revenue models and customer support, the answer is yes!

Learn how Concerto can help you.

 
ACollyerCommented:
I don't think you should have it connected to a trunk port - usually trunk ports are only used between switches.... Try connecting it to an "access" port (on the relevant VLAN, of course) and see what happens.
0
 
ollygdAuthor Commented:
I have the trunk port set as I have multiple MAC addresses associated with the network adapter on the server (due to all the virtual machines running on it)

I am open to this as a problem - unfortunately, it has to stay this way! :-)
0
 
ACollyerCommented:
Are you sure it has to be a trunk port? I was under the impression that trunk ports were only used for trunking multiple VLANs between switches.

If all your VMs are on the same VLAN, I'd have guessed you wouldn't need the trunk port.
0
 
ollygdAuthor Commented:
AH! I discovered the problem. Thanks for all your help, but it turned out to be down to the individual machine running out of space on their virtual HDDs. 3 or 4 of them were maxed out, and caused the server to hang - probably trying to thrash the paging files or something.

Cheers

Olly
0

Featured Post

 The Evil-ution of Network Security Threats

What are the hacks that forever changed the security industry? To answer that question, we created an exciting new eBook that takes you on a trip through hacking history. It explores the top hacks from the 80s to 2010s, why they mattered, and how the security industry responded.

  • 4
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now