1 00:00:02,100 --> 00:00:05,390 So we defined our host Persistent Volume 2 00:00:05,390 --> 00:00:07,810 and that's the first part of the equation. 3 00:00:07,810 --> 00:00:10,070 The second part is to claim 4 00:00:10,070 --> 00:00:13,550 because this just defines the volume in the cluster, 5 00:00:13,550 --> 00:00:15,090 but in order to use it, 6 00:00:15,090 --> 00:00:17,360 we now also need a claim 7 00:00:17,360 --> 00:00:21,210 and that claim then needs to be made by the pods 8 00:00:21,210 --> 00:00:23,712 that want to use this volume. 9 00:00:23,712 --> 00:00:26,220 Therefore, we gotta do two things. 10 00:00:26,220 --> 00:00:29,400 We first of all have to configure this claim 11 00:00:29,400 --> 00:00:31,813 and then we have to use it on all the pods 12 00:00:31,813 --> 00:00:33,970 that do wanna use it. 13 00:00:33,970 --> 00:00:38,400 So I'll start by adding the host-pvc.yaml file, 14 00:00:38,400 --> 00:00:39,760 which is to claim 15 00:00:39,760 --> 00:00:41,610 that's what the C stands for, 16 00:00:41,610 --> 00:00:43,793 for this host Persistent Volume. 17 00:00:44,640 --> 00:00:47,180 And again, we could merge this all into one 18 00:00:47,180 --> 00:00:49,580 or multiple bigger yaml files, 19 00:00:49,580 --> 00:00:53,300 I'm just splitting it up here across multiple yaml files 20 00:00:53,300 --> 00:00:55,283 to keep this a bit more manageable. 21 00:00:56,180 --> 00:01:00,770 So what should go into this claim file now? 22 00:01:00,770 --> 00:01:03,230 As always, we started with the apiVersion 23 00:01:03,230 --> 00:01:05,209 and that should be v1 here again, 24 00:01:05,209 --> 00:01:08,113 just as for the Persistent Volume itself, 25 00:01:08,965 --> 00:01:12,670 then we need a kind, to make Kubernetes aware 26 00:01:12,670 --> 00:01:15,480 of which kind of resource we are creating here. 27 00:01:15,480 --> 00:01:19,910 And a claim is also a resource, everything is a resource 28 00:01:19,910 --> 00:01:21,900 in a Kuberenetes world. 29 00:01:21,900 --> 00:01:26,900 So here we add PersistentVolumeClaim, written like this. 30 00:01:27,790 --> 00:01:30,850 Again, make sure you have no typo in here, 31 00:01:30,850 --> 00:01:32,513 otherwise this will not work. 32 00:01:33,660 --> 00:01:37,610 Next, we add metadata and we give this claim a name. 33 00:01:37,610 --> 00:01:42,610 This name will later be important for using it in a pod. 34 00:01:42,850 --> 00:01:46,443 And here I'll name it, host-pvc. 35 00:01:47,640 --> 00:01:49,380 Now thereafter, you might've guessed it, 36 00:01:49,380 --> 00:01:51,700 we have to spec, the specification, 37 00:01:51,700 --> 00:01:54,590 the configuration for this claim. 38 00:01:54,590 --> 00:01:56,910 And here we of course need to specify 39 00:01:56,910 --> 00:02:00,330 which Persistent Volume we wanna claim. 40 00:02:00,330 --> 00:02:01,960 Now, here we only have one, 41 00:02:01,960 --> 00:02:04,290 but we could have multiple Persistent Volumes, 42 00:02:04,290 --> 00:02:07,810 which also could use different types, If you needed that. 43 00:02:07,810 --> 00:02:09,880 And in real big applications, 44 00:02:09,880 --> 00:02:14,030 you will of course often have multiple Persistent Volumes. 45 00:02:14,030 --> 00:02:17,800 To connect our claim to this Persistent Volume, 46 00:02:17,800 --> 00:02:22,370 in the spec of the claim, we can add a VolumeName key, 47 00:02:22,370 --> 00:02:25,590 and then simply use the name we used here 48 00:02:25,590 --> 00:02:30,030 for the Persistent Volume, in my case, host-pv, 49 00:02:30,030 --> 00:02:32,483 so that this volume will be claimed here. 50 00:02:33,840 --> 00:02:35,370 Now, I will say when it comes 51 00:02:35,370 --> 00:02:37,264 to claiming Persistent Volumes, 52 00:02:37,264 --> 00:02:41,300 Kubernetes, again offers you a lot of flexibility. 53 00:02:41,300 --> 00:02:43,720 You cannot just claim by name, 54 00:02:43,720 --> 00:02:47,330 there are other ways of claiming by resource, 55 00:02:47,330 --> 00:02:48,180 you could say. 56 00:02:48,180 --> 00:02:51,870 So that you don't claim one specific Persistent Volume, 57 00:02:51,870 --> 00:02:53,690 but if you had different volumes, 58 00:02:53,690 --> 00:02:54,607 you could just say, 59 00:02:54,607 --> 00:02:57,500 "I want the volume with that much space 60 00:02:57,500 --> 00:03:00,242 and this and that configuration." 61 00:03:00,242 --> 00:03:03,110 It would be beyond the scope of this course 62 00:03:03,110 --> 00:03:07,390 and all the more and more kind of an administrator's task 63 00:03:07,390 --> 00:03:08,460 to dive into that. 64 00:03:08,460 --> 00:03:10,520 Therefore, we're not going to do that, 65 00:03:10,520 --> 00:03:13,110 but if you dive into the official documentation 66 00:03:13,110 --> 00:03:14,910 for Persistent Volumes, 67 00:03:14,910 --> 00:03:18,330 and also in the examples you find on the Kubernetes' page, 68 00:03:18,330 --> 00:03:20,310 you will also see different ways 69 00:03:20,310 --> 00:03:23,290 of claiming a Persistent Volume. 70 00:03:23,290 --> 00:03:24,740 Claiming it by name however, 71 00:03:24,740 --> 00:03:26,640 is of course very straightforward 72 00:03:26,640 --> 00:03:29,633 and therefore all the perfect for this use case here. 73 00:03:30,700 --> 00:03:32,000 So now we made it clear 74 00:03:32,000 --> 00:03:34,670 which Persistent Volume we wanna use. 75 00:03:34,670 --> 00:03:37,750 Now we need to provide more information here. 76 00:03:37,750 --> 00:03:40,710 For example, we also need accessModes again, 77 00:03:40,710 --> 00:03:44,280 and now we need to define the modes which we wanna use here 78 00:03:44,280 --> 00:03:47,137 for claiming this volume. 79 00:03:47,137 --> 00:03:50,150 Now you could again, list multiple modes here, 80 00:03:50,150 --> 00:03:53,250 but I only want to claim by the only supported way, 81 00:03:53,250 --> 00:03:56,688 which is ReadWriteOnce access mode here. 82 00:03:56,688 --> 00:04:00,000 But if you would use this claim on different pods 83 00:04:00,000 --> 00:04:03,070 and some pods only need read access, 84 00:04:03,070 --> 00:04:05,070 other need read-write access. 85 00:04:05,070 --> 00:04:07,760 You could specify multiple access modes here 86 00:04:07,760 --> 00:04:09,770 for the claim as well. 87 00:04:09,770 --> 00:04:13,005 Next, we can add the resources key here, 88 00:04:13,005 --> 00:04:17,089 which is basically the counterpart to the capacity here. 89 00:04:17,089 --> 00:04:20,930 And we can now specify which resources we wanna get 90 00:04:20,930 --> 00:04:22,710 for this claim. 91 00:04:22,710 --> 00:04:27,160 The resources object has a request key nested below it, 92 00:04:27,160 --> 00:04:31,100 and there you then can request storage. 93 00:04:31,100 --> 00:04:34,390 Again, that's the counterpart to this capacity. 94 00:04:34,390 --> 00:04:37,330 So since I have a storage of one gigabyte 95 00:04:37,330 --> 00:04:40,570 as a capacity in this Persistent Volume, 96 00:04:40,570 --> 00:04:44,320 that's also the maximum I can request here. 97 00:04:44,320 --> 00:04:47,800 So I will actually request one gigabyte of storage here, 98 00:04:47,800 --> 00:04:51,590 but we could also request less for this claim. 99 00:04:51,590 --> 00:04:55,100 If we had multiple claims to the same Persistent Volume, 100 00:04:55,100 --> 00:04:57,950 we might wanna request less, so that in total 101 00:04:57,950 --> 00:05:01,010 we don't request more than we got available. 102 00:05:01,010 --> 00:05:02,590 This could lead to errors, 103 00:05:02,590 --> 00:05:04,740 if we then try to use all these claims 104 00:05:04,740 --> 00:05:07,770 on our pods and we started writing more data 105 00:05:07,770 --> 00:05:09,830 than we have capacity. 106 00:05:09,830 --> 00:05:13,980 And that is it, this is a Persistent Volume Claim. 107 00:05:13,980 --> 00:05:15,640 Now, as you can clearly tell 108 00:05:15,640 --> 00:05:20,210 this still doesn't establish a connection to a pod though. 109 00:05:20,210 --> 00:05:25,070 Instead does this now a claim which can now be used by pods 110 00:05:25,070 --> 00:05:29,740 to make that claim to that Persistent Volume. 111 00:05:29,740 --> 00:05:32,620 So we need to go back to the deployment yaml file 112 00:05:32,620 --> 00:05:36,420 and now connect our pod here to this claim 113 00:05:36,420 --> 00:05:40,100 so that ultimately it reaches out to this Persistent Volume 114 00:05:40,100 --> 00:05:42,780 and gets its volume. 115 00:05:42,780 --> 00:05:46,240 And we established this connection still through 116 00:05:46,240 --> 00:05:49,440 the volumes key in our pod spec. 117 00:05:49,440 --> 00:05:52,883 So in the deployment yaml file, this volumes key. 118 00:05:53,840 --> 00:05:58,580 We can keep our name here, but now the type is different 119 00:05:58,580 --> 00:06:00,400 instead of hostPath, 120 00:06:00,400 --> 00:06:04,340 which created a host path volume just for this pod, 121 00:06:04,340 --> 00:06:08,253 we now use the persistentVolumeClaim type here. 122 00:06:11,770 --> 00:06:13,890 And this does what the name implies, 123 00:06:13,890 --> 00:06:18,210 now the volume type for this pod is this claim. 124 00:06:18,210 --> 00:06:20,540 So we will use the volume created 125 00:06:20,540 --> 00:06:25,540 or accessed by this claim as a volume for this pod now. 126 00:06:25,780 --> 00:06:28,430 And you could use this in different pods of course, 127 00:06:28,430 --> 00:06:30,530 you can use one of the same claim 128 00:06:30,530 --> 00:06:32,423 in different types of pods. 129 00:06:33,430 --> 00:06:35,250 So here we were making this claim 130 00:06:35,250 --> 00:06:38,040 and we now just need to specify the claim name 131 00:06:38,040 --> 00:06:42,350 as a nested option below this persistentVolumeClaim type, 132 00:06:42,350 --> 00:06:44,740 because otherwise, of course, Kubernetes doesn't know 133 00:06:44,740 --> 00:06:46,940 which claim we might wanna make. 134 00:06:46,940 --> 00:06:49,300 Here, we only have one, but again, 135 00:06:49,300 --> 00:06:51,890 in bigger projects and bigger clusters, 136 00:06:51,890 --> 00:06:55,000 you of course might have multiple claims. 137 00:06:55,000 --> 00:06:58,453 So here it's this claim with this name here. 138 00:06:58,453 --> 00:07:01,830 And therefore I will use host-pvc 139 00:07:01,830 --> 00:07:04,083 as a claim name for this volume. 140 00:07:05,310 --> 00:07:06,980 We don't need to change anything 141 00:07:06,980 --> 00:07:10,560 about the volumeMounts here because we still mounted 142 00:07:10,560 --> 00:07:12,220 in the same way as before. 143 00:07:12,220 --> 00:07:15,550 Now, it's just a volume not created specifically 144 00:07:15,550 --> 00:07:19,283 for this pod, but instead for the entire cluster. 145 00:07:20,150 --> 00:07:22,310 And that of course was a lot of talking 146 00:07:22,310 --> 00:07:23,870 and a lot of work. 147 00:07:23,870 --> 00:07:26,610 But with that, if we now apply everything 148 00:07:26,610 --> 00:07:28,190 and start everything, 149 00:07:28,190 --> 00:07:30,470 we should be able to have our application up 150 00:07:30,470 --> 00:07:31,730 and running again. 151 00:07:31,730 --> 00:07:36,730 But now data should be fully independent from our pods. 152 00:07:36,810 --> 00:07:40,360 Just as before with the host path as I will admit, 153 00:07:40,360 --> 00:07:44,640 but now theoretically all is independent from our nodes. 154 00:07:44,640 --> 00:07:48,570 Not for this host path demo here, as I explained. 155 00:07:48,570 --> 00:07:51,470 There we still have this node dependency, 156 00:07:51,470 --> 00:07:53,740 but as I also mentioned multiple times, 157 00:07:53,740 --> 00:07:56,770 what you learned about setting up Persistent Volumes 158 00:07:56,770 --> 00:08:00,330 and persistentVolumeClaims will also hold true, 159 00:08:00,330 --> 00:08:03,170 if you would use a different volume type here. 160 00:08:03,170 --> 00:08:07,000 And then you would have real node independence. 161 00:08:07,000 --> 00:08:08,593 So let's give this a try now.