Comprehensive data protection for all workloads
Post Reply
jeremipdx
Lurker
Posts: 2
Liked: never
Joined: Jun 28, 2012 5:05 pm
Full Name: Jeremi Green
Contact:

Failing over to replicated site

Post by jeremipdx »

Hey everyone,

I'm a Veeam newbie and just started backing up and replicating our production environments with Veeam. So far so good, but one thing I'm having trouble understanding is the process of failing over to our DR site in a disaster situation. All the documentation says you need to initiate a fail over from the Veeam application. I figure this is so tit can keep track of all the changes on the replicated VM side and make it easy to fail back without losing data. However, if your primary data center is leveled by an earthquake you aren't going to have access to your Veeam server. Is it okay to just power on your VM's at your DR site? I assumed it would be, but there was a sentence in the deployment guide that said "If possible, avoid powering on a replica manually as it may disrupt further replication operations or cause loss of important data. It is strongly recommended to use Veeam Backup & Replication functionality to perform fail over operations." That doesn't make me feel very warm and fuzzy about my DR solution :shock:

I spoke with Miguel in support who advised you have to have access to the Veeam server, which doesn't sound right. He advised placing the server in our DR site but It does all our production backups as well so that seemed kind of odd. Any real world experience and suggestions are appreciated!
Gostev
Chief Product Officer
Posts: 31457
Liked: 6648 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

Re: Failing over to replicated site

Post by Gostev »

Hi Jeremi, did you read the sticky FAQ topic before posting? One of the questions there answer what specifically do you lose in case of manual failover. By the way, if your Veeam server is a VM, consider simply replicating it to your DR site along with other VMs. Thanks!
jeremipdx
Lurker
Posts: 2
Liked: never
Joined: Jun 28, 2012 5:05 pm
Full Name: Jeremi Green
Contact:

Re: Failing over to replicated site

Post by jeremipdx »

Thanks for the reply Gostev. I did read through the sticky, I must have missed it. I'll go back right now. Our veeam server is a VM, I'll explore that option. Thanks!
jeremyh8
Enthusiast
Posts: 81
Liked: 11 times
Joined: Jun 17, 2012 1:28 am
Full Name: Jeremy Harrison
Contact:

Re: Failing over to replicated site

Post by jeremyh8 »

to do replication you should have a backup server at your remote site correct? simply create the replication jobs on that backup server (instead of prod) and have it pull the data from your prod site. This way if you have a disaster your replication server is already there and waiting for you to click failover.
Gostev
Chief Product Officer
Posts: 31457
Liked: 6648 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

Re: Failing over to replicated site

Post by Gostev »

jeremyh8 wrote:to do replication you should have a backup server at your remote site correct?
No, actually this is not necessary. All you need at the remote site is backup proxy.
jeremyh8
Enthusiast
Posts: 81
Liked: 11 times
Joined: Jun 17, 2012 1:28 am
Full Name: Jeremy Harrison
Contact:

Re: Failing over to replicated site

Post by jeremyh8 »

then simply install the backup server software on the machine acting as your proxy and use it for replicaiton :)
ZachW
Enthusiast
Posts: 68
Liked: 10 times
Joined: Aug 02, 2011 6:09 pm
Full Name: Zach Weed
Contact:

Re: Failing over to replicated site

Post by ZachW »

As long as you are using version 6.0+ then you will have Veeam B&R installed on a server at the source site either acting as a proxy ( by default it is a proxy listed as "Default VMware proxy" ) or another server at the source site acting as a proxy [this just means it has been specified in the "backup proxies pane" Veeam is not installed on any server other than the "Veeam" server]. Keep in mind, there isn't much manual intervention that needs to come into play when you are designating a server as a proxy as all work can be done from the Veeam server it's self. Also keep in mind that Veeam does not have to be installed on the server for it to act as a proxy ( see the User Guide for more information ). You will follow the same steps for the remote server and designating it as a proxy.

It's really quite simple and there are much more detailed instructions in the User Guide. If you still have any questions after giving it a look then feel free to check back with us.
johnlong
Enthusiast
Posts: 74
Liked: 11 times
Joined: May 12, 2012 3:37 pm
Contact:

Re: Failing over to replicated site

Post by johnlong »

I'm in the same position as the OP.

"to do replication you should have a backup server at your remote site correct? simply create the replication jobs on that backup server (instead of prod) and have it pull the data from your prod site. This way if you have a disaster your replication server is already there and waiting for you to click failover."

This really sounds like a logical approach. If you have your B&R server in the production site and a proxy at the DR site and the production site is offline you will have to install a B&R server in the DR site to perform any operations on the replicated VM's, except for just turning them on. The reverse is also true. If you loose your DR site you will need a B&R server in the prod site to perform any backup or failover operations.

Is there any harm in having 2 B&R servers with separate jobs?
vMO
Expert
Posts: 176
Liked: 21 times
Joined: Jul 08, 2011 8:16 am
Contact:

Re: Failing over to replicated site

Post by vMO »

no, but there is a more easy way..
you have the proxy on both sites, disaster and production, simply replicate the backup server to your dr site, in case of disaster, just power on backup server manualy and then initiate the failover for all your important vm's.
johnlong
Enthusiast
Posts: 74
Liked: 11 times
Joined: May 12, 2012 3:37 pm
Contact:

Re: Failing over to replicated site

Post by johnlong »

I have a physical B&R server now because it's an ideal setup with the resources I have. I use direct san access to keep the proxy on that physical box. Offsite I only have 2 servers, 1 physical and 1 ESXi 5. On the physical box I can install B&R but it's not an option to run it on the ESXi host because of RAM constraints.

That is why I would like to run 2 B&R servers with separate jobs protecting the same VM's.
foggy
Veeam Software
Posts: 21069
Liked: 2115 times
Joined: Jul 11, 2011 10:22 am
Full Name: Alexander Fogelson
Contact:

Re: Failing over to replicated site

Post by foggy »

johnlong wrote:Is there any harm in having 2 B&R servers with separate jobs?
Just a note (in case backup proxy servers and/or backup repositories are shared between different Veeam backup servers in your setup): the number of concurrent tasks per proxy and repository and maximum data ingestion rate per repository are counted within each backup server independently.
johnlong
Enthusiast
Posts: 74
Liked: 11 times
Joined: May 12, 2012 3:37 pm
Contact:

Re: Failing over to replicated site

Post by johnlong »

I will keep that in mind. I wonder if 2 jobs on different B&R servers start at the same time will one wait for the other to finish or will they both run at the same time?
foggy
Veeam Software
Posts: 21069
Liked: 2115 times
Joined: Jul 11, 2011 10:22 am
Full Name: Alexander Fogelson
Contact:

Re: Failing over to replicated site

Post by foggy »

Yes, the job waits for another one to finish before starting to process the same VM.
Post Reply

Who is online

Users browsing this forum: Google [Bot], Ivan_B and 167 guests