babbageclunk | anastasiamac: ok, https://github.com/juju/juju-restore/pull/1 is ready for another look (I guess after standup) | 00:13 |
---|---|---|
anastasiamac | babbageclunk: will do soon \o/ | 00:13 |
babbageclunk | thanks | 00:13 |
wallyworld | kelvinliu: we can land that packaging PR on libjuju | 01:02 |
kelvinliu | wallyworld: yep | 01:02 |
thumper | PR ready for the introduction of the model summary watcher to the cache: https://github.com/juju/juju/pull/11208 | 01:44 |
thumper | and a much simpler one to update a dependency: https://github.com/juju/juju/pull/11209 | 01:51 |
* anastasiamac lloking at 11209 | 01:57 | |
timClicks | thumper: did you want to catch up for a few minutes now that those PRs have landed? | 01:58 |
timClicks | wallyworld, kelvinliu, babbageclunk: found the other issue that a user has had w/ zone contraints https://discourse.jujucharms.com/t/using-vmware-vsphere-with-juju/1099/3?u=timclicks | 02:00 |
timClicks | perhaps that's the same one | 02:00 |
thumper | timClicks: sure, I have 15 minutes now | 02:02 |
kelvinliu | yeah, the same one | 02:02 |
thumper | jam: with you shortly, just trying to finish up a code review | 02:59 |
jam | ok | 03:03 |
babbageclunk | does anyone know of other repos of ours that use go modules? | 04:02 |
wallyworld | babbageclunk: charm.v4 i think | 04:02 |
wallyworld | some of the upstream juju ones | 04:02 |
wallyworld | charm.v6 i mean | 04:03 |
wallyworld | charmrepo.v4 | 04:03 |
babbageclunk | ok, thanks wallyworld | 04:04 |
babbageclunk | weirdly, our build script for charm.v6 still installs go 1.10 | 04:05 |
babbageclunk | I don't get how that works | 04:05 |
wallyworld | we'd still use dependencies.tsv in that case i'd expect? | 04:06 |
babbageclunk | wallyworld: oh yeah - dependencies.tsv's still there | 04:09 |
babbageclunk | aha, the charmrepo one is more like what I want | 04:10 |
wallyworld | good | 04:10 |
_thumper_ | wallyworld, tlm: PR 11203 introduced confict markers into the acceptance tests python file | 04:27 |
=== _thumper_ is now known as thumper | ||
thumper | it is what is causing many stop the line failures... | 04:28 |
tlm | taking a look | 04:29 |
tlm | i'll put up a pr thumper, sorry about that | 04:32 |
thumper | tlm: thanks | 04:33 |
tlm | https://github.com/juju/juju/pull/11210 | 04:37 |
hpidcock | tlm: lgtm | 04:47 |
hpidcock | same lgtm that let it through in the first place sorry | 04:47 |
anastasiamac | babbageclunk: \o/ | 04:50 |
vultaire | New MR against the interface:elasticsearch layer, need a review if possible: https://github.com/juju-solutions/interface-elasticsearch/pull/14 | 05:02 |
nammn_de | manadart: around for another review round on: https://github.com/juju/juju/pull/11186 I incorporated your feedback. On those things where I wasn't 100% sure I wrote some comments how and why I did it that way | 09:04 |
manadart | nammn_de: Will look in a minute. | 09:06 |
nammn_de | stickupkid: around for a quick talk about the test I mentioned yesterday? | 09:09 |
stickupkid | nammn_de, yeah give us 5 | 09:25 |
nammn_de | stickupkid: sure take your time | 09:25 |
stickupkid | nammn_de, daily | 09:46 |
stickupkid | manadart, just grabbing a drink, but once I've done that want to HO about openstacky | 10:08 |
manadart | stickupkid: Yep, got to pick-up/drop-off my daughter just before 12, but ping me. | 10:09 |
stickupkid | what time is it there now? | 10:09 |
stickupkid | 11? | 10:09 |
stickupkid | manadart, ^ | 10:10 |
manadart | stickupkid: Yeah, 10 past. | 10:10 |
stickupkid | manadart, ping | 10:21 |
nammn_de | stickupkid: up for a smaller cr? https://github.com/juju/juju/pull/11212 only updates tests, nothing customer facing | 10:29 |
nammn_de | manadart: was planning to look into update-space. Was planning to spec that out again. Time for a HO later? | 12:05 |
manadart | nammn_de: Maybe, looking at Prodstack issue right now. | 12:07 |
nammn_de | manadart: sure, just gimme a ping else I start specing myself and you can take a look at the network cli doc and give a comment | 12:07 |
hml | stickupkid: it looks like what is missing… the juju-<model-name> profile. | 13:17 |
stickupkid | hml, but why? | 13:18 |
hml | stickupkid: the 64 thousand dollar question. :-D | 13:18 |
stickupkid | hml, do we set it? | 13:18 |
rick_h | morning | 13:18 |
hml | stickupkid: yes, we should. it should go on every container | 13:18 |
hml | stickupkid i’ll play with it some today | 13:19 |
hml | stickupkid: wait… brain fart - they only go on machines that are containers…. not containers. | 13:20 |
hml | lack of coffee brain | 13:20 |
hml | looks to me like the charm profile is being applied. | 13:20 |
stickupkid | hml, that's very strange, we do remove it in certain places | 13:25 |
hml | stickupkid: that profile was never applied to #/lxd/# machines, only #. I’d have to look at the profiles to maybe remember why | 13:27 |
hml | it’s from the being of lxd in juju i think | 13:27 |
hml | stickupkid: it allows us to create a container in a container | 13:31 |
hml | the juju-default etc profiles | 13:31 |
hml | not sure we’re ready for a container in a container … 0/lxd/0/lxd/0? | 13:31 |
stickupkid | yeah... errr | 13:31 |
achilleasa | hml: 11206 has been approved | 14:09 |
hml | stickupkid: we would have seen errors if the container couldn’t be deployed. | 14:09 |
hml | achilleasa: ty | 14:10 |
hml | stickupkid: ping me when you’re back. found something interesting | 14:24 |
rick_h | manadart: if you get a sec can you update your "changes requested" on https://github.com/juju/juju/pull/11195 please? | 14:29 |
rick_h | the code there in question is gone now anyway | 14:30 |
hml | stickupkid: reproduced locally perhaps | 14:32 |
manadart | rick_h: Approved it. | 14:45 |
rick_h | manadart: ty | 14:50 |
achilleasa | hml: I got my framework patches to play nicely with the new state tools :-) | 14:52 |
hml | achilleasa: sweet! | 14:54 |
hml | anyone noticed that subordinate unit numbering maybe not increment by 1, I got units of 1,8,14,15. ; after a deploy and add-unit 3 times. | 15:11 |
hml | stickupkid: reproducing i’m at 50% fail rate. tracking down the root cause | 15:59 |
IOstars | Hey there everyone, does anyone here know when nova-compute charm will pull release with the current fixes in milestone 20.02? | 16:40 |
rick_h | thedac: ^ ? | 16:44 |
rick_h | IOstars: I think the charm release is immenent but defer to them on their changes getting out the door | 16:44 |
thedac | Yes, charm release is immanent but waiting on QA | 16:45 |
IOstars | Getcha, just curious. Have a fresh install that got hit by the critical bug being fixed there. | 16:45 |
IOstars | Grreat thanks :) | 16:45 |
rick_h | IOstars: :( glad there's a fix. Yea coming soon | 16:45 |
hml | stickupkid: https://pastebin.canonical.com/p/TKdw4mpwkT/ line 2 ; care to take bets on the cache? :-D | 16:49 |
stickupkid | sigh | 17:06 |
stickupkid | hml, quick ho? | 17:06 |
hml | stickupkid: i think you’re pr from jan will fix. | 17:07 |
hml | stickupkid: sure | 17:07 |
thumper | morning team | 19:23 |
rick_h | morning thumper | 19:23 |
hml | morning thumper | 19:23 |
vultaire | hello - hoping to wake up a MR which has been stuck without review for some time. interface:elasticsearch, adding "version" field to share MongoDB version. MR: https://github.com/juju-solutions/interface-elasticsearch/pull/13 | 19:51 |
vultaire | If anyone could take a look, would be quite appreciated. Thanks. | 19:51 |
vultaire | erm | 19:53 |
vultaire | I totally botched my statement there ;) | 19:54 |
vultaire | "version" field shares the elasticsearch version ;) | 19:54 |
vultaire | (I have a different MR re: mongodb as well; mixed up in my brain a bit.) | 19:54 |
timClicks | what's the simplest way to install juju_engine_report? We make oblique reference to it in places, but I can't find a place where we publicise it at all. | 21:46 |
wallyworld | it's automatically installed by jujuitself when it creates a machine | 21:48 |
wallyworld | all those diagnostic scripts are put in place, of which engine report is one | 21:49 |
wallyworld | /etc/profile.d/juju-introspection.sh | 21:50 |
timClicks | wallyworld: could you please check the note here for accuracy? https://discourse.jujucharms.com/t/what-are-your-tips-for-running-juju-in-production/2573/10?u=timclicks | 22:03 |
wallyworld | will do, just finishing something first | 22:03 |
timClicks | np | 22:04 |
wallyworld | also works on k8s, but you need to source the script when you exec in, there's a discourse post on it | 22:04 |
timClicks | and some release notes ;) https://discourse.jujucharms.com/t/new-features-and-changes-in-juju-2-7/2268 | 22:06 |
wallyworld | timClicks: were you going to mention the other functions that are available? | 22:14 |
timClicks | wallyworld: eventually.. simon's post prompted me to think about how discoverable juju_engine_report is | 22:14 |
wallyworld | i think we could build the summary into the engine report as a stadard thing | 22:16 |
* timClicks nods | 22:17 | |
timClicks | How do you train your staff to use Juju? Some suggested tasks for people learning Juju https://discourse.jujucharms.com/t/-/2639 | 23:14 |
timClicks | ^ would love anyone's input on this thread! | 23:14 |
anastasiamac | timClicks: love i | 23:18 |
anastasiamac | timClicks: for advance, i'd recommen adding model migration and CMR stuff too | 23:18 |
timClicks | :D | 23:18 |
timClicks | great idea | 23:18 |
timClicks | anastasiamac: do you want to add a comment there yourself? | 23:20 |
anastasiamac | and mayb at each laevel, how do u troubleshoot problem with each task? like usage of logs, debug-log command etc | 23:20 |
Sp3nc3r | any free mailer | 23:20 |
Sp3nc3r | or smtp | 23:20 |
anastasiamac | timClicks: feels like stepping on ur toes.... but may :) | 23:20 |
timClicks | no no, comments in the thread are welcome - otherwise it's always me talking to myself | 23:21 |
anastasiamac | and that is fun to watch too :D | 23:22 |
timClicks | Sp3nc3r: are you asking if you can deploy a mail server like postfix with juju? | 23:22 |
timClicks | evhan: sorry I couldn't be more helpful in discourse | 23:45 |
evhan | timClicks: Not at all! That background is already helpful. | 23:48 |
timClicks | evhan: there's actually some evidence that the majority of deployed charms are "private charms" because many people use the public charms as their base, then make tweaks | 23:49 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!