[00:02] sure [00:22] babbageclunk: i think the PR needs a new test? see my comment [00:24] thumper: which bug? there's nothing on the 2.2 beta3 milestone. i could search but if you have the number handy.... [00:24] wallyworld: thanks [00:24] * thumper looks for wallyworld [00:24] https://bugs.launchpad.net/juju/+bug/1669540 [00:24] Bug #1669540: state: storeManager dies permanently when it should not [00:25] oh that one [00:25] see related comments at the end of the bug [00:25] i wasn't intending to look at it [00:25] i have oracle provider stuff queded up [00:25] i want to get that done for beta3 [00:26] i could look tomorrow [00:26] wallyworld: where would I get the tools sha256 from? [00:27] there's a tools doc in mongo or are you asking where's the code to calculate the hash? [00:27] I can see the calculation in the unpack tools [00:27] but I'm doing the 1.25 upgrade [00:27] perhaps a 5 minutes Hangout would be better [00:27] sure [00:27] got time? [00:27] 1:1 [00:27] wallyworld: 1:1 [00:45] wallyworld: there was already a test for updating - I tweaked it to make it clear that it preserves arbitrary firewall rule names. [00:46] babbageclunk: ok. was there also an issue where adding a new rule with updated cidrs was an issue? [00:46] or applying a new cidr to an existing rule? [00:47] i thought there was a gap like that you mentioned? [00:49] wallyworld: That test does add a CIDR to an existing rule. Do you mean handling swapped orders of CIDRs? I've got a test at the ruleSet level. [00:50] yeah, ok, sounds good. i didn't digest all the new tests [00:50] wallyworld: sweet. [00:52] wallyworld: Looks like the error swallowing I thought I saw was actually because I was watching log messages from juju.apiserver.remotefirewaller and juju.worker.firewaller, and the error was reported by juju.worker.dependency, so I didn't see it. Checking that now [00:53] ok [02:20] babbageclunk: can you look at this trivial PR to get Oracle provider working again? [02:20] https://github.com/juju/juju/pull/7269 [02:21] wallyworld: sure [02:23] wallyworld: LGTM'd [02:24] babbageclunk: tyvm [02:35] thumper: can you recall where the code to shorten the lxd instance ids lives? was that in a utils somewhere or was it a bespoke lxd bit of code? or similarly for maas [02:36] I think it is in the instance package [02:36] tail of a UUID [03:11] wallyworld: It looks like this check is the cause of the ports not being closed: https://github.com/juju/juju/blob/staging/worker/firewaller/firewaller.go#L1124 [03:12] wallyworld: Removing it makes my new test pass, doesn't break any of the other ones. Any idea why it's there? [03:12] babbageclunk: hmmm, yes, that was to fix something else [03:12] um [03:13] i can't recall exactly what now, but it showed up in manual testing maybe [03:13] when you deleted a relation, the ports didn;t close [03:14] maybe with your work, it's now fixed implicitly [03:14] i think it was that - create a remote relation. see the ports open, delete the relation, the ports should close [03:15] wallyworld: ok, I'll check that manually. [03:15] babbageclunk: or [03:15] i think you then need to add the relation again perhaps, and see the ports open again [03:15] ah, ok [03:15] so yeah, just a small manaul test to check everything [03:15] wallyworld: sounds good [04:23] wallyworld: review plz? https://github.com/juju/juju/pull/7270 [04:23] sure [04:23] thanks! [04:24] as a reward, some delightful doggo pics: https://www.reddit.com/r/rarepuppers/ [04:24] lol [04:24] babbageclunk: so the manual testing worked. i winder why i put that line in [04:25] babbageclunk: before you land, we need to get the vsphere thing fixed or else it won't go through, so do try yet [04:25] *don't [04:27] ok, shan't [04:27] * babbageclunk goes for a wee run [04:31] babbageclunk: that means something different here in australia [04:31] just saying [04:31] wallyworld: that's a wee walk [04:31] wee runs are anything under 5km [04:31] depends on how much beer [04:39] axw: your PR is on it's way.... [04:40] wallyworld: thanks [05:29] menn0: here's a small PR for the watcher bit https://github.com/juju/juju/pull/7272 [05:29] i wish the bug only contained one issue rather than 2 [05:29] wallyworld: looking [05:33] wallyworld: done [05:33] ta [06:41] jam: unless i am wrong, containers have never inherited their host machine's constraints, and that's not something that has been changed in the last week or so, right? [06:42] wallyworld: it has never been the intent, but I do believe I've seen containers inheriting the host machines's space constraints for a reason I don't understand [06:42] oh, i see. that is unintended i'm sure [06:43] i saw the email to the list and it seems quite a bad issue for him [07:55] juju add-model has a --credential flag; is there a way to set this after the model has been added? [07:55] or show which credentials are in use for a model? === admcleod_afk is now known as admcleod [11:04] Hi, if there's somebody awake now: [11:05] I am writing new provider for juju. When I implement the provider interfaces, I am contastnly getting "method Id should be ID [go/golint]" from golint. do you guys have any recommendation how to turn this particular warning off in golint? [11:06] I am using vim, vim-go, YouCompleteMe [12:24] rick_h: new bug here https://bugs.launchpad.net/juju/+bug/1685782 [12:24] Bug #1685782: juju deployed LXD inherit constraints from the host causing lxd deploys to fail [12:25] please put some heat on that [12:25] bdx: rgr === petevg_afk is now known as petevg === bdx_ is now known as bdx [15:07] Anyone got a spare cycle for troubleshooting steps on a model migration issue in #1680392? [15:07] Bug #1680392: Model migration fails on large model === bdx_ is now known as bdx [21:19] Can I get a review of https://github.com/juju/juju/pull/7273?