/srv/irclogs.ubuntu.com/2010/11/05/#launchpad.txt

EagleScreenhi01:06
EagleScreenis there any channel to talk about launchpad development?01:06
persiaSome stuff is on-topic here, details have #launchpad-dev01:06
EagleScreendoes it envolves soyuz dvelopment?01:07
wgrantSoyuz is part of Launchpad, so it belongs in #launchpad-dev with the rest.01:08
persiaWhat is your question?01:08
EagleScreenI only want to talk about this wishlist bug https://bugs.launchpad.net/soyuz/+bug/18856401:11
ubot5Launchpad bug 188564 in Soyuz "Build also packages for Debian in PPA's (affected: 41, heat: 310)" [Low,Triaged]01:11
EagleScreensome people think that adding unstable or testing Debian suits is as easy as adding a new Ubuntu suit, is it true?01:11
persiaNot precisely, no.01:12
wgrantIt's mostly an issue of build resources.01:12
EagleScreenwhat is the difference?01:12
wgrantNot a technical problem.01:12
persiawgrant, Well, there's the virtual/non-virtual game, and questions of build-environment, etc.  But yeah, only minor technical: mostly financial.01:12
EagleScreenif a new Ubuntu suit is added each 6 months, can't be a Debian siut added as the same way?01:13
EagleScreenat least for i386 and amd64 archs01:15
persiaNon-trivially, and not without significantly more machines (as people would want to use them)01:17
EagleScreendo you think that more servers would be needed? more cost?01:18
persiaYes.01:18
EagleScreenreally? it would be just one or two new suits..01:19
Nafallowe deprecate an ubuntu release every 6 months as well (usual case for non-LTS)01:19
wgrantEagleScreen: It would probably be two new series. And if most people are building for only two or three Ubuntu series already, that doubles the amount of hardware required.01:20
lifelessEagleScreen: we regularly use all our machines with a significant backlog01:21
lifelessEagleScreen: as wgrant says adding a new target would also need machines to build it to be feasible01:21
EagleScreenthen I understand that it is a economic issue01:21
persiaEagleScreen, I suspect that if you approached Canonical with significant financial incentive, you may be able to make that happen.  I do not believe that attempting to get the bug closed will achieve much, as the operators probably wouldn't enable the additional suites even if they were turned on.01:22
persia(alternately, if you run a private LP instance, it oughtn't be that hard to implement, to run on your machines)01:22
EagleScreenI see..01:23
EagleScreenis already natty available for PPA upload?01:23
persiaHappens the same time as natty appears anywhere else.01:24
EagleScreenI have receibed this email:01:24
EagleScreenLaunchpad failed to process the upload path '~<EagleScreen>/ppa/ubuntu/natty':01:24
wgrantEagleScreen: '<EagleScreen>' is not a valid Launchpad username.01:25
EagleScreenremove < > ?01:25
wgrantRight, but a Launchpad username is also lowercase.01:26
wgranteg. 'eaglescreen'01:26
EagleScreenat upload:01:33
EagleScreenUnable to find distroseries: experimental01:33
EagleScreenthis package was obtained from Debian experimental repo01:33
EagleScreenbut doesn't launchpad allow upload it if I specify the subfolder to natty?01:34
paultagOne last thing to bug the LP admins in here about -- I need to set the Bug Supervisor to the LoCo Council -- I am not an admin on the team ( council members are members of the team ), and I asked persia ( on the CC, which owns the LC ), and he was not able to see the pages. What should I do?01:34
wgrantEagleScreen: Yes. I suspect that you didn't specify it properly.01:34
EagleScreenok01:34
EagleScreenwill pastebin my dput.cf01:35
persiaEagleScreen, it's not about dput: it's about your changelog entry.01:36
EagleScreenI think changelog can be experimental or unstable if I specify the Ubuntu suit upload subfolder01:37
wgrantpersia: The dput path can override the changelog entry.01:37
persiawgrant, Sure, but does Soyuz process that?  I thought it just followed .changes01:37
wgrantpersia: It respects the path in preference to .changes.01:38
wgrantThis is, however, fairly widely regarded as evil.01:38
persiaAh, then I'll keep telling folk it works the way it should, and just wait for some Soyuz developer to unexpectedly fix it one day.01:39
EagleScreenthis is http://pastebin.ca/198238201:39
wgrantpersia: The way it should?01:40
wgrantpersia: It obeys the .changes unless you explicitly specify a distroseries in the upload path.01:40
EagleScreenI have uploaded packages to PPA with unstable in the chnagelog01:40
EagleScreenbut some time ago01:41
wgrantEagleScreen: That's fine. But why are you regularly uploading packages with the wrong series in the changelog?01:41
persiawgrant, And about the evil bit?01:41
wgrantpersia: It is often abused to upload unmodified Debian source packages, when this is not in fact an appropriate thing to do.01:42
EagleScreenwgrant: why not? time ago they were accepted, any change about this?01:42
paultagEagleScreen, it wastes time01:42
wgrantEagleScreen: It will be accepted, but it's not always the right thing to do.01:42
persiawgrant, Right, which is why I don't like path overrides.01:42
paultagEagleScreen, and you suck up the build farm time01:42
paultagEagleScreen, for a package that can be found upstream01:43
persiapaultag, Not necessarily: binary package differences can be extreme even with the same source.01:43
paultagpersia, if there is a distro check, yeah -- but then it would have been caught in the sync with Debian for Ubuntu it's self01:43
EagleScreenthen usntable is not longer accepted?01:43
paultagright?01:44
persiapaultag, Consider build-dependencies01:44
paultaghummm, true.01:45
EagleScreenNot permitted to upload to the RELEASE pocket in a series in the 'CURRENT' state. <-- what does this mean??02:01
wgrantEagleScreen: You tried to upload to Ubuntu itself.02:03
wgrantNot a PPA.02:03
EagleScreenhow is it possible?02:04
EagleScreenthis is my dput.cf: http://pastebin.ca/198239202:04
wgrantEagleScreen: Why are there capital letters in your Launchpad username?02:05
wgrantLaunchpad usernames cannot contain capital letters.02:05
wgrantI would also strongly encourage you to rethink how you are creating your packages.02:05
wgrantUsing the path override should *not* be a normal occurrence.02:06
EagleScreenwgrant: this last error was with natty and maverick in changelog02:06
EagleScreenChecking now the capital letters ..02:06
wgrantEagleScreen: Which command did you use to upload?02:07
wgrantI suspect you omitted the target.02:07
EagleScreenlol; I put the targer after the .changes02:08
EagleScreenthe target go first, right?02:08
wgrantYes.02:09
EagleScreenone package didn't build due to a missing build-dependency, if I upload that dependency to my PPA, will the package build later if I retry?02:12
wgrantIt should.02:14
EagleScreennice02:16
=== bac changed the topic of #launchpad to: Launchpad: https://launchpad.net/ | Read https://help.launchpad.net/ for help | On-call help contact: bac | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: https://dev.launchpad.net/
bacHello, I am the Launchpad help contact for today.  Please ping me if you have questions.02:49
MTecknologybac: https://launchpad.net/~hjgf06:01
MTecknologybac: I suspect spamming - adding all teams in LP ?06:01
MTecknologywgrant: you might wanna look too?06:02
wgrantMTecknology: Not another one :(06:05
MTecknologysorry..06:05
wgrantspm: ^^06:05
spmuser suspended, team deleted06:08
* spm wields the sledgehammer.06:08
MTecknologyspm: yay06:17
MTecknologyspm: that means now you help me with my package?06:18
spmhm?06:18
MTecknologyI finally made it build without fail - but what's produced isn't at all what I was expecting06:19
StevenKThat sounds like a packaging issue, rather than something that a LOSA can help with ...06:21
MTecknologyStevenK: it is.. but you can't blame me for asking :D06:22
bacthanks MTecknology.07:00
MTecknologybac: :) Death to spam!07:01
wgrantspm: Could you also delete ~vishnu.team?07:05
wgrantspm: It's the same user, and similar members.07:05
wgrantAlso ~ab2.team.07:06
wgrantWTF07:06
lifeless?07:06
wgrantThose three spam teams all with similar members.07:07
MTecknologywgrant: probably the same person too?07:07
wgrantYes.07:08
wgrantThey're all owned by the suspended user.07:08
spmbah07:08
MTecknologyI wonder what the goal of that was.. it's not exactly 'that' disruptive07:08
spmno idea07:09
spmwgrant: both removed07:10
wgrantspm: Thanks.07:10
wgrantOhhh.07:11
wgrantIt's this same guy again.07:11
wgrantHe was at it a few months ago.07:11
wgrantAnd some of the teams from back then are in this web.07:11
spmlovely07:15
wgrantI thought that one of the teams was legit.07:16
wgrant(~india.launchpad.team)07:16
wgrantBut most of its 100 members were added within a few minutes of each other.07:16
wgrantI suspect involuntarily.07:16
MTecknologyperhaps a feature added that doesn't allow you to add a team to a team unless either a) you're a member (admin?) of both or b) an admin approves it?07:17
wgrantMTecknology: That's already the case.07:17
MTecknologyoh..07:18
wgrantThose teams he added are also owned by him.07:18
MTecknologyoh..except ~kernel-bugs - that got added07:18
wgrantAh, true.07:18
lifelesswe need rate limiting heuristics07:19
lifelessfor many many things07:19
MTecknologyoh!... I remember when he was doing that last - I think I even pointed it out that time :P07:19
wgrantYou did, yes.07:19
MTecknologyI need to toss in a little bit more spam control on wiki.nginx.org too...07:21
MTecknologylifeless: perhaps.. spam detection that.. once is 95% confident you're spamming.. 1) blocks you 2) warns losa 3) losa confirms spam 4) deletes teams 5) suspends user 6) reaches through the internet and blows up computer 7) if shrapnel causes extra dammage, oh well  ??07:24
spmwe have orbital deathrays that a nameless person installed when he was in orbit, we can use.07:25
MTecknologyvader?07:25
spmon the grounds I like my job very much thanks, I will strenuously deny that it's vader or a vader like person07:26
twbFYI, Debian's w3m maintainer has patched w3m to handle <BUTTON>, which was a show-stopped for launchpad logins.07:26
MTecknology:P07:26
twbLP: #628755, closes: #13681007:26
spmtwb: good news! (have hit that one msyelf doing internal stuff)07:27
MTecknologyg'night everyone07:34
=== jtv1 is now known as jtv
mrevellMorning09:00
Takhello09:00
=== LinuxJedi|away is now known as LinuxJedi
=== Guest26427 is now known as jelmer
=== bac changed the topic of #launchpad to: Launchpad: https://launchpad.net/ | Read https://help.launchpad.net/ for help | On-call help contact: - | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: https://dev.launchpad.net/
=== matsubara-afk is now known as matsubara
=== yofel_ is now known as yofel
ari-tczewwhy this branch has been created? https://code.launchpad.net/~ubuntu-branches/ubuntu/natty/therion/natty-20101104031212:52
gnomefreaki keep getting OOPS 1770g114213:02
gnomefreaki keep getting OOPS 1770G114213:03
gnomefreakanyway its a time out OOPS13:04
=== mrevell is now known as mrevell-lunch
gnomefreakok the bug koads fine just when i try to add xulrunner-2.0 to it it times out13:09
gnomefreaks/koads/loads13:09
falktxhi there13:14
falktxi just noticed a bug on PPAs13:14
falktxwhen I try to copy many packages at once, I get a server timeout error13:14
gnomefreaki get one when trying to add a package to a bug filed on PPA13:21
* gnomefreak finally gave up13:22
geserari-tczew: this happens when the package importer notices a difference between an upload and an bzr push; james_w can probably explain it better14:29
ari-tczewthat is a reason for me to not using bzr14:30
tumbleweedI'm guessing we don't have /dev/shm mounted in launchpad-buildd, right? bug 67144114:39
ubot5Launchpad bug 671441 in joblib (Ubuntu) "joblib FTBFS in Natty (affected: 1, heat: 6)" [Undecided,New] https://launchpad.net/bugs/67144114:39
shane4ubuntuI setup a ppa and imported a key (for signing and uploading my package) when I tried to import my ppa to my system I get this error:  Error: can't find signing_key_fingerprint14:41
shane4ubuntuIs that the same key as my uploading thing?  or is it different?14:42
=== mrevell-lunch is now known as mrevell
=== LinuxJedi is now known as LinuxJedi|away
achiangthis is a rather stupid question, but can i add more um, entries, to a lazr Collection object?15:04
achianglooking at dir() doesn't tell me much15:04
jmlachiang: it's not a stupid question15:10
jmlachiang: but I don't know what the answer is15:10
jmldir() isn't very helpful on lazr objects in general.15:10
achiangjml: hi! you're back in england now?15:10
jmlachiang: indeed I am.15:10
achiangjml: the weather can't be worse than here in boston. solid rain for the past few days15:11
jmlachiang: it's been a little better than that.15:11
* achiang is getting a little homesick, wants to return to http://bit.ly/a0gWxk15:11
achiangjml: my real issue is that i've discovered that in launchpadlib, searchTasks() by default won't tell you about tasks that are closed15:13
jmlahh yes15:13
jmlachiang: you have to specify every status to get that15:13
achiangso my thought was i would do a pass with searchTasks(status="Fix Committed") and then add on the results from default searchTasks()15:14
jmlachiang: here's one I prepared earlier: http://paste.ubuntu.com/526361/15:14
=== Ursinha is now known as Ursinha-lunch
achiangjml: that seems easy enough, but i want to avoid some DRY in my code. i guess it won't be so terrible if i just globally define a list of the statuses and just reference that list in each call to searchTasks15:15
achiangjml: your method is simple enough that even I can understand it though. :)15:15
jmlachiang: yeah, that sounds right.15:15
jmlachiang: it's arguably a bug that you have to have such a list15:16
achiangjml: i'm doing a little hacking on my tool that helps me write my status reports: tells you about all launchpad activity in the last week and dumps it out to copy/paste into an email15:16
achiangjml: the tool is not quite as useful if it doesn't tell you about bugs you've closed. ;)15:17
jmlachiang: ahh, nice.15:17
achianghttps://launchpad.net/laika15:18
achiangstill in the extremely hacky stage right now. trying to find the proper home for it at some point.15:18
achiangperhaps it'll need a rewrite and integration into bughugger15:18
jmlachiang: yeah, there's not really a good home for tools like that.15:22
jmlmaybe lptools15:22
achiangjml: yeah, that was my other thought15:23
=== matsubara is now known as matsubara-lunch
=== Meths_ is now known as Meths
=== beuno is now known as beuno-lunch
=== vila is now known as Bernardo
=== Bernardo is now known as vila
jcastroderyck: any idea why I can't linkify this here to the upstream url? https://bugs.launchpad.net/wajig/+bug/54074016:20
ubot5Launchpad bug 540740 in wajig (Ubuntu) "wajig installfoo gives misleading error message (affected: 1, heat: 6)" [Undecided,New]16:20
deryckhi jcastro.  looking now...16:21
jcastrothe upstream bug is in google code so I don't think I need to register it?16:21
deryckjcastro, yeah, you'll still have to register their tracker on google code.16:22
jcastrok16:23
jcastrothat wasn't so obvious16:23
deryckjcastro, and looking at https://bugs.launchpad.net/wajig it looks like the project is set up to have lp track its bugs.16:23
* jcastro can fix that16:23
sladencould somebody ban  braulioareis  until we work out what's going on16:24
sladenthey've been changing statuses and bug assignments all over the place16:25
derycksinzui, ^^16:25
jcastroderyck: ok I just need to know where to link lp/wajig to the external tracker? I don't see anything in administer16:26
deryckjcastro, I think only the project maintainer can setup the tracker.  that project has an individual listed as the maintainer.16:27
jcastroah ok16:27
jcastrobut now that a bug tracker is registered I can still set a bug watch right?16:27
sladenhggdh: I see you've been clearing up after  braulioareis  too16:29
sladenhggdh: sinzui: are we able to do this in a more pragmatic way---I presume there are audit logs16:30
sinzuisladen, no there are no audi logs16:30
=== LinuxJedi|away is now known as LinuxJedi
sladensinzui: joke or seriously?16:32
sinzuisladen, true16:32
=== matsubara-lunch is now known as matsubara
hggdhsladen: yes, I have, but I cannot work on the upstream projects16:32
sinzuisladen, deryck. I do not think the activity log can every be used to roll anything back16:32
jcastroderyck: hmm, still no place to paste a URL16:33
sladenhggdh: yeah, I've hit the same issue... not all the changes are reversible16:33
deryckjcastro, the upstream has to use the tracker you registered first16:33
jcastrook so in this case, pasting it in a comment will probably work?16:33
derycksinzui, sladen -- the activity log has the data, but we don't currently have a way to roll back based on that data16:34
hggdhsladen: yup. But at least we can clean those we have access to. I also emailed the braulioareis guy about his acitons16:34
sinzuisladen, deryck, allenap. I see there is already a question about braulioareis? If he has already been contacted I am tempted to suspend him right now16:34
hggdhsinzui: I just emailed him16:34
derycksinzui, we haven't contacted him.  I thought you were, for some reason.16:35
hggdhsinzui: and I did that as a bugsquad/control admin, not as a LP person16:35
allenapsinzui: https://answers.launchpad.net/malone/+question/13239616:35
sinzuideryck, I do not see my involvement in https://answers.edge.launchpad.net/malone/+question/13239616:35
sinzuideryck, I worked with a user who lost control of his email this week. Maybe that is what you recall16:36
derycksinzui, yes, I think so.  Sorry16:37
deryckI assumed that was this guy.16:37
sinzuiI really wish we had standing working, or a CoC requirement to work across communities16:37
=== deryck is now known as deryck[lunch]
sladensinzui: did we get them suspended yet?16:45
sinzuiI have the screen open, I was going to wait 15 more minutes to hear if the user apologised.16:46
allenapI think we should suspend his account. I haven't seen any activity on his account that doesn't look like experimentation.16:46
sinzuisladen, suspend is very drastic. We cannot undo it. an Lp admin can and arranging that often takes weeks16:47
allenapsinzui: What is the difference between "Deactivated" and "Suspended"?16:48
sinzuiallenap, users are reactivated on login. is SSO/authentication worked, it would also restore your email address and launchpad id16:49
sinzuiallenape, deactivated is really governed by the user, not us.16:49
allenapsinzui: Ah, okay. The only hammer we have is a very very big one.16:50
sinzuiallenap, suspend locks the email address, sets the user to http status to 410 and breaks SSO too. It is a great way to shutdown anyone using ubuntu or launchpad SSO16:51
sinzuiallenap, this issue qualifies as "poor standing" Nothing uses standing, but the idea is that users with "poor" should not be permitted to cross communities. Users with great standing could cross post to lists without being members16:52
allenapsinzui: It would be nice to manually just stop the user from using Launchpad for a while, without breaking SSO.16:55
sinzuiWe were given access to suspend users to deal with spammers, and it does exactly what we intend. Standing was designed to solve community permission issues. My last two attempts to raise its priority failed :(16:57
* sinzui suspends ~braulioareis16:57
=== deryck[lunch] is now known as deryck
=== beuno-lunch is now known as beuno
=== oubiwann is now known as oubiwann-away
=== oubiwann-away is now known as oubiwann
alkisgHi, can I build-depend on something that is on universe when uploading to my launchpad PPA?17:33
alkisgSorry, should have googled before asking - found it at https://help.launchpad.net/Packaging/PPA/BuildingASourcePackage - it says "yes" :)17:34
=== Ursinha-lunch is now known as Ursinha
=== zyga is now known as zyga-food
alkisgDo packages on launchpad have internet access while they're building? E.g. debian win32-loader uses wget to download vmlinuz as part of its build process...18:59
=== zyga-food is now known as zyga
micahgalkisg: no19:04
alkisgmicahg: thank you, so I have to patch it and include a vmlinuz inside the source tree19:05
micahgalkisg: or package it as a separate source19:06
alkisgmicahg: sorry, can you elaborate?19:06
alkisgAh, you mean the whole kernel as a dependency?19:06
micahgalkisg: yes19:07
alkisgNo I'm looking to use gpxe.kdn to provide a "boot from network" option for windows PCs19:07
alkisg*krn19:07
alkisgSo vmlinuz == gpxe.krn in my case, and that's not on debian/ubuntu yet19:07
alkisg(it's on fedora but not on debian)19:07
alkisgSo I'll be forced to use a binary blob...19:08
alkisgHm. Or not. You're right19:08
alkisgI've already packaged that to my ppa, so I can use it. Thanks!19:08
cody-somervilleIs there a PRODUCTION_SERVICE_ROOT constant in launchpadlib or something? How do I access the production service root? :P19:32
james_wcody-somerville, LPNET_SERVICE_ROOT IIRC19:32
james_wnot in old old versions of launchpadlib19:32
cody-somervilleah19:33
cody-somervillejames_w, was it added in 1.6.5? I have 1.6.4 and it doesn't seem to have it.20:07
james_wcody-somerville, I don't remember20:07
james_wcody-somerville, does your version have a launchpadlib.uris module?20:07
cody-somervilleyup20:08
cody-somervilleah20:08
james_wcody-somerville, then you can just use "lpnet" or something and it will translate for you20:08
james_wplus the variable can be imported from there20:08
derkshello... is it possible for a non subscriber to send to a launchpad team mailing list at all?  for example.. we have a hudson build system and want to send failure emails to a launchpad list...20:18
=== almaisan-away is now known as al-maisan
=== zyga_ is now known as zyga
=== al-maisan is now known as almaisan-away
=== matsubara is now known as matsubara-afk
=== JanC_ is now known as JanC
=== Ursinha is now known as Ursinha-afk
wgrantlamont: Did you hear about the buildd-manager fun yesterday?23:12
=== LinuxJedi is now known as LinuxJedi|away

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!