Skip to content

Weekly Meeting 2016 07 07

Jump to bottom Edit New page
Mark Ryan edited this page Jul 14, 2016 · 4 revisions

Agenda

##Minutes

#ciao-project Meeting

Meeting started by AmyLee7 at 16:02:17 UTC. The full logs are available at ciao-project/2016/ciao-project.2016-07-07-16.02.log.html .

Meeting summary

Meeting ended at 17:00:16 UTC.

Action Items

  • tcpepper will add 6/23 minutes manually since meeting bot broke/ disk space ran out
  • obedmr will add the full Clear RPMs spec in a packaging dir and on the ciao wiki
  • mrkz will help obedmr with RPM bits
  • markusry create an issue for the version proposal in github
  • leoswaldo will address feedback and rework issue 98
  • AmyLee7 add the API compatability testing and API gaps for Openstack compliance to our map day discussion
  • obedmr and leoswaldo will analyze and identify gaps for Ciao Openstack compatability for our map day 7/11
  • markusry will create an issue to track the ansible features/demo ideas

Action Items, by person

  • AmyLee7
    • AmyLee7 add the API compatability testing and API gaps for Openstack compliance to our map day discussion
  • leoswaldo
    • leoswaldo will address feedback and rework issue 98
    • obedmr and leoswaldo will analyze and identify gaps for Ciao Openstack compatability for our map day 7/11
  • markusry
    • markusry create an issue for the version proposal in github
    • markusry will create an issue to track the ansible features/demo ideas
  • mrkz
    • mrkz will help obedmr with RPM bits
  • obedmr
    • obedmr will add the full Clear RPMs spec in a packaging dir and on the ciao wiki
    • mrkz will help obedmr with RPM bits
    • obedmr and leoswaldo will analyze and identify gaps for Ciao Openstack compatability for our map day 7/11
  • tcpepper
    • tcpepper will add 6/23 minutes manually since meeting bot broke/ disk space ran out
  • UNASSIGNED
    • (none)

People Present (lines said)

  • markusry (42)
  • AmyLee7 (41)
  • tcpepper (34)
  • sameo_ (31)
  • obedmr (24)
  • albertom (14)
  • mrkz (13)
  • kristenc (12)
  • leoswaldo (4)
  • ciaomtgbot (2)
  • ciaoslackbridge (1)
  • _erick0zcr (1)
  • mcastelino (1)

Generated by MeetBot_ 0.1.4

.. _MeetBot: http://wiki.debian.org/MeetBot

###Full IRC Log

16:02:17 <AmyLee7> #startmeeting
16:02:17 <ciaomtgbot> Meeting started Thu Jul  7 16:02:17 2016 UTC.  The chair is AmyLee7. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:02:17 <ciaomtgbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:02:22 <AmyLee7> o/
16:02:26 <sameo_> o/
16:02:47 <tcpepper> o/
16:03:02 <AmyLee7> #link Agenda- https://github.com/01org/ciao/wiki/Weekly-Meeting-2016-07-07
16:03:23 <obedmr> o?
16:03:27 <mrkz> o/
16:03:28 <obedmr> o/
16:03:28 <leoswaldo> o/
16:03:36 <_erick0zcr> o/
16:03:38 <kristenc> o/
16:03:51 <AmyLee7> #topic opens
16:04:05 * tcpepper has none
16:04:13 <kristenc> I have one.
16:04:26 <mrkz> I do have just a request/question for team
16:04:38 <mrkz> can you please take a look on https://github.com/01org/ciao/pull/340 ?
16:04:53 <kristenc> actually, I changed my mind - I don't want to waste meeting time on it, I'll discuss later.
16:05:01 <tcpepper> oh I do have one: https://github.com/01org/ciao/wiki/Weekly-Meeting-2016-06-23  ...bot broke.  Should we copy/paste in from an individual's log?  or kristenc did bot capture anything?
16:05:12 <mrkz> the Idea of having makefile it's for easier packaging across distros
16:05:37 <kristenc> tcpepper, the bot was out of disk space, so no.
16:05:41 <tcpepper> mrkz: we should add a "packaging" subdir.  could put deb and rpm spec files in there for example
16:05:41 <markusry> o/
16:05:42 <AmyLee7> #help look at PR 340
16:06:00 <tcpepper> kristenc: ok...I'll see if I can munge my pidgin log into the wiki in a readable form
16:06:01 <mrkz> the issue I do have is that I have no idea why travis failed :/
16:06:15 <kristenc> tcpepper, slackbridge would have saved it there as well.
16:06:17 <sameo_> mrkz: I will look at it, but a Makefile is probably not the right answer.
16:06:24 <markusry> Maybe the make file in the top directory
16:06:25 <obedmr> I just have an update
16:06:31 <markusry> Maybe travis tries to run it
16:06:36 <sameo_> mrkz: As tcpepper said, if we want to help distros we should provide packaging metadata.
16:06:45 <mrkz> tcpepper, sameo_: correct, problem is to start tacking the problem :)
16:06:45 <AmyLee7> #action tcpepper will add 6/23 minutes manually since meeting bot broke/ disk space ran out
16:06:49 <obedmr> we have CIAO in ClearLinux documentation with Ansible
16:06:51 <mrkz> so any input is good
16:06:51 <obedmr> #link https://clearlinux.org/documentation/ciao-deploy.html
16:07:54 <sameo_> mrkz: What about putting Clear RPMs spec in a packaging dir ?
16:08:15 <sameo_> obedmr: Could we have that one in the ciao wiki ?
16:08:38 <mrkz> sameo_: I guess it would make more sense yeah
16:08:51 <obedmr> sameo_: sure, you mean, like a link to that doc or the full thing?
16:09:01 <sameo_> obedmr: The full thing.
16:10:19 <obedmr> sameo_: sure, will do that
16:10:35 <AmyLee7> #action obedmr will add the full Clear RPMs spec in a packaging dir and on the ciao wiki
16:10:42 <markusry> Does it create a keystone VM or does it assume keystone is already running?
16:10:56 <obedmr> ?
16:11:18 <albertom> markusry: installs keystone too
16:11:24 <markusry> Very fancy
16:11:25 <obedmr> AmyLee7: it's an action that mrkz and I will do
16:11:44 <AmyLee7> ok thank you!
16:11:50 <mrkz> AmyLee7: I'm the one looking at the RPM bits, but as obedmr said ^, team effort :)
16:12:05 <albertom> markusry: that is an example of a playbook, you can make your own to use an existing keystone with minimal effort (no coding)
16:12:17 <AmyLee7> #action mrkz will  help obedmr with RPM bits
16:12:23 <markusry> albertom:  Got it.  Very nice.  Does it work on Ubuntu as well?
16:12:51 <albertom> markusry: not yet but that is already on my queue
16:13:09 <albertom> ubuntu + binaries compiled from github master yayyy
16:13:26 <albertom> right now is just for clearlinux and the packaged binaries
16:13:38 <markusry> Okay thanks
16:13:57 <sameo_> #info albertom to extend the playbook to support Ubuntu
16:14:06 <AmyLee7> okay anymore opens?
16:14:19 <markusry> I had one about version numbers?
16:14:35 <markusry> Are we going to add version numbers to our binaries as part of the release process
16:14:44 <markusry> Or are we already doing this?
16:14:57 <kristenc> which binaries are you talking about?
16:15:06 <kristenc> we don't release binaries.
16:15:07 <markusry> ciao-launcher, ciao-controller
16:15:34 <obedmr> markusry: you mean something like ' ciao-launcher --version '
16:15:39 <markusry> Exactly
16:16:05 <markusry> Or maybe even report the version number back to the scheduler
16:16:24 <albertom> oh
16:16:30 <mrkz> +1 ^
16:16:34 <albertom> sounds cool
16:16:36 <tcpepper> ssntp has the means to do that
16:16:37 <markusry> I was just wondering about this yesterday when playing witht he clusters
16:16:38 <albertom> for upgrades
16:16:47 <albertom> where you have mixed versions during the process
16:16:50 <markusry> I realized I had no way of knowing what version of the binaries where installed
16:17:01 <sameo_> markusry: I think I have an issue opened for that.
16:17:08 <markusry> Ah perfect.
16:17:12 <albertom> cia-launcher --version ?
16:17:14 <albertom> :P
16:17:17 * sameo_ verifies
16:17:47 <sameo_> markusry: https://github.com/01org/ciao/issues/14
16:18:07 <tcpepper> also, a good distro packaging would have the NVR so you could, eg: "rpm -qf $GOBIN/ciao-scheduler" and should see something like "ciao-8-5.fc24.x86_64"
16:18:28 <markusry> Okay, and what about --version.  Should we do this too?
16:18:30 <sameo_> markusry: Versions are already carried through the SSNTP frames, we don't do anything with it yet.
16:18:55 <markusry> But the versions do not match the versions of our releases to they?
16:19:02 <markusry> do they?
16:19:04 <sameo_> markusry: Right now they don't.
16:19:07 <tcpepper> correct
16:19:43 <AmyLee7> #link https://github.com/01org/ciao/issues/14 issue is open for version checking
16:19:44 <ciaoslackbridge> <> Pssst! I didn’t unfurl <https://github.com/01org/ciao/issues/14> because it was already shared in this channel quite recently (within the last hour) and I didn’t want to clutter things up.
16:19:44 <tcpepper> I like the idea of plumbing the release tooling to plunk the "N" into the repo in a way that building causes a --version to work, but....
16:20:01 <markusry> Yes me too.
16:20:13 <tcpepper> how do we differentiate a build of "real" release N, versus a build of HEAD after N but before N+1?
16:20:15 <sameo_> tcpepper: Yes, it may be tricky with something we don't build.
16:20:37 <sameo_> tcpepper: N-g<reduced SHA1>
16:20:40 <tcpepper> we'd need "go install" and friends to understand this magic version constant and include it, else the commit-ish
16:21:35 <tcpepper> http://stackoverflow.com/questions/11354518/golang-application-auto-build-versioning
16:22:20 <sameo_> we'd need a Makefile then :)
16:22:33 <albertom> mrkz: ^
16:23:47 <sameo_> AmyLee7: We can continue that discussion after the meeting, if you want to move forward.
16:23:51 <tcpepper> yeah
16:23:55 <mrkz> yes please
16:23:59 <AmyLee7> ok
16:24:06 <sameo_> markusry: Could you please create an issue for the --version stuff ?
16:24:17 <AmyLee7> #topic bugs
16:24:43 <AmyLee7> #action markusry create an issue for the version proposal in github
16:24:46 <mcastelino> o/
16:24:48 <markusry> sameo_: Sure
16:25:00 <AmyLee7> #info No P1's this week
16:25:06 <tcpepper> and same P2's as last week
16:25:26 <AmyLee7> yes the only one I want to cover is #98
16:25:35 <AmyLee7> #link https://github.com/01org/ciao/issues/98
16:25:49 <AmyLee7> assigned to leoswaldo and tcpepper
16:26:01 <AmyLee7> the rest of the P2's are scheduled for Sprint 2
16:26:29 <leoswaldo> yes, I haven't had time to make the rework, I'll move  it up in my priority list to address the feedback
16:26:41 <AmyLee7> ok thank you leoswaldo
16:27:08 <AmyLee7> #action leoswaldo will address feedback and rework issue 98
16:27:35 <AmyLee7> #link P3's https://github.com/01org/ciao/issues?q=is0X0P+0open+is0X0P+0issue+label0X0P+0bug+label0X0P+0P3
16:28:01 <sameo_> leoswaldo: Could you please also address the comment in your pending PR ?
16:28:16 <leoswaldo> yes, that's my pending
16:28:19 <tcpepper> we swept through the P3's last week and tidied up, but yeah..mostly just lower prio
16:29:19 <AmyLee7> some have been in there for awhile, do we want to make a policy on P3's that if they are there for x amount of time they are re-assigned as P4's?
16:31:14 <AmyLee7> any thoughts?
16:31:31 <tcpepper> do we then close P4's after a while?
16:31:34 <sameo_> AmyLee7: That would not be an incentive to actually go and fix them :)
16:31:48 <tcpepper> it's a slippery slope imho that discourages recording issues and getting around to working on them
16:31:50 <sameo_> AmyLee7: imho we should keep them as P3.
16:32:03 <tcpepper> I don't have a problem with us having 2000 open P3's and P4's
16:32:18 <AmyLee7> ok lets keep as is for now
16:33:08 <AmyLee7> #topic OpenStack Big Tent
16:34:30 <AmyLee7> tcpepper: I think you had this topic
16:34:38 <markusry> Last week, tcpepper summarized the situation
16:34:45 <tcpepper> yeah.  I spewed a bunch of conversation starters into the log last week...
16:35:07 <tcpepper> anybody have any thoughts or opinions?  keep watching?  make a choice to deliberately go one way or the other?
16:35:17 <markusry> The minutes are here:
16:35:18 <markusry> https://github.com/01org/ciao/wiki/Weekly-Meeting-2016-06-30
16:36:05 * sameo_ reads
16:36:06 <AmyLee7> I think keep watching and pursuing compliance with API's
16:36:44 <obedmr> something like: "being compatible instead of Big Tent" ?
16:36:56 <markusry> sounds good to me.
16:37:17 <tcpepper> we need to nail that down to a specific action plan...who's going to do what by when
16:37:19 <AmyLee7> for now I think that is the right approach, while keeping connected to Big tent conversations
16:37:34 <AmyLee7> ok - shall we discuss in our map day?
16:37:35 <sameo_> tcpepper: Have we figured out what "being part of openstack without going through big tent" means ?
16:37:42 <kristenc> i agree - seeking compatibility only for now.
16:37:54 <obedmr> it will be easier and flexible for letting other Cloud projects talk with CIAO
16:38:18 <tcpepper> API compatibility at a minimum.  It sounds like we could use their infra, given the discussions around golang and things being "in ecosystem" instead of "in BigTent"
16:38:33 <tcpepper> there's little apparent benefit though to us using their infra imho
16:38:54 <markusry> Me too.  It would just create a lot of extra work
16:38:58 <tcpepper> markusry: did you get any pings or signs of follow up on golang test support from the O.S. infra people?
16:39:01 <sameo_> Agreed.
16:39:06 <markusry> And make things harder for contributers as well
16:39:15 <markusry> tcpepper: No
16:39:24 <tcpepper> I'm guessing since they said no to go, the need for infra to support it in test became low prio and the pathfinding probably stopped at that, without implementation
16:39:26 <markusry> Let me check the etherpad
16:39:34 <tcpepper> if there was implementation, I'd like us to engage on it
16:40:01 <markusry> No, no updates
16:40:07 <tcpepper> we've learned a lot about go tooling these past months and have gotten a lot of benefit from it...would be good to share our learnings if there are folks open to collaboration
16:40:14 <tcpepper> ok so it's a wait a see there too
16:41:49 <tcpepper> so the main AR I see here is to define the api compat testing and our current api gaps
16:42:28 <sameo_> tcpepper: And which parts of the API we want to cover.
16:42:43 <AmyLee7> #action AmyLee7 add the API compatability testing and API gaps for Openstack compliance to our map day discussion
16:44:00 <kristenc> i was going to say "define compatibility" :)
16:44:03 <markusry> Are there any commonly used tools that sit ontop of this API, apart from horizon and rally?
16:44:36 <albertom> yes
16:44:37 <albertom> heat
16:44:37 <markusry> Or are we aiming at supporting custom scripts of people who've deployed open stack
16:44:50 <obedmr> Shade
16:44:53 <obedmr> and many more
16:45:08 <tcpepper> heat's the one we've talked about mostly, but people have mentioned some api compatibility test suite or certification process?
16:45:32 <markusry> So maybe if we aim at supporting some of these, this will help us decide what's really needed
16:45:38 <sameo_> yes, some OpenStack certification thingy.
16:46:12 <AmyLee7> http://www.openstack.org/brand/interop/
16:46:17 <AmyLee7> #link http://www.openstack.org/brand/interop/
16:47:29 <AmyLee7> does someone want to take the AR to review this and come to the Map day with what we need to do :)?
16:47:43 <tcpepper> obedmr: have you looked at this already?
16:47:50 <obedmr> sure
16:47:58 <leoswaldo> we can take it
16:48:18 <AmyLee7> obedmr and leoswaldo thank you!
16:48:27 <obedmr> we have been looking on Heat and Rally research, so, we have some stuff from there also
16:49:15 <sameo_> leoswaldo: Thanks.
16:50:10 <mrkz> obedmr: nice
16:50:48 <AmyLee7> #action obedmr and leoswaldo will analyze and identify gaps for Ciao Openstack compatability for our map day 7/11
16:51:48 <AmyLee7> ok anything else before we end the meeting?
16:52:02 <tcpepper> semi-related / open:  somebody noted we had a recent spike in github *'s...I was wondering if that might have been O.S. summit presentation proposal reviewers.  Do we have any sense of when we'd hear about Barcelona acceptance?
16:52:28 <sameo_> tcpepper: Let me double check.
16:52:35 <sameo_> tcpepper: Let me check.
16:52:36 <obedmr> how many submissions we have?
16:53:07 <kristenc> one really awesome one from our team :)
16:53:07 <sameo_> From the ciao team, one afaik.
16:53:40 <obedmr> I had one that wasn't accepted on last summit, I was planning to submit it again, it's about the OpenStack improvements journey we did in Clear Linux until Ciao
16:53:58 <obedmr> what do you think?
16:54:00 <kristenc> obedmr, I think you might have missed the deadline?
16:54:07 <sameo_> tcpepper: The call for proposal is open until 07/14.
16:54:09 <obedmr> it's 13th
16:54:21 <obedmr> right sameo_
16:54:27 <sameo_> obedmr: How long does the selection process take ?
16:54:50 <obedmr> dont remember, it's like 1 or 2 months
16:54:55 <obedmr> let me check
16:55:15 <kristenc> ah. well - I would personally not submit it again if it got rejected already. It's data it's based on just gets older and less relevant. that is my opinion though.
16:55:23 <AmyLee7> just chatted with Sabrina and OS summit folks said it will not be until August
16:56:37 <markusry> One question about the ansible stuff.
16:56:46 <markusry> Can you also use it to tear down the cluster?
16:57:01 <markusry> Return your cluster to a prestine pre-ciao state?
16:57:18 <markusry> I'm thinking about demos here.
16:57:35 <obedmr> markusry: +1 ^^ use Ansible for dev cluster
16:57:52 <markusry> We're you might want to install ciao over and over again on a cluster
16:57:56 <markusry> i.e., show how easy it is
16:58:02 <markusry> Where I mean
16:58:14 <markusry> to lots of different people
16:58:38 <tcpepper> as long as we know what needs torn down, an ansible script can do it
16:59:06 <tcpepper> ansible isn't much different than those tmux scripts I made....effecitively a series of commands run across a set of machines via ssh's
16:59:13 <markusry> Does it make sense to create an issue for this?  WHere is the ansible stuff tracked?
16:59:53 <mrkz> markusry: https://github.com/clearlinux/clear-config-management/
17:00:05 <AmyLee7> #action markusry will create an issue to track the ansible features/demo ideas
17:00:12 <AmyLee7> ok I have to run everyone to another meeting
Add a custom footer

Meetings

Development

Architecture

Usage

HTTPS API

Clone this wiki locally