[07:01] good morning [08:32] hi [08:32] morning [08:33] somebody know where are can find a snappy ubuntu core iso to install it my bare metal servers? BTW what is the difference between snappy ubuntu core and ubuntu core (without snappy)? [08:34] I try to ask, but I am affraid everybody is sleeping right now [09:04] Good morning all; happy Friday, and happy No Socks Day! 😃 [09:09] Halacs: just a new user, but afaiu just core is deb based, while snappy core has all that stuff for app separation etc and no debs [09:10] Halacs: if your bare metal machine has traditional BIOS, then you can just use the x86_64 image [09:10] Halacs: if it only has UEFI, then you'll have to wait until they fix that [09:11] and by 'just use the image' I mean you can just dd it to a drive [09:12] I have an IBM Balde Center H, and to be honest, currently I dont know whether is is UFI or not, but I will try this [09:12] but, could you give me a link for the image you mentioned? [09:12] just for sure [09:14] http://releases.ubuntu.com/15.04/ubuntu-15.04-snappy-amd64-generic.img.xz [09:14] thanks tbr! [09:15] just uncompress & write that to a disk or a usb device and boot it [09:17] okay, thanks, I will try it [11:59] anyone have a dhcp-server snap app? [13:06] dduffey, i would take a look at webdm ... not dhcp but avahi ... [13:06] shouldnt be hard to deduct how to roll a dhcp server snap from that [13:06] ogra_, what I want to do is setup a small snappy dhcpd server so all my other snappy devices will get an IP address from that server [13:07] ogra_, ah, I see [13:17] do we have an apparmor guide to debugging on Snappy? I'm having problems with NodeJS not playing well. [13:30] sergiusens: hi! is the webdm package only released on armhf? I can't find it with "snappy search" on my amd64 kvm [13:31] Chipaca: too ^ [13:32] alecu: it's only released for rolingas [13:32] * alecu gets a new haircut [13:33] Chipaca: how can I enable rolingas? [13:34] is it a separate channel, or an installation option.... [13:34] alecu: not sure whether you can easily switch; at image creation time, it's something like: sudo ubuntu-device-flash core rolling --channel edge --output 386.img --developer-mode --enable-ssh --device generic_i386 [13:34] great, thanks. [13:34] kyrofa: ^ [13:44] Chipaca: no luck. I created the image with that command, booted it in kvm, but still snappy search can't find webdm [13:45] Chipaca: I check the channel.ini, and it says: [13:45] channel: ubuntu-core/rolling/edge [13:45] any ideas? [13:49] maybe it's only available for armhf? [13:49] * beuno looks [13:50] Supported architectures: [13:50] 64-bit x86, ARM [13:50] so no [13:50] Supported releases: [13:50] rolling-core [13:52] alecu: I have no idea why you can't find it. [13:53] alecu: https://search.apps.ubuntu.com/api/v1/package/webdm [13:54] beuno: did me pushing a new version for only rolling automatically unpublish 0.5 for 15.04? It was one of the things I wanted to test [13:54] sergiusens, it did, because the store only shows one version of an app atm [13:54] beuno: is that by design? [13:55] or just something missing? [13:55] sergiusens, by design originally, and now that's changed with snappy [13:55] so there's an item in the backlog to change that [13:56] beuno: that's fine then, or else we will be living confusing times [13:56] beuno: I guess I can just tick 15.04 then or else I'll get a bunch of people asking for the link [13:56] today you wouldn't be able to install a specific version, for example [13:56] sergiusens, indeed [13:56] we'll have channels soon to allow you to do releases better ) [13:57] * sergiusens creates an image from scratch to see if he sees what alecu is seeing [13:57] beuno: \o/ [13:57] sergiusens: I used the cmdline that chipaca pasted earlier, then started it with: [13:57] kvm -m 768 -redir :8090::80 -redir :8022::22 386.img [13:58] then I sshd into it, sudo snappy update, snappy search [13:58] alecu: you don't need to run update to search though ;-) [13:59] sergiusens: I tried search first, and when it didn't find anything I muscle memoried apt-get update [13:59] but I see what you mean [13:59] alecu: alias kvm_snappy [13:59] alias kvm_snappy='kvm -m 1500 -redir :8022::22 -redir :8080::8080 -redir :4200::4200' [13:59] that's what I use ;-) [13:59] nice [14:01] alecu: I can't reproduce http://paste.ubuntu.com/11026375/ [14:01] just created a fresh image [14:02] boo [14:02] alecu: I know, Chipaca told you to create an i386 image; there's no webdm for 386! [14:02] sergiusens: what command did you use to create it? [14:02] ah! [14:02] my snappy as vmware image workflow seems to be broken now :( [14:02] alecu: sudo ubuntu-device-flash core rolling --output amd64.img --developer-mode --channel edge [14:02] beowulf: how so? [14:02] sergiusens: I'll try that, thanks a bunch! [14:03] np [14:03] sergiusens: you don't use "--enable-ssh"? [14:04] sergiusens: converting i get a error message/warning: http://pastebin.ubuntu.com/11026387/ [14:04] sergiusens: and vmware fails to ubuntu-core boot with that vmdk [14:05] PXE-E53: No boot filename received [14:05] beowulf: you are seeing kpartx issues [14:05] beowulf: clear the mappings (quickest is to reboot ubuntu) [14:06] sergiusens: does that make me special? do i get a badge [14:06] sergiusens: yes, I'll reboot ubuntu.... *whistles* [14:35] @reviewlist [14:35] https://code.launchpad.net/~chipaca/ubuntu-core-launcher/unshare/+merge/258367 | No reviews (1 day old) [14:35] https://code.launchpad.net/~stephen-stewart/webdm/disabled-install-buttons-for-oem-types/+merge/258621 | No reviews (less than a day old) [14:35] https://code.launchpad.net/~jamesodhunt/snappy/install.yaml/+merge/256925 | Needs Information: 1, Needs Fixing: 1 (16 days old) [14:40] beowulf: added a comment to your MP [14:43] sergiusens: good spot, ta [14:57] beowulf: Chipaca https://code.launchpad.net/~sergiusens/webdm/queryPackageNames/+merge/258639 [14:58] hmmm [14:58] * sergiusens resubmits [14:59] sergiusens: you want me to take that and put search back? [15:00] sergiusens: which might not happen today, as i'm still trying to get a working ubuntu-core [15:01] beowulf: no worries, I also have an improvement to make to that MP [15:13] @reviewlist [15:13] https://code.launchpad.net/~chipaca/ubuntu-core-launcher/unshare/+merge/258367 | No reviews (1 day old) [15:13] https://code.launchpad.net/~sergiusens/webdm/queryPackageNames/+merge/258640 | No reviews (less than a day old) [15:13] https://code.launchpad.net/~jamesodhunt/snappy/install.yaml/+merge/256925 | Needs Information: 1, Needs Fixing: 1 (16 days old) [15:13] sergiusens: know what this means? http://pastebin.ubuntu.com/11027355/ [15:14] beowulf: do you have the latest snappy? [15:14] beowulf: how did this all start to happen? [15:14] mvo: ideas? ^ [15:15] sergiusens: on 14.04 i did udf, and vmware couldn't but the vmdk build from that img, so i've upgraded... and now i see this [15:15] beowulf: trusty? [15:15] now on 15.04 [15:15] oh [15:17] sergiusens: which ppa should i be using? [15:17] beowulf: snappy-dev/tools [15:18] jdstrand: hullo hullo hullo! [15:20] Chipaca: hi! [15:21] unfortunately, I need to step away for a few minutes. but ask in backscroll and I'll get back to you in a bit [15:23] jdstrand: sure. It's about https://code.launchpad.net/~chipaca/ubuntu-core-launcher/unshare/+merge/258367 [15:24] sergiusens: meh, I thought we had fixed all the "exit status without further info " issues, is this u-d-f build against latest stable (or unstable) of u-d-f ? I wonder if that helps [15:24] jdstrand: apparmor complains, not about unshare(2), but about mount [15:25] mvo: I think beowulf is on whatever is on vivid [15:25] mvo: and yeah, we didn't fix all the exit status messages which kind of sucks a bit :-/ [15:26] sergiusens: mvo: i initially added ppa:snappy-dev/beta, i don't know if that matters (looked at wrong docs) [15:26] mvo: maybe he is just on an out of date u-d-f (I hope it's that) [15:26] sergiusens: mvo: i hope/plan to get to that next week [15:26] beowulf: using beta should be fine [15:26] sergiusens: mvo: i expect your first few days back are going to be spent reviewing :) [15:27] sergiusens: mvo anything i can try to get a working img? [15:29] beowulf: mvo I suspect snappy unpack is failing for some reason [15:29] but that is just gut feeling [15:31] beowulf: hm, I am not sure, maybe a rebuild against a up-to-date snappy [15:34] mvo: does it fail for you using what's in the ppa? [15:34] sergiusens: against what version is u-d-f in the ppa build? [15:34] sergiusens: I need to check, give me a sec [15:35] mvo: Built-Using: golang (= 2:1.3.3-1ubuntu4), golang-ar (= 0.0~git20150512-0ubuntu2), golang-go-flags (= 0.0~git20141007-1), golang-gocheck (= 0.0~bzr20131118+85-2), golang-goconfigparser (= 0.2-0ubuntu1), golang-juju-loggo (= 0.0~git20150318-0ubuntu1), golang-pb (= 0.0~git20131219-1), golang-yaml.v2 (= 0.0~git20150225-0ubuntu1), ubuntu-snappy (= 1.0-1+424~ubuntu15.04.1) [15:36] 424 is not latest [15:36] Chipaca: no, it was until last night [15:36] i mean, 439 is what's in the ppa from last night [15:36] *ancient* :-p [15:39] sergiusens: I think r436 has some important error reporting fixes [15:39] having that would be good [15:39] mvo: I can trigger a rebuild, we have recipes for everything now ;-) [15:39] that would be cool [15:39] mvo: Chipaca to bookmark https://code.launchpad.net/~snappy-dev/+recipes ;-) [15:40] yay! you found it! [15:45] mvo: btw https://code.launchpad.net/~sergiusens/goget-ubuntu-touch/bump_upstream/+merge/258651 [15:45] or the recipe will fail... [15:46] mvo: and I've been thinking about snappy build and releases and rolling and haven't come up with anything smart yet, maybe we need to add the base release to packages.yaml [16:30] Chipaca: what is the denial? I may be surprised if seccomp likes it too [16:31] [Fri May 8 16:31:01 2015] audit: type=1400 audit(1431102661.774:53): apparmor="DENIED" operation="mount" info="failed mntpnt match" error=-13 profile="/usr/bin/ubuntu-core-launcher" name="/tmp/" pid=1928 comm="ubuntu-core-lau" flags="rw, private" [16:31] jdstrand: sneakily, I do it before seccomp [16:32] Chipaca: is it possible to do it before apparmor? right now things are quite clean wrt apparmor rules-- we don't allow any mounts, umounts or remounts [16:32] jdstrand: bah. i do it before loading seccomp for the target executable, but i guess it's run under a seccomp of its own? [16:32] Chipaca: that said, we could probably craft a rule that is acceptable [16:33] jdstrand: i think the launcher is run under apparmor already [16:33] the launcher is not [16:33] the launcher does the equivalant of an 'aa-exec' [16:34] jdstrand: then what is debian/usr.bin.ubuntu-core-launcher ? [16:34] iirc, it do an aa_change_onexec [16:34] Chipaca: oh, yes [16:34] :) [16:34] Chipaca: right, the launcher does run under a profile, for some reason I read the above denial as an app denial, but it is a launcher denial [16:34] sorry [16:35] yep yep, it's the launcher itself [16:35] yep, I'm there now [16:35] Chipaca: mount rules are tricky, hold on a sec [16:36] jdstrand: i do two mounts, fwiw [16:36] well, you can see 'em in the mp [16:38] mount options=(rw private) -> /tmp/, [16:38] Chipaca: can you add the above to /etc/apparmor.d/usr.bin.ubuntu-core-launcher, then do: sudo apparmor_parser -r /etc/apparmor.d/usr.bin.ubuntu-core-launcher [16:42] Chipaca: actually we should change that (feel free to test it for now while I get the other other one) [16:44] [Fri May 8 16:43:50 2015] audit: type=1400 audit(1431103430.829:55): apparmor="DENIED" operation="mount" info="failed flags match" error=-13 profile="/usr/bin/ubuntu-core-launcher" name="/tmp/" pid=1942 comm="ubuntu-core-lau" srcname="/tmp/snaps.1000/hello-world.canonical/1.0.15/tmp/" flags="rw, bind" [16:44] ^ advanced to the second mount :) [16:45] Chipaca: mount options=(rw bind) /tmp/snaps.[0-9]*/**/tmp/ -> /tmp/, [16:46] jdstrand: \ooooooooo/ [16:46] yay! [16:46] so feel free to add those to the profile with a comment [16:47] fyi, I asked Tyler to review the mp. he'll be back on monday [16:47] (he looked really hard at the launcher before) [16:48] jdstrand: excellent; i'll wait for his review [16:48] Chipaca: for Tyler's context, in the commit for the above profile changes, can you add the denials? [16:48] the commit log that is [16:48] i'll get mvo or sergiusens to +1 the basic code [16:48] jdstrand: sure [16:49] cool [16:49] thanks [16:50] jdstrand: thank you! [16:51] 'night all o/ [16:52] beowulf: have a good weekend! [16:52] jdstrand: there [17:03] Chipaca: great, thanks again :) [17:22] mvo: sergiusens: log juggling done, for now at least [17:22] mvo: perhaps now the parallel between logger and interacter is more obvious :) [17:22] mvo: and then again, perhaps you were right and i was overthinking it :) [17:25] Chipaca: heh, I like what I see in the branches! simple and does all we need for now, I guess one more for the error logging will come later(?). but yeah, the similarities are striking now, I wonder how to combine the log and the progress, maybe we can have a "human-friendly" log that omits all the crazy timestamps(?) [17:25] mvo: i think the one for error logging is up already [17:25] Chipaca: but I may talk nonsesens, haven't really thought this through yet [17:25] mvo: but maybe you're talking about a different one [17:26] oh [17:26] oooooh [17:26] Chipaca: I was thinking if all the logging in lp:~chipaca/snappy/die-LogError-die is covered in the followups [17:26] mvo: I'D ALREADY FORGOTTEN [17:26] * Chipaca hides in shame [17:26] ha! at least I'm not the only one with a memory like a goldfish :P [17:26] * mvo hugs Chipaca [17:31] mvo: there :) [17:33] mvo: is the target for https://code.launchpad.net/~mvo/snappy/selftest-failover/+merge/258655 the right one? [17:54] is this what people are using for pip and snappy? http://www.wefearchange.org/2015/04/creating-python-snaps.html [18:44] Chipaca: are you going to be around for a while longer? [19:04] Chipaca: ups, its not, I updated the MP, silly me [19:24] mvo: Chipaca mind taking a look at https://code.launchpad.net/~sergiusens/webdm/oemBrokeFix/+merge/258681 ? [19:36] sergiusens: I am going to be around a while, now [19:37] Chipaca: heh, if you would of replied sooner I would of taken the complicated approach to fix https://code.launchpad.net/~sergiusens/webdm/oemBrokeFix/+merge/258681 :-P [19:37] now I await the needs fixings ;-) [19:37] sergiusens: ah, i haven't top-approved [19:37] sergiusens: what was the complicated approach? [19:38] Chipaca: extend packageYaml in snappy and add an interface to get the branding [19:38] sergiusens: at this stage, what's another spot on the liger [19:38] ... wait, ligers don't have spots [19:39] sergiusens: top-approved [19:39] Chipaca: thunder cats you mean? [19:40] sergiusens: http://en.wikipedia.org/wiki/Liger [19:42] holly cow, I learned something new today (again!) [19:56] mvo: no, that's Auðumbla [19:56] or maybe Gavaevodata [20:02] mvo: what is this version number's granularity btw ? 1.4.0.0.2 ? [20:02] sergiusens: we have support for "-" so 1.4.0-2 [20:02] sergiusens: maybe I'm not understanding the question actually [20:03] mvo: so many dots, that's all :-) [20:03] I think its a artifcat from the old days when we did not have a "-" [20:03] so many dots where used to simulate the packaging version [20:04] * Chipaca is going to make packages versioned “.”, “..”, “...”. Because the docs say I can. [20:06] mvo: does selftest have tarmac? [20:06] or otherwise auto-merge? [20:06] I don't think so [20:07] mvo: i don't think i'm going to manage to review your selftest branches today though [20:07] not with any kind of proper testing [20:08] mvo: I think it does, but it doesn't any sanity checks [20:10] confirmed, it's there [20:10] aha, ok [20:10] having the selftest in tarmac and building a image etc would be cool, but I'm too tired to even think about this [20:10] (i.e. as test run it on a fresh image) [20:11] * mvo will vanish into bed soon [20:13] mvo: by "snappy install typo", you mean a snappy install of a package that can't be found? [20:13] mvo: if so, yes, we log those [20:14] will reply on the mp, go to bed :) === erkules_ is now known as erkules [20:16] Chipaca: one last one please https://code.launchpad.net/~sergiusens/webdm/policies/+merge/258686 [20:17] sergiusens: go [20:18] Chipaca: yes, that, its fine, just wanted to check :) [20:18] * mvo waves [20:30] Chipaca: i'll bbl, which means you should nt even be here btw ;-) [22:25] does snappy require root to install? I thought we got away from that with click.