WEBVTT

1
00:02:04.860 --> 00:02:05.710
Ingolf Kuss: Hello.

2
00:02:11.100 --> 00:02:11.890
Florian Kreft (LRZ): Hello.

3
00:02:19.300 --> 00:02:20.310
Florian Kreft (LRZ): Hello.

4
00:02:21.520 --> 00:02:22.510
Ingolf Kuss: Nope.

5
00:02:24.070 --> 00:02:25.040
Shelley Doljack: Hi.

6
00:02:40.840 --> 00:02:43.509
Shelley Doljack: So, how is everybody?

7
00:02:45.860 --> 00:02:48.110
Ingolf Kuss: Oh, fun. Thank you.

8
00:02:49.040 --> 00:02:53.019
Shelley Doljack: Is your internet in golf? Your internet's all fast now?

9
00:02:53.290 --> 00:02:54.220
Shelley Doljack: Instable.

10
00:02:54.220 --> 00:02:56.009
Ingolf Kuss: No, nothing happened.

11
00:02:56.120 --> 00:02:57.250
Shelley Doljack: I don't know.

12
00:02:57.730 --> 00:03:02.009
Shelley Doljack: Weren't you getting, fiber installed or something?

13
00:03:03.820 --> 00:03:09.810
Ingolf Kuss: Not yet, I didn't, Start yet.

14
00:03:09.920 --> 00:03:10.560
Ingolf Kuss: Hmm.

15
00:03:13.000 --> 00:03:13.770
Ingolf Kuss: Message.

16
00:03:21.600 --> 00:03:26.009
Ingolf Kuss: Oh, Spring it in the air.

17
00:03:32.230 --> 00:03:36.720
Florian Kreft (LRZ): Yeah, last weekend was really nice here, and Friday, Saturday.

18
00:03:37.030 --> 00:03:38.210
Florian Kreft (LRZ): In Bavaria.

19
00:03:38.550 --> 00:03:39.770
Florian Kreft (LRZ): Sunshine.

20
00:03:39.880 --> 00:03:40.770
Florian Kreft (LRZ): form.

21
00:03:44.510 --> 00:03:45.919
Ingolf Kuss: Oh, it's still nice.

22
00:03:50.380 --> 00:03:52.550
Florian Kreft (LRZ): It's a bit foggy right now here.

23
00:03:53.900 --> 00:03:59.289
Florian Kreft (LRZ): I think it will get better towards the weekend, which is perfect.

24
00:04:02.200 --> 00:04:03.390
Jason Root: Hello, everybody.

25
00:04:03.390 --> 00:04:03.830
Ingolf Kuss: Hello, gentlemen.

26
00:04:03.830 --> 00:04:05.220
Florian Kreft (LRZ): Hello, Jason.

27
00:04:05.850 --> 00:04:11.429
Florian Kreft (LRZ): I actually have a question for you, because we are now starting with Eureka.

28
00:04:12.190 --> 00:04:12.540
Ingolf Kuss: Oh.

29
00:04:12.540 --> 00:04:29.610
Florian Kreft (LRZ): like, trying to set up one example, environment, and luckily I have a colleague with some opinions about setting up Vault and stuff like this, and we tried to do it in a more, like, productive, ready way already, like the secret management side.

30
00:04:29.610 --> 00:04:39.199
Florian Kreft (LRZ): And the last thing we now stumbled over, so I think most of the infrastructure stuff is now working, including the manager applications.

31
00:04:39.270 --> 00:04:47.130
Florian Kreft (LRZ): But then we ran into the problem on what's the actual source of truth, what, like…

32
00:04:47.550 --> 00:05:02.090
Florian Kreft (LRZ): applications to use, so what module versions in the end. So the first step is obviously to deploy them in the backend, or maybe it doesn't even matter what the first step is, maybe we could also, like, have the application descriptors put in there.

33
00:05:02.090 --> 00:05:08.439
Florian Kreft (LRZ): But I tried to, like, that's the official… I think the official release notes are still in the wiki, right? And…

34
00:05:08.440 --> 00:05:16.129
Florian Kreft (LRZ): For copy-based systems, I guess the platform complete was simply the source of truth, like the JSONs there.

35
00:05:16.130 --> 00:05:17.159
Jason Root: Right. And…

36
00:05:17.160 --> 00:05:28.749
Florian Kreft (LRZ): there are still release notes, and there are still, like, issues and stuff like this, but how do you actually get the versions? For example, if there's a new Sunflower CSP, how do you get the version.

37
00:05:28.750 --> 00:05:29.190
Jason Root: understood.

38
00:05:29.190 --> 00:05:31.690
Florian Kreft (LRZ): And, exactly for that release.

39
00:05:31.690 --> 00:05:35.259
Jason Root: So it's… it's 3 things now, which is a little annoying.

40
00:05:35.550 --> 00:05:40.620
Jason Root: So, for platform complete, like we all knew and loved.

41
00:05:40.780 --> 00:05:46.519
Jason Root: There is now a… let me put it in the chat here, let me pull… Yeah, perfect.

42
00:05:47.320 --> 00:05:48.370
Jason Root: Pull up my…

43
00:05:48.850 --> 00:05:56.710
Jason Root: web browser. It is like a three-stage process, which is really annoying. Give me a second, I'm trying to find out which…

44
00:05:57.050 --> 00:06:00.449
Ingolf Kuss: I'm taking some notes, just for information.

45
00:06:00.730 --> 00:06:03.669
Jason Root: Alright, yeah, so let me… can I share my screen, actually? It'd be easier.

46
00:06:03.670 --> 00:06:04.530
Ingolf Kuss: Yeah.

47
00:06:04.530 --> 00:06:05.130
Jason Root: ice cream.

48
00:06:11.980 --> 00:06:13.919
Jason Root: Alright, can everybody see this?

49
00:06:14.520 --> 00:06:15.230
Florian Kreft (LRZ): Yes.

50
00:06:15.380 --> 00:06:16.470
Jason Root: Okay. Hmm.

51
00:06:16.800 --> 00:06:17.180
Ingolf Kuss: Yes.

52
00:06:17.180 --> 00:06:19.880
Jason Root: So… You know, we used to have…

53
00:06:20.150 --> 00:06:24.800
Jason Root: the good old platform complete, and… oh yeah, I'm on the right one.

54
00:06:24.990 --> 00:06:27.669
Jason Root: So now there is Platform LSP.

55
00:06:29.310 --> 00:06:32.490
Jason Root: And Platform LSP has tags.

56
00:06:34.360 --> 00:06:39.729
Jason Root: And these tags, relate to the release. So this is Sunflower CSP4.

57
00:06:39.880 --> 00:06:41.140
Jason Root: Is what this is.

58
00:06:41.530 --> 00:06:49.900
Jason Root: And so if you want the entire platform LSP, there is this, file called, install applications.

59
00:06:51.220 --> 00:06:57.900
Jason Root: Which gives you the application versions that make up LS… platform LSP.

60
00:06:58.050 --> 00:06:59.250
Florian Kreft (LRZ): Okay.

61
00:06:59.250 --> 00:07:08.470
Jason Root: So, for example, App Platform Minimal is right here at 2038, which is the baseline for everything you need.

62
00:07:08.850 --> 00:07:12.009
Jason Root: I see Shelly is smirking because she thinks it's awful, too.

63
00:07:12.870 --> 00:07:13.280
Shelley Doljack: Yes.

64
00:07:14.150 --> 00:07:18.379
Shelley Doljack: But also, like, if you notice, there are some apps that are not there that you.

65
00:07:18.380 --> 00:07:19.340
Jason Root: Yes.

66
00:07:19.450 --> 00:07:20.780
Jason Root: Yes.

67
00:07:20.930 --> 00:07:21.270
Florian Kreft (LRZ): Okay.

68
00:07:21.270 --> 00:07:21.590
Jason Root: It's…

69
00:07:21.590 --> 00:07:22.689
Shelley Doljack: We'll get to that.

70
00:07:22.690 --> 00:07:27.879
Jason Root: Yeah, yeah, we'll get to that. This is a very baseline type of thing.

71
00:07:27.880 --> 00:07:28.560
Florian Kreft (LRZ): Yep.

72
00:07:28.560 --> 00:07:36.040
Jason Root: So, so here are, here are, you know, App Platform Minimal, for example. So if you go to the App Platform Minimal repo.

73
00:07:36.400 --> 00:07:38.580
Jason Root: at 2038.

74
00:07:38.580 --> 00:07:39.230
Florian Kreft (LRZ): Yep.

75
00:07:39.700 --> 00:07:43.880
Jason Root: You'll see that it includes these template files, but also this app.

76
00:07:43.880 --> 00:07:44.200
Florian Kreft (LRZ): Okay.

77
00:07:44.200 --> 00:07:45.750
Jason Root: descriptor file.

78
00:07:45.750 --> 00:07:46.370
Florian Kreft (LRZ): Yes.

79
00:07:46.490 --> 00:07:50.400
Jason Root: And this application descriptor file gives you the modules.

80
00:07:51.530 --> 00:07:53.770
Jason Root: At the versions you need deployed.

81
00:07:53.770 --> 00:07:54.510
Florian Kreft (LRZ): Yes.

82
00:07:55.300 --> 00:08:08.649
Jason Root: And as Shelly pointed out, the list of applications in this is not complete, technically. They're still working on some things, I think, to get this…

83
00:08:09.220 --> 00:08:17.320
Jason Root: I want to say splayed out a little more. There are… there's a site… Here it is.

84
00:08:18.060 --> 00:08:18.900
Florian Kreft (LRZ): Yep.

85
00:08:18.900 --> 00:08:24.710
Jason Root: Yeah, yeah. So this gives you, I think, all of the current applications.

86
00:08:24.710 --> 00:08:25.810
Florian Kreft (LRZ): Okay, yeah.

87
00:08:25.810 --> 00:08:30.000
Jason Root: That exists, and what version they started existing at.

88
00:08:30.400 --> 00:08:37.110
Jason Root: I don't know if this is… how… I don't know how frequently this is kept up to date, so let me… let me refresh here.

89
00:08:37.470 --> 00:08:42.039
Jason Root: So, yeah, updated January 29th, okay.

90
00:08:42.330 --> 00:08:46.259
Florian Kreft (LRZ): So it's relatively recent, but maybe not completely up-to-date.

91
00:08:46.260 --> 00:08:51.169
Jason Root: Yeah, yeah, so it gives you all of the applications that can, in theory, be a part of

92
00:08:51.280 --> 00:08:52.760
Jason Root: your LSP.

93
00:08:53.430 --> 00:09:00.019
Jason Root: and what versions they're released at, if they're even released yet. Like, some of these, I don't even see that they were released yet, so…

94
00:09:00.020 --> 00:09:24.450
Florian Kreft (LRZ): Yeah, but… so in general, I think this is fine, like, there's an overview of what's available, since when might this be available, or maybe if there's, like, a future target release, which is not currently. So, I think, in general, I like that document, and it's maybe even fine to have platform LSP, but I don't think in the release notes, for example, for a CSP, this is linked at all, which is

95
00:09:24.450 --> 00:09:25.629
Florian Kreft (LRZ): really weird.

96
00:09:25.910 --> 00:09:34.419
Jason Root: Yeah, so what you have to do is, like, when a CSP is released… let's see, which tab was it? Not that tab. You have to go back to your platform LSP.

97
00:09:34.420 --> 00:09:35.320
Florian Kreft (LRZ): Yeah.

98
00:09:35.320 --> 00:09:40.760
Jason Root: And then pull down the tags and look at, like, now there's a CSP5, for example.

99
00:09:41.790 --> 00:09:46.210
Jason Root: So then you can look at CSP5 for sunflower, which is this one.

100
00:09:46.610 --> 00:09:56.940
Jason Root: And it gives you all, you know, some things may have changed, so it gives you, like, for example, at platform minimal is now at 2049 instead of 2038.

101
00:09:57.060 --> 00:10:01.300
Jason Root: So you can now go back to App Platform Minimal and go to 204.9.

102
00:10:02.020 --> 00:10:08.360
Jason Root: And then see what has changed as far as these mod-whatever versions are. Yep.

103
00:10:08.360 --> 00:10:08.920
Florian Kreft (LRZ): Okay.

104
00:10:08.920 --> 00:10:10.179
Jason Root: It's pretty terrible.

105
00:10:10.470 --> 00:10:15.669
Jason Root: And then, you know, like, to get a robot to do this.

106
00:10:16.070 --> 00:10:30.739
Jason Root: You know, like, it can be done, of course, you can use whatever JQ or parse all this stuff out, and do curl requests to get all this stuff out, and then said whatever you want to said, but it's… I haven't found a good way yet.

107
00:10:31.010 --> 00:10:37.390
Jason Root: To organize this… robotically Yeah. You know… Okay.

108
00:10:37.620 --> 00:10:38.300
Florian Kreft (LRZ): But…

109
00:10:38.380 --> 00:11:03.320
Florian Kreft (LRZ): So I don't… so I think we stumbled a bit around about this, and I think I've seen this list, and I was, like, the Platform LSP one, but I couldn't remember. Is this even linked anywhere? So I'm really… it's weird to me that we didn't stumble over Platform LSP, because all of the others, like, I think in general, I even like that the applications are actually just semantically versioned, and then we have, like.

110
00:11:03.320 --> 00:11:14.920
Florian Kreft (LRZ): flower release, we have this list of which application and which version. Okay, in general, I think this might be fine, maybe, but it's still, like, yeah.

111
00:11:14.920 --> 00:11:19.449
Shelley Doljack: I don't know, I asked the AI agent how to cobble this together.

112
00:11:19.770 --> 00:11:26.749
Shelley Doljack: Because it wasn't clear. But yeah, okay, you have highlighted there. It's… it's in that Kubernetes example.

113
00:11:26.750 --> 00:11:35.400
Florian Kreft (LRZ): Okay, it's in the example one, under additional information somewhere. Okay, maybe that's exactly the problem, because of how the hell…

114
00:11:35.400 --> 00:11:36.240
Shelley Doljack: Yeah.

115
00:11:36.440 --> 00:11:48.570
Florian Kreft (LRZ): Okay, current release? I mean, I guess I could have, searched the page for, like, current release sunflower, or anything like that, and found it, but, yeah, unless you know it's there.

116
00:11:48.570 --> 00:11:49.090
Jason Root: Yeah.

117
00:11:49.090 --> 00:11:50.920
Florian Kreft (LRZ): When for some reason, it's at the bottom.

118
00:11:50.920 --> 00:11:51.320
Jason Root: Yeah.

119
00:11:51.320 --> 00:11:53.429
Florian Kreft (LRZ): before everything else.

120
00:11:53.430 --> 00:12:00.339
Jason Root: I agree, I agree that the release notes should point to… the platform…

121
00:12:00.340 --> 00:12:00.890
Florian Kreft (LRZ): Yep.

122
00:12:00.890 --> 00:12:02.889
Jason Root: At whatever tag, you know.

123
00:12:03.140 --> 00:12:03.759
Florian Kreft (LRZ): I mean, the release…

124
00:12:03.760 --> 00:12:04.339
Jason Root: I agree.

125
00:12:04.340 --> 00:12:10.259
Florian Kreft (LRZ): even still have, I think, the release tag header, and there's simply nothing there.

126
00:12:10.260 --> 00:12:11.000
Jason Root: I mean, that would be…

127
00:12:11.000 --> 00:12:18.220
Florian Kreft (LRZ): obvious part to just put in the LSP, just put in the LSP with the correct tag there.

128
00:12:18.220 --> 00:12:19.000
Jason Root: Yeah.

129
00:12:19.140 --> 00:12:35.249
Jason Root: Yeah, so I go back to this a lot, these Eureka applications to see which ones are new and up and coming, and all of that fun stuff. But, yeah, I mean, it's a three-stage process, you know, to get from, you know, LSP, whatever, to…

130
00:12:35.490 --> 00:12:36.880
Jason Root: Where you want to be at.

131
00:12:37.590 --> 00:12:45.570
Jason Root: And I wish… You know, they did take my advice early on in the project, this… application descriptor?

132
00:12:46.130 --> 00:12:47.749
Jason Root: was not provided.

133
00:12:48.240 --> 00:12:48.920
Shelley Doljack: Yeah.

134
00:12:48.920 --> 00:13:00.739
Jason Root: When this whole thing started. Yep. So I got onto them, and I was like, guys, you're making these releases anyways, and you're having me run this whole application generator thing, which was really brittle at the time.

135
00:13:00.740 --> 00:13:11.039
Jason Root: I was like, can you just… I know you have a robot doing this. Can you just provide the actual output that you're using? But I'll give you… I'll give you some advice, though.

136
00:13:11.230 --> 00:13:16.429
Jason Root: So I've noticed this… some of these have, like, for example, a URL.

137
00:13:16.610 --> 00:13:17.989
Shelley Doljack: You gotta take it out.

138
00:13:17.990 --> 00:13:33.509
Jason Root: Take the URL out. You can make it null, null is fine, but if it is a UI module, you cannot have a URL here. It's not valid. It's not valid for the application descriptor. And you have to take… you have to take it out also.

139
00:13:33.670 --> 00:13:40.119
Jason Root: So you can, you know, for the mod-whatevers, you know, you can make it null, but for the UI-whatevers.

140
00:13:40.510 --> 00:13:43.240
Jason Root: I don't remember if it's…

141
00:13:44.000 --> 00:13:46.909
Jason Root: Down here… I can't remember which one I saw that at.

142
00:13:50.060 --> 00:14:03.890
Jason Root: I don't think it was… I don't think it was minimal, it was one of the other ones. But they had… they had a URL in it, and I ran, you know, I ran the, entitlement phase, and it was like, nope, I'm sorry, this doesn't meet our

143
00:14:04.060 --> 00:14:20.259
Jason Root: you know, validation, blah blah blah, I forget what the exact error was, and I had to go in there and be like, oh yeah, there's… I didn't know I had to go look at that. I knew about the mod-whatever URLs to take out, I just want to use the Folio, repo. I don't… I don't know why they,

144
00:14:20.550 --> 00:14:23.090
Jason Root: put… let me go back up to the top of this…

145
00:14:24.100 --> 00:14:26.240
Jason Root: I don't know why they'd do this.

146
00:14:26.240 --> 00:14:28.979
Florian Kreft (LRZ): I'm not sure if this is something that…

147
00:14:29.450 --> 00:14:33.679
Jason Root: Their developers just forgot to take out when they published the release.

148
00:14:33.790 --> 00:14:34.860
Jason Root: Or what?

149
00:14:35.090 --> 00:14:38.479
Jason Root: Because, yeah, I mean, they run their own registries.

150
00:14:38.790 --> 00:14:47.040
Jason Root: You know, for a lot of the artifacts. But the Docker Hub one should be the formal GA release of whatever the module is.

151
00:14:47.040 --> 00:14:47.610
Florian Kreft (LRZ): Yo.

152
00:14:47.610 --> 00:14:51.590
Jason Root: And so I don't know how this is making it in here. I really don't.

153
00:14:53.480 --> 00:14:57.739
Jason Root: And I don't know why you'd want to… yeah, I don't know why you'd want to change it anyways, like…

154
00:14:57.970 --> 00:15:10.539
Jason Root: Maybe if, you know, if you're doing a snapshot testing and you have your own internal Docker registry, but you should still, just at that point, be using the Folio CI one from Docker Hub, instead of the Folio Org one.

155
00:15:10.540 --> 00:15:11.130
Florian Kreft (LRZ): Yeah, yeah.

156
00:15:11.130 --> 00:15:11.999
Jason Root: I don't know.

157
00:15:12.360 --> 00:15:12.970
Florian Kreft (LRZ): Yep.

158
00:15:12.970 --> 00:15:14.960
Jason Root: I will stop sharing my screen now.

159
00:15:15.120 --> 00:15:17.130
Florian Kreft (LRZ): Yeah, but perfect, thank you.

160
00:15:17.960 --> 00:15:21.280
Shelley Doljack: Sorry, maybe show the… before you stop sharing, show the…

161
00:15:21.540 --> 00:15:27.639
Shelley Doljack: The other ones, like the, manager install file.

162
00:15:28.170 --> 00:15:29.359
Jason Root: Oh, yeah, okay.

163
00:15:30.480 --> 00:15:33.249
Jason Root: Can y'all… y'all can still see the screen share, yeah? Okay, I didn't.

164
00:15:33.250 --> 00:15:33.720
Shelley Doljack: Yeah.

165
00:15:33.720 --> 00:15:37.930
Jason Root: Okay, so there's, hold on. Do-do-do-do-do…

166
00:15:39.630 --> 00:15:41.820
Shelley Doljack: management modules JSON.

167
00:15:42.430 --> 00:15:43.390
Jason Root: Yeah, yeah.

168
00:15:43.390 --> 00:15:46.370
Shelley Doljack: And also, probably, Stripe's Config.js.

169
00:15:47.730 --> 00:15:49.279
Jason Root: Go back to the LSP.

170
00:15:49.490 --> 00:15:56.309
Jason Root: So, yeah, so every tagged release in LSP also gives you their recommended

171
00:15:56.600 --> 00:16:02.580
Jason Root: backend infrastructure versions for Kong, for KeyCloak, and for the manager apps.

172
00:16:02.760 --> 00:16:07.150
Jason Root: What I have found is this… you don't want to do this either.

173
00:16:07.550 --> 00:16:10.800
Shelley Doljack: Yeah, take out the NB. Yeah, so any, any, I don't know…

174
00:16:10.800 --> 00:16:29.360
Jason Root: I don't know how… yeah, I don't know how this gets in here either, but what I have found is go to the actual Folio.org, module, you know, FolioModule sidecar Docker Hub repo, and see what the real release versions are. There is a… just a flat 3.18, but there were some bug fixes recently, and I think they're

175
00:16:29.470 --> 00:16:35.290
Jason Root: they're coming out with it soon. There was some bug fix recently, which now I think this is up to 3.19 and newer.

176
00:16:35.990 --> 00:16:42.290
Jason Root: And so that's what I'm using. So if you see anything with an NB in it, Take that out.

177
00:16:42.700 --> 00:16:49.400
Jason Root: And, like Shelly pointed out, this does give you the front end, like the traditional platform complete does.

178
00:16:50.340 --> 00:16:53.350
Jason Root: Oop, that's the… that's part of it.

179
00:16:53.350 --> 00:16:54.170
Shelley Doljack: I'm big.

180
00:16:54.170 --> 00:16:54.910
Jason Root: Yeah.

181
00:16:55.680 --> 00:16:59.199
Jason Root: So, it gives you the Stripes platform, like we're used to seeing.

182
00:16:59.400 --> 00:17:17.989
Jason Root: And you will need to variable substitute your URLs in here. And then I actually left this the same, you know, this modules thing where it says populated by Stripes, modules, JS, that's what this is down here, and it just… it just plops it all in there like the traditional one used to.

183
00:17:17.990 --> 00:17:19.920
Florian Kreft (LRZ): We could have just one file in the end.

184
00:17:19.920 --> 00:17:26.239
Jason Root: You could have just one file in the end. I, for simplicity's sake, just left it the same in my own environment.

185
00:17:26.240 --> 00:17:26.819
Florian Kreft (LRZ): Yep.

186
00:17:26.829 --> 00:17:32.429
Jason Root: But you can, you can leave it there, or you can combine the two. It doesn't really matter.

187
00:17:32.739 --> 00:17:45.649
Jason Root: And then the package JSON is how we're used to seeing the package JSON. This hasn't changed much, thank goodness. But they're still doing, they're still doing referencing at Folio Platform Complete up here, you know what I mean?

188
00:17:45.650 --> 00:17:46.950
Florian Kreft (LRZ): Yeah, yeah, that's…

189
00:17:47.720 --> 00:17:57.240
Jason Root: So it's like, why are you making all these unnecessary changes when it doesn't really gain you anything? It just makes it more complex to decipher what the heck is going on, in my opinion.

190
00:17:57.650 --> 00:17:58.220
Florian Kreft (LRZ): Yup.

191
00:17:58.220 --> 00:17:58.740
Jason Root: But.

192
00:17:58.740 --> 00:18:06.050
Shelley Doljack: We found that, like, in order to build this, we had to mess with that package JSON at the very bottom.

193
00:18:06.050 --> 00:18:07.280
Jason Root: Yes, the resolution…

194
00:18:07.280 --> 00:18:15.659
Shelley Doljack: with, like, Babel helpers pinned to 726. We had to just, like, pin it to 7 or something for it to build.

195
00:18:15.660 --> 00:18:16.570
Jason Root: Yeah, let me see…

196
00:18:16.590 --> 00:18:18.009
Shelley Doljack: There's some weird thing. Let's see what…

197
00:18:18.010 --> 00:18:22.019
Jason Root: Let me see what mine was real quick.

198
00:18:25.290 --> 00:18:26.900
Jason Root: Give me a second…

199
00:18:26.900 --> 00:18:30.349
Florian Kreft (LRZ): Is the index data stuff here… no, it's private.

200
00:18:30.350 --> 00:18:36.560
Jason Root: It's all private, yeah. Pay no attention to the man behind the court now. So, let's see here.

201
00:18:38.820 --> 00:18:40.269
Shelley Doljack: You still have that.

202
00:18:40.810 --> 00:18:43.350
Jason Root: Was it… maybe it was Final Form?

203
00:18:45.220 --> 00:18:52.610
Shelley Doljack: I don't know, there was something that we messed with there in order for… we had to, do in order to get it to build.

204
00:18:52.610 --> 00:18:53.320
Jason Root: Yeah.

205
00:18:53.320 --> 00:19:01.770
Florian Kreft (LRZ): Okay, so, but you get errors on build, and then from then on, you can maybe figure out what to actually change, but…

206
00:19:01.770 --> 00:19:02.440
Shelley Doljack: Yeah.

207
00:19:02.660 --> 00:19:08.260
Jason Root: And what I did… I kind of cheated, and what I did is I…

208
00:19:09.130 --> 00:19:12.579
Jason Root: We have a pipeline that builds the front end for us.

209
00:19:12.580 --> 00:19:13.230
Florian Kreft (LRZ): You know?

210
00:19:13.230 --> 00:19:28.050
Jason Root: GitHub Action is what it is, and I cheated, and I had it update the yarn during the build process, after it pulled in the resolutions. I had it update the yarn lock file, and then re… and then build from that to… to…

211
00:19:28.200 --> 00:19:32.909
Jason Root: Resolve, whichever one it was down here. I want to say mine was like…

212
00:19:32.910 --> 00:19:35.900
Florian Kreft (LRZ): they'd actually fix this. Yeah, yeah.

213
00:19:35.900 --> 00:19:42.469
Shelley Doljack: Yeah, I don't know, ours… so I just looked at ours, put a link in it. We… it was the Babbel stuff, we had a…

214
00:19:42.890 --> 00:19:44.260
Shelley Doljack: I'm in.

215
00:19:44.260 --> 00:19:47.420
Jason Root: I want to say mine was the final form,

216
00:19:48.360 --> 00:19:51.270
Jason Root: It's very… let me go to the action here, actually, I can… I can…

217
00:19:51.750 --> 00:19:56.910
Jason Root: Let's see… these are not… yeah, I had a bunch of manual ones. I don't see it in there.

218
00:19:57.400 --> 00:19:59.279
Jason Root: Oh, that's the wrong one, Jason.

219
00:20:00.860 --> 00:20:02.259
Jason Root: This is the right one.

220
00:20:05.020 --> 00:20:07.520
Jason Root: Yeah, you see all the failed ones I had.

221
00:20:07.830 --> 00:20:09.339
Jason Root: But let me, let me look.

222
00:20:11.020 --> 00:20:15.549
Jason Root: Yeah, it's the dev branch. Okay, good, let me make sure I'm on the dev branch, then.

223
00:20:16.040 --> 00:20:18.780
Jason Root: And then I can, yeah, I can look at the workflow.

224
00:20:18.990 --> 00:20:23.590
Shelley Doljack: Oh, I can't find our failed ones, because we don't change the tag name.

225
00:20:23.660 --> 00:20:24.540
Jason Root: Yeah.

226
00:20:24.540 --> 00:20:27.810
Shelley Doljack: It's just overriding the… whatever.

227
00:20:27.810 --> 00:20:32.000
Jason Root: Yeah, I did a… Do-do-do-do-do…

228
00:20:32.000 --> 00:20:35.330
Florian Kreft (LRZ): A yarn install, and then a yarn build.

229
00:20:35.330 --> 00:20:40.099
Jason Root: Yep, then a yarn build, and then the deploy output, yeah. And we have it pushed to a bucket.

230
00:20:40.830 --> 00:20:44.039
Jason Root: And we front the… we front the bucket with, Cloud front.

231
00:20:44.560 --> 00:20:50.020
Jason Root: But that fixed it. I had to add that stuff in to update it. Okay.

232
00:20:50.550 --> 00:21:04.149
Jason Root: Actually, heads up, I mean, I had to mess with quite a bit of this with integration of the reshare stuff and our own front-end modules, so sadly, I know a bit about how to troubleshoot problems here, but…

233
00:21:04.150 --> 00:21:12.989
Jason Root: Yeah, that's fair. And you can see, like, I don't even have all the apps in yet, I've just been putting them in bit by bit and updating my source of truth here.

234
00:21:13.220 --> 00:21:18.690
Jason Root: But some of, you know, some of these files I got from the platform LSP, like this backend modules one.

235
00:21:19.330 --> 00:21:25.299
Jason Root: And it's not everything, you know what I mean? Like, they're putting this file into the branch.

236
00:21:25.430 --> 00:21:30.379
Jason Root: Or the tag of the branch, and it doesn't include every… like, this would have been helpful.

237
00:21:30.490 --> 00:21:38.830
Jason Root: if they appended this each time, but they didn't, so I'm not sure what this file does on their actual release.

238
00:21:38.830 --> 00:21:54.350
Jason Root: But I've combined them all, I've named the descriptors. I don't have a robot doing this yet, I need to figure out how to get a robot to do this. But every application descriptor I've found that I'm actually enabled for the platform, I move it into my own platform

239
00:21:54.350 --> 00:22:02.909
Jason Root: repo that's private, and I have it all here as the source of truth, and I make sure it's cleaned up and it actually works, you know, all my URLs.

240
00:22:03.030 --> 00:22:07.789
Florian Kreft (LRZ): Why do you need to do that yourself? This should be the official one. I really don't.

241
00:22:07.790 --> 00:22:18.429
Jason Root: Right. This was my… this was my test environment that I knew was going to work every time and be sane, so I wanted one place for all of my sanity to live.

242
00:22:19.100 --> 00:22:25.520
Jason Root: And I can't. I have tore down and reached stood up this system, like, 4 or 5 times now, and it works now, every single time.

243
00:22:25.700 --> 00:22:26.090
Florian Kreft (LRZ): outfit.

244
00:22:26.090 --> 00:22:29.350
Jason Root: So I know this works, and I know God.

245
00:22:29.350 --> 00:22:32.220
Florian Kreft (LRZ): Don't make that public, right? That would be extremely helpful.

246
00:22:32.220 --> 00:22:35.400
Jason Root: Yeah, sadly, yeah, I cannot make this public.

247
00:22:35.400 --> 00:22:36.270
Florian Kreft (LRZ): Devin.

248
00:22:36.270 --> 00:22:37.760
Shelley Doljack: Ours is public.

249
00:22:37.760 --> 00:22:38.670
Jason Root: Yeah, this…

250
00:22:38.670 --> 00:22:41.409
Shelley Doljack: But we did the same thing like you did here.

251
00:22:42.450 --> 00:22:46.620
Jason Root: Yeah, and you just made it all in a single… a single repo? Yeah. Yeah.

252
00:22:46.620 --> 00:22:50.680
Florian Kreft (LRZ): Can you link that to me? I should probably already have that link.

253
00:22:50.680 --> 00:23:01.430
Jason Root: And I think… I don't know if anything in here is actually, like, index data proprietary. I can go through it and find out, and then… I could probably zip all this up, because, I mean, these are all…

254
00:23:01.530 --> 00:23:14.959
Jason Root: you know, other than our scripts and our workflows, everything else in here is, from, Eureka, release, you know, from the actual public. You know, and I even pulled in their README, you know, and everything, so…

255
00:23:15.340 --> 00:23:20.030
Jason Root: It's all… It's all basically… out there.

256
00:23:20.030 --> 00:23:27.989
Florian Kreft (LRZ): the same with our Stripes build, and I mean, we got a bit sloppy with the actual module lists we use, because we used so many in different versions, we built.

257
00:23:27.990 --> 00:23:45.480
Jason Root: We make ours public in our good, but why should every self-host do this themselves? There should be, like, a sane minimal thing, at least, with, like, a possibility to, okay, here are the other things you might need, and… Right, it would be nice…

258
00:23:45.480 --> 00:23:48.489
Shelley Doljack: And you would have to rely on the… I don't know.

259
00:23:48.490 --> 00:23:57.969
Jason Root: It would be nice if this platform LSP included all of that, you know, every app other than an application descriptor. Like, because I guess you can curl this stuff in based on…

260
00:23:58.240 --> 00:24:01.210
Jason Root: What's in… not that one, Jason, we'll go back.

261
00:24:01.450 --> 00:24:08.829
Jason Root: I guess you could curl this stuff in, what's in the platform file, right? Oh yeah, you have to go back to the tag.

262
00:24:08.960 --> 00:24:09.670
Jason Root: Gosh.

263
00:24:11.440 --> 00:24:12.850
Jason Root: Oop. Come on.

264
00:24:13.070 --> 00:24:15.599
Shelley Doljack: Yeah, but if they, like, include…

265
00:24:15.860 --> 00:24:22.329
Shelley Doljack: malformed JSON that we can't… Yeah. Or invalid JSON that doesn't work.

266
00:24:22.330 --> 00:24:23.309
Florian Kreft (LRZ): Do they do that?

267
00:24:23.310 --> 00:24:26.529
Shelley Doljack: title, then… You can't use it, right?

268
00:24:26.530 --> 00:24:41.309
Jason Root: Yeah, well, like the JSON they gave us for one of the apps, and I still can't remember which one it was, like you said, if it's not… if it doesn't meet the validation spec, you know, it doesn't work. And there is. There are applications out there that have

269
00:24:41.420 --> 00:24:44.350
Jason Root: A URL for the UI modules, and it's like.

270
00:24:44.350 --> 00:24:53.700
Florian Kreft (LRZ): So, in that case, if you have descriptors like this, if you try to, what's the word, add them to the manager, whatever.

271
00:24:53.700 --> 00:24:54.679
Jason Root: You get an arrow?

272
00:24:54.950 --> 00:24:59.840
Florian Kreft (LRZ): That tells you at least what's wrong, or only that it's malformed?

273
00:25:00.350 --> 00:25:05.719
Shelley Doljack: There's actually a validation endpoint for one of the manager modules that…

274
00:25:05.930 --> 00:25:10.939
Shelley Doljack: you could use if you wanted to. And I think it does tell you, like, which module.

275
00:25:10.940 --> 00:25:11.480
Florian Kreft (LRZ): Okay.

276
00:25:11.480 --> 00:25:17.420
Shelley Doljack: It's erroring on, because we had a problem with Like, the wrong interface.

277
00:25:17.740 --> 00:25:23.700
Shelley Doljack: we put in, because we had to put together app reporting, or app Z3950.

278
00:25:24.250 --> 00:25:30.469
Shelley Doljack: I think it was AppZ… ApZ3950.

279
00:25:30.670 --> 00:25:33.400
Shelley Doljack: Doesn't have, like, we had a…

280
00:25:33.890 --> 00:25:36.130
Shelley Doljack: Put that together ourselves, or something.

281
00:25:36.130 --> 00:25:45.160
Jason Root: Yeah, there's… that was a mistake that came about with Sunflower, and I think they're working with Mike Taylor to get that resolved properly.

282
00:25:45.280 --> 00:25:49.620
Jason Root: They didn't… Yeah, they didn't have that released right.

283
00:25:50.960 --> 00:25:58.479
Shelley Doljack: Yeah, so whatever we put here, we got here, it… we got it to, entitle and validate.

284
00:26:01.750 --> 00:26:02.800
Florian Kreft (LRZ): Okay, okay.

285
00:26:03.510 --> 00:26:11.009
Florian Kreft (LRZ): Okay, so the Folio Eureka SULDLSS, there are the application descriptors as well, somewhere?

286
00:26:11.440 --> 00:26:12.750
Shelley Doljack: Yeah, so.

287
00:26:12.750 --> 00:26:13.680
Florian Kreft (LRZ): Oh, okay.

288
00:26:13.680 --> 00:26:17.499
Shelley Doljack: Let me… let me share my screen.

289
00:26:20.450 --> 00:26:22.230
Shelley Doljack: Let me go back to…

290
00:26:24.180 --> 00:26:32.250
Shelley Doljack: Okay, so we have, we got it split out for each of our environments, and, so folio test, we put

291
00:26:32.790 --> 00:26:38.900
Shelley Doljack: all the application descriptor files, and these are all for Sunflower CSP4.

292
00:26:41.260 --> 00:26:46.990
Shelley Doljack: Under here, we have our YAMLs for the infrastructure.

293
00:26:47.540 --> 00:26:48.189
Shelley Doljack: Yeah, I was like…

294
00:26:48.190 --> 00:27:01.879
Florian Kreft (LRZ): So, we've looked at this report, I think you shared the link, thankfully, at some point. I think this is also… so, sadly, my colleague is not here. I think next time, I need to make sure he's here and can maybe ask direct questions, because

295
00:27:01.880 --> 00:27:08.430
Florian Kreft (LRZ): Luckily, I'm not the only person working on Folio, and I'm still working on Okapi Sunflower upgrades at the moment, which…

296
00:27:08.750 --> 00:27:24.790
Florian Kreft (LRZ): work fine now, but, like, quite some tenants I need to upgrade, and we need to, like, make sure all our environments are clean and stuff like this, so I'm not too involved with Eureka yet, more theoretical. But he definitely looked at this, repository.

297
00:27:24.790 --> 00:27:25.240
Shelley Doljack: Yeah.

298
00:27:25.240 --> 00:27:25.720
Florian Kreft (LRZ): Yep.

299
00:27:26.590 --> 00:27:30.579
Florian Kreft (LRZ): He also has some feedback that might be helpful. Yeah.

300
00:27:30.580 --> 00:27:41.109
Shelley Doljack: I want to point out that the vault example from the Folio Wiki Kubernetes example deployment is… uses ephemeral storage.

301
00:27:41.660 --> 00:27:42.580
Florian Kreft (LRZ): Yeah, yeah, so, so that…

302
00:27:42.580 --> 00:27:44.470
Shelley Doljack: Not for pronounce it.

303
00:27:45.010 --> 00:28:01.859
Florian Kreft (LRZ): My colleague has worked with Vault, and he has invested in the last two weeks quite some time to get a sensible Vault set up for us, and also he has already made quite some changes with how we want to deploy Vault, and even how we want to, like.

304
00:28:01.860 --> 00:28:15.459
Florian Kreft (LRZ): structure the secret, which is also a complete mess in the Kubernetes example for Eureka. Like, there's, like, keys that are copied all over the place and whatever. It simply does not make sense at all. We can just tell the pods to use

305
00:28:15.460 --> 00:28:28.169
Florian Kreft (LRZ): like, have… going back to the idea of a single source of truth, have, like, one secret with the actual value, wherever you get that, you could even pull that in from Vault, or whatever, and then the point

306
00:28:28.170 --> 00:28:28.699
Florian Kreft (LRZ): What's canceled.

307
00:28:28.700 --> 00:28:52.040
Florian Kreft (LRZ): simply reference, you just need to be able, in the port deployments, whatever, maybe in the hand charts, to actually name the secret key you want to load in for that variable, and then you can have as many variable names. If, like, your different applications need different environment variables with the same value, that's fine, but in Kubernetes, there should still be only one secret with that

308
00:28:52.040 --> 00:28:55.089
Florian Kreft (LRZ): value. Otherwise, it's just not…

309
00:28:55.090 --> 00:28:55.480
Shelley Doljack: Yeah.

310
00:28:55.990 --> 00:28:58.679
Florian Kreft (LRZ): Handle… you can't handle that, yeah.

311
00:28:59.070 --> 00:29:11.910
Shelley Doljack: So, this file here, we have two vaults. We have a vault that runs outside of our cluster that we use for doing this vault static secret.

312
00:29:11.910 --> 00:29:12.540
Florian Kreft (LRZ): Yep.

313
00:29:12.540 --> 00:29:17.989
Shelley Doljack: And then we have the vault that's inside the cluster that's specific to Folio.

314
00:29:17.990 --> 00:29:20.110
Florian Kreft (LRZ): Yep, we will have the same setup.

315
00:29:20.460 --> 00:29:30.670
Shelley Doljack: Because it has the… the way that it has the paths structured for secrets… master… Or secrets, right?

316
00:29:30.670 --> 00:29:31.159
Florian Kreft (LRZ): Yep, yep.

317
00:29:31.160 --> 00:29:43.100
Shelley Doljack: So this is where we are keeping… like, Eureka Common, DB Credentials… What else? We also have…

318
00:29:44.140 --> 00:29:47.550
Shelley Doljack: Under modules, we have all of the modules.

319
00:29:49.350 --> 00:29:53.620
Shelley Doljack: We have the… this is our overrides file.

320
00:29:54.670 --> 00:29:55.380
Florian Kreft (LRZ): Joe.

321
00:29:55.380 --> 00:30:06.580
Shelley Doljack: And we're using FolioHelm V2, charts… actually, it's a fork of it. And this is, we're deploying all of our stuff using Argo CD.

322
00:30:07.400 --> 00:30:16.469
Shelley Doljack: And so, Argo CD puts together the values file that it finds in this repo, plus

323
00:30:16.840 --> 00:30:21.080
Shelley Doljack: this, folio V2 chart that we have forked.

324
00:30:23.770 --> 00:30:24.370
Florian Kreft (LRZ): Yep.

325
00:30:24.650 --> 00:30:25.580
Shelley Doljack: What else?

326
00:30:26.850 --> 00:30:32.340
Shelley Doljack: Yeah, some other stuff that… was added…

327
00:30:33.200 --> 00:30:35.920
Shelley Doljack: Let me just go back to the main page.

328
00:30:36.550 --> 00:30:41.070
Shelley Doljack: this part. So we discovered that…

329
00:30:41.110 --> 00:30:57.210
Shelley Doljack: when you're not using ephemeral storage, you… and you're using… and maybe your colleague already knows this about Vault, you have to… there's this whole process that you have to do to, like, unseal the vault, and say that you're using…

330
00:30:57.520 --> 00:30:58.730
Shelley Doljack: Version 2.

331
00:30:58.730 --> 00:31:04.640
Florian Kreft (LRZ): Actually, now that you show me this one again, we've talked about this, actually, my colleague,

332
00:31:04.810 --> 00:31:11.990
Florian Kreft (LRZ): Thought that, like, 4 put statements in a row, like this, should actually break things.

333
00:31:13.240 --> 00:31:22.719
Florian Kreft (LRZ): Like, it shouldn't be put, it should be update for all of the ones below the first one. Like, the last four lines in this box,

334
00:31:24.090 --> 00:31:25.870
Florian Kreft (LRZ): So, I mean, I guess it needs…

335
00:31:26.210 --> 00:31:33.759
Florian Kreft (LRZ): work for you, but he thought, like, we can't, like, test this out right now, but he thought this.

336
00:31:33.760 --> 00:31:38.610
Shelley Doljack: These are separate… these are separate keys to manage your applications.

337
00:31:38.610 --> 00:31:52.339
Florian Kreft (LRZ): So as long as you actually put, like, even if you have the same secret folio master and have separate keys, this actually works. Okay, perfect. It doesn't really matter, it's just something that came to mind. If it works for you, it should be fine, but…

338
00:31:52.340 --> 00:31:58.840
Shelley Doljack: Yeah, no, if it's, like, if your secret contains… You know what, I don't…

339
00:31:59.240 --> 00:32:02.029
Shelley Doljack: I'm… I'm not sure now.

340
00:32:02.740 --> 00:32:03.570
Florian Kreft (LRZ): Yeah, so…

341
00:32:03.570 --> 00:32:05.640
Shelley Doljack: I'm pretty sure that these are…

342
00:32:06.160 --> 00:32:10.060
Shelley Doljack: I don't know, but I know you have to do patch, or else you're overriding stuff.

343
00:32:10.060 --> 00:32:16.880
Florian Kreft (LRZ): Exactly, and there are four put statements in this README, which is… Yeah.

344
00:32:17.880 --> 00:32:33.940
Shelley Doljack: Let's see, I recently added this Python script to make entitlement, like, not… right now, we were, like, copying and pasting and copying and pasting, and got tired of that. So, this Python script…

345
00:32:34.270 --> 00:32:41.009
Shelley Doljack: we build inside, a pod that we put inside the cluster, and then we run this stuff.

346
00:32:41.010 --> 00:32:44.140
Florian Kreft (LRZ): have the script available inside the namespace here. Makes sense.

347
00:32:44.140 --> 00:32:53.819
Shelley Doljack: Yeah, so, and I kind of have it, like, these flags are meant to be the steps that you have to do to entitle. You first register the applications.

348
00:32:53.820 --> 00:32:54.640
Jason Root: Oh, yeah.

349
00:32:54.640 --> 00:33:00.469
Shelley Doljack: Then you register the modules for discovery, and then you create the tenant, and then you entitle applications.

350
00:33:01.140 --> 00:33:05.970
Shelley Doljack: Perfect. And then this is meant to monitor the entitlement flow.

351
00:33:06.040 --> 00:33:08.039
Jason Root: This is cool. Yeah, that's really cool.

352
00:33:08.040 --> 00:33:10.310
Florian Kreft (LRZ): This is really helpful for us as well.

353
00:33:10.310 --> 00:33:10.750
Shelley Doljack: Oh, no.

354
00:33:10.750 --> 00:33:12.259
Florian Kreft (LRZ): Have a look at that, yeah.

355
00:33:14.190 --> 00:33:20.970
Shelley Doljack: And then, so this is just examples of each of the steps that you would do, register the applications…

356
00:33:21.400 --> 00:33:41.399
Shelley Doljack: This actually could be done… whoops, sorry. You could deploy it whenever, really. You don't have to do it in this order. I mean, you could deploy the backend modules before you register the app… manage your applications, it doesn't matter. But you have to have them there, in order to, you know, create entitlements.

357
00:33:42.450 --> 00:33:53.880
Shelley Doljack: And then this part is still up in the air, whether this is needed or not. Apparently, Jason doesn't… Jason was able to entitle stuff without having these…

358
00:33:54.120 --> 00:33:56.249
Shelley Doljack: Key Cloak system users in.

359
00:33:57.520 --> 00:33:59.360
Shelley Doljack: And I think maybe…

360
00:33:59.840 --> 00:34:06.310
Shelley Doljack: Maybe we've just figured out the bugs in our deployment so that we don't need this anymore.

361
00:34:07.060 --> 00:34:09.310
Shelley Doljack: I don't… I don't know.

362
00:34:09.310 --> 00:34:14.319
Jason Root: Well, and I've found that even in the reference environments that are out there for Eureka, that the…

363
00:34:14.630 --> 00:34:19.710
Jason Root: teams use, they don't have that either. None of those users exist in their key clones.

364
00:34:19.710 --> 00:34:34.200
Shelley Doljack: Yeah, it could be the case where we figured out where we need a set OCopy URL to localhost 8082, and that was the problem, and that we actually didn't need these users at all.

365
00:34:34.310 --> 00:34:47.419
Shelley Doljack: Let's see… This is an example of monitoring the entitlements flow, we've discovered you have to…

366
00:34:47.620 --> 00:34:50.560
Shelley Doljack: You have to entitle app platform minimal.

367
00:34:50.820 --> 00:34:53.560
Shelley Doljack: First, before you can entitle everything else.

368
00:34:53.969 --> 00:34:56.199
Ingolf Kuss: Is that your experience, Jason?

369
00:34:56.719 --> 00:34:59.590
Shelley Doljack: Or can you entitle them all at the same time?

370
00:34:59.590 --> 00:35:08.649
Jason Root: I have… I… I've been doing platform minimal first, and then updating the entitlement file.

371
00:35:08.840 --> 00:35:13.040
Jason Root: to have them all. And I… I do a,

372
00:35:14.230 --> 00:35:16.839
Jason Root: Let me see here if I can find my notes…

373
00:35:20.220 --> 00:35:23.850
Jason Root: There's a way to do…

374
00:35:24.040 --> 00:35:28.700
Jason Root: Instead of a new entitlement to just override what you have currently?

375
00:35:29.330 --> 00:35:29.980
Shelley Doljack: Hmm.

376
00:35:30.740 --> 00:35:32.089
Jason Root: And what was that endpoint?

377
00:35:32.090 --> 00:35:39.150
Ingolf Kuss: do that. Has anyone ever done a de-entitlement, a delete on an entitlement, and then entitle again?

378
00:35:39.500 --> 00:35:40.460
Shelley Doljack: Yes.

379
00:35:40.460 --> 00:35:41.130
Ingolf Kuss: Okay.

380
00:35:41.450 --> 00:35:43.029
Shelley Doljack: We, we did that…

381
00:35:43.240 --> 00:35:56.789
Jason Root: A reinstall. I'm doing a re… so when I add a new application, I don't unentitle and re-entitle, I just do a reinstall, and then I have an updated Eureka entitlement JSON file that I… that I put to it.

382
00:35:58.400 --> 00:36:10.699
Florian Kreft (LRZ): If you want to do upgrades, just a question here, if you do upgrades, does this also trigger the modules to do, like, database migration stuff and stuff like this? Is this the entitlement step, or is this something I'm thinking ahead?

383
00:36:11.110 --> 00:36:15.100
Jason Root: So, like, in, like, the historical,

384
00:36:15.410 --> 00:36:18.619
Jason Root: OCopy environment when you do, like, a reinstall.

385
00:36:18.620 --> 00:36:19.170
Florian Kreft (LRZ): Yep.

386
00:36:19.170 --> 00:36:27.480
Jason Root: and it puts in, the appropriate, like, sample data and reference data and database-type stuff. That's kind of what this… this is.

387
00:36:27.480 --> 00:36:32.120
Florian Kreft (LRZ): Okay, so it's, like, should be roughly equivalent in what it should do.

388
00:36:32.120 --> 00:36:43.620
Jason Root: Yes. Okay. Yeah, it's the… it's… but it's the… it's the reinstall endpoint, and then you give it, reinstall, I think it's reinstall slash applications, let me find my command.

389
00:36:43.620 --> 00:36:57.989
Shelley Doljack: When we started, we were going… our original plan was to upgrade… no, to go Okapi Ramson's to Eureka Ramson's, and then upgrade to some Eureka Sunflower.

390
00:36:58.390 --> 00:37:06.940
Shelley Doljack: And we found that when we tried to upgrade from Eureka Ramsons to Eureka Sunflower, we had…

391
00:37:07.440 --> 00:37:15.889
Shelley Doljack: We did a put to the entitlement's endpoint, and it gave us problems, because… app acquisitions.

392
00:37:16.140 --> 00:37:17.250
Jason Root: Was a new app.

393
00:37:18.710 --> 00:37:21.000
Florian Kreft (LRZ): Yeah, okay, that seems really weird.

394
00:37:21.000 --> 00:37:28.490
Shelley Doljack: And so, the recommendation was to delete entitlements, and then post entitlements. Okay.

395
00:37:30.010 --> 00:37:40.629
Shelley Doljack: And we looked at the logs for the modules, and we saw that there was liquid-based migrations happening, so it was kind of like, okay, they upgraded themselves. Okay, so…

396
00:37:41.200 --> 00:37:42.130
Florian Kreft (LRZ): But the…

397
00:37:42.660 --> 00:37:49.210
Jason Root: It seems like a bug in general, it's like a good heads up, but I think this will be fixed at some point, I hope. Yeah, certainly.

398
00:37:49.210 --> 00:38:04.439
Jason Root: The core of the bug is still being addressed, and what the bug effectively is, is you cannot have the same versions of modules belong to two separate applications, or even

399
00:38:04.540 --> 00:38:08.820
Jason Root: to… Separate versions of the same application.

400
00:38:10.750 --> 00:38:22.080
Jason Root: And when Ramson's was released, not all of the applications were formalized yet, so they threw a bunch of those modules under a more broad application, and they're still splitting them out.

401
00:38:22.140 --> 00:38:28.050
Florian Kreft (LRZ): Oh, which makes sense, but then there has to be, like, a path to actually change to the better, improved.

402
00:38:29.120 --> 00:38:29.690
Jason Root: Yeah.

403
00:38:29.690 --> 00:38:30.610
Florian Kreft (LRZ): questioning, yeah.

404
00:38:30.870 --> 00:38:44.109
Jason Root: What we're seeing that is working so far is people are getting their local environments to sunflower okapi, and then moving that to sunflower Eureka.

405
00:38:44.110 --> 00:38:54.809
Florian Kreft (LRZ): Yeah. Okay, that will be, I think, our approach, in maybe some months, but, yeah. The first step is, like, to get, like, maybe a running Eureka,

406
00:38:55.000 --> 00:38:58.029
Florian Kreft (LRZ): Front-end where you can log in and then see the actual…

407
00:38:58.040 --> 00:38:58.630
Jason Root: Hmm, it's stop.

408
00:38:58.630 --> 00:39:09.980
Florian Kreft (LRZ): without data, that would be the first step. So, on that note, actually, so since we were not involved at all in the beginning, what are the currently used proper communication channels in Slack?

409
00:39:10.680 --> 00:39:21.409
Florian Kreft (LRZ): We weren't sure about that, and maybe we need some invitations here, but because I think now we actually have capacity to, like, work on troubleshooting stuff, and.

410
00:39:21.410 --> 00:39:25.890
Jason Root: Yeah, let me invite you to that channel. There's a private Eureka Adopters channel.

411
00:39:26.180 --> 00:39:40.319
Florian Kreft (LRZ): Yeah, yeah, please invite me to that, and I will invite my colleague, he will definitely join this session, the SysOps one. I'm not sure if there are, like, other meetings that might be helpful for us to join, since we're now, like, trying stuff out as well.

412
00:39:42.880 --> 00:39:45.179
Jason Root: Yeah, let me see here. Sorry.

413
00:39:45.180 --> 00:39:47.180
Ingolf Kuss: Moving on to the phone.

414
00:39:47.940 --> 00:39:55.359
Shelley Doljack: I wanted to say that for, for us, we were… our…

415
00:39:55.850 --> 00:40:02.849
Shelley Doljack: We were able to upgrade to Sunflower Eureka from Okapi Ramson's.

416
00:40:03.240 --> 00:40:07.579
Shelley Doljack: In our dev environment, and that's what we plan to do for production.

417
00:40:07.580 --> 00:40:08.180
Jason Root: Oh, good.

418
00:40:08.180 --> 00:40:10.589
Shelley Doljack: We're not gonna do this interstitial…

419
00:40:10.840 --> 00:40:18.990
Shelley Doljack: Eureka Ramsons to Sunflower Ramsons. We're just going Okapi Ramsons to… Sunflower eureka.

420
00:40:18.990 --> 00:40:24.859
Jason Root: Okay. And doing the… Role and permissions migration during that whole…

421
00:40:25.450 --> 00:40:25.850
Shelley Doljack: Fees.

422
00:40:25.910 --> 00:40:26.680
Jason Root: Yeah, yeah.

423
00:40:26.680 --> 00:40:27.290
Shelley Doljack: Yeah.

424
00:40:28.990 --> 00:40:36.700
Jason Root: I think… I think we are going to do Sunflower Ocopy to Sunflower Eureka, just because we won't have to redeploy

425
00:40:36.870 --> 00:40:39.739
Jason Root: Back-end modules again, if that makes sense?

426
00:40:40.470 --> 00:40:41.310
Shelley Doljack: Yeah.

427
00:40:42.530 --> 00:40:43.310
Shelley Doljack: And that's.

428
00:40:43.310 --> 00:40:50.410
Jason Root: why we decided to do it that way, in the same namespace, right? We didn't want to overflow all of the namespaces with Ramson's and…

429
00:40:50.620 --> 00:40:53.680
Jason Root: sunflower versions of modules, and how.

430
00:40:53.680 --> 00:40:54.300
Shelley Doljack: Yeah.

431
00:40:54.300 --> 00:40:56.860
Jason Root: Or worry about resourcing and all of that jazz.

432
00:40:59.140 --> 00:41:03.730
Shelley Doljack: I could continue showing this

433
00:41:04.040 --> 00:41:23.000
Shelley Doljack: I guess the last thing really is that there's also… it's similar to that bootstrap super user PerlScript that we've all used. It's this one, to, you know, create your admin user and create an admin role and get all the capabilities.

434
00:41:23.000 --> 00:41:34.799
Florian Kreft (LRZ): Yeah, but this is, like, tenant, based, right? Or is this, like, like, the super user bootstrapping thing was, like, not tenant-specific, which would be, I think…

435
00:41:35.100 --> 00:41:40.640
Shelley Doljack: This is actually… It is tenant-specific.

436
00:41:40.640 --> 00:41:41.680
Florian Kreft (LRZ): Yeah, okay, okay.

437
00:41:41.680 --> 00:41:48.309
Shelley Doljack: I don't think there is… your Key Cloak admin user, that is your.

438
00:41:48.310 --> 00:41:48.770
Florian Kreft (LRZ): give.

439
00:41:48.770 --> 00:41:49.900
Shelley Doljack: specific user.

440
00:41:49.900 --> 00:41:59.069
Florian Kreft (LRZ): Makes sense, that's… but I think the bootstrapping thing was for the super user, right? Maybe I'm misremembering the Okapi script, actually, because we are doing our own thing for so long.

441
00:41:59.390 --> 00:42:08.440
Jason Root: It can be… it can be any tenant, quote-unquote, admin user it can be used for. It can be for the superintendent, or it can be for whatever tenant you.

442
00:42:08.440 --> 00:42:25.819
Florian Kreft (LRZ): Okay, because… okay, our bootstrapping… we've since moved to our own bootstrapping scripts for our own admin users, which have quite a lot of stuff we need, like, what kind of permissions to actually add and stuff like this, so, yeah, okay.

443
00:42:26.450 --> 00:42:27.470
Shelley Doljack: Yeah.

444
00:42:28.160 --> 00:42:32.039
Shelley Doljack: Sorry, I was just scrolling through here to see if there was anything else.

445
00:42:32.180 --> 00:42:50.859
Florian Kreft (LRZ): No, so in general, we've seen this README, and the script seems really helpful, then we can, like, compare what we need to do. Like, like I said, we have, like, now, just earlier this week, set up the, like, general infrastructure stuff, including key clock, vault, and stuff like this, and setting up the secrets and the manager.

446
00:42:50.940 --> 00:42:59.120
Florian Kreft (LRZ): pods, I think, at least are running. I need to check that again, but I will probably get to that next week, and then…

447
00:42:59.190 --> 00:43:18.219
Florian Kreft (LRZ): like I said, we are now then also just, like, bootstrapping one very small sample user in the beginning, and then we will, like, try to formalize what are the next steps to take, or what do we want. So I'm pretty sure we want, at some point, to upgrade from Sunflower Carpi to Sunflower Eureka, so I think this will be…

448
00:43:18.220 --> 00:43:23.239
Florian Kreft (LRZ): Basically, for all the tenants we have that we want to migrate will be the, like.

449
00:43:23.670 --> 00:43:28.589
Florian Kreft (LRZ): Version flow, and so that's definitely something we need to test.

450
00:43:31.930 --> 00:43:34.190
Florian Kreft (LRZ): Okay, thank you.

451
00:43:35.300 --> 00:43:45.239
Florian Kreft (LRZ): Thank you, Jason. You were very helpful. That's a lot, that's a lot to take in, Florian. Good luck. Yeah, so we've looked at the, like,

452
00:43:45.370 --> 00:43:46.670
Florian Kreft (LRZ): the… the…

453
00:43:46.920 --> 00:44:11.799
Florian Kreft (LRZ): repository Shelly provided again, and we've definitely looked at all the, like, the Kubernetes example deployment, and I think we've figured quite some stuff out, and like… like I said, we restructured the secrets for us, and our wall setup is also different, like, that's one of the main things, like, my new colleague really complained about, like, the Kubernetes example. Okay, it says it's an example.

454
00:44:11.800 --> 00:44:23.989
Florian Kreft (LRZ): and it's a development setup, but, like, there's, like, a root key, which is simply root and stuff like this. It's, like, really, really bad. It's also not obvious what needs to be changed here for this to be production-ready, and it's, yeah.

455
00:44:24.050 --> 00:44:26.770
Florian Kreft (LRZ): It's not a good documentation at all.

456
00:44:26.770 --> 00:44:27.120
Jason Root: Yeah.

457
00:44:27.120 --> 00:44:28.220
Florian Kreft (LRZ): And it was…

458
00:44:28.220 --> 00:44:32.440
Jason Root: Yeah, all of the default passwords definitely change.

459
00:44:32.630 --> 00:44:34.620
Jason Root: Yeah, yeah. Except there's.

460
00:44:34.620 --> 00:44:36.009
Shelley Doljack: One that you can't.

461
00:44:36.600 --> 00:44:37.470
Jason Root: Yeah.

462
00:44:37.470 --> 00:44:51.829
Florian Kreft (LRZ): Which one can't you change? I remembered that, and then I thought, okay, if we run into that, we will run into that and see it. Do you remember which secret you can't change? It's one of the Keycloak admin ones, I think, right, Shelly?

463
00:44:52.190 --> 00:44:57.800
Shelley Doljack: Yeah, I'm trying… I'm li- I'm trying to find what it is.

464
00:44:57.800 --> 00:45:04.109
Jason Root: You can… I think you can technically change it, but you'll have to update, like, all of the secrets everywhere, and then all of the…

465
00:45:04.600 --> 00:45:07.170
Shelley Doljack: No, it's something with its built-in with.

466
00:45:07.170 --> 00:45:09.469
Jason Root: Yeah, it's one of the scripts, right?

467
00:45:09.470 --> 00:45:10.220
Shelley Doljack: I'm old.

468
00:45:10.220 --> 00:45:11.229
Florian Kreft (LRZ): That bootstraps it.

469
00:45:11.710 --> 00:45:16.179
Florian Kreft (LRZ): Is it in the folio image of Keycloak, then, or something?

470
00:45:16.180 --> 00:45:16.630
Shelley Doljack: Yes.

471
00:45:16.930 --> 00:45:20.020
Florian Kreft (LRZ): Oh yeah, okay, that makes sense. We've looked at that as well.

472
00:45:21.000 --> 00:45:29.869
Jason Root: And you… it's supposed to grab the password out of the secret that gets passed in and put it there in the script, but it does not do that.

473
00:45:30.000 --> 00:45:31.729
Florian Kreft (LRZ): Oh, interesting, thank you.

474
00:45:32.710 --> 00:45:36.469
Shelley Doljack: Casey HTTPS key store password.

475
00:45:36.980 --> 00:45:37.560
Florian Kreft (LRZ): Yep.

476
00:45:37.770 --> 00:45:39.180
Shelley Doljack: That's the one that is…

477
00:45:39.180 --> 00:45:40.450
Jason Root: Yeah, and it's just set to, like…

478
00:45:40.450 --> 00:45:43.110
Shelley Doljack: It's really hard-coded as secret passwords. Yeah.

479
00:45:43.110 --> 00:45:50.920
Florian Kreft (LRZ): Yes, and I think this is one of the things we already maybe changed and got it to work, actually, but yeah.

480
00:45:50.920 --> 00:45:55.150
Shelley Doljack: Yeah, but have you restarted Key Cloak a couple times since you got.

481
00:45:55.150 --> 00:45:55.580
Florian Kreft (LRZ): So…

482
00:45:55.580 --> 00:45:55.910
Shelley Doljack: Right.

483
00:45:55.910 --> 00:46:14.139
Florian Kreft (LRZ): And this will also be interesting because, like, we don't want to use the Bitnami key clock thing at some point, so currently we already do that, but we need to figure out a way to not rely on Bitnami stuff at all, so maybe we might end up, like.

484
00:46:14.140 --> 00:46:36.900
Florian Kreft (LRZ): forking somehow the startup scripts, and maybe building our own key clock image to use with the, like, operator, at some point, so that's why… so we've actually looked at that, and also looked at what actually happens with the… what's, like, custom about the custom folio key clock image, because that's actually good to know for troubleshooting reasons, yeah, okay.

485
00:46:37.240 --> 00:46:44.829
Jason Root: Yeah, there's a few repos that you can point at, that are in the KeyCloak, like, Helm chart that we're using.

486
00:46:45.720 --> 00:46:50.110
Jason Root: And it can, it can change the,

487
00:46:51.120 --> 00:46:56.889
Jason Root: Docker Hub repo out of Bitnami, or into, like, the Bitnami legacy one, or whatever you want.

488
00:46:56.890 --> 00:46:57.400
Florian Kreft (LRZ): Yeah, yeah, yeah.

489
00:46:57.400 --> 00:46:57.940
Jason Root: it.

490
00:46:58.150 --> 00:47:01.539
Florian Kreft (LRZ): Yeah, exactly, so there's, like, quite a few possibilities, and I think

491
00:47:01.730 --> 00:47:06.580
Florian Kreft (LRZ): With all the Bitnami stuff, we have, already alternatives we need

492
00:47:06.580 --> 00:47:25.569
Florian Kreft (LRZ): to check out, so we are currently… I think we have done… I'm not sure what our state with open searches, but for open search, we use the operator, not the Bitnami ones, and then for KeyClock, we will probably end up using the operator, the official one as well, and then there's also Kafka, which has also an operator. You need to.

493
00:47:25.570 --> 00:47:25.910
Jason Root: Yeah.

494
00:47:26.080 --> 00:47:35.100
Florian Kreft (LRZ): move away from that as well. So this is all stuff we need to test, and… but we… interestingly enough, these are all technologies we use outside of Folio as well.

495
00:47:35.100 --> 00:47:35.580
Jason Root: Oh, good.

496
00:47:35.580 --> 00:47:48.400
Florian Kreft (LRZ): I guess. So we have, like, open search running for logs of our Kubernetes setup, and we have KeyCloak for OIDC, which we use to log into our Kubernetes, actually, so to access the API.

497
00:47:48.400 --> 00:47:49.689
Shelley Doljack: That's what we do, too.

498
00:47:49.690 --> 00:47:50.090
Florian Kreft (LRZ): Yeah, yeah.

499
00:47:50.090 --> 00:47:54.659
Jason Root: Yeah, yeah. I think the only Bitnami repo that Keycloak is using

500
00:47:54.920 --> 00:48:04.649
Jason Root: is the CLI for KeyCloak Config, and you can… I mean, you can pull that old… your own image down and change that repository, or you can use…

501
00:48:04.870 --> 00:48:16.189
Jason Root: The other official one that they have out there, but it's still… for us, it's still pulling it down, I believe, even though it's not in the legacy repo, and it's a specific tag.

502
00:48:16.300 --> 00:48:20.549
Jason Root: Right? They were… Bonami said they were moving everything to legacy, except latest.

503
00:48:20.750 --> 00:48:26.630
Jason Root: But this specific tag in the key cloak column chart still is working, and still is getting pulled in. Interesting, yeah.

504
00:48:26.630 --> 00:48:38.160
Florian Kreft (LRZ): Yeah, all of this is really weird. I think half a year it prompted us to install our own Harba instance, which we can use to, like, simply at least pull through all the images once.

505
00:48:38.160 --> 00:49:02.439
Florian Kreft (LRZ): We've not done that with everything because there's no need, but at some point, we want to have, like, all the image versions of stuff we run also in our own infrastructure, so obviously we still have a problem if there's, like, security fixes we need to apply, and then we don't have, like, an upstream source, but at least that prevents us from getting locked out of, like, the stuff being removed that we are still running, and then we run into problems and stuff like this.

506
00:49:02.700 --> 00:49:04.820
Florian Kreft (LRZ): It's a… Yeah.

507
00:49:05.330 --> 00:49:20.389
Shelley Doljack: I was gonna say that my colleague Josh has been really interested in, you know, rolling our own KeyCloak and, you know, just taking the customizations from Folio Key Cloak and applying them.

508
00:49:20.440 --> 00:49:27.290
Shelley Doljack: But I'm… I'm really against going that direction, because it's just more work. Yeah, it's one more thing.

509
00:49:27.290 --> 00:49:29.290
Jason Root: And you've got to care and feed, yeah.

510
00:49:29.290 --> 00:49:30.040
Shelley Doljack: Yeah.

511
00:49:30.040 --> 00:49:48.540
Florian Kreft (LRZ): Obviously, so for first thing, we definitely will use just the official image, that's, yeah, but it's still good to… so it seems like it's not really complicated, what they are doing to the key clock image, which also, like, makes sense, just, like, some additional stuff that's… that's needed for bootstrapping? Okay, fine.

512
00:49:48.570 --> 00:49:51.469
Florian Kreft (LRZ): Well, I don't think it's,

513
00:49:51.610 --> 00:50:07.269
Florian Kreft (LRZ): too much hassle to actually… if this is, like, properly in a pipeline, it's easy to upgrade, but you need to set that up once, at least, for your environment, which is additional work, yeah. If you can just use the official image, that's fine also.

514
00:50:08.370 --> 00:50:21.240
Shelley Doljack: Yeah, the thing about the folio key cloak image that I recently discovered, and it was related to trying to figure out this key store… or no, it was… we tried to change our admin password, and then we got into some…

515
00:50:21.640 --> 00:50:25.680
Shelley Doljack: issue where we were locked out. And,

516
00:50:25.860 --> 00:50:29.050
Shelley Doljack: Apparently, like, if you look at those scripts, they…

517
00:50:29.200 --> 00:50:37.110
Shelley Doljack: they always configure… when KeyCloak… when Folio KeyCloak starts up, it's always configuring the master realm.

518
00:50:37.390 --> 00:50:40.259
Shelley Doljack: Or it's trying to, or it's always using…

519
00:50:41.040 --> 00:50:45.570
Shelley Doljack: an admin user with that password.

520
00:50:46.210 --> 00:50:54.340
Shelley Doljack: I'm not sure if I'm explaining this correctly. There's, like, a way to start up KeyCloak that doesn't always configure the realm.

521
00:50:57.150 --> 00:50:58.410
Florian Kreft (LRZ): Set up realm roles.

522
00:50:58.410 --> 00:51:04.340
Shelley Doljack: I have to remind myself what I'm talking about by looking at the… Bolio Key Cloak.

523
00:51:05.260 --> 00:51:11.499
Shelley Doljack: It was something… Anyways… Where was that?

524
00:51:17.160 --> 00:51:19.090
Shelley Doljack: When it starts up.

525
00:51:24.740 --> 00:51:28.549
Shelley Doljack: One of the scripts it does is this Configure Realms script.

526
00:51:29.110 --> 00:51:30.880
Shelley Doljack: Before it starts.

527
00:51:31.180 --> 00:51:33.140
Shelley Doljack: Or a part of the startup.

528
00:51:35.390 --> 00:51:36.439
Florian Kreft (LRZ): No, no, no.

529
00:51:37.100 --> 00:51:41.850
Shelley Doljack: Yeah, so… It just seems weird, I guess.

530
00:51:43.680 --> 00:51:53.789
Florian Kreft (LRZ): So, that's something we will play around with, definitely, like, restarting this… maybe changing passwords, maybe we also still have problems, I think, with something…

531
00:51:54.410 --> 00:52:04.239
Florian Kreft (LRZ): regarding those tokens. Like I said, my colleague could explain that more. He has actually spent quite some time with, like, Vault, at least, and then also the KeyCloak setup.

532
00:52:04.770 --> 00:52:11.249
Florian Kreft (LRZ): Right. Already. I mean, that's probably something we won't get around to actually understanding.

533
00:52:12.150 --> 00:52:21.430
Florian Kreft (LRZ): So, even if, at some point, the process of installing Folio with Eureka might get better, it's still good to understand all the moving parts.

534
00:52:24.990 --> 00:52:25.830
Florian Kreft (LRZ): Okay.

535
00:52:26.170 --> 00:52:37.119
Florian Kreft (LRZ): like I said, thank you. Even also for the password, I will talk to my colleague about this, and then maybe he can join us next time. When will be the next meeting in Gulf?

536
00:52:37.750 --> 00:52:38.559
Ingolf Kuss: Two weeks.

537
00:52:38.560 --> 00:52:39.800
Florian Kreft (LRZ): In two weeks.

538
00:52:43.200 --> 00:52:47.390
Shelley Doljack: I'm gonna be… I'm not gonna be, here in 2 weeks.

539
00:52:48.390 --> 00:52:50.240
Shelley Doljack: I'm just saying that.

540
00:52:50.240 --> 00:52:51.030
Florian Kreft (LRZ): No. Okay.

541
00:52:51.520 --> 00:52:53.430
Jason Root: vacation, hopefully?

542
00:52:53.430 --> 00:52:54.540
Shelley Doljack: No, my…

543
00:52:54.540 --> 00:52:57.689
Florian Kreft (LRZ): My husband's getting his gallbladder taken out.

544
00:52:57.690 --> 00:52:59.320
Jason Root: Oh, no, yeah, oh, Godspeed.

545
00:52:59.790 --> 00:53:00.450
Jason Root: Yikes.

546
00:53:00.450 --> 00:53:01.620
Shelley Doljack: zone, yes.

547
00:53:01.620 --> 00:53:05.329
Florian Kreft (LRZ): I didn't understand it, but I'm not asking what it is.

548
00:53:05.330 --> 00:53:05.950
Shelley Doljack: Okay.

549
00:53:10.590 --> 00:53:14.880
Jason Root: It's one of those… It's one of those organs that no one really knows what it does.

550
00:53:14.880 --> 00:53:15.640
Shelley Doljack: Yeah.

551
00:53:19.040 --> 00:53:20.970
Ingolf Kuss: Gaian Blazer, okay.

552
00:53:20.970 --> 00:53:22.970
Shelley Doljack: Yeah, gallbladder. I don't know what it is.

553
00:53:22.970 --> 00:53:23.320
Florian Kreft (LRZ): Yep.

554
00:53:23.550 --> 00:53:24.730
Ingolf Kuss: Yeah, God bless you.

555
00:53:25.250 --> 00:53:26.000
Shelley Doljack: Yeah.

556
00:53:26.950 --> 00:53:29.830
Ingolf Kuss: Do you have stones in it?

557
00:53:30.350 --> 00:53:32.280
Shelley Doljack: It's not fun, apparently.

558
00:53:33.650 --> 00:53:34.330
Shelley Doljack: So…

559
00:53:35.160 --> 00:53:37.929
Jason Root: Yeah, any major surgery like that is never fun.

560
00:53:38.510 --> 00:53:39.310
Shelley Doljack: Yeah.

561
00:53:40.520 --> 00:53:45.619
Shelley Doljack: Yeah, so I won't be here in 2 weeks, but I'll be here the 2 weeks after, probably.

562
00:53:45.620 --> 00:53:46.190
Ingolf Kuss: Huh.

563
00:53:46.630 --> 00:53:47.750
Jason Root: You can always send Josh.

564
00:53:47.750 --> 00:53:48.190
Ingolf Kuss: Good.

565
00:53:48.190 --> 00:53:49.610
Jason Root: stead, right?

566
00:53:49.610 --> 00:53:51.710
Shelley Doljack: Yeah, he'll show up sometimes.

567
00:53:51.710 --> 00:53:52.650
Florian Kreft (LRZ): Okay.

568
00:53:54.450 --> 00:53:56.360
Ingolf Kuss: That's the 1st of April.

569
00:53:57.930 --> 00:53:59.090
Jason Root: In two weeks?

570
00:53:59.090 --> 00:54:00.580
Ingolf Kuss: Jokes out there only.

571
00:54:00.580 --> 00:54:02.360
Jason Root: Oh, oh, okay, yeah.

572
00:54:04.040 --> 00:54:08.889
Shelley Doljack: And maybe by that time, we'll be on, Eureka Sunflower.

573
00:54:09.110 --> 00:54:16.410
Shelley Doljack: We're still not sure… we're still tentative plan, because we're having an issue with mod and SIP.

574
00:54:16.550 --> 00:54:19.120
Shelley Doljack: The integration. Oh, right. Yeah, yeah.

575
00:54:20.120 --> 00:54:23.189
Jason Root: Yeah, there's… he put in that ticket, I think, too, right?

576
00:54:23.640 --> 00:54:23.980
Shelley Doljack: Yeah.

577
00:54:23.980 --> 00:54:27.280
Jason Root: I see at least some people are looking at it now, so that's good.

578
00:54:27.590 --> 00:54:28.090
Shelley Doljack: Yeah.

579
00:54:28.090 --> 00:54:32.810
Florian Kreft (LRZ): of issue? Just, can you maybe summarize that quickly, or is that not possible? I've not…

580
00:54:33.140 --> 00:54:42.450
Shelley Doljack: I can't really summarize it, because I didn't really work on it. But there's a… I'm trying…

581
00:54:43.390 --> 00:54:48.060
Shelley Doljack: It's mods… I'm trying to put the ticket in the…

582
00:54:48.540 --> 00:54:50.310
Florian Kreft (LRZ): Chat here.

583
00:54:52.040 --> 00:54:53.010
Ingolf Kuss: There.

584
00:54:54.440 --> 00:54:55.809
Florian Kreft (LRZ): Because we definitely…

585
00:54:55.810 --> 00:55:01.780
Shelley Doljack: Exocopy permissions header is empty on… sometimes on a request.

586
00:55:02.360 --> 00:55:03.900
Shelley Doljack: So…

587
00:55:03.990 --> 00:55:05.070
Florian Kreft (LRZ): Interesting.

588
00:55:07.170 --> 00:55:10.789
Shelley Doljack: It could be that our… our Edge NSIP

589
00:55:10.940 --> 00:55:14.570
Shelley Doljack: User doesn't have all the capabilities it needs.

590
00:55:15.320 --> 00:55:16.769
Shelley Doljack: Maybe that's it?

591
00:55:17.390 --> 00:55:19.310
Shelley Doljack: I gotta look into that.

592
00:55:21.090 --> 00:55:29.589
Shelley Doljack: Although that wouldn't make sense if it, like, works sometimes, and then it doesn't… it's intermittent that it is missing this permissions header, so I don't know.

593
00:55:29.590 --> 00:55:35.830
Jason Root: Hey, yeah, that's the thing that's throwing me off, because if it didn't have the right capabilities, it wouldn't work any of the time.

594
00:55:36.120 --> 00:55:36.900
Shelley Doljack: Yeah.

595
00:55:36.900 --> 00:55:43.560
Jason Root: And I tried, I tried reinstalling mod settings and mod NSIP. That's a good idea, yeah.

596
00:55:43.560 --> 00:55:48.080
Shelley Doljack: And it didn't fix the problem. It reinstalled, apparently, but it didn't fix the problem.

597
00:55:48.230 --> 00:55:48.990
Jason Root: Okay.

598
00:55:49.130 --> 00:55:50.060
Jason Root: Good to know.

599
00:55:51.550 --> 00:55:54.819
Shelley Doljack: Unless it didn't really reinstall, I don't know, I tried to…

600
00:55:55.260 --> 00:55:59.790
Shelley Doljack: Figure out, like, how do you know it really reinstalled?

601
00:56:00.210 --> 00:56:01.509
Shelley Doljack: Because the law is…

602
00:56:01.510 --> 00:56:07.580
Jason Root: Yeah, usually there's a tenant init log in Kong, I think.

603
00:56:08.000 --> 00:56:19.659
Shelley Doljack: Yeah. No, I was looking at some log, and it was like, okay, I sent the request, but it never returned 201 or 204, because you know how when you entitle applications, you see, like.

604
00:56:19.850 --> 00:56:23.729
Shelley Doljack: Oh, it re… this application's entitled, it returned 204.

605
00:56:24.650 --> 00:56:26.749
Shelley Doljack: I didn't get that response back.

606
00:56:26.750 --> 00:56:28.260
Jason Root: Oh, that is interesting, yeah.

607
00:56:28.260 --> 00:56:35.580
Shelley Doljack: But when you, go to the endpoint to see what's entitled for the tenant, it… it's entitled.

608
00:56:35.850 --> 00:56:39.429
Shelley Doljack: For the tenant, so… I don't know.

609
00:56:44.570 --> 00:56:46.580
Shelley Doljack: Anyways, Eureka.

610
00:56:47.380 --> 00:56:49.519
Jason Root: Eureka! I know, what a name.

611
00:56:51.090 --> 00:56:55.489
Ingolf Kuss: Oh, no one shared the comic with me.

612
00:56:56.110 --> 00:56:57.040
Ingolf Kuss: Let's see, like…

613
00:56:58.600 --> 00:57:05.520
Jason Root: I guess in golf, next time we meet, we should probably talk about WolfCon sessions for SysOps to be involved with.

614
00:57:06.660 --> 00:57:09.010
Ingolf Kuss: Yeah, of course, sounds good.

615
00:57:09.680 --> 00:57:11.059
Florian Kreft (LRZ): I think there's one question.

616
00:57:11.320 --> 00:57:22.520
Florian Kreft (LRZ): Jason, my new colleague, if I… I don't know if he even has a Slack account yet, I can just hit you up with a message to be added to the Eureka, thing, right? It's not like…

617
00:57:23.580 --> 00:57:26.499
Jason Root: Yes, yeah, as soon as his… as soon as his…

618
00:57:27.090 --> 00:57:36.579
Jason Root: You know, the way the Slack integration works, you know, there's the invite thing and all of that into the org, which gets approved. I think Peter Murray mainly handles that.

619
00:57:36.700 --> 00:57:47.500
Jason Root: And then for the GitHub access to all of the different things, for Folio.org, I handle those requests. You know, if y'all have a team, you know, a GitHub team.

620
00:57:47.960 --> 00:57:53.610
Jason Root: If he has GitHub handle, I can add him to that. And then, yeah, the last piece is I can just invite him to the channel.

621
00:57:54.440 --> 00:57:59.530
Florian Kreft (LRZ): Yeah, okay, I don't think we even ever contributed to the GitHub.

622
00:57:59.530 --> 00:58:02.560
Jason Root: Okay, okay, no worries. Yeah, they're not linked, but yeah, all of.

623
00:58:02.560 --> 00:58:19.709
Florian Kreft (LRZ): Yeah, so I think that maybe it would be a good idea at some point, but we had our own GitLab for version control anyway, and if we have something to share, we can do that as well, but it might make sense to maybe, like, combine efforts more at some point in the future.

624
00:58:19.710 --> 00:58:25.599
Jason Root: Okay. Oh, good. Ingolf, this is great. The opposite of Eureka. Nice.

625
00:58:27.530 --> 00:58:28.389
Jason Root: That's pretty good.

626
00:58:28.390 --> 00:58:30.409
Ingolf Kuss: I did a colleague shout this.

627
00:58:30.410 --> 00:58:31.870
Florian Kreft (LRZ): the perfect…

628
00:58:37.110 --> 00:58:40.199
Jason Root: Alright, well, I'm gonna hop off, everybody. I think…

629
00:58:40.420 --> 00:58:55.720
Florian Kreft (LRZ): Yeah, thank you, thank you for all the information and stuff like this. Now we can properly get started on, like, testing stuff out, maybe finding some issues you might be not aware of, but probably mostly benefiting from all your troubling… troubleshooting.

630
00:58:55.720 --> 00:59:08.789
Jason Root: I definitely think the more eyes on this and more hands involved, the better off we'll be in the long run, because we can get issues filed and make sure they get addressed by the appropriate teams, and so on and so forth, so…

631
00:59:09.190 --> 00:59:10.379
Florian Kreft (LRZ): Yep, perfect. Thanks.

632
00:59:10.380 --> 00:59:10.890
Jason Root: Cool.

633
00:59:11.280 --> 00:59:13.030
Shelley Doljack: Alright. Okay.

634
00:59:13.030 --> 00:59:14.280
Ingolf Kuss: Thank you.

635
00:59:14.280 --> 00:59:15.869
Jason Root: Oh, thank you. Adios, everybody.

636
00:59:16.440 --> 00:59:18.009
Ingolf Kuss: In 2 weeks, but…