[06:33] <lordievader> Good morning
[07:36] <stunts513> any idea why i'd have network access from the terminal but not in the gui? i setup some network bridges and i can't seem to access any resources on one in a browser but i can wget the index page just fine from a terminal. i was doing this by ip too, not hostname.
[09:13] <lordievader> That sounds quite impossible. (Unless the browser is using some other protocol/or doh)
[15:15] <mason> Ah, he's gone.
[15:15] <mason> lordievader: Possible if it's a per-application block on the client side.
[15:17] <lordievader> If that is the case I'd sort of expect stunts to know about it. Guess we'll never know.
[15:26] <TJ-> if stunts512 is using IPv6 and the addresses are link-local, most browsers won't work
[15:35] <lordievader> Ah, additional details :)
[15:37] <frickler> cpaelzer: seems I rediscovered https://lists.gnu.org/archive/html/qemu-devel/2020-07/msg05235.html with https://gitlab.com/qemu-project/qemu/-/issues/693
[15:37] <frickler> doesn't look like it has really been fixed, though. did you something about it for Ubuntu pkgs?
[15:37] <frickler> +do
[15:41] <frickler> this will likely hit OpenStack CI once we start consuming Jammy. so far discovered it only via Bullseye, because Focal still has qemu-4.2
[18:23] <zerosum> anyone's ever encountered this error while getting a lets encrypt cert? certbot: error: unrecognized arguments: --dns-google-credentials creds/google-dns-api-cred.json
[18:23] <zerosum> I installed certbot via snap 
[18:44] <tomreyn> zerosum: probably anyone who tried to use this plugin provided functionality which does not come with certbot by default. see the Note (blue box) at https://certbot-dns-google.readthedocs.io/en/stable/
[18:47] <zerosum> tomreyn I see, thanks for that reference doc. Do I need DNS authentication to be with my DNS provider or can it be with anyone?
[19:24] <Avago_Broadqual> Are there any gotchas with doing a upgrade install from 18.04->20.04? I am running a ppc64le server mostly using it for LXC/KVM and MySQL
[19:26] <Avago_Broadqual> upgrade using do-relase-upgrade*
[19:32] <sarnold> Avago_Broadqual: that may bring along an entirely new version of mysql, it might be worth looking at that specifically before starting
[19:34] <Avago_Broadqual> but the LXC and KVM stuff should be seamless?
[19:36] <sarnold> I hope so, those feel like upgrade paths that ought to have been well-worn by now
[19:36] <Avago_Broadqual> right? i just wanted to ask someone before going ahead
[19:37] <Avago_Broadqual> I have had some issues in the past with do-release-upgrade on GUI based ubuntu systems
[19:37] <sarnold> most problems seem to come from third-party repositories
[19:40] <Jackal-o-lantern> Hello, I've recently had an issue with two completely different clusters of machines. one on 18.04 and one on 20.04. Both have NFS for /home which has encountered some issue. This prevents users from logging in as the system will stall trying to read from their homedir. While this is understandable, the case is that I also created accounts under /admin_home (for example) of another user who would not have their home directory mounted, and 
[19:40] <Jackal-o-lantern> the login still freezes for *this* user too. Freezes over ssh or even directly at the terminal with a screen and keyboard. How can I make a correct failsafe user where login won
[19:40] <Jackal-o-lantern> won't freeze then NFS dies somehow?
[19:40] <Jackal-o-lantern> when*
[20:38] <tomreyn> zerosum: i haven't really looked at this plugin, but i assume it sets up dns records for validation on you googles' authoritative dns server (which they probably provide as part of their domain selling business). you couldn't just use any dns server, you'd probably need a domain name you registereed through google, but certainly a way to set dns records ni googles' dns servers.
[20:42] <tomreyn> the other option is to use certbot on a linux system where the apex domain points to. certbot would then spawn a webserver and automates the the vallidation that way.
[20:44] <tomreyn> but then, there's documentation which explains all of this, and much better than i could,
[22:26] <rbasak> sdeziel: thank you for https://code.launchpad.net/~sdeziel/ubuntu/+source/ceph/+git/ceph/+merge/410908, but the importer can't respond to MPs. Could you add a suitable team to review it please? Or subscribe ~ubuntu-sponsors to the bug or something.
[22:26] <rbasak> I'd like to adjust the sponsoring report to display ~ubuntu-sponsors MP slots, but that's not done yet.
[22:31] <sdeziel> rbasak: oh, thanks for the hint. Let me know if something else needs fixing :)
[22:33] <rbasak> sdeziel: to be clear, adding ~ubuntu-sponsors to the MP won't work.
[22:33] <rbasak> Best to poke the OpenStack team for a review I guess?
[22:37] <sdeziel> rbasak: OK, will ask them for review but ATM, Launchpad doesn't want me to change the reviewer :/
[22:40] <sdeziel> when searching for ubuntu-openstack, it suggests ubuntu-openstack-dev
[22:41] <rbasak> Just another one
[22:42] <rbasak> But even then, make sure to follow up with an actual person please. There's not necessarily anyone or any process at the other end that will cause anything to happen in response to adding someone as a reviewer to an MP.
[22:43] <rbasak> The only general thing that works is an action that will cause it to be added to the general sponsoring queue (~ubuntu-sponsors subscription to a bug) but patch pilots are scarce nowadays.
[22:45] <sdeziel> rbasak: I've ask jamespage for a review and will poke him tomorrow
[22:45] <rbasak> Thanks!
[22:45] <rbasak> Just trying to make sure it doesn't get lost ;)
[22:47] <sdeziel> rbasak: yeah and I appreciate it although I would have preferred a simpler process. I guess I'm spoiled by how easy it is to do drive-by contributions via GitHub ;)