17:01:22 <igordcard> #startmeeting network_common_flow_classifier
17:01:23 <openstack> Meeting started Tue Feb 14 17:01:22 2017 UTC and is due to finish in 60 minutes.  The chair is igordcard. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:01:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:01:26 <openstack> The meeting name has been set to 'network_common_flow_classifier'
17:01:46 <igordcard> hi davidsha, all
17:02:19 <davidsha> Hi
17:02:27 <igordcard> let's go straight into the agenda
17:02:34 <igordcard> #link https://wiki.openstack.org/wiki/Neutron/CommonFlowClassifier#Discussion_Topic_14_February_2017
17:02:48 <igordcard> #topic Approach A - PoC status
17:03:34 <igordcard> davidsha you've been working on an initial PoC for the ccf
17:03:44 <igordcard> davidsha: what's the status?
17:04:30 <davidsha> Ya, at the moment I've the service plugin working and I'm working on the resources and database models,
17:05:01 <igordcard> davidsha: is the service plugin exposing an API?
17:05:04 <davidsha> I've defined a resource for ipv4, ipv6, ethernet, tcp, udp and neutron ports/subnets and networks
17:05:24 <davidsha> igordcard: yes, to the previously mentioned resources
17:05:46 <davidsha> At the moment I'm only testing on ipv4 though.
17:05:48 <igordcard> great, those are a nice set of classification types to start with
17:06:16 <igordcard> for the neutron resources, is that a single type or three different types?
17:06:26 <davidsha> A single type
17:06:53 <davidsha> but you can specify source and destination too
17:07:05 <igordcard> cool
17:07:32 <igordcard> are you publishing the "initial commit" before the ptg?
17:08:15 <davidsha> I'm hoping to
17:08:43 <igordcard> davidsha: alright
17:09:07 <davidsha> I should for just IPV4, It would need more time for the classification groupings and other classifications that had been discussed.
17:09:10 <igordcard> are you looking at any neutron project to demo the consumption of classification resources from the ccf?
17:10:04 <igordcard> davidsha: sounds good.. I would say a subset of IPv4 attributes is enough for this first PoC
17:10:19 <davidsha> Yes, I was looking to try and use QoS dscp
17:10:53 <davidsha> kk
17:11:11 <igordcard> great
17:11:36 <igordcard> I will work on the spec this week
17:11:43 <igordcard> #link https://review.openstack.org/#/c/333993/10/specs/ocata/common-classifier.rst
17:11:44 <davidsha> ack
17:12:25 <igordcard> planning to trim it and stick to approach A (the one used in the PoC we are discussing)
17:12:34 <davidsha> +1
17:12:47 <igordcard> will also add your name to the assignees list
17:13:00 <davidsha> :D
17:13:45 <igordcard> and looking forward to exchange some information after the new patchset, to further align the spec with the PoC's technical details
17:14:43 <davidsha> +1, I'd have liked to break it into smaller patches but it all makes the service plugin.
17:14:50 <igordcard> I should continue to stay very busy with the sfc encapsulation work, but will find time to update and align the spec
17:15:08 <davidsha> each patch would not provide anything without the others.
17:15:18 <davidsha> ack
17:15:39 <igordcard> davidsha: could you maybe split the db resources and methods from the plugin's methods (just replace them with stubs at the most)?
17:16:53 <davidsha> igordcard: True, the database migrations don't necessarily have to be in the service plugin patch.
17:16:53 <igordcard> anyway, it's not like having smaller patches will make it easier to debug a regression or something like that
17:17:04 <davidsha> true
17:17:06 <igordcard> as this is a whole new thing
17:18:12 <igordcard> alright, this is all about the PoC for now
17:18:17 <igordcard> moving on...
17:18:25 <igordcard> #topic PTG information
17:19:04 <igordcard> myself, you, and sean-k-mooney are going to the PTG
17:19:12 <davidsha> yup
17:19:40 <igordcard> we all have context about the CCF and area hoping to give the community more information and discuss at the PTG
17:19:48 <igordcard> and are *
17:20:20 <igordcard> I've added the topic to the etherpad
17:20:28 <igordcard> #topic https://etherpad.openstack.org/p/neutron-ptg-pike
17:20:33 <davidsha> kk
17:20:34 <igordcard> oops
17:20:52 <davidsha> there is an undo action ;P
17:21:32 <igordcard> davidsha: yeah but I wonder if it's in general about the previous line.. or the previous topic
17:21:36 <igordcard> let's try
17:21:38 <igordcard> #undo
17:21:40 <openstack> Removing item from minutes: #topic https://etherpad.openstack.org/p/neutron-ptg-pike
17:21:47 <igordcard> oh cool
17:21:59 <davidsha> Lol
17:22:16 <davidsha> It's the last action the bot took.
17:22:52 <igordcard> I'm looking at the live log but it's not being very helpful
17:22:55 <igordcard> anyway, back to the point
17:23:11 <igordcard> I've added the CCF topic to the neutron pike ptg etherpad
17:23:16 <igordcard> #link https://etherpad.openstack.org/p/neutron-ptg-pike
17:23:45 <igordcard> when the PoC is published I'll add the link to it too (to the etherpad ccf bullet)
17:23:53 <davidsha> kk
17:24:11 <igordcard> anything else about the PTG?
17:24:42 <davidsha> No, I'm good.
17:24:48 <igordcard> alright, moving on...
17:24:51 <igordcard> #topic Open discussion
17:24:54 <igordcard> anything else we need to discuss?
17:26:03 <igordcard> I'm good too
17:26:21 <igordcard> alright davidsha, have a nice evening
17:26:28 <davidsha> igordcard: you too.
17:26:33 <igordcard> #endmeeting