[01:26] wow. such merge. very conflict [01:29] davecheney: is that the output of "git merge --doge"? [01:31] thumper: you got time for a quick hangout? [01:31] wallyworld: hey, I should be doing my overdue 360 review for fwereade [01:32] thumper: well, do that first [01:32] heh [01:32] maybe ping me when you are free? [01:32] wallyworld: I have 1:1 calls for 1.5 hours starting in 30 minutes [01:32] :-( [01:32] sure you don't want to chat now? [01:32] ok [01:32] https://plus.google.com/hangouts/_/canonical.com/foo [01:45] wallyworld: how come we have gopkg.in depdenendcies in juju now [01:45] I thought it was very clear they were not going to be added without further discussion [01:45] davecheney: cause someone addd it? i have no idea. i didn't agree but it was ut of my hands [01:45] i was outvoted [01:45] fine [01:46] sorry [01:47] not yourfault [02:24] davecheney: we're going to have to anyway, labix.org/v2/mgo is not going to be updated anymore [02:24] and the github repo underneath can't really be used, because it refers to gopkg.in anyway [02:24] (I'm in the process of updating mgo) [02:32] axw: davecheney it was also brought up around the charmv2 stuff helping us version the api for that stuff [02:32] axw: davecheney though let me know if we need to work with IS to get something on prodstack. I'm all for making sure we run the stuff 100% correct [02:39] heh, then again I see you're referring to the charmstore [02:39] axw: just a heads up, we've got an in progress next generation in the branch v4 (which will be renamed to v2) https://github.com/juju/charmstore/tree/v4 [02:39] axw: not sure if you can/makes sense to make the same pr there as well [02:40] or we can make sure to pull it across, but fyi [02:42] rick_h__: yup, rogpeppe let me know, I have done so [02:42] thanks [02:42] axw: ty kind sir [02:43] rick_h__: as for hosting repo on prodstack; there's no point unless all of gopkg.in is on there [02:44] I don't think it's going to be any less reliable than labix.org, which hasn't really been a problem (except for a hiccup recently) [02:44] axw: ok, I know I had a debate with roger last week about 'well it's not been down' being a bit different than 'it's hosted to our standards' [02:44] axw: cool [03:02] axw: /me grumbles, loudly [03:03] thumper: hangout? [03:20] axw: when you have a moment, can we do a hangout? [03:20] waigani: sure [03:21] now's fine [03:21] axw: cool give me a sec [03:22] https://plus.google.com/hangouts/_/gu5bp7izdb5uz2sv6uzpz3fppaa?hl=en-GB [03:38] axw: could you take a peek at bug 1347715 to see any recent work we did may be responsible (one of the state server instance commits is blamed) [03:38] <_mup_> Bug #1347715: Manual provider does not respond after bootstrap

[03:38] okey dokey [03:38] ta [03:39] wallyworld: when you have a sec, https://code.launchpad.net/~axwalk/gomaasapi/update-mgo-test-dependency/+merge/228442 [03:39] sure [03:39] and there's a bunch off PRs on various juju repos [03:39] ok [03:43] thumper: thanks. don't suppose you have any better ideas about how to configure the mongod to use? [03:43] not really [03:43] seems like a solid enough approach [03:44] wallyworld: I just manually bootstrapped, worked fine.. :/ [03:45] axw: hmmm. i did think the error described in the bug (the upgrade quit one) was more noise than anything that would stop the agent working [03:46] the jenkins job linked in the bug is still red sadly [03:46] needs more investigation [03:49] yep. I'll see if I can run the CI script locally to repro [04:04] wallyworld: thanks [04:04] np [04:17] a review from someone who knows the machine agent would be appreciated: https://github.com/juju/juju/pull/407 [05:50] wallyworld: missed one before https://github.com/juju/blobstore/pull/11 === uru_ is now known as urulama [06:15] https://github.com/juju/juju/pull/408 [06:15] anyone [06:15] trivial [06:16] davecheney: done [06:18] thumper: ta [06:18] that had me scratching my head for an hour [06:19] due to a merge conflict that put that type back in a signature [06:36] davecheney: thanks. I'll have a look about fixing version in a followup [06:50] axw: ta [06:54] // TODO(dfc) wtf [06:54] if !canWatch(nil) { [06:54] return result, common.ErrPerm [06:54] } [06:55] ............................................________ [06:55] ....................................,.-'"...................``~., [06:55] .............................,.-"..................................."-., [06:55] .........................,/...............................................":, [06:55] .....................,?......................................................, [06:55] .................../...........................................................,} [06:55] ................./......................................................,:`^`..} [06:55] .............../...................................................,:"........./ [06:55] ..............?.....__.........................................:`.........../ [06:55] ............./__.(....."~-,_..............................,:`........../ [06:55] .........../(_...."~,_........"~,_....................,:`........_/ [06:55] ..........{.._$;_......"=,_......."-,_.......,.-~-,},.~";/....} [06:55] ...........((.....*~_......."=-._......";,,./`..../"............../ [06:55] ...,,,___.`~,......"~.,....................`.....}............../ [06:55] aww, where did the rest of it go [07:08] axw: sorry, was doing scholl pickup and dinner shopping [07:35] wallyworld: nps (sorry, was doing school pickup ;)) [07:35] :-) [07:36] wallyworld: thanks for approving leave. gotta do some moving/painting [07:36] np, have "fun" [08:00] morning [08:30] morning all [08:31] Some issues setting up a workstation in Romania, but I'm online [08:31] Should be better tomorrow [09:13] *gnah* [09:13] still fighting with my IPv6 routing between containers on differen nodes [09:24] axw: for future reference, we have CI enabled on the charm store repo. to merge, add a review comment containing :shipit: [09:26] oops, thanks rogpeppe1 [09:26] axw: and... you didn't update the dependencies.tsv file [09:26] gah [09:26] axw: so the branch is actually broken, i'm afraid [09:26] I will fix it [09:27] sorry [09:27] axw: thanks [09:27] remove me from committers after please ;) [09:27] axw: perhaps you could roll back tip, assuming nothing has landed in the meantime [09:27] sure [09:28] axw: ta. [09:44] wtf did I just do. reverted my revert. well done [09:44] axw: what *did* you just do? [09:45] I have no idea. not enough sleep. I'll do it slowly this time [09:46] rogpeppe1: so, CI triggers on :shipit:? [09:46] axw: yup [09:46] axw: but please wait for a review first :-) [09:46] sure [09:46] rogpeppe1: https://github.com/juju/charmstore/pull/37 [09:47] rogpeppe1: and https://github.com/juju/charmstore/pull/38 for v4 === psivaa is now known as psivaa-bbib [10:00] morning [10:01] perrito666: morning [10:09] trivial PR for juju/testing. review anyone? https://github.com/juju/testing/pull/25 [10:09] perrito666, mgz, TheMue: ^ [10:10] rogpeppe1: yup, taking a look [10:11] TheMue: ta [10:12] rogpeppe1: LGTM, it’s pretty small ;) [10:12] TheMue: yup :-) [10:13] TheMue: do you know if there's CI set up on juju sub-repos yet? [10:13] rogpeppe1: sorry, no [10:13] * TheMue still fights with IPv6 routing between containers on different nodes [10:17] rogpeppe1: not sure if you saw this... https://github.com/juju/charmstore/pull/38 [10:18] axw: ah, sorry, lost in a sea of reversions :-) [10:18] axw: looking [10:18] :) [10:19] axw: i thought i just approved a branch that looked just like that [10:19] axw: (that you pointed me at) [10:19] axw: ah, v4! [10:19] axw: gotcha [10:20] axw: LGTM [10:20] thought I'd be proactive and break both branches [10:20] thanks rogpeppe1 [10:20] axw: :) [10:20] sorry for that [10:20] axw: that's fine [10:20] axw: CI is not consistently applied to juju sub-repos... [10:20] axw: do you know if juju/testing is CI'd [10:20] ? [10:20] nope. coming soon, I think [10:20] mgz is working on it [10:22] axw: cool. i'll just push then. [10:23] Any Minute Now [10:38] axw: further update: please don't land that branch - frankban has fixed everything :-) [10:38] rogpeppe1: which branch? [10:38] rogpeppe1: the one to update dependencies.tsv? [10:38] axw: confusion - perhaps you have already landed it [10:39] rogpeppe1: the reversion already merged [10:39] axw: right, i *thought* so [10:39] frankban: ^ [10:39] frankban: do you want to do the changes to use gopkg.in/mgo then? [10:40] rogpeppe1: not for now, I just need to merge my branch, will merge trunk again [10:40] frankban: ok [10:41] axw: it's still up to you then :-) [10:42] rogpeppe1: I'll take a look tomorrow [10:42] axw: ta [10:57] axw: so, Revert "Revert "Revert "Update... what's tha changing exactly :P [11:01] !!!false [11:01] :) [11:14] TheMue: are you around now, or is this worker time? [11:18] jam: I’m here, tracing around a bit. ;) [11:18] TheMue: I have to go check on the dog, brb [11:18] jam: opening 1:1 hangout [11:20] axw: wallyworld fyi we have the charmstore stuff in ci at http://ci.jujugui.org:8080/ and it uses the lander. We'll look into why the branch/test didn't get picked up today [11:20] axw: wallyworld so no big green merge buttons in that please [11:20] rightio [11:20] rick_h__: yep sorry, I wasn't aware. rogpeppe1 pointed out after I broke it [11:20] axw: gotcha, ok cool. Still catching up this morning. Sorry for the dupe pointer. [11:21] nps [11:21] rick_h__: can you remove the hackers as a contributor to the branch then? [11:21] +1 [11:21] and add bots [11:21] mgz: we could, but our team's always been a trust everyone all permissions kind of team. [11:21] mgz: so understand the issue, it's a bit of a shift from normal work but we'll look into it [11:22] TheMue: omw [11:22] works well for people working on the charmstore regularly, not so much for people like me :) [11:22] e.g. all team members have ci access at same level, manager on leankit, etc. [11:22] axw: right, the pool is people is growing, culture shift probably on the way I guess. :( [11:22] I guess I have been conditioned a bit tho [11:22] rick_h__: pretty much it just removes the footgun [11:22] I see a button, I think there's no CI [11:23] understand, like I said, diff culture. === psivaa-bbib is now known as psivaa [11:51] * TheMue wants a command to see how often he rebooted his testing VMs today === Ursinha-afk is now known as Ursinha [13:21] TheMue: try: last === gsamfira1 is now known as gsamfira [13:22] gsamfira: oh, nice, thx. didn’t know that [13:23] you are welcome :) [13:24] gsamfira: and grep and wc are my friends too ;) [13:26] oh, of course. these are a sys admins best friend, along side lsof, sed, awk and bash one liners :D [13:29] strace. never forgt. [13:29] I was just tinking about this [13:29] :)) [13:56] ericsnow: seen nate? [13:56] perrito666: nope (I just got on) [13:57] sinzui: are you on the habit of making trick PRs now? :) [13:58] No. Just frustrated. I may not merge that PR since I don't know what will happen this week with 1.20.2 [13:59] sinzui: which reminds me [14:00] * perrito666 goes to check restore CI [14:00] * perrito666 raises an eyebrow [14:00] is that running in hp or ec2? [14:01] I think it is running in Hp because ec2 was failing to provide instances when we needed them [14:01] sinzui: I see a great deal of failure bootstraping :| [14:02] aws is healthy right now [14:04] sinzui: I see roughtly 50% of the tests failing and all before even beginning for what seems to be machine not coming up on bootstrap [14:04] perrito666, I reset the test. It will play on ec2 [14:05] perrito666, this test left a machine that we could delete last week. maybe it has happened again. [14:05] sinzui: fwiw I see no more apt related errors [14:06] yep [14:26] dmesg says my network adapters keep exiting with status 1 [14:26] by keep, I mean after reboot and after replugging my USB Ethernet adapter [14:29] natefinch: ...? [14:29] your eth is usb? [14:29] lsusb ? [14:30] natefinch: if you pastebin the dmesg I might be able to understand what exit means for an eth [14:30] also, might help to get the ethernet card type and loaded kernel modules :) [14:30] that'll be tricky [14:30] natefinch: ? [14:31] lsusb doesn't show it. but dmesg says it gets recognized and then it exits [14:32] natefinch: sudo? [14:32] pastebin is tricky with no internet [14:32] natefinch: I take you are not on that compuer [14:32] it should show up in lsusb, or at least lspci [14:32] I could sneaker net it to my wife's laptop I guess [14:32] I am on my tablet [14:32] if it does not, you can stop there :). [14:32] and order a new one [14:32] natefinch: use the pendrive luke [14:32] ...or a different USB port [14:41] it's not just the Ethernet, the built in WiFi does the same thing [14:42] natefinch: at this point I would recommend the following [14:42] boot with a live CD/USB [14:42] if that works backup your hd, fast [14:44] I have off site backup regardless, but yeah, annoying. [14:45] natefinch: anyway, what I meant is, if booting with live media works is your hd or at least the fs that is fried [14:46] https://github.com/juju/juju/pull/398 === Ursinha is now known as Ursinha-afk === ubuntu is now known as Guest8506 === Guest8506 is now known as natefinch [15:06] ug ok, so live disk works fine [15:06] natefinch: you have now 2 opts, fs corruption or hd breakage [15:06] I think its just a bad kernel update :) [15:07] I'm with gsamfira [15:07] gsamfira: well if it where so we should hear axw yell on the same direction soon [15:07] and ericsnow [15:07] :)) [15:07] natefinch: 14.04? [15:07] all three of us have the same laptop [15:07] yep [15:08] * ericsnow makes mental note to check HD [15:08] natefinch: well you can always chroot and install an older kernel [15:08] natefinch: keep in mind that mine is a newer model (with SD) [15:08] I don't have the standard HD... hada big SSD before I bought it [15:09] ok, someone's going to have to help me install the older kernel... please and thank you... [15:10] is there a way to just load up a backup from a couple days ago or something like that? I don't think I've done a kernel update recently [15:10] you still have the old kernel [15:10] after rebooting, keep shift pressed [15:10] you will get the grub menu [15:10] just select an alder version [15:10] :) [15:10] ok [15:10] and since you are there you can always do an fsck on your sdd [15:10] ssd [15:11] and if you don't already have trim, you should put that in a cronjob (considering you rarely reboot) :) [15:12] * perrito666 gets cold called by directv and his phone company with bad offers [15:12] man, these people get really pushy [15:12] :)) === Ursinha-afk is now known as Ursinha === urulama is now known as urulama-afk === Ursinha is now known as Ursinha-afk === ubuntu_ is now known as natefinch [15:48] well that was fruitless [15:48] I knew I hadn't updated the kernel in ages [15:48] sigh [15:49] the HD is only a year old, and it's a top of the line samsung... but certainly something is fubared === Ursinha-afk is now known as Ursinha === jw4 is now known as jcw4 [15:57] if anyone is willing, I would appreciate a review for: https://github.com/juju/juju/pull/189 [16:27] natefinch: hi! Is there any news on my issue? [16:32] hackedbellini: no, sorry. We had a couple other issues come in that I had to look at as well. I tried to write a fix for your problem, but had some problems getting it to work with the mongo driver we're using. It'll take a little more time. I might be able to help you tweak the database directly to let the upgrade work correctly. Unfortunately, right now I'm trying to fix my laptop as it crashed this morning and now h [16:32] as a ton of problems [16:34] natefinch: no problem :). It will be very nice if you can help me to tweak my database to make the upgrade work correctly! There's no problem that you don't have the time to do that right now, I can wait until you can. === Ursinha is now known as Ursinha-afk === urulama-afk is now known as urulama === Ursinha-afk is now known as Ursinha [18:44] thumper: hey do you have time to do a review on https://github.com/juju/juju/pull/398? [18:48] er i guess it's the middle of the night for him. who is the on-call reviewer? [18:49] katco: menn0 is the other one... probably not around either :) [18:49] jcw4: yeah... also out of NZ =/ [18:49] hmmmm [18:51] natefinch is fighting HD issues... perrito666 or axw ? [18:51] axw is out of australia [18:51] dang [18:51] perrito666: is in the western hemisphere :) just very south of me! [18:52] :) [18:54] katco: the code looks pretty and well thought out, but I don't know much about the subject matter :/ [18:55] jcw4: lol, well i at least appreciate the compliments. it was a team effort :) [18:55] :) === Ursinha is now known as Ursinha-afk [19:06] where is the event queue stored on the unit? is it in a data directory or in memory or? === Ursinha-afk is now known as Ursinha [19:17] marcoceppi, pretty sure we don't store anything on disk [19:17] natefinch: then where does the pending queues get stored? [19:17] marcoceppi, I sort of assumed we just snagged each one from mongo one at a time [19:17] natefinch: so it lives in mongo then? [19:18] marcoceppi, everything lives in mongo :) [19:18] marcoceppi, but I don't know for sure [19:18] I would look at the code, but my computer is currently hosed. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [22:01] sinzui: hey, can you remember how to deal with GhostRevisionsHaveRevNo errors when pushing a bzr branch? [22:01] * sinzui thinks [22:02] wallyworld, It is a staking order [22:02] sinzui: ok. katco did a go get in order to get trunk, then did a bzr branch and a commit, then pushed [22:03] so maybe bzr checkout lp:golxc would have been better [22:03] * katco mopes around looking guilty. [22:04] thumper, menn0: what is the 'url' field in jenv for? [22:04] although bzr merge lp:golxc said nothing to do [22:04] maybe thumper knows [22:04] wat? [22:04] * thumper otp now [22:04] wallyworld, katco The problem is usually the project changed the default branch, but other branches weren't restacked [22:05] katco, I think you need to run [22:05] bzr reconfigure --unstacked [22:05] bzr reconfigure --stacked-on:lp:golxc [22:05] juju destroy-environment local --force [22:05] ERROR cannot read environment info for "local": error unmarshalling "~/.juju/environments/local.jenv": YAML error: line 138: did not find expected key [22:06] line 138 is: url: "" [22:06] sinzui: i just checked lp and katco's branch is there with the changes she did [22:06] even though push failed [22:06] "failed" [22:07] I think lp:~rogpeppe/golxc/golxc was the original trunk, then lp:~juju/golxc/golxc became trunk [22:09] hmph, simply deleting that line allowed me to force destroy [22:11] sinzui: the mp worked without any restacking :-) so the push error wasn't fatal as such. thanks for help. [22:24] waigani: I've just had a quick look in to your url question [22:24] waigani: that message is complaining that the .jenv has "url" in it but that isn't a known field [22:25] waigani: if you look in environs/config/config.go you'll see that "url" isn't described there [22:25] waigani: so the question really is: how did "url" get in that .jenv file? [22:26] menn0: thanks. yes, indeed - that is what I'm trying to work out now. [22:33] waigani: looking through recent-ish commits for config.go I don't see the "url" ever being defined [22:33] s/ ever/ field ever/ [22:33] waigani: has the file been manually monkeyed with perhaps? [22:33] menn0: I haven't touched it - promise! [22:34] menn0: though I have been manually provisioning lxc containers and it seems to happen during that process ... [22:35] menn0: just trying to reproduce it now [22:37] waigani: ok. hope you figure it out. [22:38] menn0: I'm about to attempt reproducing the bug that I *think* killed my network config over the weekend. If I go offline and miss our standup - that is why! [22:38] waigani: ok :) [22:43] menn0: okay this is really interesting. I manually provisioned the lxc, then destroyed it via lxc-destroy (i.e. not via juju). Now the network menu on the top bar of my desktop has disappeared. [22:43] menn0: I'm guessing if I restart I will not be able to connect anymore [22:45] and he's gone ... [23:04] wallyworld: privisioners_test.go failed but it looks unrelated. any objections to merging? [23:04] katco: what was failure error? [23:04] error message didn't match; "Failed to get machine 1" vs "failed to get machine 0" [23:05] ah, one of those map ordering ones [23:05] yep, good to go [23:05] ahhh [23:05] ok pushing up now [23:05] thanks, don't forget 1.20 :-) [23:05] starting with that :) [23:12] https://github.com/juju/juju/pull/410 if you're interested in LGTM [23:35] wallyworld: landed in v1.20 [23:35] \o/ [23:35] thanks [23:35] wallyworld: testing trunk [23:35] wallyworld: no problem! [23:36] wallyworld, when you have a moment I could use a few minutes of your time [23:36] alexisb: sure, now? [23:36] hangout? [23:36] sure, same as our 1xq [23:36] 1x1