12:01:22 #startmeeting barbican 12:01:23 Meeting started Tue Jun 19 12:01:22 2018 UTC and is due to finish in 60 minutes. The chair is redrobot. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:01:24 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:01:26 The meeting name has been set to 'barbican' 12:01:32 #topic Roll Call 12:01:34 o/ 12:01:42 o/ 12:02:29 hi Luzi! 12:02:37 hi redrobot 12:02:42 Let's wait a couple of minutes to see if anyone shows up 12:02:57 I'm filling in for alee, as he is on vacation for a couple of weeks. 12:03:28 okay, I am relativly new in here :) 12:04:29 Here is the agenda link: 12:04:33 #link https://wiki.openstack.org/wiki/Meetings/Barbican 12:04:42 aaaand it looks like it hasn't been updated in ages. 12:04:48 so we're just going to wing it. 12:05:53 Ok, I don't think anyone else is coming... 12:05:57 #topic New Meeting Time 12:06:14 I like this new meeting time :D 12:06:41 I do too! 12:06:43 #link http://lists.openstack.org/pipermail/openstack-dev/2018-June/131509.html 12:07:07 link above is for the ML message. Hopefully everyone saw it. Just linking it here for folks who read the meeting minutes after the fact 12:07:50 maybe it should be updated here: http://eavesdrop.openstack.org/#Barbican_Meeting 12:08:02 i also did read the ML 12:09:14 Hmm.. 12:09:20 Ade did send an update for that 12:09:22 #link https://review.openstack.org/#/c/576177/ 12:09:37 it looks like it's merged, but for some reason the website didn't update 12:09:56 I can follow up with the infra team to figure out why the website didn't update with that patch. 12:10:14 #action redrobot to follow up with infra team regarding the meeting time change on the eavesdrop website 12:10:29 ok, moving on 12:10:36 #topic Castellan as a base service 12:11:27 Looks like the TC has a good proposal for adding a Castellan-compatible key store as a base service 12:11:30 #link https://review.openstack.org/#/c/572656/ 12:11:38 I expect the current patch to be merged 12:11:54 although I don't remember of the top of my head how long the TC waits to merge these 12:14:03 Luzi, any questions about the Castellan base services patch? 12:14:10 * redrobot waves at raildo 12:14:12 no 12:14:20 o/ 12:14:27 hi raildo 12:14:53 hey Luzi :) how you doing? 12:15:17 ok, moving on 12:16:04 #topic Code Reviews 12:16:06 #link https://review.openstack.org/#/q/project:openstack/barbican+status:open 12:16:19 looks like the next patch in the OVO series is ready for review 12:16:25 please take some time to look over it 12:17:02 #link https://review.openstack.org/#/q/project:openstack/python-barbicanclient+status:open 12:17:11 there's a few barbicanclient patches ready for review as well 12:17:31 * redrobot needs to figure out how to get a dashboard with all projects in a single page on gerrit 12:17:57 nothing new in castellan to review, so I won't link that 12:18:45 #topic Bug Triage 12:19:43 just a reminder that every project except for Castellan is being tracked on Storyboard 12:19:51 #link https://storyboard.openstack.org/#!/project_group/81 12:20:40 #link https://bugs.launchpad.net/castellan 12:20:46 I did add a new bug for Castellan 12:21:24 after talking to raildo and reading the proposed Castellan-keystore base service spec, I'm starting to think that we should probably do away with the credentials factory in Castellan 12:21:55 and instead update the Barbican backend to get its credentials directly from the conf like the Vault backend does now. 12:22:24 any thoughts on that? 12:23:18 * redrobot hears crickets 12:23:35 well, imo the credentials factory make sense if it useful for the backends 12:24:20 if we currently have 2 backend options, barbican/vault, and it's only useful for barbican, well, that would be a sign that we need to fix/improve that 12:24:46 the problem I see with it is that people are likely to continue to pass end-user oslo-contexts into the backends. With the Barbican backend that has the side effect of making the user the owner of the secret, which is explicitly a bad thing if you read the Castellan-base-service proposal. 12:26:00 >>> Note that in the context of the base services set Castellan is intended only to provide an interface for services to interact with a key store, and it should not be treated as a means to proxy API calls from users to that key store. 12:26:35 We don't have to make a decision right now, but it's something to think about... 12:26:47 I don't have a final position at this point yet :P 12:26:52 haha 12:26:54 good 12:27:00 ok, moving on 12:27:04 #topic Open Discussion 12:27:09 anything else y'all want to talk about? 12:27:26 aes xts 512 12:27:45 nothing from my side 12:27:51 Luzi, what about it? 12:28:04 as far as i have read the code - barbican can only generate AES keys with a size of 256, right? 12:28:54 so when using aes xts the key is split, and a key with a size of 256 would only be effectiva as 128 12:29:47 Hmm... I can't remember off the top of my head. What happens when you set the bit length in an order to 512? 12:30:03 barbican cannot generate it 12:30:12 it just doesn't work 12:30:23 lame. seems like something Bbarbican should do 12:30:49 especially since aes keys of arbitrary lengths are easy to generate 12:30:51 exactly, when xts is choosen barbican should be able to generate 512 keys 12:31:05 Luzi, do you want to file a bug report and work on that? 12:31:11 I would like to try to contribute in that case 12:31:14 yes 12:31:18 awesome! 12:31:42 #action Luzi to add a story to Storyboard for adding AES 512 keys to barbican 12:31:50 but i am quite new, and it would be nice, to know how exactly storyboard works 12:32:52 you should be able to sign in with your Ubuntu One account 12:33:10 after that navigate to the Barbican project and add a new story 12:33:15 redrobot: what times are you here in IRC? so, when i have have questions i would come back here 12:33:22 okay, thank you 12:33:32 #link https://storyboard.openstack.org/#!/project/980 12:34:06 Luzi, ☝ 12:34:50 Luzi, I'm typically on ~7am-5pm CST 12:34:58 thank you 12:35:01 I also have a bouncer set up, so I'm always listening 12:36:01 any other topics for Open Discussion? 12:37:14 not from my side 12:38:44 ok, let's call it a day, then. 12:39:16 we all get 20 minutes back 😄 12:39:26 #endmeeting