[01:43] blackboxsw: thank you. i'm not sure why that appears sometimes and is not crippling. === shardy is now known as shardy_lunch === shardy_lunch is now known as shardy === shardy is now known as shardy_mtg === shardy_mtg is now known as shardy [17:22] blackboxsw: https://bugs.launchpad.net/cloud-init/+bug/1742494 [17:22] Launchpad bug 1742494 in cloud-init "network reporters generate WARNING in local stage" [Medium,Confirmed] [17:22] ahh thx smoser [17:41] smoser: maas code claims it's adding apt configuration (even though they set preserve_sources=False) for earlier cloud-init versions...hmm [17:41] # Add APT configuration for new cloud-init (>= 0.7.7-17) [17:42] * blackboxsw checks how 0.7.7-17 behaved regarding apt.... maybe this is an issue on trusty for some reason? [17:56] the nwere apt format is more expressive [17:56] and allows them to say some thigns that they could not say previously [17:57] but curtin udnersands the new format [17:57] and writes it into the target for everything (i'm pretty sure) [20:56] landed https://code.launchpad.net/~smoser/cloud-init/+git/cloud-init/+merge/335108 [21:01] blackboxsw: if you are looking at merges, want to respond to smoser on https://code.launchpad.net/~powersj/cloud-init/+git/cloud-init/+merge/335774 [21:02] powersj: yeah, that was on his list. [21:02] ok :) [21:08] umm, https://code.launchpad.net/~powersj/cloud-init/+git/cloud-init/+merge/335774 so, I'm forgetting something here josh. I thought I had seen a changeset at some point that did still provide a mechanism by which we could install system packages via 'make ci-ubuntu-deps' [21:08] but looks like maybe we rebased the cii-requirements branch or something [21:08] read my latest comment [21:09] powersj: I've conveniently forgotten our conversation. and will read the latest comment [21:12] hrm.. yeah I guess I thought at the end of our conversation you were working up a solution to handle the outliers which couldn't be installed bit pypi or system packages. looks like I misunderstood/misremembered [21:13] powersj: you talked at standup about maybe pulling a commit rev or git hash on certain dependencies. [21:15] yeah I should update pylxd==2.2.4 to be something like git://github.com/lxc/pylxd.git@0722955260a6557e6d2ffde1896bfe0707bbca27 [21:15] were you thinking about that as 'futures'? we might be able to extend read-dependencies script to handle some of that lookup for us. /me digs back into context here to look for shortsighted/quick-wins o [21:15] because I'm all about the short-sighted solutions [21:16] and yes I was thinking about doing something special for integration tests, but it just didn't make sense because 1) it is only ubuntu and 2) I would need some mix of system packages + pypi packages [21:16] and I don't think I want to go installing pypi packages on anyone's system [21:19] powersj: what if we left that packages/pkg-deps.json rename and the bzr+lp system package translation in tools/read-dependencies... then at least our "make ci-ubunt-deps" could still additionally call "read-dependencies --requirements-file integration-requirements.txt" [21:19] ... with the exception of pylxd right? [21:20] or you feel that's muddying the waters too much in read-dependencies script? [21:21] yeah trying to remember why we have all these exceptions seems like asking for a nightmare [21:21] don't get me wrong though, having a install-deps for integration tests would be nice, and I would like it a lot, especially since I want to get two new jenkins systems up, I don't want to have to worry about deps for everything [21:25] i was just fine to push 'install c-i requiremenst' to a later date. [21:25] * smoser has to run [21:27] yeah that's fine there too. will land powersj as is. we can bikeshed on making make ci-deps-ubuntu work for integration tests in a subsequent branch [21:45] powersj: landed, https://code.launchpad.net/~powersj/cloud-init/+git/cloud-init/+merge/335774. Thanks for the rehash. We'll go in circles about this again later :) [21:46] blackboxsw: thanks :) and yes we will [21:47] powersj: did you say pylxd was snapped/ [21:47] blackboxsw: no, lxd is [21:47] ahh ok. gotcha [21:48] and to use the lxd snap we needed newer version of pylxd [21:49] which we can only easily get from here right? https://github.com/lxc/pylxd [22:05] blackboxsw: correct [23:19] powersj: tox env change for pylxd? http://paste.ubuntu.com/26363025/ [23:19] I can put up a branch unless you already have that [23:19] https://www.irccloud.com/pastebin/uIisHdgJ/ [23:20] just tested with: tox -r -e citest -- run --os-name=artful --platform=ec2 --preserve-data --data-dir=../results --verbose -t modules/runcmd [23:20] sorry post that irccloud got in the way [23:21] 44.9 78.1 47:21.26 chrome [23:21] geez man. [23:21] lol [23:21] 78 % MEM for chrome [23:21] my machine is almost locking up [23:22] #toomanytabs [23:22] I found chrome recently was taking all my swap too [23:22] I was out of memory, locking up [23:22] anyway... trying your patch, but I think that works [23:24] yeah that looks good, can you change the comment to https://paste.ubuntu.com/26363038/ [23:24] blackboxsw: that way I know why and can update as necessary [23:36] funny I had realized that after I posted the pastebin.... I had # lxd backend master tp from 01/10/2018 ..... I like yours better [23:36] pushing a silly branch for this, just so we all get karma (I mean so it can be tracked) [23:44] https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/335963 powersj [23:45] blackboxsw: thx! [23:46] +1ed [23:46] thx [23:46] will land it [23:46] fastest branch ever [23:46] if only I'd review your branches that quickly [23:47] * blackboxsw waits for ci [23:47] just because [23:50] :)