Veeam Backup And Replication Physical Server

[Jamie Swearengin]

Is it possible to restore (Backup using Veeam Backup and replication) the entire system state backup to a physical server, if the system is crashed. I am having the having the incremental and full back up files and can restore as a VM. But any possible ways to restore to a physical machine.

Moving from provider to another or hypervisor to another should be fairly straightforward and any IT person should be able to explain to their boss/stakeholder etc. why going back to physical has many pitfalls.

As a side note, yes you probably can do this, however it wont contain the physical drivers the hardware needs, so it may restore with no network, may fail to boot due to missing RAID drivers, may crash after booting due to incorrect drivers and so on.

It would be better to build a new machine and restore the data. If, before you say it, the system holds an application that would be prohibitively expensive to re-install, then we go back to point one - why put it back on physical.

I am having a dell poweredge R720 server with windows server 2019 OS and the hardware is faulty , due to this the server is not booting up.
So i am having another dell poweredge R740 server and if possible i want to restore the same os in that hardware.

To recover the entire computer system on new or existing hardware, you can use bare metal recovery. Bare metal recovery can be helpful in the following cases: You want to recover your computer from scratch...

A better option would be installing a hypervisor that will care about RAID, drivers, and hardware compatibilities and restore your backup to a virtual machine, especially since Veeam B&R supports that and does that perfectly in almost all the cases, so you will surely succeed in restoring your workload. You can use Hyper-V Hyper-V Server 2019 Microsoft Evaluation Center or VMware -free-buy-esxi-anyway/ for that purpose. Both are free (with some limitations).

I wanted to ask the community on here to get a discussion going about whether using VMs for Veeam is good, bad or indifferent? It is based on a question from the forum here - VM as repo - Veeam R&D Forums

I added my thoughts to this post as I disagree with some of the comments about having to use physical servers as the BP site mentions both physical and virtual. I also work at an MSP and everything we build is virtual for Veeam other than a few tape servers we have which are physical boxes since they perform better this way with direct connect to the FC fabric.

In general, we recommend whenever possible to use physical machines as repositories, in order to maximize performance and have a clear separation between the production environment that needs to be protected and the backup storage. It is also recommended combining this with the proxy role if backup from Storage Snapshots is possible. This keeps overheads on the virtual environment and network to a minimum.

Great points Joe. Now in our case the Veeam virtual infrastructure is separate from anything client related as we have our own management stack and cluster that is used whereas the clients use vCloud and have clusters allocated there. So we do ensure security between things.

Everyone decides for themselves how much risk they need to mitigate to get a good night sleep, and I see physical servers as a great step when done right, and if virtual, only in an isolated platform.

Especially in smaller environments I like to use a dedicated physical server as it makes both securing and restoring much easier. In bigger environments with different locations, it's can make sense to virtualize the VBR server role; especially if you replicate it or maintain a cold standby instance.

Most of my deploys are 2 to 4 ESXi hosts, and most of the time a Prod Cluster and Dev Cluster, and my Veeam Server Replicated (Veeam Replicas) between both Clusters, so in case of disaster, I can run my Veeam Server, and then run Replicas from it, of restore what I need.
;)

For my clients with physical servers, that means I have a Dell server in place with a battery-backed RAID card, enterprise-grade local drives with enterprise support agreements when a drive fails rather than having to figure out what the current model drive most closely matches the failed drive or figuring out if there is still a warranty from the drive manufacturer, and then finding a replacement while being down a drive for a week or more. For those physical servers, there is the caveat of having some sort of on-host proxy server for VM access, or, my preference, using direct storage access via ISCSI or SAS/Direct Attach connections depending on what the primary storage is. Makes for some really fast backups. Also, if I have direct storage access, it makes it easier to segment the backup environment from the production environment.

This allows for quick recovery at the secondary site with the backup data already mirrored to the recovery storage array. One of the great things about Veeam is that it is possible to recover from a backup with just the Veeam B&R Console by importing the backup data.

The other option is to have 2 Veeam instances running; one at the Primary and one at the Secondary site. So in case of failover, the data can get backed up when VMs are brought back online at the Secondary site. One less thing to worry about during a DR scenario. The secondary instance could also be connected to an Object Storage provider to recover from cloud, as required.

@dloseke & @dips on your points around SMB based setups, I had a call with Object First last week, and I honestly see that their platform could become a massive game changer for SMB/ROBO deployments especially.

@dloseke& @dips on your points around SMB based setups, I had a call with Object First last week, and I honestly see that their platform could become a massive game changer for SMB/ROBO deployments especially.

Yeah I attended the webinar a few weeks ago. Really want to look in to this as we are moving off Windows to Linux servers for the Immutability but would be great to have a cluster of these to use in specific use cases at our DCs or clients. Looking forward to the GA release. ?

Well put, once physical servers reach capacity i.e RAM for example, it is hard to allocate more resources to it which means budgeting for a new physical server. On the virtual side of things, if there is some spare capacity available, it can easily be allocated.

I Have been learning about veeam products and recently for practice, I setup a windows server VM & installed veeam backup & replication in that. I was trying to backup a physical machine and the backup job failed. I feel the reason is veeam is not able to identify details such as hostname and credentials. I looked various documents related to this but still I am missing out something. I even installed veeam agent on physical machine during initial setup. Can anyone help me out with this?

@Anandu , can you tell us a bit more about the setup? You deployed the agent using VBR? You have setup the agent job on the physical machine or the VBR server? Have you created the DNS A and PTR record for all components and the physical server?

What do you mean by : how can I check that? That the servers are domain-joined or that the credentials you are using are having local administrator permissions? I suggest to try at first the credentials of the local administrator account of the physical machine.

Finally, do you have an internal DNS server (via a domain or otherwise), if yes, your devices should have A/CNAME records as @Nico Losschaert has mentioned above, failing this you COULD use the hosts file within Windows to statically create records of the devices.

The best way to start this is to explain what you have in your test lab, how many machines, infrastructure layout, networking, domain situation and relationships to machines or anything that could influence the interaction between machines. Then state the error you see on backup failure (from the screen) and the point of failure when testing credentials. this will help in a big way.

My VBR is not domain joined and I am talking about credentials having local administrator permissions. i read in article which u showed saying that it have permissions in target machine. Should I give the permissions in the target?

@Anandu , at first I should suggest doing first following 2 things : use the ip-address of the target server, disable the Windows firewall on the target-machine and use the built-in local administrator credentials and try what this gives as a result

I still not able to perform backup. My VBR backup server was setup using virtualbox. Can u tell me on the network settings, what kind of an adapter(i.e, NAT/Bridged/internal/generic driver/host-only adapter) should choose so that I could integrate with the physical machine and perform backup?

There is no need to deploy Veeam Agent before using this new feature. If you have already deployed Veeam Agent for Windows / Linux on several servers, you can use them in Veeam Backup & Replication without uninstalling.

Then configure the discovery schedule and the installation/update of the agent automatically. For example, if you have chosen an organizational unit as an included object when you add the computer in this OU, Veeam is able to discover and install agent automatically on this machine.

c80f0f1006