SBS Server 2008 recovery - Redundant server

Hi guys

Please can you shed some light for me on a current situation I have.

I have a client in a remote African country where support is very limited.

I have the following environment :

1 SBS server 2008 premium running exchange mail 2007 and about 150GB's total user and data shares being backed up daily using Backup Exec 2010 with a system recovery snapshot every Sunday. We make a daily backup at 12 pm mindnight which runs for approximately 12 hours including verification.

I have installed a secondary server which does DNS etc but this has proven not to be any sort of failover to rely on as it doesnt have exchange and is also server standard.

My question is how can one have another fully redundant backup SBS server, running as a full blown copy of my SBS 2008 server so that in the event of server crash we can immediately swop over to the second server. Is this possible? I am concerned that if my backup takes so long every day how long will recovery to otehr hardware take should it crash.

Has anyone ever ventured into this type of backup scenario?

Many thanks
DeonLubbeAsked:
Who is Participating?
 
Philip ElderConnect With a Mentor Technical Architect - HA/Compute/StorageCommented:
For "Cold Backup" situations where an identical box with the OS in the off state is standing ready to take over one requires Open License.

Perhaps a third party sanitation and continuity provider (we use ExchangeDefender) would be a better option? It's very inexpensive relative to what is offered.

+ Offline access to legitimate e-mail via the provider's OWA.
+ Service holds onto e-mail until the SBS is back up and running then delivers to SBS.
+ Service santizes e-mail of any rogue attachments _before_ they enter the network.

There is a lot more to it than that, but having a service like that in place makes the redundancy question a little more palatable cost wise for the business since e-mail is probably the most important aspect of running a business today.

Philip
0
 
Chris MillardCommented:
I don't think that this is possible with SBS because you can only have one domain controller.
0
 
DeonLubbeAuthor Commented:
Isnt there someway that I can have another server, running say Server 2008 with Hyper V on it or some sort of virtualization so that if my SBS goes down, we can take a snapshot (or virtual SBS 2008 copy) and run it to take over? How would one go about to do this?
0
Simplify Active Directory Administration

Administration of Active Directory does not have to be hard.  Too often what should be a simple task is made more difficult than it needs to be.The solution?  Hyena from SystemTools Software.  With ease-of-use as well as powerful importing and bulk updating capabilities.

 
Philip ElderConnect With a Mentor Technical Architect - HA/Compute/StorageCommented:
Another option that would use both servers:

2x 1U or 2U servers (or pedestal) with dual external SAS connectors.
1x Direct Attached Storage with redundant controllers and intelligent internal RAID, LUN, LUN Masking, and Controller Affinity management.

Connect the two servers to the DAS via 2x independent SAS connections per server per controller on the DAS.

Install Hyper-V Server 2008 R2 + SP1 on both servers.
Configure MPIO software (server's MPIO or DAS MPIO depending on make/model).
Configure storage on DAS with needed LUNs + Map to Server SAS IDs.
Configure networking on both servers (5 NICs preferred. 1 Management and 2 teams).
   Independent dual port virtualization acceleration NICs (4 ports total).
Configure shared storage on NODE 1 and set offline.
Configure shared storage on NODE 2 by hitting refresh in Disk Mgmt & set Offline.
Run the Cluster Configuration Wizard to test the configuration.
When successful use the wizard to stand the cluster up.

You can then have your SBS 2008 and any other VMs running on the Hyper-V cluster.

If one node fails things keep humming along while the node is repaired.

Philip
0
 
Cliff GaliherCommented:
You could change your backup/restore software. Options like DoubleTake, ShadowProtect, or a HeroWare device would get you closer to your goal. The first two would still be cold backups, but restore quickly and can even be started as VMs in a pinch for near immediate uptime. The latter causes more replication traffic, but has a more automated fail-over story.

None of the above replaces the need for off-site backups though. These would be part of a larger DR plan.

Alternatively, you could set up a virtualization cluster (2+ node Hyper-V cluster, for example) connected to centralized storage such as a small SAN. You could then have true HA should a single node fail. Again, not a replacement for DR, but provides more fault tolerance for local failures.

-Cliff
0
 
DeonLubbeConnect With a Mentor Author Commented:
Guys thanks so much for the awesome input.

Surely there must be some basic way of using the other server and running VMware or Similiar, then captuing a virtual snapshot of the domain controller once a week when the plant is shut down, say Sunday, and run this on the second server.

Then once the primary server goes down, you would log onto the Secondary domain controller, start the virtual server and have a restored network, well at least from the previous Sunday.

Then from Backup Exec restore the Exchange database and user files?

Please excuse me if I seem ignorant here, I really need some advice. I live in another country and I am trying to make sure I have something to fall back onto remotely as the skills here are so limited I cannot afford to rely on local resources to bring the site back up.

I am thinking the more virtual servers as backups I can run the better.

Is this realistic approach?
0
 
Cliff GaliherCommented:
Snapshotting a domain controller is always a bad idea. Snapshotting your ONLY DC is even worse. Don't do it.

-Cliff
0
 
Philip ElderTechnical Architect - HA/Compute/StorageCommented:
You can close the Q by selecting the cited answers. :)

Thanks,

Philip
0
 
DeonLubbeAuthor Commented:
Thanks for this input guys, I am looking into Double take or similiar software. I'd hate to say but Im not skilled enough yet to install the clustering etc.

I was hoping to find a simple way of recovering from disaster such as the server motherboard dying etc.

I have already spent sooo much on backup exec, remote agents, exchange agents, etc etc. It just doesnt seem practical to restore quickly with BE. The daily backups and veriifications take almost 8 hours. 8 Hours to restore a server and still deal with driver issues etc is a disaster to this organization. Surely there must be an easier way to have a backup server?
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.