[01:37] <wingarmac> Is it possible to also mirror the ubuntu pro packages ?
[01:38] <wingarmac> When I tried to add these repos to mirror.list I've got apt-mirror: can't open index security.ubuntu.com/ubuntu-advantage
[01:46] <JanC> you'd need some authentication key; not sure if the mirror software supports that...
[01:55] <wingarmac> it's soved, I remove the deb-src from the mirror.list, and it did the trick.
[01:57] <wingarmac> Now, still find a way to add the Google repos to my mirror with apt-mirror. I did try a lot, and that is the last line I've found woring in my normal source file to apt-cache search the google package, so I did add it to apt-mirror, but it seems to packages are not added to my repo.. Here's the repo line for the google-chrome packages to mirror: deb [arch=amd64 signed-by=/usr/share/keyrings/google_linux_signing_key.gpg] 
[01:57] <wingarmac> http://dl.google.com/linux/chrome/deb/ stable main
[01:58] <wingarmac> now I get no error when launching apt-mirror, but no copy of the google-chrome-stable or else are done to my mirror
[01:59] <wingarmac> other third party software did work well, and are added to my repo, like webmin or eid.
[02:28] <wingarmac> nox I've suddenly Processing indexes: [SSPPPPPPapt-mirror: can't open index security.ubuntu.com/ubuntu-advantage/ubuntu-jammy-esm-infra-security//main/Packages in process_index at /usr/bin/apt-mirror line 800.
[02:28] <wingarmac> apt-mirror: can't open index security.ubuntu.com/ubuntu-advantage/ubuntu-jammy-esm-infra-updates//main/Packages in process_index at /usr/bin/apt-mirror line 800.
[02:28] <wingarmac> PPPPP]
[02:28] <wingarmac>  Because Iadded the google repo another way into mirror.list before launching apt-mirror again
[08:33] <inscw00table> I'm seeing some failures during an autoinstall of ubuntu22.04; failing to install 'efibootmgr', 'grub-efi-amd64', 'grub-efi-amd64-signed', 'shim-signed', and whining about the bad checksums of these files which seems due to the ubuntu.iso having been deleted:  "/ubuntu.iso (deleted) on /cdrom type iso9660"
[08:37] <inscw00table> This seems odd to me; this is an unattended network install; I would have expected the initram to copy the iso into ram or something so it doesn't go away.
[09:23] <rbasak> inscw00table: that suggests to me that your installation media was lost somehow - eg. loose USB connector. How is it being made available?
[09:27] <inscw00table> It's network booted from ipxe.
[09:28] <inscw00table> I was doing 5 of them in parallel and they all failed this way, 2 times in a row.
[09:30] <inscw00table> But this has worked before, reliably, so I expect something has changed in the infrastructure that is not yet apparent to me.
[09:31] <inscw00table> I don't know if it's supposed to be pulling these files from the "cdrom" iso, I've given it a network repository where it's supposed to be installing things form.
[09:32] <inscw00table> I have seen issues in the past where the "auto-detect if network install" logic goes awry and I wind up seeing goofy errors like this, where it's trying to install from an installation source I don't want it to; but nothing quite like this.
[12:58] <tomreyn> stgraber: i'm just going through the LXD demo / tutorial at https://linuxcontainers.org/lxd/try-it/ - and notice that for the "limited" guest which was created using "lxc launch images:ubuntu/20.04 limited -c limits.cpu=1 -c limits.memory=192MiB" (supposedly limited to one CPU core), the "lxc exec limited -- nproc" reports "2" rather than "1". So it seems that CPU core limitation does not actually work?
[13:02] <tomreyn> it does, however, apply after i ran all of "lxc config set limited limits.cpu=1", "lxc exec limited -- nproc" (still reporting "2"), "lxc restart limited", "lxc exec limited -- nproc" (now reporting "1").
[13:03] <tomreyn> so apparently, the limits applied by the 'launch' command do get registered, but are not taken into account for the first launch, but only after reboot.
[13:17] <tomreyn> interestingly, switching back and forth between 1 and 2 spu cores works fine / applies immediately after the guest has rebooted once.
[21:23] <teward> bryceh: RE: NGINX and backports, if Server wants to do a backport in the server special backports ppa or whatever i'm fine with it, but unless that is regularly available and easy to set up in the repos it might be more prudent for me to just backport $DEVEL_VERSION -> LTS via standard -backports
[21:29] <sdeziel> tomreyn: thanks for reporting, this is seemingly a bug, I'll check what's up with it
[22:39] <bryceh> teward, ok sounds good
[23:16] <sdeziel> tomreyn: https://github.com/lxc/lxd/issues/11462
[23:16] -ubottu:#ubuntu-server- Issue 11462 in lxc/lxd "`limits.cpu` provided at `lxc launch` time is not applied until the container is rebooted" [Open]