1
00:00:03,365 --> 00:00:05,735
- Welcome to episode 360
2
00:00:05,795 --> 00:00:08,255
of the Microsoft Cloud IT Pro podcast
3
00:00:08,895 --> 00:00:11,575
recorded live on November 3rd, 2023.
4
00:00:12,285 --> 00:00:14,815
This is a show about Microsoft 365
5
00:00:14,815 --> 00:00:17,055
and Asher from the perspective of it pros
6
00:00:17,055 --> 00:00:20,295
and end users where we discuss
a topic or recent news.
7
00:00:20,395 --> 00:00:22,615
And how about relates to you today,
8
00:00:23,315 --> 00:00:26,215
Ben Laments about a change
in the data processing
9
00:00:26,275 --> 00:00:29,175
for Azure Sentinel, which side note
10
00:00:29,725 --> 00:00:32,215
less than two weeks later
has now been reversed.
11
00:00:32,635 --> 00:00:35,295
And then Scott provides some suggestions
12
00:00:35,295 --> 00:00:37,215
around Microsoft's policies
13
00:00:37,235 --> 00:00:38,775
and how to keep up to date on them.
14
00:00:39,325 --> 00:00:41,895
Then they wrap up the
show by talking about ways
15
00:00:41,955 --> 00:00:44,095
to track resource ownership over time
16
00:00:44,275 --> 00:00:45,695
for your Azure resources.
17
00:00:48,065 --> 00:00:49,695
Scott, I feel like we're maybe sort
18
00:00:49,695 --> 00:00:51,135
of back on schedule for a little bit.
19
00:00:51,185 --> 00:00:53,695
We've been kind of recording
all over the place lately. Uh,
20
00:00:53,995 --> 00:00:54,995
- It is what it is.
21
00:00:54,995 --> 00:00:56,575
We'll get there. Holidays are coming
22
00:00:56,615 --> 00:00:57,615
- Up.
23
00:00:57,615 --> 00:00:58,135
Yeah, that'll throw us for a loop.
24
00:00:58,345 --> 00:00:59,895
We're only 15 minutes late today,
25
00:01:00,535 --> 00:01:01,495
- . Oh
26
00:01:02,005 --> 00:01:05,695
- Yeah, I got caught in a
meeting about some stuff coming
27
00:01:05,795 --> 00:01:09,255
to the business, future
announcements coming down the road,
28
00:01:09,555 --> 00:01:10,815
but I have an announcement.
29
00:01:11,025 --> 00:01:12,095
Let's just dive right into it.
30
00:01:12,095 --> 00:01:14,175
Scott, I have an announcement
I got from Microsoft
31
00:01:14,175 --> 00:01:15,975
that actually was kind of disturbing
32
00:01:16,275 --> 00:01:19,775
and it was a little ironic
33
00:01:20,125 --> 00:01:22,015
with some other
conversations I was having.
34
00:01:22,315 --> 00:01:24,695
So the other day I
should find, I don't know
35
00:01:24,695 --> 00:01:28,295
where this conversation was,
I was talking to a friend
36
00:01:28,295 --> 00:01:30,575
of mine and we were discussing,
37
00:01:31,215 --> 00:01:33,735
I think we were actually
discussing browsers
38
00:01:34,275 --> 00:01:36,415
and our preference for browsers
39
00:01:36,415 --> 00:01:41,335
with Firefox versus
Edge versus A and Safari
40
00:01:41,555 --> 00:01:43,895
and all the options. Brave all
41
00:01:43,895 --> 00:01:45,175
- Those, all the fun things. Yes.
42
00:01:45,315 --> 00:01:47,815
- And he made a comment about trying
43
00:01:47,835 --> 00:01:50,175
to start transitioning
from Edge to Firefox.
44
00:01:50,795 --> 00:01:53,215
One, he's been having some
issues playing videos and Edge,
45
00:01:53,475 --> 00:01:55,775
but the other comment he
made was he's, I'm starting
46
00:01:55,775 --> 00:01:59,175
to get a little worried
about this AI stuff with some
47
00:01:59,175 --> 00:02:01,455
of these companies training LLMs,
48
00:02:02,035 --> 00:02:05,775
and them using my data
to train their models
49
00:02:05,775 --> 00:02:08,575
because there's also been
some debate that we've had
50
00:02:08,575 --> 00:02:12,935
around this, about like training
models off of website data.
51
00:02:13,315 --> 00:02:15,375
If some of these companies,
52
00:02:15,515 --> 00:02:17,735
not necessarily specifically Microsoft
53
00:02:17,875 --> 00:02:22,015
but the OpenAI models the
bar, like all of these,
54
00:02:22,355 --> 00:02:23,775
how they're learning is are they
55
00:02:24,405 --> 00:02:27,855
potentially scraping like
copyrighted material off
56
00:02:27,855 --> 00:02:31,095
of websites to train these
Lang large language models?
57
00:02:31,765 --> 00:02:35,255
Like who owns the stuff on my
blog if they start using my
58
00:02:35,685 --> 00:02:39,535
blog and answers for these
large language models?
59
00:02:39,715 --> 00:02:42,575
All of that. So it was kind
of a security discussion
60
00:02:42,575 --> 00:02:45,575
around AI and access to
data and who you trust
61
00:02:45,635 --> 00:02:46,975
and who you don't trust and all of that.
62
00:02:47,285 --> 00:02:49,095
This was literally like three
63
00:02:49,095 --> 00:02:53,575
or four hours later I got an
email titled Microsoft
64
00:02:53,695 --> 00:02:57,455
Sentinel will begin collecting
security research data on
65
00:02:57,765 --> 00:02:59,525
forward December, 2023.
66
00:02:59,585 --> 00:03:00,925
So December 4th of this year,
67
00:03:01,305 --> 00:03:04,565
and this is not for my tenant,
this is actually for one
68
00:03:04,565 --> 00:03:05,765
of my customer's tenants,
69
00:03:06,065 --> 00:03:07,845
but it says we're changing
70
00:03:07,905 --> 00:03:09,565
how we collect data Starting on this date.
71
00:03:09,865 --> 00:03:11,045
You are receiving this message
72
00:03:11,045 --> 00:03:13,845
because one of your Microsoft
Sentinel workspaces,
73
00:03:13,905 --> 00:03:17,765
so log analytics essentially is opted out
74
00:03:18,145 --> 00:03:20,765
to share the data with
Microsoft security research.
75
00:03:21,195 --> 00:03:23,125
This was turned on for a reason.
76
00:03:23,665 --> 00:03:26,285
We did not want this particular data being
77
00:03:26,305 --> 00:03:27,685
shared to Microsoft Research.
78
00:03:27,925 --> 00:03:30,765
. Turns out Microsoft
security research protects our
79
00:03:30,965 --> 00:03:32,285
customers against threats and attacks
80
00:03:32,285 --> 00:03:34,325
by delivering built-in
state-of-the-art detections,
81
00:03:34,765 --> 00:03:36,525
investigating tools, attack disruptions
82
00:03:36,865 --> 00:03:38,205
and mitigations,
83
00:03:39,125 --> 00:03:42,005
building more accurate models, AI models.
84
00:03:42,565 --> 00:03:44,405
I am reading between the lines there.
85
00:03:44,425 --> 00:03:47,805
It doesn't specifically mention
AI requires diverse data
86
00:03:47,805 --> 00:03:49,525
sets and signals because of this.
87
00:03:50,245 --> 00:03:51,925
Starting December 4th, 2023,
88
00:03:52,285 --> 00:03:55,765
Microsoft security research
will have access to your data
89
00:03:55,785 --> 00:03:57,725
to help build, test and optimize analytics
90
00:03:57,725 --> 00:03:58,765
models and detections.
91
00:03:59,195 --> 00:04:02,405
Okay, so I turned this on specifically
92
00:04:02,405 --> 00:04:04,365
because I wanted this data to stay private
93
00:04:04,705 --> 00:04:05,965
and now you're telling me
94
00:04:06,225 --> 00:04:07,805
- And by turned it on, you mean turned
95
00:04:07,805 --> 00:04:11,205
- It off - Explicitly opted
out Out of this thing. Yeah,
96
00:04:11,425 --> 00:04:12,425
- Out of this.
97
00:04:12,705 --> 00:04:16,125
And now they're just saying,
sorry, too bad. We don't care.
98
00:04:16,185 --> 00:04:17,165
We need to build our models
99
00:04:17,225 --> 00:04:18,445
so we're gonna go grab your data.
100
00:04:18,445 --> 00:04:20,565
Anyways, they do go on to say,
101
00:04:20,665 --> 00:04:23,005
so then they have some more information
102
00:04:23,065 --> 00:04:26,925
and I haven't dug into this
too far yet, but it's privacy
103
00:04:27,025 --> 00:04:28,445
and data handling commitments
104
00:04:28,505 --> 00:04:30,485
and the Microsoft online services terms
105
00:04:30,625 --> 00:04:33,445
and online services data
protection addendum,
106
00:04:33,495 --> 00:04:34,845
their legally applies.
107
00:04:35,385 --> 00:04:39,405
It also says customers using
customer managed keys CMK won't
108
00:04:39,405 --> 00:04:40,605
be affected by this change.
109
00:04:40,945 --> 00:04:43,685
So I am theorizing that by
110
00:04:43,685 --> 00:04:45,725
that single sentence
down towards the bottom
111
00:04:45,725 --> 00:04:48,445
that if I go switch everything
to customer managed keys,
112
00:04:48,915 --> 00:04:50,205
it's not gonna get shared
113
00:04:50,205 --> 00:04:53,165
because potentially this is encrypted then
114
00:04:53,185 --> 00:04:54,725
beyond their ability to grab it.
115
00:04:54,825 --> 00:04:55,925
And then it also has a help
116
00:04:55,925 --> 00:04:58,125
and support if this change
affects your environments.
117
00:04:58,515 --> 00:05:02,205
Well yeah, 'cause you're
actually changing settings on me
118
00:05:02,585 --> 00:05:03,925
and you'd like to learn more about it.
119
00:05:04,145 --> 00:05:05,445
Please contact support
120
00:05:05,585 --> 00:05:08,805
and then it just walks
through like the path to go
121
00:05:08,805 --> 00:05:11,445
through under your issue
type and support NA Azure.
122
00:05:11,445 --> 00:05:13,805
Yep. And then it gives
the account information,
123
00:05:14,165 --> 00:05:15,205
subscription name, all of that in the
124
00:05:15,205 --> 00:05:16,685
Azure subscription that this is using.
125
00:05:17,025 --> 00:05:18,045
I'm not a big fan of this.
126
00:05:18,685 --> 00:05:21,045
I get if they wanna do
this and they send an email
127
00:05:21,185 --> 00:05:26,125
and ask permission or say
would you like to help?
128
00:05:26,785 --> 00:05:29,205
Or even saying,
129
00:05:29,605 --> 00:05:32,005
I mean it would probably be
borderline is if you don't do
130
00:05:32,205 --> 00:05:33,365
anything, this will be turned on.
131
00:05:33,385 --> 00:05:34,525
But you can go back in
132
00:05:34,525 --> 00:05:36,725
and toggle this checkbox to leave it off
133
00:05:36,905 --> 00:05:38,725
or something along those lines.
134
00:05:39,185 --> 00:05:42,485
But just kind of blatantly
saying, we don't care
135
00:05:42,485 --> 00:05:44,405
what you said before
about sharing this data,
136
00:05:44,405 --> 00:05:45,605
we're gonna go get it anyways
137
00:05:45,605 --> 00:05:48,245
because we want to train our
large language models better.
138
00:05:48,765 --> 00:05:50,765
I totally get the how
it helps with security,
139
00:05:50,865 --> 00:05:52,085
how it helps building the models
140
00:05:52,515 --> 00:05:54,445
with security co-pilot coming out.
141
00:05:54,625 --> 00:05:57,290
I'm assuming that also has
something to do with this this,
142
00:05:57,585 --> 00:05:59,605
but just going in and blatantly saying
143
00:06:00,305 --> 00:06:01,405
now we're gonna access it.
144
00:06:01,575 --> 00:06:04,125
Especially when they do tend
145
00:06:04,125 --> 00:06:06,565
to focus a little bit more
on security than maybe some
146
00:06:06,565 --> 00:06:07,765
of the other big players.
147
00:06:08,315 --> 00:06:11,325
I've tended to trust Microsoft
more than Google when it
148
00:06:11,325 --> 00:06:14,645
comes to keeping my data secure
and the privacy of my data.
149
00:06:15,145 --> 00:06:16,605
But this just kind of,
150
00:06:17,065 --> 00:06:19,205
it rubbed me the wrong way when I saw this
151
00:06:19,765 --> 00:06:20,765
- .
152
00:06:22,075 --> 00:06:25,245
Yeah, so I'm, I'm over here
shaking my head and
153
00:06:25,305 --> 00:06:26,325
and just rubbing my eyes.
154
00:06:26,785 --> 00:06:29,245
So there's a couple of things.
155
00:06:30,065 --> 00:06:33,325
So um, I think part of it is the
156
00:06:34,115 --> 00:06:37,125
just abject fear that's
been instilled in everybody.
157
00:06:37,235 --> 00:06:41,205
Like I would not want an
LLM hallucinating on top
158
00:06:41,205 --> 00:06:43,565
of my data either back to me
159
00:06:43,865 --> 00:06:46,525
or certainly in context of
potentially like my data
160
00:06:46,995 --> 00:06:49,245
with other customer
usage, things like that.
161
00:06:49,945 --> 00:06:51,725
So like reading between the lines
162
00:06:52,345 --> 00:06:53,565
and I'm not a lawyer,
163
00:06:54,085 --> 00:06:56,045
I don't work on the sentinel
team, anything like that.
164
00:06:56,405 --> 00:06:59,085
A lot of sentinel isn't
necessarily AI driven
165
00:06:59,225 --> 00:07:02,605
but it is ML driven which
ultimately like these large
166
00:07:02,885 --> 00:07:05,325
language models Yeah those
are actually ML as well.
167
00:07:05,325 --> 00:07:06,485
Like we've conflated the whole
168
00:07:07,045 --> 00:07:08,685
ML AI thing, blah blah
blah, all that stuff.
169
00:07:08,705 --> 00:07:11,565
Yep. I think it's good to
have the signals in there.
170
00:07:11,685 --> 00:07:14,405
I think it's bad to kind of
revert on a decision like that.
171
00:07:14,795 --> 00:07:19,285
It's a good reminder to
everybody that when you sign up
172
00:07:19,305 --> 00:07:23,765
for these services, like
there is a just absolute
173
00:07:24,275 --> 00:07:27,325
crap ton of legalese under
underneath the covers
174
00:07:27,435 --> 00:07:28,965
that drives these things
175
00:07:29,625 --> 00:07:31,885
and I don't think enough customers
176
00:07:32,485 --> 00:07:34,845
actually go and read those.
177
00:07:35,515 --> 00:07:39,125
I've had the opportunity a
couple times in over the course
178
00:07:39,185 --> 00:07:41,205
of my time at Microsoft to actually have
179
00:07:41,205 --> 00:07:45,245
to go read up on
the legal agreement for
180
00:07:45,965 --> 00:07:49,525
Microsoft customers when it
comes to the what's known
181
00:07:49,525 --> 00:07:53,965
as the mosa or the Microsoft
Online subscription agreement.
182
00:07:54,705 --> 00:07:58,525
So that governs things
for MCA customers, sorry,
183
00:07:58,965 --> 00:08:00,285
Microsoft customer agreement MCA
184
00:08:00,785 --> 00:08:04,165
and you know EA customers
enterprise agreement customers
185
00:08:04,165 --> 00:08:05,685
have a slightly different set
186
00:08:05,685 --> 00:08:07,645
of terms even if you don't use Azure,
187
00:08:07,645 --> 00:08:10,645
there's always just the general
like Microsoft terms of use
188
00:08:10,645 --> 00:08:14,485
that you're using in tools,
things like that along the way.
189
00:08:14,785 --> 00:08:16,205
So there's really two things
190
00:08:16,435 --> 00:08:18,645
that even if you're not into the legalese
191
00:08:18,645 --> 00:08:20,125
of the stuff you're doing, it's kind
192
00:08:20,125 --> 00:08:21,125
of interesting if you can go
193
00:08:21,125 --> 00:08:22,205
and stomach your way through it.
194
00:08:22,945 --> 00:08:26,205
One is that most Microsoft
online subscription agreement
195
00:08:26,465 --> 00:08:30,605
and then the thing that everybody
always forgets is the DPA,
196
00:08:30,815 --> 00:08:34,125
which is the data protection addendum.
197
00:08:34,745 --> 00:08:39,565
So the DPA in many cases
is actually the thing
198
00:08:40,155 --> 00:08:42,965
that drives the contract.
199
00:08:43,075 --> 00:08:44,885
Like the legal contract between you
200
00:08:44,885 --> 00:08:48,005
and Microsoft particularly with regard to
201
00:08:48,945 --> 00:08:52,765
the processing and the
security of your customer data
202
00:08:52,945 --> 00:08:54,685
and your personal data in there.
203
00:08:55,225 --> 00:08:57,765
In your case here. Yeah,
you've got a couple options.
204
00:08:58,425 --> 00:09:02,245
One is go familiarize
yourself with the DPA.
205
00:09:02,745 --> 00:09:05,245
And again I encourage every
customer to actually do that.
206
00:09:05,275 --> 00:09:07,445
Like not just your lawyers
but folks should go
207
00:09:07,445 --> 00:09:08,885
and read it, go read the DPA
208
00:09:08,885 --> 00:09:10,805
and see what Microsoft
does with your data.
209
00:09:10,845 --> 00:09:13,885
I don't think Microsoft is
like, they're definitely not
210
00:09:13,885 --> 00:09:15,285
as clear as companies like
211
00:09:15,285 --> 00:09:16,445
Apple when it comes to this stuff.
212
00:09:16,715 --> 00:09:19,725
Like Apple's very clear, hey
we do processing on device,
213
00:09:20,025 --> 00:09:23,685
here's why we do it on device,
blah blah blah privacy, all
214
00:09:23,685 --> 00:09:27,085
that kind of stuff with
providers like Google, Microsoft,
215
00:09:27,625 --> 00:09:30,565
Amazon, like all that
stuff happens in the cloud
216
00:09:31,185 --> 00:09:34,605
or the the other option like
you said is yeah flip over
217
00:09:34,785 --> 00:09:36,085
to CMK.
218
00:09:36,385 --> 00:09:38,205
So CMK is customer managed key
219
00:09:38,335 --> 00:09:41,565
where effectively you're
managing your encryption key
220
00:09:42,275 --> 00:09:45,525
including things like rotation
of that encryption key
221
00:09:46,025 --> 00:09:49,725
inside something like Azure
key vault so that you are in
222
00:09:50,835 --> 00:09:54,165
full control end to end and
you have that capability.
223
00:09:54,305 --> 00:09:56,805
So I run into a lot of customers, I don't,
224
00:09:56,805 --> 00:09:58,005
like I said I don't work on cental
225
00:09:58,005 --> 00:09:59,525
but particularly in like storage land,
226
00:09:59,525 --> 00:10:01,725
like people are very sensitive
to this kind of thing.
227
00:10:02,105 --> 00:10:04,725
So we do like customer provided key.
228
00:10:04,945 --> 00:10:06,845
So we actually we we
do customer managed key
229
00:10:07,015 --> 00:10:10,605
where customers can manage
keys inside of KeyVault
230
00:10:10,865 --> 00:10:12,845
but we also have this
capability called customer
231
00:10:13,205 --> 00:10:15,485
provided key where per request,
232
00:10:15,795 --> 00:10:19,325
like per upload you can actually
send a different encryption
233
00:10:19,345 --> 00:10:23,525
key for every single
object if you're like really
234
00:10:23,735 --> 00:10:27,285
super kind of paranoid about that thing
235
00:10:27,545 --> 00:10:30,325
and how it comes together
and what that looks like.
236
00:10:30,705 --> 00:10:34,445
The other thing that
the DPA is helpful for
237
00:10:35,185 --> 00:10:39,325
is it also doesn't just govern
how Microsoft uses your data.
238
00:10:39,785 --> 00:10:43,045
It governs how things
like data destruction go
239
00:10:43,625 --> 00:10:44,765
and you know
240
00:10:44,765 --> 00:10:48,405
what happens when you know
you issue a delete request
241
00:10:48,625 --> 00:10:50,605
or you delete a particular resource
242
00:10:50,945 --> 00:10:53,285
or something like that along the way.
243
00:10:53,945 --> 00:10:57,125
And then for this one you're
actually kind of lucky in
244
00:10:57,155 --> 00:11:00,925
that you are in maybe like
an Azure service here.
245
00:11:01,155 --> 00:11:02,885
Like your encryption capabilities
246
00:11:02,885 --> 00:11:04,925
between services are going to vary.
247
00:11:05,555 --> 00:11:09,285
It's a little bit different
over in Office 365 M 365 land.
248
00:11:09,435 --> 00:11:11,205
It's different over in Dynamics land.
249
00:11:11,825 --> 00:11:14,525
You might not necessarily
be able to come to
250
00:11:15,045 --> 00:11:17,325
consensus across all of those things,
251
00:11:17,695 --> 00:11:19,525
especially when it comes to some
252
00:11:19,525 --> 00:11:21,285
of the large language models
you were talking about.
253
00:11:21,355 --> 00:11:24,565
Like those rolling out and
getting to where they need to be.
254
00:11:24,565 --> 00:11:27,085
Like they're just gonna be
baked into experiences in some
255
00:11:27,085 --> 00:11:31,965
places and customers might
not have the level of control
256
00:11:32,235 --> 00:11:34,205
that they necessarily want there upfront.
257
00:11:34,985 --> 00:11:37,445
And that becomes part of
the rationalization exercise
258
00:11:37,465 --> 00:11:39,365
of just trying to figure out like is this
259
00:11:39,585 --> 00:11:40,725
the thing that I want to do?
260
00:11:40,945 --> 00:11:42,445
Is it the right thing
261
00:11:42,945 --> 00:11:46,365
for me at this given place
time? That kind of thing.
262
00:11:46,365 --> 00:11:48,765
- Definitely and maybe it has this
263
00:11:48,865 --> 00:11:52,765
and I have not probably spent
as much time reading these
264
00:11:53,765 --> 00:11:57,765
DPA and legalese as I should,
265
00:11:57,865 --> 00:12:00,925
but the other thing that
would've been nice to include
266
00:12:01,505 --> 00:12:02,845
in either the email
267
00:12:02,865 --> 00:12:07,645
or have a link to it is what
data is actually being sent?
268
00:12:07,955 --> 00:12:10,165
Like is it row for row
269
00:12:10,795 --> 00:12:13,005
data from my tables in Sentinel
270
00:12:13,185 --> 00:12:17,525
or is it things like when
Sentinel detects a brute force
271
00:12:17,585 --> 00:12:21,125
attack, the IP addresses of a
potential brute force attack
272
00:12:21,705 --> 00:12:25,725
and where is it almost
anonymized I guess in sense
273
00:12:25,785 --> 00:12:29,685
or sent a summary of the
data that's coming in versus
274
00:12:30,715 --> 00:12:35,085
this is sending like IP
addresses and usernames
275
00:12:35,425 --> 00:12:37,485
and countries
276
00:12:37,625 --> 00:12:42,485
and I don't know 100% at
least from this email,
277
00:12:43,035 --> 00:12:45,285
what does it mean when this data's being
278
00:12:46,205 --> 00:12:50,085
provided now to the Microsoft
security research in terms
279
00:12:50,085 --> 00:12:53,565
of the level of data or the
details around what that is.
280
00:12:53,625 --> 00:12:54,845
- That's an interesting one.
281
00:12:55,205 --> 00:12:59,325
I actually, I pulled up the DPA
while you were chatting just
282
00:12:59,325 --> 00:13:01,965
to see is there anything specific to
283
00:13:02,865 --> 00:13:04,645
to Sentinel in there two that's called out
284
00:13:04,825 --> 00:13:08,125
and yeah there's not anything in that one.
285
00:13:08,385 --> 00:13:11,165
The other thing it's gotten
kind of harder over time
286
00:13:11,265 --> 00:13:14,245
to find some misinformation
like the DPA a used
287
00:13:14,245 --> 00:13:15,445
to be like an online thing
288
00:13:15,985 --> 00:13:18,285
or one of 'em did the online service terms
289
00:13:18,585 --> 00:13:20,165
and even like SLAs
290
00:13:20,165 --> 00:13:21,965
and things now they're
like turning into like PDFs
291
00:13:21,965 --> 00:13:24,645
and word docs so you have
to like explicitly go out
292
00:13:24,745 --> 00:13:28,805
and download them and
figure all that stuff
293
00:13:28,905 --> 00:13:29,905
- Out.
294
00:13:29,905 --> 00:13:30,245
Yeah it's interesting.
295
00:13:30,245 --> 00:13:34,165
And then this is a conversation
kind of going along side
296
00:13:34,165 --> 00:13:37,685
of this in Discord too is
you mentioned this is Azure,
297
00:13:37,755 --> 00:13:39,445
this also comes into play as
298
00:13:39,675 --> 00:13:41,845
what this would mean potentially
299
00:13:41,905 --> 00:13:43,925
for Microsoft 365 down the road.
300
00:13:44,435 --> 00:13:47,885
This particular case isn't
applicable to Microsoft 365
301
00:13:48,065 --> 00:13:51,685
and Azure the customer managed key
302
00:13:52,215 --> 00:13:56,085
isn't necessarily a expensive paid
303
00:13:56,105 --> 00:13:59,925
for service versus Microsoft
365 I believe if you wanna do
304
00:13:59,925 --> 00:14:03,085
like a customer managed
key and Microsoft 365
305
00:14:03,465 --> 00:14:05,965
but you need to be on an E five license
306
00:14:06,185 --> 00:14:08,525
and I can't remember which E five it is,
307
00:14:08,665 --> 00:14:10,165
if it's the EMSE five
308
00:14:10,225 --> 00:14:12,045
or the Office 365 E five,
309
00:14:13,005 --> 00:14:15,085
Microsoft 365 E five
would definitely cover it
310
00:14:15,205 --> 00:14:16,525
'cause it has both of those in there.
311
00:14:16,945 --> 00:14:18,205
But if I remember right,
312
00:14:18,525 --> 00:14:21,045
customer managed keys in
Microsoft 365 or E five.
313
00:14:21,345 --> 00:14:23,445
So if you ended up in this boat
314
00:14:23,515 --> 00:14:26,885
with say Microsoft 365
copilot as that's coming out
315
00:14:26,885 --> 00:14:29,245
where you got an email
saying you've elected
316
00:14:29,385 --> 00:14:33,605
to not share certain information
with the Microsoft 365
317
00:14:34,285 --> 00:14:36,205
security research and the only way
318
00:14:36,205 --> 00:14:37,645
to block it is customer managed keys.
319
00:14:38,105 --> 00:14:41,445
You could find yourself in a
position where the only way
320
00:14:41,445 --> 00:14:44,765
to stop that would be to upgrade from like
321
00:14:44,765 --> 00:14:46,925
that E three five plan 2 85 plan.
322
00:14:47,345 --> 00:14:49,325
The whole large language model.
323
00:14:50,055 --> 00:14:52,125
Again, to your point, it it's my data
324
00:14:52,425 --> 00:14:54,245
but it isn't Microsoft's data center.
325
00:14:54,515 --> 00:14:55,685
They're hosting it all.
326
00:14:55,945 --> 00:14:58,445
And again I haven't
read all of the legalese
327
00:14:58,445 --> 00:15:00,845
around when they can or
can't do this type of stuff
328
00:15:00,865 --> 00:15:05,125
but it definitely does lead
to some interesting things
329
00:15:05,465 --> 00:15:09,805
to consider as everybody
is now scrambling for data
330
00:15:10,105 --> 00:15:12,165
for these large language models where
331
00:15:12,265 --> 00:15:15,365
before part of my thing was
Google always wanted your data
332
00:15:15,425 --> 00:15:17,205
and I'm gonna call out Google specifically
333
00:15:17,205 --> 00:15:19,125
because they're in the
advertising business.
334
00:15:19,965 --> 00:15:21,725
Facebook or Meta always wanted your data
335
00:15:21,965 --> 00:15:23,405
'cause they're in the
advertising business.
336
00:15:23,475 --> 00:15:26,725
They benefit from getting as
much data as they can about you
337
00:15:27,105 --> 00:15:29,525
to better advertise and market to you.
338
00:15:29,855 --> 00:15:32,085
Where let's face it, that
wasn't Microsoft's business
339
00:15:32,425 --> 00:15:33,645
but now with AI
340
00:15:33,705 --> 00:15:35,525
and everybody kind of battling
341
00:15:35,525 --> 00:15:37,285
for these large language models
342
00:15:37,425 --> 00:15:40,085
and the training of 'em, all of a sudden
343
00:15:40,645 --> 00:15:44,245
I find Microsoft wanting a lot
more data than maybe they did
344
00:15:44,245 --> 00:15:46,645
before so that they
can train these models.
345
00:15:46,815 --> 00:15:50,325
Which definitely leads
to, like someone said,
346
00:15:50,325 --> 00:15:51,965
it's some interesting
days, interesting times
347
00:15:52,185 --> 00:15:54,765
to think about some of
the implications of this
348
00:15:54,765 --> 00:15:57,485
beyond just copilot can
help me with my work. Now
349
00:15:57,685 --> 00:15:59,485
- I hate to be a stickler for language
350
00:15:59,545 --> 00:16:02,205
but I, I kind of would
continue to encourage you to
351
00:16:03,385 --> 00:16:06,365
not think of everything in ML land as
352
00:16:06,885 --> 00:16:08,045
a large language model.
353
00:16:08,465 --> 00:16:11,965
So like large language models
are specific things like in
354
00:16:11,965 --> 00:16:15,005
many cases like in the case
of chat GPT we're dealing
355
00:16:15,005 --> 00:16:16,645
with language transformers.
356
00:16:16,705 --> 00:16:19,205
So these are things, these
are machine learning models
357
00:16:19,795 --> 00:16:22,685
that are built around
natural language processing.
358
00:16:23,145 --> 00:16:26,085
So if you're ever played
around with say like Lewis
359
00:16:26,425 --> 00:16:29,525
and sorry the language
understanding intelligence service
360
00:16:30,065 --> 00:16:33,165
and building out like chatbots in Azure
361
00:16:33,165 --> 00:16:34,605
or something like that, you're dealing
362
00:16:34,605 --> 00:16:35,925
with natural language processing.
363
00:16:36,075 --> 00:16:41,045
There's all sorts of
other ML models that can
364
00:16:41,385 --> 00:16:43,845
and do exist out there.
365
00:16:44,505 --> 00:16:48,565
So that could be things like
you said like maybe maybe I'm a
366
00:16:49,005 --> 00:16:53,525
somebody like Google and I'm
trying to figure out spread of
367
00:16:54,285 --> 00:16:56,245
advertising and cohorts of customers
368
00:16:56,465 --> 00:16:59,365
and so I can give my
advertisers the best experience
369
00:16:59,365 --> 00:17:00,405
where they can target their ads
370
00:17:00,405 --> 00:17:01,965
to people in a certain demographic.
371
00:17:01,995 --> 00:17:06,685
Like I only want to target
25 to 30 year olds that live
372
00:17:07,225 --> 00:17:09,965
in Spain or even a certain region.
373
00:17:10,115 --> 00:17:12,085
Like you can get down to that level
374
00:17:12,085 --> 00:17:13,165
of granularity sometimes
375
00:17:13,195 --> 00:17:14,485
depending on what's going on out there.
376
00:17:14,485 --> 00:17:17,725
And that's very different than
hey I'm gonna throw a large
377
00:17:18,085 --> 00:17:19,405
language model against this thing
378
00:17:19,785 --> 00:17:22,845
or some type of transformer
against it so it can ingest
379
00:17:22,845 --> 00:17:24,725
and do a bunch of natural
language processing.
380
00:17:24,725 --> 00:17:27,805
So I would bet in the case
of something like Sentinel,
381
00:17:28,525 --> 00:17:29,605
'cause Sentinel is a product
382
00:17:29,605 --> 00:17:32,645
that like you can bring your
own ML models to as well.
383
00:17:32,715 --> 00:17:36,085
Like you can do your own
inference on top of it even on top
384
00:17:36,085 --> 00:17:37,445
of like your own custom data
385
00:17:37,755 --> 00:17:40,205
that you potentially bring into Sentinel,
386
00:17:40,385 --> 00:17:42,445
say you're doing like
taxi or things like that
387
00:17:42,505 --> 00:17:44,765
and kind of side curing
your own data in there.
388
00:17:45,065 --> 00:17:46,165
So I would bet
389
00:17:46,165 --> 00:17:48,925
that it's more about all
up signals especially
390
00:17:49,205 --> 00:17:51,685
'cause if you look at the
way things have been going
391
00:17:51,715 --> 00:17:55,645
with security in general like
across the tech companies,
392
00:17:56,045 --> 00:17:57,765
I don't think Microsoft's
any exception here.
393
00:17:58,065 --> 00:18:00,845
The more signals you have the
better off you can be. Yep.
394
00:18:01,185 --> 00:18:04,205
What's the sensitivity like
just a devil's advocate,
395
00:18:04,205 --> 00:18:08,365
like what's the sensitivity
to a signal in Sentinel versus
396
00:18:09,515 --> 00:18:13,845
something in your sign-in
logs in Azure AD like being
397
00:18:14,015 --> 00:18:17,445
anonymized and correlated
across the service, right?
398
00:18:17,885 --> 00:18:19,125
Microsoft always talks a lot about
399
00:18:19,825 --> 00:18:22,045
hey why does security work so well in
400
00:18:22,745 --> 00:18:24,965
ID the artist formerly known as Azure ad?
401
00:18:25,315 --> 00:18:27,925
It's because they have billions of logins
402
00:18:27,985 --> 00:18:31,125
and billions of signals that
they can run that inference on
403
00:18:31,185 --> 00:18:33,565
and figure out the next thing there.
404
00:18:37,305 --> 00:18:38,925
- Do you feel overwhelmed by trying
405
00:18:38,925 --> 00:18:41,005
to manage your Office 365 environment?
406
00:18:41,105 --> 00:18:42,885
Are you facing unexpected issues
407
00:18:42,885 --> 00:18:44,965
that disrupt your company's productivity?
408
00:18:44,995 --> 00:18:47,805
Intelligent is here to help
much like you take your car
409
00:18:47,805 --> 00:18:50,485
to the mechanic that has
specialized knowledge on how
410
00:18:50,485 --> 00:18:53,245
to best keep your car
running Intelligent helps you
411
00:18:53,245 --> 00:18:54,965
with your Microsoft cloud environment
412
00:18:54,965 --> 00:18:56,565
because that's their expertise.
413
00:18:56,675 --> 00:18:58,005
Intelligent keeps up
414
00:18:58,005 --> 00:18:59,845
with the latest updates
in the Microsoft cloud
415
00:18:59,865 --> 00:19:01,845
to help keep your
business running smoothly
416
00:19:01,945 --> 00:19:03,085
and ahead of the curve.
417
00:19:03,115 --> 00:19:05,005
Whether you are a small organization
418
00:19:05,005 --> 00:19:07,485
with just a few users
up to an organization
419
00:19:07,485 --> 00:19:10,765
of several thousand employees
they want to partner with you
420
00:19:10,785 --> 00:19:14,205
to implement and administer
your Microsoft Cloud technology,
421
00:19:14,775 --> 00:19:18,045
visit them at intelligent.com/podcast,
422
00:19:18,425 --> 00:19:23,085
that's I-N-T-E-L-I-G-I-N
423
00:19:23,165 --> 00:19:26,765
k.com/podcast for more information
424
00:19:26,825 --> 00:19:28,405
or to schedule a 30 minute call
425
00:19:28,425 --> 00:19:29,765
to get started with them today.
426
00:19:30,605 --> 00:19:33,245
Remember intelligent focuses
on the Microsoft cloud
427
00:19:33,345 --> 00:19:35,165
so you can focus on your business.
428
00:19:37,645 --> 00:19:40,765
I think that's the part that
I get where to your point,
429
00:19:40,765 --> 00:19:44,285
Sentinel would absolutely,
I was trying to find a tweet
430
00:19:44,285 --> 00:19:46,685
but I think I lost it
in our chats, being able
431
00:19:46,685 --> 00:19:47,685
to just pull those signals
432
00:19:47,685 --> 00:19:50,325
and if they're anonymized
enough, I totally get it.
433
00:19:50,355 --> 00:19:53,525
Like you look at the amount
of signals that Microsoft gets
434
00:19:53,625 --> 00:19:57,685
and I've talked to red team companies
435
00:19:57,685 --> 00:20:01,405
before that are actually
writing viruses trying
436
00:20:01,425 --> 00:20:02,965
to bypass things like
437
00:20:03,685 --> 00:20:05,845
Microsoft Defender from
an antivirus perspective
438
00:20:06,545 --> 00:20:08,485
and they've told me,
439
00:20:08,585 --> 00:20:10,485
and I don't have any validation
440
00:20:10,485 --> 00:20:12,445
for this other than just
word of mouth talking
441
00:20:12,465 --> 00:20:15,565
to these people that Microsoft
Defender antivirus is one
442
00:20:15,565 --> 00:20:19,405
of the hardest antivirus
software to bypass
443
00:20:20,035 --> 00:20:22,085
when they're just writing
a virus from scratch
444
00:20:22,545 --> 00:20:24,805
and that once they are able to write one
445
00:20:25,405 --> 00:20:27,365
Microsoft Defender is also one of the ones
446
00:20:27,435 --> 00:20:29,125
that shuts it down the quickest.
447
00:20:29,125 --> 00:20:31,565
They said I essentially get
one shot at it, I deploy
448
00:20:31,565 --> 00:20:34,605
that virus, I use it, I get
in, I never works again.
449
00:20:35,425 --> 00:20:38,365
And I think some of
that is to your point of
450
00:20:38,885 --> 00:20:40,525
Microsoft is able to take all
451
00:20:40,525 --> 00:20:43,605
of these different signals
from their products
452
00:20:44,145 --> 00:20:48,485
and they use them to
protect other tenants.
453
00:20:48,585 --> 00:20:52,485
So myself, anybody else
listening to this podcast
454
00:20:52,595 --> 00:20:55,805
that uses Microsoft 365 is
benefiting from the signals
455
00:20:55,805 --> 00:20:58,685
that these Microsoft security
products are getting from
456
00:20:58,685 --> 00:21:01,205
people also trying to attack Microsoft.
457
00:21:01,785 --> 00:21:04,405
You look at what Microsoft 365 is, 90%
458
00:21:04,405 --> 00:21:06,565
of the Fortune five hundreds, how many
459
00:21:06,565 --> 00:21:09,045
of them are deploying Defender Sentinel?
460
00:21:09,045 --> 00:21:11,885
Some of those. But there's
signals coming in from those
461
00:21:11,885 --> 00:21:14,925
that Microsoft uses to protect
the small guys as well.
462
00:21:15,105 --> 00:21:19,445
So I also do understand that
if those are anonymized enough,
463
00:21:19,445 --> 00:21:21,125
if they can figure out those signals
464
00:21:21,265 --> 00:21:23,885
and this is where I, maybe you have
465
00:21:24,125 --> 00:21:26,565
to trust Microsoft a
little bit that they're not
466
00:21:27,085 --> 00:21:29,605
grabbing a bunch of PII type of data
467
00:21:29,865 --> 00:21:33,005
or really sensitive internal data
468
00:21:33,225 --> 00:21:34,805
but looking at some of those signals
469
00:21:34,825 --> 00:21:36,285
to really help strengthen that.
470
00:21:36,645 --> 00:21:40,005
I think there absolutely
is a benefit to that.
471
00:21:40,525 --> 00:21:41,605
I think it was a tweet from Merrill
472
00:21:41,605 --> 00:21:43,925
that I saw about the amount of signals
473
00:21:43,925 --> 00:21:45,925
that Microsoft does get and analyze a day
474
00:21:45,945 --> 00:21:49,605
and it's like in the trillions I believe
475
00:21:49,825 --> 00:21:53,805
of signals a day that
Microsoft gets around
476
00:21:54,345 --> 00:21:58,965
people trying to breach the
Microsoft products essentially.
477
00:21:59,025 --> 00:22:01,165
Or maybe it was even
specific to Entra. It's
478
00:22:01,445 --> 00:22:02,445
- A ton of data, right?
479
00:22:02,445 --> 00:22:04,805
If you consider like
particularly in the case of intra
480
00:22:04,945 --> 00:22:09,605
and login data 'cause
Microsoft is able to aggregate
481
00:22:09,625 --> 00:22:13,085
and correlate signals across
the public service as well.
482
00:22:13,425 --> 00:22:16,805
So like your endpoint
when you log to enterra
483
00:22:17,535 --> 00:22:21,045
login do microsoft online.com,
like that's the same thing
484
00:22:21,045 --> 00:22:25,005
that MSAs and all that use
and all that use as well.
485
00:22:25,565 --> 00:22:27,565
I don't know if you
have any options here so
486
00:22:28,075 --> 00:22:29,725
I'll put a link in the show notes.
487
00:22:29,825 --> 00:22:31,805
So one thing that does come up from time
488
00:22:31,805 --> 00:22:34,685
to time is handling
personal data in things
489
00:22:34,715 --> 00:22:35,805
like log analytics.
490
00:22:35,995 --> 00:22:38,085
Like I haven't seen specific
guidance for Sentinel
491
00:22:38,085 --> 00:22:40,565
and I've been kind of like
binging Duck Tuck going
492
00:22:40,585 --> 00:22:42,165
and then googling in the background here.
493
00:22:42,365 --> 00:22:43,405
uh, I couldn't find much
494
00:22:43,425 --> 00:22:46,685
but there is specific guidance
out there for log analytics
495
00:22:46,985 --> 00:22:48,485
and app insights customers.
496
00:22:48,665 --> 00:22:50,645
So maybe you want to take a
look through that as well.
497
00:22:50,685 --> 00:22:52,765
I don't know if there's things
that can potentially like
498
00:22:52,835 --> 00:22:55,245
make it a little bit better
by potentially getting
499
00:22:55,245 --> 00:22:58,205
that data over to a different
log analytics workspace once
500
00:22:58,205 --> 00:23:00,525
you're done with analysis in sentinels
501
00:23:00,525 --> 00:23:02,005
so you can still retain it for yourself.
502
00:23:02,105 --> 00:23:05,925
Are there potential retention
things you wanna do At the end
503
00:23:05,925 --> 00:23:08,085
of the day the data's
the data though, right?
504
00:23:08,145 --> 00:23:09,165
So you've gotta kind
505
00:23:09,165 --> 00:23:11,645
of figure out your path
to navigate through there.
506
00:23:11,645 --> 00:23:14,645
Whether that's saying well
sentinels not the service
507
00:23:14,745 --> 00:23:18,485
for me given a set of
constraints that exist
508
00:23:19,065 --> 00:23:21,725
or it it is and I'm gonna live with it
509
00:23:21,865 --> 00:23:24,685
or sentinel's not the thing
for me in its current form.
510
00:23:24,905 --> 00:23:26,045
So I need to do something
511
00:23:26,045 --> 00:23:28,045
and potentially upgrade to some form
512
00:23:28,045 --> 00:23:30,405
of my own encryption key
like customer managed key
513
00:23:30,905 --> 00:23:34,365
and then live with any
licensing implications that come
514
00:23:34,365 --> 00:23:38,325
with it to have that associated
level of privacy. Yeah
515
00:23:38,605 --> 00:23:40,965
- I found the tweet, it
was not about signals,
516
00:23:41,075 --> 00:23:43,605
this was just about
password attacks per second
517
00:23:44,275 --> 00:23:48,445
blocked 4,000 password
attacks per second this year.
518
00:23:48,675 --> 00:23:52,645
Like that's just an insane
amount of data coming in
519
00:23:52,645 --> 00:23:54,325
because it's not just
that they blocked 'em,
520
00:23:54,345 --> 00:23:55,965
I'm sure they were looking
where they're coming
521
00:23:55,995 --> 00:23:57,245
from, all of that.
522
00:23:57,465 --> 00:24:01,485
But any who, I'll have to
go look through that article
523
00:24:01,545 --> 00:24:06,045
as well and if we want to make
any changes to log analytics
524
00:24:06,585 --> 00:24:08,485
and how that data's in there, the
525
00:24:08,485 --> 00:24:10,285
- Nice thing on your side is that you work
526
00:24:10,285 --> 00:24:13,405
with enough customers that
it's also like the all learning
527
00:24:13,405 --> 00:24:16,205
that you can carry
forward for them as well
528
00:24:16,705 --> 00:24:20,005
and say okay here's
potential considerations.
529
00:24:20,625 --> 00:24:22,845
We might wanna think about
this as we're turning it on.
530
00:24:23,225 --> 00:24:25,685
So you can kind of front
load the conversation
531
00:24:25,995 --> 00:24:30,125
with new customers in
this like crazy always
532
00:24:30,275 --> 00:24:31,725
ever changing landscape.
533
00:24:32,075 --> 00:24:35,085
- Exactly. So interesting scenario.
534
00:24:35,305 --> 00:24:36,525
Do you have other, you had
535
00:24:36,525 --> 00:24:37,725
something else you wanna talk about today?
536
00:24:38,065 --> 00:24:39,325
We spent a long time on that
537
00:24:39,665 --> 00:24:41,925
but let's dive into your topic. We
538
00:24:41,925 --> 00:24:45,525
- Did, so mine's a little bit
of a a logging thing as well.
539
00:24:46,065 --> 00:24:47,845
So this is one that came up
540
00:24:47,845 --> 00:24:51,125
with in in a customer
conversation with me recently
541
00:24:51,225 --> 00:24:54,845
and I think it impacts a lot
542
00:24:54,845 --> 00:24:57,205
of customers in Azure or
543
00:24:57,205 --> 00:25:00,125
or there's the potential for
customers to ask this question.
544
00:25:00,165 --> 00:25:01,725
I don't know, it's like
a big impacting issue
545
00:25:01,785 --> 00:25:05,645
but so have you ever been in
the boat where you're working
546
00:25:05,835 --> 00:25:08,205
with somebody and you sit there
547
00:25:08,425 --> 00:25:09,845
and you do a bunch of deployments
548
00:25:10,065 --> 00:25:11,845
and you kind of turn stuff on
549
00:25:11,845 --> 00:25:13,765
and then you walk away
from it for a while
550
00:25:13,865 --> 00:25:16,085
and then you move on to the next thing
551
00:25:16,345 --> 00:25:18,085
and then the next person
comes in behind you
552
00:25:18,085 --> 00:25:20,525
and they say, huh, what is this thing?
553
00:25:20,825 --> 00:25:22,925
Who created it? Do I
still need it anymore?
554
00:25:23,235 --> 00:25:26,045
Does it need to be on, can I delete it?
555
00:25:26,745 --> 00:25:29,005
And it's one of those
things that you came up I,
556
00:25:29,125 --> 00:25:31,365
I was talking with a customer
funny enough about some
557
00:25:31,365 --> 00:25:33,165
storage accounts and they're
like, Hey I have all these,
558
00:25:33,435 --> 00:25:35,605
they had a couple hundred
different accounts out there
559
00:25:35,985 --> 00:25:37,325
and they're like, can I delete them?
560
00:25:37,595 --> 00:25:39,765
Like well do they have activity ?
561
00:25:40,115 --> 00:25:41,765
Have you checked just the metrics
562
00:25:41,985 --> 00:25:45,125
to see if you're driving
any transactions in there?
563
00:25:45,825 --> 00:25:48,845
Oh no I haven't done that. All
right, well let's go do that.
564
00:25:49,015 --> 00:25:50,685
Let's spin up an easy mod workbook
565
00:25:50,825 --> 00:25:52,325
and uh, Azure monitor workbook
566
00:25:52,325 --> 00:25:53,765
and we'll look at some of this stuff and
567
00:25:53,825 --> 00:25:55,045
and see what's going on.
568
00:25:55,425 --> 00:25:57,285
Oh I see these 10 accounts over here
569
00:25:57,335 --> 00:25:58,645
don't have any activity.
570
00:25:58,715 --> 00:26:00,165
They go can I delete 'em?
571
00:26:00,405 --> 00:26:03,165
I don't know what, what data's
in there. Well I don't know.
572
00:26:03,335 --> 00:26:06,845
Maybe I should go and talk
to the person who manages it.
573
00:26:06,985 --> 00:26:08,325
So the crux of the question is
574
00:26:08,625 --> 00:26:10,565
who actually manages your re Yeah,
575
00:26:10,565 --> 00:26:12,525
who manages resources in Azure
576
00:26:13,025 --> 00:26:14,565
and how do you go back in time
577
00:26:15,065 --> 00:26:17,765
and figure that out for resources that
578
00:26:18,595 --> 00:26:23,285
have potentially existed for
years and years and years?
579
00:26:23,645 --> 00:26:25,845
- A hundred percent. And
it's funny you bring this up
580
00:26:26,005 --> 00:26:28,885
'cause I had a customer
doing this exercise as well
581
00:26:28,885 --> 00:26:32,765
that I'm supposed to actually
go do some work trying to sort
582
00:26:32,765 --> 00:26:36,725
through resources and
figure some of this out
583
00:26:36,725 --> 00:26:37,805
because they're in the boat
584
00:26:37,805 --> 00:26:39,485
where they have everything
in five subscriptions
585
00:26:39,625 --> 00:26:42,245
and we need to break it
out into 14 subscriptions
586
00:26:42,505 --> 00:26:45,005
but we don't wanna move stuff
that's not being used anymore.
587
00:26:45,325 --> 00:26:46,845
I mean this is not an insignificant
588
00:26:46,845 --> 00:26:47,965
amount of resources either.
589
00:26:48,015 --> 00:26:50,245
We're talking like thousands of resources.
590
00:26:50,715 --> 00:26:52,085
It's not quite, I don't think it's
591
00:26:52,085 --> 00:26:53,205
up in the tens of thousands.
592
00:26:53,425 --> 00:26:55,445
It might be definitely
not hundreds of thousands
593
00:26:55,465 --> 00:26:58,005
but to your point it's
okay we have these five
594
00:26:58,285 --> 00:27:01,605
subscriptions, thousands and
thousands of resources first
595
00:27:01,985 --> 00:27:03,765
before even who owns them
596
00:27:03,785 --> 00:27:07,365
or who would know is how
do we even go through
597
00:27:07,365 --> 00:27:10,525
and figure out which of these
resources are still used,
598
00:27:10,525 --> 00:27:13,285
which can be deleted,
which maybe can we archive?
599
00:27:13,905 --> 00:27:17,765
Are there resources that
we could change the SKU on,
600
00:27:17,835 --> 00:27:20,565
have SKUs changed so that we
can get some price savings
601
00:27:20,745 --> 00:27:22,285
by making some changes to 'em?
602
00:27:22,585 --> 00:27:25,765
And to your point then,
who actually is in charge
603
00:27:25,765 --> 00:27:27,005
of making those decisions?
604
00:27:27,005 --> 00:27:30,405
Because some of these are
production services, some
605
00:27:30,405 --> 00:27:32,605
of those are dev, some of those are test
606
00:27:32,785 --> 00:27:34,205
and what's going on.
607
00:27:34,385 --> 00:27:37,845
So since I have that question,
what's the answer Scott? I
608
00:27:37,845 --> 00:27:39,365
- Had a couple of thoughts here
609
00:27:39,585 --> 00:27:42,245
and some of it is we're
eventually we're gonna hit a
610
00:27:42,245 --> 00:27:44,125
roadblock and we can't go
back in time all the way.
611
00:27:44,425 --> 00:27:48,285
So first thing is, is this resource being
612
00:27:48,805 --> 00:27:50,045
actively used right now?
613
00:27:50,805 --> 00:27:52,845
I think one of the best ways to do that is
614
00:27:53,655 --> 00:27:56,245
start over in Azure Monitor.
615
00:27:56,745 --> 00:28:00,565
So I often refer to that
in shorthand as azon
616
00:28:01,345 --> 00:28:05,685
and over in Azon there's a
bunch of built-in workbooks
617
00:28:05,945 --> 00:28:08,405
around kind of core service metrics.
618
00:28:08,505 --> 00:28:10,085
So if you're dealing with a core service,
619
00:28:10,505 --> 00:28:14,245
so I'm thinking things that
fall into kind of compute,
620
00:28:14,245 --> 00:28:17,925
networking, storage, many of
the data analytics services,
621
00:28:17,925 --> 00:28:20,165
things like that, they all have a set
622
00:28:20,165 --> 00:28:24,365
of default metrics which
often drive back to things
623
00:28:25,035 --> 00:28:27,685
like transactions in the service, right?
624
00:28:27,685 --> 00:28:30,165
If I'm dealing with a storage
account, am I seeing puts
625
00:28:30,165 --> 00:28:31,965
and gets actively to that storage account?
626
00:28:32,265 --> 00:28:36,085
And if I am, I I I know
somebody out there is using it
627
00:28:36,085 --> 00:28:39,045
and potentially that helps
me take the next step
628
00:28:39,305 --> 00:28:40,805
to go and answer the question.
629
00:28:41,035 --> 00:28:44,445
Okay, now I know it's on who is using it
630
00:28:44,665 --> 00:28:47,605
and who is using it can go
a couple different ways.
631
00:28:47,985 --> 00:28:50,925
If I'm dealing say with if virtual machine
632
00:28:51,305 --> 00:28:55,245
and who is using it, I might want to go
633
00:28:55,585 --> 00:28:58,245
and just check out the
activity log for that VM
634
00:28:58,465 --> 00:29:00,085
or spit out some diagnostics
635
00:29:00,085 --> 00:29:01,525
and do something like log analytics
636
00:29:01,545 --> 00:29:05,485
or a storage account, some
service like that and take a look
637
00:29:05,485 --> 00:29:07,485
and see okay great, I know
this thing's powered up
638
00:29:07,485 --> 00:29:09,805
and it's on am I seeing
active logins to it?
639
00:29:10,185 --> 00:29:14,085
Can I actually infer
anything from a login name
640
00:29:14,225 --> 00:29:16,245
or like we talked about baston last week,
641
00:29:16,445 --> 00:29:18,245
a bastion connection, things like that.
642
00:29:18,865 --> 00:29:21,005
If it's a storage account I have resource
643
00:29:21,115 --> 00:29:22,125
logs that I can turn on.
644
00:29:22,555 --> 00:29:25,205
Most customers tend to
complain once we get down
645
00:29:25,205 --> 00:29:26,245
to this step
646
00:29:26,245 --> 00:29:28,325
where we're saying like well
go look at like the fine
647
00:29:28,325 --> 00:29:30,245
grained resource logs for the thing.
648
00:29:30,805 --> 00:29:32,645
'cause they go Oh well
I gotta turn that on,
649
00:29:32,685 --> 00:29:34,125
I gotta configure it and it costs me money
650
00:29:34,125 --> 00:29:35,125
for the logs .
651
00:29:35,125 --> 00:29:36,405
Yeah I get it. Time is money
652
00:29:36,665 --> 00:29:38,845
and turns out when you use additional
653
00:29:39,045 --> 00:29:40,365
resources, those cost money as well.
654
00:29:40,615 --> 00:29:41,805
We're not, not talking about keeping
655
00:29:41,805 --> 00:29:43,045
this stuff on forever, right?
656
00:29:43,105 --> 00:29:46,005
You could even turn on
diagnostic logs on a service for
657
00:29:46,785 --> 00:29:48,645
10 minutes, whatever a period of time is
658
00:29:48,785 --> 00:29:50,925
and just kind of see
what you get in there.
659
00:29:50,945 --> 00:29:54,045
It doesn't need to be like
tens of thousands of dollars
660
00:29:54,385 --> 00:29:57,285
or hundreds or even tens
of dollars to to spend
661
00:29:57,285 --> 00:29:59,245
that stuff up and turn it on
662
00:29:59,985 --> 00:30:01,325
if none of that gets you anywhere.
663
00:30:01,945 --> 00:30:06,365
The next place that I
would kind of go off to is
664
00:30:06,945 --> 00:30:09,125
the Azure activity log.
665
00:30:09,705 --> 00:30:14,645
So by default at the
subscription level there is a set
666
00:30:14,645 --> 00:30:17,805
of information that's
retained when management
667
00:30:17,965 --> 00:30:19,325
plane operations happen.
668
00:30:19,625 --> 00:30:22,725
So basically when you like
fire off a request against the
669
00:30:22,725 --> 00:30:25,565
arm APIs, which the portal
is doing all the time,
670
00:30:25,755 --> 00:30:28,205
like a create event for a
resource, an update event
671
00:30:28,205 --> 00:30:29,325
for a resource, things like that.
672
00:30:29,815 --> 00:30:33,445
Those will all be in
the activity log there.
673
00:30:33,445 --> 00:30:34,605
There's some limitations there.
674
00:30:34,995 --> 00:30:37,725
Like the activity log
only logs three months
675
00:30:37,725 --> 00:30:38,845
of data by default.
676
00:30:38,945 --> 00:30:40,165
Unless you've gone ahead
677
00:30:40,165 --> 00:30:42,845
and configured like
diagnostic export for that
678
00:30:43,145 --> 00:30:45,605
and potentially sent it
out to another place.
679
00:30:45,785 --> 00:30:48,285
But I look at that as a little
bit of a learning exercise.
680
00:30:48,705 --> 00:30:50,525
Hey here's another
opportunity for you to go
681
00:30:50,525 --> 00:30:52,125
through if this is something
that's important for you
682
00:30:52,125 --> 00:30:54,405
to audit in your environment,
you really might want
683
00:30:54,405 --> 00:30:57,405
to think about turning
on activity log exporting
684
00:30:57,585 --> 00:31:00,605
for the actual Azure monitor activity log
685
00:31:00,945 --> 00:31:02,205
and getting that data out
686
00:31:02,225 --> 00:31:05,325
to something like a log
analytics workspace so
687
00:31:05,395 --> 00:31:07,925
that you can have a richer interaction
688
00:31:08,185 --> 00:31:11,485
to go back in time over
that resource. I would
689
00:31:11,485 --> 00:31:12,485
- Agree.
690
00:31:12,485 --> 00:31:13,805
And that's kind of the
approach we've taken
691
00:31:13,945 --> 00:31:16,645
as like you said it's log analytics,
692
00:31:16,835 --> 00:31:19,285
it's Azure activity logs,
it's really just going
693
00:31:19,285 --> 00:31:22,085
through all the logging and
looking at what's there.
694
00:31:22,555 --> 00:31:23,965
This is gonna help with the activity.
695
00:31:24,485 --> 00:31:26,925
I don't know that it
always helps with the owner
696
00:31:27,445 --> 00:31:28,525
and I think this is a,
697
00:31:28,665 --> 00:31:32,765
- So the nice thing is with
activity often comes identity
698
00:31:32,995 --> 00:31:34,925
because users have to authenticate
699
00:31:34,925 --> 00:31:36,645
to these things like hey we have
700
00:31:36,645 --> 00:31:37,965
to figure out if you're authorized and
701
00:31:37,965 --> 00:31:38,885
before we figure out if you're
702
00:31:38,885 --> 00:31:40,125
authorized we need to know who you are.
703
00:31:40,125 --> 00:31:43,325
Yep. To do it. So lots of that
stuff is just happening in
704
00:31:43,425 --> 00:31:44,445
inside of OAuth
705
00:31:44,585 --> 00:31:47,365
and it's all authenticated
against Azure AD anyway.
706
00:31:47,665 --> 00:31:51,285
So AD potentially becomes
a source of logs for you.
707
00:31:51,355 --> 00:31:53,485
Okay, am I seeing sign in logs like
708
00:31:53,485 --> 00:31:55,165
during a a given period given time
709
00:31:55,825 --> 00:31:59,165
but as long as you have the
log like it'll have a goid
710
00:31:59,185 --> 00:32:02,205
or something else in it
that ties you back to
711
00:32:02,725 --> 00:32:06,805
a user identity over in
Azure active directory which
712
00:32:07,325 --> 00:32:09,485
ultimately like in inferences and
713
00:32:09,585 --> 00:32:13,485
and gets you down to ownership
Now all this stuff only helps
714
00:32:13,485 --> 00:32:15,245
you like if you have the data there.
715
00:32:15,705 --> 00:32:18,045
The other thing that I had mentioned
716
00:32:18,225 --> 00:32:20,445
and it came up in the
chat over here on Discord
717
00:32:21,145 --> 00:32:22,205
was there's other things
718
00:32:22,205 --> 00:32:23,765
that you should just think
about maybe from like a
719
00:32:23,765 --> 00:32:26,645
governance perspective, like
there's some best practices
720
00:32:26,665 --> 00:32:29,525
around things like tagging your resources,
721
00:32:30,225 --> 00:32:32,005
be it either tagging a resource group
722
00:32:32,545 --> 00:32:35,805
or tagging individual resources
in resource groups in Azure.
723
00:32:36,075 --> 00:32:38,285
Like you might wanna tag those with a,
724
00:32:38,455 --> 00:32:39,885
maybe not an individual user
725
00:32:40,065 --> 00:32:42,645
but like a responsible group for it.
726
00:32:43,225 --> 00:32:45,365
Or quite often like I encourage a lot
727
00:32:45,365 --> 00:32:47,245
of customers in like
multi-home departments,
728
00:32:47,355 --> 00:32:48,765
like they run multiple departments,
729
00:32:48,765 --> 00:32:50,405
they wanna do chargebacks
and stuff like that.
730
00:32:50,705 --> 00:32:52,365
Hey let's rationalize not just the
731
00:32:52,365 --> 00:32:53,445
environment this thing sits in.
732
00:32:53,545 --> 00:32:56,085
Is it prod, is it dev test, is it staging?
733
00:32:56,745 --> 00:32:58,485
But let's actually tie this back
734
00:32:58,485 --> 00:33:00,565
to something like an IO code
735
00:33:00,905 --> 00:33:04,765
or some type of like internal
finance code that you have
736
00:33:04,825 --> 00:33:07,485
or even like just a department,
does this belong to finance?
737
00:33:07,675 --> 00:33:09,605
Does it belong to you Azure?
738
00:33:09,915 --> 00:33:12,805
Does it belong to finance,
hr, anything like that. Yeah,
739
00:33:12,965 --> 00:33:15,165
- I was gonna bring that up as well.
740
00:33:15,165 --> 00:33:16,605
Going through this exercise
741
00:33:16,705 --> 00:33:18,925
and what we talked about
logs helps you maybe find
742
00:33:19,035 --> 00:33:20,405
what happened before.
743
00:33:20,795 --> 00:33:23,605
This is also a good opportunity
I think to go through some
744
00:33:23,605 --> 00:33:26,485
of what you just said is
maybe this is a good time
745
00:33:26,485 --> 00:33:28,645
to start thinking about tagging
746
00:33:28,645 --> 00:33:31,645
and even using Azure
policies to enforce that
747
00:33:32,605 --> 00:33:35,245
a certain resource is tagged with a owner.
748
00:33:35,345 --> 00:33:39,045
And again whether that be a
person, a department tagged with
749
00:33:39,045 --> 00:33:41,245
what type of resource
this is this something
750
00:33:41,245 --> 00:33:43,165
that's getting stood up for test or dev
751
00:33:43,185 --> 00:33:45,765
or is this a production level resource?
752
00:33:46,155 --> 00:33:50,205
I've seen customers tag dates
of when this was created,
753
00:33:50,825 --> 00:33:54,005
but thinking through maybe some
of that additional metadata
754
00:33:54,005 --> 00:33:55,605
that you should start adding
755
00:33:55,625 --> 00:33:58,845
or do wanna be adding onto
your Azure resources so that
756
00:33:59,065 --> 00:34:01,325
as you encounter this scenario again
757
00:34:01,345 --> 00:34:03,565
or as you move down the road
758
00:34:03,585 --> 00:34:05,405
and you wanna six months from now,
759
00:34:05,565 --> 00:34:08,045
a year from now do a regular type
760
00:34:08,045 --> 00:34:09,405
of audit of what's out there.
761
00:34:09,545 --> 00:34:11,445
Do we still need it?
Is it still being used?
762
00:34:11,985 --> 00:34:16,005
Having some of those tags
can also be very beneficial
763
00:34:16,305 --> 00:34:18,045
but you may not have them
the first time you do.
764
00:34:18,045 --> 00:34:19,045
It might be something
765
00:34:19,045 --> 00:34:20,925
that says maybe we
should start doing this.
766
00:34:21,165 --> 00:34:23,445
- I think it's one of those
things like when you run into it
767
00:34:23,445 --> 00:34:24,725
you're like oh maybe I should start
768
00:34:24,725 --> 00:34:26,005
doing this now kind of thing.
769
00:34:26,025 --> 00:34:28,525
Yep. I see a lot of customers
when it comes to tags
770
00:34:28,715 --> 00:34:33,365
that they tend to just
get stuck in the rutt
771
00:34:33,385 --> 00:34:36,645
of trying to rationalize
like all the options.
772
00:34:36,965 --> 00:34:39,485
'cause it's basically
just name value pairs
773
00:34:39,745 --> 00:34:41,605
and you can be right,
774
00:34:41,715 --> 00:34:43,765
very freeform in your name
values, things like that.
775
00:34:43,835 --> 00:34:46,405
There's actually some decent
guidance out there in the cloud
776
00:34:46,645 --> 00:34:49,565
adoption framework about just
like getting started with tags
777
00:34:49,565 --> 00:34:51,365
and like defining your strategy
778
00:34:51,415 --> 00:34:52,965
where it calls out a lot of these things.
779
00:34:53,275 --> 00:34:56,645
Okay, you might want to
tag like the uh, department
780
00:34:56,645 --> 00:34:59,085
that owns the resource you
might wanna tag like the
781
00:34:59,135 --> 00:35:00,885
operations team that owns it.
782
00:35:01,245 --> 00:35:02,845
I think that's very
helpful like in the context
783
00:35:02,985 --> 00:35:05,205
of even if I don't have
the individual owner,
784
00:35:05,355 --> 00:35:07,165
hopefully the people who operate the
785
00:35:07,285 --> 00:35:08,405
thing have a contact right?
786
00:35:08,405 --> 00:35:09,405
And they can get back to it
787
00:35:09,405 --> 00:35:12,645
and do it cost codes as well, like whether
788
00:35:12,645 --> 00:35:14,205
that's like an IO code
789
00:35:14,205 --> 00:35:16,565
or so like you know whatever
it is that ties you back
790
00:35:16,565 --> 00:35:18,965
to a cost center within your
791
00:35:18,965 --> 00:35:20,485
organization is also super helpful.
792
00:35:20,585 --> 00:35:21,605
Not just for like chargeback
793
00:35:21,605 --> 00:35:24,525
but for hey where's the
throat that I need to go choke
794
00:35:24,665 --> 00:35:27,365
to figure out the answer to this thing.
795
00:35:27,875 --> 00:35:29,805
I've also seen customers that do things
796
00:35:30,305 --> 00:35:32,005
and it is in the adoption framework
797
00:35:32,005 --> 00:35:35,365
where they tag things like
the owner, the requester
798
00:35:35,365 --> 00:35:37,165
and they actually put
email addresses in there.
799
00:35:37,665 --> 00:35:39,605
I'm kind of sensitive
to that stuff sometimes.
800
00:35:39,805 --> 00:35:42,325
I don't know that you should
'cause then you're pumping like
801
00:35:42,445 --> 00:35:44,125
PII out potentially into things
802
00:35:44,665 --> 00:35:46,445
and then we're back to that whole
803
00:35:46,635 --> 00:35:48,205
what does Microsoft do with my data?
804
00:35:48,355 --> 00:35:51,725
Like did I pump those activity
logs out to log analytics?
805
00:35:51,865 --> 00:35:53,525
Are there like Sam a company
806
00:35:53,625 --> 00:35:54,765
and I'm doing business in Europe.
807
00:35:54,865 --> 00:35:57,485
Are there GDPR implications for me or
808
00:35:57,545 --> 00:35:58,845
or things like that there.
809
00:35:58,945 --> 00:36:01,845
So your mileage may vary in like how far
810
00:36:01,945 --> 00:36:03,445
and how granular you wanna go,
811
00:36:03,825 --> 00:36:05,765
but I think it's definitely
a good idea to sit down
812
00:36:05,825 --> 00:36:08,685
and just do that initial rationalization.
813
00:36:09,105 --> 00:36:10,525
And if you haven't done it today
814
00:36:10,545 --> 00:36:12,965
and you're just onboarding, hey
now's a great time to do it.
815
00:36:13,385 --> 00:36:14,405
If you haven't done it
816
00:36:14,405 --> 00:36:15,325
and you're already onboarded,
817
00:36:15,625 --> 00:36:17,005
hey now's a great time to do it.
818
00:36:17,325 --> 00:36:18,445
- . Sounds good. I don't think
819
00:36:18,445 --> 00:36:19,525
I have anything else on that topic.
820
00:36:20,045 --> 00:36:21,685
Anything else you wanna cover on tagging,
821
00:36:21,965 --> 00:36:24,365
cleaning up Azure resources,
deleting everything.
822
00:36:24,445 --> 00:36:26,725
I mean there's always the
approach of you delete it all
823
00:36:26,725 --> 00:36:27,885
and see who screams, right?
824
00:36:28,115 --> 00:36:29,885
Make sure you backups on delete it
825
00:36:29,945 --> 00:36:31,965
and whoever screams is the owner there
826
00:36:31,965 --> 00:36:32,965
- Is that as well?
827
00:36:32,965 --> 00:36:34,925
Yeah, to just take it and
828
00:36:35,025 --> 00:36:37,205
and turn this thing over a different
829
00:36:37,205 --> 00:36:38,325
way and see what happens.
830
00:36:39,065 --> 00:36:41,805
And you might have some
options there if you're dealing
831
00:36:41,835 --> 00:36:44,565
with a website, like you
might not have to delete it,
832
00:36:44,565 --> 00:36:46,285
you can just shut it off or a database.
833
00:36:46,545 --> 00:36:49,205
That's certainly a way to
do those kinds of things.
834
00:36:49,275 --> 00:36:51,325
Just make sure you
leave it off long enough
835
00:36:51,915 --> 00:36:54,325
that you give people
enough time to scream.
836
00:36:54,395 --> 00:36:58,085
Like I've seen people do
things like shut off a database
837
00:36:58,545 --> 00:37:00,405
and they leave it off for a day or two
838
00:37:00,405 --> 00:37:01,285
and they're like, well nobody
839
00:37:01,285 --> 00:37:02,525
said anything so I'll delete it.
840
00:37:02,525 --> 00:37:04,965
But it turns out that database
was only used on like month
841
00:37:04,975 --> 00:37:07,125
close or like closing out
a quarter, even the end
842
00:37:07,125 --> 00:37:10,925
of the year and three months
later they get, they get
843
00:37:10,925 --> 00:37:11,725
that pinging that says,
844
00:37:11,725 --> 00:37:12,885
oh hey, what happened to blah blah blah.
845
00:37:12,885 --> 00:37:14,365
And you're like, oh crap, I deleted that.
846
00:37:14,365 --> 00:37:15,565
Nobody ever talked to me.
847
00:37:15,625 --> 00:37:17,005
So yeah, you do kind of have
848
00:37:17,005 --> 00:37:18,525
to think about those things as well
849
00:37:18,625 --> 00:37:19,625
- For sure.
850
00:37:19,625 --> 00:37:20,525
Awesome. Well thanks Scott.
851
00:37:20,825 --> 00:37:23,485
It was interesting discussion today.
852
00:37:24,005 --> 00:37:25,205
- I got one more for you. Yes.
853
00:37:25,205 --> 00:37:29,965
If anybody stuck around this
long, we are back in November.
854
00:37:30,225 --> 00:37:32,965
So last November we did
this thing where we tried
855
00:37:32,985 --> 00:37:35,645
to raise some money for Girls Who code.
856
00:37:35,745 --> 00:37:38,205
- Oh yeah.
- And we had uh, a bunch
857
00:37:38,205 --> 00:37:42,085
of awesome listeners donate
and we raised $1,500 this year.
858
00:37:42,085 --> 00:37:45,405
I'd like to see if we can
raise 2000 and get out there
859
00:37:45,985 --> 00:37:49,725
and help the next generation of IT pros
860
00:37:49,825 --> 00:37:51,725
or developers, whatever that happens to be
861
00:37:52,035 --> 00:37:53,525
with Girls who code.
862
00:37:53,665 --> 00:37:55,405
So I'm gonna be harping on this one
863
00:37:55,505 --> 00:37:57,885
for the next couple weeks just
864
00:37:57,885 --> 00:37:58,885
to see if we can get people
865
00:37:58,885 --> 00:38:00,445
going during this season of giving. Yes.
866
00:38:00,585 --> 00:38:01,725
- So we should do that. Should we
867
00:38:01,725 --> 00:38:02,925
extend this longer than November?
868
00:38:02,925 --> 00:38:07,245
Because we are also recording
with some of our schedule
869
00:38:07,625 --> 00:38:09,325
for those that are still sticking around.
870
00:38:09,945 --> 00:38:12,645
You actually won't hear
this until November
871
00:38:13,475 --> 00:38:16,125
16 if I'm doing my date math right. So
872
00:38:16,125 --> 00:38:18,645
- Here's the thing, it's
open all year , so
873
00:38:18,645 --> 00:38:19,965
- Should, that's true. You
874
00:38:19,965 --> 00:38:21,565
- Don't have to wait and come in. But we
875
00:38:21,565 --> 00:38:23,045
- Need to set a sense of urgency.
876
00:38:23,105 --> 00:38:26,165
We need to set a time like
we wanna get it by this time.
877
00:38:26,385 --> 00:38:28,205
Should we see if end of the year,
878
00:38:28,345 --> 00:38:29,845
should we go through the end of the year?
879
00:38:30,085 --> 00:38:31,885
- I forgot about the lag in release.
880
00:38:31,885 --> 00:38:33,845
Yeah, we'll go to the end of
the year. That works for me.
881
00:38:34,025 --> 00:38:37,285
- See if we can get
$2,000 by the end of 2023.
882
00:38:37,585 --> 00:38:41,485
By the time people first hear
this, that gives you a month
883
00:38:41,485 --> 00:38:45,005
and a half through the
Thanksgiving holiday season. Yep.
884
00:38:45,145 --> 00:38:47,005
- Should you celebrate, but yes,
885
00:38:47,005 --> 00:38:48,045
during November, December,
886
00:38:48,325 --> 00:38:49,405
whatever that happens to be. So yes,
887
00:38:49,405 --> 00:38:50,405
- That's true.
888
00:38:50,405 --> 00:38:51,645
Thanksgiving, I was talking
to my kids about that.
889
00:38:51,675 --> 00:38:55,125
They were like, what holidays
are worldwide versus just the
890
00:38:55,185 --> 00:38:59,085
us And there's really only a
couple I can think of that are,
891
00:38:59,515 --> 00:39:01,765
well there's no, there's a handful,
892
00:39:02,225 --> 00:39:04,525
but there's a lot of them
that are just localized
893
00:39:04,945 --> 00:39:06,245
to countries.
894
00:39:06,245 --> 00:39:09,005
Thanksgiving being one
of those. Yes. Yeah.
895
00:39:09,005 --> 00:39:11,765
Anyways, we do not need to
go on the holiday rabbit hole
896
00:39:11,965 --> 00:39:14,445
today, but through the end of 2023,
897
00:39:14,545 --> 00:39:16,885
$2,000 for Girls Who Code. Sounds good.
898
00:39:17,105 --> 00:39:18,205
- All right, let's do it.
899
00:39:18,205 --> 00:39:19,645
Alright, thanks as always for the time.
900
00:39:19,865 --> 00:39:23,685
- Yes, thank you. And we will
talk to you again next week.
901
00:39:23,905 --> 00:39:25,205
- All right, perfect. Thanks Ben. Yep.
902
00:39:25,205 --> 00:39:29,125
- Bye-Bye Scott. If you
enjoyed the podcast,
903
00:39:29,745 --> 00:39:31,845
go leave us a five star rating in iTunes.
904
00:39:32,025 --> 00:39:33,365
It helps to get the word out
905
00:39:33,385 --> 00:39:36,925
so more IT Pros can learn
about Office 365 and Azure.
906
00:39:37,665 --> 00:39:40,205
If you have any questions you
want us to address on the show
907
00:39:40,345 --> 00:39:42,805
or feedback about the show, feel free
908
00:39:42,825 --> 00:39:45,725
to reach out via our website,
Twitter, or Facebook.
909
00:39:45,985 --> 00:39:48,085
Thanks again for listening
and have a great day.