/srv/irclogs.ubuntu.com/2009/09/30/#launchpad-dev.txt

* mwhudson afk for a couple of hours (but working, at least a bit)00:01
lifelesshi thumper00:03
lifelesssolved world peave yet?00:04
masomHi, any tips or howto properly setup launchpad for a local use (non-hosted)? It kinda runs fine with the debug branch but is there anything that should be modified to have a ... let say not devel launchpad instance?00:24
intellectronicamasom: there isn't really any reason to run launchpad in any capacity other than for development, so nobody puts any effort into that00:35
masomintellectronica: thats what i figured out since nobody made a howto on how to host it by himself.00:36
masomintellectronica: but technically, from dev to "production", the main changes are in the config files?00:36
masomintellectronica: (such as the ability to actually send outbound emails)00:36
intellectronicamasom: the production environment for launchpad is a _very_ complex affair (consider that there are several administrators and engineers working full-time just to keep launchpad.net going). it's the configuration and a setup of quite a few machines and many running processes00:38
intellectronicamasom: it's not impossible to replicate it if you're willing to commit the same resources, but it would seem a pointless exercise, given that launchpad.net is already there for you to use, and that you can help develop it, it being free software00:39
masomintellectronica: yeah, i understand that :) But it's just for the kick of actually having it running as the real deal. If nobody outside canonical is able to reproduce it, its dangerous.00:40
intellectronicamasom: there's nothing stopping you from running 'the real deal'. many people run it locally for development, and it's essentially the same software00:41
masomintellectronica: yeah thats what i figured out when it launched locally, i just wanted to poke around to know how hard it would be to make the devel react like the launchpad.net deployment. And i'm glad that it looks like its mostly config files to modify.00:42
lifelessdoes anyone else find that malone bug pages run _very_ slowly once you have 30 to 40 open?01:28
lifelessjust typing in the comment field feels like molasses01:28
masom mmm stupid question, what would trigger this: lazr.config.interfaces.ConfigErrors: ConfigErrors: launchpad does not have a localhost http key. ?02:02
masomOh also a very weird thing03:01
masommake run_all with a differnt host (let say dev.local instead of dev), apache tries to connecto port 8086, moving apache to proxy 8085 works, any reason why it's attempting to use 8086 ?03:02
lifelessany new on the red border problem?03:14
lifeless[6 edits so far on this bug summary :(]03:17
al-maisanGood morning06:58
thumperhi hi09:02
mrevellhi09:04
thumpermrevell: dude, you're across the table!09:07
mrevell:)09:07
jtvspm: system reboot, distractions...  I think we'll have to let the jp weirdness lie for the moment.09:19
spmjtv: :-) oki09:19
jtvspm: thanks though!09:21
spmjtv: anytime* (* subject to core hours, laptop availability and other terms & conditions. please see your contract)09:23
jtvspm: would that be the contract kept in the data centre managed by you untrustw^^^NO CARRIER09:24
* spm returns the lart to it's place of rest09:24
* jtv hates to think what place that is09:24
spmpile of shrunken pc's from devs that were .... larted09:27
stub>>> None < 4411:41
stubTrue11:41
stubGrrr...11:41
lifeless\o/11:41
=== jtv is now known as jtv-afk
=== jtv-afk is now known as jtv
=== matsubara-afk is now known as matsubara
thumperstub: ping13:22
stubthumper: pong13:22
thumperstub: it seems we are *still* missing a db permission https://lp-oops.canonical.com/oops.py/?oopsid=1368BM213:23
thumperstub: I thought it was fixed13:23
thumperstub: can you add it to your pending branch and fix on production?13:23
stubWe granted select permissions. looks like it is trying to update.13:24
stub(?)13:24
stubno....13:25
thumperhuh?13:25
thumperoutgoing email shouldn't update anything13:25
stubthumper: Is this the branchscanner?13:25
thumperstub: no this is the revision email sending script13:26
stubthumper: What db user does it connect as?13:26
* thumper looks13:26
thumpersend-branch-mail13:27
stubsend-branch-mail?13:27
thumperyes13:27
maxbWhat bit of LP actually requires memcached? "make run" seems happy without13:28
thumpermaxb: nothing just yet13:28
stubchicken/egg13:28
stubthumper: perms granted on production. Updating my branch now.13:30
thumperstub: thanks13:31
pooliethumper: http://bazaar-vcs.org/SmartPushAnalysis1.413:45
pooliesudo tc qdisc add dev lo root netem delay 500ms13:45
pooliego on, paste in sudo commands from irc :)13:45
poolieyou know it makes sense13:45
Fly-Man-Who handles the bugs related to the source pulls?14:51
spivFly-Man-: https://bugs.launchpad.net/launchpad-code14:54
Fly-Man-spiv: Is that also the place for this bug: rocketsetup has wrong keyserver mentioned ?14:55
spivJust file a bug on the launchpad project for that, IIRC.14:55
Fly-Man-spiv: Already did14:55
barryreviewers and lurkers -> #launchpad-meeting in 4m14:56
stubHow to I get the launchpad-developers-dependencies package rebuilt? I got a change landed to the branch meta-lp-deps branch but from there I am clueless.15:06
Fly-Man-./rocketfuel-setup: line 372: 19632 Segmentation fault      bzr branch http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/ $LP_TRUNK_NAME15:15
Fly-Man-ERROR: Unable to create local copy of Rocketfuel trunk15:15
Fly-Man-Anyone want to elaborate on that one ?15:27
Fly-Man-This keeps happening over and over again with the bzr 2.0.0.1 from the ppa15:28
barryFly-Man-: i've never seen that before.  can you gdb the core file and see what it says?15:28
Fly-Man-gdb-python or gdb itself ?15:29
barryFly-Man-: gdb i'm not sure what gdb-python is15:29
Fly-Man-python variant15:29
Fly-Man-Any command lines I need to add ?15:30
barryFly-Man-: actually, start with 'file core'15:30
Fly-Man-file core ?15:31
barryFly-Man-: that will tell you which binary is crashing, though i'm going to bet it's python.  and if that's the case, it's almost certainly a bug in some bzr extension15:32
barryFly-Man-: what platform are you on?15:32
Fly-Man-Vmware Jaunty 2 Gb mem, 15 Gb hard drive15:32
barryFly-Man-: i'm on karmic now and bzr 2.1.0dev.  i'm having other problems but dumping core isn't one of them ;)15:33
maxbRandom question: what exactly is rocketfuel? Is it just a cute name for devel?15:34
barryFly-Man-: try running that bzr command manually at the shell and see what happens15:34
Fly-Man-maxb: it's the script that installs the devel branch on a machne and setup the dependencies15:35
barrymaxb: basically yes.  it's historical15:35
barrymaxb: the branch used to be called rocketfuel (get it, launchpad... rocketfuel... soyuz... there's a theme :) but the only traces now are in the setup scripts15:36
maxb<Fly-Man-> Who handles the bugs related to the source pulls?15:37
maxbFly-Man-: In general, people don't handle bugs (on any open source project). Groups of people pick up various bugs depending on their availability. Hence the use of bug trackers. :-)15:37
Fly-Man-maxb: true15:43
Fly-Man-maxb:that's why ppl always are impatient when a bug is still not fixed ;)15:46
maxbstub: You build the package in the normal way of working with Debian source packages and upload it to the launchpad PPA. (Or, if you like, I'll do it and find someone in ~launchpad to copy it there)15:49
stubGiven I don't know where to start with working with Debian source packages, you or someone who understands had better do that part.15:50
maxbstub: I'm afraid trunk doesn't build - the trailer line on your changelog entry is malformed15:56
jtvabentley: looking into the workaround for the branch commits... whaddya think, should I do it right there in DirectBranchCommits?  Or in the script that calls it, and add a check in DBC?15:56
jtvabentley: The advantage of the latter is that the branch conversion may take a while, and it's polite to keep it in a separate transaction15:56
maxbstub: There must be two spaces between the email and the date, and the date format must exactly match the format demonstrated in other changelog entries15:56
abentleyjtv: Do what?  Convert to unstacked?15:56
jtvabentley: right15:57
stubmaxb: fixed15:59
abentleyjtv: I would do it in the script that calls it.  I'm not really happy with converting to unstacked, though.15:59
jtvabentley: I'm not completely happy about it either, but I think it beats alternatives like reverting to regular branch/commit/push or putting the burden on the user.16:00
maxbstub: not fixed - must be day month year, not month day year16:00
stubmaxb: and again?16:02
maxbstub: that's good, built source, uploaded to ~maxb/+archive/launchpad, waiting for process-upload and build16:09
stubta16:10
Fly-Man-barry: Strangely when I try it with the command line, it doesn't fail so far ...16:13
barryFly-Man-: very weird!16:27
jtvabentley: I don't suppose you guys have a helper function to say whether a branch is stacked?16:43
abentleyjtv: That information is stored in the database, and can be easily retrieved with bzrlib.16:45
jtvabentley: it turns out get_stacked_on_url is a very roundabout way of finding out.  :)16:45
jtvabentley: Branch.stacked_on is None, right?16:46
abentleyjtv: Yes.16:47
abentleyjtv: An alternative is to embrace "Easier to ask forgiveness than permission", and handle the exception generated by committing to a stacked branch.17:01
jtvabentley: wayyyy too much code for the "nothing's going on" case.  :)17:02
jtvI tried.17:02
* Ursinha looks around for bugs people17:05
Ursinhaintellectronica, hi :)17:05
Ursinhaintellectronica, could you please set the importance on bug 439449?17:05
mupBug #439449: On a bug, entering +edit page shows the wrong bug number <Launchpad Bugs:Triaged> <https://launchpad.net/bugs/439449>17:05
Ursinhathanks :)17:06
intellectronicaUrsinha: done17:07
Ursinhaintellectronica, thanks again17:07
intellectronicaUrsinha: thanks for bringing this to my attention17:07
Ursinhaintellectronica, no problem. I found this when trying to workaround bug 42392417:13
mupBug #423924: Entity-body was not a well-formed JSON document when updating bug description <post-3-ui-cleanup> <Launchpad Bugs:Triaged by deryck> <https://launchpad.net/bugs/423924>17:13
Fly-Man-barry: It's stopping at the same place everytime17:37
intellectronicajtatum: rejoice, your patch has landed!17:53
intellectronicajtatum: thanks again for picking this up, you rock17:53
jtatumintellectronica: thank you very much! the celebrations have begun here in my lair. champagne for everyone.17:53
=== matsubara is now known as matsubara-lunch
=== EdwinGrubbs is now known as Edwin-lunch
ScriptRipperhi kiko18:30
=== matsubara-lunch is now known as matsubara
Fly-Man-barry: *ping*19:45
barryFly-Man-: *ponngggggg!!!* :)19:46
Fly-Man-barry: What repo are you usin for 2.119:46
barryFly-Man-: bzr?19:46
Fly-Man-yes19:46
barryit's either:19:47
barrydeb-src http://ppa.launchpad.net/bzr-beta-ppa/ubuntu karmic main19:47
barrydeb http://ppa.launchpad.net/bzr-beta-ppa/ubuntu karmic main19:47
barry 19:47
barryor19:47
barrydeb http://ppa.launchpad.net/bzr-nightly-ppa/ppa/ubuntu karmic main19:47
barrydeb-src http://ppa.launchpad.net/bzr-nightly-ppa/ppa/ubuntu karmic main19:47
barry 19:47
barryprobably the latter and i probably shouldn't be using the nightlies ;)19:47
Fly-Man-Let's see if that fixes the segmentation fault19:50
Fly-Man-barry: nope, same issue20:01
Fly-Man-wth is happening here20:02
Fly-Man-it worked with 1.16 and 1.1820:02
Fly-Man-then the ugrade to 2.x makes it shakey20:02
barryFly-Man-: are you on a 64bit system perhaps? (i am tho and haven't seen this)20:03
Fly-Man-x86 machine20:03
Fly-Man-it's latest vmware20:03
Fly-Man-machine has 4 gb20:03
barryi can't imagine vmware has anything to do with it, but you never know20:04
barryFly-Man-: you might ask around on #bzr though20:04
Fly-Man-They blamed the mem20:04
Fly-Man-the memory is brand new on that machine20:04
Fly-Man-build that machine 4 days ago20:04
Fly-Man-Hmm, and there's no tree where there's 1.18 bzr anymore ....20:05
=== Edwin-lunch is now known as EdwinGrubbs
barryi always get suspicious when the hardware is blamed.  so much more rare than a missing incref in my experience ;)20:22
Fly-Man-that's my idea as well20:44
Fly-Man-why would it work on 1.1820:44
Fly-Man-and not on 2.x20:44
Fly-Man-barry: Could you try something for me ?20:46
Fly-Man-Could you create a tarball from devel ?20:46
Fly-Man-or someone else20:46
barryFly-Man-: why not try to grab the latest bzr trunk and build it from source?20:47
Fly-Man-I just pulled the nightlies20:47
Fly-Man-same segmentation fault20:47
Fly-Man-so it seems to be something in 2.x20:47
Fly-Man-which wasn't there in 1.1820:47
barryFly-Man-: maybe abentley or rockstar knows?20:48
Fly-Man-Good idea, abentley , rockstar : Any ideas ?20:48
rockstarFly-Man-, reading backchat20:48
Fly-Man-rockstar: Thanks20:48
rockstarFly-Man-, the nightlies are working fine for me.  Upgraded just this morning.20:50
Fly-Man-rockstar: In short: I've been using Launchpad.dev for some time, until the latest release of the bzr 2.x. After that, during reinstalls the bzr just quits with a segmentation fault while trying to pull the latest20:50
rockstarFly-Man-, what do you mean by "reinstalls" ?20:52
Fly-Man-rockstar: virtual machine reinstallations20:52
Fly-Man-Every 5 days I need to clean up Vmware machines20:54
Fly-Man-while the server is getting too burdened20:54
Fly-Man-so i cleaned up last Saturday20:54
Fly-Man-after that, it's a mess while trying to get the initial tree20:54
rockstarFly-Man-, it might very well be "hardware" because VMware's memory is still software.20:54
rockstarFly-Man-, I suggest branching bzr.dev, building the C extensions, and trying that.20:55
Fly-Man-rockstar: is there a trunk for that ?20:57
rockstarFly-Man-, bzr.dev is trunk20:57
Fly-Man-rockstar: Same error21:15
rockstarFly-Man-, try it outside a VM.21:15
Fly-Man-rockstar: I'll have a look when there's a machine available in the datacenter21:16
Ursinhahey cprov, are you there?21:23
rockstarbarry, so, I think I'm the only one around for the AsiaPac reviewer meeting today.  Is it even worth having one?21:25
barryrockstar: probably not, the most meaningful thing we did in ameu was to wish cprov well :)21:25
cprovUrsinha: yup21:26
Ursinhacprov, hi :) can you help me with https://answers.edge.launchpad.net/launchpad/+question/84264, please?21:33
cprovUrsinha: sure, in a bit21:34
Ursinhathanks cprov21:34
cprovUrsinha: done21:47
Ursinhacprov, thanks :)21:52
barrybac: ping21:57
bachi barry21:57
barrybac: hi, are you busy?21:57
bacbarry: i'm always busy and productive.  but i have a few moments to spare.21:57
barrybac: ;) can we skype?21:58
bacsure21:58
* bac launches21:58
maxbThanks whoever copied launchpad-dependencies 0.55 :-) I was just about to send an email requesting it :-)21:58
maxberr21:58
bacbarry: ready21:58
maxbwhoever copied it to hardy, shouldn't have21:58
barrybac: calling...21:59
barrybac: bug 40360621:59
mupBug #403606: ExpatError errors should be handled to not generate the OOPSes <oops> <Launchpad Registry:In Progress by barry> <https://launchpad.net/bugs/403606>21:59
barrybac: https://lp-oops.canonical.com/oops.py/?oopsid=1306XMLP221:59
barrybac: http://paste.ubuntu.com/282468/22:03
maxbHmm. The launchpad PPA for hardy is currently uninstallable.22:08
maxbIs there anyone who needs to know this?22:08
barrybac: eggs/zope.publisher-3.5.6-py2.4.egg/zope/publisher/xmlrpc.py22:09
Fly-Man-rockstar: *ping*22:09
Fly-Man-A new machine just got delivered in the datacenter22:10
Fly-Man-so I;m setting that one up with Jaunty now22:10
Fly-Man-to see if it works then22:10
rockstarFly-Man-, if you continue to get the segfault, #bzr is going to be able to help much more than I can.22:14
maxbI wish the meta-lp-deps branches were somewhere I had write access :-/22:15
maxbPlease can I enlist a member of ~launchpad to add this missing tag:22:18
maxbbzr tag -d lp:~launchpad/meta-lp-deps/trunk -r revid:maxb@f2s.com-20090910075602-i7m5vu8dicck9xcu 0.5422:18
maxbbzr tag --force -d lp:~launchpad/meta-lp-deps/trunk -r revid:stuart.bishop@canonical.com-20090930150127-02ubq4ceny88higo 0.5522:19
maxbalso22:19
maxband yes I do mean --force in the second22:19
=== cprov is now known as cprov-afk
mwhudsongood morning22:34
=== matsubara is now known as matsubara-afk
maxbgood evening :-)22:46
maxbTo what extend do we care that the launchpad PPA is not currently installable on hardy?22:47
mwhudsonmaxb: it's pretty bad22:48
maxbI'm in no position test anything more than the installability of packages on hardy, but I could prepare something that is at least better than what is there.22:51
maxbCan I enlist you to run those two bzr tag commands from scrollback ~30 mins ago before they get forgotten about, or shall I email them to the list?22:53
mwhudsonmaxb: ok22:56
mwhudsonmaxb: what's the problem on hardy currently?22:57
maxbSomeone copied the new jaunty/karmic launchpad-dependencies build to hardy too, but hardy was using a branched version22:57
maxbso the hardy-specific changes were removed22:57
mwhudsonoh right22:58
maxbOn a related note, we should probably change the description of the PPA to say that Intrepid is no longer supported.22:58
=== Ursinha is now known as Ursinha-afk
maxbOK, in two minutes, once it's published, please could someone copy-with-binaries 0.55hardy1 from maxb/launchpad to launchpad/ppa (hardy)23:23
maxbAnd also push lp:~maxb/meta-lp-deps/hardy into lp:~launchpad/meta-lp-deps/hardy23:24
maxbright, it's published23:25
beunomaxb, I think I have the power to do so23:29
maxbthanks23:29
Fly-Man-rockstar: It seems that there's an issue with the latest Vmware server23:32
Fly-Man-I have alerted the Vmware staff to have a look at it23:32
beunomaxb, done23:33
maxbThanks, hardy lp devs will thank you :-)23:36

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