| babbageclunk | wallyworld: https://github.com/juju/juju/pull/11384 | 01:57 |
|---|---|---|
| wallyworld | ok | 01:57 |
| babbageclunk | way more in the description and QA steps than the code change! | 01:57 |
| wallyworld | good :-) | 01:58 |
| wallyworld | babbageclunk: what happens if you got to juju/description and try and run the tests? | 01:58 |
| babbageclunk | checking | 02:04 |
| babbageclunk | um, fails - is that me? | 02:06 |
| babbageclunk | (This is on the tip of master) | 02:06 |
| wallyworld | it fails for me due to registering a http protocol twice | 02:06 |
| wallyworld | nfi how stuff got landed | 02:06 |
| babbageclunk | yeah, seems pretty weird | 02:07 |
| wallyworld | babbageclunk: =1 on review with a request for a change | 02:07 |
| babbageclunk | go version diffs? | 02:07 |
| babbageclunk | ok, thanks! | 02:07 |
| babbageclunk | wallyworld: for me the juju/description failure is missing quotes around a time value | 02:21 |
| babbageclunk | which might be a yaml version diff | 02:21 |
| wallyworld | for me it's a godeps issue | 02:21 |
| wallyworld | the upstream all have incompatible deps | 02:21 |
| wallyworld | it's a mess | 02:21 |
| babbageclunk | ugh | 02:21 |
| wallyworld | so you ned to godeps -u dependencies.tsv | 02:22 |
| wallyworld | to get the deps as specified | 02:22 |
| wallyworld | then stuff fails | 02:22 |
| wallyworld | and trying to fix one breaks something else | 02:22 |
| wallyworld | thanks for looking though | 02:22 |
| babbageclunk | ok now the tests pass for me | 02:23 |
| babbageclunk | (I'd forgotten how to run godeps!) | 02:23 |
| wallyworld | hmmm | 02:24 |
| wallyworld | i'll try again after resetting everything | 02:24 |
| babbageclunk | that's the only way to be sure | 02:24 |
| babbageclunk | (bar nuking from orbit) | 02:24 |
| wallyworld | nup, still get the protocol already registered :-( must be a transitive dep somewhere | 02:25 |
| wallyworld | might be go 1.14 | 02:26 |
| babbageclunk | no, just tried with go 1.14 - maybe something with gomodules? | 02:28 |
| wallyworld | godeps should write out all the upstream repos to the specified sha, no gomodules involved. i've updated deps to get past that and now have juju/clock vs utls/clock compile errors, so will just need to keep at it | 02:29 |
| wallyworld | babbageclunk: maybe past you EOD, if so let me know... here's a juju/description PR, works with juju state tests with operations added. jenkins acting up pulling repos so still getting merge check to pass | 04:14 |
| wallyworld | https://github.com/juju/description/pull/74 | 04:14 |
| babbageclunk | wallyworld: nope - looking now | 04:14 |
| tlm | has the ACME code in juju ever worked ? | 04:14 |
| wallyworld | i think so, there was a bug filed recently that hpidcock fixed | 04:15 |
| tlm | ah yep I see it now | 04:16 |
| tlm | missed a line and was scratching my head | 04:16 |
| hpidcock | tlm: wallyworld: I only updated x/crypto to support acme v2. Not sure if the whole thing works end to end. I do remember testing to see if the error happened anymore and didn't see it when I setup a domain etc | 04:17 |
| hpidcock | and a controller using that domain | 04:18 |
| tlm | yeah not it's fine hpidcock, I would kill for go 1.14 at the moment | 04:18 |
| wallyworld | "soon" | 04:18 |
| wallyworld | :-) | 04:18 |
| wallyworld | next week with luck | 04:18 |
| tlm | there is a helper function that would save me an hour of work | 04:19 |
| tlm | https://golang.org/src/crypto/tls/common.go?s=34175:34244#L911 | 04:19 |
| wallyworld | cut and paste? :-) | 04:19 |
| babbageclunk | +1 | 04:20 |
| tlm | yeah could do | 04:20 |
| hpidcock | tlm: hah yeah.. we ain't vendoring that for you | 04:20 |
| tlm | but where do I dump the cut and paste | 04:20 |
| wallyworld | maybe with the worker that uses it (or whatever it is) with a TODO to remove next week | 04:20 |
| hpidcock | Yeah I'd rather you just todo it and come back to it next week like wallyworld suggests | 04:21 |
| babbageclunk | Or a test that fails in go 1.14! | 04:21 |
| tlm | checking out how they did it in 1.10 and see if there is a quick win with that first | 04:22 |
| wallyworld | thamks babbageclunk | 04:23 |
| babbageclunk | wallyworld: approved - it's a pity there's so much boilerplate to add a new entity there | 04:23 |
| tlm | ok going for the cut and paste | 04:23 |
| tlm | sorry juju | 04:23 |
| babbageclunk | I feel like we could do with some codegen | 04:23 |
| wallyworld | yeah | 04:24 |
| kelvinliu | wallyworld: +1 plz a small PR to disable some CMDs for daemonset https://github.com/juju/juju/pull/11386 | 04:42 |
| wallyworld | looking | 04:43 |
| kelvinliu | ty | 04:43 |
| wallyworld | kelvinliu: looks nice. ty. here's a small one also https://github.com/juju/juju/pull/11387 | 04:50 |
| kelvinliu | ty looking now | 04:51 |
| pmatulis | can the value of JUJU_AVAILABILITY_ZONE be inspected somehow? | 04:59 |
| wallyworld | i think you can use juju run and look at the env vars in a hook context that way | 04:59 |
| kelvinliu | wallyworld: lgtm ty | 05:02 |
| wallyworld | tyvm | 05:02 |
| wallyworld | kelvinliu: can you plese comment on https://discourse.juju.is/t/k8s-spec-v3-changes/2698/2 | 05:03 |
| kelvinliu | sure | 05:03 |
| elox | Good morning | 05:28 |
| wallyworld | o/ | 05:29 |
| hpidcock | wallyworld: this bug fix is going to take at least another day. Hopefully I have a pr up this afternoon for the implementation so you can sanity check it, but the unit tests are going to be fun. | 05:35 |
| wallyworld | yeah, i bet | 05:35 |
| wallyworld | manadart: what's O7k? | 06:07 |
| manadart | wallyworld: OpenStack | 06:57 |
| wallyworld | ah, derp | 06:57 |
| wallyworld | all this fance txt speak | 06:57 |
| wallyworld | *fancy | 06:57 |
| * manadart gestures with pipe. "We we was comin' up, we talked proper like." | 06:58 | |
| wallyworld | when i was a boy, people did speak properly with grammer rules and punctuation all followed | 07:00 |
| stickupkid | manadart, got a second? | 08:08 |
| manadart | stickupkid: Yeah. | 08:09 |
| stickupkid | manadart, https://github.com/juju/juju/pull/11388 | 08:25 |
| manadart | stickupkid: Yup. | 08:25 |
| === tmhoang4 is now known as tmhoang | ||
| elox | I'm trying to deploy a centos charm in vsphere. But I have no idea as how to get a working image in there and how to get vsphere to understand "--series centos7". Anyone that has a clue? The same is true for AWS https://discourse.juju.is/t/deploying-centos7-in-aws/2839/4 | 10:01 |
| manadart | stickupkid: HO? | 11:06 |
| stickupkid | manadart, of course | 11:06 |
| hml | manadart: updated the pr with caas qa test | 14:05 |
| manadart | hml: Ack. | 14:06 |
| stickupkid | manadart, so I have an issue with trying to pass a new type of header func through the client, in that it gets very messy and by messy, each goose.Client holds a http.Client and we create lots of different types of goose.Clients. | 14:32 |
| stickupkid | manadart, we then end up having lots of different ways to construct this goose.Client and non of them seem right | 14:33 |
| stickupkid | manadart, it feels like there should be a factory to get me a client and depending on my parameters | 14:34 |
| manadart | stickupkid: Yes. | 14:35 |
| stickupkid | manadart, but nova and neutron share the same http client between them both | 14:35 |
| jam | guild https://github.com/juju/juju/pull/11389 does now have some tests of both the CLI parsing and putting the args into the right spot, and more direct unit tests of how args get serialized. | 14:35 |
| jam | feedback appreciated. | 14:36 |
| manadart | hml: Getting this on the integration test. https://pastebin.ubuntu.com/p/pq6m8kfVhy/ | 14:40 |
| hml | manadart: huh, i’ll take a look. | 14:41 |
| stickupkid | manadart, can I grab you before you EOD quickly? | 14:59 |
| manadart | stickupkid: I'm in Daily. | 15:03 |
| hml | manadart: i’ll have to put up a new pr to fix the integration test. it’ll run if you replace `_` with `-` | 15:54 |
| stickupkid | hml, manadart that because juju has that restriction | 15:55 |
| hml | stickupkid: np, it just looks like i replaced only one. not sure where that happened | 15:59 |
| stickupkid | yeah, :) | 15:59 |
| achilleasa | anyone wants to review https://github.com/juju/juju/pull/11390? | 16:28 |
| rick_h_ | achilleasa: can you please make sure that you ping an email to thumper to peek at it as well | 16:30 |
| achilleasa | rick_h_: I have already @commented him in the PR | 16:30 |
| achilleasa | rick_h_: btw, if you have some time can you try to poke any holes in the implementation? | 16:30 |
| rick_h_ | achilleasa: awesome ty | 16:40 |
| rick_h_ | achilleasa: will look ty | 16:40 |
| achilleasa | hml: left a few comments on 11376; overall it looks good but I would recommend getting a +1 from someone who has worked with storage before | 16:57 |
| hml | quick pr review anyone? https://github.com/juju/juju/pull/11391 | 17:51 |
| rick_h_ | hml: +1 | 18:15 |
| hml | rick_h_: ty | 18:16 |
| hml | another quick on :-) https://github.com/juju/juju/pull/11392 | 18:19 |
| hatch | I'm having issues running the gui simplestreams tests....it requires the values in export_test which aren't loaded when running the test file directly `go test environs/gui/simplestreams_test.go` | 19:53 |
| === mruffell_ is now known as mruffell | ||
| tlm | mongo pem handling doesn't support headers :( | 23:37 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!