=== majduk__ is now known as majduk_
[07:53] <mup> PR operator#393 closed: adds Harness.hooks_disabled, a context manager <Created by chipaca> <Merged by chipaca> <https://github.com/canonical/operator/pull/393>
=== ChanServ changed the topic of #smooth-operator to: general discussion of the operator framework || github.com/canonical/operator || ops 0.9.0 || charmcraft 0.4.0
[09:02]  * bthomas updated to ops 0.9.0
[09:06]  * bthomas updated kernel rebooting now
[10:02] <Chipaca> moon127: remind me, were you still working on your charm this week, or were you off to more exiting topics?
[10:02] <Chipaca> exciting*
[10:03] <Chipaca> wondering if i should remove the event from the calendar
[10:05] <moon127> Chipaca: my 2 week stint was up end of last week.  As far as I know I'm onto a new assignment this week.
[10:05]  * moon127 catches up after the UK public holiday and yesterday as PTO
[10:05] <Chipaca> moon127: ok! removing the events from my calendar
[10:05] <Chipaca> moon127: the last changes were good fwiw
[10:06] <Chipaca> i commented on something that you'd already fixed, is all
[10:08] <moon127> ah yes I see, thanks!  That bit of code was rather subject to change running up to an end of week, end of rotation deadline, so good to have the syntax as intended rather than first implemented indeed
[10:09] <moon127> thanks for the review, I'll check if Tom is happy our side and get that landed
[10:18] <moon127> now merged
[10:55] <facubatista> ¡Muy buenos días a todos!
[11:00] <bthomas> 𝐌𝐨𝐫𝐧𝐢𝐧𝐠 facubatista
[13:26] <bthomas> Prometheus has a huge number of configuration options in its config file !
[13:32] <mup> PR operator#396 opened: add build dist and egg-info to gitignore <Created by chipaca> <https://github.com/canonical/operator/pull/396>
[13:51] <Chipaca> davigar15_: sorry to pester you, who was it that had the issue that i pointed to the environment juju bug and workaround? (looking for the irc handle)
[13:52] <davigar15_> DominikF
[13:52] <Chipaca> davigar15_: thanks!
[13:52] <Chipaca> DominikF: let me know if that bug+workaround isn't the thing you were hitting, please
[13:57] <DominikF> Chipaca: It is exactly what I was having issues with, the workaround fixed it for me. But it is still failing for me on 2.8.2 candidate, I commented it in the bug. Thanks!
[13:57] <Chipaca> ooh, interesting
[13:57] <Chipaca> DominikF: thanks
[13:58] <Chipaca> makes me think maybe we should push the workaround into our k8s opslib
[13:58]  * Chipaca waits to see what people say on the bug
[13:59] <DominikF> Chipaca: there is a k8s opslib?
[13:59] <Chipaca> DominikF: https://github.com/chipaca/ops-lib-k8s/
[13:59] <Chipaca> DominikF: most of it is mark maglana's work
[13:59] <Chipaca> hoping to have it grow over time
[13:59] <Chipaca> there's natalya's work but sadly she didn't write tests, so i need to find time for taht
[14:00] <Chipaca> and then we also want to do the same with the oci resources thing
[14:00] <Chipaca> so much TODO, so little DONE
[14:03] <DominikF> Chipaca: Interesting, we are looking into doing something similar as our "proxy charms" with deployments done with Helm, so there might be a moment where we could contribute that part there.
[15:46]  * facubatista -> lunch
[16:32]  * facubatista is back
[16:42] <mup> Issue operator#397 opened: Get container utility for easy pod_spec updates <enhancement> <Created by justinmclark> <https://github.com/canonical/operator/issues/397>
[16:56] <justinclark> I self-assigned myself to this new issue ^ but I won't start working on it until someone confirms we should have it (Chipaca, facubatista, jam)
[16:56]  * justinclark lunch
[17:04] <facubatista> justinclark, ack
[17:39]  * justinclark back
[18:49] <dstathis> I'm looking at the code in tests/commands/test_init.py
[18:49] <dstathis> the code here https://pastebin.com/raw/Hz47pzqQ
[18:50] <dstathis> I'm wondering what '(env_path,) = env_paths' is supposed to do
[18:50] <dstathis> Is it just trying to remove the item from a single element list?
[18:51] <dstathis> because when I run it that list has more than one element and it breaks
[19:06] <facubatista> dstathis, can you show me your sys.path in that test? thanks!
[19:06] <facubatista> there should be only one 'env/lib/python' in there, from the virtualenv, but you may have a more complex setup
[19:09] <dstathis> https://pastebin.com/raw/xhE7EDLS
[19:09] <dstathis> I have a fix for it but it
[19:09] <dstathis> but it is weird to me that this was working before
[19:12] <dstathis> I'll just open a PR with my change
[19:16] <facubatista> dstathis, how did you create that virtualenv? because 'venv' is a virtualenv, right?
[19:16] <dstathis> https://github.com/canonical/charmcraft/pull/147
[19:16] <mup> PR charmcraft#147: fix PYTHONPATH correctly <Created by dstathis> <https://github.com/canonical/charmcraft/pull/147>
[19:17] <dstathis> yes created with "virtualenv -p python3 venv"
[19:18] <facubatista> dstathis, and then how are you running those tests?
[19:18] <dstathis> ./run_tests
[19:19] <facubatista> dstathis, can you do this very same thing in a fresh dir (outside any current venv)? https://paste.ubuntu.com/p/FTsFJ8tSBQ/
[19:23] <dstathis> https://paste.ubuntu.com/p/h76bKVMdbq/
[19:24] <dstathis> yup same issue exists. multiple paths with "test123/lib/python"
[19:26] <facubatista> dstathis, yeap... it's bionic, right?
[19:26] <dstathis> yes
[19:27] <facubatista> dstathis, ack, thanks
[19:27] <dstathis> ah I see on focal this doesn't happen
[19:30] <dstathis> Well I thill think the change I proposed is better as this issue could come up
[19:30] <dstathis> also if you are running outside of a virtualenv that list will have 0 elements and it will also fail
[19:33] <dstathis> I'm not really sure how it's been working on travis since that runs on bionic
[19:39] <facubatista> +1 to be more robust anyway