[00:26] TJ-: any good news? :) [00:27] plm: I have an 18.04 armhf LXD container running. I'd need to do a lot more tests to discover any potential problems though. [00:28] TJ-: hmm good. Can I to do the same with 16.4? [00:28] TJ-: do you have a simple tutorial to me replicate here with 16.4? [00:29] TJ-: so reaaly works with LXD arm on x86_64 =D [00:29] *really [00:29] plm: I'll do some thorough testing because Stefan indicated some tooling may have problems. Once it seems OK I'll publish a script to automate creating the local hybrid image [00:30] TJ-: great TJ- =D [00:31] TJ-: do you think that tomorrow until afternoon do you will publishthat script? [00:31] plm: so far what I had to do is copy in the x86 systemd package contents, and the supporting libraries [00:31] TJ-: will be great have a simple way to run a full ubuntu armhf on x86 [00:32] plm: if you want to help test it I'll publish it via a git repository so you can pull it in and test also [00:32] TJ-: of course [00:32] TJ-: please, after you publish the script, just tell me git repos urls I I can test it. [00:33] TJ-: can I use 16.4 in my test? [00:34] plm: I'll write the script to handle whatever base container(s) you prefer. [00:34] TJ-: great =D I'm glad for that ;) [00:35] TJ-: I need to go now. What is the git repos url do will put the script? So tomorrow early I will check and test it [00:36] plm: basically it fetches both the x86 and arm images for the same release, exports and extracts them, copies files from the x86 into the arm, adds qeum-arm-static, repackages it then imports it as a local LXD image that you can then create containers from [00:36] plm: I'll send you info via the memoserv here on freenode [00:36] TJ-: momoserv is like as private message? [00:37] "private" [00:37] plm: yes. server will tell you there are messages when you log-on, or whilst online [00:37] TJ-: all right [00:37] TJ-: thank you! [00:37] TJ-: see you later [00:37] plm: then you can use "/msg memoserv read new" to read them [00:37] TJ-: all right [02:24] can i have a GUI with xfce4 on a ubuntu 16.04? installed xfce4 and lightdm but user doesnt login, what am i missing? [02:25] wr: you may have better luck in #ubuntu [02:47] sarnold, i mean ubuntu server 16.04 [02:48] wr: hmm that would indeed be a more awkward question.. #ubuntu folks are more accustomed to starting from *something* and #ubuntu-server folks aren't accustomed to doing gui things beyond their laptops.. :) [02:49] wr: well, okay, the Usual Troubleshooting checklists.. what happens? what do you expect to happen? do you get error messages in the log files or dmesg? [02:49] sarnold, yes but i need to test a thing and use webmin [02:51] sarnold, i did a fresh install of a minimal ubuntu server 16.04, then i added xfce4 for a test, but on login wasn't logging in, havent looked on dmesg yet, by the the way the ubuntu is on a virtual machine [02:52] "wasn't logging in" -- did you get a specific error message? from getty? gdm? lightdm? whhat? [02:56] sarnold, no error, how do you install xfce4 on the server from core ubuntu? maybe i missed a package [02:58] sarnold, i just did a apt-get install xfce4 lightdm on server maybe missed something, no? [03:00] wr: hrm. that's a pretty good starting point. did xorg get installed too? [03:00] sarnold, didnt install that package, need that one too i guess [03:09] sarnold, xorg is there [03:09] wr: when the machine comes up do you get a getty-style prompt to login or an xdm-style window to log in? [03:19] sarnold, forgot to say to on login have error "failed to start session", i get this https://imgur.com/Vc0QRGO [03:26] *too [03:27] wr: hmm, good luck. I dunno much about the gui-land once you've already gotten that far.. [05:43] rbasak: will you consider and reply to the last update in 1781529 ? [06:51] cpaelzer: ack [06:56] rbasak: and I think I owe you a link still [06:56] rbasak: was this the one about the force-badtest MP showing the two versions? [06:56] https://code.launchpad.net/~paelzer/britney/hints-ubuntu-ocfs2-cosmic-bump/+merge/355482 [07:11] Good morning [07:21] hi lordievader, how are you? [07:22] Doing good here :) [07:22] How are you, cpaelzer [07:22] ? [07:23] good as well === tomaw- is now known as tomaw [08:47] rbasak: since I saw you around, if you have a few minutes for that MP of yesterday [08:47] rbasak: it got an extra fix for the FTBFS that I encountered on tests [08:47] rbasak: would be at https://code.launchpad.net/~paelzer/ubuntu/+source/virt-manager/+git/virt-manager/+merge/356334 === Ussat-1 is now known as Ussat [12:10] good morning [12:25] cpaelzer_: rbasak: fwiw, mps are indeed switching automatically from approved to merged after an upload migrates. I remember we were wondering about that a while ago. Maybe it even wasn't working a while ago, but it is now [12:26] hi ahasenack [12:27] hi ahasenack [12:27] er [12:27] hehe [12:27] thanks for the info that they auto-merge [12:27] that is nice === cpaelzer_ is now known as cpaelzer [12:39] ahasenack: I think the reason it didn't work before was because the importer had stopped [12:40] could be [12:46] rbasak: is https://code.launchpad.net/~paelzer/ubuntu/+source/virt-manager/+git/virt-manager/+merge/356334 on your radar again, given cpaelzer's comment and latest change? [12:49] ahasenack: I missed it. I'll look at it now. [12:52] OK, not enough coffee this am, getting the ubuntu 18.04 live server amd is the correct one right to install server ? [12:53] * cpaelzer realized he need to ask ahasenack to ping rbasak more efficiently as direct pings are ignored :-P [12:53] ICMP over IRC is notoriously unreliable ;) [12:53] Oh, sorry. [12:53] I had only scrolled back to the most recent ping. Didn't realise the highlight in the window was for more pings :) [12:53] ping overflow! [12:54] can be abbreviated as Powerflow which makes it sound much nicer [13:09] cpaelzer: virt-manager +1 [13:09] thanks rbasak [13:20] rbasak: ahasenack: trying to split the external dependency lane I realized this isn't an issue for any of you [13:21] which makes me reconsider the split being useless for the team [13:21] if it is just me I can handle it with a few labels and be good for now [13:21] thereby aborted the split of that lane [13:21] although I cleaned up a few cards that I found hanging around [13:50] ahasenack: I can confirm that merged is auto-set after leaving it on approved this time [13:50] nice [13:51] ahasenack: and I see a time-matching mail of the "Ubuntu Git importer" [13:51] things are working! :) [13:51] ahasenack: so maybe when we wondered in the past that things didn't get merged, the importer was down for a bit [13:51] making it a more reliable service will help on that end as well [13:51] let's take that as a hint that the importer could be down [13:51] yes [13:52] one of our canaries :) [13:52] sprint on canary islands? [13:52] not quite :) [13:52] that would have a reasonable distance for you I'd think [13:53] there are very nice beaches here within a 400km radius [13:53] https://bit.ly/2NAe4gk ~250km [14:22] ahasenack: ticket 3465 was still egenrting diff the last time I checked [14:22] I set the signoff [14:22] but IIRC there are no good tests we can rely on [16:36] Hi, we've had our ubuntu 14.04 server x64 preseed break recently with linux-signed-generic-lts-xenial unmet dependencies. Has anyone seen this problem? [16:39] kierank: is it reproducible? [16:40] rbasak: yes [16:40] kierank: a possible failure mode is to do with a race condition. Which in theory shouldn't hit the updates pocket, but that's what I want to rule out. [16:40] kierank: hours apart? [16:40] kierank: can you confirm you don't have proposed pockets enabled? [16:41] rbasak: no proposed pockets [16:41] I don't have a date when it started failing but ~1-2 months ago we did our last preseed [16:42] Do you have the error log? [16:42] rbasak: yes, just need to find a way of getting it off system [16:43] can provide the preseed file as well [16:43] * kierank sets up ftp server [16:44] the pastebinit package and tool may be more convenient [16:44] there's also http://termbin.com/ [16:45] can't install anything [16:45] that's the nice thing about termbin, it can use nc or telnet or whatever you're already likely to have [16:45] didn't know that [16:46] very useful [16:46] http://termbin.com/5wbv [16:46] rbasak: ^ [16:46] Thanks [16:46] kierank: do you have a shell there available? [16:46] https://pastebin.com/fcp4XSxb [16:46] that's the preseed [16:46] rbasak: only the shell once it fails [16:47] That's fine. [16:47] Where it says "but it is not going to be installed", you can request it directly with apt [16:47] apt or apt-get doesn't seem to be installed [16:47] It will be in the chroot I expect. [16:48] Is it /target I think? [16:48] I can't remember if in-target is a binary available to you [16:48] I think it is [16:48] try "in-target apt-get install linux-signed-generic-lts-xenial" [16:48] Hopefully that will give you the same error? [16:49] just exists silently [16:49] immediately [16:49] exits* [16:49] ah errors in the shell [16:49] yes same error [16:50] try "in-target apt-get install linux-signed-generic-lts-xenial linux-signed-image-generic-lts-xenial=4.4.0.31.21 linux-headers-generic-lts-xenial=4.4.0.31.21" [16:50] Hopefully you'll see what I'm doing there. [16:50] apt-get will give you more information as to why not. [16:50] And you can follow down the tree by continuing to be more specific on the command line [16:51] thanks, we are trying this now [16:51] Eventually you should get something more helpful, like something specific not being available. [16:51] Then we can look into why that is. [16:53] There is also "E: Unable to locate package linux-headers-signed-generic-lts-xenial" in your pastebin, which seems odd. [16:54] I don't see that package either. [16:55] That doesn't exist - it doesn't make sense to sign the headers package like that - and I don't see where a dependency on that might be coming from. [16:55] what should the `ls -1` of `ubuntu/pool/main/l/linux-lts-xenial/` look like? [16:56] (January is working with me on this) [16:57] here's what my home mirror looks like http://paste.ubuntu.com/p/PQxfH7RNkt/ [16:57] I don't know. Take a look at archive.ubuntu.com? [16:57] we run `rename 's/~.*$/~14.04.1_amd64.udeb/' *` on `ubuntu/pool/main/l/linux-lts-xenial` since the ISO seems to be always broken [16:58] That sounds very wrong. [16:59] "ISO seems to be always broken" -> if it's a bug, then we can look into that, but you can't hack around like that and expect things not to break eventually. [17:12] Well unsure under which conditions filenames are broken but it was definitely an issue, after switching to 14.04.5 instead of 14.04.4 and removing the hack seems to work now [17:15] rbasak: thanks anyway [17:15] all sorted? [17:16] seems to be sorted, though the server wants to pxe boot first [17:16] fixed that and it's good [17:17] woot [17:17] did the failed kickstart https validation cause any trouble? [17:49] anyone ever fiddled with a kubernetes cluster on LXD? [17:49] and *not* get it to detect the local LXD storage pool even though it's there? [17:57] oops solved it, LXD version conflictions [18:38] sarnold: did you ever get a response on the ngx_brotli issues you opened? [18:40] teward: no [18:47] thought not :p [18:47] but meh they're probably "not caring" [18:47] :P [18:48] yeah. "not my code bug google" or similar [19:55] what package would you advise for transmission? [20:20] anyone see TJ- today? =D [20:47] no, he dropped off around the time of the huge netsplits.. [21:38] Not sure if I ask this here, but its a start, I have 18.04 installed and trying to get lxc to work and I'm having issues getting it up and running. i keep getting failed to start as the message. I am running this as non-root user as unprivileged [21:39] anything in dmesg? how about the logs? lxc or lxd? [21:40] I'm using the lxc-start command if that helps [21:40] Really new to lxc, used docker a bit and usually I'm using something like virtual box [21:40] did you pick lxc for a specific reason? lxd is a lot neater imho :) [21:41] I was told LFCE has a section on the Linux Containers, not sure if its lxc or lxd [21:42] malgorath: what's the exact command you're typing that fails? [21:42] malgorath: separately there's #lxcontainers that has the experts. We're a little more generalist here. [21:43] lxc-create -t download -n 100 works fine, then I run lxc-start -n 100 -F [21:43] Error I get is: The container failed to start. [21:44] If you're starting new, I suggest you use the lxc command, rather than the older lxc-* interface [21:45] Start with https://linuxcontainers.org/lxd/getting-started-cli/ [21:45] ah maybe the tut I was using was old [21:45] https://linuxcontainers.org/lxc/getting-started/ [21:45] ah I was close [21:45] IMHO that should be less prominent. [21:45] stgraber: ^ [21:47] I missed something or was that to someone else? [21:50] rbasak, thanks for that link, already a big help [21:50] That was to a colleague to whom I'm giving feedback about the docs having misled you (IMHO) :) [21:53] Thanks again ^.^ [21:53] You're welcome :) [23:47] I'm seeing weird errors building gnupg on ubuntu-rolling. Anyone know anything about that? [23:50] there's some warnings in the most recent build logs .. https://launchpad.net/ubuntu/+source/gnupg2/2.2.8-3ubuntu1/+build/15328452 [23:50] but it hasn't been rebuilt for over a month, maybe there's something newly busted? [23:52] seems like it, might be transitory. Just started seeing issues today . Here's an example: https://travis-ci.org/sobolevn/git-secret/jobs/439807252 [23:54] thats a lot of red [23:54] yeah