1
00:00:01,040 --> 00:00:02,060
In this demonstration,

2
00:00:02,060 --> 00:00:04,230
I'm going to teach you how to set up and configure

3
00:00:04,230 --> 00:00:07,040
Quorum for our Windows Server cluster.

4
00:00:07,040 --> 00:00:09,240
We'll also cover Cluster‑Aware Updating,

5
00:00:09,240 --> 00:00:12,670
and by the end of this demo, we will have set up also,

6
00:00:12,670 --> 00:00:16,980
but wait, there's more, we'll also have set up Storage Spaces Direct,

7
00:00:16,980 --> 00:00:18,980
as well as a highly available VM.

8
00:00:18,980 --> 00:00:21,660
Yeah, a lot to do for sure, so let's get right to it.

9
00:00:21,660 --> 00:00:24,540
I am still on my cluster node MEM1,

10
00:00:24,540 --> 00:00:27,410
and hopefully you've got your screen set at a

11
00:00:27,410 --> 00:00:29,810
resolution to where this isn't too small.

12
00:00:29,810 --> 00:00:33,040
Unfortunately, we record at 1080 resolution,

13
00:00:33,040 --> 00:00:35,810
which can make things a bit small for old

14
00:00:35,810 --> 00:00:39,890
non‑responsive applications like MMC consoles.

15
00:00:39,890 --> 00:00:44,190
But I created a custom console where I've brought in a whole bunch of

16
00:00:44,190 --> 00:00:47,540
snap‑ins that I think are relevant to our interests.

17
00:00:47,540 --> 00:00:50,920
So first of all, let's open up my Failover Cluster Manager,

18
00:00:50,920 --> 00:00:56,000
and let's take a look at my az801cluster cluster.

19
00:00:56,000 --> 00:00:59,000
And specifically, I'm going to right‑click that cluster node,

20
00:00:59,000 --> 00:01:02,460
and notice that you can always validate your cluster,

21
00:01:02,460 --> 00:01:06,740
view the most recent validation report, add a node.

22
00:01:06,740 --> 00:01:09,590
But what I want to do is go to More Actions,

23
00:01:09,590 --> 00:01:14,940
Configure Cluster Quorum Settings, and this launches yet another wizard.

24
00:01:14,940 --> 00:01:19,560
And what I'm going to do here is not let the cluster decide what the majority

25
00:01:19,560 --> 00:01:23,440
should be or just jump directly to configuring a witness.

26
00:01:23,440 --> 00:01:27,510
I'm going to go to Advanced quorum configuration so we can have the

27
00:01:27,510 --> 00:01:31,130
most flexibility over the entire Quorum setup process.

28
00:01:31,130 --> 00:01:32,820
So let me click Next.

29
00:01:32,820 --> 00:01:36,330
First, we're asked nodewise who is going to have a vote.

30
00:01:36,330 --> 00:01:40,290
In this case, I just have two nodes, mem1 and mem2.

31
00:01:40,290 --> 00:01:40,830
That's fine.

32
00:01:40,830 --> 00:01:41,840
They're both up.

33
00:01:41,840 --> 00:01:45,020
The big question here is what kind of witness do I want?

34
00:01:45,020 --> 00:01:47,730
I think I already told you I want to do a cloud witness,

35
00:01:47,730 --> 00:01:51,540
but notice that we've got disk and file share and do not

36
00:01:51,540 --> 00:01:53,950
configure as other possible options.

37
00:01:53,950 --> 00:01:59,740
Now when we configure a cloud witness, we need to connect to a storage account.

38
00:01:59,740 --> 00:02:02,910
So that presumes, let me jump over to the Azure portal,

39
00:02:02,910 --> 00:02:07,130
that you have a general purpose v2 storage account created,

40
00:02:07,130 --> 00:02:12,540
and you also need to use the access key authorization method.

41
00:02:12,540 --> 00:02:16,250
Now the good news about access keys is that they're

42
00:02:16,250 --> 00:02:18,540
convenient and that you can rotate them,

43
00:02:18,540 --> 00:02:19,490
as you can see.

44
00:02:19,490 --> 00:02:25,140
This would invalidate any leaked keys that may be out there in the world.

45
00:02:25,140 --> 00:02:28,860
And you can also, oh well, I noticed this for the first time,

46
00:02:28,860 --> 00:02:30,410
set a rotation reminder.

47
00:02:30,410 --> 00:02:31,670
I've never seen that before.

48
00:02:31,670 --> 00:02:32,740
Cool.

49
00:02:32,740 --> 00:02:36,340
But I was looking up here because there's normally a bit of verbiage

50
00:02:36,340 --> 00:02:40,340
that you can also store your access keys in Key Vault.

51
00:02:40,340 --> 00:02:42,180
But anyway, those are some points.

52
00:02:42,180 --> 00:02:43,760
Good news is convenience.

53
00:02:43,760 --> 00:02:47,960
Bad news is that if the plain text of an access key leaks,

54
00:02:47,960 --> 00:02:51,040
you really want to remediate that immediately.

55
00:02:51,040 --> 00:02:53,990
Doesn't matter whether you use key1 or key2 here,

56
00:02:53,990 --> 00:02:56,140
but we'll need one of those values.

57
00:02:56,140 --> 00:02:58,080
So we need the name of the storage account,

58
00:02:58,080 --> 00:03:03,110
the key, and then the service endpoint is core.windows.net.

59
00:03:03,110 --> 00:03:07,840
That's the public DNS that Microsoft uses in Azure for Azure Storage.

60
00:03:07,840 --> 00:03:11,080
So if I go over to my PowerShell script file,

61
00:03:11,080 --> 00:03:14,570
I can verify here because this is doing the same thing in PowerShell,

62
00:03:14,570 --> 00:03:16,700
Set‑ClusterQuorum‑‑CloudWitness.

63
00:03:16,700 --> 00:03:20,940
Timstorage001 is the name of my storage account.

64
00:03:20,940 --> 00:03:24,550
I'm going to copy out the key, and yes,

65
00:03:24,550 --> 00:03:29,210
I'm regenerating the key after I'm finished here because I certainly

66
00:03:29,210 --> 00:03:33,310
don't want a real‑live key in my Pluralsight course.

67
00:03:33,310 --> 00:03:37,670
So my storage account name, timstorage001.

68
00:03:37,670 --> 00:03:45,060
Let me Ctrl+V that key in there and click Next, Next, and here we go.

69
00:03:45,060 --> 00:03:50,540
So now we've got node majority plus witness as our forum type.

70
00:03:50,540 --> 00:03:50,940
Good.

71
00:03:50,940 --> 00:03:54,510
Now we're not going to really be able to do much of anything

72
00:03:54,510 --> 00:03:57,050
on this cluster until we've got storage,

73
00:03:57,050 --> 00:04:00,380
so we might as well take care of that next actually.

74
00:04:00,380 --> 00:04:03,750
So, I'm going to do Storage Spaces Direct.

75
00:04:03,750 --> 00:04:06,780
And I know that we haven't gotten to that module yet,

76
00:04:06,780 --> 00:04:09,590
so I'm kind of letting the cat out of the bag,

77
00:04:09,590 --> 00:04:14,030
but that's good because I found that when I first learned Storage Spaces Direct,

78
00:04:14,030 --> 00:04:16,940
it was kind of difficult to wrap my mind around it.

79
00:04:16,940 --> 00:04:19,080
So I'm just going to have us jump into the pool,

80
00:04:19,080 --> 00:04:22,400
and then when we circle back and formally cover the subject,

81
00:04:22,400 --> 00:04:25,140
you'll already have some familiarity with it.

82
00:04:25,140 --> 00:04:26,450
Now, what I've got here,

83
00:04:26,450 --> 00:04:30,470
notice I've got my Disk Management console set up for both mem1,

84
00:04:30,470 --> 00:04:34,660
the local box, as well as mem2, the other member of this cluster.

85
00:04:34,660 --> 00:04:36,780
With Storage Spaces Direct,

86
00:04:36,780 --> 00:04:42,260
the value prop is that you can use the server's own disks as shared storage.

87
00:04:42,260 --> 00:04:43,440
You see what I mean?

88
00:04:43,440 --> 00:04:48,300
The catch to that though is that the disks need to be unallocated,

89
00:04:48,300 --> 00:04:53,620
in other words, there can be no partition or volume structure on those disks.

90
00:04:53,620 --> 00:04:58,940
And number two, they need to be online and initialized.

91
00:04:58,940 --> 00:05:03,680
So as you can see, I've got two disks, 127 GB each.

92
00:05:03,680 --> 00:05:06,740
And when I connected them, of course, they were offline,

93
00:05:06,740 --> 00:05:09,510
so I right‑clicked in here and went Online.

94
00:05:09,510 --> 00:05:12,420
Then I right‑clicked, and I did initialize.

95
00:05:12,420 --> 00:05:15,910
And you can use either Master Boot Record or GUID

96
00:05:15,910 --> 00:05:17,900
Partition Table for your disk layout.

97
00:05:17,900 --> 00:05:19,350
I'm using MBR here.

98
00:05:19,350 --> 00:05:20,830
I'm kicking it old school.

99
00:05:20,830 --> 00:05:24,840
And I've got that same configuration from mem2.

100
00:05:24,840 --> 00:05:26,750
As a good best practice,

101
00:05:26,750 --> 00:05:31,010
you want to do everything you can to make sure that the storage topology,

102
00:05:31,010 --> 00:05:34,410
that is the disk geometry and topology,

103
00:05:34,410 --> 00:05:37,580
whatever term you want to use for your cluster nodes,

104
00:05:37,580 --> 00:05:38,080
matches.

105
00:05:38,080 --> 00:05:39,740
It's easiest that way.

106
00:05:39,740 --> 00:05:40,520
All right,

107
00:05:40,520 --> 00:05:43,750
so we've got a total of four disks that we're going to

108
00:05:43,750 --> 00:05:46,390
give over entirely to the cluster.

109
00:05:46,390 --> 00:05:50,090
And we can do that with one simple line of PowerShell.

110
00:05:50,090 --> 00:05:53,870
Let me show you right here, Enable‑ClusterStorageSpacesDirect.

111
00:05:53,870 --> 00:05:58,600
And then for CimSession, it's 801cluster.

112
00:05:58,600 --> 00:06:01,900
And the command does a lot of heavy lifting,

113
00:06:01,900 --> 00:06:06,640
and it's got a high impact here because by answering yes here,

114
00:06:06,640 --> 00:06:09,540
I'm actually going to do A, whoops, not in here,

115
00:06:09,540 --> 00:06:11,470
down in the output, I'm going to type A,

116
00:06:11,470 --> 00:06:13,620
meaning Yes to all.

117
00:06:13,620 --> 00:06:18,170
This is not only enabling Storage Spaces Direct on the cluster,

118
00:06:18,170 --> 00:06:23,030
but the cmdlet now is going to go out to each node and try

119
00:06:23,030 --> 00:06:27,630
to grab hold of every eligible disk and bring those disks

120
00:06:27,630 --> 00:06:31,040
in for use as a shared pool.

121
00:06:31,040 --> 00:06:35,560
And this means that going forward, if you add more storage to your servers,

122
00:06:35,560 --> 00:06:40,310
those disks are going to be snapped up by Storage Spaces Direct as well.

123
00:06:40,310 --> 00:06:44,850
So again, it's a pretty high‑impact operation here, as you can see.

124
00:06:44,850 --> 00:06:47,310
So that didn't take long to do.

125
00:06:47,310 --> 00:06:49,940
I know that in future lessons,

126
00:06:49,940 --> 00:06:52,900
I'll show you how to report on Storage Spaces Direct,

127
00:06:52,900 --> 00:06:55,210
but we don't need to worry about that right now.

128
00:06:55,210 --> 00:06:57,580
I just wanted that operation to complete.

129
00:06:57,580 --> 00:07:00,120
And you'll notice that if we go back to Failover

130
00:07:00,120 --> 00:07:08,180
Cluster Manager and if I go to Storage, we now have Cluster Pool 1 defined.

131
00:07:08,180 --> 00:07:12,120
It has the logical name S2D on az801cluster.

132
00:07:12,120 --> 00:07:14,630
And if I go to Physical Disks,

133
00:07:14,630 --> 00:07:19,440
you can see that it's enlisted both disks from each server.

134
00:07:19,440 --> 00:07:20,630
Now then from here,

135
00:07:20,630 --> 00:07:23,630
we can then start to create virtual disks that have

136
00:07:23,630 --> 00:07:28,470
different types of redundancy, like a mirror or a stripe set with parity.

137
00:07:28,470 --> 00:07:30,340
But again, that's all for the future.

138
00:07:30,340 --> 00:07:31,310
One thing at a time.

139
00:07:31,310 --> 00:07:33,930
I just wanted to make sure to have that here.

140
00:07:33,930 --> 00:07:38,780
Now about Cluster‑Aware Updating, let's go back to the cluster node here,

141
00:07:38,780 --> 00:07:44,230
right‑click, More Actions, and go to Cluster‑Aware Updating.

142
00:07:44,230 --> 00:07:49,640
This invokes a separate graphical user interface, as you can see here.

143
00:07:49,640 --> 00:07:53,540
And the idea here is that we can connect to our cluster, and you

144
00:07:53,540 --> 00:07:55,690
might have more than one cluster available,

145
00:07:55,690 --> 00:07:59,360
so you can use the single GUI to manage a multi‑cluster

146
00:07:59,360 --> 00:08:01,180
infrastructure like a cluster set,

147
00:08:01,180 --> 00:08:05,240
for example, and we can preview updates for this cluster.

148
00:08:05,240 --> 00:08:08,460
Now the plugin is either going to be WindowsUpdate,

149
00:08:08,460 --> 00:08:14,340
which is really Microsoft Update, just Microsoft hotfixes,

150
00:08:14,340 --> 00:08:16,220
or there's RollingUpgradePlugin.

151
00:08:16,220 --> 00:08:20,070
And depending upon whether you want to use the internet,

152
00:08:20,070 --> 00:08:22,490
that is Microsoft Update online,

153
00:08:22,490 --> 00:08:25,290
or if you want to use Windows Server Update Services,

154
00:08:25,290 --> 00:08:27,840
you also could do that as well.

155
00:08:27,840 --> 00:08:30,690
Now I'm just going to run an Update preview against the

156
00:08:30,690 --> 00:08:34,080
standard internet‑based Microsoft Update,

157
00:08:34,080 --> 00:08:39,300
and this is going to show any outstanding updates that those nodes don't have.

158
00:08:39,300 --> 00:08:41,660
And it looks like they're pretty well patched.

159
00:08:41,660 --> 00:08:45,320
It looks like we just have a Security Intelligence Update for

160
00:08:45,320 --> 00:08:47,800
Microsoft Defender on each of those nodes.

161
00:08:47,800 --> 00:08:49,540
Not too bad.

162
00:08:49,540 --> 00:08:52,740
Now we can apply updates to this cluster.

163
00:08:52,740 --> 00:08:55,570
We can configure cluster self‑updating options,

164
00:08:55,570 --> 00:08:58,570
but the catch with that is in order to configure

165
00:08:58,570 --> 00:09:00,630
your cluster to do self‑updating,

166
00:09:00,630 --> 00:09:05,720
you need to have the Cluster‑Aware Updating role running in your cluster.

167
00:09:05,720 --> 00:09:08,520
And I've found that works better with PowerShell.

168
00:09:08,520 --> 00:09:13,010
So here on line 23, we could do Add‑CauClusterRole,

169
00:09:13,010 --> 00:09:19,390
providing the name of the cluster, and then you define the auto‑update schedule.

170
00:09:19,390 --> 00:09:24,940
This example says that every Tuesday and Saturday for every three weeks,

171
00:09:24,940 --> 00:09:29,110
the cluster will do an automatic cluster run,

172
00:09:29,110 --> 00:09:32,880
and this highly available role is that Engine of Creation,

173
00:09:32,880 --> 00:09:34,760
if that makes sense.

174
00:09:34,760 --> 00:09:36,340
So let me right‑click.

175
00:09:36,340 --> 00:09:40,530
And this is going to create our first highly available role on the cluster.

176
00:09:40,530 --> 00:09:43,960
And it's going to take care of self‑updating this

177
00:09:43,960 --> 00:09:50,000
cluster on two days every three weeks. Well that was pretty easy, wasn't it?