[14:03] <Odd_Bloke> smoser: claudiupopa: harlowja: Any takers for this cloud-init 2.0 call?
[14:04] <smoser> Odd_Bloke, i'm in a library. lets do it here.
[14:04] <Odd_Bloke> smoser: Ack.
[14:04] <smoser> and claudiupopa agrees.
[14:05] <Odd_Bloke> So I was sprinting last week, and therefore didn't get any cloud-init work done.
[14:05] <claudiupopa> hey folks.
[14:05] <Odd_Bloke> I did some reviews Monday morning, then was off yesterday.
[14:05] <Odd_Bloke> I'm hoping to spend Thursday/Friday on cloud-init 2.0 stuff but a week spent, effectively, off last week means I have some normal work stuff that might overwhelm that.
[14:06] <smoser> ok, Odd_Bloke.
[14:06] <smoser> I will be in tomorrow (likely) but out friday as that is US holiday.
[14:07] <smoser> lets try this...
[14:07] <smoser> we'll look at http://bit.ly/cloudinit-reviews and try to quickly go through what is there.
[14:07] <smoser> and then we can talk about other things that shoudl be done.
[14:07] <smoser> sound ok ?
[14:07] <Odd_Bloke> Sounds good to me.
[14:08] <claudiupopa> Sounds perfect.
[14:09] <smoser> harlowja, sleeping still ?
[14:09] <Odd_Bloke> harlowja: WAKE UP ITS A BEAUTIFUL MORNING
[14:10] <smoser> west coast is lazy and often uses the excuse of the sun not rising on them yet.
[14:10] <smoser> ok. so i guesi 'll just go top down in my view
[14:10] <smoser>  https://review.openstack.org/#/c/195800/
[14:11] <claudiupopa> yeah, that looks fine excepting the overuse of properties.
[14:11] <claudiupopa> As both me and Daniel observed.
[14:12] <smoser> we want one for the decoded_buffer, right ?
[14:12] <smoser> as thats the whole intent of this
[14:13] <Odd_Bloke> Yeah, just not the others.
[14:13] <smoser> ok. so i'll just do that quick after this meeting
[14:15] <smoser> next.
[14:15] <smoser> https://review.openstack.org/#/c/170257/
[14:16] <smoser> i'll look over Odd_Bloke's responses after this meeting
[14:16]  * smoser didn't know about compiled regex.
[14:16] <smoser> caching
[14:17] <Odd_Bloke> smoser: http://ak-hdl.buzzfed.com/static/2015-02/1/20/enhanced/webdr02/anigif_enhanced-buzz-20392-1422840785-34.gif
[14:18] <smoser> :)
[14:18] <smoser> https://review.openstack.org/170252
[14:19] <smoser> i think we want to support rhel6. so if we can get by with 2.6 without massive pain, then we shoudl do that.
[14:20] <Odd_Bloke> I'm +1 on that change.
[14:20] <smoser> wish that harlowja would not have tagged along the logging change :)
[14:20] <Odd_Bloke> Do we think that cloud-init 2.0 is actually going to be used in RHEL6?
[14:20] <smoser> in its life time i ope so.
[14:21] <smoser> from epel i'd guess.
[14:21] <smoser> we can discuss dropping it if it becomes painful.
[14:21] <smoser> reasonable ?
[14:22] <Odd_Bloke> Yeah, definitely.
[14:22] <claudiupopa> yeah, sounds reasonable enough.
[14:22] <smoser> then we just have Odd_Bloke's https://review.openstack.org/#/c/191144/
[14:23] <smoser> i dont really want to pick up a dependency (especially one not pakcaged in ubuntu)
[14:23] <smoser> (wrt mini-lib and taskflow.types.notifier.Notifier )
[14:23] <Odd_Bloke> Yeah, I think harlowja or claudiupopa and I discussed not picking up a dependency until we thought we needed it.
[14:23] <Odd_Bloke> s/discussed/discussed on IRC/
[14:26] <smoser> Odd_Bloke, one thing i do see is that i report_event does not take a status code
[14:26] <smoser> for exit
[14:26] <smoser> pass/fail
[14:26] <Odd_Bloke> smoser: Yeah, I'm going to flesh out different event types next.
[14:27] <Odd_Bloke> Ah, claudiupopa and I agreed that it wasn't worth pulling in a library for our limited use case.
[14:27] <Odd_Bloke> harlowja just ignored me. ;.;
[14:27] <smoser> ok. well lets go with that then, but i know we need status / exit code
[14:28] <smoser> last is https://review.openstack.org/#/c/195631/
[14:29] <smoser> which i'd just pull other than Odd_Bloke wanted me to use tox -l, which makes sense in some contexts but not in others.
[14:29] <smoser> unless i was going to do a 'tox --recreate --notests' or something to create it if it was there and didnt exist
[14:29] <smoser> i just use that for thing slike:
[14:29] <Odd_Bloke> I was wondering if there's a way to get tox to just create the virtualenv, without running the commands configured.
[14:30] <smoser>  tox-venv py27 nosetests %
[14:30] <smoser> there is.
[14:30] <smoser> --recreate --notests
[14:31] <smoser> or something to that effect
[14:31] <smoser> yeah.
[14:31] <smoser> the recreate is just confusing. it will create if does not exist.
[14:32] <smoser> if you want to get that, we can.
[14:32] <smoser> so.. we can come back to that.t
[14:33] <smoser> the other 2 things ithat i have is
[14:33] <smoser> a.) need of a main / runner... so we can actually run the stuff
[14:33] <smoser> b.) i was looking some at trying to make sure we can easily test on trusty using packaged versions.
[14:34] <smoser> because as we add pip things, its very easy to not realize that you dont work on older pip versions of things.
[14:35] <Odd_Bloke> (In another meeting now, but it seems like I'll be done with it pretty soon)
[14:35] <claudiupopa> +1 for the main / runner.
[14:35] <smoser> ok. claudiupopa do you ahve anything else that you think needs to be on short list ?
[14:36] <smoser> i'm definitely ok with prioritizing the main over the 'b' above, but i'd love to be able to do 2 things:
[14:36] <smoser>  1.) run "get-dependencies --with-build" : to get dependencies installed via packages on trusty or other ubuntu (and other dists too, rather than using pip)
[14:37] <smoser>  2.) use tox to provide me with pip at versions on trusty.
[14:37] <smoser>   ie, so on my 15.10 laptop i could do: tox -e trusty-py34
[14:37] <smoser> or :
[14:38] <smoser>  tox -e rhel6
[14:38] <smoser> which would then use versions of pip that you'd find there.
[14:38] <smoser> what that means i hink si that we have to have data that includes
[14:39] <smoser>  pip package name, distro_python26_name, distro_python26_version, distro_python34_version and such.
[14:39] <smoser> which quite admittedly is a pita to maintain
[14:39] <smoser> but i think worth it.
[14:40] <smoser> ok, well, since i'm the only one talking, then i'll get to trying to review those things  and a main
[14:40] <smoser> claudiupopa, if you have some cycles, what would you / could you work on ?
[14:50] <Odd_Bloke> smoser: You're thinking that 'tox -e rhel6' would run the tests in a RHEL6-like environment?
[14:51] <smoser> yes. and 'install-dependencies' would read/determine that i'm on rhel6 yum install any deps
[17:24] <openstackgerrit> Scott Moser proposed stackforge/cloud-init: pip2distro: work in progress: build tox env for specific targets  https://review.openstack.org/197671
[17:28] <harlowja> what u guys talking about, ha
[17:30] <smoser> harlowja, welcome.
[17:30] <harlowja> sup
[17:30] <harlowja> my damn irc client not setting me away
[17:30] <harlowja> need to figure that out
[17:30] <smoser> glad you decided to roll out of bed this morning<cough>afternoon
[17:30] <harlowja> lol
[17:30] <harlowja> anytime man
[17:30] <harlowja> got up just for u smoser
[17:32] <harlowja> u guys get up to early, lol
[17:33] <harlowja> sleep in, get your rest, let your brain rest and all
[17:33] <harlowja> its the CA way :-P
[17:33] <smoser> :)
[17:34] <smoser> you had one thing that could get merged if you just dropped properties
[17:35] <smoser> https://review.openstack.org/#/c/195800/
[17:35] <harlowja> :-P
[17:35] <smoser> can you do that quick ? just to make happy ?
[17:35] <harlowja> i like making the happies
[17:36] <harlowja> happy happy
[17:36] <harlowja> in progress
[17:36] <smoser> and i will respond to templater ones
[17:37] <harlowja> happy happy
[17:38] <openstackgerrit> Joshua Harlow proposed stackforge/cloud-init: Expose api response properties and cache buffer decoding  https://review.openstack.org/195800
[17:38] <harlowja> ^^ happy
[18:04] <smoser> harlowja, thanks
[18:04] <harlowja> np
[18:35] <smoser> harlowja, how did you make py26 fail ?
[18:35] <smoser> https://review.openstack.org/#/c/195800/
[18:35] <smoser> i saw you do that somewhere.
[18:35] <harlowja> oh man
[18:35] <harlowja> u killed 26
[18:35] <harlowja> lol
[18:36] <openstackgerrit> Scott Moser proposed stackforge/cloud-init: tools/tox-venv: support running other than ./tools/tox-venv  https://review.openstack.org/195631
[18:37] <harlowja> ya, hmmm, don't think there is a importlib on 2.6, lol
[18:37] <harlowja> we probably need https://pypi.python.org/pypi/importlib in the requirements for 26 then
[18:38] <smoser> http://stackoverflow.com/questions/9418064/conditionally-installing-importlib-on-python2-6
[18:38] <harlowja> ya, pbr now supports requirement markers so should probably just use those
[18:38] <harlowja> ie crap like https://github.com/openstack/requirements/blob/master/global-requirements.txt#L37
[18:39] <harlowja> so we probably want a importlib;python_version=='2.6'
[18:39] <smoser> wow. that i ugly.
[18:39] <smoser> :)
[18:39] <harlowja> :-P
[18:39] <harlowja> https://www.python.org/dev/peps/pep-0426/#environment-markers
[18:40] <smoser> tox calls pip, how does pip get pbr ?
[18:41] <harlowja> https://github.com/stackforge/cloud-init/blob/master/setup.py
[18:41] <smoser> i see. its the first in that list, and the pip like re-executes itself or something
[18:41] <harlowja> something like that
[18:41] <smoser> oh. ok.
[18:42] <openstackgerrit> Joshua Harlow proposed stackforge/cloud-init: Add importlib for python 2.6 *only*  https://review.openstack.org/197691
[18:42] <smoser> so you want to fix taht ?
[18:42] <smoser> there yagoo
[18:42] <harlowja> ha
[18:42] <harlowja> let's see if it works
[18:46] <smoser> looks like no dice ?
[18:48] <harlowja> hmm
[18:49] <smoser> maybe. i thought it ran and failed. but hasnt ran
[18:49] <harlowja> https://jenkins03.openstack.org/job/gate-cloud-init-python26/2/console
[18:49] <harlowja> its ran
[18:49] <harlowja> but didn't get installed :-/
[18:49] <smoser> whered that link come from ?
[18:49] <smoser> not run on https://review.openstack.org/#/c/197691/ yet
[18:49] <harlowja> http://status.openstack.org/zuul/
[18:49] <harlowja> search for cloud-init
[18:50] <harlowja> then click the box
[18:51] <harlowja> ^ gives u the jenkins links before they all finish
[18:52] <harlowja> i'll ask some pbr folks when that all finishes (with real links)
[18:52] <harlowja> unsure why it didn't get installed
[18:52] <smoser> ok. thanks.
[18:53] <harlowja> np
[19:15] <smoser> Odd_Bloke, still around ?
[19:16] <harlowja> so thats weird
[19:16] <harlowja> http://logs.openstack.org/91/197691/1/check/gate-cloud-init-python26/eae73ee/console.html shows different errors than what jenkins had, lol
[19:16] <harlowja> seems importlib did get installed
[19:16] <harlowja> weird
[19:16] <harlowja> ha
[19:37]  * harlowja fixing these tests up
[19:49] <openstackgerrit> Scott Moser proposed stackforge/cloud-init: Bring over the 'templater' from bzr  https://review.openstack.org/170257
[19:53] <smoser> harlowja, thanks.
[19:53] <harlowja> np
[19:53] <smoser> i just did ^
[19:53] <harlowja> it will prob still break until 26 resolved
[19:55] <openstackgerrit> Joshua Harlow proposed stackforge/cloud-init: Add importlib for python 2.6 *only*  https://review.openstack.org/197691
[19:55] <harlowja> ok let's see if that passes
[19:56] <harlowja> *passed local 2.6
[23:43] <openstackgerrit> Joshua Harlow proposed stackforge/cloud-init: Add importlib for python 2.6 *only*  https://review.openstack.org/197691