[00:14] blackboxsw: files cloudinit/apport.py, cc_snap.py also affected [00:20] thanks! [02:18] thx miminal, added to the bug [16:52] holmanb: falcojr or aciba, minor PR up to ensure Oracle testing on Jammy can succeed https://github.com/canonical/pycloudlib/pull/200 [16:52] Pull 200 in canonical/pycloudlib "util: update UBUNTU_RELEASE_VERSION_MAP for oci" [Open] [16:53] with this in place I can now run aciba's integration test on Oracle w/ VM shape VM.GPU2.1. will test VM.GPU2.3 [16:53] with this in place I can now run aciba's integration test on Oracle w/ VM shape VM.GPU2.1. will test VM.GPU3.2 [16:57] blackboxsw: +1 [20:44] ok sorted my oracle test setup woes. 2nd pycloudlib fix to handle configuring instances on the right subnet for oracle based on availability_domain https://github.com/canonical/pycloudlib/pull/201 [20:44] Pull 201 in canonical/pycloudlib "oci: select proper subnet matching availability_domain" [Open] [21:02] hey guys, i am trying to deploy a RHEL 8 VM to VMware vSphere 7.0. I think I have the right metadata and userdata. cloud-init 22.1-1 says that everything is valid and creates a correct network configuration. But the interface is not brought up. If i go in and do "ifup" the interface comes up. I got a python traceback, I have added the info here, [21:02] together with the userdata and metadata. Any comments or ideas would be very appreciated. thanks in advance! https://gist.github.com/jbfriedrich/348e988ba26a5a8631d7b33395a0ca90 [21:32] ryland: thanks for the ping looking "grep render /var/log/cloud-init" tell you? [21:33] ryland: thanks for the ping looking "grep render /var/log/cloud-init.log" tell you? [21:34] blackboxsw: "Selected renderer 'sysconfig' from priority list: None" [21:35] blackboxsw: i think this might be related: https://bugs.launchpad.net/cloud-init/+bug/1894837. i am not using the RHEL provided version as I need the VMware datasource that is not included in the old RHEL provided cloud-init version. so "NM_CONTROLLED=no" is in there [21:35] Launchpad bug 1894837 in cloud-init "sysconfig: NM_CONTROLLED=False should not be set on RHEL8" [Medium, Triaged] [21:36] ryland: another question is what cloud-init "saw" from network config too "grep Applying network configuration /var/log/cloud-init.log" [21:38] blackboxsw: it has the right network config, but it says "bringup=False" [21:39] blackboxsw: I have added the screenshot to the gist [21:40] I have not seen any configuration that would allow me to chose if i want to bring up the interface or not. maybe i am missing something [21:47] bringup=False generally means network confg files are being detected during cloud-init's init-local boot stage before the system networking in officially brought up. So, cloud-init will rely on systemd-networkd or NetworkManager etc to bring up the written network config once cloud-init finished the init-local datasource and network-config writing operation https://cloudinit.readthedocs.io/en/latest/topics/boot.html#local. [21:49] blackboxsw: thanks for the clarification! that would make sense then and bring me back to the patch that I need to apply, as NM is ignoring the device [21:54] ryland: I'm rereading the bug you referenced and I think that is the patch you want https://bugs.launchpad.net/cloud-init/+bug/1894837/comments/8 [21:54] Launchpad bug 1894837 in cloud-init "sysconfig: NM_CONTROLLED=False should not be set on RHEL8" [Medium, Triaged] [21:55] yes, and I see that that patch still hasn't upstreamed to main for cloud-init. I know we recently landed a patch from rhel and centos folks that upstreamed some of RedHat's downstream patches, but this didn't get promoted to tip of main [22:09] Blackboxsw: thank you! I will apply that patch tomorrow and rebuild the template and then try again. Thats a relief that I now know what to look for. Thanks again!