| menn0 | thumper: can you have another look at http://reviews.vapour.ws/r/1975/ pls when you're back? | 00:16 |
|---|---|---|
| === anthonyf is now known as Guest70961 | ||
| davecheney | thumper: james fixed the worker/provisioner race | 00:35 |
| davecheney | https://github.com/juju/juju/pull/2566 | 00:35 |
| davecheney | so i guess that means we don't need to argue about gocheck maintainership | 00:35 |
| thumper | menn0: ack | 01:16 |
| thumper | hmm... looks like reviewboard host is out of disk space | 01:27 |
| davecheney | rup row | 01:28 |
| davecheney | score one for the cloud | 01:28 |
| thumper | the cloud, where everything just works | 01:28 |
| davecheney | yup, as well as it does right here in your home | 01:34 |
| davecheney | or your money back | 01:34 |
| axw | wallyworld: I'm not really sure what we need to say about iaas resource tagging beyond what's in the release notes. do you have any ideas? | 01:40 |
| wallyworld | axw: from memory the release notes seemed to cover it. they explained what was done and how to ass custom tags etc. so maybe just copy to a PR on the doc project | 01:41 |
| axw | wallyworld: ok | 01:41 |
| wallyworld | we just need to make sure the doc work is scheduled - hence the PR / bug being done | 01:42 |
| axw | wallyworld: do I specifically need to do a PR? I'm not sure where this would best go, so ok if I just create an issue on the project with the text, and then keep an eye on it? | 01:43 |
| wallyworld | yeah, that will be fine | 01:43 |
| menn0 | thumper: is the RB machine also the main Jenkins host? | 01:45 |
| thumper | NFI | 01:45 |
| menn0 | thumper: just checked, they are | 01:46 |
| axw | menn0: reviews. and juju-ci. both resolve to the same | 01:46 |
| menn0 | axw: yep, i just checked the same thing :) | 01:46 |
| axw | :) | 01:47 |
| menn0 | so if RB is having trouble them Jenkins will be as well | 01:47 |
| thumper | hazaah | 01:47 |
| menn0 | thumper: i'm on the host now... root volume is certainly full | 01:48 |
| thumper | menn0: go delete some stuff | 01:48 |
| thumper | pretty sure we don't need /var | 01:49 |
| thumper | :-) | 01:49 |
| menn0 | i'm just looking for where the space is being used | 01:49 |
| * thumper chuckles to himself | 01:49 | |
| menn0 | it's painfully slow | 01:49 |
| thumper | hmm | 01:49 |
| thumper | need coffee | 01:49 |
| menn0 | thumper: you there? | 01:58 |
| thumper | yup | 01:58 |
| thumper | ugh... snow | 01:58 |
| davecheney | thumper: ready when you are | 01:58 |
| menn0 | thumper: the culprit is the logs for the juju env that hosts the various CI services (reviews, CI proxy, reports) | 01:58 |
| thumper | hah | 01:59 |
| thumper | no rotation? | 01:59 |
| menn0 | thumper: the disk isn't really big enough to support the way Juju rotates the logs | 01:59 |
| thumper | heh, oh the irony | 01:59 |
| menn0 | thumper: there's several units each with 2 backups of 300MB plus the current log file | 01:59 |
| menn0 | the disk is only 7GB total | 02:00 |
| menn0 | thumper: also the logs are full of: exited "rsyslog": x509: certificate signed by unknown authority (possibly because of "crypto/rsa: verification error" while trying to verify candidate authority certificate "juju-generated CA for environment \"rsyslog\"") | 02:00 |
| menn0 | weee | 02:00 |
| * thumper stabs rsyslog | 02:01 | |
| menn0 | wallyworld/thumper/davecheney: do you remember if the above has been fixed/ | 02:01 |
| menn0 | ? | 02:01 |
| thumper | I keep seeing it... | 02:01 |
| wallyworld | hmmm | 02:01 |
| wallyworld | i thought it had been | 02:02 |
| * menn0 is compressing the backup logs | 02:02 | |
| menn0 | that's better... 3.5GB free | 02:03 |
| menn0 | wallyworld, thumper: this env is 1.21. I don't think the problem has been fixed there. | 02:04 |
| menn0 | regardless, the disk is never going to be big enough | 02:05 |
| menn0 | i'll shoot out an email | 02:05 |
| anastasiamac | menn0: tyvm for fixing/cleaning up the machine :D | 02:05 |
| menn0 | anastasiamac: np | 02:06 |
| mup | Bug #1467362 opened: utils/ssh: data race in test <juju-core:New> <https://launchpad.net/bugs/1467362> | 02:45 |
| davecheney | thumper: menn0 git push --set-upstream origin fixedbugs/1465115 | 03:36 |
| mup | Bug #1465115: api: data race in test <intermittent-failure> <race-condition> <unit-tests> <juju-core:In Progress by dave-cheney> <https://launchpad.net/bugs/1465115> | 03:36 |
| davecheney | thumper: menn0 https://github.com/juju/juju/pull/2618 | 03:36 |
| davecheney | you'll love this one | 03:36 |
| * menn0 looks | 03:37 | |
| menn0 | davecheney: LGTM | 03:40 |
| menn0 | davecheney: nasty problem | 03:40 |
| davecheney | yeah, that is a terrible footgun in the http api | 03:42 |
| mwhudson | davecheney: so the situation with rugby is that it doesn't have any kind of useful outbound network access at all, right? | 03:55 |
| mwhudson | not even via proxy | 03:55 |
| mup | Bug #1467372 opened: api/cleaner: data race in test <juju-core:New> <https://launchpad.net/bugs/1467372> | 04:00 |
| davecheney | mwhudson: yup | 04:03 |
| davecheney | it broke a few weeks ago | 04:03 |
| davecheney | this isn't the first time it broke | 04:03 |
| mwhudson | davecheney: nice | 04:03 |
| davecheney | but this was the first time I no longer had the stregth to complain on #is | 04:04 |
| davecheney | mwhudson: if you would take up the charge this time, I would be indebted | 04:04 |
| mwhudson | davecheney: i'm happy to (tomorrow) if you can tell me what should be working | 04:04 |
| davecheney | there is a proxy | 04:05 |
| davecheney | it runs on batuan | 04:05 |
| davecheney | well it used to | 04:05 |
| davecheney | but it doesn' tnow | 04:05 |
| mwhudson | davecheney: https_proxy=http://squid.internal:3128/ ? | 04:05 |
| davecheney | this proxy isn't monitored by is | 04:05 |
| davecheney | so it shits tiselfe every now and then | 04:05 |
| mwhudson | i see | 04:05 |
| davecheney | and needs to be manually unshitted | 04:05 |
| mup | Bug #1467372 changed: api/cleaner: data race in test <juju-core:New> <https://launchpad.net/bugs/1467372> | 04:12 |
| mup | Bug #1467372 opened: api/cleaner: data race in test <juju-core:New> <https://launchpad.net/bugs/1467372> | 04:15 |
| mwhudson | davecheney: you got some arm64 hw recently, right? | 04:20 |
| mwhudson | that iirc you weren't very impressed with | 04:20 |
| davecheney | yes, and yes | 04:27 |
| mup | Bug #1447234 changed: juju prints "error" when deploying yet no units are in error <deployer> <lxc> <reliability> <juju-core:Expired> <https://launchpad.net/bugs/1447234> | 04:30 |
| mup | Bug #1467374 opened: worker/uniter/filter: ci test failure <juju-core:New> <https://launchpad.net/bugs/1467374> | 04:30 |
| mwhudson | davecheney: what was it? | 04:34 |
| * mwhudson disappears, will read backlog later | 04:34 | |
| davecheney | mwhudson: xgene | 04:40 |
| mup | Bug #1467379 opened: "attachmentcount" field not set when upgrading from 1.24 <storage> <juju-core:Triaged> <https://launchpad.net/bugs/1467379> | 05:21 |
| === mthaddon` is now known as mthaddon | ||
| === jamespag` is now known as jamespage | ||
| jam | dimitern: I'm going to miss standup, I have to run an errand | 08:40 |
| dimitern | jam, ok, np | 08:46 |
| voidspace | dimitern: ping | 09:06 |
| voidspace | dimitern: standup? | 09:06 |
| fwereade | is anyone free to take a look at RB? [Errno 28] No space left on device: '/tmp/reviewboard.pcPtS2' | 09:09 |
| voidspace | dimitern: https://github.com/juju/juju/pull/2598 | 09:12 |
| === jcsackett_ is now known as jcsackett | ||
| === anthonyf is now known as Guest35024 | ||
| axw | evilnick: I possibly just inadvertently switched a checkbox on https://github.com/juju/docs/issues/444 | 10:56 |
| axw | evilnick: which one left as an exercise to the reader (I don't know which one it was :/) | 10:56 |
| evilnick | axw hehehe. Thanks! | 10:56 |
| axw | axw: sorry about that. didn't realise clicking them did things. | 10:57 |
| evilnick | those things are a mixed blessing | 10:57 |
| axw | untracked things at that | 10:57 |
| evilnick | it's okay, we are nearly done with them anyhow - it will be pretty easy for me to tell what is changed | 10:57 |
| axw | cool | 10:58 |
| tasdomas | is the reviewboard server out of disk space? I see this message: '[Errno 28] No space left on device: '/tmp/reviewboard.iG8Eys'' on http://reviews.vapour.ws/r/1963/diff/# | 11:50 |
| === coreycb` is now known as coreycb | ||
| Muntaner | hi jujuers | 12:32 |
| Muntaner | I'm having problems with a bootstrap | 12:32 |
| Muntaner | http://paste.ubuntu.com/11755548/ | 12:33 |
| dooferlad | dimitern / TheMue: could you take a look at https://github.com/juju/juju/pull/2621 please? ReviewBoard hasn't found it (still out of disk space?) so please review on Github. | 12:39 |
| mgz | Muntaner: what did you do to add an ubuntu image to your deployment and register it with simplestreams? | 12:39 |
| mgz | Muntaner: I'm presuming you've read and followed jujucharms.com/docs/stable/howto-privatecloud | 12:41 |
| dimitern | dooferlad, looking | 12:41 |
| TheMue | dooferlad: *click* | 12:46 |
| Muntaner | mgz, does juju work with the new vivid cloud ubuntu images? | 12:47 |
| Muntaner | I mean, the 15.04 | 12:47 |
| Muntaner | because it worked flawlessly with the old ones (14.04), but with the new I'm getting strange errors | 12:47 |
| dimitern | dooferlad, done | 12:48 |
| mgz | Muntaner: yes, but in that bootstrap it's not finding the image stream at all | 12:48 |
| Muntaner | mgz, it was an error in the metadatas | 12:48 |
| Muntaner | mgz, now I managed to go on: I get another error that I'm pasting | 12:49 |
| Muntaner | mgz, http://paste.ubuntu.com/11756595/ | 12:49 |
| Muntaner | seems like it is searching for metadatas of a 14.04 version, why does it? | 12:49 |
| Muntaner | sorry, I pasted the same stuff for two times | 12:51 |
| TheMue | dooferlad: agreeing to dimitern comments ;) | 12:52 |
| mgz | Muntaner: looks like you are trying to bootstrap trusty and have no trusty images | 12:53 |
| Muntaner | mgz, mmmh | 12:53 |
| Muntaner | I'm not trying to bootstrap trusty: I wanna vivid | 12:53 |
| Muntaner | in my environments.yawl, I got a default-series: vivid | 12:54 |
| mgz | what's default-series in your environments.yaml? | 12:54 |
| Muntaner | mgz -> http://paste.ubuntu.com/11756623/ | 12:54 |
| Muntaner | I got vivid | 12:54 |
| Muntaner | mgz, also with juju bootstrap --debug --series=vivid --upload-tools I got the same result | 12:58 |
| mgz | Muntaner: yeah, --series doesn't do that | 13:00 |
| Muntaner | mgz, sooo ... | 13:01 |
| Muntaner | mgz, maybe I solved | 13:03 |
| Muntaner | via tools-metadata-url: https://streams.canonical.com/juju/tools/ | 13:03 |
| mgz | yeah, you also want to be able to access vivid tools, but doesn't seem to be where it's getting stuck from the logs | 13:04 |
| Muntaner | mgz, are you a developer? | 13:04 |
| Muntaner | aw yes, we are in dev chan | 13:05 |
| Muntaner | I think that juju logs need to be revisited | 13:06 |
| Muntaner | in other situations, I'm having a lot of troubles in understanding what isn't working | 13:06 |
| mgz | simplestreams is unreadable junk | 13:06 |
| mgz | and why we're still logging config contents >_< | 13:06 |
| Muntaner | mgz, will juju never work with containers? :) | 13:07 |
| mgz | ? it does. | 13:07 |
| mgz | so, can you now bootstrap or are you still stuck on simplestreams? | 13:08 |
| Muntaner | mgz, sorry, got confused! seems to be bootstraping, but now it's stuck at Installing package: cloud-image-utils | 13:08 |
| Muntaner | maybe I've got some issues in my openstack | 13:09 |
| dooferlad | dimitern: replied to that review (https://github.com/juju/juju/pull/2621) | 13:09 |
| dimitern | dooferlad, replied | 13:12 |
| dooferlad | dimitern: thanks. Will fix up as suggested | 13:13 |
| dimitern | dooferlad, cheers | 13:14 |
| dooferlad | dimitern: by the way, I will probably be a bit late for the networking knowledge sharing because I need to pick my daughter up. | 13:19 |
| Muntaner | maybe I'm having problems with security groups... | 13:23 |
| Muntaner | guys, shall I open some ports in my default security group? | 13:23 |
| Muntaner | because the environments gets bootstraped | 13:24 |
| Muntaner | but it stucks at the apt-get upgrade... | 13:24 |
| Muntaner | can't ssh to machines, can't get ssh status | 13:24 |
| Muntaner | juju status* | 13:24 |
| dimitern | dooferlad, ok, no worries | 13:24 |
| Muntaner | DEBUG juju.api apiclient.go:337 error dialing "wss://192.168.0.97:17070/environment/46eefeea-bd6a-43a9-8571-23a841643c0f/api", will retry: websocket.Dial wss://192.168.0.97:17070/environment/46eefeea-bd6a-43a9-8571-23a841643c0f/api: dial tcp 192.168.0.97:17070: connection refused | 13:24 |
| Muntaner | guys, anybody can help me in understanting why it's getting stuck at Installing package: cloud-image-utils | 13:32 |
| Muntaner | ? | 13:32 |
| Muntaner | mmmh... | 13:42 |
| Muntaner | where can I find the tgz of the tools? | 13:46 |
| natefinch | Muntaner: I haven't done this personally, but I Think this section shows the relevant topics: https://jujucharms.com/docs/stable/howto-privatecloud#image-metadata | 13:49 |
| Muntaner | natefinch, I think I fixed the metadata problem | 13:50 |
| natefinch | Muntaner: cool | 13:51 |
| Muntaner | natefinch, the host machine got lost in this: | 13:52 |
| Muntaner | in the /var/log/cloud-init-output.log, I got this: | 13:52 |
| Muntaner | http://paste.ubuntu.com/11756909/ | 13:53 |
| Muntaner | a lot of | 13:53 |
| natefinch | Muntaner: hmm | 13:54 |
| natefinch | Muntaner: do the machines have outside access to ubuntu package archives? | 13:59 |
| Muntaner | natefinch, naturally, yes | 14:00 |
| Muntaner | natefinch, now It works, maybe I got some network hic-cups | 14:02 |
| Muntaner | a non juju-related question: do skype work for you? for me, it crashes after 5 seconds under xubuntu and fedora | 14:02 |
| natefinch | Muntaner: I don't use skype, just google hangouts. Works very reliably. Not sure if they use similar technology. | 14:04 |
| Muntaner | natefinch, I got a problem with endpoints | 14:18 |
| Muntaner | who is telling to the juju machine 0 the endpoints? | 14:18 |
| Muntaner | because it is looking for "controller:8774" | 14:19 |
| Muntaner | but naturally, the vm can't know who is "controller" | 14:19 |
| natefinch | Muntaner: I'm not sure where that information is coming from. That's not hardcoded or anything (neither is that port, I don't see 8774 in the code at all) | 14:21 |
| Muntaner | natefinch, I think it is asking to my openstack "what are your endpoints"? | 14:22 |
| mgz | Muntaner: it's in your keystone config | 14:22 |
| Muntaner | so I probably neet to change them | 14:22 |
| Muntaner | need* | 14:22 |
| Muntaner | it's fine :) | 14:22 |
| voidspace | dimitern: dooferlad: network problems are due to ethernet-over-power hardware problems | 14:30 |
| voidspace | still seeing if they can be resolved or if I need new hardware | 14:30 |
| Muntaner | guys | 14:31 |
| Muntaner | I'm trying to deploy juju-gui | 14:31 |
| Muntaner | on my fresh vivid environment | 14:31 |
| Muntaner | does it exists for vivid? | 14:31 |
| Muntaner | 'cos I'm getting a sad "ERROR juju.cmd supercommand.go:430 cannot resolve charm URL "cs:vivid/juju-gui": charm not found" | 14:31 |
| natefinch | Muntaner: I don't think the gui charm exists for vivid. Most charms don't exist for vivid. rick_h_ would know ^ | 14:35 |
| Muntaner | aw! | 14:36 |
| natefinch | Muntaner: you could always copy the charm and put it on launchpad under your own name for vivid... in fact, I wouldn't be surprised if someone else already had | 14:37 |
| mgz | you can always download a trusty charm, rename the series, and try deploying from local: | 14:37 |
| natefinch | that too ^ | 14:37 |
| rick_h_ | natefinch: correct, we're only LTS | 14:37 |
| natefinch | It's sort of unfortunate that we tie charms to series so tightly, when a lot of the time they work fine in other series. | 14:38 |
| dimitern | voidspace, oh boy :/ one drawback of ethernet-over-power | 14:48 |
| voidspace | dimitern: yeah | 14:48 |
| voidspace | dimitern: a router or network card can just as easily fail too though | 14:48 |
| voidspace | dimitern: I don't think they're inherently unreliable, just one extra piece that can go wrong | 14:48 |
| voidspace | dimitern: looks like the remote one has just died | 14:48 |
| voidspace | dimitern: a system reset on the main one hasn't helped and according to the diagnostic tool I have the remote one isn't working | 14:49 |
| voidspace | isn't being detected at all | 14:49 |
| Muntaner | hey guys | 14:50 |
| Muntaner | last thing :) | 14:50 |
| dimitern | voidspace, but what's the problem you've discovered? | 14:50 |
| Muntaner | I should deploy a private local bundle on my fresh juju | 14:50 |
| Muntaner | I've got my bundle.yawl there | 14:50 |
| voidspace | dimitern: well, as the remote unit doesn't work I have no network | 14:50 |
| Muntaner | what was the command? | 14:50 |
| voidspace | dimitern: and my networking configuration for the machine requires eth0 to be connected | 14:50 |
| dimitern | voidspace, remote unit being the other end of the EoP ? | 14:50 |
| voidspace | dimitern: yep | 14:51 |
| mgz | Muntaner: I should have mentioned earlier, but you'd really be better off in #juju rather than here | 14:51 |
| voidspace | dimitern: the end my desktop is connected to | 14:51 |
| dimitern | voidspace, right | 14:51 |
| dimitern | voidspace, can you use a cable instead? | 14:51 |
| mup | Bug #1467556 opened: TestMachineAgentRunsEnvironStorageWorker fails <ci> <intermittent-failure> <test-failure> <juju-core:Incomplete> <juju-core 1.24:Triaged> <https://launchpad.net/bugs/1467556> | 14:53 |
| Muntaner | mgz, sorry | 14:53 |
| voidspace | dimitern: my desktop is upstairs, so no | 14:59 |
| voidspace | dimitern: I can go back to wifi and order a replacement - | 14:59 |
| * TheMue is shortly afk to ride back home in a dry moment ;) | 15:01 | |
| dimitern | voidspace, I see, well I hope you figure out how to fix it :) | 15:02 |
| voidspace | dimitern: I can probably set up a virtual maas with maas 1.8 to test code against until the replacement arrives | 15:03 |
| voidspace | dimitern: I can't run my "real maas" without working ethernet | 15:03 |
| dimitern | voidspace, sounds good, and I can give you a hand with testing on both my maas-es | 15:03 |
| voidspace | dimitern: cool | 15:04 |
| voidspace | dimitern: thanks | 15:04 |
| dimitern | dooferlad, can you update bug 1463480 if there's anything you've missed ? | 15:07 |
| mup | Bug #1463480: Failed upgrade, mixed up HA addresses <blocker> <canonical-bootstack> <ha> <upgrade-juju> <juju-core:Triaged> <juju-core 1.22:Triaged> <juju-core 1.24:Triaged> <hacluster (Juju Charms Collection):New> <https://launchpad.net/bugs/1463480> | 15:07 |
| === kadams54 is now known as kadams54-away | ||
| mup | Bug #1467590 opened: Running out of disk space blocks interacting with env on cli <juju-core:New> <https://launchpad.net/bugs/1467590> | 16:26 |
| === ericsnow__ is now known as ericsnow | ||
| katco` | ericsnow: natefinch: planning meeting | 17:02 |
| === katco` is now known as katco | ||
| natefinch | katco: coming | 17:03 |
| === kadams54-away is now known as kadams54 | ||
| === anthonyf is now known as Guest16072 | ||
| === kadams54 is now known as kadams54-away | ||
| === anthonyf is now known as Guest26733 | ||
| === kadams54-away is now known as kadams54 | ||
| === kadams54 is now known as kadams54-away | ||
| === bcsaller_ is now known as bcsaller | ||
| === kadams54 is now known as kadams54-away | ||
| katco | ericsnow: wwitzel3: so i've never actually worked on a hook before. where are those defined? | 21:29 |
| katco | ericsnow: wwitzel3: gh.com/juju/charm? | 21:30 |
| katco | /hooks? | 21:30 |
| ericsnow | katco: what do you mean by "hook"? | 21:30 |
| katco | ericsnow: this is for the launch command | 21:30 |
| ericsnow | katco: we aren't adding any hooks | 21:31 |
| perrito666 | katco: hooks are in the charm package | 21:50 |
| katco | ericsnow: k, think i've found it: uniter/runner/jujuc/. however, where should ours live under process? | 22:00 |
| ericsnow | katco: we already wrote all that | 22:00 |
| ericsnow | katco: process/context | 22:00 |
| ericsnow | katco: see register.go for a hook context command | 22:01 |
| ericsnow | katco: launch be very similar | 22:01 |
| katco | ericsnow: awesome. ty | 22:01 |
| ericsnow | katco: :) | 22:01 |
| === mup_ is now known as mup | ||
| thumper | wallyworld: I thought you said this was fixed? http://reports.vapour.ws/releases/2801/job/run-unit-tests-precise-i386/attempt/2149 | 22:31 |
| wallyworld | it was - i checked the commits | 22:32 |
| wallyworld | if it's still broken there's maybe a regression or another problem? | 22:32 |
| wallyworld | in a meeting, will check soon | 22:32 |
| mup | Bug #1467690 opened: inconsistent juju status from cli vs api <canonical-bootstack> <juju-core:New> <https://launchpad.net/bugs/1467690> | 22:36 |
| mwhudson | davecheney: rugby can talk to its proxy again | 22:41 |
| wallyworld | thumper: i had a look at the logs for that test - i call bullshit | 23:05 |
| wallyworld | i can't see an int overflow there | 23:05 |
| thumper | cmd/juju/addrelation_test.go:18: undefined: CmdBlockHelper | 23:05 |
| wallyworld | there's a uniter test failure relating to an upgrade test | 23:05 |
| thumper | I'll look, but curious that it doesn't fail all the time | 23:05 |
| wallyworld | yeah, go figure | 23:06 |
| axw | wallyworld: conn died | 23:26 |
| davecheney | mwhudson: thanks | 23:35 |
| davecheney | mwhudson: actually | 23:36 |
| davecheney | it's not working for me | 23:36 |
| mwhudson | hm | 23:36 |
| mwhudson | i managed to clone go a few minutes ago | 23:36 |
| davecheney | are you still in #is | 23:36 |
| mwhudson | i never leave! | 23:36 |
| davecheney | welcome to canoincal, when raising RT's is for the weak | 23:46 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!