| cpaelzer | teward: double thanks - for http2 info and for the nginx merge, lets see how it passed proposed migration and behaves then | 06:31 |
|---|---|---|
| lordievader | Good morning | 07:02 |
| cpaelzer | hi lordievader | 07:08 |
| lordievader | Hey cpaelzer How are you doing? | 07:08 |
| cpaelzer | already planning how to reach conclusion of all my tasks before christmas (=impossible) :-) | 07:08 |
| cpaelzer | so good I'd say | 07:08 |
| teward | cpaelzer: indeed. | 07:08 |
| teward | rbasak: cpaelzer: FWIW as this is an LTS, I'mma touch base with the TB / infinity to determine if we can get an after-release update to the stable branch of nginx | 07:09 |
| teward | which *should* be a minimal diff at that point. | 07:09 |
| teward | similar to what we had for the 16.04 cycle | 07:09 |
| cpaelzer | teward: do you want a single agreement or a general MRE to do this for the full time of 18.04 ? | 07:10 |
| teward | cpaelzer: single-case | 07:10 |
| cpaelzer | ok | 07:10 |
| teward | cpaelzer: we had a 'single case' agreement for 16.04 up through release, with a special-case SRU for the version bump only in -updates, when Xenial was released. | 07:11 |
| cpaelzer | sound fine to me to do it the same way again | 07:12 |
| teward | such that we did standard FFes through release, and then SRU after that. | 07:12 |
| cpaelzer | yep | 07:12 |
| teward | yeah there's precedent but I want to reach out at least to the Release team / infinity who helped a lot with this in 16.04 | 07:12 |
| lordievader | cpaelzer: Good luck 👍 | 07:12 |
| teward | cpaelzer: and it's only during the LTS-cycle that I work on this, because getting to a stable branch is better than sticking to mainline :P | 07:13 |
| cpaelzer | absolutely reasonable for fixes throughout the cycle | 07:13 |
| teward | blurgh i'm tired but can't sleep | 07:13 |
| cpaelzer | I already wondered to see you here still | 07:13 |
| teward | yeah i'm not usually online at this hour, 02:15 here :P | 07:15 |
| rbasak | teward: thanks. An email to ubuntu-release@ should be sufficient I think. No need for the TB. | 07:24 |
| teward | yep, was thinking that recently. I need sleep though so... gonna try and sleep now :p | 07:26 |
| fstoltz | Regarding the 'ss' command for investigating sockets. Is it possible to get a live feed for example 'ss -t', getting the same view, but a "dynamic" view? Anyone knows about this? | 11:45 |
| ahasenack | I didn't know about ss, nice | 11:46 |
| fstoltz | ahasenack: :) | 11:49 |
| fstoltz | Just stumbled upon 'tcptrack', maybe that could be something | 11:49 |
| ahasenack | do you just want to see existing connections and their traffic perhaps? | 11:49 |
| ahasenack | try iftop in that case | 11:50 |
| fstoltz | Sory of ye, essentially just having a live TCP feed with info | 11:51 |
| fstoltz | sort of ye* | 11:51 |
| ahasenack | that will do it | 11:53 |
| ahasenack | you can pass it filters in the tcpdump format | 11:53 |
| ahasenack | like | 11:54 |
| ahasenack | iftop -i eth0 -n -f "port not 22 and not udp", which would skip all udp traffic and ssh, in case you are remotely logged in on the machine | 11:54 |
| fstoltz | ahasenack: Just installed iftop and having a look, it looks really good thank you | 11:58 |
| ahasenack | nice :) | 11:59 |
| ahasenack | cpaelzer: do you know if this is a valid use case? https://bugs.launchpad.net/ubuntu/+source/samba/+bug/1737534 | 12:18 |
| ubottu | Launchpad bug 1737534 in samba (Ubuntu) "smbd/nmbd don't restart after upgrade if started but disabled" [Undecided,New] | 12:18 |
| ahasenack | cpaelzer: tl;dr : | 12:18 |
| ahasenack | cpaelzer: systemctl disable smbd.service nmbd.service | 12:18 |
| ahasenack | cpaelzer: but the service is running still | 12:18 |
| ahasenack | (and systemctl start works just fine) | 12:18 |
| ahasenack | cpaelzer: then the package is upgraded | 12:18 |
| ahasenack | cpaelzer: after the upgrade, the service isn't running anymore | 12:18 |
| ahasenack | presumably because postinst does restart, or stop+start, and since the service is disabled, it doesn't start | 12:18 |
| cpaelzer | hi ahasenack | 12:19 |
| ahasenack | I can reproduce it, I just couldn't locate yet exactly in which maintainer script the "stop" happens | 12:19 |
| cpaelzer | hmm | 12:20 |
| cpaelzer | at first I wanted to say no | 12:20 |
| cpaelzer | but reading more it sounds interesting | 12:20 |
| ahasenack | let me see what systemctl restart does in that case | 12:21 |
| cpaelzer | in any case if we say "yes this is a problem" it is not a samba but a generic problem | 12:21 |
| ahasenack | restart works just fine | 12:21 |
| cpaelzer | which needs much (much++) wider discussion consideration | 12:21 |
| ahasenack | might be something related to the sysv compatibility layer | 12:22 |
| ahasenack | as samba as sysv initscripts shipped in /etc/init.d | 12:22 |
| cpaelzer | and as service? | 12:22 |
| ahasenack | and upstart | 12:22 |
| cpaelzer | or only syssv? | 12:22 |
| ahasenack | and systemd | 12:22 |
| ahasenack | what a mess :/ | 12:22 |
| ahasenack | all 3 are in the package | 12:22 |
| cpaelzer | that is often done for backportability | 12:22 |
| cpaelzer | only the systemd wins in that case | 12:22 |
| cpaelzer | but if there are hardcoded calls in the *inst files that might be an issue | 12:22 |
| ahasenack | systemd seems incomplete/not used | 12:23 |
| ahasenack | there is just /lib/systemd/system/samba.service | 12:23 |
| cpaelzer | I once had a case where the drop of a sysv script made the systemd service fail - as dh helpers added an invoke.rc to the sysv script which was mapped to start the systemd | 12:23 |
| cpaelzer | just to encourage the feeling of https://xkcd.com/1172/ being eveywhere | 12:23 |
| cpaelzer | ahasenack: if you think this is an interesting caseI tihnk you'd want to trace all calls to start/stop maintainer scripts are doing in this case | 12:25 |
| cpaelzer | ahasenack: will be a lot as the sysv will call but map to systemd if the name matches) | 12:26 |
| ahasenack | how can I do that? If I modigy the script inplace, but use --reinstall, the new one will overwrite my changes | 12:26 |
| cpaelzer | once summarized and reproducible without actual install this might be an interesting mail to ubuntu-devel to discuss in general | 12:26 |
| cpaelzer | ahasenack: there is a trick | 12:26 |
| ahasenack | I'm sure :) | 12:26 |
| cpaelzer | wanted to talk about it in the standup a while ago actually | 12:26 |
| cpaelzer | but people complained my time is over :-P | 12:26 |
| cpaelzer | ahasenack: https://trello.com/c/covf8RG6/543-virt-stack-for-1804#comment-5a2fd28ff56266c5c7aec3b4 | 12:28 |
| cpaelzer | ahasenack: I'm about to go to lunch, pleae check if this works for you | 12:28 |
| ahasenack | ok | 12:31 |
| ahasenack | hm | 12:33 |
| ahasenack | I thought it would be export DEB_MAINT_SCRIPT_DEBUG=1 or something sensible like that :) | 12:33 |
| ahasenack | deb packages are so easy, why would they need something sensible like that, right | 12:35 |
| cpaelzer | ahasenack: actually such a env var exists | 13:09 |
| cpaelzer | ahasenack: but it depends on the deb_helper which one (if any) they follow | 13:09 |
| ahasenack | in the meantime I suggested the person use policy-rc.d | 13:10 |
| cpaelzer | ahasenack: with the unpack/modify/pack I got most hard cases solved th ebest way | 13:10 |
| cpaelzer | does it work for his case ahasenack? | 13:10 |
| cpaelzer | policy I mean | 13:10 |
| ahasenack | he said he needed to check a disk, and then decrypt it (mount it probably), and only then start sambe | 13:10 |
| ahasenack | the decrypt bit might be impossible to do automatically, depending on his security policy | 13:11 |
| ahasenack | but he could at least avoid the error of trying to start samba at boot | 13:11 |
| ahasenack | if disk not mounted, exit | 13:11 |
| ahasenack | and the restart would work just fine | 13:12 |
| ahasenack | another option I have under my sleeve is to add the share at the same time his scripts mount the disk | 13:12 |
| ahasenack | it doesn't have to be hardcoded in smb.conf | 13:12 |
| ahasenack | from the getgo | 13:13 |
| jamespage | coreycb: I uploaded a snapshot of ceilometer - that's going to have some charm impact as both -api and -collector services have gone from upstream and the packaging! | 14:13 |
| coreycb | jamespage: oh wow. that is the advantage to releasing milestones. glad you did a snapshot early. | 14:14 |
| coreycb | jamespage: we should be all caught up on queens packaging now | 14:14 |
| coreycb | jamespage: probably need some promotions from staging | 14:14 |
| jamespage | coreycb: I've been sweeping those throught hourly | 14:19 |
| coreycb | jamespage: great, thanks | 14:19 |
| cpaelzer | rbasak: could you check if 1737984 is also hash-abi-break tag worthy? | 14:24 |
| cpaelzer | it is on pristine-tar not the actual package content | 14:24 |
| jamespage | coreycb: I'll do the same with aodh and panko | 14:24 |
| jamespage | coreycb: oh there was a bit of brokenness around swift as well - that should be fxied(needed to backprot swift-plugin-s3) | 14:25 |
| coreycb | jamespage: ok | 14:28 |
| coreycb | jamespage: horizon is partially py3 now. moving openstack_auth in tree forced me to do it for that code. | 14:29 |
| jamespage | coreycb: ok | 14:30 |
| rbasak | cpaelzer: thank you for the report. I'm not sure about hash-abi-break because a change to pristine-tar won't change the main imported commit graph, but a change to upstream/ might. So I'll tag it for now. | 14:35 |
| cpaelzer | that is exactly why I was unsure | 14:39 |
| cpaelzer | rbasak: and right chan :-) | 14:39 |
| * rbasak hands cpaelzer a sticker :) | 14:58 | |
| cpaelzer | yay | 14:59 |
| jamespage | coreycb: ok aodh snapshot uploaded | 15:31 |
| jamespage | coreycb: fwiw I'm seeing some behavioural diff between stestr and testr | 15:32 |
| jamespage | had to switch heat and aodh back to using testr directly | 15:32 |
| coreycb | jamespage: hmm ok | 15:33 |
| cpaelzer | rbasak: stgraber: and other ipv6 experienced users around - I'd like to ask for expertise on https://bugs.launchpad.net/ubuntu/+source/ntp/+bug/1737998/comments/2 | 15:34 |
| ubottu | Launchpad bug 1737998 in ntp (Ubuntu) "trying to bind on all interfaces is a good default, but fails on ipv6 link local" [Undecided,Incomplete] | 15:34 |
| cpaelzer | rbasak: stgraber: the actual issue that got me there doesn#t exist atm, but I wonder in general if ntp binding on those is ever wanted/useful | 15:34 |
| DevilTiger | i've downloaded a xenial cloud image; how do i go about logging into it? | 15:42 |
| Nafallo | DevilTiger: ssh -l ubuntu ip :-) | 15:44 |
| Nafallo | password is also ubuntu | 15:45 |
| DevilTiger | oh, ok. well that was easier than i thought. was overthinking it. now if i can just get it to boot | 15:45 |
| DevilTiger | stuck on blk_update_request: I/O error, dev fd0, sector 0 | 15:45 |
| cpaelzer | ahasenack: on the reply of powersj will you file a/the bug(s) for the discussion or do you want me to file my "extras" that came up separately | 15:46 |
| cpaelzer | ? | 15:46 |
| ahasenack | I think it's two issues | 15:46 |
| ahasenack | one a bug, one an enhancement request | 15:47 |
| ahasenack | each files his own | 15:47 |
| ahasenack | as soon as I understand the difference between sponsor and signer | 15:47 |
| ahasenack | in another topic | 15:48 |
| powersj | thank you both :) | 15:48 |
| ahasenack | why do all the samba bugs turn into support requests so easily | 15:48 |
| cpaelzer | ok then I'll file my extra parts | 15:48 |
| cpaelzer | ahasenack: actualyl let me file it in one | 15:50 |
| cpaelzer | it really is the detection that is broken | 15:50 |
| cpaelzer | while writing I found another issue | 15:50 |
| cpaelzer | I'll start with one issue for it | 15:51 |
| ahasenack | ok | 15:51 |
| cpaelzer | and whoever works on it can decide to break out pieces if one is easier to sovle than the other | 15:51 |
| cpaelzer | ahasenack: powersj: https://github.com/canonical-server/dev-summary/issues/7 | 15:55 |
| cpaelzer | powersj: no PR at hand (yet?) - sry | 15:55 |
| powersj | cpaelzer: haha shucks :) | 15:56 |
| powersj | cpaelzer: and of course, didn't expect one, but didn't want to loose your request | 15:56 |
| cpaelzer | sure | 15:56 |
| cpaelzer | the urgency is low | 15:56 |
| cpaelzer | but I think especially the "invite to test from proposed" might be good | 15:57 |
| cpaelzer | participation in this is too low anyway | 15:57 |
| cpaelzer | so every bit helps | 15:57 |
| cpaelzer | rbasak: can we all get another sticker please ? | 15:57 |
| powersj | totally agree | 15:58 |
| cpaelzer | when LP has its short hickup once a day it is nice to get all updates missed in that ~10 minutes at once | 16:03 |
| cpaelzer | I always feel like "WTF happened to my inbox" | 16:03 |
| DevilTiger | @Nafallo: ubuntu/ubuntu doesn't work | 16:04 |
| * rbasak hands out stickers all round. But only to people active in _this_ channel :) | 16:08 | |
| rbasak | DevilTiger: Ubuntu cloud images have no default password. Otherwise they'd be insecure in production. | 16:09 |
| rbasak | DevilTiger: you can either modify the image, or boot it via a tool that provides cloud-init with the desired authentication mechanism (usually a ssh public key but a plaintext password can also be used if you insist) | 16:09 |
| rbasak | DevilTiger: to help you further, please explain how you're booting the image. | 16:10 |
| rbasak | DevilTiger: if you want to hack the image locally for dev/test purposes, there's a handy tool "mount-image-callback". | 16:10 |
| rbasak | That lets you mount the image in place. | 16:11 |
| DevilTiger | i'm using hyper-v to mount vdk file that i converted from a ova file | 16:16 |
| DevilTiger | @rbasak: how would i go about doing this? | 16:41 |
| rbasak | DevilTiger: if not on Ubuntu or another Linux, then I'm not sure, sorry. | 16:42 |
| rbasak | Perhaps the easiest way is to create a config drive. | 16:43 |
| DevilTiger | i have linux subsystem on windows. if that wont work i could i could fire up 14.04 in a VM | 16:43 |
| rbasak | Sorry I think that's technically "NoCloud" | 16:44 |
| rbasak | DevilTiger: https://cloudinit.readthedocs.io/en/latest/topics/datasources/nocloud.html | 16:44 |
| DevilTiger | im not sure what to do with that information | 16:44 |
| rbasak | DevilTiger: two steps. | 16:45 |
| DevilTiger | i dont see any steps listed anywhere | 16:45 |
| rbasak | DevilTiger: 1) create a cloud-init metadata file that supplies your ssh public key or ubuntu password in the correct format | 16:45 |
| rbasak | DevilTiger: 2) make that file available to the booting cloud image | 16:46 |
| rbasak | DevilTiger: for step 1, see https://cloudinit.readthedocs.io/en/latest/topics/examples.html | 16:46 |
| DevilTiger | i don't think i can do this. i don't understand the underlying concept. i think i'll just start with a non-cloud image | 16:46 |
| rbasak | DevilTiger: the underlying concept is that all cloud images start off the same. That provides production reliability. | 16:47 |
| rbasak | DevilTiger: but a consequence is that on first boot the image needs to find the stuff that'll differentiate it into what you want it to be. | 16:48 |
| DevilTiger | i get that. i don't understand what cloud init is or what a metafile is or where to put it or how or the syntax. the link has several code examples that i have no idea which one is what i need | 16:48 |
| rbasak | DevilTiger: cloud-init is the component inside the cloud images (of most distributions) that turns the image into what you want. | 16:48 |
| rbasak | It might be easier to learn this if you start with a real cloud rather than hyper-v. | 16:49 |
| rbasak | Since most people here won't know much about hyper-v's tooling. | 16:49 |
| dpb1 | DevilTiger: you need to inform that image about your ssh ID, the cloud image isn't something that you can boot and use like that. | 16:49 |
| rbasak | Whereas on Ubuntu as a host we have tooling that does this all automatically by default./ | 16:49 |
| DevilTiger | i get that. i don't get how to tell it what my ssh keys are. | 16:49 |
| dpb1 | DevilTiger: you are on windows? | 16:49 |
| DevilTiger | especially since i'm using a vhd file, not an img file. i am on windows but i have access to linux in a VM | 16:50 |
| DevilTiger | i guess i'll just try to create my own image | 16:50 |
| dpb1 | DevilTiger: you should look at vagrant then? I mean, if the WSL stuff isn't doing what you need. | 16:50 |
| DevilTiger | tried vagrant, too much of a PITA to get running on server 08 R2 | 16:50 |
| dpb1 | ok | 16:51 |
| dpb1 | ya, I mean, basically, what vagrant is doing is interacting with cloud-init to set that ssh key inserted | 16:52 |
| DevilTiger | i'm trying to get xenial running in a VM so i can have pi-hole doing its thing on our windows server. i figured using a cloud image was the easiest way to get going fast without having to install it. | 16:53 |
| dpb1 | DevilTiger: or, you use the standard 16.04 server .iso that has a UI that walks you through options. | 16:53 |
| DevilTiger | thats what i'm doing now | 16:53 |
| dpb1 | DevilTiger: ya, really, that is what I would do, unless it's at scale, then I would look into automation | 16:53 |
| dpb1 | but if it's a one off, that's what the server .iso is for | 16:53 |
| DevilTiger | oh no this is just for a small office of 3-5 people. tired of ads | 16:53 |
| DevilTiger | installing now, i should've just done this at first. silly me | 16:54 |
| dpb1 | DevilTiger: fwiw, for 18.04, we are making a better single-system installer called subiquity that doesn't have the miriads of questions that the 16.04 one does. doesn't help you now, but figured I would let you know about it | 16:55 |
| DevilTiger | oh sweet | 16:55 |
| Ussat | dpb1, when you plan to have Aplas out of that, I would be interested in testing | 17:18 |
| rbasak | dpb1: I prefer even one-off installs to be reproducible if they are going to be used for some production purpose. | 17:21 |
| dpb1 | rbasak: can't argue, but there is also something to be said, for just want to install this one thing and use it. | 17:27 |
| Nafallo | time permitting, I'll build my own minimal image for lxd, and then use ansible to get it where I want it :-) | 17:28 |
| Nafallo | so "cloudy", but for production. | 17:29 |
| Nafallo | *shrugs* | 17:29 |
| rbasak | Nafallo: very easy to trip up badly if you try hand rolling that kind of thing. | 17:29 |
| Nafallo | rbasak: how so? | 17:30 |
| rbasak | For example, ssh host keys. | 17:30 |
| rbasak | cloud-init is the tool where all the distros put all the knowledge on how to do it right. | 17:30 |
| Nafallo | rbasak: pre-installed ssh, host keys erased and a tiny systemd unit for running ssh-keygen -A oneshot :-) | 17:30 |
| rbasak | Nafallo: and every other gotcha that I haven't mentioned | 17:31 |
| dpb1 | Ussat: it's ready to test now, if you are interested: http://cdimage.ubuntu.com/ubuntu-server/daily-live/current/ | 17:31 |
| rbasak | Nafallo: well done. You've reinvented cloud-init :) | 17:31 |
| Ussat | Thanks, will do | 17:31 |
| Nafallo | rbasak: nah. that and pre-creating the ansible user is pretty much it. ;-) | 17:31 |
| Ussat | got a full esxi test system to myself | 17:31 |
| rbasak | /etc/hosts? | 17:32 |
| rbasak | /etc/hostname? | 17:32 |
| Nafallo | rbasak: also, I said small... debootstrap --include=python,netbase,iproute2,sudo,lsb-release,openssh-server --variant=minbase rootfs xenial http://se.archive.ubuntu.com/ubuntu/ | 17:32 |
| Ussat | dpb1, any particular way you all want feedback ? carrie pigeon, smoke signals ? | 17:32 |
| Nafallo | rbasak: them two is lxd templates :-) | 17:32 |
| rbasak | Anyway, it's Free Software. You're Free to reinvent the wheel :) | 17:32 |
| dpb1 | Ussat: haha | 17:32 |
| rbasak | I just wouldn't recommend that to others to use in production. | 17:33 |
| rbasak | Since it's quite error-prone, even if a good learning experience. | 17:33 |
| Nafallo | rbasak: I'm not doing it to reinvent the wheel. I'm removing stuff I don't need to get a smaller attack vector, amongst other things :-) | 17:33 |
| rbasak | Most people don't compile their binaries in production either. | 17:33 |
| dpb1 | Ussat: probably here: https://bugs.launchpad.net/subiquity | 17:33 |
| rbasak | This is just the same thing but one level up. | 17:33 |
| dpb1 | rbasak: oh come on, gentoo ftw | 17:34 |
| Ussat | done deal.....I work at a research hospital, and our labs are exclusively Ubuntu, so I might have some help with the testing | 17:34 |
| rbasak | Nafallo: by deviating from everyone else you're also introducing attack surface: everything you've done differently is subject to your mistakes. | 17:34 |
| * dpb1 waits for X to compile | 17:34 | |
| rbasak | Nafallo: we make mistakes too, but we have the benefit of a large number of people looking and examining what we're doing. | 17:34 |
| Ussat | marking your calendar dpb1 ? | 17:34 |
| rbasak | Someone finds a problem and we fix it for everyone. You'll get left out. | 17:34 |
| Nafallo | ehrm. its not like I won't use packages... | 17:35 |
| dpb1 | Ussat: what's that? | 17:35 |
| Ussat | waiting for X to cmpile | 17:35 |
| Ussat | have you marked on the calendar a target date | 17:35 |
| Nafallo | what sort of problem would you fix outside of a package in this scenario? :-) | 17:36 |
| rbasak | Nafallo: it's not just the packages. Problems, including security problems, can also get introduced in the interactions between the moving pieces. | 17:36 |
| rbasak | ssh host keys is just one example of that. | 17:36 |
| rbasak | It's an example of an entire class of problems. You can't say that you're fine because you've covered one instance of that class. | 17:36 |
| dpb1 | Ussat: haha | 17:37 |
| dpb1 | Ussat: no kidding | 17:37 |
| dpb1 | you know, I ran gentoo for a while for fun. for me, I knew it was a problem when I started scheming ways to set up distcc so I could compile faster and update packages faster. | 17:37 |
| Ussat | dpb1, yea I did also, needed a space heater :) | 17:38 |
| Nafallo | heh. I remember gentoo as well. ran it for almost a year solid before moving back to Debian :-) | 17:39 |
| Ussat | stage1 installs FTW | 17:39 |
| Nafallo | that was before they started with pre-compiled binaries and installers and stuff though :-P | 17:39 |
| dpb1 | Ussat: :) | 17:39 |
| * Ussat remembers doin a stage 1 on a Pentium4 | 17:45 | |
| * Ussat cries | 17:45 | |
| Nafallo | I did my first gentoo install on a P200 with 16MB memory IIRC. took a week to get Fluxbox :-) | 17:45 |
| Nafallo | this would have been around 98-99 maybe? | 17:46 |
| Ussat | yup | 17:46 |
| Ussat | My first linux was with Gentooo, thats also about the time I started drinking :) | 17:47 |
| Ussat | go figure | 17:47 |
| Nafallo | 26 July 2000; 17 years ago ← a little later :-) | 17:47 |
| Nafallo | initial release date, btw | 17:47 |
| teward | rbasak: FWIW i think the nginx autopkgtests are 'working', but... http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#nginx - looks like some arm64 tests are hung, they've been in that state for a while now | 18:33 |
| rbasak | cpaelzer: regarding 1737998, I was under the impression that NTP over local link IPs should be fine in principle, and was going to ask if there was a specific case where that doesn't work. | 18:48 |
| rbasak | cpaelzer: but I think you've already asked that? | 18:48 |
| hdon__ | hi all :) i've been using linux since i was like 13, ten years later, i'm starting to identify some foolish notions i still harbor. i used to think that by deleting password with passwd -d <user> that i was disabling password authentication for that user. that my be so, but the user can also reset their password without knowing their existing password at that point. | 18:57 |
| hdon__ | my question is: what's the correct procedure for password reset? | 18:58 |
| Ussat | passwd <username> | 18:59 |
| metastable | hdon__: passwd -l <username> will disable password authentication. | 21:07 |
| DevilTiger | installing ubuntu server with only 512MB of RAM takes forever | 21:28 |
| dpb1 | DevilTiger: uh, ya | 21:32 |
| sarnold | I'll be curious if it even completes.. | 21:32 |
| DevilTiger | heh i had to restart it once due to grub failing. unplugged ethernet (why that has anything to do with grub failing is beyond me) and its gotten past that. at "running update-initramfs right now | 21:33 |
| DevilTiger | annoyed me enough to order 3gb more ram for the hypervisor its under. 1 GB just ain't cuttin it | 21:34 |
| sarnold | that 'unplug ethernet' thing sounds vaguely familiar, but from ages ago... | 21:34 |
| DevilTiger | says install is completed. waiting on reboot now | 21:35 |
| DevilTiger | since i didn't autoconfig my lan adapter what would i do to set that up? dhclient wlan0 ? | 21:40 |
| sarnold | /etc/network/interfaces .. mine is entirely too simple: | 21:41 |
| sarnold | auto enp5s0f0 | 21:41 |
| sarnold | iface enp5s0f0 inet dhcp | 21:41 |
| sarnold | (with lo of course, same as yours :) | 21:41 |
| DevilTiger | i've added iface wlan0 inet dhcp to my interfaces file. that isn't going to be enough | 21:43 |
| DevilTiger | as there is no wlan0 device | 21:43 |
| sarnold | right, either rename the device or live with the name it has :) | 21:43 |
| DevilTiger | uh maybe i'm confused but there is no device | 21:43 |
| sarnold | and no idea about how to manage a wifi card via /e/n/i, sorry | 21:43 |
| DevilTiger | like i said, there is no wlan0 | 21:44 |
| DevilTiger | err excuse me, i'm not adding wifi. eth0 | 21:44 |
| sarnold | so, "ip l" doesn't show you the nic you expect to be there? | 21:45 |
| sarnold | maybe you need to load a kernel module for your nic | 21:45 |
| DevilTiger | the NIC is there, yes. eth0 | 21:45 |
| DevilTiger | pihole | 21:47 |
| DevilTiger | err wrong channel for the last one | 21:47 |
| DevilTiger | adding eth0 to the interfaces file did it. ty. figured there was more to the process | 21:51 |
| hdon__ | metastable: ahhh thanks :) | 22:02 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!