| meena0 | goneri approves: https://github.com/canonical/cloud-init/pull/4159 | 11:16 |
|---|---|---|
| -ubottu:#cloud-init- Pull 4159 in canonical/cloud-init "BSD: remove datasource_list from cloud.cfg template" [Open] | 11:16 | |
| meena0 | all of these are ready to go: https://github.com/canonical/cloud-init/pulls/igalic | 12:09 |
| blackboxsw | meena0: I'll grab 4159.... I'd like to know what change that actually allows for efficient detection in BSD now... because I thought FreeBSD didn't run ds-identify which prunes down the applicable datasource_list to things that are likely. As far as I'm aware ds-identify is only run by sytemd environments in systemd generator timeframe before cloud-init stages actually run. | 15:20 |
| blackboxsw | meena0: is there something we can point to in *BSD that triggers limiting datasource_list via ds-identify? without it, what you'll get on Azure or other clouds is that OpenStack datasource will still be a potetnially active datasource to discover (ordered before Azure)... and openstack python discovery logic will checkout for about 10 minutes retrying potential openstack endpoints trying to see if the datasource comes up. | 15:22 |
| blackboxsw | it'd be sweet if *bsd had a service that we new ran before cloud-init local that would just trigger /usr/lib/cloud-init/ds-identify to pre-seed the limited datasource_list based on environmental artifacts. | 15:23 |
| blackboxsw | or if we know that something like that already exists in BSD | 15:24 |
| blackboxsw | I'll fire up a BSD KVM image to see what's up over there. | 15:24 |
| blackboxsw | falcojr: holmanb fix for building rockylinux/8 on tip of main with ./tools/run-container https://github.com/canonical/cloud-init/pull/4161 | 15:28 |
| -ubottu:#cloud-init- Pull 4161 in canonical/cloud-init "tools: fix run-container builds for rockylinux/8 git hash mismatch" [Open] | 15:28 | |
| blackboxsw | this may solve James' rockylinux/9 build issues too. Wrote out redhat package '/home/ci-test/cloud-init/cloud-init-23.2+16.g7f8de36c-1.el9.src.rpm' | 15:31 |
| blackboxsw | Wrote out redhat package '/home/ci-test/cloud-init/cloud-init-23.2+16.g7f8de36c-1.el9.noarch.rpm' | 15:32 |
| falcojr | https://git.centos.org/rpms/cloud-init/blob/c8s/f/SOURCES/ci-Revert-Add-native-NetworkManager-support-1224.patch | 15:49 |
| falcojr | https://git.centos.org/rpms/cloud-init/blob/c8s/f/SOURCES/ci-Revert-Use-Network-Manager-and-Netplan-as-default-re.patch | 15:49 |
| falcojr | caribou: without knowing the specifics of how rocky packaging works, if these two patches are part of rocky 9, you're still going to see the same issue you pasted here earlier. Your original paste has expired, but I'm wondering what your network setup looks like that triggered the issue. We probably have a bug to fix in the sysconfig renderer. | 15:49 |
| falcojr | thanks blackboxsw, your fix worked for me | 15:53 |
| blackboxsw | great. | 15:54 |
| meena0 | the only test failing here is the docs thing that's fixed https://github.com/canonical/cloud-init/pull/2146 | 17:24 |
| -ubottu:#cloud-init- Pull 2146 in canonical/cloud-init "FreeBSD fix parsing of mount and mount options" [Open] | 17:24 | |
| meena0 | (i rebased it) | 17:28 |
| holmanb | meena0: TIL regex101 has permalinks | 20:42 |
| meena0 | holmanb: it's pretty sweet stuff | 20:49 |
| meena0 | that reminds me, i have some baklava in the fridge | 20:49 |
| * holmanb loves baklava | 21:04 | |
| meena0 | Funnily, this is Balkan style, and even tho I'm from the Balkans, I prefer Turkish or Arabic style... | 21:25 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!