20070726
Logs
TZ UTC +1 thanks PriceChild for the log
15:54:35 * agoliveira waves all
15:55:26 <cjwatson> good afternoon
15:55:54 * asac waves
15:56:00 <cjwatson> mdz: are you around?
15:56:40 <evand> hello
15:56:49 <bryce> heya
15:57:00 <heno> hey bryce
15:57:16 <pitti> hey hey
15:57:18 <dendrobates> hey all
15:57:26 <cjwatson> calc, doko, Riddell, mvo_, Mithrandir, soren: ping
15:57:35 <doko> pong
15:57:35 <Riddell> hi cjwatson
15:57:42 <doko> ohh crap, my report ...
15:57:43 <cjwatson> are kees or mathias around today?
15:58:02 <calc> pong
15:58:02 <cjwatson> mathiaz, rather
15:58:44 <dendrobates> soren is around. I have not heard from mathiaz or kees today.
15:59:10 <pitti> I talked to kees some minutes ago
15:59:11 <agoliveira> cjwatson: I'm not sure but I think he's flying back with Chris Kenyon.
15:59:19 <seb128> kees was talking on #ubuntu-devel 10 minutes ago
15:59:33 <keescook> my clock must be behind. :)
16:00:07 <mvo_> hello cjwatson
16:00:11 <tkamppeter> hi
16:00:22 <cjwatson> agoliveira: ok
16:00:30 <cjwatson> heno: I guess it's you and me then
16:00:38 <heno> yep
16:00:48 <soren> I'm here now.
16:00:56 <soren> I just got a -ENOINTERNET from Mathiaz.
16:01:06 <cjwatson> ok
16:01:16 <cjwatson> we have a relatively full agenda today, so let's move right along
16:01:27 <soren> I haven't seen the agenda..
16:01:36 <cjwatson> apparently I missed several apologies, which is to be expected just after Ubuntu Live
16:01:39 <bdmurray> here
16:01:40 <cjwatson> https://wiki.ubuntu.com/DevelTeamMeeting20070726
16:01:54 <cjwatson> no new starters this week
16:02:07 <cjwatson> any extra items for the agenda
16:02:08 <cjwatson> ?
16:02:13 * agoliveira is amazed by that!
16:02:49 * heno waves to bdmurray
16:03:26 <cjwatson> agoliveira: by which?
16:03:49 <agoliveira> cjwatson: "no new starters this week" :)
16:04:01 * Hobbsee waves
16:04:12 <cjwatson> I'll take that as no extra items
16:04:16 <soren> mathiaz: You made it!
16:04:23 <mathiaz> soren: yop :)
16:04:24 <cjwatson> mathiaz: welcome
16:04:24 <cjwatson> #
16:04:24 <cjwatson> (pitti) What's the schedule for a new kernel upload? We need to get fixed dailies to test #126964 soon, just in case the /usr/bin/id (& friends) file corruption is yet another bug.
16:04:29 <cjwatson> #
16:04:32 <cjwatson> BenC: you had a plan for this, please share :-)
16:04:48 <pitti> my primary concern about the fs corruption has been preemptively fixed by pkl
16:04:49 * pitti hugs pkl_
16:04:53 <cjwatson> pitti: also pkl's activity report is pretty relevant
16:04:54 <cjwatson> indeed
16:04:54 <BenC> Got an upload comming tomorrow
16:05:09 <pitti> but it would still be interesting, since we didn't manage to get it into T3, and there was a kernel (almost) ready apparently
16:05:09 <BenC> pkl_: are your fixed in lum now?
16:05:15 <BenC> if so, we can probably do a build of that now
16:05:26 <Meyvn> hey people
16:05:30 <pkl_> BenC: I'll push the fixes immediately after the meeting.
16:05:35 <pitti> BenC: I'm not implying particular urgency (any more), just curious about the timing
16:05:39 <BenC> ok, I'll do an upload after that
16:05:55 <BenC> pitti: would be nice to have a working daily without waiting for tribe-4
16:05:56 <pitti> BenC: just to avoid having the same awkward situation as with t3
16:06:04 <cjwatson> pitti: as I think was mentioned earlier, from here on in, the plan is to schedule kernel releases for the Friday before a milestone
16:06:05 <pitti> BenC: I agree
16:06:21 <cjwatson> which means they should all have built by Monday, and we have a few days for emergency patch-up sessions
16:06:24 <pitti> cjwatson: that's still pretty tight to sort out major problems
16:06:42 <pitti> but if that works best for the kernel team, I'm fine with it
16:06:58 <cjwatson> I don't want to squash the kernel cycle up too tight
16:07:06 <cjwatson> (and I had real problems not typing "squashfs" there)
16:07:09 <pitti> maybe s/for the Friday/by Friday/
16:07:37 <BenC> pitti: we could do uploads my time Thur night, so it's building on everyone's Friday morning
16:07:42 <BenC> but generally it will be on Friday
16:08:01 <cjwatson> TBH I think it's better to suck up that buildd time over the weekend when it's less contended
16:08:24 <pitti> ok, good for me
16:08:24 <Hobbsee> cjwatson: then agian, if it happens to be in european night, relatively few uploads are being made anyway...
16:08:39 <BenC> pitti: and with new deps in linux-meta, we can now do a full upload of everything at once, and not wait for kernel build to be done
16:08:42 <Hobbsee> seeing as most of the team is on european, or close to european timezone.
16:08:47 <cjwatson> Hobbsee: Ben's night is early European morning
16:08:57 <cjwatson> which would mean a kernel build lasting through most of the European day
16:09:03 <pitti> BenC: that's great, decoupling archive admin steps from upload
16:09:22 <Hobbsee> cjwatson: erk. i'm messing my people in countries again. apologies
16:09:23 <BenC> pitti: right, hoping that will alleviate some of the hand-holding for the process
16:09:39 <cjwatson> we need to be careful not to accept ABI-changing kernel binaries until all the architectures have built
16:09:57 <pitti> ack
16:09:58 <cjwatson> Ben doesn't want the subsequent stages to happen until that's done, for internal ABI-management reasons
16:10:38 <pitti> makes sense
16:10:43 <cjwatson> BenC: seconded on having working dailies, will help with live CD installer wok
16:10:46 <cjwatson> work
16:10:49 <cjwatson> ok, next
16:10:49 <cjwatson> #
16:10:51 <cjwatson> (pitti) Please feel urged to have a look at the dapper point release TODO:
16:10:54 <cjwatson> *
16:10:57 <cjwatson> [WWW] https://launchpad.net/ubuntu/+milestone/ubuntu-6.06.2 We need to get these SRUs in the next two weeks to release it on time.
16:11:00 <cjwatson> #
16:11:16 <pitti> oh, did I write that as agenda? was meant as a general ping in the report
16:11:38 * BenC acks kernel work
16:11:42 <pitti> well, yeah, I already annoyed some of you recently, and I'm afraid I have to do more status inquiries
16:11:51 <BenC> after tomorrow's upload, I'll be shifting to dapper point work
16:12:17 <heno> I've looked at the highest priority 2.6.15 bugs for that list, should we dig deeper in that list?
16:12:29 <cjwatson> that's quite a few bugs to cope with over the next couple of weeks
16:12:34 <cjwatson> pitti: I think so
16:12:35 <heno> (iow, has someone else / can someone else have a look too?)
16:12:42 <pitti> right, I wouldn't extend it much further
16:12:47 <Hobbsee> particularly with a tribe over the next couple of weeks too...
16:12:58 <pitti> heno: unless you stumble upon something that's relatively easy and would truly improve dapper considerably
16:13:05 <asac> will we do something like a dapper.2 beta?
16:13:15 <mvo_> pitti: this means a lot of sru-verification for the next 2 weeks?
16:13:19 <cjwatson> asac: I expect we'll have daily builds for a while leading up to .2
16:13:22 <pitti> asac: test candidates
16:13:46 <pitti> mvo_: yes, probably; a lot of hw specific issues, though, so we'll need the support team and DC guys, too for that
16:13:51 <heno> pitti: ok
16:13:55 <pitti> since the focus is on making it work on new server hw
16:13:57 <cjwatson> for those who weren't in the small meeting in London about this, the current plan is only to update the alternate/server images
16:14:05 <pitti> mvo_: so we'll have to change the verification procedure a bit
16:14:05 <heno> I agree that the list is already long
16:14:17 <cjwatson> for the reason pitti cites
16:14:28 <mvo_> pitti: ok
16:14:31 <cjwatson> if you install dapper desktop now, you aren't getting very much longer support lifetime than if you install edgy anyway
16:14:40 <pitti> we currently have two or three important desktop fixes, but they'll reach dapper-updates, too, and can be done independently
16:14:44 <cjwatson> er, feisty
16:15:23 <cjwatson> this will keep the workload down, and make it more feasible for the point release to happen without affecting gutsy development too seriously
16:16:13 <Hobbsee> you're aiming for .2 at the end of august, presumably?
16:16:20 <keescook> btw, is there an "easy" way for someone with LTS to determine if a package is part of the "server" seed?
16:16:21 <cjwatson> we'll need to figure out how to handle d-i uploads to dapper; I have a feeling we might need two, in order that we can test the hw-detect change before having new kernel
16:16:24 <pitti> Hobbsee: current goal (set by mdz) is Aug 25
16:16:25 <cjwatson> Hobbsee: thereabouts, yes
16:16:42 <cjwatson> keescook: "easy"? dunno, but http://people.ubuntu.com/~ubuntu-archive/germinate-output/ubuntu-dapper/server
16:16:42 <pitti> keescook: look at the server CD package .list file?
16:16:47 <cjwatson> or what pitti said
16:16:59 <cjwatson> (server also includes server-ship though)
16:17:09 <pitti> cjwatson: can we build test CDs with -proposed?
16:17:10 <keescook> pitti, cjwatson: yeah, that's the closest I've seen.
16:17:13 <Hobbsee> brilliant. directly when tribe 5 is due. cd testing might get interesting, then.
16:17:14 <cjwatson> I meant http://people.ubuntu.com/~ubuntu-archive/germinate-output/dapper/server
16:17:25 <cjwatson> Hobbsee: yes, it's always going to collide with something
16:17:32 <cjwatson> not doing so is essentially impossible
16:17:37 <cjwatson> pitti: yes
16:18:12 <cjwatson> case $DIST in
16:18:12 <cjwatson> dapper)
16:18:12 <cjwatson> #export PROPOSED=1
16:18:12 <cjwatson> ;;
16:18:13 <pitti> BenC: oh, did you already see a change which will require an abi bump in all cases?
16:18:14 <cjwatson> esac
16:18:17 <cjwatson> just uncomment that ...
16:18:23 <Hobbsee> cjwatson: that's a point, when you get that late in the release cycle
16:18:29 <calc> btw is gutsy+1 (lts) planned for 8.04 or 8.06 ?
16:18:36 <pitti> cjwatson: ok, let's discuss the technical bits out of band
16:18:42 <BenC> pitti: for dapper or gutsy?
16:18:46 <pitti> BenC: dapper
16:18:47 <BenC> for dapper, not yet
16:19:05 <cjwatson> calc: I think currently 8.04, though delaying isn't being ruled out; I don't particularly like the idea of a delay personally unless we have a really strong reason for it
16:19:11 <pitti> BenC: I remember one fix which required an abi change, but it was only a 'bonus' type of thing; it would be utterly cool to avoid it, of course :)
16:19:14 <cjwatson> I think dapper's delay had quite a negative effect on edgy
16:19:34 <pitti> BenC: so far many bugs were actually quite simple, like adding modules to the initramfs and such
16:19:35 <calc> cjwatson: ok, i'm going to be off on a cruise sometime next spring/summer so need to determine when it is safe to plan that ;)
16:19:43 <cjwatson> calc: noted
16:19:57 <BenC> pitti: yeah, that's what I've gotten too...couple updated drivers and one or two new ones
16:20:12 * calc doesn't want to be gone at an important time
16:20:37 <pitti> ok, I'm done with that meeting topic
16:21:06 <cjwatson> rest -> #ubuntu-devel or #ubuntu-kernel or whatever's appropriate, and if you have items assigned to you please look at them forthwith
16:21:09 <cjwatson> (tkamppeter) Decide on which printer setup tool should be default in Gutsy (gnome-cups-manager or system-config-printer).
16:21:21 <pitti> ah
16:21:24 <cjwatson> pitti: do you think your visual objections to s-c-p are substantially difficult to solve?
16:21:33 <pitti> so, I quickly tried s-c-p today
16:21:48 <cjwatson> I must admit that tkamppeter makes a compelling argument
16:21:52 <pitti> it has a significantly harder UI than g-c-m, but as Till says, it's at least maintained upstream
16:22:03 <pitti> it still has a lot of warts, but nothing unfixable, of course
16:22:26 <tkamppeter> And very important: It fixes some of my and pitti s specs from Mountain View:
16:22:27 <pitti> apart from the serious bugs, my main concern is that it is much harder to install a printer with current s-c-p than with g-c-m
16:22:55 <pitti> with g-c-m, it's "new printer", "ack the one from autodetect list", "press ok on the details page"
16:23:18 <Mithrandir> cjwatson: delaying> I think the current plan is to stretch the release for hunky a little bit, but not so much that itchy gets as painful as edgy.
16:23:21 <pitti> with s-c-p, you don't see the list of detected printers at first and have to invent a name, location, description, etc., and the detected list is very confusing
16:23:37 <cjwatson> Mithrandir: please don't tell me those are the actual names :P
16:23:49 <pitti> that could be rectified with fully automatic printer installation, of course
16:24:01 <Mithrandir> cjwatson: no idea, but I would be surprised if they were. :-P
16:24:11 <pitti> tkamppeter: I didn't see how to call s-c-p for auto-configuration, but if that's possible in general, let's talk about that in #u-d later
16:24:20 * agoliveira would *not* be surprised...
16:24:29 <calc> hunky he-man ;)
16:24:42 <pitti> Mithrandir: of course it'll be "Harry Hogwarts" :-P
16:24:58 <Mithrandir> pitti: I don't even know if we'll be skipping H because of 5.04
16:25:22 <cjwatson> I don't think we ought to switch to s-c-p straight away if it has known UI regressions that make it harder to install printers; however it is in main already and we still have a couple of weeks before feature freeze
16:25:33 <tkamppeter> pitti, hal-cups-utils has a script which auto-configures the printers, it only has a little bug that HAL does not actually call the script when plugging the printer. We need a HAL expert to look into that.
16:25:44 <cjwatson> tkamppeter: do you think you can tackle the most urgent concerns as a prerequisite for switching to s-c-p by default?
16:25:45 <pitti> tkamppeter: I can have a look at that
16:27:01 <tkamppeter> cjwatson, I can look into some, like filtering the list of auto-detected printers somewhat and taking care that HPLIP is always used if the printer supports it
16:27:02 <pitti> (examples: enabling browsing or sharing messes up cupsd.conf and doesn't restart cups, s-c-p only runs as root without an apparent reason, detected printer list is not filtered for human consumption)
16:27:04 <cjwatson> assuming we switch, I recommend that at beta time (when Canonical folks other than the distro team are liable to upgrade) we get non-technical folks in the London office to set up the printer there using s-c-p
16:27:32 * boredand1logging is now known as boredandblogging
16:27:40 <pitti> and the UI is still quite complex in general, it's nothing for novices
16:27:59 <pitti> lots of good features, of course
16:28:02 <tkamppeter> pitti, for handling the server config options correctly something on your non-root mode of CUPS needs to be improved. Web interface has the same problem.
16:28:29 <pitti> tkamppeter: right, cupsd doesn't know how to reload its configuration without dropping the tcp port binding
16:28:49 <pitti> tkamppeter: a mere /etc/init.d/cupsys force-reload works fine
16:29:26 <tkamppeter> pitt, if you could improve the CUPS patches, we could perhaps fix both s-c-p and web interface.
16:29:26 <pitti> so, I didn't see anything that isn't fixable, as I said, but someone's got to do it
16:29:50 <pitti> tkamppeter: I doubt it's easy to fix that properly; a complete restart is much easier
16:30:12 <tkamppeter> pitti, how does g-c-m fix the problem?
16:30:43 <pitti> tkamppeter: /etc/init.d/cupsys force-reload
16:31:04 <tkamppeter> This works as non-root?
16:31:07 <pitti> so, just to keep the meeting short, I propose that some of us install s-c-p and play around with it
16:31:20 <pitti> ACTION: pitti to file the worst s-c-p bugs in LP
16:31:21 <cjwatson> who volunteers?
16:31:34 <tkamppeter> Then we could perhaps make a simple patch on s-c-p to do the same thing.
16:31:34 <pitti> o/ (for testing and bug filing)
16:31:37 <cjwatson> (one of these days I must plug in our printer)
16:31:37 <seb128> I can test on deskjet printers (local and windows share)
16:31:39 <pitti> tkamppeter: no, as root
16:32:05 <seb128> o/ also testing and bug filing
16:32:19 <pitti> tkamppeter: currently s-c-p runs as root anyway (it shouldn't need to just to configure printers, that's another issue)
16:32:37 <tkamppeter> pitti, this one could probably also do in s-c-p, with a SUID helper or asking for the password to do a "sudo" call.
16:32:43 <heno> perhaps post a request to the gutsy forum section
16:32:47 <heno> ?
16:32:48 <pitti> tkamppeter: gksu, yes
16:32:59 <cjwatson> heno: I think that's the second step, after the distro team
16:33:09 <cjwatson> but yes, that should be done
16:33:09 <heno> (or dev link section if that's active at all)
16:33:18 <pitti> it's the default frontend for xubuntu, so maybe there are even some bugs already
16:33:45 <tkamppeter> pitti, I am runninh s-c-p always as normal (privileged) user and I can set up queues but not configure the server (as I also can run lpadmin).
16:33:51 <cjwatson> tkamppeter: could you post any necessary instructions to ubuntu-devel@ and call for testers?
16:34:00 <pitti> tkamppeter: I cannot connect to the server as normal user
16:34:07 <cjwatson> including any particular items that are known, and areas that need attention
16:34:09 <pitti> tkamppeter: (and yes, I'm in lpadmin, too)
16:34:45 <soren> pitti: What exactly can't you do without root? Does it not start at all?
16:34:56 <tkamppeter> pitti, once you need to update to the newest version, I did a small patch
16:35:05 <pitti> soren: it starts, but it doesn't display any printers or my localhost server, and manually connecting to it fails
16:35:07 <cjwatson> ACTION: tkamppeter to post s-c-p call for testing and instructions to ubuntu-devel@
16:35:12 <cjwatson> let's move on, as time is getting short
16:35:13 <cjwatson> (tkamppeter) Should we stay with CUPS 1.2.x in Gutsy or advance to 1.3.x (Red Hat has already switched over in Rawhide)?
16:35:15 <soren> pitti: Ah, ok.
16:35:22 <pitti> tkamppeter: I'm up to date in gutsy
16:35:41 <cjwatson> tkamppeter: do you have a summary of changes? (preferably a link rather than all pasted in here)
16:35:46 <tkamppeter> second, I can only add, modifiy and delete queues, not any changes on cupsd.conf
16:35:51 <pitti> based on my experience with migrating to 1.2 from 1.1 late in the process, I'd recommend keeping 1.2 at this stage
16:36:19 <cjwatson> http://www.cups.org/articles.php?L479
16:36:27 <pitti> the early 1.2.x broke a lot of stuff, and we played catchup with SRUs a lot in dapper
16:36:39 <tkamppeter> pitti, after the meeting I will help you to set up s-c-p.
16:37:02 <pitti> tkamppeter: later, preferably, I gotta leave immediately after the meeting (dinner invitation); tomorrow?
16:37:10 <pitti> tkamppeter: what would we get with 1.3?
16:37:18 <tkamppeter> cjwatson, so perhaps we keep 1.2 then and give 1.3 half a year to mature.
16:37:19 <pitti> in terms of Great Features?
16:37:50 <pitti> tkamppeter: it also depends on what we gain with it, of course
16:37:52 <tkamppeter> pitti, lets move it out to tomorrow.
16:37:54 <seb128> maybe better to push a new version this cycle than for lts?
16:38:20 <pitti> mdns support is cool, of course
16:38:37 <cjwatson> pitti: but apparently avahi support isn't quite there yet
16:38:43 <pitti> tkamppeter: ^ that obsoletes the cups specific browsing protocl, I take it?
16:38:54 <cjwatson> see the comments at the bottom of the article I posted above
16:39:07 <tkamppeter> http://www.cups.org/documentation.php/whatsnew.html
16:39:19 <pitti> ldap and kerberos are not so terribly urgent, since main ubuntu doesn't support it OOTB very well
16:39:19 <tkamppeter> This is a summary of the new features.
16:40:08 <tkamppeter> I think avahi is to find network printers, less to connect CUPS servers, or not?
16:40:45 <pitti> tkamppeter: right
16:41:09 <pitti> tkamppeter: avahi tells clients where to find the server, and the correct ipp:// address, and those clients can then use that address in their local cups spooler for printing to the remote queue
16:41:42 <pitti> tkamppeter: very roughly, it has the same task than the cups browsing protocol
16:42:42 <cjwatson> is the peer credentials stuff used by s-c-p?
16:43:27 <pitti> ah, that looks like replacing the cert hack in /var/run/cups/certs/
16:43:48 <tkamppeter> cjwatson, I do not know, as I did not try rawhide.
16:44:01 <pitti> tkamppeter: is that cert -> peer cred change exposed at all? this should be completely opaqued by libcups
16:44:38 <tkamppeter> The new cupsctl command is also interesting, it makes it much easier to write printer setup tools with server config support.
16:45:05 <pitti> tkamppeter: is that something s-c-p supports?
16:45:05 <tkamppeter> pitti, I do not know about the architecture of this one.
16:45:14 <cjwatson> so, I think the answer to your question is that there doesn't seem to be anything compelling that *requires* an upload, but some potentially useful features, and the main concern is stability; ultimately, since we're not at upstream version freeze yet, I think it's between you and your sponsor
16:45:16 <pitti> --w-r----- 1 cupsys lpadmin 32 2007-07-26 15:30 /var/run/cups/certs/0
16:45:33 <pitti> in 1.2, everyone who can read this (i. e. lpadmins) are allowed to access cups without password
16:45:34 <tkamppeter> I do not know, I tested s-c-p only on Gutsy with CUPS 1.2..
16:45:55 <pitti> which is pretty much same effect like unix socket peer credential check for lpadmin group
16:46:27 <tkamppeter> Side-Channel API is also interesting to make drivers with -bi-di support and better error reporting.
16:47:08 <pitti> so, if you guys want, I can try to build an 1.3 test package and see how much work it is to port our patches over
16:47:28 <pitti> very few have been accepted by upstream, but we have to carry most of them ourselves unfortunately (even bug fixes)
16:47:37 <cjwatson> pitti: seems like something Till should prepare, and you should review
16:47:47 <pitti> WFM
16:48:23 <pitti> for a mere test package we don't even need to port the derooting patches, that'll be much easier
16:48:26 <pitti> tkamppeter: ^
16:48:34 <cjwatson> ACTION: tkamppeter to prepare test packages of cupsys 1.3, pitti to review
16:48:36 <pitti> for a production package I would be very sad to see them go, though
16:48:45 <pitti> cupsys offers an awful lot of attack vectors
16:49:02 <asac> pitti: does upstream refuse to use our patches? or didn't we submit them?
16:49:05 <tkamppeter> I think carrying over the patches is better done by pitti, as he has created the non-root mode and so knows better about for what the patches are good ofr.
16:49:07 <cjwatson> I don't think it's worth reviewing an incomplete patchset
16:49:22 <cjwatson> tkamppeter: it's subject to pitti having time
16:49:22 <pitti> asac: I submitted a lot, but upstream keeps insisting that cupsd needs to run as root
16:49:44 <pitti> asac: none of his arguments were compelling, though
16:49:47 <asac> pitti: maybe there are valid arguments? why would they insist otherwise?
16:50:01 <pitti> tkamppeter: as I said, for a test package we can probably do without the derootign
16:50:31 <pitti> asac: they presented a list of five-ish reasons, and I explained them why each of those doesn't need root *shrug*
16:50:46 <asac> pitti: did discussion just stopped then?
16:50:48 <cjwatson> so Till can port the patches other than derooting, and Martin can finish it off?
16:50:52 <cjwatson> I guess that's ok ..
16:50:56 <asac> pitti: maybe its worth reviving that?
16:51:01 <pitti> of course the derooting patch is not perfect, but as long as they don't even support the general approach, I'm not very encouraged to clean them up
16:51:05 <tkamppeter> pitti, so I will make test packages without non-root.
16:51:10 <pitti> asac: stopped> pretty much, yes
16:51:48 <cjwatson> let's take the rest to -devel, as we still have a reviewers meeting to squeeze in here
16:51:55 <cjwatson> any other business for today's meeting?
16:51:57 * jsgotangco (n=JSG@ubuntu/member/jsgotangco) has joined #ubuntu-meeting
16:52:07 <iwj> I'd just like to mention consistent-login-screen.
16:52:28 <iwj> We (seb128, Amaranth and I) had some discussion about it earlier and it looks like we might need to change our plans.
16:52:34 <iwj> (to put it mildly)
16:52:48 <pitti> iwj: oh, that's why my gdm screen is now blue, to match KDE? :-)
16:52:52 <iwj> No.
16:52:53 <cjwatson> iwj: can you give a quick summary?
16:52:54 <kwwii> please don't tell me we need to use the stupid splash screen again
16:53:07 <iwj> I'm going to write up the situation for a decision.
16:53:08 <bryce> heh, I wondered about that
16:53:24 <iwj> Summary is upstream are doing / have done something quite different already in the last few weeks.
16:53:35 <pitti> bryce: ♥ gnome flower :-)
16:53:37 <seb128> cjwatson: basically FC7 fixed most of the usuability issues using consolekit and the code is to upstream GNOME now
16:53:40 <iwj> So (a) if you would like to receive this mail let me know now.
16:53:43 <cjwatson> from what I heard earlier it seemed to be dependent on gdm3?
16:53:47 <cjwatson> iwj: o/
16:53:47 <iwj> Or (b) say it should be on ubuntu-devel.
16:54:02 * pitti votes for (b)
16:54:03 <iwj> cjwatson: No, but the gdm2 patches I have don't work in my tests.
16:54:04 <cjwatson> ubuntu-devel seems entirely reasonable ..
16:54:10 <iwj> OK, (b) it is.
16:54:13 <iwj> Expect a mail.
16:54:26 <cjwatson> ok, gdm3 was the bit that scared me
16:54:40 <cjwatson> (most)
16:54:46 <seb128> cjwatson: that's planned for next cycle
16:55:04 <kwwii> if we need a splash screen again someone let me know as I am not on that list
16:55:13 <cjwatson> kwwii: this is, AIUI, nothing to do with splash screens
16:55:20 <kwwii> ahhh, cool
16:55:24 <cjwatson> kwwii: you should be on ubuntu-devel surely - it's moderated so relatively quiet
16:55:46 * jsgotangco has quit (Client Quit)
16:55:48 <kwwii> hm
16:55:52 <iwj> cjwatson is correct; the splash screen is a different bug/situation.
16:56:23 <cjwatson> anything else before we close?
16:56:42 <Hobbsee> only that there are too many bugs. *drowns in more bugs*
16:57:17 <iwj> So do you approve of my automated tester ? :-)
16:57:23 * bryce pimps brian's bug plots - http://people.ubuntu.com/~brian/testing_graphs/
16:57:31 <iwj> It seems to have churned through main already. I wonder if I should set it going on universe.
16:57:32 <heno> Hobbsee: please welcome Pedro our new triager next week :)
16:57:40 <Hobbsee> heno: oh nice!
16:57:50 <pitti> cjwatson: can I be excused now? I have a RL appointment now
16:57:59 <iwj> We're done now surely.
16:58:11 <cjwatson> pitti: yes
16:58:12 <heno> lovely per-package plots :)
16:58:18 <pitti> thanks to everyone
16:58:29 <cjwatson> I think we're drifting, so yes, we seem to be done
16:58:38 <cjwatson> anyone prepared to run a reviewers meeting? dholbach isn't here
16:58:47 <Hobbsee> bryce: nice!
16:59:04 <cjwatson> if not, let me pimp http://daniel.holba.ch/sponsoring/
16:59:13 <cjwatson> dholbach will clean up the UI later
16:59:23 <seb128> cjwatson: what is the reviewers meeting about?
16:59:36 <cjwatson> seb128: standing item to deal with unassigned code review items
16:59:37 <asac> seb128: ack?
16:59:51 <seb128> asac: ack what?
17:00:08 <asac> confirmed your question
17:00:21 <seb128> ah ;)
17:01:54 <cjwatson> unassigned review items seem to be 96586 (inkscape), 95852 (adept), 63506 (adduser), 62396 (manpages), 128137 (espeak), 128028 (mysql-dfsg-5.0), 127273 (laptop-mode-tools), 125384 (linux-source-2.6.22), 119821 (dh-make)
17:02:04 <cjwatson> volunteers?
17:02:15 <cjwatson> I may have missed some where patch submitters have assigned themselves
17:02:46 <Hobbsee> cjwatson: Riddell should grab adept, but i thought there were a couple more patches on launchpad
17:03:00 <asac> cjwatson: wasn't the idea that we assign them to people we think are capable to do the review?
17:03:07 <seb128> I would say bryce for the inkscape one since he's upstream ;)
17:03:10 <cjwatson> asac: yes, but volunteers are always better first
17:03:18 <cjwatson> seb128: indeed, but he's not in core-dev yet ...
17:03:29 <keescook> I can take 96586
17:03:38 <cjwatson> I was just about to suggest you. :)
17:03:45 <bryce> I reviewed the inkscape one yesterday
17:03:49 <seb128> I'm looking at 62396
17:03:52 <cjwatson> I'll assign Riddell for adept
17:04:01 <Riddell> yes, there's a bunch of adept patches I hope to look at today
17:04:53 <cjwatson> I'll do the trivial sync ones
17:05:08 <cjwatson> (I extirpated my sponsorship guilt by dealing with grub today)
17:05:35 <keescook> I can take 119821 too; but I'm unclear if the XSBC stuff is right.
17:05:59 <keescook> if you're dh_make'ing for ubuntu, shouldn't the person doing it be the list (@u.c) maintainer?
17:06:07 <bryce> keescook: for 96586, the backporter team bounced it back to the sru team before they'll do a backport. All the information is there to do either sru or backport, it just needs done.
17:06:10 <iwj> cjwatson: I'll take adduser (63506).
17:06:22 <keescook> bryce: but the patch looks complete?
17:07:08 * kwwii moves on to the next meeting....thanks everyone
17:07:17 <Hobbsee> bug 96586
<ubotu> Launchpad bug 96586 in inkscape "Correct multiple inkscape issues in Feisty (SRU and Backport)" [Low,Fix released] https://launchpad.net/bugs/96586
17:07:23 <bryce> yes
17:07:35 <iwj> I keep finding these rather trivial manpage bugs.
17:07:35 <keescook> bryce: cool; I'll start shoving it.
17:08:00 * beuno has quit ("Ex-Chat")
17:08:03 <cjwatson> iwj: thanks. yes, I'm not sure how much effort they're worth ...
17:08:06 <Hobbsee> oh darn, wrong bug
17:08:12 <Hobbsee> bug 119821
<ubotu> Launchpad bug 119821 in dh-make "dh_make does not follow blueprint" [Undecided,Incomplete] https://launchpad.net/bugs/119821
17:08:24 <iwj> cjwatson: Yes. Ideally I'd just write a Debian bug report and let it follow through in the normal course of events.
17:08:28 <cjwatson> (and grr, it DEFINITELY should be hash not pound)
17:08:34 <iwj> (Not sure where upstream for adduser is.)
17:08:37 <cjwatson> damn americanisms
17:08:47 <cjwatson> iwj: Debian I think
17:08:52 <seb128> iwj: I'm going to send it to Debian now
17:09:02 <iwj> seb128: If you've got the mail already written by all means.
17:09:19 <iwj> I'm happy to do it though.
17:09:31 <cjwatson> I'll take espeak, Luke's good so that should be straightforward
17:09:37 <seb128> iwj: yes
17:09:46 <Hobbsee> keescook: assuming that the package is going to universe first, that XSBC stuff should be fine
17:09:52 <iwj> Yes you've got the mail written I take it. OK.
17:10:05 <keescook> Hobbsee: okidoky
17:10:12 <Hobbsee> keescook: and it clearly wont go to main without a MIR anyway, so can be changed then. or can be changed if the user overrides it, knowing that it's going to go to main, etc.
17:10:14 <cjwatson> BenC: can you look after that kernel bug (125384)? it's not appropriate for the normal sponsorship process
17:10:49 <cjwatson> (or ping it back to lirc, it's not immediately clear to me where it belongs)
17:10:54 <keescook> Hobbsee: I was thinking that if a new package was being made, the uploader would be the maintainer, which is allowed in the maintainer blueprint.
17:11:08 <keescook> but, generally, I guess the diff is correct
17:11:14 <BenC> cjwatson: ok
17:11:17 <cjwatson> thanks
17:11:24 <Hobbsee> keescook: this is true, but they are trying to get to group maintainership.
17:11:48 <cjwatson> that leaves manpages, laptop-mode-tools, dh-make
17:12:14 <BenC> cjwatson: IMO, it can be closed because a) it's lirc's fault, and b) we have lirc modules in lum now
17:12:15 <keescook> cjwatson: I'm snagging dh-make (discussed above)
17:12:40 <iwj> dh-make has dholbach's name on the bottom as a comment.
17:12:42 <cjwatson> keescook: oh yes, thanks
17:12:55 <cjwatson> manpages should have a Debian bug filed rather than deviating; I'll take care of that
17:13:07 <cjwatson> BenC: thanks, please do whatever's appropriate then
17:13:15 <BenC> cjwatson: done
17:13:16 <seb128> iwj: manpages bug with the diff sent to the BTS
17:13:20 <iwj> cjwatson: ^
17:13:23 <cjwatson> aha
17:13:45 <cjwatson> I recall Matthew Garrett saying recently that laptop-mode wasn't really particularly desirable, so am not sure what to do with that bug
17:13:54 <cjwatson> mjg59: could you look at bug 127273, please?
<ubotu> Launchpad bug 127273 in laptop-mode-tools "laptop-mode init script links not created" [Undecided,Confirmed] https://launchpad.net/bugs/127273
17:14:14 <cjwatson> I think that's all then
17:14:18 <keescook> bryce: odd, is 128551 sane?
17:14:43 <Hobbsee> you can always do the universe sponsorships too, if you wish
17:14:52 <cjwatson> Hobbsee: not especially :)
17:15:03 <Hobbsee> cjwatson: awww...was worth a try...
17:15:18 * Hobbsee should get the Long Pointy Stick of DOOM!!!!!!!!!!!!!!! ™ out, and try again
17:16:42 <bryce> keescook: hmm, I don't seem to have permission to look at that one
17:17:28 <cjwatson> bryce: you do now
17:18:00 <keescook> cjwatson: thanks. bryce: the mentioned CVE was fixed quite a long time ago.
17:18:00 <bryce> aha thanks
17:18:13 <keescook> (http://www.ubuntu.com/usn/usn-448-1)
17:18:14 <cjwatson> it's clearly not appropriate for a sync in any case
17:18:35 <bryce> right
17:18:48 <cjwatson> the changelog version is ancient
17:18:49 <keescook> agreed, I just wanted to get other eyes on it in case I was crazy. :)
17:19:47 <cjwatson> anyway, I think we're done with this meeting
17:19:49 <cjwatson> thanks allMeetingLogs/UbuntuDev/20070726 (last edited 2008-08-06 16:38:39 by localhost)