00:00 < knurd> | Meeting ping dgilmore, knurd, mmcgrath, nirik, stahnma, quaid and everyone interested in EPEL -- EPEL meeting in #fedora-meeting now! 00:00 --- | knurd has changed the topic to: EPEL Sig meeting -- Meeting rules at http://www.fedoraproject.org/wiki/Extras/Schedule/MeetingGuidelines -- Init process 00:00 < knurd> | Hi everybody; who's around? 00:00 * | Jeff_S here 00:00 * | nirik is here. 00:00 * | dgilmore is here 00:00 --> | notting (Bill Nottingham) has joined #fedora-meeting 00:00 * | mmcgrath is here 00:01 < knurd> | no quaid yet, but let's start 00:01 --- | knurd has changed the topic to: EPEL Meeting -- repotags -- all 00:01 < knurd> | so, the never ending story :) 00:01 < knurd> | feelings? 00:01 < knurd> | opinions on the last proposal? 00:02 < nirik> | did you see axel's email to the list about it? 00:02 < knurd> | opinions on Axels mail where he said "but adding repotags to epel will not help anymore in this - it's too late, repotags are dead and the pain has been inflicted, in fact for me the revival of repotags would imply the same transition pain again." 00:02 < knurd> | nirik, yeah, saw it 00:02 < knurd> | but I read it after I send the last proposal out 00:03 < nirik> | I really just want the issue to go away... I would be fine with pushing the proposal forward or just dropping it (again). 00:03 < knurd> | I'm wondering if we should try a different route: ask the FPC for a official statement if repotags are fine for them 00:03 < knurd> | (in general) 00:04 < knurd> | get that answer from their next week meeting 00:04 < knurd> | and we decide how to move on next week 00:04 < knurd> | I'm unsure 00:04 < knurd> | even I would likely vote "+/- 0" for my own proposal currently 00:04 * | quaid is aqui 00:05 < knurd> | other opinions? 00:05 < nirik> | ok, anyone else? opinions? votes? 00:05 < mmcgrath> | knurd: we really need to just decide and move on. 00:05 < spot> | we're not going to give that statement. 00:05 < spot> | just as an FYI 00:05 < nirik> | I lean toward just dropping the subject... moving on with our lives. 00:05 < knurd> | spot, ohh, I didn#t expect you would be around 00:05 < knurd> | spot, why? 00:05 < spot> | if you really want repotags vote on it, and we'll figure out how to implement them. 00:05 < Jeff_S> | I support your proposal, but with the idea that we can use that in the meantime and work on other (better?) ways of co-existance with other repos going forward 00:06 < dgilmore> | knurd: all i can say is drop it 00:06 < spot> | its not our domain to decide whether they're ok or not, thats FESCo 00:06 < dgilmore> | currently i would vote -100 00:06 < knurd> | spot, would you share your opnion on repotags? 00:07 < spot> | I don't know what technical problem they solve. 00:07 < dgilmore> | spot: none 00:07 < spot> | They clutter up the namespace. 00:07 < knurd> | dag tried to explain one two me, and it made a bit of sense 00:07 < nirik> | they allow end users to trivially see what repo packages are causing them conflicts/problems... or so the theory goes. 00:07 < knurd> | say there is clamav in both epel and dags repo 00:07 < dgilmore> | i dont know how any sane person can expect to mix repos providing the same packages and get a consistent and sane result 00:07 < knurd> | with different sub-packages 00:07 --> | smooge (Stephen J Smoogen) has joined #fedora-meeting 00:08 < spot> | i'd tend to agree that mixing repos with the same packagesets == BOOM 00:08 < spot> | and repotags won't resolve that. 00:08 --> | Karl_le_Rouge (Karl) has joined #fedora-meeting 00:08 < knurd> | then it can happen that a clamav subpacakge from epel tracks in the clamav main pacakge from dag 00:08 < Jeff_S> | knurd: yeah, repotags don't solve that issue 00:08 < knurd> | as long as there is no strict version-release dep 00:08 < spot> | all it will do is make the remains of your system more identifiable 00:08 < knurd> | and a repotag could make that strong version-release dep 00:09 < spot> | knurd: if the subpackages don't already have that strict version-release dep, it won't matter. 00:09 --> | danieldk (Daniel de Kok) has joined #fedora-meeting 00:09 < spot> | and if they do, the likely hood of the same e:n-v-r in both repos? 00:10 < spot> | I think you guys need to decide whether third-party repo simultaneous operation is a goal. 00:10 < spot> | If it is a goal of EPEL, then repotags are only part of a much bigger effort that needs to be undertaken. 00:10 < knurd> | third-party repo simultaneous operation is not a goal afics 00:10 < spot> | If it isn't, then repotags don't really have a point. 00:10 --> | Foolish (Sindre Pedersen Bjordal) has joined #fedora-meeting 00:11 < spot> | That's just my opinion. :) 00:11 < knurd> | thx spot 00:11 --> | stahnma (Michael Stahnke) has joined #fedora-meeting 00:11 < knurd> | I must say: I don't know how to move on 00:12 < mmcgrath> | knurd: After the in person meeting I thought we were going to do repo tags but if axel himself doesn't want them now I'm not sure what good we accomplish by turning them on. 00:12 < nirik> | ok, so can we move on now? or is there any more dead horse beating we need to do? 00:12 < stahnma> | sorry, IM late 00:12 < knurd> | okay, so lets somehow vote 00:13 < knurd> | A=no repotags; B=investigate repotags further; C=undecided 00:13 * | knurd votes C 00:13 --> | RemiFedora (Unknown) has joined #fedora-meeting 00:14 < mmcgrath> | B and C both imply that we're not moving on. I strongly feel the repo tags discussion does more harm by continuing to dicsuss it then either implementing it or disallowing it would. 00:14 < dgilmore> | A 00:14 * | quaid votes C with a clear statement, "Sorry, didn't mean to piss in anyone's breakfast cereal, we pledge to be more careful in the future." 00:14 < stahnma> | +1 quaid 00:14 <-- | rdieter_away has quit (Read error: 104 (Connection reset by peer)) 00:14 < knurd> | in case anyone is wondering: I'm interest in all opinions 00:14 --> | rdieter (Rex Dieter) has joined #fedora-meeting 00:15 < nirik> | mmcgrath: yeah, I want this to find a end... somehow. 00:15 < stahnma> | agreed 00:15 < quaid> | unless we can make "A=no repotags currently" or something, so we aren't having to reverse even more decisions 00:15 < knurd> | so people not in the steering committee can share their opinion by saying A, B or C, too 00:15 * | nirik votes A I guess. 00:15 < knurd> | quaid, "A=no repotags (as the case currently)" 00:16 < knurd> | quaid, so going for A would just end that topic as long as nobody brings it up again 00:16 < stahnma> | A 00:16 < quaid> | knurd: yeah 00:16 < quaid> | Cheers, Jeff. 00:16 < stahnma> | if that's the case, A 00:16 --- | rdieter is now known as rdieter_away 00:16 * | quaid pastes strangeness from his buffer, sorry 00:16 < f13> | oh! it's the weekly repotag conversation again. 00:16 * | Jeff_S votes D - go ahead with the proposal and start looking for alternatives as well 00:16 < quaid> | f13: don't kick a puppy when it's down! 00:16 < f13> | seriously, who do we have to shoot to get this topic dropped like we do every week? 00:16 < Jeff_S> | quaid: cheers ;) 00:17 < nirik> | f13: yes, it's a standing item it seems anymore. 00:17 * | mmcgrath notes 15 minute mark 00:17 < knurd> | okay; thl and quaid voted C; stahnma dgilmore and nirik A 00:17 < stahnma> | any other votings ? or move on? 00:17 < knurd> | mmcgrath ? 00:17 < quaid> | you have to ask/ 00:17 < smooge> | ok now that I have typed all the stuff in the wrong window 00:18 < Jeff_S> | everyone see #epel for smooge's ideas... 00:18 < smooge> | sorry 00:19 < f13> | I have no vote, but I"d vote A like always 00:19 < f13> | (and I'd vote that way in FESCo too FYI) 00:19 < smooge> | The main issues I deal on a volunteer support level are people who have followed a lot of Howto's and have multiple repo's or dont know they have multiple repos because their ISP was too nice 00:19 < nirik> | someone mentioned that this discussion did in fact happen with fedora.us/early fedora-extras days and was shot down there... just FYI... 00:20 * | knurd still waits for mmcgrath's vote 00:20 < smooge> | Yes.. there was a lot of discussion back then 00:20 < mmcgrath> | knurd: we're just waiting on me? 00:20 < mmcgrath> | why? 00:20 < knurd> | I didn#t see a A, B or C (or something else) 00:21 < knurd> | or did I miss it? 00:21 < mmcgrath> | A 00:21 < nirik> | smooge: I understand that this would give you more info, but I think that info can be gotten pretty easily other ways... 00:21 < knurd> | mmcgrath, k, thx 00:21 < smooge> | and it was put off of saying clearly that the group then was working with the older packaging community or not. Instead it was a sort of quiet no we are not while saying we will take it up later 00:21 <-- | RedKarl has quit (Connection timed out) 00:21 < knurd> | that makes four A (continue with no repotags) two C (undecided) and Jeff_S as potential steering committee member as "D" (proposal by thl) 00:22 --- | rdieter_away is now known as rdieter 00:22 < Jeff_S> | knurd: if you want my vote in terms of A/B/C, I suppose B would be the closest 00:22 < knurd> | Jeff_S, noted 00:22 < knurd> | so, that means no repotags 00:22 < nirik> | Jeff_S: splitter! :) 00:22 < f13> | for real this time? 00:22 < quaid> | smooge: as Axel said on list, maybe we need to handle this stuff at a higher level -- agree to work together with other repos, and work out shared scripts/tools that help in tech support. 00:22 < knurd> | or does anyone change his opinion after smooge's comments? 00:23 < nirik> | f13: well, until next meeting I bet. ;) 00:23 < mmcgrath> | quaid: I don't know how we can do that though, I think we need to work with axel and dag and co on much lower levels. The manitainers. 00:23 < knurd> | f13, I suppose there will be cries on the list 00:23 < mmcgrath> | its the packages that are incompatible. 00:23 < stahnma> | I have agreed with his comments, but am so tired of this, I just want to move on. Especially since Axel already said they are not worth it anymore. 00:24 < f13> | there is clearly a need to better identify where packages came from. repo tags aren't the solution IMHO 00:24 < smooge> | Ok, could there be a public reasoning statement about this to make it final. Basically, that EPEL is not meant to work with other repo's and that mixing/matching repo's is considered too much of a support burden for EPEL volunteers. 00:24 < stahnma> | hopefully a better techincal solution will be found soon and we can lead the way with that 00:24 --> | mdomsch (Matt_Domsch) has joined #fedora-meeting 00:24 < quaid> | smooge: +1 00:24 < mmcgrath> | smooge: don't confuse repo tags with collaborating with other repos. 00:24 < f13> | making it easier to pick out the gpg key would be best, hard to fake. %PACKAGER is a also usable, fakeable, but usable. 00:24 < knurd> | mmcgrath, +1 00:24 < smooge> | A web-page etc that says that will be a better line than the silent treatment that came out of fedora.us 00:24 < mmcgrath> | smooge: would you ask axel to do the same? He's not using repo tags. 00:25 < nirik> | perhaps we could get everyone to use yum-priorities. ;) But thats for another day... 00:26 * | knurd waits for 1 minute idle time before moving on to the next topic 00:26 < smooge> | mmcgrath, I could do so.. but his reasoning statement is that EPEL told him to get stuffed 00:26 < smooge> | one more second please knurd 00:26 < knurd> | smooge, that's why I said "one minute idle time" 00:26 < lancelan> | I am surprised that you have listened to axel ... 00:26 < mmcgrath> | If we told him to get stuffed, why have we been discussing it for 3 months? 00:27 < mmcgrath> | lancelan: he's the only one that bothered joining the SIG. That means a lot (to me at least) 00:27 < lancelan> | I thought that Dag gave good reason for repotags ... 00:27 < lancelan> | but hey - if you guys dont intend for folks to be able to mix repos then the discussion is pointless :) 00:28 < mmcgrath> | lancelan: I still didn't follow what he was talking about but then Axel goes on the list and says he doesn't want repo tags. 00:28 < smooge> | mmcgrath, the issue is that there is a big ego and communication gap going on... and this is a layer 8 problem that precludes any co-operation 00:28 < mmcgrath> | so I'm getting mixed signals from the other repos and it pisses me off because I thought we came to an agreement. 00:28 < lancelan> | well the main point was in mixed repo - even with stuff like priorities etc going on 00:28 < lancelan> | I th0ught anyway :) 00:29 < f13> | lancelan: Dag gave reasons why identifying where a package came from would be handy. repodtags are a hacky way of accomplising that goal (kind of), but not one that FEdora is willing to use. We'd rather find a more robust solution to the problem. 00:29 * | notting is confused. we have repo tags already. %{DISTRIBUTION} and %{PACKAGER} 00:29 < mmcgrath> | smooge: the problem is that in the dag and axel world, the guys that run the show are also THEE packagers. Thats not the case in EPEL. I firmly believe that to fix this problem the packages have to change. And the EPEL SIG just can't do that. 00:29 < f13> | notting: +1 00:29 < smooge> | mmcgrath, and we get mixed signals from the EPEL group and Fedora people.. some say that EPEL is doing this, others say that. 00:30 < lancelan> | well it seems to me that the fedora camp is firmly entrenched against repotags ... 00:30 < mmcgrath> | lancelan: but would you say "the fedora camp is firmly entrenched against 3rd party repos" ? 00:30 < lancelan> | and that something else _may- come along in the future 00:30 < smooge> | mmcgrath, so we end up with everyone taking the message that they want to hear or not to hear and just more and more conflicts 00:30 < knurd> | smooge, one of the downsides of a real community project with lots of volunteers afaics :-/ 00:31 < quaid> | minute has passed 00:31 < lancelan> | mmcgrath, I thought that was said before ... 00:31 < smooge> | knurd, I agree but at some point an organization has to take a stand and people have to fall to one side or the other.. otherwise nothing gets done 00:31 < mmcgrath> | smooge: but you're binding repo tags and collaboration and thats a fallacy. 00:32 < knurd> | smooge, your have a point, but well, it's a quite controversial topic.... 00:32 < mmcgrath> | repo tags will not fix the clamav problems, dag working with the clamav packager will. And that has nothing to do with the SIG. 00:32 < mmcgrath> | or the clamav packager working with dag will. 00:33 <-- | rwmjones has quit ("Closed connection") 00:33 * | knurd will move on soon if nothing happens anymore 00:33 < smooge> | mmcgrath, since I have joined it has been mentioned on both sides as being part of collaboration at some point or another. 00:33 < lancelan> | I actually dont like repotags - and agree that yum/rpm should be able to indicate repo - but they dont ... 00:33 < smooge> | I would like to see a definition in the end of what is collaboration and what can be done if anything to meet it 00:33 < mmcgrath> | smooge: but collaboration does not mean "do what the other guy says" 00:34 < lancelan> | unless of course you specifically ask 00:34 < f13> | lancelan: they can, but it requires more query flags. Now whether or not those query flags are made more prominent, or... 00:34 < mmcgrath> | knurd: what else is on the schedule today? 00:34 < f13> | lancelan: (: 00:34 < lancelan> | f13, sure, some user tools to show stuff would be good - and they are coming in to yum ... 00:35 <-- | londo has quit (Read error: 110 (Connection timed out)) 00:35 < knurd> | does anybody really think opinions will cahnge if we discuss this further? 00:35 < wwoods> | rpm -qi isn't that hard 00:35 < lancelan> | I would actually like to see mixed repos - but yum isnt clever enough yet IMHO 00:35 < f13> | smooge: working with 3rd party repos is going to be pretty hard in any case. THe EPEL project cannot reference nor link to these 3rd party repos for software, as those repos host illegal content. SO we have to package everything we want in EPEL proper, which immediatly creates 'conflicts' with these 3rd party repos. 00:35 < knurd> | otherwise I suggest we move on 00:35 < knurd> | f13, +1 00:35 < f13> | smooge: if these repos would just drop the illegal content then it would be much easier. 00:35 < lancelan> | f13 - define illegal for me please :) 00:35 < smooge> | f13, but it might not be illegal where they are 00:36 < f13> | smooge: and that doesn't matter to us 00:36 < Jeff_S> | f13: kbs extras has something RH would consider "illegal"? 00:36 < dgilmore> | lancelan: illegal in the US mp3 dvd etc 00:36 < f13> | Jeff_S: what is 'kbs' ? 00:36 < dgilmore> | Jeff_S: he only built Fedora Extras for RHEL 00:36 < Jeff_S> | sorry, centos.karan.org 00:36 < f13> | ah. 00:36 < Jeff_S> | dgilmore: yes, exactly 00:36 < f13> | in that case, it would be obsolete 00:37 < f13> | as we want to use the same infrastructure framwork and people to do the builds. 00:37 < Jeff_S> | I'm just saying that you're making quite a blanket statement 00:37 < dgilmore> | Jeff_S: AFAIKT he has not updated any packages since we dropped FC-3 support 00:37 < f13> | Fedora Packager Joe can't get to the source control for kbs, to fix things, or whatever. 00:37 < smooge> | Jeff_S, its ok.. this I have learned is f13's normal mode of talking in meetings. 00:37 * | Jeff_S ignores it then 00:38 < notting> | actually, speaking of centos, do we have whomever's doing the /extras/ stuff on the main centos mirrors working in epel? 00:38 < f13> | Jeff_S: the people that are complaining loudly about 'interoperability' with 3rd parties are the ones that are carrying things we can't reference. 00:39 < nirik> | notting: not the Xfce people at least.... They are using my Xfce specs tho, so if I push Xfce for EPEL it should be fine as long as we keep the centos-extras in step. 00:39 < knurd> | notting, I don't think there are much people that are working in Fedora and Centos Extras 00:39 < f13> | Jeff_S: so yes, I'm blanket stating that working with "3rd party" is going to be difficult on many levels. As we found with FEdora itself, just having an official 3rd party was a lot of unnecessary trouble. 00:39 < Jeff_S> | notting: it is a small group of people AFAIK, and I think the answer is no to most or all of them 00:39 < smooge> | notting, no. I understand that there are contractual issues with them signing the Fedora License stuff 00:39 < notting> | *sigh* 00:39 < mmcgrath> | notting: we tried though. 00:40 < Jeff_S> | f13: well if it is necessary or not is debatable IMO 00:40 < nirik> | smooge: if you know who the people(s) that are doing the Xfce for centos-extras are, I would love a contact email... I haven't been able to figure out who it is. 00:40 < Jeff_S> | f13: but, yes, it is trouble, that much is clear 00:40 < stahnma> | I think I have to get to a work meeting 00:40 < stahnma> | so, I am parting early 00:41 < Jeff_S> | bye stahnma 00:41 < knurd> | shall we move on with the other topics of the EPEL meeting? 00:41 < knurd> | by stahnma 00:41 < Jeff_S> | knurd: I would like that 00:41 < smooge> | knurd, I have spoken my piece on this 00:41 < knurd> | smooge, thx for your input 00:41 --> | donavan (CtrlProxy User) has joined #fedora-meeting 00:41 < knurd> | so let's move on 00:41 --- | knurd has changed the topic to: EPEL Meeting -- EPEL mock configs in Fedora's mock package 00:42 < knurd> | seems it has happened 00:42 < knurd> | does anyone know if the EPEL4 configs have been updated? 00:42 * | mmcgrath hasn't done it 00:42 < mmcgrath> | dgilmore: ? 00:42 < Jeff_S> | I haven't gotten around to figuring out git yet... 00:42 < knurd> | s/mirror.centos.org/mirrorlist.centos.org/ 00:42 < quaid> | Dennis fell asleep 00:42 < Jeff_S> | knurd: I think that has been fixed, but I'm not positive 00:43 < knurd> | Jeff_S, okay, can you keep an eye on it if it really has been fixed when the new package comes out? 00:43 < f13> | we have a mock update coming out soon 00:43 < knurd> | that would be enought for now afaics 00:43 < f13> | I _think_ it has epel configs 00:43 < f13> | you can check the git tree though 00:43 < f13> | git://git.fedoraproject.org/git/hosted/mock 00:43 < Jeff_S> | knurd: yes, I'll check to see if he's sent it through koji yet 00:43 < knurd> | Jeff_S, thx 00:43 < f13> | well, there was some bugs found, so we're fixing those 00:43 < Jeff_S> | f13: can that be accessed via http somehow? 00:43 --- | knurd has changed the topic to: EPEL Meeting -- comps.xml for EPEL -- dglimore 00:43 < f13> | yes 00:44 < Jeff_S> | f13: yeah, Michael mailed the epel list earlier today 00:44 < f13> | git clone http://git.fedoraproject.org/git/hosted/mock 00:44 < f13> | or jsut browse http://git.fedoraproject.org 00:44 < Jeff_S> | f13: yes, the 2nd option is what I meant, thanks 00:44 < knurd> | dgilmore, are you still around? 00:44 < dgilmore> | knurd: yeah sorry 00:45 < knurd> | dgilmore, np 00:45 < knurd> | any progress on teh compx.xml stuff? 00:45 < dgilmore> | knurd: there was a post that epel configs will be in mock 0.7.1 00:45 < smooge> | ok on mock configs.. will EPEL just be focusing on EL4 and EL5? Will it be doing anyting for EL-BRIC? 00:45 < knurd> | smooge, BRIC? 00:45 < dgilmore> | knurd: i started i need to clean up some of the packages not in EPEL 00:46 < Jeff_S> | knurd: the updated EL4 configs look good in the git repo, so they should get pushed w/ the newest mock package 00:46 < smooge> | RHEL-BRIC a 2 year desktop offering that RH is supposed to be doing 00:46 < dgilmore> | smooge: we are only looking at EL4- and 5 00:46 < dgilmore> | we have no current plans to do 2.1 or 3 00:46 < knurd> | smooge, ohh, that one, sorry, i didn#t know that acronym yet 00:46 < f13> | smooge: that's not RHEL 00:46 < dgilmore> | but if there is demand and people to do work there is no reason we cant 00:46 < f13> | rather, that's Red Hat Global Desktop 00:46 < f13> | and no, I don't think RHGD is a good target for EPEL 00:47 < f13> | (yet) 00:47 < smooge> | f13, sorry.. the RH marketing literature put them together 00:47 < f13> | well, RHGD is based on RHEL, but we're not calling it RHEL (: 00:47 < smooge> | so I was confused 00:47 < f13> | tell me about it 00:47 < f13> | *sigh* 00:47 < knurd> | dgilmore, shall we talk about the comps.xml stuff fiurther next week? 00:47 < quaid> | smooge: if you have a pointer to that marketing lit, let me know 00:48 < knurd> | dgilmore, or do you need help with it? 00:48 < smooge> | quaid, sorry it was in the Summit stuff I was reading. They were mentioning about the offerings and I must have mixed/matched in my head like a bad techie 00:48 < quaid> | smooge: no worries, you weren't the only person 00:48 < smooge> | on the mock/build infrastructure, I have a question focused for 3rd party groups 00:48 < smooge> | is this a good time for it or another section? 00:49 < knurd> | please wait until the free discussion section 00:49 --- | knurd has changed the topic to: EPEL Meeting -- bodhi/testing repo/final repo layout/ koji for epel 00:49 < knurd> | did anybody ( dgilmore ? ) talk to mbonnert? 00:49 < dgilmore> | knurd: well part of it is making sure the groups are sane and that we have the packages 00:50 < dgilmore> | Ill check something into cvs soon and then post to the list asking people to look over it 00:50 < knurd> | dgilmore, thx 00:50 < dgilmore> | knurd: i spoke to him 00:50 < dgilmore> | no timeframe yet 00:50 * | dgilmore does not have the cycles to work on it 00:50 < knurd> | k, I suppose we have to live with that for now 00:50 --- | knurd has changed the topic to: EPEL Meeting -- finish the wiki docs and remove the warnings by end of may 00:51 < knurd> | quaid, ? 00:51 < knurd> | lala 00:51 < knurd> | I sometimes think we need to make the status updates mail based 00:51 < quaid> | *sing* 00:51 < knurd> | they take to much time in the meetings 00:51 < quaid> | eek, that was end of May? 00:52 < quaid> | sorry for the delay 00:52 < knurd> | quaid, that once was the plan :-) 00:52 < quaid> | I actually have it up in my to-do list as a high priority, so I should be able to by end of week 00:52 < knurd> | quaid, k, thx 00:52 < quaid> | yeah, well, plan and then the wind blows 00:52 --- | knurd has changed the topic to: EPEL Meeting -- Free discussion around EPEL 00:52 < knurd> | quaid, happens :-) 00:52 < quaid> | question on that is -- we are generally ready to be all announced? 00:52 < knurd> | smooge, shoot 00:53 < nirik> | we need bodhi I think before that... which needs koji. 00:53 < smooge> | Ok is the infrastructure done in a way that can be duplicated by 3rd parties. When I worked for US Government.. we could not directly pull stuff from any non-RH RHN provided repo without rebuilding the packages ourselves 00:54 < smooge> | So basically Ihad a small mock system that rebuilt various extras items that were needed on the RHEL system 00:54 < knurd> | smooge, it should be possible to duplicate it 00:54 < knurd> | smooge, we just use open souce tools for it 00:54 < dgilmore> | smooge: right now its all done in plague 00:54 < nirik> | yeah, everything should be available... 00:54 < Jeff_S> | smooge: yeah, it's all just mock under the hood anyway 00:54 < knurd> | plague (soon koji) with mock and stuff like that 00:55 < quaid> | there however is no guide/how-to yet, right? 00:55 < smooge> | the issues are the build parameters and such that can be fed underneath the cooker 00:55 < mmcgrath> | smooge: I'll put it this way, anyone that can't duplicate it should come talk to us so we can correct that :) 00:55 < dgilmore> | smooge: one of our goals is the same as fedora. use only open source tools 00:55 < knurd> | smooge, it's all in the spec files afaik 00:55 < knurd> | smooge, we don#t use magic "--with foo" stuff 00:55 < dgilmore> | smooge: standard mock settings 00:55 < smooge> | Ok from experience with the RHEL rebuild there were various things that needed to be fed to get packages to 'match' via libraries and such 00:56 < smooge> | I wanted to make sure that if these were done.. it was documented in 'meta-spec' files or some such 00:56 < f13> | smooge: well you're going to run into the same things that centos and oracle run into 00:56 < f13> | the public RHEL binaries aren't the result of every package built against the GA package 00:56 < knurd> | smooge, it should be possible without to much hassle (you know, there are always some bugs on the way....) 00:57 < f13> | so the GA packages were built against packages that may have been older than what is in GA 00:57 < f13> | to extend this... 00:57 < smooge> | f13, I know with the RHEL that is the case.. but RH-EPEL I wanted to know if that was the case or if I needed to tell my Gov replacement he was cool with what was there 00:57 < f13> | if EPEL builds using RHEL binaries, the results may actually be different than EPEL packages built against CentOS binaries. 00:58 < mmcgrath> | smooge: I'm actually writing a RHM article right now about how people can duplicate our environment. 00:58 < f13> | and EPEL builds built against "GA" epel builds may be different than the GA epel builds. 00:59 < f13> | it's just due to how build systems work, well sane ones (although some people will differ). Building a package doesn't automatically make every package that is associated with it rebuild as well, ad infinium. 00:59 < f13> | you're always going to have cases where the "released" package wasn't built with the "released" tree 00:59 < smooge> | Ok thanks.. the guys at LANL will be happy about that.. they need to build stuff against 1.0,2.1,3,4,5 (1.0 being RHL6.2) 00:59 < knurd> | k, anything else? 01:00 * | knurd will close the meeting in 30 01:01 * | knurd will close the meeting in 10 01:01 < knurd> | -- MARK -- Meeting end 01:01 --- | knurd has changed the topic to: http://fedoraproject.org/wiki/Communicate/FedoraMeetingChannel -- Meetings often get logged -- see schedule in the wiki for next meeting 01:01 < knurd> | thx everyone 01:01 <-- | RemiFedora has quit ("A bientôt...")
/me again
On 13.06.2007 20:04, Thorsten Leemhuis wrote:
00:00 < knurd> | Meeting ping dgilmore, knurd, mmcgrath, nirik, stahnma, quaid and everyone interested in EPEL -- EPEL meeting in #fedora-meeting now! 00:00 --- | knurd has changed the topic to: EPEL Sig meeting -- Meeting rules at http://www.fedoraproject.org/wiki/Extras/Schedule/MeetingGuidelines -- Init process
Just want to highlight this: the decision once again (for the third time (¹) IIRC) was to not do repotags in EPEL. See the log for all the details; it's the first topic in the meeting and was discussed in the first 40 minutes.
CU thl
(¹) -- and I suppose the EPEL Steering Committee doesn't want to touch the topic again; if you think the EPEL Steering Committee did a bad decisions you are of course free to go to higher Committee's like FESCo or the Fedora Board. But I suppose we all have to find some way to live with this decision now.
On 13-jun-2007, at 20:04, Thorsten Leemhuis wrote:
00:42 < knurd> | does anyone know if the EPEL4 configs have been updated? 00:42 * | mmcgrath hasn't done it 00:42 < mmcgrath> | dgilmore: ? 00:42 < Jeff_S> | I haven't gotten around to figuring out git yet... 00:42 < knurd> | s/mirror.centos.org/mirrorlist.centos.org/ 00:42 < quaid> | Dennis fell asleep 00:42 < Jeff_S> | knurd: I think that has been fixed, but I'm not positive 00:43 < knurd> | Jeff_S, okay, can you keep an eye on it if it really has been fixed when the new package comes out?
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=239981
Ruben
On Wed, 13 Jun 2007, Thorsten Leemhuis wrote:
00:24 < f13> | there is clearly a need to better identify where packages came from. repo tags aren't the solution IMHO
True, but any solution proposed will not hit RHEL until RHEL6.
This is not Fedora where we can fix something for the next release a few months away. This Fedora mentality is really what is wrong with EPEL in my opinion. How many of you people are still using RHEL2.1 or RHEL3 ?
RHEL5 is the RHEL3 of 2010. No customer is interested in RHEL5 now if it will be ignored like RHEL3 is today.
00:24 < smooge> | Ok, could there be a public reasoning statement about this to make it final. Basically, that EPEL is not meant to work with other repo's and that mixing/matching repo's is considered too much of a support burden for EPEL volunteers.
Please do. Because that's what we have seen being discussed in the EPEL meetings, but on the list the same viewpoint has not been defended. In fact it has been rebutted many times.
00:29 < f13> | lancelan: Dag gave reasons why identifying where a package came from would be handy. repodtags are a hacky way of accomplising that goal (kind of), but not one that FEdora is willing to use. We'd rather find a more robust solution to the problem.
So we won't have a solution for RHEL2.1, RHEL3, RHEL4 or RHEL5. Because looking at new development is excluding the currently supported distributions. I'm willing to look at any alternative if we have one, but I'm more interested to implement something that we can use today. And repotags have been that solution.
f13 was not in the discussion at LinuxTag and apparently did not read some of my mails concerning repotags:
http://www.redhat.com/archives/epel-devel-list/2007-May/msg00134.html http://www.redhat.com/archives/epel-devel-list/2007-May/msg00135.html
repotags make the package-name unique, which is important for a depsolver to understand what packages belong together.
00:29 * | notting is confused. we have repo tags already. %{DISTRIBUTION} and %{PACKAGER}
Those do not make a package unique. Welcome to the discussion. A depsolver does not use the information and even if it could be made to do that, we won't have it in RHEL2.1, RHEL3, RHEL4 or RHEL5.
Please move any design work to RHEL6 in 2009. We are talking about the past and the presence.
00:29 < mmcgrath> | smooge: the problem is that in the dag and axel world, the guys that run the show are also THEE packagers. Thats not the case in EPEL. I firmly believe that to fix this problem the packages have to change. And the EPEL SIG just can't do that.
No, packages do not have to change. In fact, there is no repotag in the authoritative SPEC file, they are added by the buildsystem. We _have_ discussed that and Thorsten said it was a possibility, welcome to the discussion.
00:30 < lancelan> | well it seems to me that the fedora camp is firmly entrenched against repotags ... 00:30 < mmcgrath> | lancelan: but would you say "the fedora camp is firmly entrenched against 3rd party repos" ?
/me notes that EPEL is a 3rd party repository itself.
00:31 < mmcgrath> | smooge: but you're binding repo tags and collaboration and thats a fallacy.
repotags are required for RHEL2.1 upto RHEL5 if you want to be able to mix repositories. If EPEL is against mixing repositories, please say so clearly.
And then we move back to the fact that different RHEL/CentOS users want different things, like maturity or latest. Which cannot possibly live in the same repository (unless we fix Yum to understand it) and hence we move into a subject that EPEL is desperately ignoring.
You'll have to mix repositories at some point because not all users want/need maturity. (whatever that is defined by)
00:32 < mmcgrath> | repo tags will not fix the clamav problems, dag working with the clamav packager will. And that has nothing to do with the SIG.
repotags would at least allow to upgrade from one version to another. Without repotags it wouldn't even be possible (as there may not be a distinction from RPM's point of view). I've explained this in our meeting and at least some people understood that repotags could help.
It won't fix all problems, but I doubt there is one solution that fixes all problems.
00:33 < smooge> | mmcgrath, since I have joined it has been mentioned on both sides as being part of collaboration at some point or another. 00:33 < lancelan> | I actually dont like repotags - and agree that yum/rpm should be able to indicate repo - but they dont ... 00:33 < smooge> | I would like to see a definition in the end of what is collaboration and what can be done if anything to meet it 00:33 < mmcgrath> | smooge: but collaboration does not mean "do what the other guy says"
Please come up with a workable proposal then ? Don't just refuse the idea without even understanding what I have been explaining.
And please understand that any workable proposal cannot touch RPM or Yum since we cannot replace those in RHEL5 and older.
00:34 < lancelan> | unless of course you specifically ask 00:34 < f13> | lancelan: they can, but it requires more query flags. Now whether or not those query flags are made more prominent, or...
Please, do not talk about future development. We've been there, no solution for the past and present.
00:35 < knurd> | does anybody really think opinions will cahnge if we discuss this further? 00:35 < wwoods> | rpm -qi isn't that hard
Oh, it makes a difference between replying with a solution straight away and having to ask for additional information. On IRC that often means, a solution or no solution.
Besides, as soon as we have clamav packages with exactly the same EVNR, be sure that rpm -qi may not even give proper information, because clamav-0.90.2-3 and clamav-devel-0.90.2-3 could be from 2 DIFFERENT repositories. Querying one or the other may confuse you even more.
00:35 < f13> | smooge: working with 3rd party repos is going to be pretty hard in any case. THe EPEL project cannot reference nor link to these 3rd party repos for software, as those repos host illegal content. SO we have to package everything we want in EPEL proper, which immediatly creates 'conflicts' with these 3rd party repos.
So EPEL is a 3rd party repository that does not work well with other 3rd party repositories ?
00:35 < f13> | smooge: if these repos would just drop the illegal content then it would be much easier.
What is illegal to you, is not illegal to me. I do not live in the US and I do not want to be bound to US law. Other countries may even be stricter than US law.
It's not only problematic that US is lobbying for stricter laws all around the world, but using this forum and repositories to force US law on the rest of the world I find distasteful.
00:38 < f13> | Jeff_S: the people that are complaining loudly about 'interoperability' with 3rd parties are the ones that are carrying things we can't reference.
So, we finally found an excuse for no colaboration. At least that means I wil have more time for not trying.
Kind regards, -- dag wieers, dag@wieers.com, http://dag.wieers.com/ -- [all I want is a warm bed and a kind word and unlimited power]
Dag Wieers wrote:
So, we finally found an excuse for no colaboration. At least that means I wil have more time for not trying.
Just curious, do atrpms, RPMForge, Centos Extras and KBS conflict? If not how do you guys keep them from conflicting?
Is there a neutral mailing list we can create somewhere?
-Mike
On Sat, 16 Jun 2007, Mike McGrath wrote:
Dag Wieers wrote:
So, we finally found an excuse for no colaboration. At least that means I wil have more time for not trying.
Just curious, do atrpms, RPMForge, Centos Extras and KBS conflict? If not how do you guys keep them from conflicting?
Is there a neutral mailing list we can create somewhere?
People use them together and report any problems. But I think what helps the most is that we carefully look what others have implemented before we duplicate any effort.
CentOS does not introduce anything that is not compatible with RPMforge. Karan excludes everything from Extras that causes incompatibilities.
I guess that's because people care about allowing users to enable their repository and incompatibility with someone else's repository is one way of driving users away.
What's more, I often look at Fedora packages to know what I have to Obsolete/Provide to make those package somewhat compatible. But the effort should not be completely wasted on us IMO.
Kind regards, -- dag wieers, dag@wieers.com, http://dag.wieers.com/ -- [all I want is a warm bed and a kind word and unlimited power]
On Sat, 2007-06-16 at 17:50 +0200, Dag Wieers wrote:
I guess that's because people care about allowing users to enable their repository and incompatibility with someone else's repository is one way of driving users away.
So true ...
I see two potentially shining lights we can look to help illuminate our way, with the goal of helping our users:
1. Formalizing the third-party repo agreement (thanks Tim); and, yes, I think we all agree that EPEL is another third-party repo.
2. From that, figure out a way to send people to other repos for packages in a (somewhat) automated, culturally acceptable, process-integrated, and legal-for-those-employed-by-US-corporations.
Fedora already is clear that, for example, we can refer people to fedoraforum.org as "a place to get answers" but *not* with a direct URL to how to install an illegal-in-the-US[1] package.
So, similarly, EPEL could:
* Make it easier to find compatible, other-repo packages
* Agree not to package something that is already in another repo, instead referring to that repo and package in particular (cf. fedoraforum.org and pointing at not-illegal solutions there)
* Similarly, where a package is *not* in EPEL because it is part of a RHEL layered product, we could still point people to where those packages are in other repos. It may not be considered "kind" to Red Hat, but that would be under the same bogus argument that people should pretend CentOS doesn't exist. The situation is going to arise where EPEL doesn't have a package for this reason, and we cannot just shrug our shoulders and point at a RHEL sale page.
* Work with all repos on improvements to rpmfind.net, to make it easier for users to: - Find compatible packages - Identify when different repos are likely to conflict - Figure out the origin of a package they have installed This would allow us to use a common troubleshooting methodology for all of our users. I understand what Dag is saying about removing steps (i.e., having a repotag answers a common question without asking it), and my idea here is the opposite -- add a common step for all users, but one that answers multiple questions while troubleshooting.
* Help define and advertise a common "meaning" for each repo -- "This one is steady, older packages; this one provides the latest possible packages; etc." Make it more obvious to users why they would choose one repo over another for realistic reasons.
- Karsten
[1] I'll encourage people to use nomenclature like this, if it helps, but I think we can agree that when Red Hat associates say "illegal" we are referring to what Smooge said -- trying to keep from getting our employer in hot water over US laws.
Karsten Wade wrote:
So, similarly, EPEL could:
...
- Agree not to package something that is already in another repo,
instead referring to that repo and package in particular (cf. fedoraforum.org and pointing at not-illegal solutions there)
Currently not possible/practical in many cases, since by policy, epel packages current can't (Build)Requires anything outside of epel.
-- Rex
On Mon, 2007-06-18 at 13:21 -0500, Rex Dieter wrote:
Karsten Wade wrote:
So, similarly, EPEL could:
...
- Agree not to package something that is already in another repo,
instead referring to that repo and package in particular (cf. fedoraforum.org and pointing at not-illegal solutions there)
Currently not possible/practical in many cases, since by policy, epel packages current can't (Build)Requires anything outside of epel.
Aren't there a sub-set of packages that are either:
* Likely to have multiple versions in use (stable => edge), or * Stand-alone
?
- Karsten
Karsten Wade wrote:
- Agree not to package something that is already in another repo,
instead referring to that repo and package in particular (cf. fedoraforum.org and pointing at not-illegal solutions there)
Any package that is already in a third party repository? That would essentially force users to mix and match many repositories which I don't think we should be doing. If it is just for packages we don't for legal reasons then that's doable.
Rahul
On 06/16/2007 06:36 PM, Mike McGrath wrote:
Dag Wieers wrote:
So, we finally found an excuse for no colaboration. At least that means I wil have more time for not trying.
Just curious, do atrpms, RPMForge, Centos Extras and KBS conflict?
For what it's worth, here is the situation of an admin who uses those 4 repos for a couple of years on around 200 of RH 7.x, C3, C4 machines (plus a dozen fedora>=FC1 (yes, I do have functional RH 7.2 and FC1 in production)). All of those systems are in an enterprise environment (as opposed to home/playground use), but most of the desktops have mp3/mplayer/pdf viewer/other goodies which make user life easier installed - never ever had any major issues with packages from rpmforge, centos extras and kbs. they have always played excellent together. As far as I remember I had maybe 2 or 3 conflicts during the last 5 years. Maybe the fact that I do not blindly install packages helps, but I am more then satisfied with what dag/thias/dries/karan have supplied. I will never be able to express my gratitude for them but I hope that somehow somewhere they know about it. - I've also used atrpms a couple of times and it also played nice. Centos recommends being cautious when using it and since it ships newer versions of Base stuff I think it is normal to be cautious in this case. But I am grateful that Axel saved my time suppling openswan and acl enabled kernels, newer versions of lm-sensors (for chipsets not supported by the standard kernel stack) or wifi support for my laptop which runs Centos 3 (and cannot run anything newer due to the EDA tools which are not supported on a newer OS)
On Sat, Jun 16, 2007 at 09:00:02PM +0300, Manuel Wolfshant wrote:
- I've also used atrpms a couple of times and it also played nice.
Centos recommends being cautious when using it and since it ships newer versions of Base stuff I think it is normal to be cautious in this case. But I am grateful that Axel saved my time suppling openswan and acl enabled kernels, newer versions of lm-sensors (for chipsets not supported by the standard kernel stack) or wifi support for my laptop which runs Centos 3 (and cannot run anything newer due to the EDA tools which are not supported on a newer OS)
Thanks Manuel! ATrpms gets hit that often and that hard for offering replacement packages that a kind word in retaliation justifies the effort spent on doing so. ;)
On Sat, Jun 16, 2007 at 10:36:17AM -0500, Mike McGrath wrote:
Dag Wieers wrote:
So, we finally found an excuse for no colaboration. At least that means I wil have more time for not trying.
Just curious, do atrpms, RPMForge, Centos Extras and KBS conflict? If not how do you guys keep them from conflicting?
I usually check what exists (it's easy when you have a local mirror of everything) and divert any request to the matching repo.
It's not something that has been proceduralized, but has grown as de facto behaviour over the course of the last years. Tim's recent manifesto intended for the intrarelation of all 3rd party repos posted on this list and transfered to the wiki would had been the first formalization proposal of how 3rd party repos work together. I really liked it and will pick it up in some way or another.
On 6/16/07, Dag Wieers dag@wieers.com wrote:
00:35 < f13> | smooge: if these repos would just drop the illegal content then it would be much easier.
What is illegal to you, is not illegal to me. I do not live in the US and I do not want to be bound to US law. Other countries may even be stricter than US law.
It's not only problematic that US is lobbying for stricter laws all around the world, but using this forum and repositories to force US law on the rest of the world I find distasteful.
Jesse just needs to take some classes in talking from his mouth versus other parts of his body.
People employed directly or in-directly by a US company have to remember that their company is subject to US laws. Any work that they do that could be seen as Red Hat promoting the breaking of those laws can be enforced against the company. And unless you are a lawyer.. its hard to know when you are using your first amendment right as an individual or as a paid employee of XYZ corp. And yes I think it sucks, the laws suck, and the US acting like an Empire when ti cant pay its light bills sucks.
epel-devel@lists.fedoraproject.org