[00:38] <phillw> xnox: If you have a CLI for updating precise --> Raring, I'll be more than happy to run it on a VM and report back any errors for you, same goes for lucid --> raring. I have both as VM's and after updating, I would just need to clone them.
[00:39] <phillw> before attempting to get the new ones to raring.
[03:34] <yotux> Using live image for ubuntu amd64 today gui will not load
[03:35] <yotux> Can I install via cmd?
[05:09] <vibhav> GOod Morning
[07:24] <dholbach> good morning
[08:53] <jibel> good morning
[08:53] <pitti> bonjour jibel
[08:54] <jibel> bonjour pitti , comment ça va?
[08:55] <pitti> jibel: ça va bien! I'm stuck in pygobject debugging, chasing memleaks; argh
[09:00] <jibel> pitti, chasing memleaks can be very frustrating
[09:41] <jibel> pitti, I blocked on a rather trivial problem with adt. Some tests write test results inside the adt chroot in places like debian/tests/results
[09:41] <jibel> pitti, any idea how we could gather the results before adt destroys its chroot?
[09:42] <jibel> actually the results are in $ADTTMP/<anotherdir>/<source_pkg>/debian/tests/result
[09:43] <pitti> jibel: what is the recommended place for tests to put results into?
[09:43] <pitti> jibel: could we just fix those tests?
[09:44] <jibel> pitti, there is no recommended place, but we can define one
[09:44] <jibel> I could ask to write results to /tmp/something but that'd only work for the null dirver
[09:44] <jibel> driver
[09:45] <pitti> I meant, adt-run already has an --output-dir
[09:45] <pitti> for the log files
[09:45] <pitti> but of course that doesn't apply to files created by the tests
[09:45] <jibel> yes it does, but it doesn't seem to copy all the files generated by the test
[09:45] <jibel> right
[09:45] <pitti> jibel: debian/tests/results/ or $ADTTMP/results/ seem appropriate
[09:46] <pitti> I slightly prefer the latter, as it avoids writing into the source tree
[09:46] <jibel> pitti, ok, thanks, I'll do some more tests
[09:58] <xnox> phillw: whatever jenkins is running as upgrade tests that would be useful.
[15:31] <roadmr> cprofitt: hey, are you around?
[15:36] <cprofitt> roadmr: yes
[15:37] <roadmr> cprofitt: hey :) when is the next UF meeting scheduled? I missed the last one :/ I'm terribly sorry about that
[15:38] <cprofitt> I do not have one scheduled currently, but we should work on that
[15:38] <cprofitt> I was caught off-guard about UDS being next week, but it would be good if we could get some of these items discussed at that
[15:38] <roadmr> cprofitt: ok. With virtual UDS coming next week, I wanted to propose ...
[15:38] <roadmr> cprofitt: haha you read my mind? it's obvious that it'd be a good opportunity :)
[15:40] <cprofitt> it would be... can you organize that?
[15:42] <roadmr> ara: so what's the process for getting a UDS session scheduled? a blueprint needs to be created first, right?
[15:42] <smartboyhw> roadmr, yes
[15:43] <roadmr> smartboyhw: thanks :)
[15:44] <roadmr> cprofitt: so we should create a blueprint on launchpad, and I can see that it gets a session to discuss it at UDS
[16:21] <eugene_> Hello, everyone. What should I do if there is a mistake in one of the current Cadence's testcases? Where should it be reported?
[16:26] <cprofitt> roadmr: sorry got pulled away... yes, that is what we should do
[16:29] <roadmr> cprofitt: yay, ok so since we can potentially have all interested parties there, including jedimike, we should set a few goals then use the session to flesh out the work to be done to accomplish those goals
[16:29] <roadmr> cprofitt: of course more goals can appear during the session but having an idea of where the discusion should go would be good
[16:38] <cprofitt> roadmr: yes
[16:38] <cprofitt> we need to contact jedimike
[16:39] <roadmr> cprofitt: let me see if I can get him in here
[16:41] <jedimike-> roadmr, hi
[16:42] <roadmr> jedimike-: hello! so cprofitt and I are discussing having a session at UDS next week to discuss Ubuntu Friendly
[16:43] <jedimike-> roadmr, that would be good, hopefully the next two weeks at work won't be as intense as the last month :)
[16:45] <roadmr> jedimike-: yay, so what we're going to do is create a blueprint outlining some high-level goals we'd potentially like to achieve and then discuss in more detail plus create actions/work items in the session
[16:45] <roadmr> jedimike-: of course it'd be important for you to attend since you know the UF website/backend best
[16:46] <jedimike-> great!
[16:47] <jedimike-> will be a welcome break from a project from debugging thousands of lines of BASH :)
[16:51] <roadmr> jedimike-: awesome, just wanted to know if you liked the idea.
[16:52] <roadmr> jedimike-, cprofitt : ok so I guess next step is to create the blueprint heh
[16:52] <jedimike-> roadmr, yes, absolutely, there are so many things I want to improve on the backend of UF, including having checkbox submit directly to it, and I'm happy you're involved because we can make those things happen
[16:53] <roadmr> jedimike-: yep, exactly, but we need to be able to schedule the time to do it, and it'll be easier if we have an actual blueprint with well-outlined work
[16:54] <jedimike-> absolutely
[16:54] <jedimike-> let's do it! :)
[16:54] <roadmr> jedimike-: you already know how to parse submission.xmls directly, and we know how to submit them to an endpoint, so all the pieces exist, it's just a matter of putting them together
[16:54] <roadmr> jedimike-: this will get us to being able to give much better feedback to contributors, which will make cprofitt happy :)
[16:55] <jedimike-> roadmr, fantastic, I think cprofitt knows that not being able to give that feedback is one of my major annoyances with the project as it is.
[16:56] <roadmr> jedimike-: yes, everyone's; and I think the single most frequent question is "why aren't my results shown yet?"
[16:57] <cprofitt> roadmr: yes, create the blueprint
[16:57] <cprofitt> I will have some difficulty in attending - depending on when it gets scheduled -- because I am at work during the time UDS is held
[16:58] <cprofitt> the 4pm - 11pm UTC is equivalent to 11am to 6pm my time
[16:58] <cprofitt> we also have to work on the details of how to use the G+ hangouts etc...
[17:00] <roadmr> cprofitt, jedimike- : The goal is to discuss and plan short-term improvements in Ubuntu Friendly, aiming at providing faster/better feedback for testers, so people know their score right away and are able to correct any errors due to bad testing before submitting (or at least, having a tighter feedback look so they know they need to retest and resubmit).
[17:00] <SergioMeneses> cprofitt, I could have the same problem :S
[17:00] <roadmr> ^^ this is the blueprint description
[17:00] <SergioMeneses> byw hi all!
[17:01] <roadmr> SergioMeneses: hey! what timezone are you on?
[17:01] <SergioMeneses> roadmr, utc-5
[17:01] <cprofitt> roadmr: yes, that sounds like a good generalization of what we want to do
[17:01] <roadmr> SergioMeneses: sounds like EST to me
[17:02] <cprofitt> I think we also want to address usability and features of the UF site itself
[17:02] <roadmr> cprofitt: ok let me add that as well
[17:02] <cprofitt> but we might need two different sessions to accomplish that
[17:02] <cprofitt> not sure how long the sessions are
[17:02] <SergioMeneses> cprofitt, btw I think we can edit the new UF wiki
[17:02] <roadmr> cprofitt: I think they'll be 1 hour long but don't quote me on that :p
[17:05] <roadmr> cprofitt, SergioMeneses, jedimike- : OK so it looks like you register as UDS attendants, indicate the hours you can attend, subscribe to the blueprint
[17:05] <roadmr> cprofitt, SergioMeneses, jedimike- : then the scheduler will try to put the session at a time that's convenient for all, if possible
[17:05] <roadmr> if not, it may explode :)
[17:07] <cprofitt> can we cover both topics in 1 hour?
[17:07] <roadmr> cprofitt, SergioMeneses, jedimike- : when you register for the blueprint, mark yourselves as essential so the scheduler tries its hardest to accomodate you
[17:07] <cprofitt> can you link the blueprint?
[17:07] <roadmr> cprofitt: hmm we can try and if we run out of time, schedule another session
[17:07] <roadmr> cprofitt: not sure if it's worth trying to schedule two sessions now
[17:08] <cprofitt> ok... lets go with one... we can always meet ourselves later
[17:08] <SergioMeneses> roadmr, perfect then :D thanks for your help
[17:08] <cprofitt> they gotta remove 'physically' from the options
[17:08] <SergioMeneses> btw where is the bluprint?
[17:08] <roadmr> cprofitt: yep, as soon as I finish creating it :) hehe it's the first time I do it so I'm going a bit slow
[17:08] <roadmr> SergioMeneses: it will be somewhere in blueprints.launchpad.net
[17:08] <cprofitt> sounds good... thanks for taking care of it roadmr
[17:09] <SergioMeneses> cprofitt, but will there not more uds in person ?
[17:10] <cprofitt> SergioMeneses: that is the plan for at least the next two
[17:10] <cprofitt> then they said they would re-evaluate
[17:10] <SergioMeneses> cprofitt, mmm...
[17:10] <cprofitt> I am guessing if the rolling distro gets adopted that the only real need for a UDS would be for LTS releases
[17:11] <SergioMeneses> ok guys, lunch time, I'll read the log :)
[17:11] <cprofitt> but that is pure speculation on my part
[17:16] <roadmr> SergioMenesesAFK: cprofitt : jedimike-: https://blueprints.launchpad.net/ubuntu-friendly/+spec/uds-march-2013-ubuntu-friendly-improvements
[17:16] <roadmr> cprofitt: I signed you up as drafter but don't worry, I can help you with that
[17:21] <cprofitt> thanks,... I have done that before
[17:22] <cprofitt> I subscribed and made participation essential
[17:22] <cprofitt> though jedimike- is much more important than I am
[17:25]  * roadmr goes for lunch - I'll continue this in a bit
[21:32] <phillw> xnox: ping me when you have a few minutes to spare re: the 'out of sequence' upgrades, thanks.
[21:35] <xnox> phillw: i'm not sure what you mean by "out of sequence" upgrades.
[21:36] <phillw> well, 12.04 is not scheduled to be able to 'jump' to 13.04. It is expected to be able to move to 14.04 (LTS --> LTS)
[21:37] <xnox> phillw: yes.
[21:37] <xnox> phillw: technically one can jump from any point to any point. Just change sources and go for it.
[21:37] <xnox> phillw: we take care of keeping transitional packages between lts -> lts release and always have.
[21:38] <phillw> same with 11.10 --> 13,04
[21:39] <xnox> phillw: the only difference is that we will start testing clean & bug free lts -> current-rolling (to be lts) upgrades earlier. In the past we had continues integration testing of lts -> each of non-lts releases for QA purposes and to find and fix bugs.
[21:39] <xnox> in that respect nothing changes.
[21:39] <phillw> IIRC, it is the -d flag that would invoke a 'force' to 13.04 ?
[21:40] <xnox> phillw: 11.10 -> 13.04 is not the same. Although nothing technically prevents one from doing so (apt-get dist-upgrade will happily do it) we as a community do not support upgrade bugs in such scenarios as 13.04 may have dummy transitional packages removed since precise and hence some dependency resolutions may fail and require one manually add/remove packages.
[21:41] <xnox> phillw: I am not talking about "upgrade-manager / do-release-upgrade" those are such high-level / hand-holding tools that have walled gardens against silly things =)
[21:41] <phillw> xnox: what would you like me to test?
[21:42] <xnox> phillw: is this about in general or the specific request to find obsolete_conffiles?
[21:43] <xnox> to find obsolete_conffiles: ideally install default server & desktop seeds in lucid -> do-release-upgrade to precise and then do-release-upgrade to quantal and then do-release-upgrade -d to raring.
[21:43] <xnox> and then print with dpkg all obsolete conffiles.
[21:43] <xnox> dpkg-query -W -f='${Conffiles}\n' | grep obsolete
[21:44] <phillw> xnox: anything you would think would be of help. I do have the VM's, and can back the startup one up so if it all goes horribly wrong and it will not even boot, i can still access the logs :)
[21:44] <xnox> =))) nice.
[21:45] <xnox> lucid -> raring upgrade in three stages while having ubuntu-desktop & ubuntu-server installed would be awesome. (kubuntu-desktop as well maybe to catch those bugs)
[21:45] <phillw> xnox: I have guesfish installed, which means I can use http://libguestfs.org/virt-copy-out.1.html
[21:46] <phillw> It's the reason I squealled so loudly when it stopped working under lubuntu :)
[21:52] <phillw> xnox: lucid == 10.04?
[21:52] <xnox> phillw: correct.
[21:52] <xnox> phillw: i'm trying to catch things that existing for ever now and _still_ have not been fixed /o\
[21:53] <phillw> I'll have to grab the ISO. my earliest on my current system is oneric.
[21:53] <phillw> I'll grab it over night.
[21:54] <phillw> I'm guessing you want both desktop and server?
[21:55] <phillw> xnox: can I use alternate to install as you just want the details of updates?
[21:56] <xnox> phillw: yeap, netinst or alternate or server cd will be fine.
[21:57] <xnox> phillw: just make a single vm and install ubuntu-desktop and ubuntu-server packages and then do the tripple upgrade.
[22:02] <phillw> xnox: will do :)
[22:03] <phillw> s/single/ simple as I'll create one for desktop and one for server... they are quite different beasts :)
[22:05] <xnox> cool, thanks a lot =)
[22:17] <phillw> xnox: shame you're not familiar with KVM :)
[22:22] <phillw> I'm going to pull in the server edition into my dedi server, it has a 100 MB/s back bone. my home system is 512 kb/s (4 hours for an iso to download).
[22:23] <xnox> phillw: i use kvm daily to develop installer, I only have one production machine.
[22:23] <xnox> (i386/amd64 based that is)
[22:23] <xnox> and armhf installer is often different / slower.
[22:24] <phillw> xnox: I can give you access to a 10.04 server VM on the dedi server if you want to chase it on there?
[22:25] <xnox> phillw: nah, thanks =)
[22:25] <phillw> or even, just the ability to use a VM on there for anything you want
[22:25] <xnox> phillw: i'd rather fix all the obsolete conffiles, but I don't have bandwidth / close enough mirror to ugprade 3 releases of ubuntu-desktop/-server.
[22:26] <phillw> that's why I use the dedi server where I can.. It has 10 Tb / month bandwidth allowance, needless to say I've never gone near that :)
[22:27] <phillw> Just running desktop off it via ssh -X is painfully slow on my so called b/band link :/
[22:33] <phillw> xnox: that dedi makes me sooooo envious.... ubuntu-server 10.04.4 downloaded in 76 seconds.