1
00:00:01,040 --> 00:00:04,110
And the specific recovery options, or the restoration

2
00:00:04,110 --> 00:00:07,200
options, are up here on the toolbar. We've got the Restore

3
00:00:07,200 --> 00:00:09,560
VM path and the File Recovery path.

4
00:00:09,560 --> 00:00:12,640
Let's take a look at Restore VM first.

5
00:00:12,640 --> 00:00:15,320
Now hopefully it makes sense that you can't do anything unless

6
00:00:15,320 --> 00:00:18,550
you've got a restore point. And I've talked about the different

7
00:00:18,550 --> 00:00:22,080
consistency types that we have, and you'll notice that we've got

8
00:00:22,080 --> 00:00:24,680
quite a few Crash Consistent. Now,

9
00:00:24,680 --> 00:00:28,330
I'm not going to freak out about that because we know that crash

10
00:00:28,330 --> 00:00:32,660
consistent is what happens if Azure Backup works when the virtual

11
00:00:32,660 --> 00:00:36,230
machine is stopped and deallocated. And I happen to know that this

12
00:00:36,230 --> 00:00:40,110
particular machine has been stopped and deallocated for the last

13
00:00:40,110 --> 00:00:42,550
several days, the last week or more.

14
00:00:42,550 --> 00:00:47,750
I'm going to grab the most recent Application Consistent snapshot, click OK. And

15
00:00:47,750 --> 00:00:51,550
then here's what I was talking about in the theory part of this module, do you

16
00:00:51,550 --> 00:00:55,500
want to create an entirely new virtual machine or are you going to do an

17
00:00:55,500 --> 00:01:00,280
in‑place restore of just the storage. And notice also it says that if you're

18
00:01:00,280 --> 00:01:02,420
going to do an alternate configuration,

19
00:01:02,420 --> 00:01:04,730
you're going to have to use PowerShell or the Azure

20
00:01:04,730 --> 00:01:07,050
CLI to do that restore operation.

21
00:01:07,050 --> 00:01:11,750
If we're going to do a Create new, note that there's two options here as well.

22
00:01:11,750 --> 00:01:16,050
We could create a new virtual machine by providing a name; Resource group;

23
00:01:16,050 --> 00:01:20,430
this Virtual network would definitely be one that's different from where

24
00:01:20,430 --> 00:01:25,640
web1 actually lives, you would have hopefully a dedicated DR, or disaster

25
00:01:25,640 --> 00:01:31,000
recovery, virtual net to work with; Subnet; and then you specify one of

26
00:01:31,000 --> 00:01:32,340
your storage accounts.

27
00:01:32,340 --> 00:01:37,140
I've got twdrstorage created specifically for this reason.

28
00:01:37,140 --> 00:01:42,850
Give it a name, and I could call it web1‑test just to verify that it's

29
00:01:42,850 --> 00:01:47,580
going to be a test. And then we can click Restore to send that job into

30
00:01:47,580 --> 00:01:51,890
the RS vault. First we can see that we're validating against the ARM

31
00:01:51,890 --> 00:01:56,740
schema, and assuming all of our choices are correct and valid, that job

32
00:01:56,740 --> 00:01:58,940
will be triggered. There it is.

33
00:01:58,940 --> 00:02:02,340
Alright, so while that's cooking, we can do work in parallel.

34
00:02:02,340 --> 00:02:07,080
Let's go back to Restore a VM, and let's again select a restore point.

35
00:02:07,080 --> 00:02:11,140
I'm going to choose the most recent Application Consistent restore point,

36
00:02:11,140 --> 00:02:12,090
click OK.

37
00:02:12,090 --> 00:02:12,630
And again,

38
00:02:12,630 --> 00:02:17,960
we'll do Create new, but then we may just want to extract the disks from that VM

39
00:02:17,960 --> 00:02:22,910
and put them into a storage account. And then we could use those restored disks

40
00:02:22,910 --> 00:02:28,090
to create a new VM programmatically, specifying a different configuration. Let

41
00:02:28,090 --> 00:02:30,440
me click Restore to trigger that job.

42
00:02:30,440 --> 00:02:33,470
So those are the two options for Create new.

43
00:02:33,470 --> 00:02:36,970
The last thing we'll look at after Azure Resource Manager gives us

44
00:02:36,970 --> 00:02:40,370
our control back is we'll look at the in‑place restore. Ooh,

45
00:02:40,370 --> 00:02:43,790
Validation failed. Click here for more details.

46
00:02:43,790 --> 00:02:47,930
Unable to initiate restore as another restore is currently in progress.

47
00:02:47,930 --> 00:02:52,680
Well that is a bummer. Wouldn't we expect logically that cloud computing

48
00:02:52,680 --> 00:02:58,010
supports parallel operations? Hopefully because that restore is triggered

49
00:02:58,010 --> 00:03:03,200
successfully, let's try again the second restore and see if that's all

50
00:03:03,200 --> 00:03:07,920
that the error was about, just making sure that the first job was at least

51
00:03:07,920 --> 00:03:12,010
past the trigger state and we don't actually have to sit and wait for the

52
00:03:12,010 --> 00:03:12,490
whole thing.

53
00:03:12,490 --> 00:03:13,590
No, it looks like we do.

54
00:03:13,590 --> 00:03:16,340
Wow, that is really disappointing.

55
00:03:16,340 --> 00:03:20,020
Okay, note to self. So to finish what we were talking about,

56
00:03:20,020 --> 00:03:22,250
if we go to Replace existing,

57
00:03:22,250 --> 00:03:25,650
the only option here is Replace Disks. And remember, the point with

58
00:03:25,650 --> 00:03:28,800
in‑place restore is that you don't want to scratch your VM

59
00:03:28,800 --> 00:03:31,000
configuration and all of its dependencies.

60
00:03:31,000 --> 00:03:34,150
You want to keep the shell, as it were, of the VM in place.

61
00:03:34,150 --> 00:03:37,510
You're just going to swap out the OS and any data disks that are

62
00:03:37,510 --> 00:03:41,940
attached. And all we need to do for this is Replace existing,

63
00:03:41,940 --> 00:03:45,560
assuming that you've got your snapshot, as I said before, and

64
00:03:45,560 --> 00:03:48,820
then we specify one of our storage accounts as a staging

65
00:03:48,820 --> 00:03:51,170
location, and then we hit Restore.

66
00:03:51,170 --> 00:03:52,840
That's all there is to it.

67
00:03:52,840 --> 00:03:56,630
Good, so let's come back out of the screen and now let's take a look at

68
00:03:56,630 --> 00:04:00,710
the File Recovery scenario, which is where you need to restore individual

69
00:04:00,710 --> 00:04:04,240
file system artifacts from a backup snapshot.

70
00:04:04,240 --> 00:04:08,060
So once again, step one is to choose the recovery point.

71
00:04:08,060 --> 00:04:10,940
I'm going to choose my AppConsistent one.

72
00:04:10,940 --> 00:04:14,690
And depending upon whether it's a Windows or Linux virtual machine,

73
00:04:14,690 --> 00:04:17,060
you'll have either a Python or a PowerShell script.

74
00:04:17,060 --> 00:04:21,170
This is a Windows machine, so I'm going to download an executable to my

75
00:04:21,170 --> 00:04:25,820
local system and we'll be able to decrypt that executable and pop out

76
00:04:25,820 --> 00:04:30,580
the PowerShell script that's inside of it by providing this one‑time

77
00:04:30,580 --> 00:04:38,200
password that Azure is simultaneously generating for us. The script

78
00:04:38,200 --> 00:04:42,210
file is ready to download, so I'll click Download to bring that down.

79
00:04:42,210 --> 00:04:49,260
I will also copy the password to my clipboard. And I missed the browser

80
00:04:49,260 --> 00:04:55,200
notification, so let me manually go to my Downloads directory here. Double

81
00:04:55,200 --> 00:05:00,690
left‑click, I'm going to paste in that password. That decrypts the exe,

82
00:05:00,690 --> 00:05:06,050
pops open an elevated PowerShell console, connects to the iSCSI client or

83
00:05:06,050 --> 00:05:10,120
initiator service on the machine, and it says here to Open Explorer and

84
00:05:10,120 --> 00:05:13,660
browse for files after recovery, to remove the disks and close the

85
00:05:13,660 --> 00:05:17,940
connection, please click 'Unmount Disks' in the portal.

86
00:05:17,940 --> 00:05:21,590
See, so now if we go to our File Explorer here,

87
00:05:21,590 --> 00:05:26,170
we're going to see a new disk right here, this E drive that's labeled Windows.

88
00:05:26,170 --> 00:05:30,320
This is the OS disk from that recovery snapshot.

89
00:05:30,320 --> 00:05:36,150
So the idea is that I can come in here and I can restore files and folders,

90
00:05:36,150 --> 00:05:39,930
and then when I'm finished I can close the connection. Pretty nice. And I

91
00:05:39,930 --> 00:05:42,640
mentioned this already, but let me repeat it again.

92
00:05:42,640 --> 00:05:45,750
It says here, To restore files larger than 10GB,

93
00:05:45,750 --> 00:05:50,010
restore at least the disks, but maybe the entire VM to an

94
00:05:50,010 --> 00:05:53,960
alternate location, and that the data transfer rate is not 10

95
00:05:53,960 --> 00:05:57,930
GB/Hr, but it's up to 10 GB an hour.

96
00:05:57,930 --> 00:06:02,400
So Microsoft heavily throttles the connection, so don't abuse it, okay. It's

97
00:06:02,400 --> 00:06:04,700
a pretty nifty technology though, I think you'd agree.

98
00:06:04,700 --> 00:06:08,400
Let's click Unmount Disks, and that will break the connection.

99
00:06:08,400 --> 00:06:10,290
Now, as this says up here, the drives,

100
00:06:10,290 --> 00:06:15,330
if I don't click Unmount Disks, will remain mounted on my system for 12 hours.

101
00:06:15,330 --> 00:06:19,070
Now please understand, that was an over‑the‑network, over‑the‑internet

102
00:06:19,070 --> 00:06:22,540
connection from my client machine into that snapshot.

103
00:06:22,540 --> 00:06:26,740
It's not like I downloaded that 127GB file.

104
00:06:26,740 --> 00:06:28,340
No, I didn't do that.

105
00:06:28,340 --> 00:06:32,520
It's just coming over the internet via good, old iSCSI

106
00:06:32,520 --> 00:06:37,130
and web services and web APIs. Alright, so how do I want to finish here?

107
00:06:37,130 --> 00:06:37,380
Well,

108
00:06:37,380 --> 00:06:44,740
let's go back to the vault and let's come down under Backup Jobs, and

109
00:06:44,740 --> 00:06:48,160
let's take a look at, it looks like we've got a couple in process.

110
00:06:48,160 --> 00:06:51,940
Here's that restore I was wondering about. Let's go to View details.

111
00:06:51,940 --> 00:06:55,040
Let's see how far along it might be.

112
00:06:55,040 --> 00:06:58,700
So we're recovering a VM to an alternate location.

113
00:06:58,700 --> 00:07:02,240
It looks like we've got a little while here unfortunately. Remember,

114
00:07:02,240 --> 00:07:05,760
this is one of the reasons why you have to have your Azure VMs in the

115
00:07:05,760 --> 00:07:09,900
same region as your RS vault. And I'm not sure why I'm not farther

116
00:07:09,900 --> 00:07:12,580
along than this, to be totally honest, because you know as part of

117
00:07:12,580 --> 00:07:14,240
our backup policy,

118
00:07:14,240 --> 00:07:17,850
we have that Instant Restore area. Now, speaking of which, we could

119
00:07:17,850 --> 00:07:23,020
always come over here under Manage, Backup policies, to find your

120
00:07:23,020 --> 00:07:25,340
policy and edit it just to make sure.

121
00:07:25,340 --> 00:07:28,640
So our az801‑policy, let's take a look.

122
00:07:28,640 --> 00:07:38,000
Yeah, so I'm maxing out the Instant Restore snapshots right here. So, anyway, you know the lay of the land. I think my work here is done.