| === mamarley is now known as Guest427 | ||
| === mamarley_ is now known as mamarley | ||
| === mamarley is now known as Guest91007 | ||
| === mamarley_ is now known as mamarley | ||
| xnox | i wonder if i can run LTS kernel on yakkety, such that i get CLS for my laptop | 07:50 |
|---|---|---|
| smb | xnox, you surely can run then 18.04 kernel ... at some point | 08:02 |
| xnox | hm. i'm wondering why we don't keep lts ga kernel in yakkety. Or like forward copy it. | 08:03 |
| xnox | i should be able to pin to xenial kernel package i think. things should generally work. | 08:03 |
| smb | xnox, its just that I wonder why you want that | 08:06 |
| smb | (and don't assume I have any idea at this time of day about what this CLS ramblings are) | 08:08 |
| xnox | live patchign | 08:08 |
| xnox | because it is stated that live patches are not currently built/released for yakkety kernel | 08:09 |
| smb | xnox, so what? its not like you could not reboot your laptop :-P | 08:11 |
| caribou | guys, I must be missing something obvious with my kernel rebuilds: | 09:10 |
| caribou | even when building the generic package with skipabi=true skipmodule=true, I'm not able to reboot with a rebuilt bzImage | 09:11 |
| caribou | it just doesn't find the root device upon reboot | 09:12 |
| caribou | FYI I'm working on a bisect & don't want to have to rebuild all modules each & everytime | 09:12 |
| caribou | apw: smb: any thought ? ^^^ | 09:15 |
| caribou | if you're not too busy of course :) | 09:15 |
| smb | caribou, hm... frankly for bisecting I usually (most of the time its on mainline trees anyhow) copy a /boot/config-... into .config and run "make -j32 deb-pkg INSTALL_MOD_STRIP=1" | 09:16 |
| smb | There is some auto incrementing version of the packages that way, too | 09:16 |
| caribou | smb: which Mainline tree did you use ? kernel.org one ? last time I tried, it complained that it could no find debian/control | 09:17 |
| smb | Just one needs to be careful what number is used with the latest build (they might jump back and forth) | 09:17 |
| caribou | smb: I would prefer to use the mainline as I can't use non-linear tags for the bisect | 09:18 |
| caribou | smb: let me try your command again | 09:18 |
| smb | caribou, yes the kernel.org one | 09:18 |
| smb | caribou, note its is just plain make and not dpkg-buildpackage or so | 09:18 |
| boulabiar | Hi, how to bug report the new kernel livepatch ? | 09:47 |
| boulabiar | after installing, my laptop can't sleep anymore | 09:47 |
| boulabiar | ubuntu-bug canonical-livepatch don't work, because the package is installed via snap | 09:48 |
| henrix | boulabiar: i guess you'll need to use https://bugs.launchpad.net/ubuntu/+source/canonical-livepatch/+filebug | 09:50 |
| boulabiar | henrix: something went wrong with the page you asked my to use | 09:56 |
| boulabiar | I've written a full paragraph, then when i clicked send, an error appeared, I repeated this 2 times... | 09:57 |
| henrix | boulabiar: ah, so launchpad is probably in one of those days | 10:01 |
| ogra_ | apw, did you forget to binary-new the linux-raspi2 package in xenoal-security/-updates ? | 10:01 |
| ogra_ | (seems the images that QA tries to test here at the sprint all ended up with 4.4.0.1029.29) | 10:03 |
| smb | ogra_, the version number sounds like the one I would expect | 10:05 |
| ogra_ | smb, | 10:06 |
| ogra_ | linux-raspi2 | 4.4.0-1029.36 | xenial-security/universe | source | 10:06 |
| ogra_ | vs | 10:06 |
| ogra_ | linux-raspi2 | 4.4.0.1029.29 | xenial-security/universe | armhf | 10:06 |
| ogra_ | brad is on it though, i pinged him in person | 10:07 |
| apw | ogra_, what is wrong with those versions ? | 10:19 |
| apw | remember one is source and the other is a binary from a _different_ source | 10:19 |
| ogra_ | apw, we need 4.4.0-1029.36 in the archive as binaries | 10:19 |
| ogra_ | ogra@anubis:~/datengrab/images/snappy$ rmadison -s xenial-updates,xenial-security linux-raspi2 | 10:20 |
| ogra_ | linux-raspi2 | 4.4.0-1029.36 | xenial-updates/universe | source | 10:20 |
| ogra_ | linux-raspi2 | 4.4.0-1029.36 | xenial-security/universe | source | 10:20 |
| ogra_ | linux-raspi2 | 4.4.0.1029.29 | xenial-updates/universe | armhf | 10:20 |
| ogra_ | linux-raspi2 | 4.4.0.1029.29 | xenial-security/universe | armhf | 10:20 |
| ogra_ | there is no armhf binary for 4.4.0-1029.36 anywhere | 10:20 |
| apw | ogra_, the binary is _not_ called linux-raspi2 ? | 10:22 |
| apw | you are asking rmadison for binaries only called linux-raspi2 | 10:22 |
| ogra_ | oh man | 10:22 |
| apw | try -S | 10:23 |
| ogra_ | indeed you are right, thoughh why do we only get .29 in the snaps then | 10:23 |
| ogra_ | https://launchpadlibrarian.net/290161189/buildlog_snap_ubuntu_xenial_armhf_pi2-kernel_BUILDING.txt.gz | 10:24 |
| ogra_ | thats a build log from 2h ago | 10:24 |
| apw | ogra_, the linux-raspi2 package is from linux-meta-raspi2 and that is the ight number for that | 10:24 |
| ogra_ | Get:5 http://ftpmaster.internal/ubuntu xenial-security/universe armhf linux-image-raspi2 armhf 4.4.0.1029.29 [2324 B] | 10:24 |
| apw | you should have linux-raspi2 at 4.4.0.1029.29 (note no - in the version) | 10:24 |
| apw | and linux-image-4.4.0-1029 at 4.4.0-1029.36 which is from linux-raspi2 | 10:25 |
| ogra_ | hmm | 10:25 |
| ogra_ | Get:2 http://ftpmaster.internal/ubuntu xenial-security/universe armhf linux-image-4.4.0-1029-raspi2 armhf 4.4.0-1029.36 [35.3 MB] | 10:25 |
| ogra_ | there is actually is | 10:25 |
| apw | right, so that is exactly as expected | 10:25 |
| ogra_ | then i dont get whyy all the bugs are still there :((( | 10:26 |
| apw | ogra_, which bugs, ones which are fixed in the -proposed kernel ? | 10:26 |
| smb | ogra_, because it only fixes security | 10:26 |
| smb | ? | 10:26 |
| ogra_ | (teh kernel i tested from proposed yesterday was not oopsing on boot ... the QA team now tests it and has unbootable imges again) | 10:26 |
| apw | ogra_, they are not in that kernel it was an embargoed security fix | 10:26 |
| ogra_ | err | 10:27 |
| apw | ogra_, that is exactly expected | 10:27 |
| ogra_ | so you didnt base on that one ? | 10:27 |
| apw | that is _how_ embargoed cves work | 10:27 |
| ogra_ | damn | 10:27 |
| ogra_ | well, we need the one from proposed today | 10:27 |
| apw | no we would never do that, it was an untested CVE CRD release | 10:27 |
| apw | ogra_, that is gone | 10:27 |
| ogra_ | so is there any way to get the fixes in today (we tested the hell out of the proposed kernel yesterday to make sure it can land asap) | 10:28 |
| apw | ogra_, that doesn't have the aa fix bear in mind | 10:29 |
| ogra_ | it is essential for the snappy RC image that we need to release this week | 10:29 |
| ogra_ | (and what you have in the archive is unbootable anyway on pi3 currently) | 10:29 |
| apw | ogra_, i might be able to resurect the previous version but that won't have the aa fix which we respan for | 10:30 |
| sbeattie | ogra_: jjohansen may have a kernel you can use which was based on the -proposed kernel with the raspi2 fixes + the needed apparmor fix. | 10:30 |
| ogra_ | sbeattie, for the archive ? | 10:30 |
| sbeattie | in a ppa | 10:30 |
| apw | oh yeah thats right, we could do that because there was jj's kernel in a ppa | 10:30 |
| jjohansen | ogra_: so based on kernel that was in updates yesterday + the apparmor fix | 10:30 |
| jjohansen | https://launchpad.net/~apparmor-dev/+archive/ubuntu/apparmor-devel | 10:30 |
| apw | ogra_, anyhow, i'll see where we are with the respins | 10:30 |
| sbeattie | s/updates/proposed/ | 10:30 |
| ogra_ | can we somehow fast-path that in ? (i have the QA team here and i can spin snaps for them from the PPA to get everything tested) | 10:31 |
| jjohansen | sbeattie, orgra: so again this is not based on -proposed but on -updates from yesterday | 10:31 |
| jjohansen | the xenial -proposed kernel has the apparmor backport from yakkety on it | 10:32 |
| sbeattie | jjohansen: oh, updates? | 10:32 |
| jjohansen | yeah -updates from yesterday | 10:33 |
| ogra_ | jjohansen, well, we have a kernel that oopses on boot, that was the released kernel and the one formerly in updates ... | 10:33 |
| ogra_ | the only kernel that boots is the one that was in -proposed yesterday | 10:34 |
| apw | ogra_, cna you remember the version number ? | 10:34 |
| ogra_ | i can dig it up, i have a test image but that needs some time since i need to set up gear to boot it to run uname | 10:35 |
| apw | ogra_, can you build images/snaps whatever against a PPA | 10:35 |
| apw | if not then it is no point in me trying to find it | 10:36 |
| ogra_ | apw, ppisati said it should just be >= 4.4.0-1028.34 but thats obviously not true when the former proposed kernel wasnt even used | 10:36 |
| apw | what should be ? | 10:36 |
| jjohansen | ogra_: so xenial -proposed should not be vulnerable to the aa bug, the yakkety version of apparmor had refactored that code and due to that had the fix in it already | 10:36 |
| ogra_ | apw, the one with all the fixes :) | 10:36 |
| apw | i need a specific exact version as i have to reach blind into the archive | 10:37 |
| ogra_ | yes, i know | 10:37 |
| jjohansen | the ppa has the -updates kernel with the apparmor fix based on the assumption you would a released kernel | 10:37 |
| ogra_ | as i said, that requires some work | 10:37 |
| ogra_ | jjohansen, right, but -updates is a step back and wont boot without some bootloader hackery first | 10:38 |
| ogra_ | apw, i'll ping you as soon as i can give you the version | 10:38 |
| jjohansen | ogra_: got it, so I just want to confirm for you from an apparmor pov the ppa kernel based on yesterdays -updates and the xenial -proposed kernel should be functionally equiv. ie not have the bug | 10:39 |
| apw | ogra_, oh, a thought, isn't the download of the package from -proposed in the logs of the image build which was good; rather than booting it | 11:25 |
| ogra_ | apw, \o/ ... sorry it took so long, found the package version in a build log | 12:16 |
| ogra_ | Get:5 http://ftpmaster.internal/ubuntu xenial-proposed/universe armhf linux-image-raspi2 armhf 4.4.0.1028.28 [2324 B] | 12:16 |
| ogra_ | and | 12:17 |
| ogra_ | Get:1 http://ftpmaster.internal/ubuntu xenial-proposed/universe armhf linux-image-4.4.0-1028-raspi2 armhf 4.4.0-1028.34 [35.4 MB] | 12:17 |
| ogra_ | does that help ? | 12:18 |
| apw | ogra_, yes that helps, thanks. | 12:21 |
| yoasif | hi, is there any documentation or way of getting the "extras" package for mainline kernel builds? | 16:16 |
| tjaalton | i'm hit by https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1635115 | 18:44 |
| ubot5` | Ubuntu bug 1635115 in grub2 (Ubuntu) "grub fails zfs root filesystem detection due to failed checksum" [Undecided,Confirmed] | 18:44 |
| tjaalton | my home server fails to boot after latest updates to xenial | 18:44 |
| tjaalton | oh well, no time to fix it now.. | 18:47 |
| mamarley | Ooh, that should be probably be marked Critical. | 18:55 |
| apw | i am not sure zfs is support as root/boot is it? | 18:55 |
| tjaalton | not in the installer, but i see no reason to make it harder | 19:16 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!