15:00:24 #startmeeting Stewardship SIG Meeting (2020-05-19) 15:00:24 Meeting started Tue May 19 15:00:24 2020 UTC. 15:00:24 This meeting is logged and archived in a public location. 15:00:24 The chair is decathorpe. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:24 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:24 The meeting name has been set to 'stewardship_sig_meeting_(2020-05-19)' 15:00:24 #meetingname stewardship-sig 15:00:24 The meeting name has been set to 'stewardship-sig' 15:00:24 #topic Knock Knock 15:00:34 \o 15:00:36 * cipherboy knocks back 15:00:45 who's there ;-) 15:00:48 #chair cipherboy 15:00:48 Current chairs: cipherboy decathorpe 15:01:24 decathorpe: I think Dinesh is stuck in a screen share today so he might not make it back to IRC 15:02:49 alright 15:03:16 * cipherboy pings sillebille in case he can visit 15:04:35 Okie dokie, I don't think Dinesh's joining us. 15:05:15 hey, sorry 15:05:30 hello! 15:05:32 \o hey mhroncok 15:05:33 Python 3.9.0b1 was released, I am swamped 15:05:34 #chair mhroncok 15:05:34 Current chairs: cipherboy decathorpe mhroncok 15:05:47 no problemo, python is more important than Java 15:05:48 let me know if you need some input from me 15:05:52 decathorpe: :D 15:05:56 mhroncok: Yeah! G'luck! I happen to like the new python upgrades :) 15:06:00 decathorpe: does that come on tshirts? 15:06:36 well, there is a rant on devel about my work, so I am a bit tired :( 15:08:04 FWIW, I think it would be difficult to handle python upgrades better 15:09:16 something for mood++: https://twitter.com/PratikNilange/status/1254250952535576577 15:09:42 #link https://pagure.io/stewardship-sig/issue/90 Agenda 15:10:05 #topic Botched retirement of jaf 15:10:20 #link https://pagure.io/releng/issue/9462 15:10:50 so jaf was "retired" 10 months ago but something fell over and the package remanined in fedora 32+, so I've asked releng to retire it properly 15:11:06 I didn't realize we needed it until today :( 15:11:26 If we need jaf, why are we completing the retirement? 15:11:53 because it was in a broken state :( 15:12:05 it was retired according to dist-git, but not according to koji 15:12:12 Ah ok 15:12:21 So if we drop it, nothing depends on it? 15:13:01 Guess not -- report page doesn't list it. 15:13:04 Good catch :) 15:13:15 no, we need it for jackson-module-jaxb-annotations 15:13:40 BTW I've retired around 250 packages yesterday, expect something to free up and/or break 15:13:40 decathorpe: But that's a dead package? 15:13:40 and for https://koschei.fedoraproject.org/package/dogtag-pki? 15:13:54 decathorpe: https://src.fedoraproject.org/rpms/jackson-module-jaxb-annotations/tree/master 15:14:09 mhroncok: no, I think you didn't break anything for us :) I would've complained otherwise ;) 15:14:58 cipherboy: not dead: https://src.fedoraproject.org/rpms/jackson-modules-base/blob/master/f/jackson-modules-base.spec#_54 15:15:11 just subsumed into jackson-modules-base package 15:15:40 Ahh ok 15:16:20 This makes sense :) 15:16:38 Sorry, I didn't have my morning caffeine 15:17:01 .moar coffee cipherboy 15:17:01 here cipherboy, have some more coffee 15:17:35 so I guess I'll file the requested tickets to get jaf again 15:18:19 moving on? 15:18:43 decathorpe: Works for me, thanks! Let me know if you need anything there. 15:18:52 will do 15:19:01 #topic OpenJDK 11 fedora 33 Change 15:19:50 I've tried to find a generic way of making javadoc-11 just ignore errors that were non-fatal with javadoc-8, but I came up against some walls 15:20:11 decathorpe: Yeah I checked the javapackaging repo, but they're not the ones adding xdoclint 15:20:40 no I think it's the new default with OpenJDK 11 15:20:45 which is ... just great 15:21:12 Would modifying https://src.fedoraproject.org/rpms/maven-javadoc-plugin/ to default to manually specifying -Xdoclint:None? 15:21:13 I did a rebuild of all* java packages against java 11 and it doesn't look so good https://copr.fedorainfracloud.org/coprs/g/java-maint-sig/java-11-default/monitor/ 15:21:56 Yeah, tomcatjss will get fixed soon, dogtag-pki/pki-core will get fixed when we figure out what we're doing. 15:22:05 cipherboy: I thought about it, but we'd have to patch the sources, I think. it might be worth it, though - just change one package instead of all of the broken ones 15:22:35 decathorpe: Yeah, I'd say it'd be worth it. ITs the same reason not to run, say, pylint in rpm build -- no point in it. 15:22:51 Upstream already shipped the code/docs, no point re-linting it. 15:23:07 exactly 15:23:12 I'll be devoting next month to fixing Java 11 issues, starting with Dogtag PKI and then moving down. 15:23:21 good luck! 15:23:29 I could add maven-javadoc-plugin to my todo list 15:23:43 I'll look at it after the meeting 15:23:56 Yeah, we probably want it before Jiri's next rebuild. 15:24:03 yup 15:24:24 what's nice about the COPR I set up: I can rebuild all packages from current fedora sources with two buttons :) 15:24:50 Lots of compute power tho... 15:25:33 yeah I won't do it unless necessary ;-) 15:26:03 Needs an only-if-changed... guess you could stick it in a module... :D 15:26:15 uh ... what about no? 15:26:25 :D 15:26:29 FWIW, we could also check if using xmvn-javadoc instead of maven-javadoc-plugin fixes this ... I see -Xdoclint:none in the xmvn source code 15:26:47 Wouldn't that mean modifying 100s of packages again? 15:26:50 :/ 15:26:54 nope, just xmvn package 15:26:56 but xmvn FTBFS against java 11 with test failures :( 15:27:01 :D 15:27:12 We're waiting for module stream to fix that before we update? 15:27:22 * decathorpe giggles 15:27:49 oh, the switch is actually in javapackages-tools: https://src.fedoraproject.org/rpms/javapackages-tools/blob/master/f/javapackages-tools.spec#_12 15:28:40 which has the benefit that we don't need to fix xmvn to try it 15:29:22 That's... a lovely spec file. 15:30:11 if you focus your eyes slightly behind your monitor, it helps a lot 15:30:57 Trees :) 15:31:31 that's better 15:31:43 :D 15:31:57 alright, I'll try to find out if using xmvn-javadoc instead of maven-javadoc-plugin helps with doclint issues 15:31:59 Okie dokie, moving on? Or more Java11 stuff to discuss? 15:32:17 no more 15:33:16 #topic Review Open RHBZ Bugs 15:33:23 New bugs looks fine. There's CVEs aginst log4j and ant, I think we dep on log4j so we could take a look at rebasing to fix those CVEs. Ant I think we wait for module stream and cherry-pick 15:33:32 #link https://bugzilla.redhat.com/buglist.cgi?bug_status=__open__&email1=stewardship-sig%40lists.fedoraproject.org&emailassigned_to1=1&emailcc1=1&emailtype1=substring&list_id=11076493&product=Fedora&query_format=advanced 15:33:33 Then just a bunch of package updates. 15:33:58 cipherboy: https://src.fedoraproject.org/rpms/ant/pull-request/3 15:34:21 Ah :) 15:34:24 log4j I can't update yet, because it looks like it started to require java 11 though. 15:34:43 wt 15:34:46 stupid computer 15:34:49 one second 15:35:39 Sorry I'm back :) 15:35:51 So wait for Changes/Java11 to ship before updating log4j? 15:36:11 yeah 15:36:24 or, if that won't happen, backport the fix. 15:36:35 but I think it was just medium severity? 15:36:50 I think so, sounds about right. 15:38:15 #topic Review Open Pull Requests 15:38:26 #link https://fedora-stewardship.github.io/pr-report/ 15:38:48 so there's a few more updates, including some that I ported from module branches 15:39:33 nothing urgent though, maybe except the ant 1.10.8 update fixing the CVE 15:41:56 Cool, I'll take a look at the ant update after I reboot that laptop. 15:42:31 good. moving on? 15:42:38 Moving on. 15:42:49 #topic Review (SIG) Leaf Packages 15:42:55 #link https://fedora-stewardship.github.io/report/#total-leaves 15:43:02 looks like there's nothing to see here, for now. 15:43:27 there should be some, when we start moving things to the Java SIG though. 15:44:00 Right. So what's our plan with that? You created a new SIG, are we just going to move everything there? 15:44:11 Shutter the Stewardship SIG and join the Java SIG? 15:44:23 well, certainly not immediately 15:44:30 Agreed. 15:44:43 #topic The Plan ™ for the new Java SIG 15:45:00 #link https://pagure.io/stewardship-sig/issue/89 15:45:14 #link https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/DP4RKBS5X4UGX7CNG3CNJNTHHUF3P3DD/ 15:45:29 I responded to Ankur earlier today with what I'm planning to do 15:45:40 So we're blocked on fixing all those before moving to java-maint? 15:46:21 oh, no. I'd just like to have all the brokenness documented (any maybe fixed) when moving packages. 15:46:44 What's the status of the e*c*l*i*p*s*e bug -- last update 2 months ago, and the only thing left un-crossed is myln 15:46:46 .hello2 15:46:47 x3mboy: x3mboy 'Eduard Lucena' 15:47:07 cipherboy: no idea. I think mylyn was dropped somewhere 15:47:35 and actually, it was retired yesterday during the mass FTBFS cleanup 15:47:50 hello x3mboy :waves: 15:47:58 decathorpe, o/ 15:48:29 decathorpe: I'll ping in the bug. :) 15:48:46 thanks! 15:49:28 I guess Ankur got stuck in his work meeting, because he wanted to join us if he had time 15:49:59 decathorpe: yeah, that's what held up Dinesh. 15:50:08 Well, different meetings :) 15:50:08 cipherboy: do you have any suggestions for handling the transitions differently / better? 15:50:38 decathorpe: I kinda stopped following the thread due to a certain other person's involvement. Do you have a link to the proposed migration? 15:51:24 the link to the devel list I pasted above links to the message directly, not the thread :) 15:51:38 decathorpe: Ah sorry : ) 15:51:42 * cipherboy clicks 15:52:02 yeah the thread as a whole was kinda de-railed 15:52:42 Hmmm... rails were stolen, replaced in a different place ,and the entire train moved elsewhere :) 15:53:01 sounds fun 15:53:08 I've stopped folowing the thread long ago 15:53:27 decathorpe: Yeah, I think that sounds like a good plan. I'd kinda like to see Stewardship SIG "rebranded" as the Java Maint SIG rather than leaving both in-place. 15:53:36 JUST PACKAGE THE PRE-COMPILED BUILDS!!! 15:53:46 :D 15:54:10 decathorpe: I think we've gotten some good new blood, and I'd prefer we get them hooked while they're still enthusiastic. :) 15:54:17 cipherboy: I think that'll be the outcome. though I hope that I can step back from maintaining 200 java packages eventually :) 15:54:42 decathorpe: Right. My opinion is, while people are enthusiastic, move everything, then direct that enthusiasm :) 15:55:20 that's The Plan (TM( 15:55:30 .moar caffeine decathorpe 15:55:32 here decathorpe, have some more caffeine 15:56:41 alright I don't think I have any more to say today ... 15:56:58 that's great, because I need to gooo 15:57:18 mhroncok++ 15:57:24 I have a question 15:57:33 x3mboy: \o ask away :) 15:57:33 sure 15:57:39 If is ok with you guys for me to ask 15:57:53 I'm trying to launch/create a new Spin for i3wm 15:57:53 go ahead 15:58:11 We will have the first meeting in 1 hout 15:58:14 hour* 15:58:49 oh, nice. good luck! 15:59:03 The way GNOME is going on F32, I might have to switch... 15:59:09 #topic Open Floor 15:59:17 cipherboy: AGAIN? ;-) 15:59:20 And my question is, regarding being a SIG, Are there an specific requirement to became a SIG? 15:59:35 More than create a ML, a wiki page and a paguer repo 15:59:45 I have the 3 of those already 15:59:46 x3mboy: not really. it depends on how formalized you want everything to be 16:00:29 for example, do you expect to need a group that can be assigned bugs and / or added as maintainers to packages? 16:00:37 decathorpe: Anyhow, that's all I have I think. I'll try to be more involved next month esp. with transition & JDK11 16:00:47 cipherboy++ 16:00:47 decathorpe: Karma for cipherboy changed to 1 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 16:00:48 thanks! 16:01:35 decathorpe, well, I already have the support of the maintainer of the i3 packages 16:01:53 As our final product will be a Spin, I think a formal structure could be good 16:02:12 decathorpe: Thanks! 16:02:14 decathorpe++ 16:02:14 cipherboy: Karma for decathorpe changed to 8 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 16:02:37 x3mboy: I don't think you'll need more than a ML / pagure project / Wiki page for that 16:02:46 regular meetings help, though. 16:02:50 decathorpe, Ok, so thanks 16:03:02 decathorpe, Sure, I will set the meeting cadence today in our 1st meeting 16:03:36 yeah, good idea 16:03:41 #endmeeting