WEBVTT 00:00.140 --> 00:06.500 So right now we have two node ports set up to expose our external services. 00:06.500 --> 00:12.530 However, we want to actually create a new ingress resource because behind the scenes, Google Kubernetes 00:12.530 --> 00:19.040 engine will automatically provision us a load balancer and give us a single IP address that we can use 00:19.040 --> 00:24.150 to make requests to our entire application, which is what we want to do. 00:24.170 --> 00:30.590 So to easily accomplish this, let's open up our folder and we're going to add a new file in the templates 00:30.590 --> 00:34.390 directory called Ingress dot Yaml. 00:34.400 --> 00:42.050 This will have an API version of networking.k80s.io/v one. 00:42.080 --> 00:48.680 The kind will be of type of ingress and we're going to have some metadata here where we give it a name 00:48.680 --> 00:49.880 of sleeper. 00:49.880 --> 00:58.010 And then we importantly we define the spec with a rules section and then we specify the Http paths in 00:58.010 --> 00:59.600 our application. 00:59.600 --> 01:09.180 So go ahead and specify a paths list here and we'll specify the first path, which will be slash reservations. 01:09.180 --> 01:14.610 And we'll use a wild card here to direct all traffic to our reservation service. 01:14.640 --> 01:21.810 We're going to go ahead and also specify the path type as implementation specific, which just means 01:21.810 --> 01:26.790 Google Kubernetes engine will handle all of the routing back end section here. 01:26.790 --> 01:32.160 We'll have a service and we're going to go ahead and specify the name of the service, which of course 01:32.160 --> 01:34.290 we know for reservations. 01:34.290 --> 01:40.920 We have our service Yaml where the name is reservations, so we'll simply provide the name of the reservations 01:40.920 --> 01:44.940 service and then provide the port number that we're going to access it on. 01:44.940 --> 01:47.420 So in this case it'll be 3004. 01:47.430 --> 01:53.580 Let's go ahead and do the same thing now for the slash auth routes in our application. 01:53.580 --> 01:58.770 We're going to go ahead and direct them as a implementation specific route. 01:58.770 --> 02:06.630 And the back end here will have a service with a name of auth Http because we want to be directing traffic 02:06.630 --> 02:11.190 to our Http auth service and not the TCP one. 02:11.190 --> 02:17.310 So make sure we specify the name of the auth Http here, which matches the name we've specified. 02:17.310 --> 02:21.900 And the port for Http if we take a look, is 3003. 02:21.900 --> 02:24.870 So we'll go ahead and specify the port as 3003. 02:24.870 --> 02:31.290 So make sure we actually change the port here to the number field and specify 3003. 02:31.290 --> 02:36.210 And don't forget to add the wild card path to the auth path here. 02:36.210 --> 02:42.240 So we catch all traffic that starts with auth and then proxy the rest of the path onwards to our service. 02:42.270 --> 02:50.550 We'll go ahead and run helm upgrade Sleeper to install our new ingress resource so we can also go in 02:50.550 --> 02:57.240 the Google Cloud engine UI and under the services and ingress section, we can see that there is a section 02:57.240 --> 02:58.980 for ingress here. 02:58.980 --> 03:05.220 If we click on it, we can see that our ingress is currently being created and it's picked up these 03:05.220 --> 03:11.280 routes we defined earlier on reservations and auth and we can even click into it and see a bunch more 03:11.280 --> 03:17.760 detail, including the annotations, the pods it's going to be serving and the state of the underlying 03:17.760 --> 03:18.720 deployments here. 03:18.720 --> 03:25.080 You can see our auth and reservation deployments haven't yet responded back to the health check, so 03:25.080 --> 03:27.870 we just need to wait a bit until these finish. 03:27.870 --> 03:34.920 So after a bit you should be able to run Kubectl, get ingress and see the address for our externally 03:34.920 --> 03:39.240 provisioned load balancer which is associated with our ingress. 03:39.240 --> 03:47.850 So now we can copy this external address and go into Postman and launch an Http request and paste in 03:47.850 --> 03:54.330 the IP address that you have here and hit slash reservations and send a request. 03:54.360 --> 04:01.230 If you see this 502 server error, give your application a bit of time before it ends up resolving these 04:01.230 --> 04:05.010 health checks and you should see the 502 error here go away. 04:05.250 --> 04:13.890 Now, if we launch a request at our external IP here at Http slash login and post our usual body of 04:13.890 --> 04:20.490 an email and password, we can go ahead and log in with our known email here. 04:20.490 --> 04:22.740 So I'll go ahead and log in with our known email. 04:22.740 --> 04:26.460 And of course we get our JWT cookie set from the auth service. 04:26.460 --> 04:35.430 And then on the same external URL, we'll now going to hit slash reservations to create a new reservation. 04:35.430 --> 04:41.340 I'll go ahead and copy an existing payload from one of our previous requests and paste it in here. 04:41.340 --> 04:41.790 Now. 04:41.790 --> 04:48.330 I'll then go ahead and change the amount to 35 and send off the request to create a new reservation. 04:48.330 --> 04:49.920 And you can see our 201. 04:49.920 --> 04:51.300 Created response. 04:51.450 --> 04:57.390 So before we test out our new load balancer, I'm going to go ahead and authorize our Gmail API one 04:57.390 --> 04:59.940 more time and get an up to date refresh. 04:59.970 --> 05:00.600 Token. 05:00.600 --> 05:02.930 So go back to the Oauth2 playground. 05:02.940 --> 05:06.840 Make sure you've entered your OAuth client ID and secret here. 05:06.840 --> 05:09.870 And we've selected the Mail Google API. 05:09.900 --> 05:15.390 We'll go ahead and authorize it and exchange the authorization code for tokens. 05:15.420 --> 05:24.660 Go ahead and copy the refresh token and go ahead and base 64 encoded using the echo dash n open quote, 05:24.660 --> 05:31.830 paste the refresh token, end quote, and then pipe it to the base 64 command. 05:31.830 --> 05:40.890 We'll go ahead and copy this and edit our secret Google and replace the refresh token and paste it in 05:40.890 --> 05:43.920 and then go ahead and save the up to date secret. 05:43.950 --> 05:51.480 Now, after updating the secret, make sure we restart our notifications Pod by running rollout, restart 05:51.480 --> 05:59.340 deployment notifications so we can go ahead and run Kubectl get in here short for ingress and see that 05:59.340 --> 06:04.900 we have an external load balancer IP now in the address field here. 06:04.900 --> 06:10.240 So we want to copy this and this will be the IP address that we can access all of the routes in our 06:10.240 --> 06:12.820 application inside of this ingress. 06:12.820 --> 06:16.960 So this is the address of the load balancer in Google Kubernetes engine. 06:16.960 --> 06:23.530 We're going to go ahead and paste the request to Http, colon slash, slash enter in that load balancer 06:23.530 --> 06:30.610 address and go to auth slash login and change our request to a post request here. 06:30.610 --> 06:32.710 And now you can see we're getting a 401. 06:32.710 --> 06:34.330 Because we're unauthorized. 06:34.330 --> 06:38.230 So you can see we're actually able to communicate with the auth service. 06:38.230 --> 06:47.020 Let's go ahead and provide a raw Json body here and go ahead and copy a login payload from our previous 06:47.020 --> 06:54.280 login route so we can copy this body, paste it in to the Json body field. 06:54.280 --> 06:58.600 And this is going to be the email associated with our Gmail account. 06:58.600 --> 07:06.190 So we'll go ahead and send this to log in and get back the JWT cookie here for our user. 07:06.190 --> 07:15.040 So now we can launch a request at slash reservations and copy a reservations payload from our previous 07:15.040 --> 07:18.220 requests and paste it into this body. 07:18.220 --> 07:21.100 And let's change the amount here to 45. 07:21.130 --> 07:25.270 So now we have a new reservation and we'll go ahead and send it off. 07:25.300 --> 07:26.980 You can see we have a response. 07:26.990 --> 07:32.950 404 Make sure that we add a trailing slash here to match the exact ingress route. 07:33.340 --> 07:39.370 So now if we send this off, we can see that we have a new reservation that's been created and we can 07:39.370 --> 07:46.840 send a slash get make sure you send no body for this request and we can see all of the reservations 07:46.840 --> 07:47.800 in our system. 07:47.800 --> 07:55.300 Back in Gmail, I can see I have a new notification saying our $45 payment has completed successfully. 07:55.300 --> 08:03.220 And if we go into Stripe and refresh the payments screen, I can see a new payment that succeeded for 08:03.220 --> 08:04.570 $45. 08:04.570 --> 08:09.490 So now we can see that our application is fully running on Google Kubernetes engine.