[09:11] <rbasak> hallyn: it did not :(
[09:11] <rbasak> It's difficult to fix blog posts following design changes
[09:11] <rbasak> I intend to sit down and write up current docs in sphinx at some point
[09:12] <rbasak> In the meantime, anything specific you need that I can help with?
[12:29] <LeMike> hello there. Is it possible to join a server (via Mosh or SSH) but actually be in a `screen` automagically? I need a fail-safe solution for long disconnects, so that I can continue my work after getting connected again.
[12:32] <rbasak> LeMike: try "byobu-enable". byobu can be configured to use either tmux or screen I think, defaulting to tmux.
[12:34] <LeMike> rbasak: Ah ok, so I set the user shell to byobu ? And when the user accidentially disconnects then it stays open for a day or so? I try it. Thanks :)
[12:37] <rbasak> LeMike: it's not a shell. I think it configures your shell to launch byobu when you get a new shell, if it isn't runnign already. Or something like that.
[12:37] <rbasak> If two people connect they see the same thing.
[12:40] <frickler> cpaelzer: could you please point me to the sources for UCA versions of libvirt? in particular I'm looking for the delta from 5.0.0-1ubuntu2.6~cloud1 to 5.0.0-1ubuntu2.6~cloud2.1 (stein)
[12:40] <cpaelzer> hi frickler, let me try to find those for you
[12:42] <frickler> cpaelzer: or, possibly related, do you know of a situation where nova live-migration would log "Migration running for 0 secs, memory 100% remaining; (bytes processed=0, remaining=0, total=0)" and then go on with that for a day or more without any change?
[12:45] <cpaelzer> frickler: I don't remember seeing this situation
[12:46] <cpaelzer> frickler: a long long time ago I was tying to force this case artificially (for testing)
[12:46] <cpaelzer> but it never got stuck at 100%
[12:46] <cpaelzer> that really looks more like a bug in regard to not-started migraton or loosing an event or such
[12:48] <cpaelzer> the version you refer to was created for https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1915811
[12:48] <ubot3> Launchpad bug 1915811 in libvirt (Ubuntu Groovy) "Empty NUMA topology in machines with high number of CPUs" [Undecided, Fix Released]
[12:48] <cpaelzer> but for all the delta I need a while
[12:48]  * cpaelzer is searching the right PPA ..
[12:48] <cpaelzer> that should be bionic-stein I think
[12:50] <cpaelzer> here is the staging PPA https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/stein-staging
[12:50] <cpaelzer> and thereby this filter would give you all the past builds of it https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/stein-staging/+builds?build_text=libvirt&build_state=built
[12:51] <cpaelzer> which leads you to the full builds at https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/stein-staging/+build/21134685 and https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/stein-staging/+build/20167525
[12:51] <mgedmin> I hate apt's error reporting sometimes
[12:52] <mgedmin> E:Failed to fetch http://apt.postgresql.org/pub/repos/apt/dists/focal-pgdg/InRelease  403  Forbidden [IP: 192.168.1.3 8000], E:The repository ''http://apt.postgresql.org/pub/repos/apt focal-pgdg InRelease'' is not signed.
[12:52] <cpaelzer> frickler: so it turns out that ~cloud2 an ~cloud2.1 wree about the same change
[12:52] <mgedmin> this means "your local squid proxy is not allowing that url"
[12:52] <cpaelzer> therefore the bug (an threby the change discussed in there) is the only change between the version you asked for
[12:52] <mgedmin> but my eyes focus on the 'is not signed' and I debug my ansible's apt_key invocation for thirty minutes
[12:54] <frickler> cpaelzer: thanks a lot, that looks like it would not be related to my issue. I'll talk with nova folks next
[12:56] <cpaelzer> frickler: there are a few bugs worth to read, I'll send a query to spare the channel too mcuh flooding
[13:09] <hallyn> rbasak: no, i was just wondering what the state was and what the server team wasdoing these days :)
[13:09] <hallyn> thanks.
[15:08] <raub> tomreyn: you mentioned yesterday preseed is dead. What is the replacement?
[15:12] <mgedmin> https://discourse.ubuntu.com/t/autoinstall-generator-tool-to-help-with-creation-of-autoinstall-files-based-on-preseed/21334
[15:13] <peetaur> Howdy...so what's the solution to apparmor not working in Ubuntu 18.04 nor 20.04? I tried 18.04 with default 4.15 and hwe 5.4 kernel, and 20.04 with the default 5.4 kernel, which all fail, and tried 18.04 with mainline 4.19.188 which works.
[15:14] <peetaur> audit.log doesn't show what's wrong, and what it shows is malformed (name="" or name="dev/null" (should be eg. "/dev/null" or something else probably)), so aa-logprof doesn't  update the profile to work (but don't need to probably, since 4.19.188 works with the profile)
[15:19] <peetaur> and just tried the 4.19.188 in 20.04 ...works fine too, but has some of the malformed stuff too
[15:19] <peetaur> type=AVC msg=audit(1619536717.319:196): apparmor="DENIED" operation="file_mmap" info="Failed name lookup - disconnected path" error=-13 profile="/usr/sbin/apache2" name="" pid=1665 comm="apache2" requested_mask="wr" denied_mask="wr" fsuid=0 ouid=0
[15:25] <sdeziel> peetaur: you probably need to add flags=(attach_disconnected) to your apache2 profile
[15:26] <sdeziel> otherwise it has trouble with the mount namespaces used by the apache2.service (PrivateTmp=)
[15:26] <peetaur> I tried by running apache manually without systemd and had the same issue... will that still fix it?
[15:30] <peetaur> oh it seemed to fix it ... it works, and now I only see this:   type=AVC msg=audit(1619537376.666:209): apparmor="DENIED" operation="file_mmap" profile="/usr/sbin/apache2" name="/" pid=1801 comm="apache2" requested_mask="wr" denied_mask="wr" fsuid=0 ouid=0
[15:50] <smoser> rbasak: what would you say would be the correct / most up to date information on git-ubuntu . I'm specifically interested package delta maintenance work flow
[16:17] <peetaur> thanks for suggesting attach_disconnected btw... it seems to be the only workaround for this
[16:18] <rbasak> smoser: https://github.com/canonical/ubuntu-maintainers-handbook/blob/master/PackageMerging.md maybe
[16:19] <rbasak> smoser: https://wiki.ubuntu.com/UbuntuDevelopment/Merging/GitWorkflow is the old page
[16:35] <smoser> rbasak: thank you.