Host-based backup of VMware vSphere VMs.
Post Reply
michael-fuller
Lurker
Posts: 2
Liked: never
Joined: Oct 28, 2015 9:10 pm
Full Name: Michael S
Contact:

Disaster recovery best practices

Post by michael-fuller »

We are in the process of creating our internal run book for how to recover our system using Veeam backups/replicas and were hoping you would have some kind of guide to give us a head start. Do you have a best practices guide for how to recover a network from backups/replicas in the event of a disaster?
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:

Re: Disaster recovery best practices

Post by Shestakov »

Hello Michael and welcome to the forums!
Recovery best practices depend on your needs. There are 47 restore scenarios for different purposes.
If you specify your infrastructure configuration, it can help us give a good advice.
Thanks
Vitaliy S.
VP, Product Management
Posts: 27055
Liked: 2710 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:

Re: Disaster recovery best practices

Post by Vitaliy S. »

Also please let us know about the desired RTO and RPO periods.
michael-fuller
Lurker
Posts: 2
Liked: never
Joined: Oct 28, 2015 9:10 pm
Full Name: Michael S
Contact:

Re: Disaster recovery best practices

Post by michael-fuller »

Thank you for the replies.

Via Veeam we are backing up all our VMs onto a NAS device. We also replicate all VMs to a separate subnet at a remote site. This includes our DC, email server, file servers, application servers, webservers, Sharepoint servers etc.

We replicate our Veeam backup server as well although we do not have a computer with Veeam installed at our remote site. We do have Vsphere installed at our remote site.

We are wondering what is the best setup in the event our main network goes down. What is the best practice to failover completely using replicas at our remote site or from our NAS backups? What is the best practice to eventually failback to our main network?

RPO would be the time of the last replicas or backup. Our critical servers are replicated every 2 hours and backed up once a day. The RTO would of course preferably be ASAP, but what would be a realistic timeframe to expect?
Shestakov
Veteran
Posts: 7328
Liked: 781 times
Joined: May 21, 2014 11:03 am
Full Name: Nikita Shestakov
Location: Prague
Contact:

Re: Disaster recovery best practices

Post by Shestakov »

One of the best backup+replica scenarios is backup+backup copy+replica from backup files. Doing that you take a snapshot of the source VM only once and transfer data between sites only once, at the same time you will meet 3-2-1 rule, having backup+copy of the backup offsite+replica VM.

Regarding restore, in a case your source site shut down, you make Failover to the replicated VM. Then you have a choice either to perform a permanent failover and make replica VM a production one or failback to the source site. It will let you have minimal RTO. In a case replica is not usable, like BSOD, hack etc. you can restore from backup. That`s why it`s recommended to have VBR server in the DR site, to be protected in a case of disaster in the production site.
Thanks!
Vitaliy S.
VP, Product Management
Posts: 27055
Liked: 2710 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:

Re: Disaster recovery best practices

Post by Vitaliy S. »

Just a clarification - in your case you will need to spin up the replicated Veeam backup server manually, and then do all failover operations via Veeam backup console. Also there are existing topics with the same questions, for example, check out this one > Veeam Replication Best Practice and Veeam location.
Post Reply

Who is online

Users browsing this forum: No registered users and 70 guests