=== BenC__ is now known as BenC
=== bigbash is now known as zz_bigbash
=== jj-afk is now known as jjohansen
[06:37] <_ruben> jk-: you gave the hint on ls -l /sys/classes/ata_device .. after upgrading the box to maverick and then natty, that became available, but i'm still not seeing how i'd be able to figure out which disk is on ata4?
[06:38]  * jk- looks
[06:40] <_ruben> (the upgrades themselves were tricky, plain -server install on a 2G flashdisk, 1.5G /, hadda remove certain packages to free enough space :P)
[06:40] <_ruben> well, plain + contaminations over time ;)
[06:46] <jk-> hm, I had assumed the 'id' file would contain something sensible
[07:01] <_ruben> jk-: it might, for those fluent in hex ;)
[07:07] <jk-> _ruben: oh, mine is all zeros here
[07:08] <_ruben> jk-: mine's non-zero for rougly 30$
[07:08] <_ruben> 30%
[07:09] <jk-> _ruben: could be a good question for askubuntu.com
[07:09] <_ruben> good idea, will post there
=== smb` is now known as smb
[07:36] <janimo> jjohansen, apw is the last section of https://wiki.ubuntu.com/Kernel/Dev/KernelGitGuide uptodate?
[07:36] <janimo> I'd like to import a new tree based on 2.6.38 and I am not sure if cloning the oneiric ubuntu tree is recommended or not
[07:37] <janimo> in order to save space
[07:37] <janimo> this is the linux-ac100 tree based on an upstream that is not on kernel.org 
[07:42] <apw> janimo, if the repo is going to just be a clone of mainline, then you might want to clone from the linux repo instead of ubuntu
[07:42] <apw> (there is a local mirror of that in the ubuntu/ directory too)
[07:42] <janimo> apw, not a clone of mainline, but of a gitorious based fork of the chromeos kernel
[07:43] <apw> janimo, the nearest base would be linus there i assume, not ours as we have junk on top
[07:43]  * apw makes sure our linus is up to date ... is now
[07:45] <janimo> so clone the bare repo off /srv/kernel.ubuntu.com/git/ubuntu/linux-2.6/ ?
[08:06] <jjohansen> janimo: where is the clone going to live?
[08:07] <janimo> jjohansen, in git/jani/ubuntu-ac100 . I cloned and moved it there already
[08:07] <janimo> just not pushed anything to it yet
[08:08] <jjohansen> janimo: ah okay, yeah I would clone off of something close in /srv
[08:13]  * jjohansen is tired so night *
[09:56] <tkamppeter> I have an old 32-bit HP Compaq 12-inch laptop and it does not boot with the 3.0.0-12 kernel, but only with the last Natty kernel. How to report a bug/debug this?
[11:25] <smb> tkamppeter, I suppose you still get to the grub screen where you could try to replace "quiet" with "debug" and remove the "splash" to potentially get a bit more output before it gets stuck. Then I would open the bug with "ubuntu-bug linux" from the last working natty boot and edit the description to say it is not booting with 3.0
[11:26] <smb> If you can get any messages from the failed boot, those should get added, too
[11:51] <tkamppeter> smb, thank I will try. In which files kernel boot messages get logged? And are they conserved if I boot a Natty kernel after the crash?
[11:54] <smb> tkamppeter, This depends on how far you actually get in the boot. There may be a good chance that it is not far enough (otherwise those would go into /var/log/syslog). If there is nothing but something on the screen, take a picture and attach the jpeg...
[12:03] <Kano> hi, will 3.1-rc9 build soon?
[12:03] <Kano> also i would like to know when 3.0.6 wil be merged to oneirc (+3 dvb patches)
[12:04] <tkamppeter> smb, it seems not to be able to mount the root file system. It says something like
[12:05] <tkamppeter> smb, Kernel Panic - Not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
[12:06] <tkamppeter> smb, Pid: , comm: swapper Not Tainted 3.0.0-12-generic #19-Ubuntu
[12:06] <tkamppeter> smb, sorry, Pid: 1, ...
[12:06] <smb> tkamppeter, Hm, does this actually happen quite quickly? I begin to think there may be something wrong with the initrd file
[12:07] <tkamppeter> smb, yes, it appears immediately.
[12:08] <tkamppeter> So I have the Grub menu, press 'b' after changing to the debug settings, and the crash output immediately shows up.
[12:08] <smb> tkamppeter, When you boot back into natty and check /boot, is there a initrd-3.0.0-12-generic or so
[12:08] <tkamppeter> The Natty kernel (with Oneiric as user space) bootes perfectly.
[12:09] <apw> commit 512228f0be3af44bf5cf6cc5750ddd279bbedaf3
[12:09] <apw> Author: Andi Kleen <ak@linux.intel.com>
[12:09] <smb> tkamppeter, Also could it be you got /boot on a different fs (which may be full, thus preventing to have a valid new initrd)
[12:09] <apw> Date:   Fri Aug 19 16:15:10 2011 -0700
[12:09] <apw>     Add a personality to report 2.6.x version numbers
[12:09] <apw> uggg
[12:09] <tkamppeter> till@till-laptop:~$ ls -l /boot/initrd.img-3.0.0-12-generic 
[12:09] <tkamppeter> -rw-r--r-- 1 root root 13585679 2011-10-03 11:10 /boot/initrd.img-3.0.0-12-generic
[12:09] <tkamppeter> till@till-laptop:~$ 
[12:10] <tkamppeter> smb ^is it a problem that there is only an "initrd.img" file?
[12:10] <smb> tkamppeter, No, the filename looks ok
[12:10] <smb> (just remembered it wrongly)
[12:11] <tkamppeter> smb, no, my /boot is on the root partition and has 2.5G free.
[12:11] <tkamppeter> till@till-laptop:~$ df -h /boot
[12:11] <tkamppeter> Filesystem            Size  Used Avail Use% Mounted on
[12:11] <tkamppeter> /dev/sda7              13G  9.0G  2.5G  79% /
[12:11] <tkamppeter> till@till-laptop:~$ 
[12:11] <smb> tkamppeter, ok... One thing to try: boot natty and then
[12:12] <smb> sudo update-initramfs -u -k 3.0.0-12-generic
[12:12] <smb> (I hope)
[12:12] <smb> and after that a
[12:12] <smb> sudo update-grub
[12:13] <tkamppeter> OK, did it.
[12:13] <smb> tkamppeter, Anything sounded like an error message?
[12:13] <tkamppeter> But now I have a debconf menu.
[12:13] <tkamppeter> 1. install the package maintainer's version
[12:13] <apw> tkamppeter, when you have a prompt find out which driver is running your disks
[12:13] <tkamppeter> 2. keep the local version currently installed
[12:14] <tkamppeter> 3. show a side-by-side difference between the versions
[12:14] <apw> tkamppeter, does it tell you what this version is of ?
[12:14] <smb> apw, Sounds rather like a modified /etc/default/grub
[12:14] <apw> yeah could be
[12:14] <tkamppeter> 4. show a 3-way difference between available versions
[12:14] <tkamppeter> 5. do a 3-way merge between available versions (experimental)
[12:15] <tkamppeter> 6. start a new shell to examine the situation
[12:15] <smb> tkamppeter, I'd probably look at 3 and see what is changed
[12:16] <smb> apw, Just a bit strange that this did not happen on any upgrade before...
[12:17] <tkamppeter> apw, smb: It has only two "initrd /boot/initrd.img-3.0.0-12-generic" lines added at two points.
[12:18] <apw> then its not /etc/default/grub
[12:18] <smb> tkamppeter, Hrrrm, are you by chance still using grub1
[12:18] <apw> grub1 ?!?!
[12:18] <smb> apw, If you upgraded only for a long time grub2 is not enforced
[12:18] <apw> gurgle
[12:19] <tkamppeter> smb, this is possible. The computer is 6 years old and I have never done a fresh install, only updates. It must have gone through 12 Ubuntus.
[12:19] <ogra_> iirc there is a wiki howto about manually upgrading to grub2 for such cases
[12:19] <tkamppeter> So I need to upgrade to Grub2, how do I do this?
[12:20] <smb> tkamppeter, And if you use grub one, better say take maintainers version
[12:20] <smb> (cause if one once says stick with mine, then you stick forever...)
[12:20] <ogra_> https://help.ubuntu.com/community/Grub2#Upgrading_to_GRUB_2_From_GRUB
[12:21] <tkamppeter> smb, I have done so. No more prompts, should I try to reboot?
[12:21] <tkamppeter> ogra_, thanks.
[12:21] <smb> tkamppeter, You could do a quick visual check of /boot/grub/menu.cfg
[12:21] <smb> whether it looks sane
[12:22] <smb> But I guess your problem where those two missing initrd lines
[12:27] <tkamppeter> smb, i have no /boot/grub/menu.cfg
[12:27] <smb> tkamppeter, So yes, for the future, you may want to consider upgrading to grub2. Even if that means learning a new configuration language...
[12:28] <smb> tkamppeter, Something starting with menu? Oh, maybe .lst
[12:28]  * smb ususally types menu.<tab>...
[12:29] <tkamppeter> ok, looks nice, contains many kernels.
[12:29] <smb> ... grub2 is grub.cfg, so I likely mixed them up
[12:29] <smb> tkamppeter, Then try a boot
[12:30] <tkamppeter> smb, I am doing so and it seems that it works.
[12:31] <smb> tkamppeter, Good, if uname -a of the booted kernel says 3.0.0-12-generic you are good
[12:32] <smb> So it seems for some reason you had the kernel lines but not the initrd ones. 
[12:34] <tkamppeter> smb, it booted the new kernel but the user space of Ubuntu 6.10.
[12:34] <smb> 6.10 ???
[12:34]  * smb wonders how exactly tkamppeter does updates...
[12:36] <tkamppeter> It probably booted a wrong partition. Didi not know that I had such an old partition around.
[12:36] <apw> is that _edgy_ ?
[12:37] <tkamppeter> Can be, was the first version where I worked at Ubuntu. I started mid-2006.
[12:37] <tkamppeter> Natty kernel boots Oneiric.
[12:37] <smb> If I remember the magics of grub1 correctly then, there is a commented out line in menu.lst which says root=<something> which is used as a template for all of the grub entries later...
[12:38] <smb> Though I thought that update-grub would take that and make them all boot the same partition somehow. 
[12:38] <smb> Though that could be different for anything below a certain "not automatically done" marker somehow
[12:39] <smb> But I have not really looked closely for a long time
[12:40]  * smb thinks maybe one of the magic Colin's may remember out of their head
[12:40] <tkamppeter> When I go through the boot menu entries and press e for edit on them, both Oneiric and Natty kernels have roo=(0,6).
[12:40] <apw> ouch
[12:41] <smb> apw, Yes that were those dark times without uuid
[12:42] <tkamppeter> ## default grub root device
[12:42] <tkamppeter> ## e.g. groot=(hd0,0)
[12:42] <tkamppeter> # groot=(hd0,6)
[12:43] <tkamppeter> And hd0,6 is the edgy partition.
[12:43] <smb> Hm, right, so one (well at least me) would expect all boot lines to use 0,6 now... 
[12:44] <ogra_> apw, 6.10 is dapper (the only 04 release that was delayed ;) )
[12:44] <tkamppeter> The right one is /dev/sda7
[12:44] <apw> ogra_, na that was 6.06
[12:44] <ogra_> hrm, we only had one 06 releaase ever
[12:45] <ogra_> 06.xx i should say
[12:45] <tkamppeter> Should I try s/(hd0,6)/(hd0,7)/
[12:46] <ogra_> oh, heh, ignore me
[12:46]  * ogra_ needs to not read that RTL
[12:46] <tkamppeter> apw, smb: ^^
[12:47] <tgardner> ogasawara, this 3.0.6 stable update is giving me some weird crashes on my HP mini-10
[12:47] <smb> tkamppeter, I probably would try that on a single boot instance first, then change the groot line and run update-grub again
[12:47] <smb> tkamppeter, After you tested the single changed one
[12:48] <smb> tkamppeter, It probably asks you then again whether you want to keep your version or take the new one (which then again has to be answered with "the maintainers version2
[12:50] <tkamppeter> smb, too late, now its broken down, as this old box does not contain anything important any more and has space waisted by unneeded partitions, I should do a from-scratch installation on it.
[12:51] <tkamppeter> smb, instead of trying to break Update records. 5 years are enough.
[12:51] <tkamppeter> Buth thanks anyway.
[12:52] <smb> tkamppeter, Well, at least there was not kernel problem... 
[13:05] <ogasawara> tgardner: hrm, I've been running it on my adamo and not seeing any issues.  I'll put it on my mini and see if I can reproduce.
[13:08] <tgardner> ogasawara, look through your dmesg closely. I've gotten some random oops that weren't reported right away.
[13:11] <apw> ogra_, 
[13:11] <ogra_> thats me !
[13:11] <apw> stupid paste
[13:11] <apw> there were two in 06
[13:11] <apw> dapper and edgy
[13:11] <ogra_> yeah, i read it RTL :)
[13:12]  * ogra_ must have catches a chinese bug or so
[13:12] <ogra_> *catched
[13:12] <apw> heh
[13:21] <apw> tgardner, hey did you push your hardy tree last night, there seems to be a CVE missing from it
[13:21] <tgardner> apw, Everything up-to-date
[13:21] <tgardner> whats missing ?
[13:21] <apw> tgardner, ahhh my fault, i see why its gone wrong
[13:22] <apw> just not told the tracker enough things
[13:53] <tgardner> ogasawara, rebased P to v3.1-rc9
[13:53] <ogasawara> tgardner: awesome thanks
[14:21] <sconklin> tgardner, I think I understand the versioning stuff now, and I've moved past tilting at that windmill, so we'll move on to fixing the tools.
[14:22] <tgardner> sconklin, good. I'll be bisecting the stable update before I publish it in order to find the regression oops I'm having.
[14:24] <ogasawara> tgardner: my mini panics on boot and hangs, looks to be somethings with the i915 driver for me
[14:24] <ogasawara> tgardner: am going to bisect as well
[14:24] <tgardner> ogasawara, so you've got something to keep you busy today :)
[14:25] <ogasawara> tgardner: indeed
[14:25]  * apw is glad its not uploaded :)
[14:40]  * ogasawara back in 20
[15:20] <lamont> http://pastebin.ubuntu.com/702808/ <-- which part of this is surprising (if any...): i915 vs oneiric
[15:21] <tgardner> apw, ogasawara: ^^ more fun with i915
[15:22] <lamont> that system is rather memory constrained, currently has a frozen mouse on a black screen, and thought it was running apport to report lpstat dying on sigsegv
[15:25] <lamont> and it looks like apport didn't manage to actually report the segv
=== yofel_ is now known as yofel
[17:24] <cyphermox> hi, could someone please take a look at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/867679 ; see if I'm wrong in my analysis ?
[17:24] <ubot2> Launchpad bug 867679 in linux "Network fails after suspend/resume" [Undecided,Confirmed]
[17:31] <ogasawara> apw, tgardner: bisected the panic and hang on boot with my HP-mini
[17:31] <ogasawara> commit c17bcfb52394745dd9100e042031a1eab8442f9c
[17:31] <ogasawara> Author: Andy Whitcroft <apw@canonical.com>
[17:31] <ogasawara> Date:   Fri Sep 23 16:50:27 2011 +0100
[17:31] <ogasawara>     UBUNTU: SAUCE: acpi_init: move to full RCU free for IO mappings
[17:31] <ogasawara> reverting that fixes it
[17:32] <tgardner> ogasawara, tahts about the same conclusion I was coming to
[17:34] <tgardner> ogasawara, since it was purely for performance I vote we revert it and make apw work on it a bit more
[17:34] <ogasawara> tgardner: +1
[17:35] <tgardner> ogasawara, actually, since its still in master-next we could just rebase it out of existence
[17:35] <ogasawara> tgardner: that works for me too
[17:38] <tgardner> sconklin, did you ever get versioning working so I can create a stable tracking bug for oneiric ?
[17:39] <apw> tgardner, ogasawara, yep one reason i was happy it got a bit more testing before upload
[17:40] <apw> concur with rebasing it out of the history, its never been uploaded
[17:40] <sconklin> tgardner, working on it now. LP staging system is down, so I'm going to have to test on the live LP.
[17:40] <sconklin> should have it for you soon
[17:48] <tgardner> ogasawara, pushed master-next without 'UBUNTU: SAUCE: acpi_init: move to full RCU free for IO mappings'
[17:48] <ogasawara> tgardner: ack, thanks
[17:49] <tgardner> now, back to testing stable
[18:00] <tgardner> apw, ogasawara, I've been messing with 'make dpkg-deb' in the upstream kernel. It works pretty well for bisecting.
[18:00] <sconklin> tgardner, it's a little more difficult than I thought. 
[18:01] <ogasawara> tgardner: cool, for mainline I usually use apw's mainline-build-once script when bisecting
[18:03] <tgardner> ogasawara, I like the upstream method since you don't have to rebuild the whole kernel every time.
[18:06]  * tgardner -> lunch
[18:25] <sconklin> tgardner, create-stable-tracker should be fixed now, try it. I'm about to fix the other script that is likely also broken (verify-release-ready)
[18:34] <ogasawara> cyphermox: just posted a comment to claire's bug.
[18:35] <ogasawara> cyphermox: I've got almost identical hardware with respect to wifi, and am unfortunately not seeing any issues after suspend/resume
[18:41] <ogasawara> jibel: does your team have access to a Thinkpad X200s?
[18:44] <cyphermox> ogasawara: ok. I tried briefly with a X200, but no luck
[18:45] <ogasawara> cyphermox: no luck as in suspend/resume is working?
[18:45] <cyphermox> iirc yes, it was working, and I otherwise couldn't see freezes or whatever
[18:46] <cyphermox> but I don't think it's quite the same ahrdware either
[18:47] <ogasawara> cyphermox: I looked in hexr and I think we might have access to similar (possibly identical) hw, but I don't know how to go about getting access to the hw
[18:47] <cyphermox> ogasawara: was there an ID or something for it?
[18:48] <ogasawara> cyphermox: in hexr, I'm looking at "Lenovo ThinkPad X200s - 7465CTO"
[18:48] <cyphermox> ok.
[18:50] <cyphermox> ah, looks like someone's own system :)
[18:52] <ogasawara> cyphermox: I'll email him and see if he can help test
[18:52] <ogasawara> cyphermox: I'll CC you
[18:52] <cyphermox> I'll take another look at stuff, in case I can find other hardware with teh right wifi card that sees this
[18:56] <cyphermox> ogasawara: does it make sense though to try fw_restart=0 in case it avoids crashing as it does? I found that as a possible replacement (though not really the same thing) for wd_disable. I'm also curious why that parameter isn't available :)
[18:57] <tgardner> sconklin, https://bugs.launchpad.net/bugs/868628 created with your fixes
[18:57] <ubot2> Launchpad bug 868628 in linux "Oneiric update to 3.0.6 (3.0.0-13.20) stable release" [Undecided,New]
[18:57] <sconklin> tgardner, thanks
[18:59] <ogasawara> cyphermox: I'd have to review the changes to see why wd_disable isn't available any longer.  you can try fw_restart but I didn't see any firmware issues in clan's logs, did I miss something?
[18:59] <cyphermox> no, don't think you did
[19:01] <cyphermox> but I was wondering if it wasn't due to what's in the link I added; as in if the firmware gets reloaded on resume, due to the watchdog doing it or any other reason
[19:01] <cyphermox> even if there really isn't any indication that this is being done
[19:10] <tgardner> ogasawara, pushed 3.0.6 stable onto master-next
[19:10] <ogasawara> tgardner: cool
[19:13]  * jjohansen -> lunch
=== kentb is now known as kentb-out