WEBVTT

0:00:03.300000 --> 0:00:06.940000
 Within Azure, there's really two different
 ways to protect your virtual

0:00:06.940000 --> 0:00:10.480000
 machine. One is backup and that's
 covered in other videos.

0:00:10.480000 --> 0:00:14.540000
 The other is through recovery,
 replication and recovery.

0:00:14.540000 --> 0:00:17.800000
 So really this becomes an availability
 solution and that's what we're

0:00:17.800000 --> 0:00:19.780000
 looking at in this video.

0:00:19.780000 --> 0:00:23.400000
 We're going to take a look at virtual
 machine recovery with Azure Site

0:00:23.400000 --> 0:00:26.900000
 Recovery or Recovery Services.

0:00:26.900000 --> 0:00:28.980000
 Virtual Machine Recovery with ASR.


0:00:28.980000 --> 0:00:31.220000
 We're going to look
 at recovery options.

0:00:31.220000 --> 0:00:33.120000
 So what are the different
 ways we can recover?

0:00:33.120000 --> 0:00:35.840000
 We're going to look at the virtual
 machine recovery requirements.

0:00:35.840000 --> 0:00:36.840000
 What do you need?

0:00:36.840000 --> 0:00:38.780000
 What kind of infrastructure
 do you need?

0:00:38.780000 --> 0:00:44.100000
 And then I'm going to do a quick demonstration
 of setting up Azure Virtual

0:00:44.100000 --> 0:00:45.040000
 Machine Recovery.

0:00:45.040000 --> 0:00:48.460000
 What you need to set up in
 the virtual machine itself.

0:00:48.460000 --> 0:00:56.580000
 And with that, let's go ahead and let's
 start looking at Site Recovery.

0:00:56.580000 --> 0:01:01.400000
 Now within Azure Site Recovery, there
 are a few different options.

0:01:01.400000 --> 0:01:05.080000
 So this Recovery Services Vault and
 I use those in this case a little

0:01:05.080000 --> 0:01:06.760000
 bit interchangeably.

0:01:06.760000 --> 0:01:14.140000
 I probably shouldn't because the Recovery
 Services Vault actually supports

0:01:14.140000 --> 0:01:20.900000
 two operations. I could
 spell correctly.

0:01:20.900000 --> 0:01:24.740000
 It supports both backup
 and site recovery.

0:01:24.740000 --> 0:01:28.720000
 So it's really Azure Site Recovery
 or Recovery Services Vault.

0:01:28.720000 --> 0:01:31.480000
 Now there's two options
 for recovery.

0:01:31.480000 --> 0:01:35.300000
 Backup is supported by Recovery Services
 Vault, but that's also covered

0:01:35.300000 --> 0:01:36.020000
 in other videos.

0:01:36.020000 --> 0:01:39.300000
 We're really just looking at
 the virtual machine recovery.

0:01:39.300000 --> 0:01:43.580000
 But before that, I want to
 look at the bigger picture.

0:01:43.580000 --> 0:01:47.300000
 One scenario, and it's actually multiple
 scenarios, it's supported by

0:01:47.300000 --> 0:01:51.020000
 Recovery Services Vault
 is on premises.

0:01:51.020000 --> 0:01:56.340000
 I can actually use this to protect my
 non-Azure workloads and those can

0:01:56.340000 --> 0:01:58.660000
 be running in Hyper-V.

0:01:58.660000 --> 0:02:02.800000
 They can run in VMware.

0:02:02.800000 --> 0:02:07.060000
 They can even run directly
 on physical servers.

0:02:07.060000 --> 0:02:10.740000
 So I can, if I could spell,
 having some problems there.

0:02:10.740000 --> 0:02:17.220000
 There we go. I can actually take my
 on-premises workloads and replicate

0:02:17.220000 --> 0:02:25.300000
 them into Recovery Services Vault where
 they can then be recovered into

0:02:25.300000 --> 0:02:31.000000
 Azure. In fact, they could even be recovered
 back to another on-premises

0:02:31.000000 --> 0:02:33.080000
 location under certain
 circumstances.

0:02:33.080000 --> 0:02:35.280000
 But that's not really what
 we're focusing on.

0:02:35.280000 --> 0:02:40.860000
 What we're really focusing on is the
 fact that I can take an Azure virtual

0:02:40.860000 --> 0:02:44.440000
 machine. That's in one region.

0:02:44.440000 --> 0:02:47.480000
 Let's say it's in the East US.

0:02:47.480000 --> 0:02:54.340000
 And I can configure that to replicate
 through Recovery Services Vault

0:02:54.340000 --> 0:03:01.020000
 so that if necessary, I could restore
 that virtual machine somewhere else.

0:03:01.020000 --> 0:03:08.240000
 So for example, I could restore that
 same VM and we'll just say West US.

0:03:08.240000 --> 0:03:14.160000
 And the idea is to make it as simple as
 possible to facilitate that recovery

0:03:14.160000 --> 0:03:16.140000
 capability and failover.

0:03:16.140000 --> 0:03:21.480000
 Now I will tell you there's no automatic
 failover, but the manual failover

0:03:21.480000 --> 0:03:27.360000
 can be as simple as a click of a button
 or the execution of an Azure CLI

0:03:27.360000 --> 0:03:29.760000
 command or a PowerShell
 commandlet.

0:03:29.760000 --> 0:03:33.540000
 Or even if you wanted to go very hardcore,
 you could of course do it through

0:03:33.540000 --> 0:03:35.180000
 the REST API as well.

0:03:35.180000 --> 0:03:37.080000
 But big picture, that's the idea.

0:03:37.080000 --> 0:03:41.880000
 This Recovery Services Vault is really
 supplying all of the glue that

0:03:41.880000 --> 0:03:45.780000
 is necessary for virtual
 machine recovery.

0:03:45.780000 --> 0:03:50.920000
 Now let's go ahead and take a look at
 what is required for virtual machine

0:03:50.920000 --> 0:03:57.060000
 recovery. For virtual machine recovery,
 here's what you need.

0:03:57.060000 --> 0:04:02.880000
 You're going to have of
 course your source.

0:04:02.880000 --> 0:04:11.960000
 So I've got the source and I've got
 a virtual machine, this is a VM, and

0:04:11.960000 --> 0:04:15.620000
 that virtual machine of course is
 running in a virtual network.

0:04:15.620000 --> 0:04:20.040000
 So we'll say VNet.

0:04:20.040000 --> 0:04:24.460000
 And this is a source.

0:04:24.460000 --> 0:04:27.800000
 Now just so you know, there's absolutely
 no reason that I switched back

0:04:27.800000 --> 0:04:30.340000
 and forth between handwritten
 and typing.

0:04:30.340000 --> 0:04:33.920000
 But just in case you thought
 that was important, it's not.

0:04:33.920000 --> 0:04:38.180000
 Now in addition to that, if I'm going
 to have this recover, I'm going

0:04:38.180000 --> 0:04:42.460000
 to need some kind of destination.

0:04:42.460000 --> 0:04:49.700000
 And that destination is also going
 to need a virtual network, I might

0:04:49.700000 --> 0:04:53.420000
 as well at least be consistent
 and write that out.

0:04:53.420000 --> 0:04:56.540000
 So I've got the source,
 I've got a destination.

0:04:56.540000 --> 0:05:04.460000
 Now facilitating this whole process, as
 you'd expect, is a recovery services

0:05:04.460000 --> 0:05:09.200000
 vault. Let's put volts in there.

0:05:09.200000 --> 0:05:15.560000
 Now in addition to the vault, a couple
 things to note, there is going

0:05:15.560000 --> 0:05:20.980000
 to be a storage account that
 is used for caching.

0:05:20.980000 --> 0:05:30.800000
 So we've got a caching
 storage account.

0:05:30.800000 --> 0:05:35.520000
 And so what's going to happen is this
 virtual machine is going to start

0:05:35.520000 --> 0:05:37.020000
 replicating changes.

0:05:37.020000 --> 0:05:41.720000
 And those changes are going to go into
 the caching storage account, they're

0:05:41.720000 --> 0:05:45.560000
 going to go through the vault, and
 then they're going to be replicated

0:05:45.560000 --> 0:05:47.280000
 over to the destination.

0:05:47.280000 --> 0:05:50.760000
 Now not into the VNet and there
 won't be a virtual machine yet.

0:05:50.760000 --> 0:05:54.240000
 Typically you're going to use managed
 disks, so behind the scenes it's

0:05:54.240000 --> 0:05:56.020000
 generating managed disks.

0:05:56.020000 --> 0:05:59.400000
 You can do this with unmanaged disks,
 but it's not recommended, I won't

0:05:59.400000 --> 0:06:00.840000
 even cover that.

0:06:00.840000 --> 0:06:05.920000
 One thing with this, why are we going
 through the caching storage account?

0:06:05.920000 --> 0:06:08.280000
 Why am I drawing this
 to the destination?

0:06:08.280000 --> 0:06:11.920000
 The source and the destination have
 to be in different regions.

0:06:11.920000 --> 0:06:16.720000
 In fact, the vault cannot be in the
 same region as the source either.

0:06:16.720000 --> 0:06:26.260000
 So if the source is East US, I could
 have a vault in West US, and I could

0:06:26.260000 --> 0:06:29.200000
 have the destination if I wanted to
 also be West US, as long as it's not

0:06:29.200000 --> 0:06:39.460000
 East US. And in fact, the resource group
 that the vault is in cannot be

0:06:39.460000 --> 0:06:43.000000
 in the same region as a resource
 group for the source.

0:06:43.000000 --> 0:06:47.560000
 The idea of being that if an entire region
 goes down, you're still covered.

0:06:47.560000 --> 0:06:51.900000
 So the vault, it's resource group, all
 have to be in a different region

0:06:51.900000 --> 0:06:53.220000
 than the source.

0:06:53.220000 --> 0:06:56.060000
 And that's why you have the caching
 storage account because that's going

0:06:56.060000 --> 0:06:58.440000
 to be local in the same
 region as the source.

0:06:58.440000 --> 0:07:01.940000
 Source writes very efficiently to the
 caching account, and the caching

0:07:01.940000 --> 0:07:07.080000
 account is the data is then moved
 by the vault into the destination.

0:07:07.080000 --> 0:07:11.440000
 Now in order to facilitate this, you
 have to have an account that has

0:07:11.440000 --> 0:07:17.980000
 at a minimum virtual machine contributor,
 and also site recovery contributor.

0:07:17.980000 --> 0:07:22.120000
 You would need site recovery contributor
 on the source side, and you would

0:07:22.120000 --> 0:07:24.760000
 need virtual machine contributor on the
 destination side because you need

0:07:24.760000 --> 0:07:30.020000
 to be able to create the virtual machine
 of the destination and all of

0:07:30.020000 --> 0:07:32.420000
 its various requirements.

0:07:32.420000 --> 0:07:35.380000
 Okay, so the process, that's
 what you go through.

0:07:35.380000 --> 0:07:36.440000
 Those are the components.

0:07:36.440000 --> 0:07:38.800000
 You want to know the components
 where they go.

0:07:38.800000 --> 0:07:41.580000
 Again, key thing to think about here,
 you've got this vault in the middle,

0:07:41.580000 --> 0:07:47.440000
 and it's facilitating the replication
 of your virtual machines from a

0:07:47.440000 --> 0:07:53.840000
 source. So you're going to have to have
 a virtual network for recovery,

0:07:53.840000 --> 0:07:59.420000
 and the vault, the destination, and
 the resource group for the vault,

0:07:59.420000 --> 0:08:03.680000
 all have to be in a different
 region than the source.

0:08:03.680000 --> 0:08:06.940000
 And that's the process.

0:08:06.940000 --> 0:08:10.060000
 Now what I'm going to do is go through
 a little mini demonstration on

0:08:10.060000 --> 0:08:15.360000
 setting up your recovery for
 an Azure virtual machine.

0:08:15.360000 --> 0:08:22.140000
 And to do this, I've got a
 virtual machine somewhere.

0:08:22.140000 --> 0:08:27.240000
 It looks like I had
 a virtual machine.

0:08:27.240000 --> 0:08:30.020000
 Oh, there it is.

0:08:30.020000 --> 0:08:32.640000
 It's not the exact right one.

0:08:32.640000 --> 0:08:33.840000
 I was on the wrong
 virtual machine.

0:08:33.840000 --> 0:08:35.580000
 That was why I was messed up.

0:08:35.580000 --> 0:08:38.100000
 All right, I'm going to go
 into a virtual machine.

0:08:38.100000 --> 0:08:44.180000
 Now this can be done via the
 setup for a vault as well.

0:08:44.180000 --> 0:08:48.900000
 So I could go into a vault or provision
 of vault, make sure it goes in

0:08:48.900000 --> 0:08:50.500000
 a different location.

0:08:50.500000 --> 0:08:53.360000
 This is running in the East US, the
 vault and its resource group would

0:08:53.360000 --> 0:08:55.000000
 have to go somewhere else.

0:08:55.000000 --> 0:08:58.640000
 But I'm going to make my life a little
 bit easier and just go through

0:08:58.640000 --> 0:09:02.240000
 the disaster recovery interface
 for the virtual machine itself.

0:09:02.240000 --> 0:09:08.960000
 I'm going to pop on
 disaster recovery.

0:09:08.960000 --> 0:09:12.500000
 And target. So I can target somewhere
 other than East US.

0:09:12.500000 --> 0:09:14.800000
 We'll go with West US.

0:09:14.800000 --> 0:09:15.600000
 Actually, let's see.

0:09:15.600000 --> 0:09:17.040000
 We'll go with West US.

0:09:17.040000 --> 0:09:18.960000
 I go to advanced settings.

0:09:18.960000 --> 0:09:26.080000
 And what this is going to do is this
 is going to allow me to configure

0:09:26.080000 --> 0:09:29.560000
 my restore. And this
 is kind of important.

0:09:29.560000 --> 0:09:32.080000
 So where am I going
 to target a restore?

0:09:32.080000 --> 0:09:34.760000
 I'm going to target
 INE demonstrations.

0:09:34.760000 --> 0:09:36.560000
 That's my subscription.

0:09:36.560000 --> 0:09:44.180000
 All right, I'm going to create a new
 resource group called AZ-1042-ASR.

0:09:44.180000 --> 0:09:47.240000
 Create a new virtual network.

0:09:47.240000 --> 0:09:50.980000
 OK, put this in a single instance.


0:09:50.980000 --> 0:09:56.380000
 And I can see storage settings, replication
 settings and extension settings.

0:09:56.380000 --> 0:10:02.240000
 If I show details, here's
 my cache storage account.

0:10:02.240000 --> 0:10:07.280000
 And here's my source managed disk,
 standard SSD, replicated.

0:10:07.280000 --> 0:10:09.940000
 It's a new replicated disk.

0:10:09.940000 --> 0:10:16.340000
 And I can choose standard SSD,
 standard HDD or premium SSD.

0:10:16.340000 --> 0:10:19.520000
 Now this is the OS disk,
 so it must be included.

0:10:19.520000 --> 0:10:23.720000
 If I had data disks, it would give you
 the option per data disk, whether

0:10:23.720000 --> 0:10:26.040000
 or not I wanted to include them.

0:10:26.040000 --> 0:10:31.120000
 Now for me, I'm going through this
 interface and I'm pretty much just

0:10:31.120000 --> 0:10:31.720000
 clicking things.

0:10:31.720000 --> 0:10:33.640000
 And it's pretty straightforward.

0:10:33.640000 --> 0:10:37.140000
 If you're taking an exam, you want
 to make sure you do this and you do

0:10:37.140000 --> 0:10:39.160000
 understand what these options are.


0:10:39.160000 --> 0:10:42.400000
 OK, so I'm going to go into
 the replication settings.

0:10:42.400000 --> 0:10:45.440000
 OK, so where is my vault?

0:10:45.440000 --> 0:10:47.340000
 I've got INE demonstrations.

0:10:47.340000 --> 0:10:52.660000
 Now I already have a vault over there
 in the West US, site recovery vault

0:10:52.660000 --> 0:10:57.860000
 West US. Resource group is
 site recovery vault RG.

0:10:57.860000 --> 0:11:02.980000
 And replication policy is a
 24 hour retention policy.

0:11:02.980000 --> 0:11:05.700000
 And you can actually change that and
 I'll show you some of the policy

0:11:05.700000 --> 0:11:14.120000
 settings. And next, I simply review
 and start the replication.

0:11:14.120000 --> 0:11:19.700000
 OK, and so what it's doing, it's going
 through and it is creating the

0:11:19.700000 --> 0:11:21.160000
 target resources.

0:11:21.160000 --> 0:11:28.340000
 OK, and what I want to do is go ahead
 and take a look at that site recovery

0:11:28.340000 --> 0:11:33.900000
 vault. OK, and this is
 my site recovery vault.

0:11:33.900000 --> 0:11:37.100000
 And what I'm going to do is go to site
 recovery infrastructure because

0:11:37.100000 --> 0:11:43.300000
 I want to show you, you know, it went
 to that 24 hour, which is really

0:11:43.300000 --> 0:11:46.060000
 policy. And we're going to kind
 of show you where that is.

0:11:46.060000 --> 0:11:49.840000
 OK, so I've got Azure virtual machines
 and there's a few different settings.

0:11:49.840000 --> 0:11:52.360000
 We're not going to go through all of
 them, but I do want to go replication

0:11:52.360000 --> 0:11:57.120000
 policies. OK, here's that
 24 hour retention policy.

0:11:57.120000 --> 0:12:01.020000
 And if I dive into that,
 fairly straightforward.

0:12:01.020000 --> 0:12:04.220000
 Recovery point retention
 in hours 24.

0:12:04.220000 --> 0:12:06.420000
 All right, app consistent
 snapshot.

0:12:06.420000 --> 0:12:10.780000
 Now, the what the app consistent snapshot
 does is it takes a snapshot,

0:12:10.780000 --> 0:12:18.960000
 not only of the virtual disks, but
 it will also take a snapshot of the

0:12:18.960000 --> 0:12:23.320000
 in memory. Well, or the memory
 of the virtual machine.

0:12:23.320000 --> 0:12:26.980000
 So whatever is in memory, get snap shotted
 as well as any transactions,

0:12:26.980000 --> 0:12:29.760000
 any distransactions that
 are currently ongoing.

0:12:29.760000 --> 0:12:32.360000
 And the idea being that, you know,
 if you're running an app such as a

0:12:32.360000 --> 0:12:35.900000
 database, you can get it back
 to a consistent point.

0:12:35.900000 --> 0:12:39.860000
 So you know that it is literally
 exactly as it was at that point.

0:12:39.860000 --> 0:12:45.640000
 And so that app consistent snapshot
 is occurring every four hours.

0:12:45.640000 --> 0:12:49.640000
 And there's a few different options
 for that off one, two, three, four

0:12:49.640000 --> 0:12:55.580000
 up to 12 hours. OK, so by default,
 you're taking that once every four

0:12:55.580000 --> 0:13:02.020000
 hours. Recovery point retention
 is between zero and 72.

0:13:02.020000 --> 0:13:05.900000
 OK, and so if you put it to zero,
 it doesn't mean you can't recover.

0:13:05.900000 --> 0:13:11.660000
 It just means it's not holding
 any previous recovery points.

0:13:11.660000 --> 0:13:14.720000
 All right, so those are the
 replication policies.

0:13:14.720000 --> 0:13:20.880000
 That's at 24 hours and I
 got to replicated items.

0:13:20.880000 --> 0:13:26.460000
 OK, now right now I only have one replicated
 item that is still in the

0:13:26.460000 --> 0:13:28.480000
 process of synchronizing.

0:13:28.480000 --> 0:13:30.900000
 It does take a while, the
 initial synchronization.

0:13:30.900000 --> 0:13:33.320000
 I set this up about an
 hour and a half ago.

0:13:33.320000 --> 0:13:34.940000
 Now I'm going to go ahead.

0:13:34.940000 --> 0:13:37.960000
 I can't actually recover it, but I
 just want to give you a quick look

0:13:37.960000 --> 0:13:38.740000
 at that interface.

0:13:38.740000 --> 0:13:43.540000
 And it's really a pretty much
 a point and click interface.

0:13:43.540000 --> 0:13:50.620000
 OK, so this is the recovery status,
 the recovery blade for a particular

0:13:50.620000 --> 0:13:51.600000
 virtual machine again.

0:13:51.600000 --> 0:13:54.320000
 This is a different virtual machine
 that's when I set up about an hour

0:13:54.320000 --> 0:13:57.000000
 and a half ago. It's
 80% synchronized.

0:13:57.000000 --> 0:13:58.440000
 I see it's healthy.

0:13:58.440000 --> 0:13:59.820000
 There's no errors.

0:13:59.820000 --> 0:14:04.460000
 OK, but it hasn't been
 test failover.

0:14:04.460000 --> 0:14:05.940000
 It hasn't fail over yet.

0:14:05.940000 --> 0:14:08.460000
 OK, there's options here, failover
 and test failover.

0:14:08.460000 --> 0:14:12.760000
 And when you click one of these options,
 what happens is you get essentially

0:14:12.760000 --> 0:14:17.420000
 the same thing I just showed you on
 the policy where you can check off

0:14:17.420000 --> 0:14:19.680000
 not the policy, but
 the initial setup.

0:14:19.680000 --> 0:14:21.780000
 OK, where do I want this to go?

0:14:21.780000 --> 0:14:24.400000
 What size virtual machine
 do I want to use?

0:14:24.400000 --> 0:14:28.380000
 Any storage settings that I want, and
 it will come up quickly because

0:14:28.380000 --> 0:14:30.980000
 the replication has
 already occurred.

0:14:30.980000 --> 0:14:33.680000
 You're not building a
 virtual machine new.

0:14:33.680000 --> 0:14:40.440000
 All you're really doing is you are taking
 a replicated copy that's already

0:14:40.440000 --> 0:14:40.980000
 been replicated.

0:14:40.980000 --> 0:14:45.500000
 It's been kept up to date and you're
 creating a new virtual machine on

0:14:45.500000 --> 0:14:48.920000
 top of that replicated disk set.

0:14:48.920000 --> 0:14:50.740000
 And so it's just a
 couple of settings.

0:14:50.740000 --> 0:14:52.520000
 You can walk through
 that very quickly.

0:14:52.520000 --> 0:14:57.500000
 I'm not going to hold you up on that,
 but that is your virtual machine

0:14:57.500000 --> 0:15:02.840000
 recovery. One thing to keep in mind is
 you are going from region to region.

0:15:02.840000 --> 0:15:06.640000
 There are other recovery options that
 go outside of virtual machines.

0:15:06.640000 --> 0:15:10.580000
 You've got the ability to set up recovery
 for your on-premises environment.

0:15:10.580000 --> 0:15:14.560000
 Again, remember that's Hyper-V, VMware,
 and even physical machines.

0:15:14.560000 --> 0:15:19.080000
 So this is a very powerful tool, but
 also when it comes to working just

0:15:19.080000 --> 0:15:25.440000
 with virtual machines, it is a tool with
 really a minimized set of operations

0:15:25.440000 --> 0:15:28.400000
 to make it, well, operational.