WEBVTT

0:00:02.740000 --> 0:00:07.020000
 Hey, let's take a few minutes
 to talk about alerts.

0:00:07.020000 --> 0:00:09.360000
 We're going to really go through
 a couple different things.

0:00:09.360000 --> 0:00:14.060000
 We're going to talk about alerts
 in Azure in general.

0:00:14.060000 --> 0:00:17.680000
 Then I'm going to go through a demonstration
 of how you can set up and

0:00:17.680000 --> 0:00:20.800000
 how you might trigger alerts
 and what are the components.

0:00:20.800000 --> 0:00:24.000000
 Then I'm going to talk about some of the
 for lack of a better term advanced

0:00:24.000000 --> 0:00:30.260000
 alerts and what are some
 of the built-in alerts.

0:00:30.260000 --> 0:00:35.820000
 Then I want to go ahead and talk about
 some of the alert takeaways.

0:00:35.820000 --> 0:00:39.680000
 What are some of the things that we
 want to really come away from this

0:00:39.680000 --> 0:00:45.440000
 lesson with? Let's go ahead and jump
 into this and let's talk about alerts

0:00:45.440000 --> 0:00:50.460000
 in general. The concept of alert is
 something that if you've worked in

0:00:50.460000 --> 0:00:53.320000
 IT, you're probably familiar with.


0:00:53.320000 --> 0:00:57.460000
 It's really just a process where there's
 some level of automated monitoring

0:00:57.460000 --> 0:01:02.380000
 and when something happens, usually something
 bad, you want to be notified

0:01:02.380000 --> 0:01:07.460000
 of this, whether it's by email or text
 message or maybe making a call

0:01:07.460000 --> 0:01:09.720000
 to another automated system.

0:01:09.720000 --> 0:01:15.360000
 Whatever it is, if you think about the
 core concept of alerts in general,

0:01:15.360000 --> 0:01:19.080000
 I'm not even talking about the concept
 of alerts in Azure, but if we have

0:01:19.080000 --> 0:01:24.800000
 the core concept of alerts, you've got
 this idea of something that happens.

0:01:24.800000 --> 0:01:28.540000
 There's some triggering event.

0:01:28.540000 --> 0:01:36.120000
 Let's say for example, there's
 a HTTP, HTTP error.

0:01:36.120000 --> 0:01:43.320000
 I've got a web app, there's an HTTP error,
 and then I want that to trigger

0:01:43.320000 --> 0:01:47.380000
 some response, some action.

0:01:47.380000 --> 0:01:51.720000
 Maybe multiple actions.

0:01:51.720000 --> 0:01:56.240000
 For example, most commonly,
 maybe it sends out an email.

0:01:56.240000 --> 0:02:01.280000
 At the highest level, that's what we
 have when we think about alerting.

0:02:01.280000 --> 0:02:06.900000
 I've used alerting in a wide range
 of systems, probably back in the, I

0:02:06.900000 --> 0:02:09.180000
 love saying this,
 previous century.

0:02:09.180000 --> 0:02:12.840000
 Maybe one of the first things I used
 was SQL Server and how you could

0:02:12.840000 --> 0:02:15.400000
 set up notifications there.

0:02:15.400000 --> 0:02:18.960000
 Really, I've used it in many other systems
 as well, and likely, frankly,

0:02:18.960000 --> 0:02:22.460000
 so have you. Really, what we want to
 do is say, okay, we've got this basic

0:02:22.460000 --> 0:02:25.960000
 concept, and even if you'd never seen
 it before, what we have over here

0:02:25.960000 --> 0:02:30.600000
 really is the basic concept, how is
 this implemented within the Azure

0:02:30.600000 --> 0:02:35.500000
 environment? That's actually also
 really fairly straightforward.

0:02:35.500000 --> 0:02:40.300000
 You have two components to really defining
 alerts, and the first component

0:02:40.300000 --> 0:02:43.140000
 is the alert rule.

0:02:43.140000 --> 0:02:44.080000
 That's what we have here.

0:02:44.080000 --> 0:02:49.080000
 We've got this alert rule, and with an
 alert rule, you really define three

0:02:49.080000 --> 0:02:54.560000
 things. First of all, you define what is
 the resource that you're monitoring?

0:02:54.560000 --> 0:02:59.020000
 Really, that term very clear,
 this is the Azure resource.

0:02:59.020000 --> 0:03:01.280000
 What Azure resource
 are you monitoring?

0:03:01.280000 --> 0:03:07.520000
 Then what condition do you want to
 have as the trigger condition?

0:03:07.520000 --> 0:03:10.680000
 The conditions, of course, are going
 to vary based on the alert.

0:03:10.680000 --> 0:03:14.360000
 The conditions I would have for a web
 app are going to be necessarily

0:03:14.360000 --> 0:03:18.960000
 different than the conditions I would
 have for alerting based on a virtual

0:03:18.960000 --> 0:03:21.780000
 machine, although there
 may be overlap.

0:03:21.780000 --> 0:03:27.600000
 Then the last component of the alert
 rule is what action you want to take.

0:03:27.600000 --> 0:03:32.920000
 I've got the resource, I've got what's
 going on with the resource, and

0:03:32.920000 --> 0:03:36.160000
 then I've got what I want to
 do when this gets triggered.

0:03:36.160000 --> 0:03:41.320000
 Now, the action is really defined
 by an action group.

0:03:41.320000 --> 0:03:46.400000
 An action group is a set of actions
 that are defined together that can

0:03:46.400000 --> 0:03:49.300000
 be associated with alert rules.

0:03:49.300000 --> 0:03:53.420000
 Why a group? Well, let's say you
 have one particular alert.

0:03:53.420000 --> 0:03:57.140000
 Let's say you've got an HTTP error,
 and one thing that you want to do

0:03:57.140000 --> 0:04:02.960000
 is you want to notify your
 administrators by email.

0:04:02.960000 --> 0:04:04.060000
 Makes total sense.

0:04:04.060000 --> 0:04:08.720000
 But maybe also, you've got a third
 party automation and logging system

0:04:08.720000 --> 0:04:13.960000
 that has a webhook, it has a REST API,
 and you want to send some data

0:04:13.960000 --> 0:04:15.540000
 to that as well.

0:04:15.540000 --> 0:04:18.320000
 Well, I could define two different
 actions, but tie those together as

0:04:18.320000 --> 0:04:21.380000
 an action group, and that's
 really all that is.

0:04:21.380000 --> 0:04:25.700000
 That's why I have this ability to have
 an action group, which can be made

0:04:25.700000 --> 0:04:30.340000
 up of one or more actions, and each
 action group can be associated with

0:04:30.340000 --> 0:04:31.980000
 one or more alert rules.

0:04:31.980000 --> 0:04:37.800000
 And likewise, each alert rule can actually
 have multiple action groups.

0:04:37.800000 --> 0:04:38.740000
 That's the idea.

0:04:38.740000 --> 0:04:40.320000
 Now, what are the different
 types of action groups?

0:04:40.320000 --> 0:04:43.880000
 Well, you've got communication-based
 action groups, sending out emails,

0:04:43.880000 --> 0:04:51.700000
 text messages. You've got the ability
 to link into functionality, to link

0:04:51.700000 --> 0:04:56.040000
 into logic. And there's really actually,
 if I look about pretty much everything

0:04:56.040000 --> 0:05:02.140000
 other than communication and possibly
 ITSM, those are all potentially

0:05:02.140000 --> 0:05:04.900000
 custom actions that can be taken.

0:05:04.900000 --> 0:05:10.000000
 So you can think of an alert not only as
 a way of notifying you that something

0:05:10.000000 --> 0:05:14.560000
 possibly bad has happened, but you
 might even take that as a means of

0:05:14.560000 --> 0:05:17.200000
 just automating processes.

0:05:17.200000 --> 0:05:21.100000
 You're starting to run
 out of a disk space.

0:05:21.100000 --> 0:05:24.120000
 What do you do? Well, that may not be
 a critical thing, but maybe there's

0:05:24.120000 --> 0:05:27.760000
 a whole automation process that you
 want to kick off to respond to that.

0:05:27.760000 --> 0:05:31.900000
 And so there's a lot of different
 ways that you can use alerts.

0:05:31.900000 --> 0:05:33.940000
 In fact, you can think
 of alerting.

0:05:33.940000 --> 0:05:37.280000
 I typically look at alerting from the
 standpoint of saying, OK, something's

0:05:37.280000 --> 0:05:41.520000
 happening. I want to be extremely
 fast-reactive to that.

0:05:41.520000 --> 0:05:43.060000
 So I want to get the
 alerts so I know it.

0:05:43.060000 --> 0:05:47.000000
 But you could actually look at it as
 an element of an automation strategy

0:05:47.000000 --> 0:05:50.600000
 altogether. It doesn't really matter
 as far as Azure's concerned.

0:05:50.600000 --> 0:05:53.800000
 You've got the idea of some sort of
 event, and you've got some sort of

0:05:53.800000 --> 0:05:57.000000
 response to it. As long as it falls
 within that, then it's going to work

0:05:57.000000 --> 0:06:03.520000
 out fine. Now, in terms of the actual
 condition, that's based on these

0:06:03.520000 --> 0:06:07.620000
 things we call alert signals, which
 sounds like it's a really, I don't

0:06:07.620000 --> 0:06:10.020000
 know, that sounds like
 a really complex thing.

0:06:10.020000 --> 0:06:11.100000
 It's really not.

0:06:11.100000 --> 0:06:15.600000
 It's really just, OK, what are the different
 actions, activities, events

0:06:15.600000 --> 0:06:18.320000
 that can trigger an alert rule?

0:06:18.320000 --> 0:06:20.380000
 And there's two categories
 for that.

0:06:20.380000 --> 0:06:23.100000
 One is metrics, so things
 like performance.

0:06:23.100000 --> 0:06:25.040000
 And then the other
 is activity logs.

0:06:25.040000 --> 0:06:29.380000
 So you could have an alert, for example,
 that triggers when kind of a

0:06:29.380000 --> 0:06:36.860000
 stereotypical example, your percent processor
 is exceeding 80% on a virtual

0:06:36.860000 --> 0:06:38.520000
 machine. You could
 trigger off that.

0:06:38.520000 --> 0:06:42.720000
 But you could also trigger when somebody
 tries to change certain properties

0:06:42.720000 --> 0:06:46.640000
 of a storage account, or maybe
 add things to a resource group.

0:06:46.640000 --> 0:06:52.520000
 And you can have those kinds of activity
-based, activity-log-based alerts.

0:06:52.520000 --> 0:06:58.740000
 In fact, if you're in the Azure Monitor
 and you're going through the logs,

0:06:58.740000 --> 0:07:01.820000
 which is covered in other videos, you
 can run a log query and actually

0:07:01.820000 --> 0:07:05.880000
 create an alert based on that, based on
 the kind of data you would retrieve.

0:07:05.880000 --> 0:07:11.880000
 So, again, broad concept, and what
 I want to do next is getting beyond

0:07:11.880000 --> 0:07:17.240000
 the broad concept, I'm going to go
 ahead and take a look at setting a

0:07:17.240000 --> 0:07:19.420000
 performance-based alert.

0:07:19.420000 --> 0:07:22.000000
 And for this, I'm going to
 take an existing web app.

0:07:22.000000 --> 0:07:26.020000
 And I've got the web app set up
 so that it can generate errors.

0:07:26.020000 --> 0:07:28.940000
 Not something you would normally want
 to do with the web app, certainly

0:07:28.940000 --> 0:07:31.800000
 not intentionally, but it works out pretty
 well when you're demonstrating

0:07:31.800000 --> 0:07:33.660000
 things like alerts.

0:07:33.660000 --> 0:07:37.840000
 And what I'm going to do is set up an
 alert on this web app so that any

0:07:37.840000 --> 0:07:43.840000
 time there is an error, or a certain
 number of errors, it will go ahead

0:07:43.840000 --> 0:07:45.700000
 and send me an alert.

0:07:45.700000 --> 0:07:53.020000
 So without further ado, let's go
 ahead and let's set this up.

0:07:53.020000 --> 0:08:00.060000
 I've got my portal to my standard subscription
 here, and I have deployed

0:08:00.060000 --> 0:08:04.840000
 a web app. And here's my web app.

0:08:04.840000 --> 0:08:08.520000
 And I've actually got
 the page open to it.

0:08:08.520000 --> 0:08:13.040000
 I am not a designer, so this is not the
 most amazing looking web app ever.

0:08:13.040000 --> 0:08:18.280000
 But I have the ability, for example,
 to set a network load, and I have

0:08:18.280000 --> 0:08:22.360000
 the ability, a lot of reasons I can
 use this, and I can have this kind

0:08:22.360000 --> 0:08:23.900000
 of randomly generated errors.

0:08:23.900000 --> 0:08:28.080000
 So I'm going to have 100 requests, and
 in there there will be some errors.

0:08:28.080000 --> 0:08:30.260000
 And you can see the
 errors out of this.

0:08:30.260000 --> 0:08:34.320000
 I've got some 500 errors, some
 404 errors that are generating.

0:08:34.320000 --> 0:08:35.180000
 And there we go.

0:08:35.180000 --> 0:08:40.380000
 So I've got 73 good messages,
 and then 27 errors.

0:08:40.380000 --> 0:08:42.340000
 And again, just an easy
 way to generate errors.

0:08:42.340000 --> 0:08:44.040000
 Please don't do this at home.

0:08:44.040000 --> 0:08:49.900000
 What I want to do is I want to set
 up an alert on this app service so

0:08:49.900000 --> 0:08:56.060000
 that when I do get those error messages,
 I can, well, be alerted to that.

0:08:56.060000 --> 0:08:59.900000
 So what I'm going to do is I'm going
 to pop down here simply to the alert

0:08:59.900000 --> 0:09:05.520000
 section. Now, the interface that I'm
 about to pull up is actually going

0:09:05.520000 --> 0:09:07.700000
 to be a highly standardized
 interface.

0:09:07.700000 --> 0:09:10.320000
 In fact, I'm going to
 bring up two of these.

0:09:10.320000 --> 0:09:15.260000
 I'm going to bring up one where I can
 set up an action group, and another

0:09:15.260000 --> 0:09:16.780000
 one where I create the alert rule.


0:09:16.780000 --> 0:09:20.980000
 Now, I could create an alert rule and
 an action group at the same time,

0:09:20.980000 --> 0:09:26.880000
 but I want to separate these out
 just for illustrative purposes.

0:09:26.880000 --> 0:09:31.420000
 Now, I mentioned in fact that there's
 a number of different ways that

0:09:31.420000 --> 0:09:36.520000
 you can go in and manage both
 action groups and alert rules.

0:09:36.520000 --> 0:09:41.600000
 I'm going directly through a resource
 simply because I think it's the

0:09:41.600000 --> 0:09:47.040000
 easiest way to see
 what's going on.

0:09:47.040000 --> 0:09:50.800000
 I'm going to go and add an action group,
 and we're going to call this

0:09:50.800000 --> 0:10:02.680000
 demo action group because I am
 not the least bit creative.

0:10:02.680000 --> 0:10:05.400000
 A short name of demo action is
 hopefully that's short enough.

0:10:05.400000 --> 0:10:11.660000
 It is, and I'm going to put this in,
 we'll put it in that resource group,

0:10:11.660000 --> 0:10:15.940000
 O1 tasks. All right, so that's the action
 group, and then I start to define

0:10:15.940000 --> 0:10:20.680000
 actions. And the first action that
 I'm going to define, I'll just call

0:10:20.680000 --> 0:10:24.300000
 it email because that's what
 I'm going to end up doing.

0:10:24.300000 --> 0:10:28.120000
 And then I can select an action type,
 and you'll see all these action

0:10:28.120000 --> 0:10:34.660000
 types, which are essentially the list
 that I had in the previous slide.

0:10:34.660000 --> 0:10:39.240000
 However, I had communication, which
 could either be email, SMS, push,

0:10:39.240000 --> 0:10:42.960000
 or voice, or email as your
 resource manager role.

0:10:42.960000 --> 0:10:46.000000
 So it kind of sets that
 out automatically.

0:10:46.000000 --> 0:10:50.620000
 I have secure web hook and web hook,
 which is just variations of calling

0:10:50.620000 --> 0:10:55.240000
 a REST API. But we're going to go ahead
 and set up email SMS, push, and

0:10:55.240000 --> 0:11:00.140000
 voice. And I've got four options, and
 I can choose any, I have to choose

0:11:00.140000 --> 0:11:02.440000
 at least one, but I can
 choose more than one.

0:11:02.440000 --> 0:11:13.600000
 But I'm going to simply choose email
 and send this to my email account.

0:11:13.600000 --> 0:11:19.120000
 And we will not use the common alerts
 schema, which is basically alert

0:11:19.120000 --> 0:11:22.580000
 here. And I hit OK.

0:11:22.580000 --> 0:11:26.980000
 And that's going to send an email out
 based on the, you know, anytime

0:11:26.980000 --> 0:11:27.960000
 this gets called.

0:11:27.960000 --> 0:11:29.120000
 Now this is just the action group.


0:11:29.120000 --> 0:11:31.420000
 This is not the alert rule.

0:11:31.420000 --> 0:11:32.960000
 So this isn't triggering anything.


0:11:32.960000 --> 0:11:38.000000
 It's just essentially giving an endpoint
 that can be used when alerts

0:11:38.000000 --> 0:11:44.340000
 are defined. Now I could have another
 action in this action group and

0:11:44.340000 --> 0:11:47.600000
 just select something else, for example,
 maybe a web hook, and then I

0:11:47.600000 --> 0:11:50.960000
 would go and define that web hook.


0:11:50.960000 --> 0:11:53.420000
 But I don't have a web
 hook set up for this.

0:11:53.420000 --> 0:11:56.920000
 So I'm going to delete
 that and hit OK.

0:11:56.920000 --> 0:11:59.320000
 And that gives me an action group.


0:11:59.320000 --> 0:12:02.300000
 There we go. I've got
 my action group.

0:12:02.300000 --> 0:12:07.520000
 All right. Now I'm going to go back
 and I am going to create a new alert

0:12:07.520000 --> 0:12:15.300000
 rule. Now the resource that I'm creating
 this on is Iany Alert Web.

0:12:15.300000 --> 0:12:18.820000
 This interface is the absolute
 standard interface.

0:12:18.820000 --> 0:12:24.580000
 In fact, if I were to go to monitor,
 I could create an alert exactly for

0:12:24.580000 --> 0:12:26.220000
 exactly the same scenario.

0:12:26.220000 --> 0:12:29.460000
 The only difference is that the resource
 would not be preselected.

0:12:29.460000 --> 0:12:33.300000
 So by going through the interface, the
 portal interface for a particular

0:12:33.300000 --> 0:12:37.320000
 resource, it just essentially
 pre-fills that option.

0:12:37.320000 --> 0:12:39.960000
 All right. Now I do need
 to specify a condition.

0:12:39.960000 --> 0:12:41.840000
 So I'm going to add a condition.

0:12:41.840000 --> 0:12:48.080000
 And these are all of these signals that
 are available for this particular

0:12:48.080000 --> 0:12:50.000000
 type of resource.

0:12:50.000000 --> 0:12:54.980000
 And I've got two types as I mentioned
 metrics, which are the majority

0:12:54.980000 --> 0:12:58.240000
 and activity logs, or at least
 the majority we could see.

0:12:58.240000 --> 0:13:02.140000
 All right. So for example, maybe I want
 to know every time a web app gets

0:13:02.140000 --> 0:13:08.000000
 restarted. I could have
 that as my condition.

0:13:08.000000 --> 0:13:11.480000
 But I'm going to go ahead
 and go with metrics.

0:13:11.480000 --> 0:13:14.420000
 And I'm going to come down
 here to HTTP server errors.

0:13:14.420000 --> 0:13:19.540000
 So anytime there's any server errors,
 I want to go ahead and choose that.

0:13:19.540000 --> 0:13:23.020000
 And this is going to show me, OK, here's
 what kinds of server errors you've

0:13:23.020000 --> 0:13:27.040000
 had. And you notice actually
 there are a few of them.

0:13:27.040000 --> 0:13:33.280000
 And then I can set up my instance.


0:13:33.280000 --> 0:13:37.840000
 OK, I'm not going to set up an instance saying
 that has to do with differentiating

0:13:37.840000 --> 0:13:43.980000
 the way that you are calculating
 or monitoring that.

0:13:43.980000 --> 0:13:48.640000
 OK, also you can set up dynamic
 alert, thresholding.

0:13:48.640000 --> 0:13:52.100000
 What that does is it actually uses a little
 bit of artificial intelligence.

0:13:52.100000 --> 0:13:54.860000
 It kind of monitors what's
 been going on.

0:13:54.860000 --> 0:13:59.960000
 So for example, if I've got a web app
 that just always generates a lot

0:13:59.960000 --> 0:14:03.620000
 of HTTP errors, maybe it's
 not perfectly built.

0:14:03.620000 --> 0:14:08.740000
 All right. I don't necessarily always
 want to alert that, but I might

0:14:08.740000 --> 0:14:12.640000
 want to get alert if I get something
 that's really out of the norm.

0:14:12.640000 --> 0:14:15.340000
 All right. But in this case, I'm just
 going to pick static and I'm going

0:14:15.340000 --> 0:14:20.420000
 to say, if the total is greater
 than, we'll say, five.

0:14:20.420000 --> 0:14:23.180000
 And this is over a period.

0:14:23.180000 --> 0:14:25.580000
 We have an aggregation
 period here.

0:14:25.580000 --> 0:14:27.780000
 I can aggregate every
 five minutes.

0:14:27.780000 --> 0:14:29.780000
 I can aggregate every one minute.

0:14:29.780000 --> 0:14:32.320000
 We'll go and aggregate
 every one minute.

0:14:32.320000 --> 0:14:35.600000
 And it's going to check
 every minute.

0:14:35.600000 --> 0:14:38.180000
 So we'll go back to five minutes.

0:14:38.180000 --> 0:14:43.080000
 And done. And so that
 sets up the condition.

0:14:43.080000 --> 0:14:47.680000
 Now I am going to associate
 an action group.

0:14:47.680000 --> 0:14:52.040000
 So I go in here and there's the action
 group that I just created.

0:14:52.040000 --> 0:14:54.480000
 I have another one that's just
 used for other demonstrations.

0:14:54.480000 --> 0:14:56.940000
 Actually, it'll be used
 in a few minutes.

0:14:56.940000 --> 0:15:00.580000
 And now I've got an action
 group associated with this.

0:15:00.580000 --> 0:15:01.740000
 And then there's a few details.

0:15:01.740000 --> 0:15:05.240000
 I have to give this a name, which not
 surprisingly, it's just going to

0:15:05.240000 --> 0:15:09.960000
 be demo alert. I should
 give it a description.

0:15:09.960000 --> 0:15:19.940000
 And we'll say server errors because
 that's really what it is.

0:15:19.940000 --> 0:15:22.860000
 And the severity is just
 a number that you choose.

0:15:22.860000 --> 0:15:24.560000
 And it's really just metadata.

0:15:24.560000 --> 0:15:27.240000
 Believe it has severity three.

0:15:27.240000 --> 0:15:29.740000
 And do I want this rule
 enabled upon creation?

0:15:29.740000 --> 0:15:31.760000
 Yes, in fact, I do.

0:15:31.760000 --> 0:15:34.960000
 So I'm going to go ahead
 and create that rule.

0:15:34.960000 --> 0:15:36.340000
 Wait till that's done.

0:15:36.340000 --> 0:15:37.780000
 And then I'm going to
 generate some errors.

0:15:37.780000 --> 0:15:39.600000
 And I'll come back to this.

0:15:39.600000 --> 0:15:44.500000
 And hopefully by the time I'm done covering
 our advanced errors, the alert

0:15:44.500000 --> 0:15:48.960000
 or advanced alerts, the error
 will have generated an alert.

0:15:48.960000 --> 0:15:52.220000
 For some reason, that was
 difficult for me to say.

0:15:52.220000 --> 0:15:55.420000
 All right, give this
 just a moment.

0:15:55.420000 --> 0:15:58.300000
 And we'll be back as soon as that's
 done creating the alert rule.

0:15:58.300000 --> 0:16:01.920000
 And I will generate the errors.

0:16:01.920000 --> 0:16:09.120000
 All right, so the alert rule and the
 action group have been created.

0:16:09.120000 --> 0:16:11.980000
 And really now all I need
 to do is go ahead.

0:16:11.980000 --> 0:16:14.780000
 And I'm just going to rerun this.

0:16:14.780000 --> 0:16:17.720000
 And the edge create plenty.

0:16:17.720000 --> 0:16:21.620000
 So I execute this and I start
 to get some errors.

0:16:21.620000 --> 0:16:27.140000
 All right, and fairly quickly, if all
 goes well, ironically, that should

0:16:27.140000 --> 0:16:30.720000
 trigger the alert and we should
 be able to track that.

0:16:30.720000 --> 0:16:32.860000
 In the meantime, let's go ahead.

0:16:32.860000 --> 0:16:34.580000
 So we're not waiting
 around for that.

0:16:34.580000 --> 0:16:42.260000
 And let's take a look at
 some advanced alerts.

0:16:42.260000 --> 0:16:46.080000
 And really, the advanced alerts aren't
 particularly advanced, really more

0:16:46.080000 --> 0:16:47.560000
 like built-in alerts.

0:16:47.560000 --> 0:16:51.560000
 And there's a few different types of
 built-in alerts that you can take

0:16:51.560000 --> 0:16:56.840000
 advantage of. From the cost management
 side, there are budget alerts.

0:16:56.840000 --> 0:17:01.120000
 You can set a budget and then you can
 set points at which you want to

0:17:01.120000 --> 0:17:05.180000
 be alerted. So if I've got my typical
 spin budget, I can set an alert

0:17:05.180000 --> 0:17:07.820000
 at certain percentages
 of that budget.

0:17:07.820000 --> 0:17:12.520000
 Now, I'm going to demonstrate that if
 you watched actually really covered

0:17:12.520000 --> 0:17:16.140000
 budgets before and kind of as
 a side note did the alerting.

0:17:16.140000 --> 0:17:20.260000
 But if you saw that, then you may not
 need to watch this part of the video.

0:17:20.260000 --> 0:17:21.540000
 But I've got that.

0:17:21.540000 --> 0:17:24.960000
 In addition to that, there are alerts
 that you can set up within Azure

0:17:24.960000 --> 0:17:30.260000
 AD. And the alerting capability to some
 extent depends on whether or not

0:17:30.260000 --> 0:17:33.340000
 you have the premium tier.

0:17:33.340000 --> 0:17:38.980000
 But things, for example, like password
 reset, that you can get alerted

0:17:38.980000 --> 0:17:40.720000
 on at all times.

0:17:40.720000 --> 0:17:46.880000
 You have a user's at risk alert so
 that if there are risky activities

0:17:46.880000 --> 0:17:49.800000
 that occur, then you can
 get alerted on that.

0:17:49.800000 --> 0:17:53.440000
 If you are implementing privileged identity
 management, which in and of

0:17:53.440000 --> 0:17:57.820000
 itself requires the premium tier, it
 actually requires premium P2 tier

0:17:57.820000 --> 0:18:02.020000
 of Azure AD, you can set up
 so that you're alerted.

0:18:02.020000 --> 0:18:05.960000
 Anytime, for example, somebody may
 elevate there or request to elevate

0:18:05.960000 --> 0:18:11.580000
 their permissions in order to accomplish
 some functionality within Azure.

0:18:11.580000 --> 0:18:13.160000
 And that's really what that is.

0:18:13.160000 --> 0:18:20.020000
 But the idea is that these are really
 predefined, built-in alerts.

0:18:20.020000 --> 0:18:23.260000
 And what I'm going to do is go through
 the budget and just show you how

0:18:23.260000 --> 0:18:26.340000
 you can set up an alert
 within the budget.

0:18:26.340000 --> 0:18:32.920000
 And so that demonstration, which I just
 noticed, still says create a setting

0:18:32.920000 --> 0:18:35.500000
 performance alert, which is
 what we've already done.

0:18:35.500000 --> 0:18:40.140000
 I'm just going to jump right into it
 without showing that cool preview

0:18:40.140000 --> 0:18:42.240000
 page. All right.

0:18:42.240000 --> 0:18:45.560000
 I am back in my portal.

0:18:45.560000 --> 0:18:50.000000
 And what I'm going to do is go down
 to cost management and billing.

0:18:50.000000 --> 0:18:55.020000
 And I'm going to go
 to cost management.

0:18:55.020000 --> 0:18:58.000000
 And I'm going to go to budgets.

0:18:58.000000 --> 0:19:02.940000
 And I am going to go ahead
 and add a budget.

0:19:02.940000 --> 0:19:07.020000
 And I'm going to give
 this budget a name.

0:19:07.020000 --> 0:19:08.660000
 It will probably be
 a shock to you.

0:19:08.660000 --> 0:19:11.100000
 They're called demo budget.

0:19:11.100000 --> 0:19:14.580000
 I'm set to reset monthly.

0:19:14.580000 --> 0:19:20.620000
 Going from 2019 to 2021 or 2020.

0:19:20.620000 --> 0:19:26.580000
 And as the case may be, okay,
 next, let's go previous.

0:19:26.580000 --> 0:19:30.900000
 I'm also going to actually
 set the budget.

0:19:30.900000 --> 0:19:35.300000
 Isn't it right place?

0:19:35.300000 --> 0:19:43.560000
 Yes. Ah, that's why I
 have the wrong scope.

0:19:43.560000 --> 0:19:50.040000
 Always do this. I guess I should have
 had some budget data showing there.

0:19:50.040000 --> 0:19:56.160000
 Okay. Now I'm going
 to add a budget.

0:19:56.160000 --> 0:20:00.160000
 That really wasn't supposed to be
 part of the demo, but good to see.

0:20:00.160000 --> 0:20:02.340000
 Anyways, demo budget.

0:20:02.340000 --> 0:20:05.940000
 And a monthly budget.

0:20:05.940000 --> 0:20:09.140000
 It actually suggests
 a number for me.

0:20:09.140000 --> 0:20:14.840000
 Let me say $1000 budget.

0:20:14.840000 --> 0:20:15.820000
 And it's pretty cool.

0:20:15.820000 --> 0:20:20.700000
 I can see my budget line relative to
 what my costs have been and what

0:20:20.700000 --> 0:20:22.420000
 my projected costs are.

0:20:22.420000 --> 0:20:29.160000
 Then I can go and set
 my alert conditions.

0:20:29.160000 --> 0:20:32.480000
 Really, what I'm doing here is I'm setting
 alert rules based on budget.

0:20:32.480000 --> 0:20:36.140000
 So let's say, for example, I want to
 get an early alert when I'm up at

0:20:36.140000 --> 0:20:39.540000
 25% of budget. And then what
 do I want to do with that?

0:20:39.540000 --> 0:20:42.580000
 I want to send that alert
 to demo action group.

0:20:42.580000 --> 0:20:45.620000
 And I can do it again
 at, let's say, 70%.

0:20:45.620000 --> 0:20:50.320000
 Send that to demo action
 group again.

0:20:50.320000 --> 0:20:57.040000
 Then if we get up to 100%, then I'm
 sending that to my primary budget

0:20:57.040000 --> 0:20:59.900000
 folks, which actually
 is me as well.

0:20:59.900000 --> 0:21:01.240000
 But that's the idea, right?

0:21:01.240000 --> 0:21:05.920000
 So I have this ability to actually define
 multiple alert rules or conditions

0:21:05.920000 --> 0:21:08.280000
 in this one interface
 for budgeting.

0:21:08.280000 --> 0:21:12.320000
 And that's really what that does.

0:21:12.320000 --> 0:21:14.480000
 We'll create that.

0:21:14.480000 --> 0:21:16.160000
 And the budget was created.

0:21:16.160000 --> 0:21:17.720000
 And you can see this budget now.

0:21:17.720000 --> 0:21:23.200000
 Now, if alerts are generated, I can actually
 track those here in the cost

0:21:23.200000 --> 0:21:27.660000
 alerts. Now, I don't have any alerts
 because I just created that budget.

0:21:27.660000 --> 0:21:31.560000
 And fortunately, I've never
 triggered the actual budget.

0:21:31.560000 --> 0:21:37.700000
 But that is the idea of setting up
 alerts and kind of a one-off way.

0:21:37.700000 --> 0:21:42.520000
 And again, you have similar alerting
 capabilities within Azure AD.

0:21:42.520000 --> 0:21:49.640000
 And now what I want to do is I want
 to go back and really cover some of

0:21:49.640000 --> 0:21:53.140000
 the takeaways. Just wrap this up and
 then right at the end, we'll pop

0:21:53.140000 --> 0:21:57.380000
 back over and see if our performance
 base alert has actually generated

0:21:57.380000 --> 0:22:01.200000
 an alert yet. So when you think about
 alerts, you really want to think

0:22:01.200000 --> 0:22:04.260000
 about the components, the alert
 rule and the action group.

0:22:04.260000 --> 0:22:05.260000
 Pretty straightforward, right?

0:22:05.260000 --> 0:22:09.420000
 The alert rule has three
 main elements.

0:22:09.420000 --> 0:22:11.120000
 It's got the trigger.

0:22:11.120000 --> 0:22:14.240000
 It's got the filter.

0:22:14.240000 --> 0:22:19.920000
 So the trigger is going to be the
 resource and the actual filter.

0:22:19.920000 --> 0:22:23.060000
 It's got the action group.

0:22:23.060000 --> 0:22:26.240000
 And you also have a few of those
 additional settings that we saw.

0:22:26.240000 --> 0:22:32.400000
 The action group is
 email, SMS, phone.

0:22:32.400000 --> 0:22:36.640000
 You've got function apps, logic apps,
 runbooks, webhooks, webhooks with

0:22:36.640000 --> 0:22:43.760000
 security, ITSM, and also now emailing
 directly to the administrators,

0:22:43.760000 --> 0:22:46.260000
 the owners. All right,
 alert monitoring.

0:22:46.260000 --> 0:22:47.780000
 How do you monitor alerts?

0:22:47.780000 --> 0:22:52.680000
 You've got alert monitoring at the resource
 level within a given blade,

0:22:52.680000 --> 0:22:56.680000
 for example, I can go into the portal,
 go to the blade for a resource,

0:22:56.680000 --> 0:23:00.120000
 and I can actually keep track of the
 alerts that have been generated.

0:23:00.120000 --> 0:23:05.160000
 I have an alerting window within the
 Azure monitor, so I go to that and

0:23:05.160000 --> 0:23:08.020000
 kind of have centralized
 management of my alerts.

0:23:08.020000 --> 0:23:11.220000
 You also have an API that's available where
 you could pull alert information,

0:23:11.220000 --> 0:23:16.040000
 or, of course, you have the ability within
 the alert to send it to a logic

0:23:16.040000 --> 0:23:22.220000
 app or a function app or a webhook,
 which is really any qualifying REST

0:23:22.220000 --> 0:23:26.660000
 API. And by qualifying, there are certain
 rules with a REST API that accepts

0:23:26.660000 --> 0:23:28.440000
 alert information.

0:23:28.440000 --> 0:23:33.220000
 It has to be defined in such a way to
 actually prove that it can receive

0:23:33.220000 --> 0:23:36.160000
 this information, but it's pretty
 straightforward to do so.

0:23:36.160000 --> 0:23:37.840000
 All right, and so again,
 that's what alert is.

0:23:37.840000 --> 0:23:45.020000
 Let's take a quick look back and let's
 see if my alert has been generated.

0:23:45.020000 --> 0:23:52.280000
 All right, we are over
 here in the alerts.

0:23:52.280000 --> 0:23:59.980000
 So let's go to the dashboard
 and let's go to my web app.

0:23:59.980000 --> 0:24:08.400000
 And down here under
 alerts, there we go.

0:24:08.400000 --> 0:24:13.460000
 I've got my one alert, not terribly exciting,
 but this didn't exist before.

0:24:13.460000 --> 0:24:15.520000
 Notice it's severity three.

0:24:15.520000 --> 0:24:17.700000
 I can actually click that.

0:24:17.700000 --> 0:24:19.380000
 There's the demo alert.

0:24:19.380000 --> 0:24:23.320000
 I can see that it fired on my web.


0:24:23.320000 --> 0:24:29.280000
 And if I go into the actual alert,
 I can see why it fired.

0:24:29.280000 --> 0:24:37.580000
 And criteria, and so that was an HTTP
 500 error that fired it and the

0:24:37.580000 --> 0:24:38.560000
 rest of the information.

0:24:38.560000 --> 0:24:39.660000
 So there are server errors.

0:24:39.660000 --> 0:24:43.520000
 That was my description, signal type
 metric alert rule demo alert.

0:24:43.520000 --> 0:24:45.180000
 And I can really get information.

0:24:45.180000 --> 0:24:50.800000
 In addition to that, I have also received
 an email that I'm going to ask

0:24:50.800000 --> 0:24:54.060000
 you to just trust me on because I don't
 have my email up right at the

0:24:54.060000 --> 0:24:57.900000
 moment, but that will have sent me an
 email as well that giving me details

0:24:57.900000 --> 0:24:59.200000
 about the alert.

0:24:59.200000 --> 0:25:00.960000
 And so that's really it.

0:25:00.960000 --> 0:25:04.940000
 I have spent a fair bit of time talking
 about alerts because I think there's

0:25:04.940000 --> 0:25:08.860000
 something that's important, but also
 hopefully you see that conceptually

0:25:08.860000 --> 0:25:10.760000
 they're not really
 all that difficult.

