[00:01] <penguinepimple> oh wow, that's pretty cool. so i guess that you're setting this up to manage a large scale workstation environment?
[00:18] <OerHeks> ehm, 800gb+ would be the whole mirror :-D
[00:18] <OerHeks> i think limiting to 18.04 would take less
[00:19] <OerHeks> i will use it to learn more about this stuff
[02:46] <JanC> you could probably also restrict it to the architectures you actually need, etc.
[02:50] <hggdh> OerHeks: I use apt-cacher-ng for my home network, right now only Bionic and Disco. For all machines I have (and they have different programs installed), apt-c-ng is using ~ 270G
[02:50] <JanC> how many architectures?  :)
[02:50] <hggdh> OerHeks: granted, there is a bit of cruft from previous releases
[02:51] <hggdh> i386/amd64
[02:52] <JanC> some people might want some armhf or arm64 too, but s390x is less likely to be used at home...
[04:34] <OerHeks> hggdh, thanks, noted
[12:10] <penguinepimple> a
[17:56] <jayjo> I have an application hosted in two different physical locations in order to be closer to clients. Currently my database is replicated using master-standby methods so writes can only happen in one place. Now I have to add additional sources of data from each of the regions. Is there an architecture pattern for this? I think I could keep the single source of truth application with replicated instances
[17:57] <jayjo> at different regions if I incorporated the new data stream inputs with a message broker
[17:57] <jayjo> The applications takes a data stream and performs computation and stores the result. Now there will be multiple data streams from multiple physical regions
[19:48] <lunaphyte> hi.  i have a computer that's not booting successfully after upgrading from 14 to 16.  grub loads, and starts the kernel, but it's complaining about not being able to mount /usr
[19:49] <lunaphyte> i've booted from an 18.10 server installer, and was planning to use the "rescue a broken system" function, but it seems to be gone
[19:49] <lunaphyte> am i remembering things wrong?  hadn't that function been there in the past?
[19:55] <Glorfindel> lunaphyte: was that only on the desktop iso? I can't remember if I ever saw it on ubuntu server isos
[19:55] <Glorfindel> maybe download an old iso and see if it's there?
[19:56] <lunaphyte> thanks, i was just thinking the same thing.  i'm downloading the desktop iso now
[21:12] <Glorfindel> https://i.imgur.com/qFKxkOG.png what do I set this too if I don't own a domain? I'm currently using duckdns for ease of memory, and have no plans to change this
[21:12] <Glorfindel> leave it? I don't recall ever setting this to anything
[21:12] <mybalzitch> if you leave it then any mail it attempts to send will stay local
[21:12] <mybalzitch> (to that machine)
[21:14] <Glorfindel> ok, so how do I find out what the proper setting is?
[21:15] <Glorfindel> I'm thinking I might just leave it though... I don't use this server for mail
[21:16] <Glorfindel> would gloomweaver.duckdns.org be my fqdn then?
[21:24] <mybalzitch> if that points to its external ip, then yes most likely
[21:25] <Glorfindel> ok, that's what I set it to, thanks mybalzitch
=== randomthoughts is now known as tops
[22:49] <lunaphyte> my upgrade from 14 to 16 didn't go so well - during boot, it fails to mount /usr, and drops to busybox/initramfs
[22:50] <lunaphyte> but the keyboard [usb] stops working when it does that
[22:50] <lunaphyte> how can i fix this, so i can continue troubleshooting?
[22:53] <lunaphyte> if it's of any value, here are a couple of screen shots: https://ibin.co/4TPvc6IPR9s4.png https://ibin.co/4TPvhijOsVbI.png
[23:10] <_KaszpiR_> restore from backup and better reinstall + copy files
[23:11] <TJ-> lunaphyte: still working at it?
[23:11] <lunaphyte> TJ-: yeah
[23:11] <TJ-> lunaphyte: are you sure it was /usr that could not be mounted? It is EXTREMELY unusual to have a separate file-system for /usr/, and unheard-of to be mounting it from the initialramfs :)
[23:12] <TJ-> lunaphyte: I'd expect you actually mean the root file-system
[23:12] <TJ-> lunaphyte: I'm also guessing the keyboard is connected via USB
[23:12] <lunaphyte> i'm pretty sure it's /usr, yeah
[23:12] <lunaphyte> this system does have a separate /usr, and that's what it complains about during boot
[23:12] <lunaphyte> yes, it's a usb keyboard
[23:14] <TJ-> lunaphyte: ahhhh, your 1st screenshot, I see it now. As I said it is unusual to have /usr as a separate mount. presumably that file-system cannot be found
[23:14] <TJ-> lunaphyte: let's first try to solve the lack of a keyboard
[23:15] <TJ-> lunaphyte: does the PC have both USB3 (blue) and USB2 (black) sockets?
[23:17] <TJ-> lunaphyte: the reason the USB keyboard works in GRUB is that the PC firmware/BIOS is handling it... once Linux starts it needs to load drivers for the USB controller, and the USBHID (Human Interface Device) and presumably one or both of those didn't happen. Often that is due to having keyboard plugged into a USB3 controller the kernel, or the initialramfs, doesn't have the kernel module for
[23:17] <TJ-> lunaphyte: so if possible moving the keyboard to a USB2 socket can work around that issue (most USB2 controllers are supported )
[23:19] <TJ-> lunaphyte: your screenshot already shows the usb EHCI PCI controller being managed, it also shows the Dell USB keyboard hub - so if this is an internal USB keyboard it is surprising it isn't working
[23:21] <TJ-> lunaphyte: I'd suggest rebooting, stopping at the GRUB menu, editing the kernel command line again (linux ...) as we talked about in #grub earlier, add add "nomodeset break=premount" and then press Ctrl+X to boot - that should stop the initialramfs at a busybox shell just before it starts trying to mount file-systems. If they keyboard works there you can then explore.
[23:23] <lunaphyte> thanks very much for all of this.  i'm stepping away for a bit to eat, but i'll try that as soon as i'm back