17:10:15 <matjazp> #startmeeting training_guides
17:10:17 <openstack> Meeting started Mon Jul  6 17:10:15 2015 UTC and is due to finish in 60 minutes.  The chair is matjazp. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:10:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:10:20 <openstack> The meeting name has been set to 'training_guides'
17:10:29 <matjazp> roll call
17:10:34 <rluethi> hi
17:11:09 * rluethi hears crickets chirping
17:11:31 <matjazp> I was a bit surprised, I thought you guys postponed the meeting time
17:11:45 <matjazp> pranav just said he can't make it
17:11:53 <rluethi> matjazp: I was surprised, too :-)
17:12:18 <rluethi> I knew there was talk of changing the slot, but I don't think anything has been agreed on.
17:12:27 <matjazp> you know if sayali will be comming?
17:12:34 <rluethi> no idea.
17:12:47 <rluethi> isn't she travelling?
17:12:53 <rluethi> or was that last week?
17:13:06 * rluethi shrugs
17:13:12 <matjazp> anyhow, lets just look at the agenda..
17:13:22 <matjazp> oh, a standard one :)
17:13:23 <rluethi> yeah, let's :-)
17:13:29 <matjazp> #topic reviews
17:14:05 <matjazp> pranav submitted a patch for tox that publishes training guides in the drafts
17:14:47 <rluethi> I didn't see that. Lemme check.
17:14:58 <matjazp> he still needs to polish it a bit, he has a typo and maybe he should also insert a associate guide folder
17:15:39 <matjazp> now you can just click in the gate and you can see the training guides slides
17:16:34 <rluethi> which gate?
17:17:07 <matjazp> ...publish.-checkbuild
17:17:23 <matjazp> #link http://docs-draft.openstack.org/35/197535/3/check/gate-training-guides-tox-doc-publish-checkbuild/3fd20d8//publish-docs/
17:17:25 <rluethi> sweet.
17:17:52 <matjazp> it will be easier to review content
17:18:09 <rluethi> not obvious for newcomers, but very cool for those who know the trick.
17:18:45 <matjazp> well it's the standard way of publishing drafts for the whole docs project
17:19:42 <rluethi> I always thought something like this should exist, but I didn't know that it did.
17:20:26 <matjazp> what do you think, do we need to add "associate-guide" or "associate" folder?
17:20:43 <matjazp> we will need it later, so maybe we can add it now
17:20:50 <rluethi> yes, I was about to add a comment agreeing with you.
17:21:20 <matjazp> later= when we add content for developer guide and operator guide
17:22:55 <matjazp> is there anything new about labs?
17:23:27 <rluethi> not really, we were busy celebrating the latest bug fix :-).
17:23:35 <rluethi> there was some planning for the future.
17:23:51 <rluethi> pranav and sayali worked on the python port.
17:24:18 <rluethi> I wanted to port my KVM backend to master, but it takes more time than I currently have.
17:24:45 <matjazp> I know taht we discussed this already a few times, but I stil think that maybe vagrant could be a good option... it has MANY providers already (KVM, vBox, OpenStack, etc.)
17:25:33 <matjazp> so you have KVM bash port and there's also a python/KVM labs port?
17:26:50 <rluethi> I hacked together a KVM backend for osbash, but it's only a PoC (i.e. breaks the virtualbox code) and needs work.
17:27:14 <rluethi> python is also at PoC stage.
17:27:41 <rluethi> the problem with vagrant is that it's a pretty substantial dependency.
17:27:53 <rluethi> something we've been trying to avoid.
17:28:04 <rluethi> so it would be an alternative, not a replacement.
17:28:11 <matjazp> rluethi: yes, it is, but it also solves some major pain point we have
17:28:18 <rluethi> that would mean we maintain yet another backend.
17:28:20 <matjazp> eg. different providers
17:30:03 <matjazp> I'm experimenting a bit with KVM backend and multinode devstack with vagrant
17:30:08 <rluethi> maybe. I am skeptical. I like vagrant, don't get me wrong, but right now osbash is much easier to use than setting up vagrant.
17:31:11 <matjazp> rluethi: for developer's guide labs, we will probably need devstack too, not just install form packages
17:31:41 <rluethi> ic
17:32:13 <rluethi> well, devstack already does that, doesn't it?
17:32:18 <matjazp> anyway, I just remembered about all that stuff we already discussed when I needed to reboot machine during provisioning
17:33:06 <matjazp> rluethi: yes, but it is much easier to install multinode devstack with vagrant
17:33:38 <rluethi> I'll take your word for it.
17:33:51 <matjazp> and for some things, you need multinode, allin-one is not enough (eg. showing scheduling, networking, etc
17:34:04 <rluethi> I wonder if we could leverage the devstack code instead of vagrant.
17:34:24 <rluethi> we do have a multinode setup, what we don't have is the install from git repos.
17:35:48 <matjazp> osbash that calls devstack directly? maybe
17:36:05 <matjazp> it would preserve osbash workflow
17:36:33 <rluethi> yes.
17:36:39 <matjazp> but back to reviews... thats it, right?
17:36:46 <rluethi> yes.
17:36:50 <rluethi> we're done.
17:37:17 <matjazp> #topic any other business
17:38:05 <matjazp> I'm just curious, would you prefer the meeting an hour or two later?
17:38:21 <rluethi> not really.
17:38:26 <rluethi> for me it's alright.
17:39:24 <matjazp> and others? Is the only problem Monday, or is it also the time slot?
17:39:25 <rluethi> i will let others decide and try to be there for the meeting if I can.
17:39:35 <rluethi> I dunno.
17:40:04 <rluethi> I was curious about that, too, and I expected to read more on the mailing list, but there was nothing. zilch. nada.
17:41:23 <matjazp> hmm... you have anything else?
17:41:49 <rluethi> nope. I'm good.
17:41:56 <rluethi> wrap up.
17:41:59 <matjazp> ok, so let's wrap this
17:42:04 <matjazp> #endmeeting