06:33:12 <yamamoto_> #startmeeting taas
06:33:12 <openstack> Meeting started Wed Jan 27 06:33:12 2016 UTC and is due to finish in 60 minutes.  The chair is yamamoto_. Information about MeetBot at http://wiki.debian.org/MeetBot.
06:33:13 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
06:33:15 <openstack> The meeting name has been set to 'taas'
06:33:26 <yamamoto_> #link https://wiki.openstack.org/wiki/Meetings/taas agenda
06:33:41 <yamamoto_> #topic spec
06:33:50 <yamamoto_> #link https://review.openstack.org/#/c/256210/
06:33:57 <yamamoto_> i updated the spec review
06:34:14 <yamamoto_> i think i applied all feedback.  sorry if i missed yours!
06:34:52 <yamamoto_> does anyone have anything particular about the spec to discuss?
06:35:10 <reedip> i think fawadkhaliq mentioned about Policy Based packet mirroring
06:35:11 <fawadkhaliq> yamamoto_: thanks for updating the spec
06:35:20 <fawadkhaliq> I just look a quick look
06:35:27 <fawadkhaliq> I am not sure if we covered that yet
06:35:45 <fawadkhaliq> so maybe we can discuss that and the incorporate in the spec.
06:36:00 <muawiakhan> #agreed
06:36:01 <fawadkhaliq> s/and the/and then/
06:36:14 <reedip> yamamoto_ : from a use case point of view, would policy based mirroring  be required ?
06:36:38 <fawadkhaliq> Since we are defining the API
06:36:44 <yamamoto_> do you have any idea how api for policy thing would look like?
06:36:45 <fawadkhaliq> I had the following thoughts:
06:37:14 <fawadkhaliq> We cover all the potential use cases and define the API to be able to support them.
06:37:20 <fawadkhaliq> The implementation can come in pieces.
06:37:51 <fawadkhaliq> Similar to how ipv6 works, not everything is implemented and expected to work but API definition is complete and can be consumed by different backends.
06:37:55 <reedip> fawadkhaliq +1 , but we need to formalize use-cases
06:38:02 <fawadkhaliq> reedip: Agree
06:38:10 <yamamoto_> my impression is it's better to wait neutron classifier stuff settle and try to leverage it, rather than trying to invent our own right now.
06:38:11 <fawadkhaliq> I can take a stab at adding patch to yamamoto_ spec
06:38:48 <fawadkhaliq> yamamoto_: can you please elaborate a bit about Neutron classifier support for everyone's benefit?
06:39:58 <reedip> Reference: https://github.com/openstack/neutron-classifier
06:40:34 <fawadkhaliq> #link https://github.com/openstack/neutron-classifier
06:40:39 <yamamoto_> my understanding is that it's trying to define something which can be shared among projects like fwaas sg etc
06:40:57 <fawadkhaliq> yamamoto_: yes, I think so too.
06:41:09 <fawadkhaliq> yamamoto_: so the classificaiton mechanism will be added as lib to neutron
06:41:16 <fawadkhaliq> yamamoto_: we will need the API definition
06:41:38 <fawadkhaliq> I am not sure how active this project is btw
06:41:46 <fawadkhaliq> does anyone know if this is being actively worked on?
06:42:45 <yamamoto_> i have no idea
06:42:58 <fawadkhaliq> I see this was last updated 2 months ago :(
06:43:03 <reedip> fawadkhaliq : I think this is on a slow burner ...
06:43:11 <fawadkhaliq> reedip: exactly
06:43:12 <reedip> might take some time
06:43:51 <fawadkhaliq> so maybe we move on and think about what's there today and make it happen with the existing primitives and if/when classifiers arrive, we can think about potential changes with an upgrade path
06:44:17 <muawiakhan> +1
06:44:21 <fawadkhaliq> reedip: yamamoto_ muawiakhan soichi: thoughts?
06:44:22 <reedip> Anyways, I think keeping dependancy on neutron classifier is more towards the API definition, which can be followed up once a list of definitive use-cases are finalized
06:44:47 <reedip> fawadkhaliq +1 on thinking about classifiers later
06:45:23 <soichi> fawadkhaliq: +1
06:45:37 <yamamoto_> sure, we can/should feed our requirements to neutron classifer.
06:45:59 <fawadkhaliq> reedip: soichi muawiakhan yamamoto_. perfect!
06:46:21 <fawadkhaliq> so next step should be to update the spec with use cases? I can help there.
06:46:50 <reedip> fawadkhaliq : +1
06:47:04 <reedip> I will also look into it, and if possible update the same.
06:47:45 <fawadkhaliq> okay cool. yamamoto_ you can assign me an action item on that, I will be happy to help.
06:48:01 <yamamoto_> #action fawadkhaliq update the spec with use cases
06:48:15 <reedip> :)
06:48:36 <fawadkhaliq> thanks yamamoto_ \o/ I like work :)
06:48:51 <muawiakhan> :)
06:49:11 <reedip> There are some ideas on the spec as well , like (1) Network monitoring, instead of port (2) Policy based monitoring (3) tenant based monitoring(?)
06:49:56 <fawadkhaliq> reedip: yes, nicely summarized in one line.
06:51:38 <fawadkhaliq> lots of silence in the room :-)
06:51:39 <reedip> anything else ?
06:51:42 <fawadkhaliq> *plays music*
06:51:48 <reedip> yeah, it tends to be a bit quiet :)
06:51:55 <yamamoto_> lets move on
06:52:04 <yamamoto_> #topic Open discussion
06:52:27 <reedip> L2 extnesion for taas is currently on hold, once vyadav comes back , will look into it
06:52:56 <fawadkhaliq> so from the governance perspective, can someone please summarize a bit whats the status of taas?
06:53:03 <fawadkhaliq> is it a subproject in neutron etc
06:53:29 <yamamoto_> it isn't neutron subproject.
06:53:34 <reedip> fawadkhaliq : Nope , its still under discussion
06:53:53 <reedip> #link https://review.openstack.org/#/c/229869/
06:53:58 <fawadkhaliq> yamamoto_: reedip thanks. what are your thoughts around it? I am assuming we have a proposal
06:54:02 <fawadkhaliq> oh perfect
06:54:15 <fawadkhaliq> I will take a look
06:54:16 <fawadkhaliq> thanks
06:54:19 <reedip> fawadkhaliq, its currently on hold, mainly because of the spec finalization
06:54:32 <fawadkhaliq> reedip: I see. thanks.
06:54:44 <yamamoto_> and "Evolving the stadium concept"
06:55:00 <fawadkhaliq> got it
06:55:10 <fawadkhaliq> thanks guys. that's all from my side.
06:55:49 <reedip> yamamoto_ and all, nothing from my side as well .
06:56:25 <yamamoto_> it seems we've done
06:56:31 <reedip> fawadkhaliq: the 4th comment from Armando on the governance commit is an important one, JFYI
06:56:47 <fawadkhaliq> reedip: thanks. Will take a look.
06:57:01 <yamamoto_> thank you for attending
06:57:08 <yamamoto_> #endmeeting