[04:00] <mup> PR operator#485 opened: Draft of get_services API <Created by benhoyt> <https://github.com/canonical/operator/pull/485>
[06:42]  * lewq[m]  < https://matrix.org/_matrix/media/r0/download/matrix.org/YNWtOLoQMnuIxqnVKKoBQKqe/message.txt >
[09:25] <mup> Issue operator#486 opened: Can't send environment to pebble <Created by mthaddon> <https://github.com/canonical/operator/issues/486>
[13:06] <ruivasconcelos[7> John Meinel:
[13:08] <ruivasconcelos[7>  * John Meinel: could we have some ETA for this, as it is a blocking issue. I understand that you guys have a lot on your plate, but we need to unblock @lewq
[13:08] <ruivasconcelos[7>  * John Meinel: could we have some ETA for this, as it is a blocking issue. I understand that you guys have a lot on your plate, but we need to unblock lewq soon if we want to have an AI story
=== jam1 is now known as j-a-meinel
=== j-a-meinel is now known as jam1
[13:21] <jam1> @ruivasconcelos[7, I assume you are talking about operator#485  ?
[13:21] <mup> PR #485: Draft of get_services API <Created by benhoyt> <https://github.com/canonical/operator/pull/485>
[13:21] <jam1> or are you talking about operator#486
[13:21] <mup> Issue #486: Can't send environment to pebble <Created by mthaddon> <https://github.com/canonical/operator/issues/486>
[13:21] <jam1> (i probably think the latter)
[13:22] <jam1> but there are lots of things you could be talking about :)
[13:23] <lewq[m]> jam1: can you link to the issues please?
[13:23] <lewq[m]> i never opened an issue, but i reported the problem up there ^^ with some screenshots
[13:24] <jam1> lewq[m], ah, I don't get pings if you use my full name, so I missed them. I don't think my slack is wired up to this channel anymore.
[13:24] <lewq[m]> https://matrix-client.matrix.org/_matrix/media/r0/download/matrix.org/RNaUlprHlekIXySXxrkziIvi and https://matrix-client.matrix.org/_matrix/media/r0/download/matrix.org/adqNwMlQNujqncYmjjlHGSNb
[13:25] <lewq[m]> np
[13:25] <jam1> lewq[m], as for bootstrapping an old RC, if you just "snap install juju --channel=2.9/candidate" you should get 2.9-rc6, its only because you are on edge that you have 2.9-rc7
[13:28] <jam1> lewq[m], ruivasconcelos[7 : fwiw, I thought 2.9-rc7 in edge already had a fix for this particular issue (i had asked tlmiller to look into it.)
[13:37] <ruivasconcelos[7> Thank you for the clarity jam1
[13:38] <ruivasconcelos[7> This has been a mess with communication tools
[13:43] <lewq[m]> @jam1: the screenshots I posted above were on 2.9-rc7
[13:43] <jam1> lewq[m], 2.9-rc7 has not been finish published yet, you could only get it from /edge which is constantly being updated
[13:43] <jam1> so *what* version of 2.9-rc7 matters
[13:43] <jam1> since it isn't releasde
[13:43] <jam1> :)
[13:43] <lewq[m]> oh, ok
[13:44] <lewq[m]> i shall test again, then!
[13:44] <jam1>  /edge is built from the tip of the branch
[14:29] <lewq[m]> jam1: it's now broken with ```ERROR "caas-broker-tracker" manifold worker returned unexpected error: cannot create caas broker: unable to determine legacy status for namespace controller: namespaces "controller" is forbidden: User "system:serviceaccount:controller-testfaster:modeloperator" cannot get resource "namespaces" in API group "" in the namespace "controller"
[14:29] <lewq[m]> ```
[14:29] <lewq[m]> ```
[14:29] <lewq[m]> ERROR "caas-broker-tracker" manifold worker returned unexpected error: cannot create caas broker: unable to determine legacy status for namespace controller: namespaces "controller" is forbidden: User "system:serviceaccount:controller-testfaster:modeloperator" cannot get resource "namespaces" in API group "" in the namespace "controller"
[14:29] <lewq[m]> ```
[14:30] <lewq[m]> ```
[14:30] <lewq[m]> kubectl logs modeloperator-5f57bd5b5f-lwmjd -n controller-testfaster
[14:30] <lewq[m]> ```
[14:30] <lewq[m]> latest 2.9-rc7 as of a few minutes ago
[14:30] <jam1> lewq[m], are you bootstrapping and then deploying into the controller model, or are you bootstrapping, then adding a model and deploying into it
[14:31] <lewq[m]> I'm doing this: https://github.com/mlopsworks/charms/blob/main/.testfaster.yml#L68-L78
[14:32] <jam1> so it is just the bootstrap itself that is failing?
[14:33] <lewq[m]> the bootstrap succeeds, but then it gets stuck, nothing starts up, and that's the error in the modeloperator pod
[14:33] <knkski[m]> btw lewq, were you going to move that code over to github.com/canonical/mlflow-operators?
[14:33] <lewq[m]> yeah i will do knkski
[14:35] <lewq[m]> minikube with kube `v1.17.11` if that makes any difference (https://github.com/mlopsworks/charms/blob/main/.testfaster.yml#L101)
[15:01] <jam1> lewq[m], if you can, please confirm that "snap install juju --channel 2.9/stable' does work for you, and then we'll try to address the issues with 2.9-rc7
[15:02] <jam1> also, these might be better in #juju since it is more specifically about juju vs charms
[17:07]  * lewq[m]  < https://matrix.org/_matrix/media/r0/download/matrix.org/GwnhNrMYxgJmdPqSzidInQDX/message.txt >
[17:54] <lewq[m]> still down. please lmk when the snap store is back up