15:00:16 <jlaska> #startmeeting Fedora QA Meeting
15:00:16 <zodbot> Meeting started Mon Mar 28 15:00:16 2011 UTC.  The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:16 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:23 <jlaska> #meetingname fedora-qa
15:00:23 <zodbot> The meeting name has been set to 'fedora-qa'
15:00:28 <jlaska> #topic Roll Call
15:00:35 * tflink is present
15:00:40 * kparal is here
15:00:44 * mkrizek visits
15:00:44 <jlaska> Hello gang ... show of electronic hands time
15:00:52 <jlaska> woah ... welcome back mkrizek!
15:01:00 <jlaska> Hello tflink + kparal
15:01:18 * vhumpa hello there!
15:01:39 <jlaska> howdy vhumpa
15:01:48 <kparal> mkrizek: looking forward to hearing autoqa updates? :-)
15:02:05 <jlaska> Anyone else lurking?  robatino Viking-Ice adamw ?
15:02:15 * robatino is here
15:02:23 * jlaska greets
15:02:27 <mkrizek> kparal: yep, can't wait:)
15:02:34 <jlaska> okay, let's get started
15:03:00 <jlaska> btw ... I may have to go on radio silence for a moment during the meeting.  kparal was kind enough to agree to co #chair
15:03:04 <jlaska> #chair kparal
15:03:04 <zodbot> Current chairs: jlaska kparal
15:03:20 <jlaska> I'll be walking through the agenda posted on the wiki - https://fedoraproject.org/wiki/QA/Meetings/20110328
15:03:25 * jsmith lurks
15:03:40 <jlaska> lurker!
15:03:49 <jlaska> #topic Previous meeting follow-up
15:04:10 <jlaska> We had 3 items on the list, 2 of which pertain to test days, so we can cover those when we review ucoming QA events
15:04:14 <jlaska> Hey, there's jskladan!
15:04:18 * jskladan lurks
15:04:22 <jskladan> jlaska: hey there
15:04:25 <jlaska> #info jlaska - provide updated boot.iso for testing new lorax+anaconda builds so testers can provide karma
15:04:42 <jlaska> So thanks to everyone who helped test the newer anaconda and lorax builds last week
15:04:46 * jlaska looking at robatino
15:04:51 * Viking-Ice checks in
15:05:08 <jlaska> I just updated my boot.iso to include the proposed anaconda-15.25-1 and NM-0.9 updates
15:05:16 <jlaska> Hey Viking
15:05:41 <jlaska> I'll send another request out to the list, but we have another anaconda update that needs some karma so it can land in Beta TC1
15:05:44 <jlaska> #link https://admin.fedoraproject.org/updates/anaconda-15.25-1.fc15
15:05:56 <jlaska> That's all from me on this one
15:06:00 <jlaska> anything I missed from last week?
15:06:31 <jlaska> alright then, let's dive in
15:06:41 <jlaska> #topic Rescheduling GNOME3 test day
15:06:49 <jlaska> vhumpa requested adding this topic
15:06:56 <jlaska> I believe we may need adamw around for this though
15:07:18 <vhumpa> Yes, he needs to be around
15:07:42 <jlaska> vhumpa: can you highlight the problem again, for the minutes?
15:07:53 <vhumpa> jlaska: of course
15:08:08 <tflink> :q
15:08:13 <tflink> sorry, wrong window
15:08:17 <jlaska> tflink: heh
15:08:31 <vhumpa> The test day is scheduled on 7th of April now, which colides with the GNOME3 release again
15:08:53 <jlaska> #info The test day is scheduled on 7th of April now, which colides with the GNOME3 release again
15:08:55 <Viking-Ice> the third time then right ;)
15:09:06 <vhumpa> So, like last time, there would likely be a need to push important updates to the test day image at the last moment
15:09:14 <kparal> when's the release date?
15:09:22 <vhumpa> Which seems to make problems, notably to AdamW's sleep :)
15:09:34 <vhumpa> kparal: 7th April
15:09:39 <kparal> vhumpa: ah, I see
15:09:54 <vhumpa> Thus we though to move the test day to later date
15:10:07 <vhumpa> Next week though is a virtualization test day
15:10:09 <Viking-Ice> reschedule the Gnome 3 test after the upstream release ?
15:10:28 <vhumpa> viking-ice: yes
15:10:44 <kparal> well, does it have to be Thursday? I know it's common practice, but what about Friday?
15:10:46 <jlaska> Just trying to find a good fit ... and not interfere with any other test events
15:10:48 <kparal> or other day next week?
15:10:52 <vhumpa> First free slot is on 21st, which might just be too late
15:11:01 <jlaska> vhumpa: adamw proposed the Tuesday before the virt event
15:11:04 <Viking-Ice> you dont have to stick with the slots
15:11:08 <vhumpa> kparal: it doesnt
15:11:18 <jlaska> I was just concerned that we might overload that week with two *large* test days (GNOME3 and Virt)
15:11:20 <vhumpa> Also no problem with multiple test days per week
15:11:40 <vhumpa> The issue is exactly what jlaska just mentioned
15:11:54 <vhumpa> So, what we can do...
15:11:58 <Viking-Ice> the upcoming printing test day was not set on a "schedule" slot
15:12:23 <vhumpa> Move the test day just to 8th (Friday) and hope the extra day gives time to evade trouble
15:12:26 <Viking-Ice> and date that suited the maintainer was simply found..
15:13:04 <jlaska> vhumpa: the upstream release is on Apr 7th, right?
15:13:21 <tflink> is 1 day going to be enough for the released packages to get into repos?
15:13:26 <vhumpa> jlaska: yes I think so
15:13:27 <jlaska> I don't think so
15:13:33 <Viking-Ice> just simply ask the Gnome desktop team when it's best for them to hold the test day ?
15:13:34 <jlaska> Let's at least push the test day to the following week?
15:13:36 <vhumpa> Likely Nope...
15:14:03 <jlaska> vhumpa: I think Viking-Ice has the right approach
15:14:15 <vhumpa> We could also talk to virt guys to swap
15:14:22 <Viking-Ice> you dont have to swap
15:14:23 <jlaska> definitely
15:14:38 <Venemo_N900> am I late for the QA meeting?
15:14:42 <jlaska> Venemo_N900: nope, welcome
15:15:02 <vhumpa> Let's ask the deskop guys then
15:15:17 <jlaska> vhumpa: from there we can adjust/swap if needed
15:15:22 <Viking-Ice> yup
15:15:31 <jlaska> vhumpa: you want to reach out to desktop@ ?
15:15:47 <vhumpa> jlaska: Will do
15:15:54 <jlaska> excellent, thanks
15:16:12 <jlaska> adamw noted he may have already started discussion on this topic, but I don't know if a date was ever selected
15:16:30 <vhumpa> We have a bunch of them here, so no problem asking :)
15:16:35 <jlaska> yeah
15:16:37 <jlaska> #topic Upcoming QA events
15:16:49 <kparal> #action vhumpa reach out to desktop team to disscuss best day for upcoming test day
15:16:56 <jlaska> kparal: can you take over ... ?
15:17:00 <kparal> jlaska: ok
15:17:31 <kparal> Tuesday, March 29 - Beta 'Test Compose'
15:17:45 <kparal> jlaska - will provide updated boot.iso for testing new lorax+anaconda builds so testers can provide karma
15:17:52 <kparal> I think jlaska already covered this
15:18:19 <kparal> he will send out email asking for help in testing the new builds
15:18:45 <kparal> Tuesday, March 29 - Printing Test Day
15:18:49 <Venemo_N900> may I have a few suggestions regarding the desktop in F15?
15:18:58 <Venemo_N900> I'm not sure if this is the right place to discuss
15:19:00 <kparal> tflink to check in with twaugh on preparations for printing test day
15:19:22 <kparal> Venemo_N900: in #fedora-qa or on desktop channel
15:19:33 <kparal> tflink: do you have some updates on this?
15:20:02 <tflink> I sent an email to twaugh asking about the status of the test day. I heard back from him on tuesday, saying that he hadn't forgotten but had been busy
15:20:42 <kparal> the wiki seems kind of ready
15:20:44 <tflink> he said that he was planning to send out an announcement later that day (also in qa trac #172)
15:21:10 <kparal> I think it wasn't sent yet
15:21:14 <tflink> I don't believe that I saw an announcement for the test day, though
15:21:32 <tflink> I'll poke him about sending that out today
15:21:38 <kparal> tflink: ok, can you try to contact twaugh about the announcement or send it yourself?
15:21:41 <kparal> thanks
15:22:04 <tflink> either/or. I would want to make sure that they're ready before sending out the announcement, though
15:22:15 <kparal> tflink: sure
15:22:35 <kparal> #action tflink to handle announcements for Printing test day
15:22:51 <kparal> one more thing
15:23:00 <kparal> there seems to be a change of URLs for nightly composes
15:23:11 <kparal> this doesn't work any more: http://alt.fedoraproject.org/pub/alt/nightly-composes/desktop/
15:23:21 <kparal> instead see http://alt.fedoraproject.org/pub/alt/nightly-composes/current.html
15:23:43 <kparal> so it's not as easy to link to a current image from the wiki page
15:24:06 <kparal> we will have to decide whether to link to an image from a particular day, or link to current.html page (may be confusing for people)
15:24:30 <adamw> oh fer...
15:24:36 <adamw> why do i never seem to be able to get meeting times right lately
15:24:37 <adamw> hi
15:24:48 <kparal> adamw: hello!
15:25:13 <kparal> ok, if there is nothing to add to Printing Test Day...
15:25:42 <adamw> we might want to do more about the nightly URL thing
15:25:47 <adamw> but we can put that in open discussion...
15:26:00 * jlaska back
15:26:02 <kparal> adamw: ok, let's put it there
15:26:06 <kparal> adamw: please read backlog about gnome3 test day date and comment if you have something further on your mind
15:26:30 <kparal> next on:
15:26:32 <kparal> Thursday, March 31 - ABRT/Retrace Server Test Day
15:26:45 <kparal> # kparal to check with jmoskovc what is needed about ABRT test day
15:27:06 <adamw> kparal: i already reached out to desktop team about changing the date, the thread never really went anywhere.
15:27:10 <adamw> at this point we should probably just do it.
15:27:26 <adamw> but we can come back to it later
15:27:30 <kparal> adamw: lets go back to that later, thanks
15:27:38 <kparal> so, abrt test day
15:27:57 <kparal> I did checked and was told that abrt guys were working on those test cases
15:28:02 <kparal> they don't need any wiki help
15:28:19 <jlaska> cool, [X] Check! :)
15:28:20 <kparal> but they might need to build a custom LiveCD if nightly compose is not good enough for them
15:28:31 <kparal> I offered my services in that case
15:28:47 <kparal> apart from that, they didn't ask for anything else
15:28:57 <jlaska> thanks ... we also have help from EDT and PDT if needed for building the ISO's
15:29:15 * kparal not familiar with those abbreviations
15:29:29 <jlaska> sorry, just timezones (eastern-US and pacific-US)
15:29:38 <kparal> ah, yes
15:29:55 <kparal> ok, /me giving mic back to jlaska
15:30:16 <kparal> we can go back to gnome3 test day or go  forth to the next topic
15:30:35 <jlaska> kparal: rats, you were doing good! :)
15:30:58 <jlaska> #info Friday, April 1 - F-15-Beta Blocker #4
15:31:17 <jlaska> Just a reminder ... the fourth beta blocker review will be hosted again this Friday
15:31:26 <jlaska> #link http://bit.ly/f15-beta-blocker-proposed
15:31:31 <jlaska> #link http://bit.ly/f15-beta-blocker-accepted
15:31:57 <jlaska> kparal: I think that's it for upcoming events, did I miss any?
15:32:04 <jlaska> if not ... we'll jump back to GNOME3
15:32:06 <kparal> I don't think so
15:32:16 <jlaska> okay, thanks for driving
15:32:26 <jlaska> #topic Rescheduling GNOME3 test day
15:32:44 <adamw> so yeah, i think waiting for desktop team on this may not work
15:32:49 <jlaska> okay ... with adamw here now ... is there anything else to discuss on this topic?
15:32:58 <adamw> they seem to take a 'ok, whatever' approach to scheduling - at every point i asked for their input but then just went ahead and did it
15:33:04 <jlaska> okay
15:33:21 <vhumpa> I just got exactly that from one of them :)
15:34:05 <vhumpa> Well, actually - I was wrong about the release day
15:34:07 <adamw> so, wait a sec, the open slot we have is 21st?
15:34:08 <vhumpa> It is on 6th
15:34:26 <vhumpa> adamw: yes
15:34:51 <adamw> i think i've been looking at an old calendar when being worried about that date
15:34:57 <adamw> now we have a two week slip, 21st actually should be fine
15:35:10 <adamw> final change deadline isn't till may 9th, now
15:35:33 <adamw> so i think we can probably just go ahead with 21st, unless anyone else is worried about it
15:35:49 <vhumpa> adamw: that sounds good
15:35:54 <jlaska> no worries here
15:36:32 <vhumpa> Also makes a buffer time if something goes wrong with G3 release finally
15:36:40 <adamw> true
15:37:08 <jlaska> #info Reschedule for Apr 21st - plenty of time before F-15 final change deadline
15:37:27 <jlaska> thanks vhumpa + adamw
15:37:41 <jlaska> #topic Open Discussion - <your topic here>
15:37:48 <jlaska> That's all I had on the agenda for today
15:38:03 <jlaska> Any topics folks would like to cover?
15:38:06 * kparal proposes AutoQA updates
15:38:19 * vhumpa is running home
15:38:25 <jlaska> vhumpa: cya
15:38:31 * kparal created quite some notes in advance
15:38:37 <jlaska> #topic AutoQA Updates
15:38:43 <jlaska> kparal: take it away when you are ready
15:38:57 <kparal> mkrizek: wake up, the time has come :)
15:39:02 <jlaska> hehe
15:39:11 <kparal> I'd like to shortly recapitulate the latest news about AutoQA and its "send Bodhi comments" feature. As you probably noticed, I posted the following message as a blogpost and into devel and test lists:
15:39:17 <kparal> #link https://kparal.wordpress.com/2011/03/21/autoqa-will-be-providing-comments-for-fedora-updates-really-soon/
15:39:27 <kparal> After we enabled Bodhi commenting last week, a lot of problems popped up. That was certainly expected, although not desired. We turned off the feature after only a single hour or so, because we received many reports of misbehaviour. It turned out that some mistakes were caused by AutoQA bugs, some mistakes were caused by Bodhi bugs.
15:39:44 <kparal> #info Bodhi should be fixed now, thanks lmacken for quick response.
15:40:01 <kparal> #info As for AutoQA, we're working on fine-tuning all those errors and would like to make a bugfix release soon (next week? just a guess). Then we will turn the commenting feature on again.
15:40:06 <jlaska> yeah, that was some fast action resolving that stuff
15:40:18 <tflink> it looks like there are some packages with stale tags hanging around still
15:40:42 <jlaska> tflink: do you have a script you are using to detect?
15:40:51 <kparal> tflink: if you have some links, feel free to create a ticket for bodhi and CC us, thanks
15:40:53 <tflink> nope, just looking at the upgradepath results
15:41:15 <kparal> tflink: well, there were some bugs in upgradepath
15:41:18 * tflink will put together a list of stale tags today and submit a ticket
15:41:27 <kparal> tflink: builds containing epoch were not detected correctly
15:41:48 <kparal> it's better to check manually using "koji" command or so
15:42:05 <tflink> kparal: OK, I was planning to double check with koji before submitting anyways
15:42:14 <kparal> tflink: great
15:42:24 <jlaska> #action tflink will gather a list of stale -pending tags and determine whether autoqa or bodhi bug
15:42:29 <jlaska> tflink: thx
15:42:36 <tflink> jlaska: np
15:42:43 <kparal> so, I'll post another email/blogpost when we're nearing the date when we should enable Bodhi comments again
15:42:52 <kparal> I also expect this whole cycle to repeat itself a few times untill we have it all perfectly running.
15:43:29 <kparal> That's all from the recap. A few notes about other recent updates:
15:43:35 <kparal> #info tflink added new method to our library allowing us to easily download RPMs for a specific build
15:43:39 <kparal> This is his first push to master IIRC, yay! :)
15:43:42 <jlaska> I'll stay tuned to do a new stable tag + build when requested
15:44:17 <jlaska> first of many, I'm sure
15:44:25 <kparal> definitely
15:44:31 <kparal> And similarly for vhumpa:
15:44:35 <kparal> #info vhumpa committed a patch splitting watcher scripts and event definitions into separate directories (abolishing 'hook' word). That should make the architecture more comprehensible. He's now in the process of adjusting wiki documentation to reflect it.
15:44:43 <tflink> I should hope so. I'd be worried if my only contribution to autoqa was a single utility method :-D
15:44:50 <jlaska> tflink: haha!
15:45:14 <kparal> And jskladan wasn't idle either:
15:45:20 <kparal> #info jskladan has provided patches for upgradepath, which was incorrectly reporting results to updates consisting of more than 1 build
15:45:25 <kparal> Not yet in master, it will be there any time soon.
15:45:36 <jlaska> kparal: I gather that'll also land in stable?
15:45:42 <kparal> jlaska: yes
15:45:45 <jlaska> cool
15:45:57 <jlaska> nice work gang ... looking forward to 0.4.5-1
15:46:10 <kparal> which reminds me...
15:46:17 <kparal> #info AutoQA will following a better versioning scheme from now on. Major number bump for very large changes, minor number bump for standard feature additions and standard-size changes, and revision number bump for bugfix releases. Roughly :) That should help our packages to be more predictible and easily distinguish bugfix releases from feature releases or packaging changes.
15:46:23 <kparal> #info We have even created 'stable' git branch that should always contain the latest stable code (released or about to be released).
15:46:50 <jlaska> kparal: and by better, you mean "actual"  ... and note my lazy release bump method  :)
15:46:55 <jlaska> s/note/not/
15:47:08 * jlaska got caught being lazy
15:47:32 * kparal is denying that
15:47:42 <jlaska> Anything else on AutoQA?
15:47:48 <kparal> ok, that's all from AutoQA world for today
15:47:57 <jlaska> thanks for the updates
15:48:05 <jlaska> #topic Open Discussion - <your topic here>
15:48:27 <jlaska> I had a quick question ... anyone know what the latest build status is for the NM-0.9 stuff?
15:48:41 <jlaska> I see a bodhi update ... I'm guessing we're just waiting for the repo to update?
15:49:06 <jlaska> #link https://admin.fedoraproject.org/updates/pidgin-2.7.11-2.fc15,telepathy-glib-0.14.1-1.fc15,telepathy-logger-0.2.6-1.fc15,libsocialweb-0.25.11-3.fc15,kdebase-runtime-4.6.1-3.fc15,epiphany-2.91.92-2.fc15,evolution-2.91.92-2.fc15,control-center-2.91.92-3.fc15,kde-plasma-networkmanagement-0.9-0.44.20110323.fc15,kdebase-workspace-4.6.1-5.fc15,empathy-2.91.92-2.fc15,geoclue-0.12.0-7.fc15,krb5-auth-dialog-2.91.91-3.fc15,gnome-shell-2.91.
15:49:10 <adamw> jlaska: yes, as far as the system is concerned, it's stable.
15:49:13 <jlaska> ouch
15:49:18 <adamw> so if you don't have it yet, it's just compose / mirror sync
15:49:22 <jlaska> okay
15:49:37 <jlaska> cool ... and what's up with libreoffice, is that in the same boat?
15:49:57 <jlaska> #link https://admin.fedoraproject.org/updates/libreoffice-3.3.2.2-2.fc15
15:50:19 <adamw> 3.3.2.2-2.fc15 has been submitted but not pushed
15:50:24 <Viking-Ice> I want to mention one thing
15:50:32 <adamw> the summary screen on bodhi is actually more useful when checking this, btw
15:50:35 <adamw> see https://admin.fedoraproject.org/updates/search/libreoffice
15:50:38 <adamw> it makes it really obvious
15:51:00 <jlaska> adamw: ah I see, thanks
15:51:03 <jlaska> Viking-Ice: what's up?
15:51:25 <jlaska> #info libreoffice and NM-0.9 updates working through system and should hit mirrors soon
15:51:57 <Viking-Ice> I propose that we enable we convert_html_to_plaintxt option in mailman and reject the message on the bases that it it violates the list guide lines and provide http://fedoraproject.org/wiki/Communicate/MailingListGuidelines#No_HTML_Mail.2C_Please. . for the test list
15:52:17 <Viking-Ice> one too many we there..
15:52:31 <jlaska> have a few HTML messages landed recently?
15:52:40 <Viking-Ice> yup
15:52:41 <adamw> um, can we finish discussing networkmanager update first?
15:52:47 <Viking-Ice> yup
15:52:57 <jlaska> adamw: oh, didn't realize we were still discussing it, what'd I miss?
15:53:01 <adamw> so one thing we should probably do is check and confirm other desktops are okay
15:53:18 <adamw> is knetworkmanager working okay for everyone in kde, is nm-applet actually still working in lxde/xfce
15:53:21 <jlaska> if we want to do this prior to TC1, we have a small window to build custom live images
15:53:34 <jlaska> #topic NM-0.9 update status
15:53:40 <adamw> yeah. i can build some for myself to test but uploading them for anyone else would take ages.
15:53:50 <adamw> other thing you can do, of course, is just test on an installed system
15:53:57 <adamw> yum install @desktop_of_choice
15:54:00 <jlaska> #info the bodhi update is working it's way through, and will hit compose/mirror_sync soon
15:54:09 <nirik> adamw: works in xfce here. :) had to reboot tho after the update.
15:54:27 <jlaska> #info nirik reported that XFCE works well with NM-0.9 update
15:54:28 <adamw> nirik: great
15:54:50 <jlaska> I think I saw confirmation of knetworkmanager in the bodhi update
15:55:00 <jlaska> anyone want to volunteer for LXDE confirmation?
15:55:30 <adamw> i can check it.
15:55:35 <adamw> i'll look at the others too, just in case.
15:55:43 <Viking-Ice> I could have had my T61p not broken down..
15:55:53 <jlaska> #info adamw volunteer to test LXDE + NM-0.9
15:56:13 <adamw> if it works in xfce, though, i don't expect any problems
15:56:22 <jlaska> adamw: I need to knock out testing of custom boot.iso and DVD.iso stuff ... I expect that'll chew up my afternoon
15:56:31 <adamw> funsies
15:56:40 <jlaska> anything else on this topic?
15:57:25 <jlaska> #topic Proposal to enable convert_html_to_plaintxt
15:57:34 <jlaska> okay, now to Viking-Ice's topic
15:57:41 <Viking-Ice> So basically The convert_html_to_plaintxt option tells Mailman to strip all HTML from the message and deliver it as plain text.
15:57:45 <adamw> no objections from me
15:57:53 <jlaska> yeah, this seems like a "good thing"
15:58:03 <jlaska> I'll go make the change if no one objects
15:58:07 <Viking-Ice> yeah but we should set a filter rule and reject the message
15:58:34 <jlaska> Viking-Ice: so you want to convert to plaintext, *and* reject ?
15:58:35 <Viking-Ice> referring to the mailing list guidelines
15:59:02 <Viking-Ice> jlaska: yup enforce and educate at the same time
15:59:06 <adamw> i don't see the point in converting if we're going to reject it
15:59:06 <Cerlyn> Adding the rejection seems kind of silly; many mail clients do not make it obvious how to send plain text on a per-recipient basis
15:59:14 <jlaska> yeah, that seems odd
15:59:29 <Viking-Ice> yeah my mistake
15:59:50 <Viking-Ice> not up to speed in mailman filters though you had to enable then filter for some wierd reason
16:00:10 <jlaska> +1 on the convert_to_plaintext
16:00:39 <adamw> +1
16:00:40 <Viking-Ice> by simply converting you dont educate user(s) at the same time
16:00:56 <jlaska> Viking-Ice: it was your proposal ... do you want to change it?
16:01:19 <Viking-Ice> yes rather reject but fall on converting if that fails ;)
16:01:30 <jlaska> would you rather automatically reject mails with HTML?
16:01:36 <Viking-Ice> as in fails to be agreed
16:01:57 <Viking-Ice> jlaska: yup filter rule reject referring to mailing list guidelines
16:02:19 <jlaska> I'll need to look into that with mailman, and I'll let you know
16:02:29 <jlaska> I'm sure it's possible, I just haven't done it
16:02:33 <Viking-Ice> enforcing and educate instead of just work around
16:03:06 <jlaska> if I can customize the rejection response to point to the guildeines, I have no objections
16:03:29 <Viking-Ice> great
16:03:29 <jlaska> any objections to auto-reject w/ friendly response?
16:03:55 <adamw> sure, whatever
16:03:59 <tflink> not sure that rejection is the best idea if we're trying to get new pre-release users to send email to test@
16:03:59 <jlaska> #agreed Update mailman settings for test@ to auto-reject HTML messages, and reply with list guidelines
16:04:07 <jlaska> #undo
16:04:07 <zodbot> Removing item from minutes: <MeetBot.items.Agreed object at 0x34137150>
16:04:11 * tflink is slow, aparrantly
16:04:29 <Viking-Ice> tflink: they are violating the guidelines
16:04:44 <Viking-Ice> they need to be educated
16:04:57 <Viking-Ice> instead of us working around the problem
16:05:03 <tflink> is rejecting someone's first email to a list the best way to educate them?
16:05:15 <jlaska> I'm going to enable the plaintext conversion for now
16:05:28 <Viking-Ice> tflink: and refering to the mailing list guidelines yes
16:05:31 <kparal> let's convert to plaintext and accept, but send nagmail back asking for no html next time. is that possible?
16:05:31 <jlaska> and we'll continue this discussion on #fedora-qa or on the list (without using HTML messages)
16:05:54 <jlaska> #agreed Update mailman settings to enable convert_html_to_plaintext
16:06:11 <Viking-Ice> ?
16:06:23 <jlaska> is this worth a lot of time?
16:06:34 <jlaska> meaning, do we really have a ton of test@ mails containing HTML?
16:06:51 <adamw> let's just agree and move on
16:06:56 <jlaska> let's quick fix with the conversion, and politely educate users when it happens
16:07:18 <jlaska> anything else for today?
16:07:24 <Viking-Ice> right support their broken behaviour
16:07:27 <jlaska> we're over time ... so if not, I'll #endmeeting shortly
16:07:35 <tflink> one small thing
16:07:39 <tflink> sorry, I just remembered
16:07:54 <tflink> regarding a pending website change
16:08:00 <jlaska> Viking-Ice: it's about approach/attitude ... your idea is sound, we just don't want to tell them to piss off on their first post
16:08:14 <jlaska> #topic Pending website change
16:08:17 <jlaska> tflink: take it
16:08:31 <Viking-Ice> jlaska: depends on how you phrase the reply
16:08:54 <jlaska> Viking-Ice: agreed, and when the phrasing is a mailman rejection ... that stucks
16:08:57 <jlaska> s/stucks/sucks/
16:09:11 <tflink> after a conversation with a new user on friday, I sent this email out to the test@ list about new user feedback http://lists.fedoraproject.org/pipermail/test/2011-March/098160.html
16:09:36 <tflink> A ticket was filed with fedora-websites to add a notice that all pre-release feedback should be directed to #fedora-qa or test@
16:09:44 <adamw> yay
16:09:46 <tflink> https://fedorahosted.org/fedora-websites/ticket/45
16:09:56 <jlaska> hmm, I wonder if that explains the rise in chatter
16:10:08 <tflink> that request has been accepted and the change is scheduled for beta release
16:10:20 <jlaska> cool, good to know this workflow also
16:10:31 <jlaska> tflink: thx for follow-up
16:10:47 <adamw> nice one tflink
16:11:10 <tflink> jlaska: np
16:11:24 <jlaska> #info A ticket was filed with fedora-websites to add a notice that all pre-release feedback should be directed to #fedora-qa or test@
16:11:44 <jlaska> #topic Open Discussion - last call
16:11:54 * jlaska sets the fuse at 1 minute
16:12:09 <adamw> oh
16:12:10 <adamw> one more
16:12:15 <adamw> #topic nightly URLs
16:12:16 <jlaska> darnit!
16:12:17 <jlaska> ;)
16:12:22 <jlaska> #chair adamw
16:12:22 <zodbot> Current chairs: adamw jlaska kparal
16:12:24 <adamw> #topic nightly URLs
16:12:25 <adamw> thanks
16:12:31 <jlaska> take it my good man
16:12:34 <adamw> so as we noticed earlier - nightly generation is being done by koji now
16:12:36 <adamw> so all the URLs changed
16:12:41 <jlaska> oh, good reminder
16:12:55 <adamw> the old URLs were pretty widely distributed; it'd be nice if we could get any URL under that whole tree to redirect somewhere sane
16:13:04 <jlaska> #info nightly generation is being done by koji now, so the URLs changed
16:13:05 * kparal is not happy when X restarts suddenly
16:13:06 <adamw> right now you just get a 404, which doesn't give you any clue where to look
16:13:09 <jlaska> kparal: :(
16:13:11 <adamw> nirik: any input?
16:13:33 <jlaska> does the Release/Branched wiki need an update
16:13:34 * jlaska checks
16:13:56 * nirik looks up
16:14:07 <nirik> I guess I could see about a redirect...
16:14:40 <nirik> or I could make the old dirs and put links back to the top in or something.
16:15:10 <jlaska> for the logs ... what's the new location?
16:15:11 <kparal> is it possible to have always current link to last desktop image?
16:15:16 <jlaska> koji.fedoraproject.org ...
16:15:27 <tflink> http://alt.fedoraproject.org/pub/alt/nightly-composes/current.html
16:15:29 <adamw> http://alt.fedoraproject.org/pub/alt/nightly-composes/current.html
16:15:32 <adamw> doh, i lose
16:15:36 <kparal> but direct one
16:15:40 <adamw> tflink: you realize, of course, this means war
16:15:50 <adamw> yeah, it'd be nice to have a reliable direct link like we had before
16:15:52 * tflink is worried about what he started :-/
16:16:11 <tflink> kparal: I don't think that there is a static link to the current nightly anymore
16:16:17 <tflink> since they're all koji builds
16:16:25 * nirik isn't sure what you are asking...
16:16:33 <nirik> a link to just the desktop composes?
16:16:39 <adamw> well, actually, current.html is about as good as what we had before
16:16:46 <kparal> not really
16:16:47 <adamw> before you could reliably get to the right directory but not direct to an iso
16:16:55 <jlaska> should current.html just be index.html?
16:16:56 <adamw> it's still two clicks
16:17:06 <kparal> yes, but if you link to current.html, some people might be confused. too much text there
16:17:11 <nirik> jlaska: I suppose it could be.
16:17:15 <nirik> can add a link there easily.
16:17:47 <adamw> nirik: and have any other URL under /nightly-composes just redirect there
16:17:50 <kparal> and btw finding iso on koji build page is not as easy as it used to be too
16:18:21 <adamw> it's not really hard, though. i don't wanna make a mountain out of a molehill, here...
16:18:31 <kparal> alright
16:18:35 <nirik> I'm happy to try and improve things...
16:18:39 <jlaska> nirik: I have some code that can push all this stuff into a wiki page if you want it there instead
16:18:41 <nirik> but this is much nicer IMHO than the old setup
16:19:05 <jlaska> yeah, takes the burden off of you for maintaining the build environment
16:19:06 <nirik> having composes in ~30min instead of 10hours and having 2 weeks of old ones is nice.
16:19:14 <jlaska> oh, that too! :)
16:19:27 <jlaska> #link http://alt.fedoraproject.org/pub/alt/nightly-composes/current.html
16:19:35 <jlaska> adamw: anything else on this topic?
16:19:46 <nirik> well, currently it's still a script I have to run... I'd like to get it automated so I don't have to do anything, but figured it was far enough to put in place.
16:20:11 <adamw> nirik: just cleaning up the 404s would be enough for me
16:20:23 <nirik> which 404's?
16:20:30 <nirik> oh, right.
16:20:37 <nirik> will see what I can do
16:21:18 <adamw> nirik: basically as long as when you land in the nightly-composes tree you can find where you were trying to go, i'm happy
16:22:15 <adamw> i think that's all
16:22:17 <jlaska> #action nirik will look into updating the nightly-composes directory to eliminate any 404's
16:22:30 <jlaska> #topic Open discussion - last last last call
16:22:36 * jlaska sets fuse at 1 minute
16:23:25 <jlaska> thanks all!   I'll send minutes to the list later on
16:23:32 <jlaska> Happy testing of TC1 when it lands this week :)
16:23:37 <jlaska> #endmeeting