[02:05] axw: meeting? [02:05] wallyworld: sure [02:05] rotating team meeting [02:05] see cal [02:05] oh right, core team [02:05] whoops [02:06] :-) [02:07] perrito666: if you're still around [02:37] thumper: want the 1:1 now? [02:37] wallyworld: yeah, in our normal hangout [02:38] already there :-) [03:04] * thumper goes to make a coffee [03:21] axw, I updated bug 1324255 to explain that the juju and the slave were dead from disk shortage and zombie mongos. from other tests [03:21] <_mup_> Bug #1324255: Local provider cannot boostrap

[03:22] sinzui: hmm ok. I'm pretty sure I saw it on the amd64 tests to [03:22] too* [03:24] axw, CI doesn't have a trusty local lxc test yet [03:24] sinzui: sorry, I was unclear; I meant the unit tests themselves see the "Closed explicitly" error. not in local deployment, just running the unit tests [03:24] sinzui: e.g. http://juju-ci.vapour.ws:8080/job/walk-unit-tests-amd64-trusty/485/console [03:24] it may be unrelated, I'm not sure [03:26] I see [04:22] * thumper making dinner, back in about 2 hours for a bit more === thumper is now known as thumper-afk [05:06] https://codereview.appspot.com/98700043 :) === vladk|offline is now known as vladk [06:15] axw: thanks for review, more info on reason for new interface etc is here https://docs.google.com/a/canonical.com/document/d/1J6EZ37gfiZdVMdoxdFrih4R27LCcqyBzi50wjWFjbiA [06:18] okey dokey, will have a read [06:18] wallyworld: I think calling it "err" will also be a problem, because an "err" in the body will shadow it [06:19] axw: err in body is not meant to be a new var [06:19] that's why i used = not := [06:19] wallyworld: not on the first line of the function [06:20] that's because file is new but err will not be as it's the 2nd var and there's one in scope [06:20] or have i got that wrong? [06:20] mm I think it only counts if it's declared in the same scope. will verify [06:20] I could be full of crap [06:20] the err at the top is in the same scope afaik [06:20] bbiab, gotta pick up kid [06:21] axw: i was tempted to leave the err in the defer as err := but thought it could be confusing so changed its name even though not necessary [06:21] wallyworld: you are correct [06:21] np [06:22] that's why i hate := and = [06:22] too easy to make subtle mistakes [06:22] now i'm really bbiab === Ursinha-afk is now known as Ursinha [08:08] morning all [08:30] is anyone working on the "juju doesn't bootstrap on precise" issue? [08:30] I suspect it's my fault [08:31] I'm spinning up a precise VM to look at it [09:06] voidspace: if you can reproduce it, I have a small change I'd love to test [09:07] I'll also start a precise image on AWS [09:08] I'm creating a local VM. Still downloading precise - my "old" VM was 13.10 not precise. [09:11] voidspace: from what I've found in Googling, it seems that mgo will kill sockets for servers that it doesn't think are available in the cluster [09:11] voidspace: so I've made a small change to copy (rather than clone) the session when running replSetInitiate, to force it to create a new socket [09:14] jam: ping [09:18] vladk: pong [09:19] jam: I'd like to discuss my further steps in developing Networker API [09:19] sure, can we get dimitern involved as well? [09:20] jam: of course [09:20] (well that was mostly just to ping him) [09:20] dimitern: ping [09:21] jam: can we do it on hangout? [09:21] certainly, I was just waiting to see if dimitern was actually around [09:24] jam, vladk, i'm here [09:25] k I need about 30s [09:26] we can just use our team standup hangout [09:27] https://plus.google.com/hangouts/_/canonical.com/juju-sapphire [09:27] dimitern: vladk ^^ [09:29] axw, pre-commented on https://codereview.appspot.com/102860043/ -- let me know if it sounds sane to you [09:31] axw: awesome, that sounds very useful [09:31] axw: might help test issues as well [09:31] axw: we start mongo ok, but then fail to connect to it [09:31] axw: (when using replica sets for tests) [09:31] axw: and it's not a timing issue, we wait sixty seconds [09:32] axw: I'm creating a branch with my changes backed out - so once I get the VM setup I can quickly tell if it's my changes [09:32] fwereade: thanks, seems sensible to put it on RelationUnit. in my PoC I did the address selection TODO in the uniter, which is a bit dumb I guess. [09:32] fwereade: yes, free for a chat tomorrow [09:32] voidspace: we wait sixty seconds for what? [09:33] axw: we wait up to sixty seconds (mongoDialupTimeout I believe) when trying to connect to the mongo server we have created for tests [09:33] axw: you're missing some context [09:34] axw: as well as using replica sets for local provider I wanted to switch us to using replica sets for tests too [09:34] I see [09:34] axw: so that our test mongo matches the production mongo [09:34] we're already doing that in a bunch of tests - you want to do it in *all* of the,? [09:34] them* [09:34] axw: but when I do that, many tests fail with "unreachable servers" - many of them deterministically [09:34] axw: yep [09:34] ok [09:35] JujuConnSuite *doesn't* [09:35] the driving motivator is that we want to switch to write-majority [09:35] sounds like it will add quite a bit of overhead to the tests. setting up a replica set can take quite some time [09:35] axw, no worries, we converge :) [09:36] the command to switch to "write majority" (WMode) errors if you're not using replica sets [09:36] :/ [09:36] so one fix is to use replica sets everywhere [09:36] that's what we were attempting [09:36] it maybe that because of precise we *can't* [09:42] * axw sighs [09:42] of course it bootstraps first time for me [09:42] of course it'd help if I had pulled trunk [09:42] heh [09:51] axw: this branch reverts my local provider replica set changes [09:51] axw: lp:~mfoord/juju-core/revert-local-replset [09:51] ok [09:51] or "bzr merge -r 2803..2802" [09:58] voidspace: if you manage to reproduce the error with replSet enabled, try this patch: http://paste.ubuntu.com/7542857/ [09:58] (please) [10:00] axw: yep [10:00] axw: precise installing now [10:00] axw: you couldn't reproduce? [10:00] voidspace: nope :( [10:00] odd [10:03] mgz: 1:1 time [10:13] wallyworld: whoops, not looking, joining now [10:17] voidspace: it's not just precise. http://juju-ci.vapour.ws:8080/job/local-deploy-trusty-ppc64/ [10:17] there's a bunch of failures there on r2807 [10:18] "no reachable servers" [10:28] axw: :-/ [10:28] axw: configuring my vm to try it [10:31] 215mb of updates... [10:35] hi all [10:36] menn0: morning :-) [10:37] menn0: howsit? [10:37] voidspace: alright [10:37] voidspace: how are you? [10:37] voidspace: we had a bit of household chaos here this afternoon so I'm catching up now. have one big email to send to #juju-dev today. I'm off until Tuesday. [10:39] menn0: ah, fun [10:39] menn0: I hope "household chaos" is good chaos [10:39] voidspace: not really, but all fine now [10:41] menn0: :-/ [10:41] we're all good here, lots going on [10:41] better a hectic life than a boring one though [10:42] trying to reproduce precise bugs at the moment [10:42] voidspace: I think there's a balance :) [10:42] voidspace: yeah I heard about that problem [10:42] voidspace: tricky to reproduce? [10:43] menn0: so far [10:43] menn0: I'm just updating and configuring a precise vm to see if I can reproduce [10:43] axw has failed so far === Ursinha is now known as Ursinha-afk === thumper-afk is now known as thumper [10:56] thumper: hi again :) [10:56] o/ [11:01] cmars, fwereade: joining us? [11:03] ahhhh stupid hangouts keeps crashing [11:11] natefinch: morning [11:18] voidspace: morning === Ursinha-afk is now known as Ursinha [11:51] so if I try to bootstrap with the local provider on precise it tells me that juju-local is required [11:51] but juju-local is also unavailable for precise [11:51] ahh hrm [11:51] do I need to add a ppa? [11:51] looking [11:53] natefinch: "Due to needing newer versions of LXC the local provider does require a newer kernel than the released version of 12.04." [11:53] Therefore we install Linux 3.8 from the LTS Hardware Enablement Stack: [11:55] natefinch: I've added ppa:juju/stable and juju-local is now available [11:55] looks like I need to upgrade the kernel too [11:56] voidspace: I used cloud-archive:tools [11:56] axw: is there a difference? [11:57] I guess "stable" is likely to be out of date [11:57] I mean to get the newer lxc [11:57] ah [11:57] hmmm [11:57] hm, I guess you got an older 12.04, I probably got 12.04.03 [11:57] I did a full update [11:58] ok I dunno [11:58] axw: those quotes were from the docs, not an error message [11:58] https://juju.ubuntu.com/docs/config-LXC.html [11:59] axw: I might try without the new kernel first [11:59] maybe it will fail in the same way... [11:59] voidspace: `sudo apt-add-repository cloud-archive:tools` if you've not [11:59] although Curtis is pretty switched on, so I doubt that's the problem [11:59] mgz: ok [12:00] mgz: you sound like you know what you're talking about :-) [12:00] worth checking your kernel version too, by default upgrades may not give you a newer kernel [12:00] * axw disappears [12:00] axw: o/ [12:00] night axw [12:00] mgz: they're unlikely too [12:00] *to [12:01] mgz: would you reccommend I remove the juju/stable ppa? [12:02] if you're not trying to use the *client* we have released on that machine, yes [12:02] jam. dimitern: what is your way to keep go packages up to date? [12:02] 'go get -u ./...' damages my working juju-core, so I wrote a script http://pastebin.ubuntu.com/7543496/ [12:02] vladk: yeah, go get is dangerous [12:03] vladk: I generally run just "godeps -u dependencies.tsv" and then go update the ones that are out of date [12:03] you can just "go get -u github.com/juju/errors/…" for example [12:03] vladk: ^that's what I do too [12:03] mgz: I'm trying to use trunk, thanks [12:03] well, I don't use go get at all, I pushd to the branch and git pull or whatever [12:04] mgz: sure, I've done that as well, depends on my mood, I guess [12:05] jam: and yeah, I do use go get as well, if say we've added a new dep with its own deps [12:05] ok, installing new kernel [12:05] so lunch [12:06] vladk: jam, mgz: I want to update all of packages [12:14] vladk, that's a handy script, thanks [12:14] hello [12:15] wwitzel3: morning [12:15] so, juju bootstrap on precise, with trunk, seems to work fine for me [12:15] so after not sleeping more than 5 hours the last 3 days, I finally zonked out and managed to sleep 10 hours. [12:15] dimitern: use with care, not fully tested, I'm not sure about pull options [12:15] feels good, like I'm sane again :) [12:15] going on lunch [12:15] will dig deeper after that [12:19] night folks [12:20] vladk, sure, i'll give it a try [12:21] wwitzel3: that's good news, welcome back to the land of the living [12:22] voidspace: thanks, enjoy lunch [13:03] cmars, hazmat, fwereade: omnibus review [13:05] good sort of morning [13:08] sort of morning? [13:09] mgz: I usually start around or before 7AM now its 10 [13:11] sinfully late [13:45] perrito666: it is a trend today, I didn't get on until just after 8 and I even went to bed right after the meeting last night. [13:47] wwitzel3: I did too, it was like midnight when the meeting ended, but this AM I just spent 2.5hs in a queue trying to do some bureaucracy and then an extra half an hour having to exit and re enter the city because the path to go trough the center was blocked by a cluster of strike protests [13:48] that sounds less fun than sleeping [13:48] yep [13:48] fwereade: ping [13:48] wwitzel3, pong, listening, but in meetings, so responding slowly [13:50] fwereade: np, I was pretty dense yesterday and I don't think I retained the full value of our earlier conversation. [13:51] fwereade: looking at the presence code, I'm not seeing us actually use WaitAgentPresence anywhere. We define it and have tests. [13:51] fwereade: I'm wondering if we should just be using that with or instead of SetAgentPresence in the Pinger. [13:52] wwitzel3, not quite following -- I know we don't use WaitAgentPresence except in tests, though, but I don't get the larger thread [13:53] fwereade: part of the problem is I'm still cloudy on the "Why" SetAgentPresence doesn't think the agent is there when it is called the first time. [13:53] wwitzel3, I *think* it's because it uses "2 consecutive pings" as the "yep, it's there" condition [13:54] wwitzel3, but please validate that against reality [13:54] wwitzel3, (ie if you've only had one ping it's not yet considered presence) [13:58] Wow, I just realized the Juju Core Team meeting was last night not tonight. Stupid time zones making my Thursday meeting on Wednesday :/ [13:59] yup, do we have stdup today? [13:59] hah [13:59] I just set up an alarm for the day before, so I know in advance when that meeting is [14:00] I had skipped thursdays, because we used to have the juju core meeting before it, but now that the juju core meeting is moving, I think I need to reinstate the thursday standup. Short answer: yes [14:00] fwereade: I'm not seeing the two pings, following the trail from api status -> machine.AgentPresence -> presence.Alive which which fires a watcher.sendReq for the machine key and we select over the result and return bool,err back up the stack. [14:02] fwereade: I can make the problem with a long enough wait and recalling SetAgentPresence, but that isn't a legitimate fix and I am still failing to really grok the problem in the first place. [14:03] wwitzel3, voidspace: standup? [14:03] natefinch: oops, sorry [14:03] wwitzel3: it's ok, it wasn't on the calendar for today until now === dannf` is now known as dannf [14:19] fwereade, natefinch, mgz, I'd love a review on this networks constraints CL https://codereview.appspot.com/102830044 [14:20] wow! LP now supports inline comments on MP diffs! \o/ [14:20] wow [14:20] is it too late now to not switch to github? :D [14:21] dimitern: hah [14:23] haha [14:24] fwereade: did you and Ian figure out if we should use --replset in tests? [14:24] wwitzel3, ok, I may be wrong there: is it just the case that the first ping is not landing quickly enough to be seen? [14:25] natefinch, essentially yes [14:26] natefinch, for the tests that *by their fundamental nature* require mongo, we should be using --replset, because those really are integration tests and we should mimic a real environment as closely as possible [14:27] natefinch, but we have many tests that only coincidentally require mongo, and in those cases --replset is not required, because we should be aiming to make those tests unitier anyway -- to the point where they don't depend on mongo [14:28] natefinch, in particular replset clearly needs a real mongo, and almost certainly peergrouper too; and very probably state [14:29] natefinch, other test that merely happen to use state should be migrating away from the mongo dependency anyway, and so the loss of --replset is not a big deal [14:29] natefinch, how upsetting is this from your perspective? [14:30] sinzui: ping [14:30] hi voidspace [14:30] sinzui: ah, actually ping for five minutes [14:30] fwereade: it seems very squishy. Deciding if a test "really" needs mongo is non-trivial. [14:30] I assumed you'd be slower :-) [14:31] natefinch, well, state/presence and state/watcher are the other two definite ones [14:31] fwereade: btw are you on the cross team meeting, or should I jump on? [14:31] natefinch, but generally anything we *could* back with a double for state.State is IMO a candidate [14:32] natefinch, I'm not, alexis is; your presence couldn't hurt though :) [14:32] heh ok [14:36] fwereade: thank you, I think you've given me an epiphany [14:56] I've been asked to speak at PyCon India [14:56] A good opportunity to promote juju [14:56] Have to work on a snappy demo [14:57] Cool [15:00] voidspace: sweet [15:00] voidspace: you can try the one from LAS [15:00] fwereade, did you get a chance to check out my MR? [15:01] perrito666: Las Vegas? [15:01] that was pretty IBM centric [15:01] I'm looking for direction to either dig deeper on the gojsonschema to add (id) references, or to move forward towards state and param validate [15:01] something equally impressive though I hope [15:01] voidspace: yes, but nevertheless pretty [15:01] yep [15:01] something python related would be good [15:02] voidspace: some of the material used on the os demo, minus the lego part [15:02] I'll talk to the online services guys as they're charming up their infrastructure [15:02] bodie_: I'm on it [15:02] deploying and updating a django app would be good [15:02] and scaling out [15:02] fwereade, mgz, natefinch, a gentle reminder about https://codereview.appspot.com/102830044/ [15:02] voidspace: oi! that's my planned demo for PyConUK :D [15:03] bloodearnest: awesome, I can use yours [15:03] dimitern: I'm not wild about that syntax... [15:03] oops, I mean "we can work on it together"... [15:03] mgz, it's decided from above [15:03] voidspace: sure you meant that :p [15:03] dimitern: code seems fine so far [15:03] :-) [15:04] mgz, I'll wait for jam and/or fwereade to take a look as well, because i'm not sure about a few bits, but any comments are welcome [15:04] voidspace: that is cool, I only get invited to pycons I cant afford at times I cant travel :p [15:05] voidspace: when is pycon india? [15:05] perrito666: right, thankfully they're covering the costs or I wouldn't be able to afford it either [15:05] bloodearnest: end of September [15:05] so after PyCon UK I think [15:05] kk [15:05] voidspace: so we should do a joint talk proposal for pyconuk then :) [15:06] thanks mgz, nothing too fancy really. just got your changes in to the github repo (all tests are green) but need to make a decision about gojsonschema itself [15:07] bloodearnest: heh, yeah - sounds good [15:07] fwereade indicated interest in getting it MR'd as such even if it's not a "real" merge request, just to get visibility [15:07] voidspace, http://pastebin.ubuntu.com/7544599/ [15:08] sinzui: looking, thanks [15:08] voidspace, I asked abentley to join us to discuss what the test is doing bad [15:09] sinzui: so that worked [15:10] sinzui: ok [15:10] sinzui: although revision 2802 on master worked [15:10] so *something* changed [15:10] abentley: hello [15:10] voidspace: Hi. [15:11] voidspace, yes, but the only thing I think changed is that I renamed the test to from local-deploy to loca-deploy-precise-amd64 [15:11] heh [15:12] it was failing deterministically before that [15:12] abentley, voidspace I can try to restore the test name [15:13] that *really* shouldn't make any difference [15:13] oh but ci is still running [15:13] I can copy the test to the old name [15:13] oh, voidspace, I haven't run upgrade yet. I will do that to get the new lxc [15:13] sinzui: so the initial manual bootstrap works, but we think that re-bootstrapping from the generated .jenv file doesn't? [15:14] sinzui: Are you re-bootstrapping from old jenvs? [15:14] abentley, voidspace, the env name and the pregenerated jenv are the difference [15:14] or by "generated jenv file", do you mean your pre-canned one? [15:14] abentley, I used true local with a jenv [15:15] abentley, I used true local withOUT a jenv [15:16] can I get a copy of the failing jenv file? [15:16] sinzui: And you used an old jenv for testing local-deploy-precise-amd64? [15:17] voidspace: We don't have a pre-canned jenv. [15:17] I can't see it in the workspace to get at it [15:17] voidspace, abentley. It is generated. not pre-canned. We let juju make the jenv file. We update the version in it to ensure juju does what we and enterprises mean [15:17] ah, I see [15:17] bodie_: see comments on review, one thing needs fixing and some other comments, then lets land [15:17] voidspace, we tell users to share jenvs. we are [15:17] cool [15:17] sinzui: We leave the .jenv entirely untouched. It's the yaml that we tweak. [15:18] I am tempted to do the lxc upgrade, then ask jenkins to run the test as usual [15:19] sinzui: ok [15:19] mgz, I agree about pulling out the yaml -- is there a place in the repo where test files could go? there will be a charm actions yaml loader tested separately as part of dir (I think it was) [15:20] can I get at the CI scripts? I'm reading the logs, but it looks like a lot of the work is in deploy_job.py [15:20] bodie_: it would be an improvement to just have them as top level variables so that's easy later, splits the test logic a bit but easier to pull apart later [15:21] voidspace: the branches should be public [15:21] mgz: ok [15:21] voidspace: bzr branch lp:juju-ci-tools [15:21] yep, just found it :-) [15:22] mgz: thanks [15:23] mgz, I moved them to local scope due to fwereade's comments here: https://codereview.appspot.com/94540044/diff/1/charm/actions_test.go#newcode21 [15:24] bodie_: lets not go back and forth on that then :) [15:24] voidspace: Every juju action is dumped to stdout so you should be able to reproduce the script actions from the log output. [15:24] abentley, voidspace jenkins failed again after the update [15:25] there's probably a case to be made for either approach. I'm personally kind of a fan of the way the gojsonschema tests work with external files, but in this case it's not really testing the schema as much as the YAML reader, which is simple enough [15:25] oh, I forgot the mem constraint [15:25] * sinzui run manually with the constraint [15:25] cannot initiate replica set [15:25] so it is the problem with mongo [15:25] and just as mysterious as last time we saw it apparently [15:25] frankban: reviewed https://codereview.appspot.com/92700044 [15:25] okay, going to put my kid down for a nap, then into the salt mines with me! [15:26] lick lick? [15:27] voidspace, where any of these mongo warning relevant to the failure http://juju-ci.vapour.ws:8080/job/local-deploy-precise-amd64/1371/console [15:27] rogpeppe: thanks [15:27] mgz -- after I get these tweaks in shall I start digging towards State then? [15:28] maybe start on a new branch to get dir and the loader landed [15:28] bodie_: yeah, I think so [15:28] we'll make sure this lands first :) [15:29] sinzui: unknown config field. Shouldn't be. [15:29] I'll look at the last successful build, see if those warnings are new. [15:30] nope, those warning were there before [15:30] voidspace, the warning is from a vestigial environments.yaml. I think I can remove it now, juju 1.18 doesn't need it. 1.16 did [15:33] voidspace, abentley maybe a memory issue...this is the failure from my manual run. I recalled the good bootstrap command from bash history, then appended the mem=2G http://pastebin.ubuntu.com/7544723/ [15:34] I'll try that locally [15:34] I didn't use the memory constraint [15:35] although my VM only has 1G! [15:35] machine has lots of memory [15:35] sinzui: Hmm. I wouldn't have thought that had an effect. Especially since mongo runs outside the lxc. [15:35] * sinzui tries again without mem=2G [15:35] abentley, agreed [15:36] nope, works [15:36] for me [15:36] if we can't figure out the difference we'll have to back out the replica set change [15:36] abentley, I believe 1G should be adequate for our testing on lxc anyway [15:36] I'm looking up that specific failure (cannot initiate replica set) [15:37] natefinch: ^^^ [15:37] natefinch: the precise failure is consistent when run as a CI job [15:37] natefinch: although effectively the same command run on the CI machine succeeds [15:37] natefinch: the specific failure is "cannot initiate replica set" [15:38] natefinch: so the replica set is the problem, although it's not obvious why [15:38] mongo logs probably have a better message. [15:39] that error implies that we have a mongo started *without a replica set* and then tried to add a replica set [15:39] or we tried to add a non-empty mongo to a replica set [15:40] it's the initiation which has to happen after it starts [15:40] but we start mongo with the replset argument [15:40] that's not initiation [15:41] nope, but initiation will fail if the db isn't empty [15:41] voidspace, abentley I bootstrapped again manually *without* the mem constraint and it failed. This is the relevant errors: http://pastebin.ubuntu.com/7544771/ [15:41] voidspace: you are missing a step there? [15:41] natefinch: https://jira.mongodb.org/browse/SERVER-6294 [15:41] voidspace: I don'tb think that's true. [15:42] natefinch: that's the closest I could find to a reference [15:42] natefinch: it's local.* that needs to be empty [15:42] perrito666: I'm sure I'm missing lots of step [15:42] voidspace: ahh, yeah... I don't think that's the problem [15:44] usually it's that mongo can't resolve the address that you gave it during initiate, or you don't have access to the admin database [15:44] rebootstrapping from an existing jenv worked for me [15:46] natefinch: should I back out the change to unblock CI? [15:46] natefinch: without a way to repro it's going to be hard to re-introduce though [15:47] what's the version of mongo on ci vs your vm? [15:48] sinzui: can you check the mongo version on the CI machine for me please [15:48] natefinch: I'm using 2.4.6 it looks like [15:48] voidspace, Installed: 1:2.4.6-0ubuntu5~ubuntu12.04.1~juju1 [15:49]