1
00:00:01,350 --> 00:00:05,400
Now that we've taken a look at a note port service we're going to start to take a look at cluster IP

2
00:00:05,400 --> 00:00:06,480
as well.

3
00:00:06,510 --> 00:00:14,110
Remember that the goal of a cluster IP service is to expose a pod to other pods inside the cluster to

4
00:00:14,110 --> 00:00:15,610
better explore a cluster IP.

5
00:00:15,610 --> 00:00:20,030
I want to give you a quick reminder on how parts of our application are working right now.

6
00:00:20,080 --> 00:00:22,210
The president we've got that post project.

7
00:00:22,210 --> 00:00:27,430
Anytime someone tries to create a post we are sending a request over to the event bus and the event

8
00:00:27,430 --> 00:00:33,070
bus hopefully is going to emit that event or kind of broadcast that event to everything else inside

9
00:00:33,070 --> 00:00:41,150
of rap including back over to post so let's try to create a new pod that is going to run event bus.

10
00:00:41,160 --> 00:00:46,550
We're then going to setup some services to allow posts any best to communicate with each other.

11
00:00:46,560 --> 00:00:49,710
Let me show you a diagram to help you understand what we're really going to do here.

12
00:00:51,120 --> 00:00:51,510
All right.

13
00:00:51,660 --> 00:00:57,870
So we have one node we currently have a pod running our post stuff and pretty soon we're going to set

14
00:00:57,870 --> 00:01:04,230
up a second pod to run all the event bus code these two pods are going to have to somehow communicate

15
00:01:04,230 --> 00:01:05,330
with each other.

16
00:01:05,370 --> 00:01:10,770
Now unfortunately these pods cannot just reach out directly back and forth to each other like so they

17
00:01:10,770 --> 00:01:14,500
technically can but there are a variety of reasons so we don't do so.

18
00:01:14,670 --> 00:01:19,320
We never really know ahead of time what I.P. address a pod is going to be assigned.

19
00:01:19,500 --> 00:01:26,460
And so we can't really just reach out to say local hosts whatever they 4000 like so because this pod

20
00:01:26,460 --> 00:01:28,410
will not be accessible at local host.

21
00:01:28,410 --> 00:01:33,850
Instead it gets assigned more or less a random IP address inside of our communities cluster.

22
00:01:34,050 --> 00:01:39,130
And this event bus will not know ahead of time what that IP address is.

23
00:01:39,170 --> 00:01:45,550
That's why we have to communicate between these pods through these cluster IP services the post pod

24
00:01:45,640 --> 00:01:50,170
is going to want to reach out and send a message over to the event bus and we've discussed that many

25
00:01:50,170 --> 00:01:51,130
times before.

26
00:01:51,340 --> 00:01:57,590
But as I just said we can't reach out directly so instead we're going to create a cluster IP service

27
00:01:58,070 --> 00:02:04,350
that's going to govern access to the event plus pod whenever posts wants to emit an event.

28
00:02:04,370 --> 00:02:11,060
It's going to make a request over to that cluster I.P. service in turn the event bus is going to want

29
00:02:11,060 --> 00:02:13,710
to send an event back over to post.

30
00:02:13,820 --> 00:02:18,860
It technically wants to send an event to many other pods or any other programs inside of application

31
00:02:19,100 --> 00:02:23,030
but we're not going to worry about creating anything to run say the comments pod or anything like that

32
00:02:23,030 --> 00:02:23,510
just yet.

33
00:02:23,540 --> 00:02:29,390
We're just going to set up post and event bus so event buzz is going to want to send a message back

34
00:02:29,390 --> 00:02:30,890
over to the post pod.

35
00:02:30,950 --> 00:02:35,710
But as I just mentioned we can't reach out directly or at least not very easily.

36
00:02:35,720 --> 00:02:41,930
So instead we're going to have the event bus pod reach out to the cluster I.P. service that is governing

37
00:02:41,960 --> 00:02:43,400
access to the post pod.

38
00:02:43,400 --> 00:02:50,170
Like so I know these lines are now kind of overlapped but that's look somewhere bad.

39
00:02:50,180 --> 00:02:50,600
There we go.

40
00:02:50,630 --> 00:02:57,700
That's kind of what it is actually looking like in reality so essentially what we have to do is create

41
00:02:57,830 --> 00:03:01,430
another deployment that's going to create this event plus pod.

42
00:03:01,600 --> 00:03:07,440
We're then going to create one cluster I.P. service for posts and one cluster IP service for event buzz.

43
00:03:07,480 --> 00:03:11,710
So with all that in mind let's take a pause right here and start making this stuff in just a moment.