3 server setup

[inasr]

I have 3 servers: Primary, secondary, and backup.
On the secondary we are replicating the primary, and on the backup, we are creating a backup copy.
When the primary goes down, the secondary should directly bootup and take over which is happening, but all its services are appearing down.
If the backup is turned on to take over, no issues are faced.

What is the main difference between the process of replication and back up?
What are the possible problems?

[david.domask]

Hi inasr, welcome to the forums.

I'm afraid I'm not quite understanding the situation here. Can you elaborate a bit more on the servers you're talking about? These are 3 separate Veeam Backup and Replication servers or these are servers you're protecting with Veeam?

A diagram would be appreciated if possible, as unfortunately a bit difficult to understand the situation based on your description.

If you're getting a technical issue however within the Veeam Backup and Replication software, open a Support case as is noted as a requirement when creating a topic.

[inasr]

Hello David,
Yes they are 3 different Dell servers.
To make the downtime shorter, a server is used as replica site that would monitor any changes. And to keep a fixed point of return to lessen data loss, another server is used as backup site.
Regarding the setup, I am not exactly sure how everything is connected as I currently do not have the topology map.

[david.domask]

Hi imasr,

Just to be clear, are we discussing:

- Production servers you're protecting with Veeam Backup and Replication? (e.g. you have a backup of a server, a replica, and a backup copy?)
- Several Veeam Backup Servers? (Site A VBR, Site B VBR, Site C VBR?)

Again, regrettably your inquiry is unclear so not sure how to guide you

[inasr]

Apologies for the delay and the unclearness, but I am a bit new to this.
They are several Dell servers. Veeam is pointed to primary to replicate to secondary that is also pointed to, and backs up to tertiary that is also pointed to.

[david.domask]

Think I'm starting to get the picture, but regrettably still unclear, and I think it would likely benefit from partner engagement at this stage to review the environment; it sounds like a deeper review of the configuration is required to best meet your needs

But I think I can at least answer one of your questions above:

> What is the main difference between the process of replication and back up?

Replication creates a 1:1 replica of a workload on a secondary site, so there is an actual VM on the secondary site ready to be booted and take over the production workload when activated.

A backup is a a point in time copy of the workload stored in a portable format that can be restored from or restore to production.

I unfortunately cannot comment on potential issues as the situation is not clear, and I recommend engaging a partner as noted above first to review the environment and setup, and if there are technical issues during the backup and/or replication operations, open a Support case and allow Veeam Support to review the technical issues.

[JeroenL]

I would design it like this.

Install Veeam Backup & Replicion on the third server. This will be your Backup & Replication Orchestration server. I would opt for a 4th server or cloud storage, but I'll come to that next.

The first server will be your primary production server. I don't know anything about physical locations, but this can be in location A.
The second server can be your failover server. I don't know what kind of disaster scenario's you are protection against, but this van be in location B.

If you are planning for protection against physical disaster like fire, flooding, severe storms or theft of the server I would place the failover server in another location. If it is a production location that cannot function when the building is gone it mostly don't help anything to bring the IT online in another location in a few minutes. In this case I would balance load over both servers and be sure to keep at least 50% resources free on both servers. This way you make best use of the available hardware and gaining performance profits until one of the servers fails.
The third server can be in the same location too. This way your restores in case of hardware failure or ransomeware can be performed with maximum performance. Add a fourth server in another location or use cloud storage and create an immutable backup copy on that server. That way at least one copy of your backup storage is untouchable by disaster or malicous actors. Best is to use the 32110 rule explained on many places.

Again. We don't know anything about the company or the environment you are trying to protect. What are the RPO/RTO requirements of the various IT systems you are running? Try to identify the company IT crown-jewels. What is most important and how much data can the company accept to loose? Is this 1 week, 1 day, 1 hour of maybe even 1 second? Next is how much time is there to perform a recovery? Is 1 week after a server failure acceptable of do you have 15minutes to have everything back up and running? These parameters also determine, in some ways, how to design your production environment. If you cannot affort downtime, you might design for high-availablity with replicating storage and multiple servers. What "disasters" are most likely to occur that need protection against? Are you located in a location that has frequent tornado's grounding buildings or floodings filling up buildings with meters of water? All such things determene how you should design your production, but also you DR enviroments.