1
00:00:00,980 --> 00:00:05,510
In this video we're gonna start to set up a routing rules for all the other micro services inside of

2
00:00:05,510 --> 00:00:06,620
our cluster.

3
00:00:06,720 --> 00:00:11,240
I'm inside of our ingress SRB file inside of our infra K Cade's directory.

4
00:00:11,240 --> 00:00:16,670
Right now we only have a route mapping for slashed posts which means in theory we can only attempt to

5
00:00:16,700 --> 00:00:21,650
fetch or retrieve posts is that is the only thing we're doing with our post cluster IP service.

6
00:00:21,650 --> 00:00:26,030
So we need to add in some additional parts to this thing to make sure we map of all the different incoming

7
00:00:26,030 --> 00:00:27,540
requests that might come in.

8
00:00:27,800 --> 00:00:32,150
Now I went through all of our different micro services and the react application and put together a

9
00:00:32,150 --> 00:00:37,000
quick diagram to help you remember all the different routes that we need to service.

10
00:00:37,090 --> 00:00:37,390
All right.

11
00:00:37,420 --> 00:00:42,000
So in this diagram we have some number of requests coming in from our ReACT application.

12
00:00:42,010 --> 00:00:44,300
These are all the different requests we expect to receive.

13
00:00:44,890 --> 00:00:50,530
So we're going to get a post to slash post if that occurs it should go to the post pod.

14
00:00:50,530 --> 00:00:54,470
We're also gonna have a post to slash post slashed ideas slash comments.

15
00:00:54,610 --> 00:01:01,030
That request should go to the common service a get to slash post should go to query and then finally

16
00:01:01,030 --> 00:01:04,940
a get to just slashed by itself is attempting to load up the react application.

17
00:01:04,980 --> 00:01:06,960
So we need to make sure that gets sent to the reactor.

18
00:01:08,040 --> 00:01:12,190
So we need to write out all these different routing rules inside that ingress config file.

19
00:01:12,240 --> 00:01:14,660
Now I got some bad news.

20
00:01:14,760 --> 00:01:19,390
This is something that I knew about ahead of time and I wanted you to have to go through this as well

21
00:01:19,650 --> 00:01:22,820
because I've had to deal with this on a real project too.

22
00:01:22,860 --> 00:01:27,430
The bad news here is that we have two routes right now of slash posts.

23
00:01:27,600 --> 00:01:32,850
If a request comes in with a method of post then that should go to the postal service.

24
00:01:32,850 --> 00:01:35,480
Otherwise if it is a get issue go to query.

25
00:01:35,760 --> 00:01:42,300
The bad news here is that our ingress engine X module can not do routing based upon the method of the

26
00:01:42,300 --> 00:01:43,300
request.

27
00:01:43,470 --> 00:01:49,970
So all the methods I have list on here the ingress controller has no idea that these methods exist.

28
00:01:50,190 --> 00:01:56,890
So the only thing we can really root on effectively is a route early a path so for all these things

29
00:01:56,900 --> 00:02:00,430
it really goes down to just this.

30
00:02:00,550 --> 00:02:01,500
This is all we get.

31
00:02:01,500 --> 00:02:06,190
This is what we have to route based on.

32
00:02:06,320 --> 00:02:11,090
Now the reason this is bad is that now when we look at this you'll notice that if we have a request

33
00:02:11,090 --> 00:02:15,830
coming into a slash posts we're saying it should go to this post pod but there's another one down here

34
00:02:15,830 --> 00:02:19,900
that says if a request comes in to slash post it should go to the query pod.

35
00:02:20,210 --> 00:02:23,970
We can no longer differentiate on the method of the incoming request.

36
00:02:24,020 --> 00:02:28,730
So now we don't have enough information with these requests to figure out whether or not one should

37
00:02:28,730 --> 00:02:33,390
go to the post pod or query so that's the bad news here.

38
00:02:33,390 --> 00:02:38,930
To fix this we need to make sure that we have some unique path or request that should go to posts and

39
00:02:38,940 --> 00:02:41,940
a unique path for one it should go to query.

40
00:02:41,970 --> 00:02:44,910
In other words we really just need to change the name of one of these parts.

41
00:02:44,940 --> 00:02:50,550
So maybe this right here instead of being slash post it should be something like those slash create

42
00:02:50,610 --> 00:02:52,000
or something like that.

43
00:02:52,140 --> 00:02:56,160
If we put that in now we can distinguish between requests that's going to go to the post pod and the

44
00:02:56,160 --> 00:03:00,090
query pod that actually make this reality.

45
00:03:00,110 --> 00:03:05,660
Well we need to make sure that we not only update our ReACT application to make a request to slash create

46
00:03:05,960 --> 00:03:12,150
we need to make sure we also update the post pod itself so that it should expect requests to come in

47
00:03:12,150 --> 00:03:14,260
to slash post slash create.

48
00:03:14,290 --> 00:03:19,250
So in other words we need to change two different images right now just to solve this image or this

49
00:03:19,250 --> 00:03:21,230
issue that we are running into.

50
00:03:21,230 --> 00:03:25,940
Let's take care of that right now because we're going to have to edit those two code bases and then

51
00:03:25,940 --> 00:03:28,880
rebuild the images and post them back up to Docker Hub.

52
00:03:29,330 --> 00:03:32,970
So I can first get started inside of by react application.

53
00:03:33,060 --> 00:03:38,610
I'm going to find my post create file that is where we are making a post request to slash post so I'm

54
00:03:38,610 --> 00:03:45,270
gonna change that route to post slash create again so that we can differentiate between a request that

55
00:03:45,270 --> 00:03:49,050
is supposed to go to the post pod versus the query pod.

56
00:03:49,260 --> 00:03:50,150
There's the change right there.

57
00:03:50,190 --> 00:03:58,000
I can save this file then after that I'm gonna go to my post project find the next J S file inside there.

58
00:03:58,180 --> 00:04:01,910
Here's the post request handler so post flash.

59
00:04:02,110 --> 00:04:08,580
Create you will recall that we do have a getter or a get request handler for Slash post right here.

60
00:04:08,680 --> 00:04:13,000
That was in the previous initial version of our application before we implemented that query.

61
00:04:13,030 --> 00:04:16,790
So this thing is technically unused right now and we don't really care about that handler at all.

62
00:04:16,810 --> 00:04:18,520
We can actually delete it if we want to.

63
00:04:18,520 --> 00:04:23,050
I only left it in here so we could test out our post end point once we initially set up all that ingress

64
00:04:23,080 --> 00:04:25,440
engine X stuff okay.

65
00:04:25,470 --> 00:04:34,970
So I'll say this file as well and I'll go back to my terminal I'm gonna go to my client folder and I'll

66
00:04:34,970 --> 00:04:39,930
do a docker build client Dot.

67
00:04:40,010 --> 00:04:40,400
There we go.

68
00:04:41,680 --> 00:04:45,950
And a docker push.

69
00:04:45,970 --> 00:04:51,240
There we go and we're going to repeat that same process for post as well.

70
00:04:51,260 --> 00:04:55,220
Now this one time you're probably sitting there saying wait a minute Steven wait wait wait wait.

71
00:04:55,220 --> 00:05:00,200
We have to rebuild the image manually and push it up to Docker Hub every single time we make a change

72
00:05:00,200 --> 00:05:04,680
or a code Hey wait a minute that seems a really really awkward.

73
00:05:04,750 --> 00:05:06,540
I know it is really awkward.

74
00:05:06,570 --> 00:05:09,040
This is something else we're gonna fix up very quickly.

75
00:05:09,070 --> 00:05:13,630
I really just want us to have to go through this problem distinctly one more time because very quickly

76
00:05:13,630 --> 00:05:17,320
like in the next come coming videos we're gonna take a look at a tool that's going to automate all this

77
00:05:17,320 --> 00:05:20,740
change process for us and so I just want to give you a quick reminder that yeah.

78
00:05:20,790 --> 00:05:24,910
Sound right now they want to make a change to our code we have to rebuild image push it and then update

79
00:05:24,910 --> 00:05:26,350
the deployment manually.

80
00:05:26,350 --> 00:05:28,480
All very painful.

81
00:05:28,500 --> 00:05:28,700
All right.

82
00:05:28,710 --> 00:05:30,360
Well the client is all done.

83
00:05:30,360 --> 00:05:33,690
I'm going to restart that deployment.

84
00:05:33,750 --> 00:05:38,170
They'll do a cube C.T. El.

85
00:05:38,270 --> 00:05:48,470
What is it rollout restart deployment client Client dash devil.

86
00:05:48,470 --> 00:05:49,230
There we go.

87
00:05:49,460 --> 00:05:53,480
That will restart our client deployment and we're going to go through all those same steps for our postal

88
00:05:53,480 --> 00:05:54,770
service as well.

89
00:05:54,860 --> 00:06:02,920
So I will go back to post I'll do a docker Bill dash t even greater post dot.

90
00:06:03,110 --> 00:06:07,010
I'll do a docker push even greater post

91
00:06:12,220 --> 00:06:16,540
and then once this is all done we'll go ahead and restart the post deployment as well

92
00:06:21,270 --> 00:06:21,960
quick upload.

93
00:06:21,960 --> 00:06:29,070
Come on come on it this apple it always takes slightly longer than you hope I'll.

94
00:06:29,080 --> 00:06:40,180
There you go so we'll do a cube GTL rollout restart deployment those dash double OK.

95
00:06:40,210 --> 00:06:45,250
So we've now got that new version of code or posts and for the rack client.

96
00:06:45,250 --> 00:06:49,600
So now we've got unique routes that we're going to access each of our different pods.

97
00:06:49,790 --> 00:06:51,000
So now quick pause right here.

98
00:06:51,000 --> 00:06:54,730
When come back the next video we can start to actually write out the routing configuration that we need.