[08:39] <konstruktoid> i've might have missed the discussion, but is there a reason systemd-resolved isn't installed by default in kinetic?
[10:51] <rbasak> hallyn: no I don't think it did. Sorry.
[10:51] <rbasak> I suggest wrapping it in ssh_config with a custom ProxyCommand
[10:52] <rbasak> Then not just scp, but rsync and everythign else would work
[10:52] <rbasak> Eg. catching *.uvt or something like that, and then "ssh foo.uvt", "scp ... foo.uvt:" etc would work.
[10:52] <rbasak> I do that for lxd
[12:58] <hallyn> ok thanks.  i'll try that next time
[16:02] <sergiodj> lvoytek: hey, this is one of those dreaded "too many symbolic links" bugs: https://bugs.launchpad.net/ubuntu/+source/mysql-8.0/+bug/1988202
[16:03] <sergiodj> do we already have a bug that we use for this issue (so I can dupe this one)?
[16:09] <lvoytek> sergiodj: From what I can tell this is a new one. The too many symbolic links statement is a red herring though. I've never seen the symlink config fail before, it's possible they deleted a folder prior to the install
[16:09] <sergiodj> lvoytek: gotcha.  I will take a closer look at this one, then
[16:59] <mybalzitch> whats the "proper" procedure to set/change the vmlinuz, etc. symlinks in /boot ? I tried dpkg-reconfigure for the newer kernel, but that didn't change the symlinks, they still point to the older version HWE kernel that just got released
[16:59] <mybalzitch> other than manually rm'ign them
[17:14] <znf> mybalzitch, what are you trying to do specifically?
[17:15] <znf> the symlinks will always point to the highest version number, I believe
[17:15] <znf> The script that maintains those is linux-update-symlinks
[17:17] <znf> Also, the symlinks themselves are NOT used by grub.cfg
[17:17] <znf> I'm pretty sure it's just for legacy reasons
[17:20] <mybalzitch> I have the xanmod kernel package installed, it uses 5.19.5. ubuntu just pushed a new hwe kernel, 5.15.0-47, and the os now looks like its gonna boot from 5.15.0
[17:21] <alkisg> What's the output of: ls -l /boot | nc termbin.com 9999
[17:22] <mybalzitch> termbin is slow today
[17:49] <alkisg> Works for me; but you may use any other pastebin service
[18:11] <JanC> maybe the xanmod kernel install does some manual symlinking of its own that gets overridden by the official kernel packages?
[18:14] <sarnold> does the grub cfg builder even care about those symlinks?
[18:14] <JanC> no idea
[18:16] <JanC> it could still be better if the xanmod people would fix their packaging to be more like the Ubuntu packages maybe?
[18:17] <sarnold> or somehow otherwise suggest how to have *only* their kernels and not ubuntu kernels on the system
[18:19] <JanC> that might be harder to do because of dependencies (and keeping default kernels available in case their patched kernel fails might be good ;) )
[20:04] <mybalzitch> https://termbin.com/03zq
[20:11] <alkisg> mybalzitch: edit /etc/kernel/postinst.d/xx-update-initrd-links; add `set -x; exec 2>/tmp/debug.log` near its top
[20:11] <alkisg> Then reinstall your custom kernel packages and see if they do things properly, e.g. if they set DEB_MAINT_PARAMS
[20:12] <alkisg> If not, either patch that script, or file a bug report in that custom kernel package
[20:16] <mybalzitch> hnn
[20:16] <mybalzitch> hmm
[20:42] <ahasenack> does ubuntu server have a default iptables-save/restore systemd unit somewhere?
[20:43] <ahasenack> I wanted to avoid suggesting to add an iptables command to /etc/rc.local
[20:43] <ahasenack> it's a masquerading rule, which ufw doesn't support nicely, i.e., there is no `ufw masquerade foobar` command,  you have to edit a file anyway
[20:44] <ChmEarl> ahasenack, iptables-persistent
[20:44] <ahasenack> oh, nice
[20:44] <ahasenack> too bad it's in universe, but I guess I can use it
[20:46] <sarnold> there's a few netfilter and ipsets persistent thingies too https://termbin.com/7h5v