| --- bpepple has changed the topic to: FESCo meeting -- Meeting rules at http://www.fedoraproject.org/wiki/Development/Schedule/MeetingGuidelines -- Init process | ||
| bpepple | FESCo meeting ping -- bpepple, caillon, c4chris, dgilmore, dwmw2, f13, jeremy, jwb, notting, spot, nirik, tibbs, warren | |
|---|---|---|
| ReleaseParty/F9/Ohio:Central Ohio Release Party | ||
| * nirik is here. | ||
| tibbs here | ||
| bpepple | Hi everybody; who's around? | |
| * dgilmore is here | ||
| bpepple | hmm, not many people here today. | |
| * warren here | ||
| c4chris here now :) | ||
| * caillon prefers to show presence by his being here | ||
| bpepple | The only thing I had on the agenda was a post-mortem on F9 to see what we can work on improving for F10. | |
| * dwmw2 here | ||
| --- bpepple has changed the topic to: FESCo-Meeting -- F9 - Post-mortem? - all | ||
| caillon | http://dbaron.org/log/20080513-distros | |
| * f13 | ||
| f13 | There are multiple "Lessons Learned" pages sprouting up on the wiki | |
| dwmw2 | I like that we didn't change rawhide over to F10 until F9 was actually released -- although there seemed to be a threat of changing over earlier. | |
| caillon | has a good summary i think of something we should attempt on improving for f10 | |
| f13 | some from Infrastructure, some from QA | |
| caillon | though i'm not sure how to get there yet | |
| f13 | dwmw2: the plan was to switch it as soon as F9 had gone GOLD and we weren't taking any more testing/changes of the F9 package set. | |
| dwmw2: and we had a set of updates pushed to point people at for early adoption | ||
| dwmw2: alas the updates push too a bit longer than expected and it all came about on the same day, cutting into F10 development time pretty badly | ||
| (see giant pile of rawhide updates the first day) | ||
| http://fedoraproject.org/wiki/Infrastructure/F9LessonsLearned is the Infrastructure set | ||
| wwoods | that dbaron post is all known bugs and vague bullshit | |
| caillon | yeah. i really wish we had a big enough testing community such that we could have both F9 and F10 be available simultaneously | |
| dwmw2 | f13: would people still be able to install the about-to-be-released tree, between the date we switch and the actual release? | |
| tibbs | Not having F10 stalled me out for a while. | |
| caillon | wwoods, known bad bugs that we ship with. fucked up sound last release, this release with known X/input stuff broken... | |
| f13 | caillon: honestly, I've bene using the testing community as a scape goat for the lack of hardware to make it happen. | |
| wwoods | we are always going to ship with known bugs. | |
| forevever. | ||
| f13 | dwmw2: ideally yes, just not sure how to make that happen. | |
| tibbs | Plus I'm still trying to catch my morrors up to the triple threat of F9, rawhide and 0-day updates. | |
| wwoods | with an extra 'ev'.. for extra-forever | |
| dwmw2 | you also don't want the mirrors to delete all those f9 packages, just to have to copy them again when they reappear in a different place | |
| caillon | wwoods, yes. but we are perceived as having more BAD bugs that prevent people from using the distro. | |
| jwb | bpepple, i'm here now | |
| f13 | dwmw2: not without confusing the hell out of everybody by having live pakages in the releases/9 directory. | |
| caillon | BAD bugs like not being able to install on samsung hard disks | |
| wwoods | what does that even mean. | |
| caillon | BAD bugs like input not working properly for everyone | |
| f13 | dwmw2: by that point, they would have already had the 9 pacakges in the release dir | |
| wwoods | and how do we solve this when we're already well beyond our capacity to test what we have? | |
| caillon | wwoods, like i said, i don't know how to fix it. | |
| dwmw2 | f13: the official mirrors would. Other people wouldn't though. Like the mirror I keep locally. | |
| wwoods | awesome. | |
| "I don't know how to fix it, but you guys should fix more bugs before releases" | ||
| I totally agree. | ||
| notting | caillon: samsung we just needed to catch earlier. not sure how | |
| f13 | dwmw2: why don't you become an official private mirror and thus get acl access to the tier1s or tier2s? | |
| caillon | wwoods, well... | |
| dwmw2 | is that easy for the Great Unwashed to do? | |
| f13 | dwmw2: there is a wiki page | |
| jds2001 | dwmw2: yes, im an official private mirror :) | |
| caillon | wwoods, some of it could possibly be fixed by having a longer freeze period. | |
| caillon | wwoods, or at least helped. | |
| whether that means freezing at alpha, or extending the release cycle, i don't know. | ||
| warren | a lot of the things we fixed especially in the installer was because nobody tested it until the last minute | |
| dwmw2 | that sounds like a viable solution to the problem | |
| notting | caillon: gah. the samsung thing somehow was reported in mid april and never hit the blocker list :/ | |
| wwoods | notting: yeah. | |
| warren | how prevalent are samsung disks? | |
| wwoods | and xinput stuff was reported like wild around beta | |
| * warren just wondering | ||
| caillon | notting, i hadn't seen the report date. | |
| warren | wwoods: xinput bug #? | |
| f13 | warren: or other changes invalidated prior tests. | |
| notting | caillon: 442457. april 14 | |
| wwoods | warren: ask caillon. | |
| warren | caillon: ? | |
| f13 | what we really need is an automated test system that can crank through these installer tests. | |
| wwoods | there's plenty. I'm not sure which specific one caillon's referring to. | |
| bpepple | f13: +1 | |
| caillon | warren, there's plenty of them. mostly the fact that we only fixed input for some sets of keyboards. ones that were combined keyboard/mouse type stuff didn't have a fix. | |
| f13 | wwoods: we also changed abunch of stuff related to input post-beta | |
| on again off again evdev | ||
| caillon | yeah | |
| and the fact that we like to go "hey, this feature is at 80%, well we're shipping so we need it to be 100%, let's just document what we have" | ||
| dwmw2 | f13: that kind of automation might be feasible with emulation/virtualisation. Although it doesn't cover "random hardware manufacturers use characters in their strings which we don't expect". | |
| f13 | caillon: that's a feature owner decision though right? Could have chosen to use the contengency plan. | |
| dwmw2: indeed it could. I was eluding to something that's in the works. | ||
| dwmw2 | and wouldn't cover "hey, what happens if we unplug the keyboard while the installer is running" either :) | |
| f13 | "eluding"? | |
| dwmw2 | alluding | |
| f13 | I hate spelling. | |
| dwmw2 | it eludes you? :) | |
| * jds2001 sends f13 hooked on phonics :) | ||
| caillon | f13, that is true... but given the choice of "it will be ready 2 weeks after the release, we'll push an update" vs "not have it working for another entire release cycle", tends to lead to option 1 | |
| for better or for worse | ||
| f13 | caillon: yes, and then the rest of us take the heat for all those people who install the release, and fail. | |
| I guess that's covered in the "for better or for worse" (: | ||
| maybe I'd just like to see more realistic contengency plans and more use of them when necessary | ||
| caillon | i suppose i meant "for better AND for worse" | |
| c4chris | but is it really the kind of thing (incomplete features) that leads broken sound or Samsung disk trouble ? | |
| caillon | c4chris, no, that was probably lack of triaging | |
| c4chris | that's my impression too | |
| caillon | if it was reported in time to get a fix, and nobody noticed. it's a problem, but a different one | |
| wwoods | the samsung bug is lack of testers combined with a triage failure | |
| dwmw2 | we're doing better with triage these days than we used to, aren't we? | |
| * jds2001 promises to do a better job :) | ||
| caillon | dwmw2, yeah, better definitely | |
| wwoods | yeah I expect we'll be a lot better off for the F10 cycle, far fewer bugs dropped on the floor | |
| caillon | i hope so | |
| dwmw2 | the Samsung thing... why in hell does it _matter_ what the name of the damn device is? | |
| f13 | 'triage failure' doesn't mean af failure on teh part of the triage team, just a failure in triageing the bug in general, by anybody | |
| dwmw2 | there's a certain lack of resilience there which is a concern in itself | |
| wwoods | dwmw2: that's mentioned in the bug, think davidz said he'd take a look at it, and then it was never revisited | |
| dwmw2 | isn't the point of exceptions that we can catch them? :) | |
| wwoods | either by testers or devs | |
| f13 | where is the samsung bug? | |
| caillon | i wonder why that never got put on the blocker list though... | |
| warren | http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9085160 Do we know what installer bugs this reviewer hit? | |
| notting | f13: 442457 | |
| caillon | was it because the reporter didn't have access to add bugs to F9Blocker? | |
| wwoods | bug 442457 | |
| buggbot | Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=442457 medium, low, ---, David Zeuthen, ASSIGNED , Unhandled exception in FC9 beta installation - "invalid object path" | |
| notting | caillon: anyone does | |
| caillon | ok | |
| f13 | caillon: anybody can add bugs | |
| in fact, too many do (: | ||
| warren: I think most our dualboot code is concerned with dualbooting with Windows or OSX, not necessarily other Linux operating systems (including our own) | ||
| it's just not the common/interesting case | ||
| notting | dwmw2: even if we caught it, it would make it non-installable to that drive | |
| caillon | f13, i'd rather have people add bugs that aren't blockers than not add bugs that are blockers. we can always remove bugs, but we can't guess what the blockers are :) | |
| warren | f13: sadly this particular review became a slashdot frontpage today =( | |
| f13 | caillon: I would too. | |
| wwoods | ugh, yes, that guy managed to hit every single stupid minor bug | |
| f13 | warren: I continue to not care about /. | |
| warren | f13: sadly it is a pretty accurate barometer on public perception, which we are still losing badly. | |
| f13 | wwoods: he sounds like the perfect person to invite to test our alpha/beta/prereleases for F10 | |
| wwoods | yup | |
| f13 | warren: I don't really believe that, but OK. | |
| wwoods | in a perfect world, anyway | |
| jwb | there is truth to what warren is saying | |
| * nirik notes that the biggest yelling in #fedora seems to be lack of nvidia binary only driver. | ||
| wwoods | yeah, we should really have that on the commonbugs page with a nice link to an nvidia page | |
| f13 | basically what we can learn from this is that we need dualboot coverage in our test matrix | |
| we'll get that done for F10. | ||
| jds2001 | nothing much to do about that :( | |
| bpepple | nirik: yeah, that's what I've noticed to, though there isn't really anything we can do about that. | |
| wwoods | oh and apparently nvidia is blaming us for shipping a beta xorg. ha ha ha! | |
| f13 | breaking nvidia drivers is a Fedora tradition. | |
| nirik | sure, but we could document/announce that it's the case better I think. | |
| f13 | I"d be sad if we didn't do it | |
| notting | i thought it was in the relnotes | |
| bpepple | f13: ;) | |
| nirik | notting: do people read those? :) | |
| jwb | there are inaccuracies in the relnotes too | |
| warren | f13: even if you don't care about slashdot specifically, this is a microcosm of our problem with media in general, where factless or narrow reports become what everyone understands. | |
| nirik | notting: I don't see it there. | |
| jwb | warren, that's where we need the marketing and ambassador teams to step up | |
| wwoods | It's not my job to shape perception | |
| f13 | warren: for the one bad report, I've seen a large number of very favorable reports out there | |
| notting | warren: that's a general media problem. it's not fedora-specific | |
| caillon | yeah | |
| warren | f13: me too | |
| f13 | warren: the /. crowd loves to throw rocks, so of course they're going to mod up the few stories that have problems in them. | |
| nirik | in terms of hits, digg seems more (by number) than /. now I think. | |
| jwb | perhaps we should start having one rep from each SIG attend this meeting | |
| f13 | jwb: they're all welcome | |
| but sure, we could encourage them to show up | ||
| warren | slashdot comments with its moderation system is a great way to *correct* mistakes, but I saw almost zero fedora people responding | |
| jwb | f13, i know that. i want to be a bit more forceful. then we can actually interlock with the teams that _are_ responsible for various items, like marketing and docs | |
| f13 | I would have liked more SIG representation in the QA and releng meetings too, SIGs taking on more responsiblility for the release in terms of their area of responsibility | |
| dgilmore | warren: i dont know anyone who takes any credence in slashdot | |
| warren | If you're trying to spread correct information it is worthwhile to participate there. | |
| jwb | so do that... | |
| wwoods | worthwhile for who? I don't think it's worth f13's time, f'rinstance | |
| warren | but anyway, back to the bugs | |
| notting | f13: well, perhaps instead of/in addition to a rel-eng meeting, or a FESCo meeting, we need to have a standing 'F10' meeting | |
| warren | how prevalent is the samsung issue? is it possible to fix that in updates.img? | |
| wwoods | as far as I can tell, every moment spent trying to educate the slashdot crowd would be better spent trying to improve Fedora | |
| jwb | notting, yes | |
| f13 | notting: sure, the more the merrier. | |
| notting: we /did/ have release readyness meetings, that were attended by docs, marketing, web, infrastructure, releng, qa | ||
| jwb | wwoods, for developers, QA, and releng | |
| nirik | might be worth for F10 having some folks setup to post to digg/slashdot, etc... and follow up and correct. Ie, assign some people to do so. | |
| warren | nirik: yes, that's all i'm asking for | |
| wwoods | jwb: strongly disagree! | |
| you're gonna have a tough time convincing me that my time is better spent commenting on slashdot than working on QA. | ||
| jwb | wwoods, strongly disagree with what? | |
| warren | it took hours after the release for anybody to post a fedora news item to slashdot and it turned out to be inaccurate | |
| f13 | warren: that's reasonable. | |
| jwb | wwoods, no no no no | |
| nirik | warren: yeah, I think thats a good idea... and if we assign some folks to it, no one else has to deal with it. ;) | |
| f13 | warren: I don't quite think that's true. /. has a tendency to ignore a lot of Fedora stories. | |
| jwb | wwoods, i'm agreeing with you on that part. however we do have teams that _should_ be responsible for that | |
| f13 | well a lot of stories in general. | |
| jwb | wwoods, like marketing... | |
| wwoods | ah, misunderstood you. yes. | |
| warren | f13: I saw monitoring the raw submissions | |
| f13 | warren: ok. Can you take a job item to marketing so that we get a .digg and /. submission ready to go before the release? | |
| and hit all the sites in a coordinated effort? | ||
| warren | I can't submit myself to slashdot if I'm going to moderate | |
| jwb | what? | |
| warren | it's one of their anti-abuse rules | |
| jwb | oh, i misread that | |
| warren | I used 30 mod points on the fedora 9 release article | |
| modding up correct info, modding down incorrect info | ||
| f13 | warren: I didn't say do it yourself, I said go to the marketing team and ask for somebody from there to do it. | |
| nirik | warren: did you mod any on the 'anaconda sucks' article? | |
| warren | nirik: ran out of mod points | |
| nirik | bummer. | |
| FYI, that article in case folks didn't see it: http://linux.slashdot.org/article.pl?sid=08/05/14/2145254 | ||
| caillon | there's an anaconda sucks article? heh | |
| warren | anyway this is extremely off topic, we can talk about this later, let's focus on the bugs. | |
| stickster | warren: There's a Marketing meeting at 3pm today in #fedora-mktg. | |
| warren: Please stop by. | ||
| warren | stickster: sure. | |
| stickster | thanks :-) | |
| wwoods | oh hey - Spaceman wants to sync cycles with us | |
| neat | ||
| jwb | is it? | |
| stickster | 'Cos Wall Street will LOVE that. | |
| f13 | wwoods: well, our cycles are predictable. They're welcome to match 'em | |
| warren | wwoods: he's welcome to follow us. | |
| f13 | (but I think he was more talking about RHEL and Novell SLES) | |
| stickster | Actually, he never mentions Fedora, just RHEL. | |
| f13: +1 | ||
| wwoods | what a dink. | |
| c4chris | spaceman spiff ? | |
| warren | the media seemed to get this one right. they pointed out that Canonical has to really step up to the plate in core contributions to be taken seriously | |
| they also point out UnitedLinux as a failure to sync enterprise schedules | ||
| This is REALLY OFF TOPIC | ||
| What do we do about the bugs? | ||
| bpepple | warren: agreed. | |
| jwb | c4chris, mark shuttleworth. ubuntu guy | |
| f13 | warren: add test cases to the QA tree | |
| c4chris | jwb: thx | |
| bpepple | warren: work on recruiting more people to join the QA team. | |
| wwoods | add test cases; try to get a person (people) working on a test-plan tracker | |
| making it easier to create test cases *and* submit results *and* give people credit for doing so | ||
| warren | Attempting to do QA in the beta and RC time, it seemed like the vast majority of testers didn't appear until the last minute. | |
| notting | wwoods: with test plans for each Feature, and a schedule to have them done? | |
| wwoods | will (and does) lead to increased test participation | |
| warren | simply doing releases wasn't enough | |
| wwoods | e.g. we suddenly tripled our tester pool when I put the "tested by" field in the matrix | |
| warren | wwoods: Will or will? =) | |
| wwoods | heh! small-w will. | |
| f13 | warren: I don't think beta/rc had anything to do with it, I think it was more that QA as a community effort was getting noticed to exist | |
| warren: I fully expect as much if not more people hitting F10 alpha test cases than we had for Preview. | ||
| wwoods | we have an intern this summer who's got experience with web design stuff and is learning python | |
| caillon | f13, i hope you're right | |
| wwoods | I'm thinking that working on things like this would be a very good use of time | |
| f13 | caillon: I believe I am (: | |
| wwoods | further I think we're going to repurpose the seemingly-disused 'qa' FAS group to track testers | |
| caillon | f13, is that because there was 1 person on preview and will be pretty hard to have less? ;) | |
| f13 | caillon: I also believe we can eventually sweeten the pot with honorable mentions in places, maybe USB keys sent to key QA folks, etc... | |
| wwoods | and we want to start getting metrics from bugzilla/bodhi/FutureUnnamedTestPlanTracker | |
| f13 | caillon: eh, so the nubmers on the Final tree | |
| wwoods | RHEL QE has (informally) pledged budget to give Cool Stuff to testers | |
| which is one motivation for tracking metrics | ||
| caillon | f13, i'm all in favor of recognition and reward. that's one of the ways moz got its big test community. | |
| bpepple | wwoods: sounds good. | |
| f13 | since we're talking about bringing back the slideshow during install, we can use one or more of the slides to recognize the QA team | |
| wwoods | anyone know what the FAS 'qa' group is for? | |
| bpepple | wwoods: no idea. | |
| wwoods | f13: yeah, once we have the metrics pretty solid I'd love to be able to give shout-outs to the top 20 testers or whatever | |
| wwoods | bpepple: f13 owns it and he doesn't have any idea either. heh | |
| f13 | wwoods: since I own it, and I don't use it, feel free to repurpose it. | |
| warren | wwoods: Cool Stuff or cool stuff? | |
| (the former sucks) | ||
| wwoods | warren: the latter | |
| warren | good | |
| caillon | wwoods, ETA on that? | |
| wwoods | on.. what? | |
| f13 | cool stuff | |
| caillon | metrics? | |
| f13 | caillon: Stay Tuned for FUDCon | |
| wwoods | metrics? jeez. no idea. still trying to get preupgrade out the door, haven't talked to Andy about any of this | |
| caillon | ok | |
| f13 | caillon: I fully expect something usable for the latter half of F10 | |
| wwoods | yeah I just booked my travel stuff for FUDCon | |
| caillon | f13, superb | |
| jds2001 | wwoods: super :) | |
| wwoods | lmacken sez getting bodhi metroids is easy | |
| hah! metrics | ||
| f13 | wwoods: LOL! | |
| wwoods | stupid meat-based autocomplete | |
| caillon | metroid <3 | |
| jds2001 | lol | |
| wwoods | anyway, bugzilla is trickier but doable | |
| f13 | better than bodhi hemmoroids | |
| jds2001 | bugzilla metroids are a little harder but doable | |
| lol | ||
| wwoods | what metrics would people suggest? | |
| actually maybe that conversation should happen on a list or something | ||
| but I don't want to get into bikeshed-painting | ||
| warren | a list of install failures avoided due to QA during the F9 cycle? | |
| jds2001 | right, lets work something out at fudcon | |
| warren: a lot :) | ||
| wwoods | I also don't want to hijack the meeting with a brainstorm session, so you could email me and/or we'll talk at FUDCon | |
| bpepple | wwoods: sounds good. | |
| wwoods | warren: I'm talking about per-tester metrics mostly | |
| overall QA metrics are good but have less direct benefit to QA | ||
| I can see the value, but per-tester metrics -> recognition / rewards -> more testers | ||
| f13 | wwoods: I think the first one to go after is number of test matrix items knocked off during a release. | |
| wwoods: and exclude you me, and um... | ||
| wwoods | yeah, but since that's a wiki page it's a hand-count, which doesn't scale well | |
| still, worth doing | ||
| f13 | wwoods: second would be new test cases submitted and accepted. | |
| wwoods: again, stay tuned for FUDCon... | ||
| f13 | (or, go talk to bpeck) | |
| nirik | bonus for test plans that have ks files/snake info to automate them as much as they can be. | |
| wwoods | oh def | |
| notting | can we actually start dogtailing anaconda? | |
| warren | is dogtail still alive? | |
| caillon | yeah | |
| wwoods | I can't remember if the patches were merged | |
| * warren is trying to convince evil company to use dogtail | ||
| wwoods | but ISTR some internal test cases involving using dogtail in anaconda | |
| poelcat | ubuntu references dogtail in some of their test cases | |
| f13 | yeah, I think it's there, just waiting to be exercised. | |
| otherwise, you could dogtail virt-manager and install anaconda in there (: | ||
| wwoods | I.. don't think that works | |
| wwoods | dogtail operates at the gtk widget level IIRC | |
| f13 | wwoods: I'm just being ....creative | |
| wwoods | and your VNC session is just a single canvas-style widget | |
| warren | you would need dmalcolm's OCR-based window recognizer for that | |
| f13 | nod | |
| nirik | if there was a package that would do a 'this install is good/ok/sane' we could add it to ks files with snake and have it go do a test, see it was ok, then fire off the next snake test, etc. | |
| warren | (he never did release it, it was pretty cool, although unfinished) | |
| f13 | anywho, we're drifting again | |
| bottom line, there are some QA improvements on the horizon, and there are more community interest in QA work, and we plan to grow that via various means. | ||
| caillon | sounds good | |
| * caillon approves | ||
| c4chris too :) | ||
| bpepple | ok, we're almost to the hour point. anything else we want to discuss? | |
| wwoods | clumens says dogtail is in anaconda these days | |
| jlaska | it is ... there's a big "but" there though | |
| caillon | skvidal, do you have an update on the ssl fallout? | |
| skvidal | caillon: nothing so far | |
| 150ish accounts had their ssh keys suspended | ||
| b/c they were weak keys | ||
| caillon | ok. (me notes that there may be more that the key didn't detect, which is the only sort of worrying thing for me) | |
| nirik | does FAS reject them just reuploading the weak key again? | |
| warren | skvidal: if you received an e-mail your key was suspended? | |
| tibbs | How would you know? I think I actually have a DSA key in there; were we going to toss those out as well? | |
| skvidal | warren: I haven't sent the secon email out yet | |
| tibbs: no, we're not | ||
| warren | skvidal: perhaps it isn't well enough known that simply if you had used your Fedora generated DSA key to login to a Debian/Ubuntu box then it may have been compromised. | |
| tibbs | I think RSA was still patented when I created my keys. | |
| f13 | we should warn about DSA though, perhaps. | |
| warren | yeah | |
| jwb | all of my keys are DSA.. | |
| what's wrong with DSA? | ||
| skvidal | okay | |
| here's the problem | ||
| stickster talked to a really serious crypto expert who said | ||
| "it is not obvious anything is wrong with dsa" | ||
| warren | (the early years of fedora even demanded that users use DSA keys instead of RSA because of a misunderstanding of gafton) | |
| caillon | jwb, DSA relies on the PRNG of both the client and remote hosts, is the thing causing the concern | |
| jwb | doesn't ssh-keygen default to dsa? | |
| caillon | no, rsa | |
| * stickster sighs... he made a bunch of comments earlier just as his wifi kicked the bucket. | ||
| jwb | what fedora version was that changed in? | |
| caillon | i don't know | |
| jwb | ok | |
| skvidal | jwb: no clue - but you can check your key easily enough | |
| jwb | skvidal, ok. how? | |
| stickster | jwb: rsa | |
| skvidal | open up your pub key | |
| look for the string 'ssh-dss' or 'ssh-rsa' | ||
| f13 | well, if it's named id_pub.dsa it's probably... dsa | |
| jwb | yes, it's ssh-dss | |
| i already _knew_ the were DSA keys | ||
| caillon | id_dsa.pub | |
| jwb | anyway, offtopic | |
| skvidal | I'm planning on posting a blog post that is more or less "If you do not understand the OpenSSL issue then replace your ssh keys" | |
| caillon | jwb, the other thing is that RSA defaults to 2k and DSA is limited to 1k by ISO, so it's "weaker" in general. | |
| skvidal | I'd encourage others to do the same | |
| caillon | i've already replaced mine just to be on the safe side | |
| f13 | skvidal: yeah, I've done so on the various Linux user lists I'm no | |
| on | ||
| (and yet I've failed to replace my own keys thus far) | ||
| bpepple | ok, we out of time. anything else? otherwise I'm going to wrap it up. | |
| stickster | My lost comments seem to have been covered in the conversation about needing test cases | |
| warren | the election | |
| stickster | Some other things that might help: | |
| Step-by-step instructions on how to mirror & create testing media nightly. | ||
| Guidance about what's being QA'd regularly so community people don't duplicate efforts? | ||
| eof | ||
| sorry to rehash. | ||
| jwb | nirik, were you working on a "boot and install in kvm" setup at all? | |
| jds2001 | stickster: step by step of creating media is on my blog | |
| mirroring i can write. | ||
| stickster | jds2001: OK, a wiki destination would be good for posterity | |
| nirik | jwb: I have some virt-install scripts I use to install rawhide in kvm most every day... other than that no... | |
| poelcat | it needs to be part fo our core docs | |
| bpepple | warren: what about the election do you want to discuss? | |
| jds2001 | i responsed to your mail on f-mktg this morning, too. | |
| warren | bpepple: seems nobody is up for nomination | |
| jds2001 | poelcat: agreed. | |
| warren | bpepple: we're going from 13 to 9 seats is it? | |
| jwb | yes | |
| bpepple | warren: correct. FESCo will be 9 seats after the next election. | |
| warren | I'm a bit concerned that with the usual # of people who don't show up for meetings, it might be difficult to maintain discussions with fewer overall seats. | |
| bpepple | there's still 2 weeks left for the nomination period, and I'll be sending out more reminders to the list. | |
| * nirik still hasn't decided to run again or not... | ||
| * bpepple hasn't either. | ||
| warren | So I'm thinking of not running for this next election, because I am concerned about having fresh blood in FESCo. | |
| tibbs | I haven't decided either. | |
| c4chris | warren: if this happens, the # can be increased again | |
| warren | OTOH, we need both fresh blood and existing experienced leaders in FESCo for it to be funtional. | |
| bpepple | warren: agreed. | |
| warren | thus fewer seats makes this difficult | |
| jwb | i don't really agree with that | |
| dgilmore | warren: we already decided this long ago. | |
| warren | compounded by the FOO # who don't show up to any given meeting | |
| sure | ||
| notting | well, i don't see how fewer seats -> less nominees logically follows | |
| warren | I'm just saying why I'm thinking of not running | |
| dgilmore | warren: we can revist after going for 6 months with 9 | |
| bpepple | warren: most meetings we are only missing one or 2 people. | |
| warren | bpepple: yes, and that was less problematic 2 missing people is a smaller % | |
| f13 | perhaps those 2 missing people won't rejoin. | |
| jwb | i'll be running again i think | |
| at least that's what my board nomination says | ||
| bpepple | I would hope that people woudn't run for FESCo if they can't regularly attend the meetings. | |
| jwb | i would hope they at least try, given that the meeting time can be changed... | |
| f13 | yeah, that's always been a stipulation of FESCo that after each election we can adjust meeting time to match current electees | |
| warren | I guess question for this meeting, should we more actively ask for nominations to fesco? | |
| warren | nominations need not be only self-nominations | |
| how are we going to encourage fresh blood? | ||
| foster the next generation of leaders | ||
| wwoods | maybe I'll run for a seat | |
| I'm always here anyway | ||
| jwb | you should | |
| f13 | actually a Red Hat Magazine article about this would be awesome | |
| warren | what would the point of that article be? | |
| f13 | showing the various ways that every day people can influence the direction of Fedora from various different places. | |
| FESCo, Fedora Board, Packaging, Releng, QA, Marketing, etc.. | ||
| warren | when do nominations close? | |
| f13 | all these places that have open elections | |
| bpepple | warren: June 1st, if I remember correctly. about 2 weeks. | |
| warren | ok, I guess time to end the meeting? | |
| c4chris | yup | |
| bpepple | warren: yeah. | |
| * bpepple will end the meeting in 60 | ||
| bpepple will end the meeting in 30 | ||
| bpepple will end the meeting in 15 | ||
| bpepple | -- MARK -- Meeting End | |
Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!