[03:52] wallyworld babbageclunk: would you please take a look at https://github.com/juju/juju/pull/6884 [03:53] axw: sure [03:53] ok [03:57] axw: was the change to introduce cloud name as an attribute of cloud pre 2.0 release? just thinking about 2.0.0 client bootstrapping with a 2.2 agent [03:57] wallyworld: bootstrap always bootstraps the same version [03:57] wallyworld: the addition of name was post 2.0 anyway [03:57] unless you use --agent-version [03:58] or --auto-upgrade [03:58] wallyworld: that still doesn't bootstrap a different version, it upgrades post-bootstrap [03:58] hmmm yeah, that is true [04:04] axw: lgtm, ty [04:05] axw: are you looking at this by any chance? https://bugs.launchpad.net/autopilot-log-analyser/+bug/1655783 [04:05] Bug #1655783: 2.1beta4: LXDs getting stuck pending in cloud-init during Landscape Autopilot deployment [04:06] menn0: nope [04:06] axw: ok thanks [04:08] axw: reviewed [04:08] babbageclunk: thanks (and wallyworld) [04:45] wallyworld: I've responded to comments on https://github.com/juju/juju/pull/6878, PTAL [04:45] ok [04:48] axw: yeah, ignore me, i misread the code [04:49] wallyworld: cool [04:57] anastasiamac: axw: menn0: babbageclunk: are you guys ok to do standup an hour earlier tomorrow? uros wants to pop in [04:57] wallyworld: sure [04:57] wallyworld: ok, may not be very coherent tho [04:57] wallyworld: fine with me. I imagine it's going to be hardest for axw [04:57] yeah, a bit early for axw [04:57] less coherent than usual anyway :> [04:57] axw: yeah, it will be 11pm for him [04:57] just a one off [04:58] tim was supposed to email about it last week but is seems he forgot [05:28] jam: you around? could we catch up and discuss CMR when you are free? maybe in say an hour? [05:52] wallyworld: fine by me. altho it'll conflict with release call [06:01] shit happens [06:02] yeah, and apparently when u eat it, don't nibble [06:03] ewww [06:04] not my survivor tip [06:04] * anastasiamac shrugs [06:31] Bug #1658100 changed: Deployment of a large bundle fails or hoggs the system [06:38] Hi folks, just wondering if you saw https://bugs.launchpad.net/juju/+bug/1660087 yet? Looks like a slightly different twist on some of the bugs we've been seeing lately. [06:38] Bug #1660087: Controller and all models unresponsive [06:40] Bug #1658100 opened: Deployment of a large bundle fails or hoggs the system [06:42] blahdeblah: m getting to triage now :) [06:43] anastasiamac: ta - axw is gonna love me for this one. :-) [06:44] blahdeblah: how long did it take to get unresponsive? [06:44] blahdeblah: is it 2.0.x? [06:44] anastasiamac: I don't know, sorry; logs will hopefully show that [06:44] Yes, it's 2.0.2 [06:44] blahdeblah: there was a memory leak that I believe have been fixed in later 2.0.x [06:45] blahdeblah: i wonder if it's related... [06:45] I'd have to double check graphs to find out if memory was leaking, but I don't think so [06:45] Unless it was a leak in mongodb, because it was the top user of memory [06:46] blahdeblah: if memory serves right, it was mongodb-connections-related.. [06:47] Well, hopefully the profiling data gathered will give you something to go on; if not, please let us know and we can gather more. This is happening several times per week across our public cloud services - I've seen it on AWS, Azure, and GCE. [06:49] * anastasiamac afraid to ask... [06:49] blahdeblah: and what do u do when it happens? [06:49] blahdeblah: reboot? [06:49] Restart all the things! \o/ [06:49] anastasiamac: We stop jujud-machine-0, restart juju-db & rsyslog, and start jujud-machine-0 again [06:51] blahdeblah: i'll update the bug with this info. I ma sure that axw will be thrilled indeed :D [06:52] Bug #1658100 changed: Deployment of a large bundle fails or hoggs the system [06:52] Bug #1658549 changed: Security issue: jujud is not owned by a user on the system [06:52] Bug #1659102 changed: juju status shows ip address from public-api space rather than internal-api space [06:54] blahdeblah: are u seeing similar behavours with 2.1.x? [06:55] anastasiamac: I'm sorry to disappoint, but we don't deploy non-stable versions into our production cloud mirrors. :-) [06:57] blahdeblah: i admire our definition of stable \o/ [06:58] Bug #1658549 opened: Security issue: jujud is not owned by a user on the system [06:58] Bug #1659102 opened: juju status shows ip address from public-api space rather than internal-api space [07:00] wallyworld: I just got back from giving my wife a ride, I need a few minutes but then I'm happy to meeet [07:00] ok, sgtm [07:01] Bug #1658549 changed: Security issue: jujud is not owned by a user on the system [07:01] Bug #1659102 changed: juju status shows ip address from public-api space rather than internal-api space [07:15] jam wallyworld: what's the status of merging 2.1 into develop? I'm working on 2.1 now, and want to forward-port to develop... merge brings in a whole lot of unrelated conflicts [07:16] i tried this morning but it failed [07:16] i didn't look into it after that [07:16] axw: I have the branch that should land, it had a problem in Bundle tests that seem spurious [07:16] testing here it passes, so I'll resubmit [07:16] jam: okey dokey, thanks [07:16] wallyworld: nps, i'll wait for jam's branch to land [07:17] axw: I *could* just hit that "merge now" button... but that would be an abuse of power. :) [07:17] jam: heh :) going through CI seems safest to me [07:18] http://juju-ci.vapour.ws:8080/job/github-merge-juju/10137/ says that 'grant' failed as well, but I don't see anything other than maybe an EOF in that file [07:18] and that doesn't give me any real clue as to what is wrong. [07:58] jam: it merged, thank you [08:03] jam: i'm going to grab dinner, maybe I can ping you a bit later if you're free then? [08:03] sure [08:10] jam: review done [08:11] thanks menn0, will go through it now [08:14] jam: I found a lot of stuff but it's mostly minor === frankban|afk is now known as frankban === junaidal1 is now known as junaidali [09:00] jam: free when you are; finish whatever you're doing and give me a ping [09:01] k [09:08] jam: if at some stage you have a chance, I'd appreciate your review of https://github.com/juju/juju/pull/6887. I recommend reviewing the commits separately. they're in one PR because without the upgrade step the branch is unreleasable [09:17] sure axw [09:17] wallyworld: ping [09:18] hey [09:18] ho? [09:18] https://hangouts.google.com/hangouts/_/canonical.com/a-team-standup [09:19] wallyworld: that sounds like a scary place to go [09:19] I'm not sure I'm welcome there [09:20] huh? i'm the only on ehere, not that scary [09:20] we can go elsewhere [09:20] i just had a link jandy [09:20] *handy [11:44] anyone else seeing failures in "grant" testing where it seems to be expecting a password prompt and isn't seeing it? [11:44] the code even mentions bug #1621532 [11:44] Bug #1621532: grant-revoke could not enter password to login

[11:45] about macaroons [11:49] jam: nope, inquiere rogpeppe about that he might know what is going on [11:49] brb taxes pay day [11:50] jam: it might be because the command line behaviour has changed so the command-line mode is no longer the default (because it often fails) [12:48] rogpeppe: from what I can tell the issue is that the test is doing something like "juju login" and is expected to be prompted for a password, but it is getting no prompt [12:48] the related bug says that "when there are macaroons at play, you don't always get prompted" [12:53] axw: ping about 2.1 vs 2.2 on 'handleGetRegionError' stuff. [13:55] jam: congrats on getting the branch merged to 2.1 [13:55] rick_h: thanks. still needs one more thing before it is releasable, but Nick asked that it land rather than not. as it feels better to be testing the real thing. [13:55] jam: cool, still big chunk of stuff there so <3 [13:56] yeah, and I got the SSH-keyscan stuff landed, merging it to 2.2 now [13:56] got 2.1 synced into 2.2, so we can keep getting stuff in both [13:56] doing decent [13:57] still trying to sort out the real fix for bug #1657449 [13:57] Bug #1657449: 2.1 needs to fallback to lxdbr0 for clouds that can't be bridged [13:57] and things like vSphere and Manual both *wanting* to bridge-by-default [15:14] perrito666: i've just commented on https://github.com/juju/juju/pull/6865. I wonder what you think. [15:26] rogpeppe: looking [15:27] rogpeppe: mm, interesting comment [15:32] rogpeppe: what you say is true, I like your idea, thank you [15:32] perrito666: cool, thanks [17:03] morning === petevg is now known as petevg_noms === mskalka is now known as mskalka|afk === petevg_noms is now known as petevg === frankban is now known as frankban|afk === mskalka|afk is now known as mskalka [18:58] redir: sorry that was rude from my part, morning [18:59] nah [20:11] axw, you introduced a new launchpad dependency? :-( [20:17] d'oh, never mind, wrong branch :-) [22:14] so standup's still an hour early today, right? [22:15] babbageclunk: from calendar, it looks like it has been reverted to original time [22:18] anastasiamac: thanks, should've checked there first duh [22:18] babbageclunk: nps ;D === mskalka is now known as mskalka|afk [23:12] perrito666: since you're OCR today, mind taking a look at https://github.com/juju/juju/pull/6868? [23:13] babbageclunk: our definitions of today differ a bit but glad to [23:14] * perrito666 is running tests so he can't do anything of consequence in the laptop [23:14] oh, duh - forgot [23:14] perrito666: thanks! [23:17] babbageclunk: lots of dereferencing, lovely provider API I see :) [23:18] perrito666: Yeah, the rest API client autogenerator that the azure SDK uses makes everything a pointer. Blech. [23:22] jam: pong about handleGetRegionError (sorry, was EOD) [23:23] babbageclunk: ship it, brb [23:23] perrito666: awse, thanks