=== Sarvatt_ is now known as Sarvatt
=== _LibertyZero is now known as LibertyZero
[07:17] <ppisati> morning
[08:19] <smb> morning
[08:21] <abogani> morning
[08:24] <ohsix> this can't be right ...
[08:24] <ohsix> [314951.125601] CPUFREQ: ondemand sampling_rate_max sysfs file is deprecated - used by: md5sum
[08:24] <ohsix> [317300.986027] process `md5sum' is using deprecated sysctl (syscall) net.ipv6.neigh.default.retrans_time; Use net.ipv6.neigh.default.retrans_time_ms instead.
[08:25] <ohsix> nm on the first one, it was actually reading /sys ;]
[08:51] <cjwatson> hmm, so regarding 3.0; the kernel team knows that 3.0.0 < 3.0 as far as dpkg is concerned, right?  we'll need to be careful when uploading the RCs ...
[08:52] <smb> Probably we would not have thought about that first. I guess there will also be all sorts of hardcoded 2.6 we will find in scripts...
[08:52] <_ruben> don't even wanna think what kind of packaging hell the 3.0 decision is causing in distro land :)
[08:52] <broder> wasn't that deliberately supposed to be part of the fun? :)
[08:52] <_ruben> quite likely ;)
[08:53] <cjwatson> you can use 3.0~3.0.0-<whatever> or similar, although I don't know what that will do to your scripts :-)
[08:54] <cjwatson> yeah, I already know of at least one bit of hardcoding in the installer
[08:54] <broder> wait, aren't they doing 3.0.0 now with the intent of switching to 3.<num>.<stable release> once everybody wakes up?
[08:55] <smb> cjwatson, I probably should add this in mail form not to forget... There are a lot of not workers today... Actually, isn't that a day off in the UK?
[08:55] <ppisati> smb: and usa too iirc
[08:55] <broder> yeah, today's a US holiday
[08:56] <smb> ppisati, I did know the US, I surprised a bit by the UK part. :)
[08:57] <ppisati> smb: i think apw told me that
[08:57] <ppisati> smb: but i could be worng, a lot things happe during a weekend that can help you forget :)
[08:57] <smb> ppisati, Yep. Me too. When telling me see you on Tuesday on Friday.
[08:58]  * smb plans to take off this Friday just to pay back... ;-P
[08:59] <ppisati> :)
[08:59] <ppisati> in italy we have a national holiday 2th of june
[08:59] <ppisati> but i'm not sure if i'll take it
[08:59] <ppisati> too many cve to fix
[09:00] <smb> ppisati, Have the same in Germany. And there is always many CVEs (and there will be after)
[09:00] <ppisati> yep
[09:01] <ppisati> bank holiday in uk
[09:01] <cjwatson> smb: yes, I'm not really working today
[09:01] <ppisati> memorial day in usa
[09:01] <cjwatson> doesn't stop me being on IRC now and then though :)
[09:01] <smb> ppisati, You know the saying: "nothing is so important that it cannot get more important by delaying a bit"? ;)
[09:02] <smb> cjwatson, :) Some people are crazy in a worrying way. ;)
[09:09] <smb> Hm, that probably came out more rude that intended. More tried to express that looking out of my window and if I had a free day, I could imagine so much nicer things than to be on irc...
[09:14]  * ppisati keeps hitting bugs that are marked as "Fix Released"
[09:19] <cjwatson> smb: not so much a free day as a housework day, for me
[09:21] <smb> cjwatson, I guess that changes things a bit (at least the order of appealance)
[09:21] <cjwatson> quite
[09:42] <awilkins> Is there some kind of NVRAM marker that the kernel writes to inform the BIOS that a boot occurred successfully? My BIOS keeps claiming that the last boot failed or the POST was interrupted, but only after I shut down Natty - it never did it for Maverick, and doesn't do it for Windows.
[10:19] <cdbs> Hey, any time frame for a kernel 3.0.0-rc1 release on the mainline ppa?
[10:20] <cdbs> I have an intel sandy bridge, and performance on 2.6.3x (even .39) is like that of a 2004 computer running Win7
[10:39] <smb> cdbs, It may need some effort getting that to run. Normally it should be picked up when the tag appears, but I suspect we will see some things breaking by that version change
[11:15] <soren> I'm having difficulties running the Oneiric kernel as a Xen dom0.
[11:16] <soren> Looking at http://wiki.xensource.com/xenwiki/XenParavirtOps there's a number of config options that need to be enable that aren't.
[11:16] <soren> Some of them are listed as =y, so they'll bloat the kernel by some amount.
[11:17] <soren> Is Xen dom0 something you do want to support, thought?
[11:18] <soren> -t
[11:22] <awilkins> Maybe they'd want to enable it on the server builds, or their VM server build (they have a special build for EC2, no?)
[11:32] <soren> Not dom0.
[11:33] <ogra_> soren, err, isnt that a mainline feature now anyway ?
[11:33] <soren> ogra_: The basic dom0 support, yes, but a number of extra options need to be enabled for it to actually *work*.
[11:33] <ogra_> well, what i read was that ll needed features are mainlined in 3.0
[11:34] <soren> ogra_: One wonders what the use is of a kernel with dom0 support enabled, but not all the other stuff, but meh.
[11:34] <ogra_> so why would someone disable them :)
[11:34] <soren> I don't know. I don't know why Kconfig even allows it.
[11:34]  * ogra_ thinks the oneiric kernel is just not up to date yet
[11:34] <soren> 2.6.39 has the relevant stuff, too.
[11:34] <soren> Just not all the right stuff *enabled*.
[11:35] <soren> At least that's my current working hypothesis.
[11:35] <soren> It's not booting, and that page I linked to above says that these things should be enabled, and they're not, so it seems like a good guess that that's why it's not booting for me.
[11:36] <soren> I'll know more once I get around to building a kernel with those things enabled.
[11:40] <soren> Oh, darn it, it actually works on this box.
[11:40] <soren> /ignore me
[13:22] <lool> smb: Oy
[13:22] <smb> lool, Sup?
[13:22] <lool> smb: Would you mind cherry-picking "UBUNTU: Include nls_iso8859-1 for virtual images" to natty and maverick?
[13:23] <lool> fs/nls/nls_cp437.ko was picked in the natty but not iso8859-1
[13:23] <smb> lool, Thought we sort of had. But I might confuse it or its just stuck in proposed... 
[13:23] <apw> cjwatson, thanks for the heads up on 3.0
[13:23] <lool> smb: looking at tip of natty.git, it's still missing iso8859-1
[13:23] <lool> I'm still checking maverick
[13:23] <apw> smb, yes holiday today
[13:24] <lool> (slooowwwww checkout over ecryptfs)
[13:24] <smb> apw, And another one who cannot stay without irc. :)
[13:24] <lool> smb: oh are you on leave too?
[13:24] <lool> sorry about that
[13:24] <lool> I saw Thursday is a big leave in Europe
[13:24] <smb> smb, Nah, /me is Germany. 
[13:24] <apw> smb yep
[13:24] <smb> No holiday today. Only UK
[13:24] <smb> and US
[13:25] <lool> smb: right, maverick also only has nls_cp437 and not nls_iso8859-1
[13:25] <smb> lool, for maverick it seems to be in master-next
[13:25] <lool> smb: also, will there actually be some SRU of linux in maverick and natty still?  I don't know how long the non-LTS releases get kernel SRUs
[13:25] <lool> smb: ahhh master-next gah
[13:25] <smb> lool, 79cc10ef1158acc18688a0594a7fcda9921a11ce
[13:26] <lool> smb: I see it in natty msater-next as well
[13:26] <lool> and in maverick master-next too
[13:26] <lool> I had completely forgotten about this master-next branch
[13:26] <smb> lool, Yup, so just paaaatience. :)
[13:27] <lool> smb: is there a set cadence for the SRUs in non-LTS as well?
[13:27] <smb> lool, Yes there is a cadence
[13:28] <smb> But for Maverick, we just found some problem with that and ec2
[13:28] <lool> I don't need it urgently, but we have a Linaro engineer who is setting up jenkins to start instances to do builds with a fixed kernel, and it's not trivial to arrange to upgrade the kernel and reboot (new ABI!) before kicking the build
[13:28] <lool> smb: ah, but not due to these patches I would hope?
[13:30] <smb> lool, No there was some Xen change that has unexpected results for us
[13:30] <lool> Ok
[13:31] <lool> I've told her to build the .ko locally with the old ABI and wget them on boot now
[13:31] <lool> thanks for confirming the commits are in there!  I thought the fix had fallen through the cracks
[13:31] <lool> smb: maybe I should reset the bug tasks to work in progress or triaged until there's an upload to -proposed?
[13:32] <lool> (they are fix committed which is usually the bug state after a -proposed upload happened)
[13:33] <smb> lool, Well there are uploads to proposed, just not what is in master-next necessarily. And no, fix committed always was just commited to repo. Not necessarily uploaded anywhere
[13:33] <smb> At least the way we had been using that state
[13:34] <lil_pete> hey guys i want to boot a custom kernel on 10.10, but during boot it ignores my /boot/grub/menu.lst 's 1st entry (--> custom kernel)... am i missing something here? 
[13:35] <jk-> lil_pete: I don't think menu.lst is used anymore
[13:35] <smb> lil_pete, menu.list is only used with grub1, grub2 has grub.cfg
[13:35] <lool> smb: Ok; I was suspecting that for kernel the state was slightly different; the SRU processing scripts usually set fix committed when the package is accepted into -proposed
[13:36] <lil_pete> jk-: erm... okay. so i have to edit grub.cfg? 
[13:36] <lil_pete> "do not edit this file" lol 
[13:37] <jk-> lil_pete: would be better to add a /etc/grub.d/nn-my-custom-kernel
[13:38] <smb> Yes, it is completely regenerated, so it might not be very persistent. Depending how "custom" the kernel is (ie not using the debian build scripting) you probably have to do like jk- said
[13:39] <lil_pete> jk-: examples anywhere? /etc/grub.d/* looks like scripts finding everything... strange it doesnt find my kernel though i copied it to /boot/
[13:40] <lil_pete> smb: im not sure about the "customness" :) got the sources from kernel.org, did a configure + make
[13:40] <jk-> lil_pete: I'm no grub expert, but it looks like you just create a shell script that outputs a 'menuentry' stanza
[13:41] <smb> lil_pete, That probably is custom enough. You would have to look at the existing parts in grub.d, but it may be that is looks for system map or other parts than just a kernel image to detect it as a bootable kernel.
[13:41] <jk-> the 10_linux script will pick up your kernel if you've named it according to a specific pattern though
[13:42] <jk-> list=`for i in /boot/vmlinuz-* /boot/vmlinux-* /vmlinuz-* /vmlinux-* ;
[13:42] <lil_pete> i want my oldschool-grub back :(
[13:42] <lil_pete> got a "/boot/vmlinuz-2.6.39"
[13:42] <lil_pete> i _think_ that should be recognized?! 
[13:43] <jk-> did you do a `sudo update-grub` ?
[13:43] <lil_pete> jk-: ...
[13:43]  * lil_pete is the dumbest a$$ in this channel.
[13:44] <lil_pete> i only did a grub-mkconfig
[13:44] <jk-> :D
[13:44] <lil_pete> ill be back in one 2.6.39 boot time :D hopefully... :) 
[13:47] <smb> hopefully he did not need to run update-initramfs, too
[13:55] <ppisati> CVE 2010-4655
[13:55] <ubot2> ppisati: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem.  When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2010-4655)
[13:55] <ppisati> looks like there's a duplicare
[13:55] <ppisati> duplicate
[13:56] <ppisati> lp#771445 is the right one
[13:56] <ppisati> lp#771442 is untouched
[13:56] <ppisati> lp#771445
[13:56] <ppisati> lp771445
[13:56] <smb> bug 771445
[13:56] <ubot2> Launchpad bug 771445 in linux-ti-omap4 "CVE-2010-4655" [Undecided,In progress] https://launchpad.net/bugs/771445
[13:57] <ppisati> bug 771442
[13:57] <ubot2> Launchpad bug 771442 in linux-ti-omap4 "CVE-2010-4655" [Undecided,New] https://launchpad.net/bugs/771442
[13:59] <smb> ppisati, At least the titles look duplicate. Depends a bit what/who opened them. Could it be a slight issue when doing some automatic updates...?
[13:59] <ppisati> smb: both leann
[13:59] <lil_pete> well... it does load my kernel, but it doesnt boot. ill figure that out. thx for the advice guys. :) 
[14:00] <smb> lil_pete, I was wondering whether you had installed all the modules and also run update-initramfs to generate an initrd...
[14:00] <lil_pete> nope
[14:01] <lil_pete> im trying make-kpkg now
[14:01] <smb> ppisati, I would tend to mark the one without anything done to it as duplicate of the other. You may wait till tomorrow to check back with Leann though
[14:02] <ogasawara> ppisati: I can't remember why I would have opened two bugs, I'd mark the untouched one as a duplicate of the other.
[14:02] <smb> Argh, yet another holiday zombie. :)
[14:03] <ogasawara> smb: I actually swapped today's holiday for last friday, so I'm workin today
[14:03] <ppisati> ogasawara: ok
[14:03] <ogasawara> bah, be back in a bit, lil guy woke up early
[14:03] <smb> ogasawara, Ahh. :) I was becoming a bit worried that I don't spend my holidays on irc. :-P
[14:04] <lil_pete> smb: would make modules-install install all i need? 
[14:04] <smb> lil_pete, It would install the modules, but not generate any initrd
[14:05] <lil_pete> hhmmm no rule for modules-install ? 
[14:05] <smb> was that modules_install 
[14:06]  * smb has not used it for a while
[14:06] <qwerty_> hii
[14:06] <qwerty_> i am trying to compile the latest version of ubuntu kernel
[14:07] <qwerty_> i've come across some messages "section mismatch "
[14:07] <lil_pete> smb: hehe ill try
[14:07] <lil_pete> hitting tab takes forever... :-/ 
[14:08] <lil_pete> smb: make modules_install it is. :) 
[14:11] <qwerty_> hii
[14:12] <lil_pete> smb: update-initramfs --> anything i should be looking out for? 
[14:13] <smb> lil_pete, Normally "sudo update-initramfs -u <whatever the dir for you modules is named in /lib/modules>
[14:19] <lil_pete> smb: seemed to work without that dir (that got me an error message), i had to pass -k 2.6.39 though. anyway... ill reboot. this is gonna be really thrilling :D
[14:19] <smb> lil_pete, Oh, yeah, sorry -k and the version or dir name
[14:20] <lil_pete> np :) 
[14:20] <smb> Comes if one does not call things too often by hand
[14:22] <lil_pete> smb: i got nearly everything in /boot now, except "abi-2.6..." do i need that? 
[14:23] <smb> lil_pete, I do not believe this is strictly needed
[14:25] <lil_pete> alright, ill try rebooting... cya in a min. :) 
[14:34] <lil_pete> smb: im up and running now, thx. :) 
[14:34] <smb> lil_pete, welcome :)
[14:34] <ogasawara> whoa, looks like Oneiric will be 3.0
[14:35] <smb> ogasawara, Yep, surprise! :)
[14:35]  * ogasawara wonders how many of our scripts will be messed up
[14:35]  * _ruben feels sorry for you guys... 
[14:35] <smb> Me too... There will be quite a few I expect...
[14:37] <smb> And there might be packaging issues. Though maybe not too bad as 3.0.0 seems to be < 3.0 ... but then we have stable/longterm...
[14:38] <ogra_> lovely, so i will be able to ask the kernel team all my perl questions after this cycle !
[14:39] <_ruben> haha
[14:39] <smb> ogra_, We'll rewrite everything in python... ;-P
[14:40] <ogra_> lol
[14:40] <ogra_> to speed up the buildtime, right ? :)
[14:40] <smb> Hehe, of course. 
[14:56] <vmlinuz> all, I've updated my Lucid to the latest kernel updates and it's now freezing my system. How can I start to troubleshoot and gather information to file a bug?
[15:02] <Tommeh> Regarding the packages here: http://kernel.ubuntu.com/~kernel-ppa/
[15:03] <Tommeh> If only some packages appear in a given directory (i.e. some are missing), is it because the kernel failed to build/was pulled/hasn't been done yet, or something else?
[15:05] <smb> Tommeh, Most likely because a build failed. 
[15:07] <vmlinuz> how can I start troubleshooting a system freeze on Lucid?
[15:08] <Tommeh> vmlinuz, is it just a graphical freeze? 
[15:08] <Tommeh> Or total system
[15:08] <smb> vmlinuz, probably looking into /var/log/syslog whether there is anything. Trying to ssh into it to see what Tommeh said
[15:08] <Tommeh> Of course, the most obvious thing to do would be to see if anything was logged up until the point where the system froze.
[15:08] <vmlinuz> Tommeh: I'm not sure if it's only a graphical freeze, because I'm not able to switch to text/console
[15:08] <vmlinuz> Tommeh: think this is a total freeze
[15:09] <vmlinuz> smb: right
[15:09] <Tommeh> Ok. As said, it's worth checking to see if ports/services are still available over the network. Alternatively it's also worth testing the shutdown button to see if the machine shuts down/suspends when pressed.
[15:10] <Tommeh> Also vmlinuz, you don't say at what point after boot-up the machine freezes?
[15:10] <vmlinuz> Tommeh: I don't know exactly.. random freeze
[15:10] <Tommeh> Before logging in/after ?
[15:10] <Tommeh> Or both?
[15:10] <vmlinuz> Tommeh: after loggin
[15:10] <Tommeh> Ok
[15:10] <Tommeh> Are you using compositing?
[15:11] <Tommeh> Compiz/unity, etc.
[15:11] <vmlinuz> Tommeh: compiz
[15:11] <Tommeh> I can't remember if there's an option on the lucid login screen to disable effects -- try that.
[15:11] <Tommeh> If that fixes it, you can be fairly sure it's related to your GPU/mesa in some way.
[15:12] <vmlinuz> Tommeh: makes sense.. I'll try to login without 3D effects
[15:12] <vmlinuz> Tommeh: thanks a lot!! Appreciated
[15:12] <Tommeh> Graphics stack in Lucid isn't particularly up to date, tbh.
[15:12] <vmlinuz> /var/log/syslog still saves the crash messages even after I reboot?
[15:14] <vmlinuz> brb :)
[15:20] <vmlinuz> Tommeh: I'm without composite (selected Failsafe GNOME on GDM screen)
[15:20] <vmlinuz> Tommeh: let's see if it doesn't freeze anymore :)
[17:42] <Tommeh> vmlinuz, any luck?
[17:42] <vmlinuz> Tommeh: yes, you were right... after disabling composite, it's working stable now
[17:43] <vmlinuz> Tommeh: I need to find a way to start compiz, compositing in debug mode to gather more info to file a bug
[17:43] <Tommeh> Mmm.. What GPU do you have?
[17:44] <Tommeh> Or rather, which driver are you using?
[17:44] <Tommeh> I'd suggest seeing if something like openarea causes the same lockups.
[17:45] <vmlinuz> Tommeh: 'lspci' says "00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02)"
[17:50] <Tommeh> Weird
[17:50] <vmlinuz> Tommeh: my '/var/log/Xorg.0.log' says  `LoadModule: "intel"`
[17:50] <Tommeh> Well, it's Intel-based anyway, so that narrows the potential drivers to 1 :)
[17:50] <Tommeh> If you find that GL apps consistently crash, you might have more luck in #intel-gfx (I think that's the right channel)
[17:51] <vmlinuz> Tommeh: thanks, appreciated your help!
[17:52] <vmlinuz> Tommeh: now, it's confirmed it has nothing to do with kernel update itself, but with composite/driver
=== elmo__ is now known as elmo
[19:26] <akgraner> apw is there a wiki page that walks someone through updating a kernel
[19:27] <akgraner> or would someone for the team be willing to pop into #ubuntu-locoteams to answer that question?
[19:29] <cjwatson> 14:37 <smb> And there might be packaging issues. Though maybe not too bad as 3.0.0 seems to be < 3.0 ... but then we have stable/longterm...
[19:29] <cjwatson> smb: was that a typo?
[19:29] <cjwatson> as I mentioned, 3.0.0 > 3.0 :-)
[19:58] <ogasawara> May 30 05/30/11:00:51:00 <cjwatson>     hmm, so regarding 3.0; the kernel team knows that 3.0.0 < 3.0 as far as dpkg is concerned, right?  we'll need to be careful when uploading the RCs ...
[19:58] <ogasawara> cjwatson: ^^
[19:59] <ogasawara> cjwatson: I just want to make sure I have it clear in my head, 3.0.0 > 3.0 
[20:00]  * ogasawara plans to discuss our versioning with apw tomorrow
[20:12] <cjwatson> I'm sorry, I did mean > in the line you quoted above
[20:12] <cjwatson> i.e. that if you upload 3.0.0-1 (say) then you can't subsequently upload 3.0-1
[20:12] <cjwatson> or 3.0-1000 for that matter
[20:13] <ogasawara> cjwatson: right, that makes sense to me
[20:13] <cjwatson> just jumped out at me as a really easy way to shoot oneself in the foot in this situation
[20:14] <ogasawara> cjwatson: indeed
=== yofel_ is now known as yofel
[21:02] <dupondje> Hi ! Any plans for an updated kernel on Natty ?
=== soren_ is now known as soren
=== maxb_ is now known as maxb
=== em_ is now known as em