Step-By-Step: Azure Site Recovery – Failover to Azure


Hello folks,

My last couple of posts were about protecting your physical servers with Azure Site Recovery or ASR for short.

I found this video for those of us who prefer listening to a video than reading a post to learn about ASR.

Today we will look at how we failover to Azure for Machines that are already protected.

you can’t just failover a machine just like that.  Like anything else you need to plan and configure how the cut over will happen.  You need to create a plan that includes one or more groups that contain that protected virtual machines. These machines will fail over depending on the group they’re in. Virtual machines in a particular group fail over in parallel.

To plan our failover we must define groups of machines that fail over and start up together. For example if we need to fail over a specific workload, we would group the virtual machines for that workload in the same recovery plan group. then we automate and extend failover.

Recovery plans are displayed on the Recovery Plans in the Site Recovery portal.

Now that we know what is our next step.  let’s do it…

Create recovery plans

1- On the Recovery Plans tab, click Create Recovery Plan.

image

2- In the Name field we specify a name for the recovery plan. Since we are replicating from a physical on-premises server to Azure, we will select our configuration server as the source (CANITPRODRPCFG) and Azure as the target and click the right pointing arrow.

image

3- In Select virtual machines select the virtual machines (or replication group) that you want to add to the default group – Homelab (1) in the recovery plan.

image

Run a failover

You can run different types of failovers recovery plan.

Failover

When to run

Details

Process

Test failover

Run to validate your replication strategy or perform a disaster recovery drill

No data loss or downtime

No impact on replication

No impact on your production environment

Start the failover

Specify how test machines will be connected to networks after failover

Track progress on the Jobs tab. Test machines are created and start in the secondary location

Azure – connect to the machine in the Azure portal

Secondary site – access the machine on the same host and cloud

Complete testing and automatically clean up test failover settings.

Planned failover

Run to meet compliance requirements

Run for planned maintenance

Run to fail over data to keep workloads running for known outages – such as an expected power failure or severe weather reports

Run to failback after failover from primary to secondary

No data loss

Downtime is incurred during the time it takes to shut down the virtual machine on the primary and bring it up on the secondary location.

Virtual machines are impact as target machines becomes source machines after failover.

Start the failover

Track progress on the Jobs tab. Source machines are shut down

Replica machines start in the secondary location

Azure – connect to the replica machine in the Azure portal

Secondary site – access the machine on the same host and in the same cloud

Commit the failover

Unplanned failover

Run this type of failover reactive manner when a primary site becomes inaccessible because of an unexpected incident, such as a power outage or virus attack

You can run an unplanned failover can be done even if primary site isn't available.

Data loss dependent on replication frequency settings

Data will be up-to-date in accordance with the last time it was synchronized

Start the failover

Track progress on the Jobs tab. Optionally try to shut down virtual machines and synchronize latest data

Replica machines start in the secondary location

Azure – connect to the replica machine in the Azure portal

Secondary site access the machine on the same host and in the same cloud

Commit the failover

Source: https://azure.microsoft.com/en-us/documentation/articles/site-recovery-failover/

Failover from On-Prem to Azure

1- to initiate a failover we need to Select Recovery Plans from the Azure portal, select the proper recovery Plan. Click Failover and Test Failover.

image

2- On the Confirm Test Failover page, specify how replica machines will be connected to an Azure network after failover.  And which recovery point you want to use

image

We can track failover progress on the Jobs tab. once it completes, You should be able to see the test replica machine in the Azure portal.

image

You can access replica machines in Azure using an RDP connection from the Virtual Machine tab in the azure portal. Remember, port 3389 will need to be open on the endpoint for the virtual machine.

image

That’s it!  we have now failed over to Azure.  Keep in mind that there are other scenarios that may come in play. Scenarios like keeping the same IP address on the failed over machine. By default a failed over machine will have a different IP address than the source machine.  We will look at that scenario in another post.

Cheers!

Signature

Pierre Roman
@pierreroman

Comments (0)

Skip to main content