[01:30] <ubotu> New bug: #147892 in malone "Broken links in CVE pages (to bug's c-a-p / distrotask pages)" [Undecided,New]  https://launchpad.net/bugs/147892
[01:56] <ubotu> New bug: #147898 in launchpad "Import release information from freshmeat" [Undecided,New]  https://launchpad.net/bugs/147898
[03:06] <mpt> Hobbsee, to continue our discussion from yesterday, can you give an example of a team page where you expect to see the team's contact e-mail address but don't?
[03:07] <Hobbsee> mpt: http://launchpad.net/~ubuntumembers
[03:07] <Hobbsee> mpt: but that may be because they dont have a team contact address
[03:08] <jamesh> not all teams have contact emails ...
[03:08] <mpt> Hobbsee, probably
[03:08] <mpt> Can you see the e-mail address on <http://launchpad.net/~launchpad>?
[03:09] <Hobbsee> jamesh: indeed.  i was trying to figure yesterday if it didnt show, because they didnt have a team contact email, or if LP had decided to remove it from everywhere, for some unknown reason
[03:09] <Hobbsee> jamesh: i came to the conclusion that the former was true
[03:09] <jamesh> Hobbsee: I guess this is one of the problems with omitting empty portlets
[03:09] <Hobbsee> mpt: yep
[03:10] <jamesh> although things would look very crowded if we displayed all possible portlets everywhere
[03:10] <Hobbsee> mpt: for some reason, i couldnt find a team that actually *had* a contact email, to check with
[03:10] <Hobbsee> jamesh: indeed.  i think you're usually fine, as most people dont expect you to remove features at random
[03:10] <Hobbsee> but i have recently seen that bug portal go missing for some unknown reason, so... :)

[03:11] <jamesh> I am sure we'd have a good reason for removing a feature you find useful :)
[03:11] <mpt> Probably the e-mail address should go in the contact details table
[03:12] <mpt> except that teams don't have one of those, hmm
[03:12] <Hobbsee> jamesh: havent found otu what it is yet :)
[03:12] <Hobbsee> mpt: i dont think people will look for it, to be honest.
[03:12] <Hobbsee> it was only that i was looking to blackhole a team, so wsa looking for other teams as an eg
[03:12] <mpt> blackhole a team?
[03:12] <jamesh> probably not worth working on this til Barry's work is done
[03:13] <Hobbsee> mpt: sorry, blackhole the address for a team.
[03:13] <Hobbsee> not blackhole the entire team :)
[03:19] <Hobbsee> mpt: (so that when the next nutter decides to assign a bug to kubuntu-members, not everyone gets spammed)
[03:20] <Hobbsee> ubuntu-core-dev is also a good one for that.
[04:00] <pwnguin> if update manager doesn't display changelogs to a ppa, is that a bug in the ppa system or in update manager?
[04:01] <Hobbsee> pwnguin: it's a bigger bug than that.  it'll be a bug in both libept and ppa.  but ppa first.
[04:02] <Hobbsee> pwnguin: as in, it may well be by design
[04:02] <Hobbsee> well, eventually by design
[04:02] <pwnguin> by design as in "not a bug" or as in "we'd have to redesign to fix it"?
[04:03] <Hobbsee> the latter.
[04:03] <Hobbsee> well, it begs teh question - do we only want to show the changelogs from ubuntu?
[04:04] <Hobbsee> does everyone *want* their changelogs plastered on something like changelogs.ubuntu.com?
[04:04] <lifeless> wherelse would a changelog come from ?
[04:04] <lifeless> oh, meh, missed a word
[04:04] <Hobbsee> what happens if two people upload the same package, with different changelogs?
[04:04] <lifeless> its a design bug
[04:04] <Hobbsee> lifeless: :)
[04:04] <lifeless> changelogs are centralised
[04:05] <pwnguin> so bassically, changelogs aren't in the repos
[04:05] <Hobbsee> pwnguin: correct.
[04:05] <pwnguin> unfortunate
[04:05] <lifeless> not an lp design bug, ubuntu design bug
[04:05] <lifeless> if that gets addressed, with a hint from package metadata in the packages list
[04:05] <Hobbsee> lifeless: debian design bug, you mean.
[04:05] <pwnguin> I wonder if wig and pen addresses that
[04:06] <lifeless> Hobbsee: we haven't fixed it, so its an ubuntu bug :)
[04:06] <lifeless> w&p is orthogonal
[04:06] <Hobbsee> lifeless: seeing as debian does the same, afaik...
[04:06] <Hobbsee> lifeless: oh right.  i meant introduced from debian.
[04:06] <lifeless> this is fixable by adding a field
[08:00] <spiv> lifeless: I'll start the meeting
[08:01] <spiv> BjornT, thumper, jamesh, jtv: reviewer's meeting time.
[08:01] <thumper> here
[08:01] <BjornT> here
[08:01] <thumper> spiv: and jml?
[08:01] <jamesh> here
[08:01] <spiv> jml phoned with apologies.
[08:01] <thumper> ok
[08:01] <jtv> here
[08:01] <spiv> Apparently the building he was working in is surrounded by fire trucks and is "a little bit on fire".
[08:02] <jtv> And that stops him?
[08:02] <thumper> he had wireless didn't he?
[08:02] <spiv> = Agenda =
[08:02] <spiv>   * Roll call
[08:02] <spiv>   * Next meeting
[08:02] <spiv>   * Action items
[08:02] <spiv>   * Queue status
[08:02] <spiv>   * Mentoring update
[08:03] <spiv> = Roll call =
[08:03] <Hobbsee> meh.  whoever let a bit of fire interrupt with meetings!
[08:03] <spiv> We've basically done that :)
[08:03] <spiv> I'm sure jml will share the gory details when he gets back online.
[08:03] <spiv> = Next meeting =
[08:04] <thumper> AU goes into summertime on Sun
[08:04] <thumper> NZ is there now
[08:04] <spiv> Next week, same bat-time, same bat-channel?
[08:04] <spiv> Ah, good point.
[08:04] <jtv> That's same UTC bat-time, I take it?
[08:04] <thumper> shift it by an hour?
[08:04] <spiv> +/- an hour doesn't bother me at all.
[08:04] <spiv> As it's still inside my normal workday.
[08:04] <thumper> it's7pm NZ time
[08:04] <thumper> which hits kids bed time
[08:05] <thumper> how does it impact you BjornT to be an hour earlier?
[08:05] <BjornT> for me, the same UTC time will already be at 8am, so i'd rather not having it earlier
[08:05] <thumper> keep it 0600UTC for now then
[08:06] <thumper> and I'll bitch if it becomes a problem
[08:06] <jamesh> I don't mind either way
[08:06] <spiv> thumper: sounds like a plan!
[08:06] <spiv> So, same time next week.
[08:07] <spiv> = Action items =
[08:07] <spiv> I failed to get the minutes from the last meeting sent out.
[08:07] <spiv> Let me check the log quickly to see if there were any action items
[08:08] <spiv> Nothing that I can see.
[08:08] <spiv> = Queue status =
[08:08] <lifeless> I'm +1 on shifting it
[08:09] <lifeless> (most back, but I may disappear violently)
[08:09] <spiv> Long, and lots overdue.
[08:09] <spiv> I just did my overdue review just before the meeting, but there's 14 others.
[08:09] <thumper> the oldest is actually needs-reply
[08:10] <thumper> but jamesh hasn't updated it
[08:10] <spiv> Ah, that's good news.
[08:10] <spiv> So, just 13 then ;)
[08:11] <spiv> We seem to almost always have a lot of branches overdue for reviews.
[08:11] <spiv> I'm not sure what the cause is, or how to fix it, but that seems like a serious problem to me.
[08:11] <thumper> spiv: start a penalty jar
[08:11] <thumper> $1 per day per review
[08:11] <thumper> and use it to buy drinks at sprints
[08:12] <spiv> I was going to suggest donating it to a charity for helping chronic procrastinators ;)
[08:13] <thumper> I was going to do mine today, honest
[08:13] <spiv> It seems to be mostly non-.au reviewers that are slacking right at the moment.
[08:14] <spiv> Or at least, reviewers that aren't present at this meeting.
[08:14] <lifeless> yay fingerpointing
[08:14] <lifeless> :)
[08:14] <spiv> Yeah.
[08:15] <spiv> Well, normally we nag tardy reviewers at these meetings with moderate success.  We can't do that if they aren't here :)
[08:15] <spiv> I'll send an email nag to the list.
[08:15] <spiv> = Mentoring update =
[08:16] <spiv> Any news from mentors or mentees?
[08:16] <spiv> I haven't had anything to mentor recently, but jml has a couple of overdue branches so presumably that will change shortly...
[08:17] <jtv> Is there an official "needs mentoring" state for branches?
[08:17] <jamesh> I've got to followup on jtv's reviews (I haven't been particularly good on that)
[08:17] <spiv> Heh.
[08:18] <spiv> That could be useful.
[08:18] <spiv> Although merge-conditional could serve that role.
[08:18] <jtv> It'd help a lot, I think, if "needs mentoring" showed up separately on the status overview page.
[08:18] <jtv> spiv: I beg to differ
[08:18] <jtv> merge-conditional means approval is conditional only on things the coder has to do
[08:19] <jamesh> spiv: well, merge-conditional's current meaning is that the developer can merge without any further permission from reviewers
[08:19] <spiv> Right, and the coder needs to wait for the mentor's review ;0
[08:19] <spiv> Anyway, I agree, it's a very poor substitute.
[08:20] <jtv> So, new state needs-mentoring?
[08:21] <spiv> I don't mind if we add a new state for that, although I don't have to implement it :)
[08:21] <spiv> Perhaps we should propose it for the .eu meeting?
[08:21] <spiv> Assuming we're all in favour, which we seem to be.
[08:21] <thumper> +1
[08:21] <lifeless> +1
[08:22] <jamesh> sure.
[08:22] <BjornT> +1
[08:22] <spiv> Ok, I'll put it on the wiki page.
[08:22] <spiv> = Any other business? =
[08:23] <spiv> tick tick tick...
[08:23] <spiv> 5
[08:23] <spiv> 4
[08:23] <spiv> 3
[08:23] <spiv> 2
[08:23] <spiv> 1
[08:23] <spiv> MEETING ENDS
[08:23] <spiv> Thanks everyone.
[08:23] <thumper> yay, dinner
[09:04] <carlos> morning
[09:49] <_polto_> hello all
[09:50] <_polto_>  i'm trying to upload a modified version of mplayer, but it lack all dependencies... how can i include dependencies to compile on PPA ?
[09:59] <stdin> _polto_: what do you mean? you need to specify all the build dependencies in the debian/control file
[10:00] <_polto_> stdin, i apt-get source mplayer, applied a patch and uploaded the sources
[10:00] <_polto_> debian/control did not changed :)
[10:01] <_polto_> (I'm new to PPA)
[10:01] <_polto_> never used before.
[10:01] <stdin> got the link build log?
[10:02] <_polto_> https://edge.launchpad.net/~polto/+archive/+builds?build_text=&build_state=all
[10:03] <_polto_> http://launchpadlibrarian.net/9631434/buildlog_ubuntu-feisty-i386.mplayer_2%3A1.0%7Eppa3_MANUALDEPWAIT.txt.gz
[10:05] <stdin> _polto_: seems (for some reason) it's been uploaded to the "main" section, it needs to be in multiverse, make sure in debian/control it has "Section: multiverse/graphics" 
[10:05] <_polto_> oups !
[10:05] <_polto_> :)
[10:05] <_polto_> thanks!
[10:05] <_polto_> stupid error .. :)
[10:06] <stdin> you aren't be the first to do that (and won't be the last) :)
[10:08] <_polto_> :)
[10:11] <_polto_> stdin, should i only switch from graphics to multiverse/graphics in debian/control? what's all ?
[10:11] <stdin> that's the only place it's set, so yeah
[10:13] <_polto_> :)
[10:13] <_polto_> thanks
[10:38] <geser> _polto_: don't forget to bump the version else the upload will get rejected
[10:39] <_polto_> i did :) thanks
[11:26] <ubotu> New bug: #148011 in launchpad-bazaar "doesn't propagate changed tags to http?" [Undecided,New]  https://launchpad.net/bugs/148011
[02:41] <ubotu> New bug: #148082 in launchpad "A lot of bogus 'ubuntu' projects registered" [Undecided,New]  https://launchpad.net/bugs/148082
[02:46] <ubotu> New bug: #148087 in launchpad-bazaar ""bzr break-lock bzr+ssh://bazaar.launchpad.net/..." fails to break a lock" [Undecided,New]  https://launchpad.net/bugs/148087
[02:49] <Kopfgeldjaeger> hi
[03:07] <Kopfgeldjaeger> i submitted a package to PPA yesterday, and it was accepted. its sources are already in ppa, but nothing is built yet
[03:29] <Kopfgeldjaeger> cant PPA use multiverse? 
[03:30] <Kopfgeldjaeger> ah, ok, i see
[03:53] <Kopfgeldjaeger> um... launchpad ppa rejection mail: "If you don't understand why your files were rejected, or if the override file requires editing, reply to this email." and in the mail header: "Replay-To: no_reply@launchpad.net" :d
[04:42] <sabdfl> cprov: ^^ I think that should be "Reply-To" ;-)
[04:42] <sabdfl> err
[04:43] <sabdfl> i think that whole thing should be better ;-)
[04:43] <cprov> sabdfl: ouch ...
[04:44] <bigjools> coulda sworn I fixed all that!
[04:48] <bigjools> cprov: it does not say that for a PPA rejection any more
[04:49] <bigjools> cprov: see ppa-upload-rejection.txt
[04:49] <cprov> bigjools: yes, it seems to be fixed in RF. Do you remember which revision it was done ?
[04:50] <bigjools> no, but it was released in 1.1.9
[04:52] <cprov> Kopfgeldjaeger:  are you sure you received it recently ?
[05:18] <Kopfgeldjaeger> cprov: some minutes ago
[05:19] <cprov> Kopfgeldjaeger: let me check
[05:19] <Kopfgeldjaeger> shall i make a screenshot?
[05:20] <bigjools> Kopfgeldjaeger: what is the rejection reason in the email?
[05:20] <Kopfgeldjaeger> unable to find distroseries: hardy ;)
[05:20] <bigjools> kwality :)
[05:20] <Kopfgeldjaeger> and... "Unhandled exception processing upload: 'NoneType' object has no attribute 'getPublishedReleases'". but i dont know if it has to do with hardy
[05:21] <bigjools> yes that extra message is a bug that is being fixed right now
[05:21] <Kopfgeldjaeger> ok. http://img9.myimg.de/replyToNoReplye3f6f.png ("Antwort an" ^= "Reply to"
[05:23] <Kopfgeldjaeger> do i have to specify "Component: multiverse" in debian/control to get sth. into a multiverse-ppa? 
[05:24] <bigjools> cprov: if it can't determine it's really a PPA upload yet, it will send a "normal" upload email
[05:26] <cprov> bigjools: I'm not sure what is happening, I just got a rejection that says "If you don't understand why your files were rejected please send an email to launchpad-users@lists.canonical.com for help."
[05:26] <bigjools> cprov: I wonder if he uploaded to a non ~user dir?
[05:26] <Kopfgeldjaeger> i did.
[05:27] <bigjools> Kopfgeldjaeger: you uploaded to ~you ?
[05:27] <Kopfgeldjaeger> bigjools: 
[05:27] <Kopfgeldjaeger> incoming = ~nicolai-spohrer/ubuntu
[05:27] <bigjools> ok
[05:29] <superm1> cprov, is there a particular reason that amd64 builds on the PPAs are queuing up about 10-40 times faster than i386 builds?
[05:30] <cprov> superm1: don't you mean "building" faster ?
[05:30] <superm1> cprov, no literally i queue up a source package, and it gets onto the amd64 builder (promethium) within 30 minutes
[05:31] <superm1> its been taking a few hours for it to hit i386 builders
[05:33] <cprov> superm1: ok, let's use better concepts, 1 queued (when soyuz recognize there is something to build, NEEDSBUILD build record), 2. dispatched (when a queued build records reaches a builder) and 3 built (when a builder delivers the binary packages for a given build record).
[05:33] <cprov> superm1: and you meant, amd64 dispatches 10-40 times faster than i386, right ?
[05:33] <superm1> okay using those terms, its been appearing that they are being dispatched later to i386
[05:33] <superm1> yeah
[05:33] <superm1> perhaps did i386's builder go down and now its just behind amd64?
[05:34] <cprov> superm1: that because amd64 queue is 10-40 times smaller than i386 (arch-indep)
[05:34] <superm1> ah
[05:34] <superm1> that would be it
[05:34] <superm1> wouldn't it make sense to spread arch-indep stuff among the two then rather than all of that go to i386?
[05:36] <iwj> BjornT: I'd appreciate it if you could take a look at bug 147754 and let me know what you think.  Scott suggested I ask you :-).  Thanks.
[05:36] <ubotu> Launchpad bug 147754 in malone "searching including duplicates uses wrong status" [Undecided,New]  https://launchpad.net/bugs/147754
[05:37] <cprov> superm1: lp-buildd is not able to build arch-indep in other archs than i386, yet.
[05:37] <superm1> ah i see
[05:37] <superm1> thanks cprov, makes much more sense :)
[05:38] <cprov> superm1: you're welcome (this discussion could be one item of our FAQ)
[05:46] <BjornT> iwj: i agree that bug 147754 should be fixed; it's basically that we don't handle duplicates as good as we should. i don't know when we will have time to fix it, though.
[05:46] <ubotu> Launchpad bug 147754 in malone "searching including duplicates uses wrong status" [Undecided,New]  https://launchpad.net/bugs/147754
[05:57] <iwj> BjornT: Can you have it fixed in 1.1.10 ?  Without this fix my workflow for autopkgtest is going to get worse and worse.
[05:57] <iwj> Also it means that autopkgtest will sometimes fail to test a package when it should.
[06:01] <gmb> WTF? Amarok has gone insane. No matter what I put in the playlist it won't stop playing LugRadio!
[06:01] <gmb> I'm being haunted by Jono Bacon.
[06:03] <BjornT> iwj: sorry, it won't be possible to fix it for 1.1.10, it's a bit too late for that, especially since i suspect the fix will require changes to the db schema.
[06:04] <BjornT> iwj: can you explain your workflow for autopkgtest? maybe there's some workaround
[06:07] <iwj> BjornT: I look at the list of bugs which it has filed, because that's a worklist of packages that need fixing.  I want to include duplicates because autopkgtest bugs are often duplicates of other bugs.
[06:07] <iwj> (or vice versa)
[06:07] <iwj> Obviously I don't want to include closed bugs.
[06:09] <Kopfgeldjaeger> how to get a package into MULTIVERSE in ppa? i tried "Section: multiverse/blub (or just multiverse)" or "Component: multiverse", but it always goes into main
[06:09] <iwj> BjornT: Likewise autopkgtest uses the same list to construct a set of packages which (a) should be de-preferred when choosing the next package to test and (b) to avoid filing duplicates of bugs it has filed already.
[06:10] <iwj> BjornT: As I mentioned on lp-users, I'm screenscraping for this information.  IWBNI there were a better way.
[06:15] <stdin> Kopfgeldjaeger: "Section: multiverse/whatever" works
[06:15] <Kopfgeldjaeger> stdin: what can whatever be?
[06:16] <stdin> umm, lets see if I can find a link...
[06:16] <stdin> I know there kde, gnome, x11, base, libs, devel and libdevel
[06:17] <pochu> Kopfgeldjaeger: restricted and multiverse are not allowed, afaik
[06:17] <pochu> So, you can't.
[06:17] <Kopfgeldjaeger> :/, that is _not_ good for me :( avidemux depends on liblame, for example...
[06:18] <stdin> it is allowed as long as the package you're distributing is under a free licence
[06:18] <stdin> (or so I understand)
[06:19] <Kopfgeldjaeger> it's gpl
[06:19] <BjornT> iwj: hmm. i can't think of a simple fix that would help you. the earliest we could fix this would be 1.1.12, since 1.1.10 is already too full, and 1.1.11 will be a short cycle.
[06:20] <iwj> BjornT: Surely the simple fix is to, when searching, use the status of the duplicate target for duplicate bugs to match against the search criteria, rather than the status of the duplicate itself ?
[06:20] <iwj> Is that difficult ?
[06:20] <stdin> Kopfgeldjaeger: http://www.debian.org/doc/maint-guide/ch-dreq.en.html#s-control has a couple other sections too. for most GUI apps x11 is OK, there's also multimedia  (guessing from your dependencies that'll be a good fit)
[06:20] <Kopfgeldjaeger> ill try, thanks
[06:21] <stdin> Kopfgeldjaeger: #ubuntu-motu is a great place to ask about some generic debian packaging stuff if you need help (and we all do ;)
[06:21] <Kopfgeldjaeger> stdin: /whois Kopfgeldjaeger ;)
[06:21] <stdin> so I don't whois everyone I see  :p
[06:22] <Kopfgeldjaeger> stdin: but i thought it would be the wrong place for launchpad specific questions 
[06:22] <laga> pochu: do you know where it is said that multiverse is not allowed?
[06:22] <stdin> Kopfgeldjaeger: it is, that's why I said for "generic debian packaging" questions
[06:23] <elmo> laga: https://help.launchpad.net/PPATermsofUse
[06:23] <BjornT> iwj: i don't even want to think about how that sql query would look like ;) well, i could take a quick look at it tomorrow, to see if can get something to work, that would solve your use case, and wouldn't cause any timeouts.
[06:23] <iwj> Thanks.
[06:25] <iwj> BjornT: Alternatively, since the interface between the autopkgtest machinery and the screenscraper is tiny, perhaps someone with database read access could run a suitable query out of cron for me.
[06:25] <stdin> but an app can be open-source (GPL for instance) but build-dep on a package in multiverse, so that doesn't forbid multiverse. just packages that aren't under a free license
[06:26] <laga> elmo: doesn't mention multiverse there
[06:26] <Odd_Bloke> stdin: If something build-deps on something in multiverse, it has to be in multiverse.
[06:26] <elmo> laga: err
[06:26] <stdin> Odd_Bloke: that's my point
[06:28] <Odd_Bloke> stdin: Are you saying that PPAs should be able to build multiverse packages?
[06:28] <laga> well, PPAs can already build multiverse packages. however, i was just told here that i'm not allowed to do that.
[06:29] <stdin> I'm saying that PPAs TOS doesn't forbid multiverse, as long as the package itself is under an open license
[06:29] <stdin> (or i'm saying that's how I understand it)
[06:32] <pochu> laga: from https://help.launchpad.net/PPAQuickStart : 'if your package is targeted to main your dependencies will be satisfied from main, whereas if your package is targeted to universe your dependencies will be satisfied from main and universe.'
[06:32] <pochu> laga: it doesn't say you cant upload build against multiverse, but it doesn't say you can either.
[06:32] <BjornT> iwj: yes, that might be possible. kiko would probably be a good person to talk about that.
[06:32] <Kopfgeldjaeger> stdin:      -> Component: main Section: graphics again... in my control file, there is: "Section: multiverse/multimedia"
[06:32] <kiko> mmmm
[06:33] <laga> pochu: os there's n o definite answer, but it looks like liblame etc is OK.
[06:33] <kiko> iwj, BjornT: what does the script need to do?
[06:33] <Kopfgeldjaeger> stdin: um... does "Section" have to be in the source package section?
[06:34] <stdin> Kopfgeldjaeger: as far as I know, -motu is a better place to ask to get a definite answer
[06:35] <pochu> laga: maybe cprov-lunch or Rinchen know.
[06:36] <pochu> cprov-lunch, Rinchen: ^ do you know whether it's possible to upload a package which has build-dependencies on multiverse? (the package itself is redistributable, under a GPL license).
[06:36] <Rinchen> cprov-lunch would be best for that (or kiko).  My knowledge is limited to the qsg on this unfortunately.
[06:37] <iwj> kiko: find all bugs filed by autopkgtest which are still open, and produce a list of (source package name, bug number).  Where `open' means in one of the usual open states _but_ if the bug is a duplicate, if the duplicate target is open.
[06:37] <kiko> iwj, did bdmurray get his DB access?
[06:38] <iwj> Err, I don't know.  I haven't spoken to bdmurray about this AFAIR.
[06:39] <kiko> iwj, talk to him first; I requested access and he might be able to help you first
[06:45] <cprov> pochu: according to the ogre-model, packaging in multiverse can build-dep on main + restricted + universe + multiverse. So, yes, upload it to the multiverse component.
[06:46] <iwj> kiko: Apparently he hasn't yet.
[06:46] <laga> cprov: will that comply with the ToS? eg if a package depends on liblame0?
[06:48] <kiko> iwj, unfortunate. let's try and make that happen
[06:49] <cprov> laga: IMHO, yes, you can link with liblame0 in ppa.  But I might be wrong :(
[06:49] <iwj> kiko: Thanks.
[06:51] <laga> cprov: thanks :)
[07:01] <Kopfgeldjaeger> stdin: working now... Section has to be in the source section
[07:02] <stdin> good :)
[07:35] <giga> Hi All
[07:35] <giga> Thanks from Georgia for your perfect Distro
[07:55] <ubotu> New bug: #148195 in soyuz "Sanitize Build sampledata" [High,Triaged]  https://launchpad.net/bugs/148195
[08:09] <Kopfgeldjaeger> what does "failed to upload" mean in PPA? https://edge.launchpad.net/~nicolai-spohrer/+archive/+build/401486
[08:17] <kiko> Kopfgeldjaeger, you get email about it, but it normally means that the control file in your generated binaries has an inconsistency.
[08:31] <ScottK> Kopfgeldjaeger: The one time it happened to me (on a real upload, not to a PPA), it was a Soyuz bug and had to be manually fixed, but the next upload of the same package with only very minor modifications went through fine.
[08:31] <Kopfgeldjaeger> ok
[08:33] <Kopfgeldjaeger> Verifying timestamps in avidemux-foobar_amd64.deb \n No signer, therefore ACL not processed (...) \n Upload was rejected
[08:34] <Kopfgeldjaeger> ah, ok. i see the error
[08:35] <Kopfgeldjaeger> control file failure
[08:35] <Kopfgeldjaeger> is there a list of possible multiverse/(.*) sections?
[08:39] <Kopfgeldjaeger> ok, http://packages.ubuntu.com/gutsy/ should do it
[09:11] <ubotu> New bug: #148226 in launchpad-bazaar "the branch listing should be sortable and searchable" [Undecided,New]  https://launchpad.net/bugs/148226
[09:26] <kiko> mwh, ddaa: I'm getting a bzr error when pushing to the supermirror.
[09:26] <kiko> AssertionError: end of file reading from server.
[09:26] <kiko> oh, permission denied.
[09:26] <ddaa> my first impulse would be to blame bzr
[09:26] <mwh> sounds like you're not ssh keyed up
[09:27] <kiko> yep
[09:27] <kiko> what a horrible error message!
[09:36] <jerenkrantz> hola.  any launchpad members around?  apache's subversion server is being hammered by launchpad (for qpid) - https://launchpad.net/qpid
[09:37] <mwh> ah
[09:37] <mwh> yes
[09:38] <jerenkrantz> (i'm not familiar enough with how launchpad operates.)
[09:39] <jerenkrantz> it's just that canonical is toping our abusers list.  =(
[09:39] <jerenkrantz> our next step is to add an ip block...
[09:40] <mwh> jerenkrantz: it's importing qpid into a bazaar branch
[09:40] <mwh> jerenkrantz: i can kill the job if that would be preferable
[09:40] <ddaa> sorry about the abuse, we know that our svn code is very inefficient
[09:40] <jerenkrantz> how is it doing the import?
[09:41] <jerenkrantz> (we're not opposed to doing ya'll doing imports, but it's getting absurd.)
[09:42] <ddaa> jerenkrantz: I believe the problem you are referring to is https://bugs.edge.launchpad.net/launchpad-cscvs/+bug/120992
[09:42] <ubotu> Launchpad bug 120992 in launchpad-cscvs "use persistent svn_ra sessions and reconnect on failure" [High,Confirmed]  
[09:43] <ddaa> our svn import code uses the pysvn library (instead of upstream libsvn bindings) so we use a new svn session for every single operation
[09:43] <jerenkrantz> sounds about right.
[09:43] <ddaa> such as directory listing, or cat
[09:43] <jerenkrantz> anyway to throttle it back?
[09:43] <ddaa> not really
[09:43] <ddaa> but if you can be bothered to give us svn dumps now and then
[09:44] <jerenkrantz> just gonna say
[09:44] <ddaa> we can do the initial imports from the dump
[09:44] <jerenkrantz> what can we give you to make you go away?  =)
[09:45] <ddaa> a svn dump of http://svn.apache.org/repos/asf
[09:45] <jerenkrantz> like 30GB.  =P
[09:45] <ddaa> we need the whole dump so the revision numbers are correct :/
[09:46] <ddaa> bandwidth is not an issue for us
[09:46] <jerenkrantz> i can dump a partial path and include blanks for non-qpid paths
[09:46] <ddaa> jerenkrantz: that might just do it
[09:47] <ddaa> jerenkrantz: we'll probably need to contact you again in the future though
[09:47] <jerenkrantz> ok.  need any other projects besides qpid?
[09:47] <jerenkrantz> sure.  when you need apache projects, email infrastructure@apache.org
[09:48] <jerenkrantz> (it'll take me a few days to find the cycles to do the dump/filter, but if you can cancel the imports until then...=P)
[09:49] <ddaa> mwh: I'll kill it
[09:50] <ubotu> New bug: #148240 in launchpad "file new (Report a) bug for Ubuntu ambiguous error message with package name" [Undecided,New]  https://launchpad.net/bugs/148240
[09:50] <mwh> ddaa: thanks
[09:50] <Kopfgeldjaeger> my first ppa-package was just built successfully :D
[09:51] <izm99> how should I suggest a bug be closed?  I reported it, but I don't see the option anywhere...
[09:52] <ddaa> jerenkrantz: the job has been killed
[09:52] <joes2> ddaa: thx
[09:52] <jerenkrantz> ddaa: thx.
[09:52] <ddaa> jerenkrantz: BTW, congratz for your network monitoring
[09:52] <jerenkrantz> ddaa: when i get the dump, how do you want me to get it to you?
[09:52] <jerenkrantz> ddaa: thx joes2.  =)
[09:52] <ddaa> we've never had someone complain so quickly
[09:52] <jerenkrantz> just find you here on irc?
[09:53] <jerenkrantz> fwiw, apache.org has some serial abusers, so we have scripts to monitor it
[09:54] <ddaa> jerenkrantz: upload the dump to some public webspace and send an email to feedback@bazaar-importer.canonical.com 
[09:54] <jerenkrantz> will do.
[09:55] <ddaa> btw, do not forget to bzip it
[09:55] <carlos> good night!
[09:55] <jerenkrantz> heh.  =)
[09:55] <ddaa> svn dump format is incredibly verbose...
[09:55] <jerenkrantz> if you use --deltas in the dump, it's actually pretty compressed
[09:55] <jerenkrantz> but that's not the default
[09:56] <jerenkrantz> (fwiw, i'm a core svn dev.  =P)
[09:57] <ddaa> mwh: can you think a good place to record that we should ask infrastructure@apache.org to get dumps for imports from apache.org?
[09:57] <ddaa> mostly, I want to record the email address somewhere where we'll find it easily
[09:57] <mwh> not really
[09:57] <mwh> "our heads"
[09:58] <ddaa> My head is not good at recording this sort of data.
[09:58] <jerenkrantz> anyhoo, thx for being responsive.  will get you a dump in a few days.  l8rz.
[09:58] <ddaa> jerenkrantz: you're welcome, nice to check with us first.
[09:59] <jerenkrantz> (fwiw, we also informally hang out in #asfinfra; but infrastructure@apache.org is our principal contact.)
[09:59] <jerenkrantz> take care!
[10:03] <LaserJock> kiko: thanks for the blog post
[10:20] <kiko> LaserJock, you're welcome! more to come.
[10:27] <ScottK> kiko: Are you reading responses on ubuntu-devel or just on launchpad-users?
[10:28] <kiko> ScottK, I'm reading responses to me!
[10:29] <kiko> I don't read mailing lists (not really :-P)
[10:29] <ScottK> OK.  Generally if you are going to send mail to a mail list, I'd expect you to read the responses on the list.
[10:30] <ScottK> I've just forwarded you the responses I sent to the list.
[10:36] <kiko> ScottK, if you read my reply to seb, you'll see I catered for the larger problem of "Incomplete to what?" there
[10:37] <ScottK> kiko: True.  My reply was written before his.
[10:37] <kiko> ScottK, I'm sorry for not subscribing to ubuntu-devel, but i'm wary of all the email I already get
[10:37] <ScottK> kiko: If you aren't going to read replies to the list, then I think you ought to tell people that in advance.
[10:38] <kiko> ScottK, I was hoping people would comment on the blog itself.. I even asked there
[10:39] <ScottK> Blog commenting takes more time that hitting reply to list.  I don't even know if I can comment (registration, etc.)
[10:39] <kiko> but yeah.
[10:39] <kiko> ScottK, for the record, I /hate/ Triaged
[10:39] <kiko> I think it's a bong idea
[10:39] <ScottK> Agreed, although I dislike Triaged less than auto-expiring.
[10:40] <kiko> we're going to get into a fight now!!! :)
[10:40] <radix> kiko: you asked for upstreams not using Triaged. Landscape doesn't.
[10:40] <ScottK> I think the idea is proken by design.
[10:40] <radix> (for one)
[10:40] <kiko> radix, Launchpad doesn't either (except by mistake)
[10:41] <radix> heh
[10:41] <kiko> ScottK, it takes some getting used to, that's for sure! 
[10:42] <radix> kiko: we also use "Fix Committed" to mean "in trunk", not "Fix Implemented Somewhere But Not Yet Merged" (like your post suggested)
[10:42] <radix> (and then it's moved to Fixed Released once we actually deploy or release it)
[10:42] <ScottK> kiko: The implementation aside, even the concept of what was planned to be implemented is just flat wrong.
[10:42] <radix> i.e., it goes to Fix Committed *after* we review it.
[10:43] <kiko> ScottK, you're not going to convince me over IRC, but I'd be happier if you waited until we think we have it right to weigh your definitive opinion.
[10:43] <kiko> radix, that's exactly what I described (hopefully)
[10:43] <radix> kiko: ok, I got a different impression.
[10:44] <ScottK> kiko: I'm willing to be convinced, but unless and until you get 100% concurrence from all projects on the semantics of Incomplete, whatever rule you choose will be wrong for someone.
[10:44] <radix> kiko: I guess I read "(Please Test My Fix)" as "(Please Review My Fix)"
[10:44] <ScottK> Additionally, I think it's just wrong to automatically reject someone's work (i.e. the bug report) without a human decision to do so.
[10:44] <kiko> ScottK, the janitor will help people get the semantics right. it's a bit forceful, but yeah.
[10:44] <ScottK> For you definition of right.
[10:44] <radix> kiko: oh, I see your comment in the second paragraph about review now. Ok, sorry.
[10:44] <ScottK> Not you customers.
[10:44] <ScottK> you/your
[10:45] <kiko> ScottK, I understand your philosophical issue, but I am more concerned with the practical issue. we have trillions of bugs and reporters which file them incomplete and never come back. 
[10:45] <kiko> if the reporter cares, he can come back and reopen it
[10:45] <kiko> so I don't see any philosophical problems here.
[10:45] <LaserJock> for me personally I'd rather get rejected automatically than by a person
[10:45] <ScottK> Then there are classes of bugs that are real issues even if you know enough.
[10:46] <ScottK> A crash is always a valid bug, you just may not now enough to figure out how to fix it.
[10:46] <ScottK> Crashes should never be automatically closed in my opinion.
[10:46] <kiko> ScottK, but if the crash isn't affecting anyone who cares about it, I argue it doesn't matter.
[10:46] <kiko> there are always bugs.
[10:46] <kiko> it's software
[10:46] <kiko> you wanna fix the ones which affect users.
[10:46] <ScottK> Crashes always represent a real defect.
[10:46] <ScottK> Sure.
[10:46] <kiko> the ones which aren't affecting users, you ignore.
[10:47] <ddaa> A crash bug report that is not reproducible by lack of information is not good for much.
[10:47] <ddaa> Besides, crashes are NOT always bugs.
[10:47] <ScottK> No, but several tother might be.
[10:47] <ScottK> They are always a bug in something.
[10:47] <kiko> in Launchpad, we get OOPSes triggered by crawlers that are submitting invalid UTF-8
[10:47] <ddaa> they may be caused by hardware defects too
[10:47] <kiko> do you think we care very much?
[10:47] <ScottK> kiko: No, I'm not under the impression you do.
[10:47] <kiko> I sneeze at said robots! crappy inhuman denizens of the internet.
[10:48] <kiko> anyway.
[10:49] <ddaa> ScottK: maybe your issue is more one of naming... I agree that "Incomplete" may not convey the right message.
[10:49] <ddaa> It really means "Reply expected, or else..."
[10:50] <ScottK> I think not losing information is more important than keeping some manager's bug stat report happy.
[10:50] <kiko> it's not really the manager; it's the triager who has a billion bugs to look at needs help filtering out what is more important.
[10:50] <ScottK> I also think not disrespecting bug reporter's contributions by having a bot judge their work is also important.
[10:51] <kiko> I think by not replying to a developer's request is a much greater disrespect.
[10:51] <ScottK> Then they can sort by recently changed.  That's what I do.
[10:51] <kiko> or whatever that means in english
[10:51] <ScottK> Yes, but that's a matter between the reporter and the triager.  The system shouldn't insert itself in the middle.
[10:52] <ScottK> But I need to lay off this discussion before my blood pressure gets to high.
[10:53] <kiko> you should try cycling. it is great for the heart
[10:53] <kiko> of course, you sometimes end up like this: http://www.amigosdabike.com.br/foto/bv_ft_32676_18.jpg
[10:54] <kiko> I am almost 100% healed from my last accident
[10:54] <kiko> but I have two races in the next week
[10:54] <ajmitch> time for another one then?
[10:55] <kiko> I'm praying no!
[10:55] <radix> kiko: do you show people that picture to help them relieve their fantasies of kiko-directed violence?
[10:55] <kiko> radix, in part, yes.
[10:55] <radix> good idea ;-)
[10:56] <kiko> I would be willing to offer a launchpad bugfix of choice
[10:56] <kiko> if a person explained to me
[10:56] <ScottK> I was once run over by a mini-van when on a bicycle.  It's no fun at all holding onto the front bumper trying not to go under.
[10:56] <kiko> how such clueless users subscribe to launchpad-users, confirming the email, and then can't get off.
[10:56] <kiko> it is amazing.
[10:56] <kiko> ScottK, front bumper? the person didn't brake?
[10:56] <radix> kiko: hmm
[10:57] <radix> kiko: interesting that confirming the email doesn't actually require understanding English
[10:57] <ScottK> They didn't see me until they hit me and say me going under.  I only got dragged a few meters.
[10:57] <radix> kiko: It could be that they received an email and hit "Reply" to type in "I don't understand you" ;-)
[10:57] <ScottK> Every mail list I've ever seen has that problem.
[10:57] <kiko> radix, so they could be replying "WTF ARE YOU TALKING ABOUT" in swahili?
[10:57] <radix> yeah.
[10:57] <kiko> ScottK, I think the secret might be MUA integration
[10:57] <kiko> i.e. put a button "I never want to get email from this f*cking source ever again"
[10:58] <ScottK> Well I know of people who would have trouble with that too.
[10:58] <radix> that sure would be nice.
[10:59] <kiko> I know some too. but maybe that might help some of the cases
[11:01] <ubotu> New bug: #148271 in rosetta "Downloaded tar file produces errors/warnings when extracting" [Undecided,New]  https://launchpad.net/bugs/148271
[11:05] <ubotu> New bug: #148276 in rosetta "po files should have the same name as when they were uploaded" [Undecided,New]  https://launchpad.net/bugs/148276
[11:20] <ubotu> New bug: #148286 in rosetta "multiple template export has wrong directory names" [Undecided,New]  https://launchpad.net/bugs/148286