[00:11] len_, if you answered in the last thirty minutes or so, my daughter closed xchat :/ [00:14] astraljava, the ubiquity installer background/text bug, i think i have an answer for that [00:15] this is per xubuntu: [00:16] http://bazaar.launchpad.net/~xubuntu-dev/ubuntu-seeds/xubuntu.precise/view/head:/live has the 'xubuntu-live-settings' package in the seed [00:17] http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/precise/xubuntu-default-settings/precise/files has /casper/48xubuntu_maybe_ubiquity [00:18] i believe we get both of these installed then we should fix this [00:21] micahg, TheMuso: do either of you know how i can make the -lowlatency kernel be installed on the image properly? should i ask in #ubuntu-devel? [00:21] i remember changing the seeds with persia to transition from the -rt kernel to the -generic, but that might have been that the -rt was just shipped but not installed [00:21] which makes sense, given perspective [00:22] ScottL: It was different with the alternate CD, let me look at the seeds for studio now. [00:30] thank you! very, very much :) [00:40] install (mar 14) failed same way as yesterday filed bug #955617 [00:40] Launchpad bug 955617 in ubiquity (Ubuntu) "ubiquity hangs (no activity forever) at configuring target system" [Undecided,New] https://launchpad.net/bugs/955617 [00:44] ScottL: Ok this is not something thag that needs changing in the seeds, not for a live DVD at any rate. The livecd-rootfs package needs updating to use the lowlatency kernel for studio. [00:44] TheMuso, should i ping cjwatson tomorrow morning? [00:44] ScottL: I don't have time to do that right now, as I am in the middle of something, but if I get around to it, and someone hasn't done it already, I will make the needed change. [00:45] TheMuso, thank you, i'll ping cjwatson in the morning if you haven't told me it's fixed, which is about 11 hours hence [01:18] ScottL, scott-upstairs, my main reason for trying todays iso was to file the bug report. I am not sure if the ubiquity problem is ours... I am not sure how to tell for that matter. [01:30] Comment on the webpage at staging. The download page mentions burning a DVD several times. There are getting to be more and more machines that have no dvd burner/reader. Making some mention of installing from a usb stick might be a good idea. [01:33] This ^^^ is talking about the download page. Says "Users must burn the ISO images to DVD". might add "or write to a USB memory stick" [01:43] len-nb, the ubiquity background/text issue appears to be a quantified problem, we just need to add some files to seeds and default-settings package [01:45] len-nb, i am presuming the uninstallable aspect is because of the kernel snafu (i.e. me changing the seeds for lowlatency headers), luke says he is looking at this when he has time, although i will poke cjwatson tomorrow morning if nothing is resolved by then [01:45] scott-upstairs, ok... I haven't had a background issue though. I can see all the text. I have been able to since you fixed the default session. [01:46] len-nb, good comments about the burn dvd comments, at the time we only had alternate which was a bit buggy with usb sticks [01:46] len-nb, i found that if you choose to install directly from the first menu then you see the ubiquity background/text problems [01:46] if you let the live dvd boot and then choose to install the problem doesn't manifest [01:46] Ah, ok. [01:50] scott-upstairs it looks like all the apps are there for this cycle now? [01:51] i think so, but i'm having second thoughts about digikam, it alone pulls in 375MB of dependencies, including quite a bit of qt libraries that even qjackctl or whatever else we have don't [01:53] now if i can just remember where the default-settings package actually sets the menu (by filename) :P [01:53] Can anything else do the job? I think that would be best for someone who uses linux for photo work to answer. [01:54] the menu file is /etc/xdg/xdg-ubuntustudio/menus/application.menu I think. [01:55] oh, i know where the file is, i'm just wondering where it is set [01:55] i think it's under xconf and elsewehere there, i just am not looking right now ;) [01:55] i will though after i move a few things and rename the file [01:55] scott-upstairs, xfce looks there first because that is where our session stuff is. [01:56] Are you thinking of the panel stuff? The menu is just one file. or do you mean the old one? [01:58] hmmm, you may be right then, but i really thought i remembered the menu being set somewhere [01:58] len-nb, does your cursor look like a hand when it hovers over a link? [02:02] I don't remember, sorry. I rebooted already, there is only so much I can do with a live iso [02:03] scott-upstairs, I'll look next time. [02:03] len-nb, if so, it isn't like xubuntu's cursor and i'm not sure why at this point but i want to fix it [02:08] ScottL: Yeah, I got that much from a later comment in the bug, as well as on #xubuntu-devel. I'll be working on that during the weekend. [02:08] scott-upstairs: ^^ [02:09] astraljava, working on which one? [02:09] * scott-upstairs really, really appreciates the help :-D [02:09] The maybe_ubiquity / live-settings thing. [02:10] outstanding! [02:10] thank you very much :D [02:11] scott-upstairs: Did you find any info from the bowels of your memory re: rtprio.py? [02:11] astraljava, sorry, i completely forgot about it, TBH [02:11] astraljava, i'll dig around later tonight after i make changes for the menu [02:17] scott-upstairs: No worries, I'll do a test install and see whether it's still used or not. [02:23] scott-upstairs, it would be possible (probably not a great idea) to have a different session for different work flows. with different menus... [02:32] scott-upstairs: Well, installing ubuntustudio-audio in a precise schroot tries to create an audio.conf file, so I'm assuming that means no more using rtprio.py, but it fails with calling postinst with an unknown argument `install', and like I mentioned earlier already, I can't even find the postinst script anywhere. [02:35] scott-upstairs: And I see why, now, cause in the -meta's debian/ directory, ubuntustudio-audio.install file, usage of rtprio.py is commented out, so at least one bug can be squished now. [02:38] micahg: Don't know if you missed it earlier, but I'd love some pointers to how the -meta source package is created, and subsequently on how the various binary packages are rolled from that. [02:38] TheMuso: You probably have lots of knowledge on that, too. [02:52] astraljava, yes, yes, yes. that reminds me! i couldn't find it except searching through launchpad and found what you described, and micahg commented it out but didn't remove it [02:53] astraljava, i'm glad to have that resolved (well, at least remembered) [04:00] astraljava: ScottL: it might need a preinst or prerm to remove the file beforehand, I was wondering about that [04:13] micahg: Right, but I don't know where to create one. So far, I'm seriously puzzled on where such files should be created. Like I said, I'd like some pointers to where the -meta source packages and corresponding binary packages creation is being explained. [04:13] well, that stuff never should've been in the meta in the first place, but the default settings package [04:14] Ok. But still, why is postinst called with unknown argument `install' when installing ubuntustudio-audio? [04:15] astraljava: idk, that usually points to bad "punctuation" in the file [04:15] micahg: I can start moving the audio.conf creation into -default-settings package during this coming weekend. [04:15] micahg: But which file? [04:16] there is no postinst anymore [04:16] micahg: Like I've repeatedly mentioned, I have no idea what file to correct, as I don't know where postinst lies at. There's none in -meta, and I don't even know how it gets created. [04:16] which version is this? [04:16] 0.97 [04:16] buggers [04:18] astraljava: where's the bug BTW? [04:19] micahg: There are several, hang on I'll list at least a few of them. [04:21] bug 954468, bug 946819, bug 946312 [04:21] Launchpad bug 954468 in ubuntustudio-meta (Ubuntu) "package ubuntustudio-audio (not installed) failed to install/upgrade: subprocess new pre-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/954468 [04:21] Launchpad bug 946819 in ubuntustudio-meta (Ubuntu) "package ubuntustudio-audio (not installed) failed to install/upgrade: subprocess new pre-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/946819 [04:21] Launchpad bug 946312 in ubuntustudio-meta (Ubuntu) "package ubuntustudio-audio (not installed) failed to install/upgrade: le sous-processus nouveau script pre-installation a retourné une erreur de sortie d'état 1" [Undecided,New] https://launchpad.net/bugs/946312 [04:22] The last one is a little misleading, but was revealed by looking into the dpkg log file attached. [04:23] Also bug 933722, which is filed against 0.96, but fails with the same problem. [04:23] Launchpad bug 933722 in ubuntustudio-meta (Ubuntu) "ubuntustudio-audio fails to install" [Undecided,New] https://launchpad.net/bugs/933722 [04:24] * micahg might have broke something accidentally [04:27] * micahg isn't sure, maybe TheMuso has some idea [04:31] Alright, well I could help too, if you could just briefly introduce me to the underlying fundamentals on -meta creation. I know that the seeds play a big part in it, but as to [pre|post][rm|inst]'s, I have no clue. [04:31] astraljava: well, those are the same as any other package, they're not meta specific [04:32] micahg: I understand, they would be binary package specific. But how is the source package created? [04:32] astraljava: same as any other source, but it's maintained as a source in the archive as opposed to bzr [04:33] micahg: Yes, and how do I tweak that, when I don't have access to the files in the archive? [04:33] the only thing that's special is the germinate part, otherwise, it's a normal package [04:33] astraljava: if you have internet access, you can apt-get source :) [04:34] micahg: Yes, but how/where do I upload a changed file? [04:34] astraljava: just add a debdiff to a bug and subscribe sponsors like anything else [04:34] micahg: Ok, thanks! NOW did I get the procedure on fixing those bugs. :) [04:35] micahg: It's just sometimes confusing, what with the UDD and all that. [04:35] I need to update the meta as well for ScottL's changes, but I'd prefer not to upload until the live image is switched to the lowlatency kernel [04:35] * micahg doesn't UDD if there's another option :) [04:36] micahg: Haha. :) Ok. Well, I'll look into -meta if I see no fixes on it before the weekend. [05:30] * CFHowlett Jack Bauer mode: DAMMIT [05:31] I was all set to stay with 10.04 until EOL. Now I see openshot/vlc/audacity and GIMP are coming up with serious upgrades... [05:34] CFHowlett: meh, it's only 1 year difference :) [05:34] gimp only went from 2.6.8 to 2.6.12 [05:35] micahg I might stay put until 12.04.1 [05:36] CFHowlett: always a wise idea [05:39] CFHowlett: BTW, audacity might be a good candidate for backports [05:40] if you wait 4 weeks, you can go strailght from precise -> lucid [05:40] micahg great minds ... Yeah, I just ran update and it's not there. I would definitely support backporting it. [05:40] CFHowlett: we'd just need install and run tests [05:41] micahg related question: y u no have openshot!? [05:42] Seriously, I know openshot in its present form was released AFTER 10.04. However, I find it much more intuitive and useful than pitivi or kino. [05:43] course, maybe that's just me... [05:43] CFHowlett: it's seeded, just need to be uploaded [05:43] :) [05:43] good to know. [05:43] umm, audacity 2.0 can't be backported as it need ffmpeg 0.6 [05:44] * backporteed to lucid [05:44] scheize! === saidinesh is now known as saidinesh5 [11:25] micahg, astraljava : we do not need the audio.conf file! [11:25] that was an older methodology for setting rtprio and memlock and nice settings [11:26] wait, wait....i just woke up, let me find the web page with this on there [11:27] i think the audio.conf file is necessary and is the new way, but jack creates it via debconf when jack is installed [11:27] (that is the part cjwatson was trying to fix in the preseeding) [11:27] (and BTW, jack copies the audio.conf.disabled file to the computer and debconf renames the file if answers correctly) [11:28] what we don't want is... [11:29] rtpito, memlock, and nice used to be set in /etc/security/limits.conf [11:30] which is what i believe rtprio.py is still trying to do [11:30] and if jack creates the /etc/security/limits.d/audio.conf then the system could be confused or unstable [11:31] especially if contradicting settings are used (i.e. 'rtprio 99' is set in limits.conf and 'rtprio 95' is set in audio.conf) [11:32] in my opinion (for what it is worth) i would completely remove rtprio.py and any associated install files, not just comment them out, to avoid any potential for creating an unstable system in the future [11:33] 'of course, feel free to comment the reasoning for removing rtpio.py :) [11:34] [11:34] a general observation.... [11:34] i believe that people who are serious (not necessarily even professionals) about recording music stay with an LTS release due to it's stability [11:35] it appears that CHWowlett would affirm this believe :) [12:13] ciao quadrispro , good morning :) [12:14] ehy ScottL, will you go to the next UDS? === ScottL changed the topic of #ubuntustudio-devel to: Ubuntu Studio Development Channel | Ubuntu Studio code: https://code.launchpad.net/ubuntust?gs first Sunday of each month [12:23] quadrispro, i believe i will be going, and you? [12:23] wow, not sure how i changed the topic and not really sure how to change it back to the full text :/ [12:29] yes, me too [12:30] I'm very active as upstream of laditools [12:30] great changes will come soon :) [14:02] janne, you might also look at this bug for the -rt privileges issue with audio.conf: https://bugs.launchpad.net/ubuntustudio/+bug/571688 [14:02] Launchpad bug 571688 in Ubuntu Studio "ubuntustudio-audio, ubuntustudio-controls and jackd installation scripts redundancies in 10.04" [Medium,In progress] [14:02] micahg: too ^^^ [14:03] doh, that should have been janne -> astraljava ^^^ [15:28] micahg: TheMuso : cjwatson states in the bug report that "ubiquity shouldn't hang", would it be a good idea to revert to the -generic kernel for now (which also allows us to keep testing other fixes) until ... [15:29] a) get ubiquity or livecd-rootfs package patched for lowlatency, or [15:29] scott-work: you haven't switched kernels yet, just the headers [15:29] b) can determine that the lowlatency kernel issue isn't causing ubiquity is hanging [15:29] idk if ubiquity even uses the headers or not [15:30] micahg: i suppose i should have said, "revert the headers to -generic" [15:31] * scott-work does realize that updating the meta's is not trivial [15:36] scott-work: I never uploaded the new meta, so it's just the seeds ;) [15:37] micahg: oh, good! hehe, i don't usually say that about updating meta [15:37] i'm just worried about getting a proper and functioning release [15:38] if we don't get the lowlatency kernel as default for 12.04 i'm okay with that, as long as we fix other issues and have a strong, stable, functional release :) [15:38] i would *like* to have the lowlatency kernel as well, but i don't feel comfortable about risking a (potential) LTS release for it however [15:39] * scott-work still needs to email the TB about LTS applications....so many things to do and follow up [15:48] scott-work: oh, I'm sorry, I was supposed to help you with a risk analysis of your packages [15:49] scott-work: basically, you want to get this branch lp:~ubuntu-security/ubuntu-cve-tracker/master/ and look and see if your packages historically have had lots of CVEs, the ones in universe, your team would be expected to provide diffs to the security team for [15:50] scott-work: graphical interface is here, idk if it has historical info though: http://people.canonical.com/~ubuntu-security/cve/ [18:05] if we have a chance to either use linux-latency or linux-lowlatency-pae, which should we use? [18:05] i know that the pae kernel shoudl allow 32 bit users to recognize more memory, but will there be undesirable consequences of this? === scott-work_ is now known as scott-work [18:08] going with linux-lowlatency [18:08] most new computer are 64 bit and i don't think we can adequate support everyone using an older computer [18:09] scott-work: I think you might want the pae kernel for i386 as studio is targetted to beefier machines [18:09] micahg: is there a downside to choosing pae? [18:10] the non-pae kernel on i386 means machines with >4GB RAM won't be able to address it [18:10] right, but is there a sacrifice to using the pae? [18:10] the downside to pae is older machines and older laptops netbooks won't be able to boot off the image [18:11] i wonder if we would start seeing bug reports of these machines not booting then :P [18:11] i'm still leaning towards just keeping linux-lowlatency , to be honest [18:11] scott-work: upgrades would work, it's just the live DVD won't [18:12] it depends on who your audience is [18:12] i believe our current kernel is not pae enabled, so in this regard we wouldn't be "changing" [18:12] people with more RAM just have an extra step to install the PAE kernel [18:12] BUT, we will be giving them more performance [18:12] hmmm [18:13] well, the default in 12.04 for i386 is the PAE kernel [18:13] oh, hmmm [18:13] but you have the choice here [18:13] okay, i'll tell cjwatson then [18:13] Xubuntu decided at the last meeting to take the non-pae kernel if it's easy enough and won't break the images [18:14] but xubuntu is targetting lower end systems [18:14] or rather it's one of their targets [18:14] I'm guessing Lubuntu will also try for non-PAE, but again, their target audience is lower end systems [18:15] okay, going with pae then [18:16] and worse case is you get a Xubuntu or Lubuntu ISO, install, then install the ubuntustudio-desktop package and switch kernels for the 0.01% of users who mimght care [18:17] scott-work: Ok, I'll look at that bug as well. [18:17] astraljava: which bug is that? (sorry, to many conversations going on) [18:17] scott-work: The one you linked to. [18:18] anybody familiar with iptables? [18:20] astraljava: oh yes, i remember now [18:20] knome: Just a little bit, why? What do you want to achieve? [18:20] block a subdomain [18:20] dd-wrt failed with that [18:20] Hmm... not sure, next! [18:20] the solution can be network-wide or computer-specific, i don't care. [18:21] I can look into it, if you want. [18:21] nah [18:21] astraljava: don't forget about the ubiquity background/text bug too, if you can. i would really like to get that one resolved before B2 if possible [18:21] it's not critical [18:23] scott-work: do you do a lot of bug triage? [18:23] why, oh why people think that subfolder == subdomain [18:23] >__< [18:24] scott-work: Yeah, I have that on the TODO for the upcoming weekend, or possibly tomorrow, if I don't have to work. [18:25] knome: Because most of them can't tell their right thumb from their left index finger? [18:25] yeah. it just makes googling for help soooo much harder. [18:29] micahg: not too much, usually only when it is pretty critical [18:29] bug reports is not something we are currently active within, unfortunately [18:31] scott-work: that's also a requirement for LTS status :) [18:31] watching bug reports [18:32] otherwise you won't know what to fix ;) [18:33] micahg: i have wanted to do that (watching bug reports more actively) and i think once two things happen we will be in a better position... [18:33] 1. get through 12.04 development [18:33] 2. fix the ubuntustudio-audio bug with rtprio.py causing installs not to update [18:33] scott-work: I don't mean you personally, but as a project they need to be watched [18:33] :-D [18:35] scott-work: rtprio.py doesn't seem to stop installs anymore, it's the "postinst called with unknown argument `install'". [18:35] micahg: See? Someone watches for the bugs. :) [18:36] micahg: right, i knew what you meant :) [18:36] astraljava: agreed [18:45] bloody hell, cjwatson is pretty awesome [18:46] scott-work: This comes as news to you? *blink* *blink* [18:46] there are many people in the community who are awesome. [18:53] when i met colin at last UDS i was struck by how gentle he is, but he has this incredible ability to get stuff done, "arcane stuff" as he has called it, that greatly affects ubuntu studio on a fundamental level [18:54] i kept notes on people as i met them, just little two or three work lines, just to help me remember the numerous people i met, i think i wrote "gentle red force majure" for him :P [18:54] i think it really helps a derivative if its people know the correct people in the community [18:54] lol [18:55] oh, absolutely! knowing the right people has been probably the major stumbling block of getting the lowlatency kernel [18:55] i had good luck when i was at uds, i got to know a few really cool guys [18:55] like mvo [18:55] and in london, like stgraber [18:55] it really helps if you know personally, or at least have met [18:56] "hey, do you know how to..." get MUCH more friendly [19:51] scott-work, re the background bug with ubiquity when not live. I think the fix is to use the same themes in the xfce session as in the ubuntustudio session. [19:55] This may be as simple as copying /etc/xdg/xdg-ubuntustudio/xfce4/ to /etc/xdg/xfce4 [19:56] Or even making it a link. [20:03] len_: i've been told that we need to create a package like the xubuntu-default-settings package, add it to the dvd-live seeds, and also add the /casper/48_maybe_ubiquity file [20:04] astraljava: is going to work on that today or tomorrow (i think that's what he said) [20:04] scott-work re pae kernel. my machine is petty old and has shown no issues with pae [20:04] len_: that's good to hear and reassures me with the decision :) [20:05] The only older machine I have (98 laptop) doesn't support enough memory for much audio work anyway. (p366) [20:07] scott-work, it's probably around 5 year old machines which can suffer if there is no non-pae kernel [20:07] I think there is a minimum machine that will support audio work with the software we ship. [20:07] mine is at least 8 yrs old. [20:08] some older machines do support pae, but not all [20:08] the problem is not (only) age-specific [20:10] I would guess that is true. However, by the time the DE is loaded and then jack and ardour and maybe hydrogen, other things than CPU like memory might be an issue anyway. [20:11] I am already seeing issues with my tiny 1G ram... [20:16] jack is not working great for me [20:16] Unable to stop it from qjackctl [20:16] Also, getting xruns easily with -lowlatency [20:16] I suspect it has something to do with dbus, but who knows [20:17] jackd1 is easier, but still getting xruns :/ [20:23] ailo, i had that with my ensoniq card, but ok with the envy24...but that was with PA-jack thrown in the mix. [20:23] my machines are 32bit [20:24] This is definately not a problem with the card, since I've never had problems before [20:25] -lowlatency for earlier releases were awesome [20:25] Might be a problem elsewhere [20:26] Its still pretty much black box like to me... I have a rough understanding of what happens in there, but not much more. [20:26] It's difficult to know without being able to analyze the problem [20:27] In this case experience will at least tell something [20:27] yup, been there. [20:28] I know that there has been some work under the hood for PA and alsa stuff, but who knows if it is related [20:28] Beyond the jack bridge thing [20:28] All I know is that currently, low latency audio is not working [20:29] 6 months ago, jackd + -lowlatency were awesome [20:29] So far mostly I have been doing install testing, the last audio test was 2 or 3 weeks ago now [20:30] In retrospect, it would have been good if I had put some time into testing. Been occupied with other things [20:30] We don't heat the space where my audio computer is and I have kids to watch too, so I have mostly been testing on my netbook. [20:31] The focus has been on the switch to xfce [20:31] So mostly look and feel. [20:52] I'm reporting a bug about qjackctl not being able to stop jackd2 (it happens often, but not everytime) [20:52] jackd1 is no problem [21:01] Here's the bug report, if anyone would like to test and confirm https://bugs.launchpad.net/ubuntu/+source/qjackctl/+bug/956438 [21:01] Launchpad bug 956438 in qjackctl (Ubuntu) "qjackctl unable to stop jackd2" [Undecided,New] [21:03] Now I would love to find out why xruns are rampant [21:23] Somehow qjackctl starts jackdbus and not jackd, but if you start jackd from the terminal, it's just jackd [21:27] scott-work: Please, pay attention (now why do I have to keep telling that to everyone?!), tomorrow or over the weekend! :) [21:33] That's all for today for me. Another day tomorrow [22:06] for what it is worth (and especially since micah hasn't update the meta files), i think i will remove digikam from the seeds after looking at the dependencies, including the surprising number of unrelated KDE applications http://paste.ubuntu.com/885578/ [22:11] That's install recommends by default for ya. [22:42] ScottL, you will need an UIFe for the slideshow changes. [23:36] knome, okay, thank you