14:30:06 #startmeeting Pulp Triage 2020-08-11 14:30:06 !start 14:30:06 #info fao89 has joined triage 14:30:06 Meeting started Tue Aug 11 14:30:06 2020 UTC. The chair is fao89. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:30:06 Useful Commands: #action #agreed #help #info #idea #link #topic. 14:30:06 The meeting name has been set to 'pulp_triage_2020-08-11' 14:30:06 fao89: fao89 has joined triage 14:30:31 #info ppicka has joined triage 14:30:31 !here 14:30:31 ppicka: ppicka has joined triage 14:30:52 #info daviddavis has joined triage 14:30:52 !here 14:30:52 daviddavis: daviddavis has joined triage 14:31:15 #info ggainey has joined triage 14:31:15 !here 14:31:15 ggainey: ggainey has joined triage 14:31:25 !next 14:31:26 #topic https://pulp.plan.io/issues/7299 14:31:26 fao89: 1 issues left to triage: 7299 14:31:27 RM 7299 - daviddavis - ASSIGNED - HTML in our json schema 14:31:28 https://pulp.plan.io/issues/7299 14:31:37 #info ttereshc has joined triage 14:31:37 !here 14:31:37 ttereshc: ttereshc has joined triage 14:31:46 #idea Proposed for #7299: accept and add to sprint 14:31:46 !propose other accept and add to sprint 14:31:46 fao89: Proposed for #7299: accept and add to sprint 14:31:55 +1 14:31:59 +1 14:32:00 #agreed accept and add to sprint 14:32:00 !accept 14:32:00 fao89: Current proposal accepted: accept and add to sprint 14:32:01 +1 14:32:02 fao89: No issues to triage. 14:32:08 #info dkliban has joined triage 14:32:08 !here 14:32:08 dkliban: dkliban has joined triage 14:32:08 +1 14:32:15 #info ipanova has joined triage 14:32:15 !here 14:32:15 ipanova: ipanova has joined triage 14:32:31 open floor: https://hackmd.io/@pulp/triage/edit 14:32:48 topic: How to version plugin_template? https://pulp.plan.io/issues/7273 14:33:05 yes, the date is a good way to go 14:33:20 date wfm 14:33:36 fao89: do you want to document this or should I? 14:34:00 date works, however is it compatible with our automation? 14:34:09 daviddavis, I added step 0 14:34:10 do we use the US date formate or the rest of the world? 14:34:10 or will it be done without it 14:34:14 you can review it 14:34:18 #info bmbouter has joined triage 14:34:18 !here 14:34:18 bmbouter: bmbouter has joined triage 14:34:28 #info x9c4 has joined triage 14:34:28 !here 14:34:28 x9c4: x9c4 has joined triage 14:34:31 I think yyyy.mm.dd is unambiguous 14:34:43 yes 14:34:45 let's do that 14:34:47 it's the same for the us and the rest of the world 14:34:48 +1 14:34:55 ISO stds ftw 14:35:18 ttereshc: I don't think we're actually releasing plugin_template so there's no automation 14:35:43 ah, so we won't release it, just assign a version? 14:35:51 we shouhld tag it though 14:35:58 agreed 14:36:05 Unix timestamp would also be unique And you can do it several times a day 14:36:09 ok 14:36:40 eh, unix timestamp is hard to parse 14:36:44 lol 14:36:45 we could add hour 14:36:48 x9c4: unix timestamp trips over DST (as in, there are timestamps that happen twice in one day) 14:37:19 I don't see us tagging multiple times per day but we could add another number or something 14:37:26 I think the tag should be more human friendly than the timestamp 14:37:27 i agree 14:37:31 concur 14:37:52 next topic: Spam: do we clean up comments and issues or just leave them for the spam script? 14:38:05 so like 2020.08.11 by default and then 2020.08.11-1 and so forth if needed 14:38:08 I personally clean the triage ones before the triage 14:38:30 daviddavis: +1 14:38:31 daviddavis: +1 14:38:32 that sounds good 14:38:41 daviddavis: +1 14:38:45 +1 14:38:52 fao89: if you notice them, I guess go for it, but I'm happy letting the script do its thing 14:38:57 ok so we can still clean up spam in redmine as we encounter it? 14:38:58 daviddavis: +1 indeed 14:39:42 if no one knows the answer, I can check with lubos 14:39:59 lmjachky, ^ 14:40:02 I heard from lubos that he needs this smtp thing resolved 14:40:32 you are allowed to remove them all 14:40:43 the script itself is working 14:41:02 I planned to send an email to pulp-dev that you are allowed to remove all spam content 14:41:04 lmjachky: yeah what do we need to do to get that script in production 14:41:15 but was not sure whether I am allowed or not 14:41:40 I am creating a ticket for IT whether it is possible to create for us a new email address that I can manage 14:42:11 or whether it is possible to create a temporary TXT record for me during the verification process when signing up for an SMTP relay service 14:42:17 oh man this is sounding complicated 14:42:39 do I have another option? 14:42:53 I think not; and I do not want to use my email address for sending notifications 14:43:01 may I use your email address? :) 14:43:06 probalby the official correct path, but waiting on a third part for a special case sounds like we'll be held up for a while 14:43:31 I agree to not use your email I wouldn't either 14:43:51 this is a no-reply email, I don't think it's important if it has @redhat.com at the end or gmail.com or protonmail.com 14:44:24 ok silly idea time: let's create pulpnoreply@gmail.com and configure pulp-infra@redhat.com as its backup 14:44:30 and then use gmail's smtp servers for now 14:44:49 I think it should be fine for IT to provide a no-reply e-mail, I'd just make it more specific, like pulp-no-reply@XXX 14:45:19 hmm hmm 14:45:29 yeah 14:45:30 ttereshc: I'm ok w/ that route but I don't think we should invest in cleaning up spam manually in the meantime 14:45:32 agree with ttereshc 14:45:52 I am almost done with the ticket 14:46:00 concur w/all of the above - spam has been hanging out while we trained the script, a few more days isn't going to hurt 14:46:07 meanwhile I will create pulpnoreply@gmail.com 14:46:14 and let's give IT a chance to give us a 'real' adr to use 14:46:19 and do the rest, so we can start using it now 14:46:25 lmjachky: I'm hearing we should wait for IT, if I'm ok w/ that 14:46:26 +1 not to invest time into manual cleanup 14:46:38 s/if I'm/and I'm/ 14:46:43 when we will obtain a replay from the It, then I will open a new PR against pulp-ci to chahnge the from email address 14:46:50 I'm also ok w/ that 14:47:09 ipanova: ggainey ttereshc ya'll make the decision, what do you want to advise 14:47:39 so, the consensus is -> open ticket -> register a new email address at gmail.com -> wait for the response from the IT -> update the email address if any response given 14:48:02 bmbouter: I'd say open the ticket, revisit if we don't have an answer by THurs - but I don't have a prob w/the above sequence either 14:48:14 otherwise, we are tight up for the pulpnoblablareply@gmail.com 14:48:24 the current suggestion works for me. I'm ok with any temp solution if we are still trying to get the official email, and +1 to not invest any time into manual cleanup 14:48:31 +1 14:49:00 last topic: Release 14:49:09 lmjachky: do you have everything you need to proceed? 14:49:50 lmjachky: please ensure the script runs carefully the first time somehow when it does the mass deleting 14:49:51 bmbouter, I need to add one secret to the github repository 14:49:53 and that is all 14:49:59 sure 14:50:07 I will reach out to you 14:50:18 lmjachky: sounds good let's start in #pulp-dev 14:50:36 put some nicks on the request we'll have to gpg exchange is my belief 14:51:53 last topic: Release 14:52:23 subtopic: release blockers 14:52:40 I moved the list of PRs up to that section 14:52:57 I thinkwe need a [yes] or [no] on each of these 14:53:09 https://github.com/pulp/pulpcore/pull/820 needs to be a yes due to immediate usage by galaxy_ng plugin 14:54:26 bmbouter: do you mean "yes it's a blocker", or "yes it's ready to be merged"? 14:54:46 yes it is a blocker 14:54:52 kk cool 14:55:53 820 is a blocker, 834 is a blocker because we need clear docs on SSL configs 14:56:23 I agree, and with 3.6 specifically the installer is switching to SSL only 14:56:38 so 835, 770, and 829 all have some variant of 'approved' or 'lgtm' on them 14:57:05 I think we can merge these 14:57:11 concur 14:57:24 they are not blcokers I think but +1 to merge since it's ready 14:57:24 I hear one +1 if there is a second I can click merge right now 14:57:32 ok that's two 14:57:36 * bmbouter clicks merge 14:58:01 834 - explanation looks reasonable to me 14:59:03 834 needs to be rebased 14:59:09 there are conflicts 14:59:19 dkliban, ^ 14:59:29 I'm asking for a change on https://github.com/pulp/pulpcore/pull/835/files#r468648012 so I marged it as [no] 14:59:35 I don't think it's a release blocker 14:59:38 oh bah, that must have just happened as a result of the other merges :) 14:59:45 ttereshc: i will rebase 834, but i am also waiting for the installer change to be finished so i can add more docs 14:59:57 yeah we'll need more docs I was thinking too 15:00:06 kk 15:00:10 ok, so we can mark it as blocked at the moment 15:00:15 dkliban, did you read mikedep333 status? 15:00:26 installer change won't be ready today 15:00:42 "Will not finish pulp_installer's support for letsencrypt in time for 3.6.0 release today" 15:00:44 fao89: i did ... we also met today 15:00:47 fao89: I am implementing the backup plan. 15:00:56 cool 15:01:46 I am testing / developing manually with letsencrypt's actual service on an AWS instance. Which is something we'd want instructions for, but have to test manually, anyway. 15:01:59 that sounds good 15:02:19 Rather writing CI code to sett up a test ACME protocol server. 15:02:33 mikedep333: thank you 15:02:44 let's make 7299 a blocker 15:02:52 it's an easy fix 15:03:03 +1 15:03:22 +1 15:03:43 who is fixing? fao and I are working on the uers/groups so I don't think we can 15:03:47 sure - it's sounding to me like "release on the 12th", just to get the last-minute changes merged? 15:03:55 (which is fine) 15:05:32 ggainey: I agree but if we move the goal to the 12th then what happens is it ends up being the 13th 15:05:40 I think the goal should still be EOB today 15:05:58 the last subtopic is the real question here 15:06:07 what is the plan 15:06:25 lol yeah 15:06:41 so this is a release with many significant changes, and we don't have an owner yet 15:08:25 I can do pulp_ansible and pulp_installer releases 15:08:45 you can't do those until pulpcore is done 15:08:52 bmbouter, do you plan to add rbac to the announcement email? 15:08:52 and i think the main questin is about pulpcore right now 15:09:08 i can do the pulpcore release 15:09:14 ttereshc: actually I don't, I removed it I'd like to chat up about why after we fid release owners 15:09:27 i will do the pulpcore release 15:09:46 I don't agree, IIRC you did it last time 15:10:10 i don't remember, but i really don't mind 15:10:31 ok 15:11:07 now we need a plan 15:11:22 I would volunteer but I have been on the critical path for a long time now I don't want to own the release 15:11:30 rbac has worn me out 15:11:41 def understandable 15:12:00 fao89: and I are still on it even with https://github.com/pulp/pulpcore/pull/820 15:12:10 ggainey: ty 15:12:47 any other release owners need to be identified I think maybe that's all of them? pulpcore, pulp_installer, and pulp_ansible ? 15:13:05 really the first two, but we'll need pulp_ansible asap also 15:13:06 i will do pulpcore adn pulp_file 15:13:11 I will end the open floor because it is holding me from working on the blocker 15:13:16 oh yeah pulp_file how could I forget you 15:13:36 I can help with pulp_file if needed 15:13:38 fao89: agreed I should explain the rbac positioning anyway 15:13:40 or pulpcore 15:13:48 dkliban, ping me if you prefer to share the load 15:13:53 ok 15:13:59 #endmeeting 15:13:59 !end