[09:21] hello [09:21] please can i ask you aquestion ? [09:22] i would like to pass interactively the password [09:23] of encrpted disk in ubuntu installation [09:23] i can not find any yaml file fo rthe storage in ubuntu.... [13:42] https://www.irccloud.com/pastebin/qBljot0H/ [13:42] cloud-init failed at stage init-local [13:43] this is happening on fedora32 , what could be the reason? [13:54] manuvakery1: do you know if this occurs on a Fedora release that is still maintained (i.e. F33 or F34) [13:56] holmanb: I need to test it [14:30] its happening with fedora33 as well.. I feel like something to do with the hostmachine network [14:30] iam deploying an openstack baremetal instance [15:58] manuvakery1: does that system have a nic with the corresponding mac address? (b4:92:91:c5:76:23) What is the output of `ip l`? [16:00] holmanb: yeah its there, also ips are assigned properly [16:00] If i deploy ubuntu/centos on the same host its works fine [16:02] @manuvakery1 which centos did you try on? [16:03] centos7 [16:08] falcojr: I'm rebasing your branch with paride's CI fix now. [16:10] also rebased https://github.com/canonical/cloud-init/pull/1050 [16:10] Pull 1050 in canonical/cloud-init "tox integration: pass the *_proxy and GOOGLE_* env vars" [Open] [16:10] falcojr: it's all approved and re-running CI. once we succeed there we can cut the impish release branch [16:12] @paride can you also passthrough GCP_* ?https://cloud.google.com/functions/docs/configuring/env-var#runtime_environment_variables_set_automatically [16:17] blackboxsw, done [16:27] falcojr: merged https://github.com/canonical/cloud-init/pull/1048 [16:27] Pull 1048 in canonical/cloud-init "Allow disabling of network activation (SC-307)" [Merged] [16:28] falcojr: do you want to generate the ubuntu/devel release branch please? then I can sponsor [17:29] manuvakery1: Did you use the same version of openstack with {Ubuntu,centos-7,fedora}? [21:16] woo hoo. cloud-init release accepted for impish with GCP fixes. I've queued uploads for B F and H Ubuntu series. Thanks James. === andrewbogott is now known as Guest6504 === cjp256_ is now known as cjp256 [22:21] falcojr: Hey, I can confirm it's *not* two datasources loaded simultaneously causing the network bug - https://github.com/kubernetes-sigs/image-builder/issues/712#issuecomment-938195915 [22:21] Issue 712 in kubernetes-sigs/image-builder "[Ubuntu] Cloud-init failing with KeyError: id0" [Open] [22:22] I don't know what it is yet, but likely some issue I introduced with the updated datasource. I'm not sure what as nothing really changed here. My next step is to use the old datasource with CI v21.3 to see if it fails the same way. I wonder if it's v21.3 and not the DS. Rather, I wonder if it's both DSs being incompatible with CI v21.3 :( [22:25] I have a theory... [22:26] And it was right. Crap. [22:26] falcojr: It's the "set-name" directive. When that's removed, everything works as expected. [22:26] Something in v21.3 is breaking due to set-name. [22:31] falcojr: See https://github.com/kubernetes-sigs/image-builder/issues/712#issuecomment-938200573 for more info [22:31] Issue 712 in kubernetes-sigs/image-builder "[Ubuntu] Cloud-init failing with KeyError: id0" [Open]