[01:22] <keithzg[m]> Well shucks, I rebooted an 18.04 server after applying the latest updates, and now it's falling flat on its face; can't even quite figure out what's failing since login attempts hang.
[01:30] <keithzg[m]> Okay, a login attempt I made in-person *finally* succeeded after several minutes of waiting. I only get VT1; ctrl+alt+F# just shows a blinking cursor for anything else. Trying to run `top` (trying to quicjly see if it's just something somehow hogging the CPU) just sits there forever, as does `systemctl --failed`. Interrupting with ctrl-c works fine. I suppose I should just try and let 'em run and see if they
[01:30] <keithzg[m]> eventually return something...
[01:35] <docmur> Hey guys, I had an LVM array that crashed, which consisted of three drives.  One of the drives is dead but it's volume was across all three of the drives.  Is there a way to get the data from that one volume, to the extend it's possible, back?  I know the 3TB of files that were on the dead drive are gone, and some of the other files for sure, but could the good files on the other two be recovered?
[01:37] <keithzg[m]> Well shucks, `systemctl --failed` itself failed, with "Failed to list units: Connection timed out".
[01:40] <mason> heh
[01:41] <keithzg[m]> Aha, but with sudo it succeeded, and the problem units are accounts-daemon, mariadb, and networkd-dispatcher. Hrmm.
[01:44] <keithzg[m]> Manually starting accounts-daemon seemed to work, as did manually starting mariadb, which is good 'cause this is indeed our company database server so it'd be kindof . . . awkward if people came into work tomorrow and it was down :P
[04:38] <CarlenWhite> For some reason I feel like I stumbled across something that'd let you relay software packages from one server so several servers didn't have to go out into the net to fetch one package.
[04:39] <CarlenWhite> Of course could be imagining.
[04:48] <qman__> there are essentially two options for that, a caching proxy, or setting up your own mirror
[04:49] <CarlenWhite> Then a caching proxy is what I was thinking of. It'd be overkill to host a mirror for internal purpose.
[04:49] <qman__> setting up your own mirror is more involved and hard to justify unless you have a lot of machines using it
[04:50] <qman__> for the proxy, you also have two options, set it up as a transparent proxy that "just works", or set up a normal proxy and configure your apt clients to point at it
[04:52] <qman__> the trouble with a transparent proxy is making sure it doesn't interfere with anything else; the trouble with a regular one is doing the configuration on all your clients
[04:54] <CarlenWhite> I might end up doing it the regular method. Plus if I really need to configure it for many clients I'd write a script help remove some of the repetitiveness.
[04:56] <CarlenWhite> Plus this is for my own home net. Was doing a release upgrade on a few computers and figured it'd make sense to fetch all the packages from the local home server and have them all download it from one 15Mbit DSL connection
[04:56] <CarlenWhite> [...]local home server instead have them all[...]
[06:23] <roaksoax> g/win 14
[08:07] <TJ-> Can anyone confirm that 18.04 with strongswan/xl2tpd L2TP/IPsec (direct or via network-manager-l2tp{,-gnome}) will connect to Cisco IOS (870) - got Windows and OSX clients that'll connect but Ubuntu is failing. IKEv1 proposal is accepted but then there's just resent messages and the connection fails. Using aes128-sha1-modp2048/aes128-sha1
[09:15] <CarlenWhite> Oof
[09:15] <CarlenWhite> I really put something sideways in my server
[09:15] <CarlenWhite> It's running now but it isn't running well
[10:35] <CarlenWhite> Woo. Everything's fixed. And that was on botched do-release-upgrade on my part.
[10:43] <CarlenWhite> Basic gist, grub got botched and had to manually instruct how to boot. Boot partition damaged. Learn latest kernel got corrupted. fsck. Boot again. apt install --reinstall the damaged kernel. System works without intervention now.
[15:36] <elox_> Hello, I'm trying to install python3-pip on my fresh and nice bionic but it mysteriously fails.... whats happening?
[15:38] <tomreyn> elox_: we could make some wild guesses, or you could show some output.
[15:39] <elox_> https://paste.ubuntu.com/p/hNJtWqm4Wg/
[15:39] <elox_> tomreyn: ofc, apprechiate it
[15:39] <elox_> I've battled this for some time.
[15:40] <tomreyn> this is bug 1783129
[15:40] <ubottu> bug 1783129 in subiquity "Only "main" component enabled after install" [High,Confirmed] https://launchpad.net/bugs/1783129
[15:40] <tomreyn> !info python3-pip
[15:40] <ubottu> python3-pip (source: python-pip): Python package installer. In component universe, is optional. Version 9.0.1-2.3~ubuntu1 (bionic), package size 111 kB, installed size 585 kB
[15:40] <elox_> I must be doing something VERY wrong since I use the official image but it just seems borked.
[15:41] <tomreyn> "component universe"
[15:41] <tomreyn> oh it Is borked.
[15:41] <tomreyn> thus the bug report
[15:41] <elox_> It seems that if I use this image http://cdimage.ubuntu.com/ubuntu-server/bionic/daily/current/ it mysteriously works!
[15:42] <tomreyn> it's not mysteriously, if you look at the bug report i posted
[15:42] <elox_> Oh, I'll try look that bug.
[15:42] <elox_> Thank you alot!
[15:42] <tomreyn> welcome