[06:11] <chl_> Probaly the wrong channel, but here goes anyway.. Has anyone worked with
[06:11] <chl_> ..getting leases from isc-dhcpd server parsed to a database?
[08:29] <jamespage> icey: I've uploaded your fix + 2.13.1 to https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/3534
[08:29] <icey> cool, thanks jamespage
[08:51] <icey> hey jamespage - it looks like cinder added a new dependency on zstd (https://pypi.org/project/zstd/) between the 2nd and 3rd milestones and afaict, we don't even package it, much less have it in main
[08:53] <icey> jamespage: 6 days ago: https://review.opendev.org/#/c/726765/
[08:54] <jamespage> icey: awkward but a) zlib is still the default and b) the import is only done if a use specifies zstd
[08:54] <icey> jamespage: sure, we can patch out those bits, but that also seems painful
[08:54] <icey> jamespage: "Cinder supports modern compression algorithms, unless it's used on Ubuntu" ?
[08:56] <jamespage> well c) is package zstd and just get it into universe for now
[08:56] <jamespage> we can demote to a suggests
[08:56] <jamespage> and just use during testing
[08:56] <icey> jamespage: c) seems reasonable and like the best approach to me
[08:56] <icey> I'll get something worked up
[08:56] <jamespage> zigo may be looking at it for Debian as I think he's working on victoria packaging
[09:07] <icey> jamespage: slightly related: do you have any idea why cookiecutter dropped after bionic?
[09:10] <jamespage> icey: https://launchpad.net/ubuntu/+source/cookiecutter/+publishinghistory
[09:10] <jamespage> comment from doko - was RM'ed in Debian first I think
[09:10] <jamespage> well it got removed from tesing
[09:11] <jamespage> indicating some sort of lack of maintenace
[09:11] <icey> jamespage: looks like I may be ignorable on the zstd bit - looking into that C poackage to see if it does the python bits too ;-)
[09:11] <icey> and annoying re: lack of maintenance
[09:12] <jamespage> icey: https://tracker.debian.org/pkg/cookiecutter
[09:14] <icey> ah, and nope, the zstd we have is the upstream c lib, the pypi bit seems to be from https://github.com/sergey-dryabzhinsky/python-zstd :-/
[09:28] <jamespage> icey: can you update your openvswitch bug for the SRU? it will need the normal SRU template completing including a test case etc...
[09:28] <jamespage> + subscribed to ubuntu-sru
[09:29] <icey> jamespage: this one: https://bugs.launchpad.net/ubuntu/focal/+source/openvswitch/+bug/1887177 ?
[09:29] <jamespage> OH I see that's already done - ignore me
[09:29] <icey> :)
[09:31] <jamespage> icey: right - uploaded for the SRU team to review
[09:31] <jamespage> thanks for your fix
[09:35] <frickler> jamespage: any chance of progress on https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1839592 ? or will we need to keep running pkgs from disco or eoan until we can deploy focal? which will be some time since we are only now upgrading from rocky to stein
[10:29] <rzgd> on the ubuntu server installer with pre-exiting and unlocker dm-crypt partitions, the installer is trying  to read keyfiles when not needed, and throws an error on block_meta.py line 1384, but being a preserved partition it shouldn't, error doesnt happen in the desktop version
[10:29] <rzgd> from ubuntu-20.04.1-live-server-amd64.iso when installing with existing partitions keep getting: An error occured handling 'dmcrypt-data': ValueError - encryption key or keyfile must be specified - Traceback (most recent call last): File "/snap/subiquity/1966/lib/python3.6/site-packages/curtin/commands/main.py", line 202, in main
[10:29] <rzgd> finish: cmd-install/stage-partitioning/builtin/cmd-block-meta: FAIL: configuring dm_crypt: dmcrypt-data
[15:38] <RoyK> hm… I have this 20.04 server and I use a static ipv6 address. I want to turn off automatic address assignment, but keep RA, since it gives me a router and such. I tried with auto, it works, I tried turning off RA in netplan, it doesn't, even if I set the same LL address as gateway as set on other machines on the network. Any idas?
[15:45] <RoyK> never mind - typo/PEBKAC
[23:56] <cliluw> Can an Ubuntu server use both AppArmor and SELinux?
[23:57] <sarnold> cliluw: not really; you might be able to install a fedora in an lxd instance and that might work, but I can't promise it'll work..
[23:57] <sarnold> stacking LSMs is a work-in-progress