WEBVTT 00:00.840 --> 00:01.673 Instructor: In this section 00:01.673 --> 00:02.970 we're gonna start making some changes 00:02.970 --> 00:05.250 to our deployment configuration file 00:05.250 --> 00:08.220 and just verify that all the pods that are associated 00:08.220 --> 00:11.040 with this deployment successfully get updated. 00:11.040 --> 00:13.530 So in particular, I really want to try updating 00:13.530 --> 00:15.420 the container port value down here 00:15.420 --> 00:18.330 from 3000 to some other value, 00:18.330 --> 00:19.620 because you'll recall that 00:19.620 --> 00:21.840 when we tried changing that container port 00:21.840 --> 00:22.947 inside of our client.pod, 00:22.947 --> 00:26.130 or excuse me client-pod.yahml file 00:26.130 --> 00:28.080 we very quickly got an error message saying, 00:28.080 --> 00:30.540 hey, you can't update that value. 00:30.540 --> 00:33.240 Okay, so back inside of the client deployment. 00:33.240 --> 00:35.670 So we're in the deployment file right now. 00:35.670 --> 00:37.410 I'm gonna find the container port 00:37.410 --> 00:41.160 and I'll update the port to 99999 like so. 00:41.160 --> 00:41.993 Now, just so you know, 00:41.993 --> 00:43.830 when we change this port assignment 00:43.830 --> 00:45.390 you will no longer be able to visit 00:45.390 --> 00:47.880 our application inside of the browser. 00:47.880 --> 00:49.770 So in case you apply this update 00:49.770 --> 00:51.570 and then try to test it out inside the browser, 00:51.570 --> 00:54.420 it's not going to work because the container doesn't care 00:54.420 --> 00:58.503 about port 999, it wants you to expose traffic on port 3000. 00:59.760 --> 01:02.130 Okay, so now that we changed our configuration file, 01:02.130 --> 01:04.530 I'm gonna make sure that I save the file 01:04.530 --> 01:06.390 and then I'm going to go back over to my terminal 01:06.390 --> 01:08.580 and we're gonna do the same thing we always do 01:08.580 --> 01:11.220 anytime we want to update an object. 01:11.220 --> 01:16.087 We'll do a kubectlapply-F clientdeployment.yaml. 01:19.320 --> 01:21.720 So I'll run that, and then it tells us very directly 01:21.720 --> 01:26.720 that an existing object type with this name was configured. 01:26.970 --> 01:29.880 It says configured as opposed to something being created. 01:29.880 --> 01:31.650 So we definitely just made a change 01:31.650 --> 01:33.420 to an existing deployment as opposed 01:33.420 --> 01:36.000 to creating a new one entirely. 01:36.000 --> 01:40.590 We can now do a kubectl get deployments 01:40.590 --> 01:42.060 to print out all of our deployments. 01:42.060 --> 01:43.530 Yep, it's still there. 01:43.530 --> 01:45.750 I can do a kubectl get pods, 01:45.750 --> 01:47.040 and now with this you're gonna see something 01:47.040 --> 01:48.870 a little bit more interesting. 01:48.870 --> 01:52.920 Notice how the age of this pod right here is 26 seconds. 01:52.920 --> 01:56.790 So clearly when we apply this configuration file right here 01:56.790 --> 02:00.900 Kubernetes noticed that we made a change to the template 02:00.900 --> 02:04.320 for the pods that are being controlled by this deployment. 02:04.320 --> 02:06.450 It saw that we changed the container port, 02:06.450 --> 02:09.870 and so rather than trying to update the existing pod 02:09.870 --> 02:11.130 that already existed 02:11.130 --> 02:14.430 it deleted that pod and completely recreated it 02:14.430 --> 02:16.893 with the new container port value right here. 02:20.370 --> 02:21.900 Now one thing that would be really nice to do 02:21.900 --> 02:24.060 is make sure that this new pod right here 02:24.060 --> 02:26.460 is in fact running with that new port. 02:26.460 --> 02:29.760 So I'll do a kubectl describe pods. 02:29.760 --> 02:31.590 We'll get the long form description 02:31.590 --> 02:33.420 about this pod right here and we'll probably 02:33.420 --> 02:36.720 be able to verify that port 999 is assigned to it. 02:36.720 --> 02:39.030 Now remember when you do describe pods, 02:39.030 --> 02:40.710 the name of whatever you wanna look up 02:40.710 --> 02:42.000 is completely optional. 02:42.000 --> 02:44.010 In this case, we only have one pod, 02:44.010 --> 02:46.320 so I'll just print out the information about all pods 02:46.320 --> 02:47.820 because hey, there's only one. 02:48.900 --> 02:49.830 Okay, so when I run that, 02:49.830 --> 02:51.390 I'll scroll up a little bit 02:51.390 --> 02:52.990 and then I should be able to see 02:53.850 --> 02:55.050 our list of containers. 02:55.050 --> 02:56.670 Here's the client container 02:56.670 --> 03:00.630 and it has port 999 mapped on it. 03:00.630 --> 03:02.100 So without a doubt this pod is running 03:02.100 --> 03:04.800 with the most up-to-date configuration. 03:04.800 --> 03:06.990 Okay, so that's pretty cool. 03:06.990 --> 03:08.550 Now I wanna do one other change 03:08.550 --> 03:09.990 or one or two other changes here. 03:09.990 --> 03:13.710 Let's first try scaling up our replica setting right here. 03:13.710 --> 03:17.250 So when we scale up from one to something like say five, 03:17.250 --> 03:18.900 we're now going to have our deployment 03:18.900 --> 03:21.210 attempt to create five different pods 03:21.210 --> 03:24.033 all with this identical template right here. 03:25.080 --> 03:27.780 So I'm going to make sure I update replicas to five, 03:27.780 --> 03:30.180 I'll save the file and then as usual 03:30.180 --> 03:35.180 we'll do a kubectl apply client deployment.yaml. 03:38.190 --> 03:42.210 Now if you type in very quickly kubectl get deployments 03:42.210 --> 03:44.790 we might okay a little bit too slow 03:44.790 --> 03:46.470 in that case, you'll notice that 03:46.470 --> 03:48.150 everything says five right here. 03:48.150 --> 03:49.890 If we were a little bit faster 03:49.890 --> 03:52.620 we might be able to see that there are less than five 03:52.620 --> 03:54.240 of these different pods available. 03:54.240 --> 03:55.800 We'll make another change in a second 03:55.800 --> 03:57.210 that we'll hopefully be able to 03:57.210 --> 03:59.280 if we type in this get deployments fast enough 03:59.280 --> 04:01.920 we might be able to see the update in action 04:01.920 --> 04:02.790 and we'll see something like, 04:02.790 --> 04:04.560 hey, only five or maybe one 04:04.560 --> 04:07.230 or even zero pods are currently available. 04:07.230 --> 04:09.840 So if we now do a kubectl get pods 04:09.840 --> 04:12.240 you'll notice we have five separate pods 04:12.240 --> 04:15.090 each of which are running a unique copy 04:15.090 --> 04:18.060 of that very specific client, 04:18.060 --> 04:19.530 let's see, multi client that's the idea 04:19.530 --> 04:20.363 or that's the name of it. 04:20.363 --> 04:21.480 Multi client image. 04:21.480 --> 04:23.250 So we now have five containers running 04:23.250 --> 04:25.563 each in their own very separate little pod. 04:26.760 --> 04:28.980 Let's try making one more configuration change. 04:28.980 --> 04:31.830 So how about instead of running the image multi client 04:31.830 --> 04:35.190 let's try running the image multi worker instead. 04:35.190 --> 04:37.050 So I'm gonna change the name of the image 04:37.050 --> 04:38.520 that we're running inside this pod. 04:38.520 --> 04:40.380 I'm gonna make sure I save the file. 04:40.380 --> 04:41.820 And now this time after we do the 04:41.820 --> 04:46.050 kubectl apply-F client deployment 04:46.050 --> 04:48.750 I'm gonna, very quickly right after I run this command 04:48.750 --> 04:52.380 I'm gonna very quickly do a kubectl get deployments 04:52.380 --> 04:54.960 and hopefully you'll see some different numbers 04:54.960 --> 04:56.100 of containers, excuse me, 04:56.100 --> 04:57.930 different numbers of pods inside there 04:57.930 --> 04:59.760 as the deployment is creating 04:59.760 --> 05:02.940 and destroying some of the pods inside of our cluster. 05:02.940 --> 05:04.530 Okay, so I'm gonna hit Enter here, 05:04.530 --> 05:07.803 and then very quickly I'll do a kubectl get deployments. 05:08.760 --> 05:09.910 Now that's interesting. 05:10.830 --> 05:12.870 So notice I typed in fast enough 05:12.870 --> 05:13.920 that I was able to see 05:13.920 --> 05:16.350 some different numbers in action here. 05:16.350 --> 05:18.540 So the client deployment wants to have 05:18.540 --> 05:21.900 five different pods with the very specific 05:21.900 --> 05:24.270 configuration listed right here. 05:24.270 --> 05:26.820 But at present there are seven pods. 05:26.820 --> 05:30.510 That means that maybe five of the old version exist. 05:30.510 --> 05:31.343 Or actually, you know what, 05:31.343 --> 05:33.420 we can use these other numbers here to figure it out. 05:33.420 --> 05:35.280 So there are seven pods total 05:35.280 --> 05:38.010 and four of them are up-to-date and available. 05:38.010 --> 05:41.190 That means that we have four up-to-date and available pods 05:41.190 --> 05:43.560 running the newest configuration that says 05:43.560 --> 05:46.740 that we should be running the multi worker image. 05:46.740 --> 05:49.410 That means that there are three other pods 05:49.410 --> 05:50.970 because seven minus four is three, 05:50.970 --> 05:53.640 means that there are three pods still sitting around 05:53.640 --> 05:55.440 using the old configuration, 05:55.440 --> 05:58.350 and those need to be deleted and cleaned up. 05:58.350 --> 06:00.930 If we now do kubectl get deployments again 06:00.930 --> 06:03.120 we're probably gonna see five across the board 06:03.120 --> 06:04.860 because now all the old versions 06:04.860 --> 06:06.300 of those pods have been deleted 06:06.300 --> 06:09.420 and replaced with the latest up-to-date version 06:09.420 --> 06:12.510 of the pod spec, which says that we need to have 06:12.510 --> 06:15.900 the multi worker image running inside there. 06:15.900 --> 06:16.950 Okay, so that's pretty interesting 06:16.950 --> 06:18.720 to see these deployments in action. 06:18.720 --> 06:19.860 So let's take another quick pause. 06:19.860 --> 06:21.150 We're gonna come back to the next section 06:21.150 --> 06:23.820 and I wanna point out one or two kind of interesting things 06:23.820 --> 06:25.500 about this entire deployment system 06:25.500 --> 06:27.840 that might trip you up when you start making use of it 06:27.840 --> 06:29.430 in a production environment. 06:29.430 --> 06:31.853 So quick pause and I'll see you in just a minute.