| Author |
 Topic  |
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
 Posted - 09/14/2008 : 4:28:41 PM
|
Hi,
I’m building High Available solution using Hyper-V and Microsoft Fail over cluster. My target is to double all hardware to eliminate single point of failure. I know how to provide iSCSI network redundancy, heart beat network redundancy. I also know how to set up firewall redundancy.
Unfortunately weak point in my plan is public network redundancy. I tried to find any solution for eliminating single switch and NIC point of failure (as far as I know I can use teaming due not compatibility with Hyper-V issue). In my opinion it is quite probable that one NIC or switch will be down (especially that I met Cisco switch completely stopped to work). Only advice I’ve got is to provide second, stand bay network and when I will find NIC or switch be down I can manually change virtual switch to another adapter (it is not solution I’m looking for. It has to be automatic process).
However I have still hope someone can give me answer for this question.
Best Regards
|
|
|
wkasdo
Moderator
Netherlands
6140 Posts
Status: offline |
Posted - 09/15/2008 : 02:52:20 AM
|
> as far as I know I can use teaming due not compatibility with Hyper-V issue
Depends. With recent drivers, failover teaming tends to work. Load balancing does not. |
 |
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/15/2008 : 04:16:48 AM
|
Many, many thanks for this reply. It is giving me a hope.
I’m going to use DELL servers, (probably PowerEdge 2950 III). I don’t care about load balancing but I would like to be sure fail over is working perfectly. Could you give me some additional information about this set up please? I didn’t even know it is possible to set up only fail over in timing…. Is any link relating to this problem?
|
|
|
|
wkasdo
Moderator
Netherlands
6140 Posts
Status: offline |
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/15/2008 : 07:22:50 AM
|
Basically what you are saying is when I disable Load Balancing on teaming it will work on any network adapter? It is good news for me.
What about switch configuration. Do I need special switches for this as for LB NIC teaming?
|
|
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/16/2008 : 07:31:04 AM
|
I found thisƒ¼
Hyper-V Virtual Networking not functional on Broadcom Network Adapters with
BACS-based teaming and VLANs
You can use the Broadcom Advanced Control Suite (BACS) software to team
multiple adapters to offer fault-tolerance and load-balancing capabilities.
Hyper-V virtual networks that are bound to network adapters configured with
BACS-based teaming or VLANs may not be functional and may disrupt
network traffic from virtual machines to the external network.
There is currently no workaround to support Hyper-V virtual networking on
BACS-based teams or VLANs. A resolution may be available in a future
release. Dell currently does not support BACS-based Teams or VLANs on
network adapters bound to Hyper-V virtual networks.
This post is from July so maybe something has changed. It was very interesting about his new drivers I can use and disabling load balancing. It seems to be good way¡K Could you give me some more details please? Maybe source I can read more about this¡K
|
|
|
|
joe_elway
Honorable But Hopeless Addict
Ireland
6673 Posts
Status: offline |
Posted - 09/16/2008 : 08:12:06 AM
|
| I believe you're in the same boat as the rest of us. No NIC teaming native in Windows and the OEM's have pretented that Hyper-V doesn't exisit. I'm using Broadcom NIC's on HP h/w and have no teaming. |
Aidan Finn
MCSE, MVP (Virtual Machine: Systems Administration)
IT Blog: http://www.aidanfinn.com
My Photography: http://www.aidanfinnphoto.com/
My Hyper-V Book: Mastering Hyper-V Deployment
Twitter: http://twitter.com/joe_elway |
|
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/16/2008 : 08:19:18 AM
|
Yes, it is really sad. Hopefully this boat won’t sink…
Any way did you find any method to provide NIC redundancy (I don’t care about load balancing)? I wonder also about switch fail and network teaming supposed to help in this issue.
|
|
|
|
joe_elway
Honorable But Hopeless Addict
Ireland
6673 Posts
Status: offline |
Posted - 09/16/2008 : 08:55:25 AM
|
quote:
Originally posted by PiotrIr
Any way did you find any method to provide NIC redundancy (I don’t care about load balancing)?
Nope. I'm waiting on the OEM's to provide updated software. There needs to be pressure on them from the likes of us. |
Aidan Finn
MCSE, MVP (Virtual Machine: Systems Administration)
IT Blog: http://www.aidanfinn.com
My Photography: http://www.aidanfinnphoto.com/
My Hyper-V Book: Mastering Hyper-V Deployment
Twitter: http://twitter.com/joe_elway |
|
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/16/2008 : 09:18:03 AM
|
| Please tell me how to press tem so I will do it. Any way it will take probably long time to fix this problem and I need this ASAP:-( |
|
|
|
daveyd
Here To Stay
205 Posts
Status: offline |
Posted - 09/19/2008 : 11:06:46 AM
|
I have a 2 node 2008 Hyper-V host cluster setup with an iSCSI SAN
My equipment:
2x Dell PE2950s (each with 2 embedded Broadcom NICs and each with a 3 Intel NICs)
2x Cisco 3750s for iSCSI network
2x Cisco 3750s for Public network
Dell MD3000i iSCSI SAN
My config:
PE1 [Bcom NIC1] to iSCSI3750[1] port 1
PE1 [Bcom NIC2] to iSCSI3750[2] port 1
PE2 [Bcom NIC1] to iSCSI3750[1] port 2
PE2 [Bcom NIC2] to iSCSI3750[2] port 2
Dell3000i [Controller1, NIC1, Controller2, NIC1] to iSCSI3750[1]
Dell3000i [Controller1, NIC2, Controller2, NIC2] to iSCSI3750[2]
PE1 [Intel NIC1] to Public3750[1] Port1
PE1 [Intel NIC2] to Public3750[2] Port1
PE2 [Intel NIC1] to Public3750[1] Port2
PE2 [Intel NIC2] to Public3750[2] Port2
PE1 and PE2 [Intel NIC3] cluster heartbeat
I haven't had too many issues and am pretty redundant.
|
Edited by - daveyd on 09/19/2008 11:16:01 AM |
|
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/19/2008 : 11:27:45 AM
|
Daveyd it sounds good :-)
Do you use teaming? What will happened when PE1 [Intel NIC1] will be down? Will it switch automatically Hyper-V switch to PE1 [Intel NIC2] or do you need to do this manually?
|
|
|
|
daveyd
Here To Stay
205 Posts
Status: offline |
Posted - 09/25/2008 : 2:04:44 PM
|
I do not use teaming at all.
I did find out that if I yank a cable from the active node PE1 [Intel NIC1] it will not failover. That kind of seems dumb to me cuz if a switch would go down, then the VM would not failover.
I suppose I could create an IP resource and add it to the Hyper-V application group |
|
|
|
PiotrIr
Welcome Newcomer
22 Posts
Status: offline |
Posted - 09/26/2008 : 04:24:52 AM
|
So you have the same problem like everybody :-(
Could you let me know if you will find any solution please?
|
|
|
|
daveyd
Here To Stay
205 Posts
Status: offline |
Posted - 09/26/2008 : 08:07:14 AM
|
Here is a response I got on another forum...
Basically, this is a current limitation of the cluster resource monitor for the Hyper-V resources in cluster. The resource monitor does not check all the important details like network connectivity inside of Hyper-V.
To work around this, you could add an IP address resource to your Hyper-V application group. This way, if a network connectivity issue occurs, this IP address resource would fail causing the VM to move to the other node.
|
|
|
|
Playwell
Honorable But Hopeless Addict
Netherlands
4061 Posts
Status: online |
Posted - 09/26/2008 : 08:19:58 AM
|
| Rediculous if you think about it |
'People who think they know everything are a great annoyance to those of us who do. '
Quote by Isaac Asimov |
|
|
|
daveyd
Here To Stay
205 Posts
Status: offline |
Posted - 09/26/2008 : 2:29:25 PM
|
quote:
Originally posted by Playwell
Rediculous if you think about it
100% agree. Why not have a 100% complete failover solution?? If my switch goes down or a NIC that is tied to a VM goes down, why wouldn't you want a failover to occur???
Pretty dumb. |
|
|
|
drexciya
Seasoned But Casual Onlooker
Netherlands
47 Posts
Status: offline |
Posted - 10/22/2008 : 04:07:08 AM
|
Microsoft should do something about this, if only for fail over. It's ridiculous that they leave it up to third parties whereas VMware does it out of the box.
You might get around this if you would use hardware level functionality (trunking, bonding whatever they call it), especially in blades. |
|
|
|
digg1980
Welcome Newcomer
8 Posts
Status: offline |
|
| |
Topic |
|
Hyper-V and real fail over
