=== amitk is now known as amitk-afk [07:17] moin [08:16] morning === amitk-afk is now known as amitk === rperier_ is now known as rperier [10:11] brb [10:27] apw: Is everything good with my kernels now? I realize I've made a couple of administrational misses in the past. Want to make sure it's all ok now [10:40] grub?!?! [10:45] ppisati, what about it ? [10:46] its a bootloader :) [10:46] ogra_: i was upgrading a calxeda node from Q to R [10:46] ogra_: and i tried to install grub [10:46] lovely [10:46] *it [10:46] fis the kernel deps then ? [10:46] *fix [10:46] ogra_: didn't hapend on panda&c [10:47] ogra_: shouldn't be the kernel [10:47] * ppisati notes he can't type this morning... [10:47] ogra_: is there a way to know who pulled it in? [10:48] you could dig through the redepends [10:48] or check the apt log [10:48] *rdepends ... [10:48] ah never mind [10:48] * ogra_ cant type either [10:49] i know linaro works on grub for arm ... but i doubt its even remotely usable yet [10:52] ogra_: ah never mind, wrong console, it wasn't the calxeda node [10:52] heh === stefanct__ is now known as stefanct === amitk is now known as amitk-afk [13:05] * henrix -> lunch === amitk-afk is now known as amitk [14:13] rtg_, apw, fyi. there are a lot of bug reports this morning of failures to upgrade to 3.8.0-15. bug 1161306 for example. [14:13] Launchpad bug 1161306 in linux (Ubuntu) "package linux-image-3.8.0-15-generic (not installed) failed to install/upgrade: subprocess new pre-installation script returned error exit status 128" [High,Confirmed] https://launchpad.net/bugs/1161306 [14:14] jsalisbury, hmm. what failed ? [14:15] rtg_, it looks like a failure in a python script: https://launchpadlibrarian.net/135490864/DpkgTerminalLog.txt [14:15] rtg_, maybe not related to the kernel? [14:15] jsalisbury, it looks like it might be a dpkg issue [14:16] jsalisbury, where are you seeing that ? [14:16] DpkgTerminalLog.txt [14:16] rtg_, that is in the DpkgTerminalLog.txt file [14:16] :-) [14:17] jsalisbury, yeah, doesn't look like a kernel issue. nothing changed wrt packaging. [14:18] apw, rtg_, thanks for looking. I'll chase it down with dpkg [14:18] jsalisbury, it looks to be a bit worrying if it is generic [14:19] apw, yeah. And there are a bunch of duplicate reports in a small time frame [14:19] apw, jsalisbury: last dpkg upload was 22 Mar [14:19] yeah that is a while ago ... thought [14:20] hmm [14:21] it looks like the first bug report on this was around 1AM UTC [14:21] jsalisbury, you should try to attract infinity's attention on this one [14:22] rtg_, ack [14:22] infinity, infinity help! bug 1161306 [14:22] Launchpad bug 1161306 in linux (Ubuntu) "package linux-image-3.8.0-15-generic (not installed) failed to install/upgrade: subprocess new pre-installation script returned error exit status 128" [High,Confirmed] https://launchpad.net/bugs/1161306 [14:23] jsalisbury, already poked him on #ubuntu-devel [14:23] apw, :) thanks [14:34] jsalisbury, my laptop updated correctly. it even autoremoved old kernels correctly when requested. [14:34] rtg_, ok. I'm going to see if I can reproduce it as well [14:46] apw: hi. did you have a specific patch in mind for my ivybridge link train issues? i could probably rebuild a kernel locally to test with it if so [15:00] apw, ogasawara: looks like the N7 kernel is under control for now, so I'm gonna see if I can finish the N4 kernel packaging and get it tested. [15:01] rtg_: sounds good [15:01] * rtg_ spent the morning thus far reading the various power articles and emails [15:02] rtg_: I'm gearing up for some meeting hell this morning, but I was thinking of hammering on a raring lbm this afternoon. at least looking into an updated compat wireless stack for it. [15:02] ogasawara, do you think its worth the bother for an interim release ? [15:03] rtg_: we've always done it, so I figured it wouldn't hurt [15:03] ogasawara, it's one more thing that we'd have to support [15:03] rtg_: and it seems we always get a request for some driver to land in it at the last minute [15:03] I'm thinking that folks ought to be using our backport kernels instead [15:04] rtg_: wasn't tseliot wanting it for something this cycle, I can't remember the details [15:04] ogasawara, he has a blueprint for modaliases that I've been dragging my feet on [15:04] I'm becoming less enthused about LBM in light of our kernel backports [15:05] rtg_: actually, Intel requested wifi drivers from 3.9 for 13.04 in my call last night. and I mentioned lbm would be the path of least resistance. [15:06] ogasawara, if they are new then why don't we dump them in the main kernel repo ? [15:06] rtg_: I'm fine with that too [15:07] ogasawara, did they give you some pointers as to which driver(s) were of inetrest ? [15:08] rtg_: I think they were vague saying wilkens peak support, but not direct pointers to patches [15:08] rtg_: so I'll circle back with them and then look to getting our main kernel updated and hold off on an lbm [15:09] ogasawara, only 96 patches to drivers/net/wireless/iwlwifi sine v3.8 :) [15:10] bug 1011422 [15:10] no bot? [15:10] hrmph [15:10] bug #1011422 [15:10] huh. dead bot. [15:11] https://bugs.launchpad.net/intel/+bug/1011422 [15:11] grumble mumble [15:11] private bug [15:11] the bot is probably ignoring you [15:26] kamal, please verify bug 1154238 (just a reminder) [15:26] Launchpad bug 1154238 in linux (Ubuntu Quantal) "update alx Ethernet driver" [Medium,In progress] https://launchpad.net/bugs/1154238 [15:54] sconklin: i'm having issues with maint-startnewrelease and P/omap4 [15:55] ppetraki: ok [15:55] what sort of issues? [15:55] sconklin: since the previous kernel wasn't released it can't find the pkg to extract&c to get abis&c [15:55] sconklin: i moved it manually to kernel.ubuntu.com [15:55] sconklin: as i always did [15:56] sconklin: but it keeps saying it can't find it [15:57] sconklin: what does --local do exactly? [15:57] ppetraki: we had this problem with regular precise. I fixed it by ignoring ABIs since we knew we had an ABI bump. But I think there's a better way. [15:58] oops ppisati ^^ [15:58] ppisati: I'd have to look at getabis to see how it works [15:58] doing that [16:02] local just runs it against the local repos as opposed to those on anotehr host you have ssh access to [16:04] ppisati: using --ckt-ppa should fetch them from the PPA, which should have the ones you want [16:04] I think? === kentb is now known as kentb-afk [16:10] sconklin: no luck :( [16:10] ok, let me pull your branch and have a look [16:10] I'm uploading Quantal ti-omap4 now [16:17] ppisati: I'll just fix it for you, as you're near EOD [16:17] sconklin: no prob, i'm debugging it now [16:19] sconklin: after having opened getabis, i'll leave it to you [16:19] :-) [16:36] For the record, the 3.8 dev kernel works fine on a re-mastered 12.10 desktop live cd... [16:36] (3.8 raring kernel that is) [17:13] cking, the N4 kernel builds now. [17:15] nice ! [17:15] rtg_, yay [17:15] ogra_, now to figure out how to update the touch kernel image. any thoughts ? [17:15] no, and i need a device for that [17:15] * cking in london at mo, so can't test it [17:15] you can try abootimg ... [17:16] iterate over all /dev/block/mmcblk0p* devices with abootimg -i and see if it recognizes an android bootimg signature [17:16] if you find one thats a good first step [17:16] it should be [17:17] abootimg -u /dev/mmcblk0p2 -k vmlinuz [17:17] it *should* [17:17] work [17:17] but it didnt' when i tried [17:17] ppisati, well, abootimg doesnt recognize the bootimg partition on my galaxy S2 at all here [17:17] ogra_: no no [17:17] ogra_: it recognizes it [17:17] its totally device specific [17:17] ogra_: but i'll have to tinker a bit more with it [17:17] ogra_: but i really had no time these days [17:18] ppisati, thats N4 ... LG device ... might be completely different [17:18] N7 is easy and we should get it to work ... i can even help with that since i have one here [17:18] (i have an N4 too, but thats my private phone and i definitely wont tinker with it) [17:20] I'm gonna get some lunch first before I start wrecking my gizmos [17:20] * rtg_ -> lunch [17:41] ogasawara, rtg_: cw actually got a lot more interesting recently when they started doing compat-drivers also, it builds backported drm drivers too which could potentially save some headaches backporting things into the main kernel. just plopping in cd instead of cw builds all the juicy stuff :) [17:45] its not really documented that well but i was playing with it a few days ago https://www.kernel.org/pub/linux/kernel/projects/backports/stable/v3.8.3/ChangeLog-3.8.3-2 [17:50] things like pulling most of drm changes from 3.6 into 3.5 for i915_hsw could have been avoided with that. well not really because theres no lts-quantal lbm and it wasn't around at the time but in the future.. [18:08] Hey guys I hit an issue with an install on my ideapad Y580 with secure boot cjwatson is under the impression is might be a kernel/firmware issue from the info he got from the installers syslog see http://ubuntuone.com/7AYGMhTAuDswfLCCn9y8W5 but it basically means it isn't booting into Ubuntu [18:09] I get the folowing message appear on reboot post install "Ubuntu has been blocked by the current security policy." [18:24] jsalisbury: do you know who would be the best person to talk to about this would be ^ [18:45] davmor2, just curious, is there a bug open? [18:46] davmor2, also, did this just start happening on this machine recently? Was there a prior release that worked with Secure Boot? [19:02] jsalisbury: Not looked for a bug. I did a fresh install about 2 weeks 4 days ago ready to do the for purchase app migration testing , that is complete and a machine with 500+ apps on is a bit clunky so I've done a fresh install of todays iso and got this [19:03] davmor2, so this didn't happen on this machine with images prior to todays iso? [19:07] jsalisbury: I've done maybe 20 installs on this machine testing uefi + secure boot and the result went to the canonical mailing list :) So Nope todays is the first that has played up [19:08] davmor2, thanks for the info. Can you open a bug for this, since it appears to be a regression. I can then perform a bisect to identify the commit that introduced this. [19:10] jsalisbury: let me get the the exact date I last installed I think there has only been 1 or maybe 2 kernel upgrades since then. Is there any info you would need from the installed system as I can't actually access it except from the live cd [19:11] jsalisbury: Friday the 8th was when I updated and installed ready for the testing [19:11] davmor2, just the kernel version. You might be able to get it to boot if you disable secure boot in theBIOS [19:11] jsalisbury: ah true give me 5 minutes then. [19:12] * henrix -> EOW [19:12] davmor2, thanks. and it would be great if you can open a bug so we can track all of this info. I'll post some links to test kernels in the bug for the kernel bisect. [19:13] jsalisbury: it looks like disabling secureboot has got me into the system so I'll report a bug from there now [19:13] davmor2, cool, thanks [19:24] jsalisbury: bug #1161566 [19:24] Launchpad bug 1161566 in linux (Ubuntu) "Regression on enabling the system for secureboot" [Undecided,New] https://launchpad.net/bugs/1161566 [19:24] davmor2, thanks [19:24] jsalisbury: I included the installer syslog too [19:25] jsalisbury: let me know if there is anything else you need, and I want be around now till Tuesday. [19:25] s/want/won't [19:25] davmor2, will do. I'll post any questions/requests to the bug, so you can review them when you have time [19:26] jsalisbury: no worries [19:28] davmor2, this may have been introduced by upstream 3.8.3 or 3.8.4 stable. It would be great to test them. I'll post links to the kernels in the bug [19:28] jsalisbury: actually do you know when the lastest kernel image hit the cd? If there is an older cd image about that has the kernel image before I can try that and see if it is just the latest kernel [19:30] davmor2, i'm not sure which kernels are on which cd images. The latest image you used was based off upstream 3.8.4, so testing 3.8.3 and 3.8.2 should tell us which one has the commit that introduced this [19:31] jsalisbury: 3.8.0-15-generic [19:31] jsalisbury: with uname -a the date there is that the date the kernel was built [19:33] okay so I'm grabbing the 27ths cd and I'll check what kernel is on there [19:35] davmor2 3.8.0-14.24 and 3.8.0-15.25 were based off of upstream 3.8.4, so I think they would both have this bug. 3.8.0-13.22 was rebased to upstream v3.8.3 and 3.8.0-12.21 was rebased to upstream v3.8.2. [19:48] jsalisbury: the issue I'm going to hit is figuring out how to apply the other kernels and get them to magically rebuild the missing bit [19:49] or is that some thing the kernel + grub should handle [20:02] davmor2, yes, you can do a 'sudo dpkg -i' kernel-package-name.deb and it will install the kernel and update grub [20:03] davmor2, this wiki page might be helpful as well: https://wiki.ubuntu.com/KernelMainlineBuilds [20:09] * rtg_ -> EOD [20:51] jsalisbury, ogasawara we're getting a rash of install bugs reported against the latest raring === pgraner is now known as pgraner-afk [21:31] jsalisbury: there's no need for all the bisecting on bug 1140716, comment #33 already found the bad commits [21:31] Launchpad bug 1140716 in linux (Ubuntu Raring) "[regression] 3.5.0-26-generic GPU hangs" [Critical,Confirmed] https://launchpad.net/bugs/1140716 [21:35] or maybe it's just 899b550 on the quantal kernel, would be useful to bisect with just that reverted [21:35] and then if it is, revert it from precise too === noptys_ is now known as noptys === kentb is now known as kentb-oout [23:04] bjf, yes I saw those, and pinged rtg and apw about them this morning. I've upgraded a couple of machines and couldn't reproduce it. [23:05] tjaalton, ack. Thanks for commenting on the bug. [23:08] bjf, infinity is also aware of this, in bug 1161306 It looks like it only may be reproducable with the Software Updater and not from the terminal. [23:08] Launchpad bug 1161306 in linux (Ubuntu) "package linux-image-3.8.0-15-generic (not installed) failed to install/upgrade: subprocess new pre-installation script returned error exit status 128" [Medium,Confirmed] https://launchpad.net/bugs/1161306 [23:36] infinity, bjf, jsalisbury: I'm not convinced that the problem is specific to Software Updater ... bug 1161528 shows the same problem happening from "apt-get install -f" from a terminal [23:36] Launchpad bug 1161528 in linux (Ubuntu) "unable to install linux-image, which is required by update-manager" [Undecided,Confirmed] https://launchpad.net/bugs/1161528 [23:36] (but I've not been able to reproduce it either) [23:38] That's a subtly different error than the other ones... [23:38] Hrm. [23:44] Random googling suggests that having /var/cache in a tmpfs could lead to this... [23:47] infinity: some of the bug reports include 'df' output, which show no such weirdness [23:47] Yeah, was a shot in the dark. [23:49] The fundamental issue seems to be, potentially, a corrupted debconf DB, but if a bunch of people all got corruption at the same time, that points at a package bug... Somewhere. [23:58] kamal, thanks for the update