=== 64MAAJNBW is now known as sunnydrake === _LibertyZero is now known as LibertyZero === yofel_ is now known as yofel === fairuz_ is now known as fairuz === cking_ is now known as cking === smb` is now known as smb === johanbr_ is now known as johanbr [14:14] reboot [15:42] * ogasawara back in 20 [15:43] bjf, latest update for c-o-t-d 2.6.38 seems to have been 2011-05-05 [15:43] diwic, that was the first and only, was waiting feedback [15:44] bjf, aha [15:44] diwic, i can fire off another if desired [15:45] bjf, if it's simple, please do [15:45] diwic, will do === jdstrand_ is now known as jdstrand [15:53] bjf, meanwhile I installed the package and it seems to work fine, so I think you're good to go for enabling daily builds for Natty as well [15:53] diwic, just uploaded a new one, thanks for the feedback, will start doing them automatically [15:55] bjf, thanks! [15:57] diwic, added, if you see any issues with it now, don't hesitate to mention it/them [16:38] bjf, how would I add that kernel call to my collection? [16:38] I tried by the URL, but no joy [16:38] and I am too stupid just now to figure it out :) [16:38] um, one sec [16:40] bjf, sconklin: I do not see any pending karmic kernel to be tested. Am I missing something? [16:41] hggdh, looks like someone went ahead and promoted the karmic kernel to updates [16:41] skaet, ^ [16:41] bjf, hggdh, looks like we have a process failure then. [16:41] :P [16:42] well, we can still go ahead and validate it, just in case [16:42] hggdh, please do so. [16:42] JFo, did you get some notice about that calendar? i just added you to the share list [16:42] I'll take this up with pitti and see if we can figure out what happened. [16:43] k [16:43] * JFo looks [16:43] bjf, looks like it just showed up in my main calendar now. Thank you :) [16:44] JFo, np [16:44] hggdh, please let me know if there's any problems with it, soonest. [16:45] * apw pops out to get a replacement hdmi cable [16:50] ogasawara, we going to do a status meeting this week ? [16:51] bjf: yep, that's the plan [16:52] ogasawara, always good to have a plan, do we have a list of blueprints we are going to be tracking ? [16:52] bjf: https://wiki.ubuntu.com/KernelTeam/Specs [16:53] ogasawara, we're going to want all of them on the agenda or you want me to pick the ones I think are relevant? [16:54] bjf: I'd only pick out the relevant ones. [16:54] ogasawara, ack [17:06] ogasawara, i've updated the meeting agenda, most of the blueprints are for you, let me know if there are others you think should be added [17:06] bjf: ack [17:10] bjf: I'm gonna move the one work item from the misc blueprint to the server-reqs blueprint and then we can just drop the misc blueprint from the agenda [17:12] ogasawara, wfm [17:43] * apw notes that plugging and unplugging hdmi outputs on intel leads to crashes :/ [17:46] apw: is that with the latest intel driver? [17:56] jjohansen, with whatever is latest in natty [17:56] now on a .39 kernel shall see if its any better [17:56] apw: ah there where some hdmi improvements in the latest intel drm tree, not sure what made it into .39 though [17:58] jjohansen, ok will see how this pans out [19:06] ogasawara, just saw your response :-) [19:11] JFo: pick a time when you want to start your kernel patching and git-fu, I'll put it in my schedule. not sure it'll require a full hour each week, but we should just start [19:11] Any chance that bug #424142 will get fixed in lucid's kernel? [19:11] Launchpad bug 424142 in linux "Address Collision" [Undecided,Confirmed] https://launchpad.net/bugs/424142 [19:11] I've just had a couple customers get bit by it, upgrading them to maverick kernel fixed it [19:13] ogasawara, right, my hope is that some of these will taper off and not be needed. [19:14] just wanted to get them started and indicate how serious I am about them to everyone :) [19:16] I suspect no, since linux-lts-backports-maverick is available [19:19] BenC: if it came through the next 2.6.32.y upstream longterm release, it'd get applied to Lucid. or use the backports kernel. [19:21] ogasawara: No chance of someone actively searching for the fix? Isn't 2.6.32.y inactive? :) [19:22] BenC: I thought 2.6.32.40 just came out, was that the last one? /me hasn't been following [19:22] wait, I didn't realize upstream was maintaining longterm kernels now, so hopefully they do a 2.6.32.y to fix it [19:24] BenC, GregKH still maintains .32 [19:25] BenC, ogasawara: looks like the fix was in 2.6.33 (commit 8c8def26) so it just never made its way to stable [19:25] Right, I knew it was fixed in 2.6.33, was just hoping for some love in lucid, even if 2.6.32.y isn't getting it upstream [19:26] I'm just saying that if we wait for it to show up in .32 stable it's never going to happen [19:26] That's exactly what I'm saying :) [19:27] sforshee, you want to do an SRU for it? [19:28] bjf, sure, I'll try to get some testing on it first to make sure it really fixes the problem [19:28] sforshee, that looks like a reasonable candidate [19:28] wfm [19:29] sforshee, looks like BenC might be able to help test :-) [19:29] Customers gave me sudo access to the affected machine, so I plan to abuse it as much as possible... [19:30] hehe [19:30] if you get something testable, I'm happy to toss it on there [19:30] BenC, I'll post a kernel to the bug report shortly, then I'll submit it for SRU as soon as you can get it verified [19:31] sforshee, not a rush, just information, if you can get a fix verified soonish, we are prep'ing kernels for the next cadence cycle [19:31] bjf, ack [19:32] sforshee: ok, please ping me on here, launchpad traffic doesn't get through my usual filters [19:33] BenC, will do. If you tell me whether you'll use a 32-bit or 64-bit kernel for testing I'll make sure the one you need is built first. [19:34] sforshee: Thanks, 32-bit [19:43] sforshee: this patch doesn't affect ABI, correct? [19:43] bjf, would you be so kind as to accept the lucid nomination on bug #424142 ? [19:43] Launchpad bug 424142 in linux "Address Collision" [Undecided,Confirmed] https://launchpad.net/bugs/424142 [19:44] BenC, it shouldn't [19:44] if it does, I would need to get the headers to fully test against this module [19:44] sforshee, you bet [19:44] sforshee, bjf: a huge thanks for the effort on this as well [19:44] BenC, np [19:44] BenC, I'll give you the headers, but as it only changes a function body it shouldn't affect the abi at all [19:45] sforshee, done [19:45] bjf, thanks [19:45] sforshee: I could have approved it for you :) [19:53] BenC, 32-bit test build is posted at http://people.canonical.com/~sforshee/lp424142/2.6.32-32.62~lp424142v201105231936/ [19:54] bjf: can you approve nominations for bug 787145 [19:54] Launchpad bug 787145 in linux-ti-omap4 "CVE-2011-1494" [Undecided,New] https://launchpad.net/bugs/787145 [19:55] herton, working it [19:56] herton, done [19:56] thanks bjf [19:56] np [20:26] sforshee: ok, give me 30 minutes [21:03] sforshee: not stalling...this box doesn't boot unattended so I'm just waiting for some one to kick it [21:03] sforshee: I've at least smoke tested here locally, but not on a box that exhibits the problem [21:07] BenC, cool, I've got everything ready to go for once we know it fixes the problem [22:22] sforshee: confirmed that it fixes the issue, and commented on the bug report [22:26] hi all, i'm having a weird issue trying to reboot with linux 2.6.38 (x86_64), ubuntu LTS 10.04, this is on atom D525 - dual core atom [22:26] system hangs on restart with: "will now restart" and then "[] Restarting system" [22:27] i've tried recompiling a variety of kernels [22:27] i've even compiled specifically for the atom, not using generic [22:27] same issue. but soft halt works fine [22:27] is this a known kernel support issue with atom dual core cpus ? [22:28] shankara: It sounds like an issue with the kernel you are compiling, like a panic, maybe related to your root filesystem [22:28] Sounds suspiciously like you are just compiling the kernel wrong [22:29] shankara: And specifically, it sounds like an issue you just have to figure out through more generic channels #kernelnewbies perhaps?) since it isn't an Ubuntu issue [22:29] BenC: i found this on the generic default kernel that came with lts 10.0.4 [22:29] You said it was 2.6.38 [22:29] after a couple of binary kernels, i decided to compile my own [22:29] same issue [22:29] shankara, have your tried our lts-backport kernel? [22:29] that's the most recent kernel i tested [22:30] bjf: not yet, maybe i should [22:30] i only resorted to source compiles after all the binary kernels in the apt repo gave me the same soft reboot issue [22:31] shankara: Booting with single (so without quiet) would be more helpful [22:31] The reason for the panic might give some clues to the real problem [22:32] BenC: i'm inclined to the some clues to the real problem might provide a reason for the panic [22:32] actually ... i wouldn't say it's a panic [22:32] there's no "ooops" [22:32] I would say it is a panic [22:32] it jsut doesn't actually restart the hardware, just promises to [22:32] The "will not restart" message comes right after a panic [22:32] err "will now restart" [22:33] shankara: If it fails to exec /sbin/init or similar, it will panic [22:33] panic != BUG [22:33] Benc: yes, sorry: "Will now restart" [22:33] it means "oh crap, I don't know what to do next" [22:34] BenC: it actually get's past stopping all system services, unmounting all filesystems ... then it jsut doesn't , so it seems init runs it's full course [22:34] sorry "then it just doesn't restart the hardware" [22:34] shankara: Do you mean it will actually boot, just not reboot? [22:35] it'll boot, it'll even halt. Just it wil not soft restart [22:35] ctrl-alt-delete has the same effect - no restart [22:35] Oh, I misread "trying to reboot with.." as "trying to reboot into..." [22:35] shankara: I have some tips for this problem... [22:36] yeah, it's a weird one [22:36] most people are trying to boot their boxes, I'm trying the opposite [22:36] restarting is critical for remote support on appliances [22:37] shankara: Try adding reboot=bios to your kernel command line [22:37] BenC: let me try that and get back to you [22:37] shankara: reboot=acpi might be worth trying too [22:37] ack. [22:38] BenC, just sent SRU patch to kernel-team list, thanks for testing [22:40] sforshee: np, thank you for taking on the logistics to get it in an SRU [22:43] is there anything i can do to accelerate getting help with bug 784335? i don't really have any kernel debugging-fu, but the machine in question isn't my primary machine and doesn't have any data on it, so i can re-install or re-configure it 10 ways to sunday if that helps [22:43] Launchpad bug 784335 in linux "Heavy network utilization with r8169 leads to kernel panic" [Undecided,Confirmed] https://launchpad.net/bugs/784335 [22:45] broder: I would argue that r8169 is a horrid driver and would suggest you'd have better luck wedging an e1000 card in there :) [22:45] but I'm not the official spokesperson, so I digress [22:47] broder: a pic of the on-screen panic would help [22:47] BenC: is the core dump not good enough? [22:49] broder: I'd have to set aside time to dig into that, perhaps later tonight [22:55] if i were to go and play the role of the apport retracer, would that get the bug more attention? [22:59] BenC! you're a genius! reboot=acpi worked :D [22:59] thanks! [23:02] as some people seem to be awake currently, I'd like to re-ask my question from like a week ago: http://paste.ubuntu.com/612064/ [23:02] any hints? [23:04] TheSeven, have you filed a bug? [23:06] no, as i don't have a clue what to file it against [23:06] this kinda seems to be an OMAP or ARM specific problem [23:07] but it doesn't seem like the OMAP guys would know what's the cause for it [23:08] so i'd like to check if there are any general hints for situations like this (some tunables maybe?) [23:08] so you would file it against linux which encompasses all of them and then I would go about determining if it was specific to one of them. [23:09] very simple process too... from a term run 'ubuntu-bug linux' [23:09] the maximum throughput being at a sub-sector block size and the fact that it happens both with USB and SD makes me think that this can't really be coming from the MMC/USB controller, but must be coming from some generic I/O code [23:09] presto! bug report! :-) How is that for the soul of ease? :) [23:09] very possible [23:10] but I would like to see the logging we collect from the bug report tool to dig a bit deeper. [23:10] so when it is behaving thus, if you could file that bug it would be most helpful [23:20] JFo: https://bugs.launchpad.net/ubuntu/+source/linux-ti-omap4/+bug/787246 [23:20] Launchpad bug 787246 in linux-ti-omap4 "Slow SD card and USB HDD I/O for block sizes bigger than a couple of bytes" [Undecided,New] [23:22] BenC: does http://pastebin.ubuntu.com/612078/ look like it has the information you wanted? [23:22] i'm not super-familiar with digging around in vmcores [23:41] smb: there? [23:47] hggdh: its pretty late in the day for smb to be here (00:47) for him [23:48] jjohansen: yeah, I had just remembered he is UTC+1 or 2... [23:48] jjohansen: perhaps you -- I am unable to boot the curretn maverick proposed on AWS, i386/m1.small [23:49] ugh, I can have a look [23:51] jjohansen: what do you need? [23:52] hggdh: hrmm, nothing just yet. I will boot an m1.small and install a proposed kernel and try to duplicate, and go from there [23:53] hggdh: actually, which region/zone, and even machine type if you can get that info [23:53] jjohansen: OK. What I did -- what I always do --: I booted the current AWS Maverick, then run a dist-upgrade with -proposed enabled [23:53] hggdh: right, I think I will be a little more selective and just pull the kernel [23:54] jjohansen: ec2-run-instances ami-b21de3db --instance-type m1.small --region us-east-1