09:00:23 <anteaya> #startmeeting nova-net-to-neutron-migration
09:00:24 <openstack> Meeting started Tue Mar 10 09:00:23 2015 UTC and is due to finish in 60 minutes.  The chair is anteaya. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:00:25 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:00:27 <openstack> The meeting name has been set to 'nova_net_to_neutron_migration'
09:00:30 <anteaya> hello
09:01:02 <anteaya> obondarev mikal gus
09:01:16 <obondarev> o/
09:02:07 <obondarev> anteaya:hi, how are you?
09:02:18 <anteaya> obondarev: good thanks
09:02:21 <anteaya> and yourself?
09:02:39 <obondarev> me too, thank you :)
09:02:43 <anteaya> good
09:03:03 <anteaya> might just be you and I
09:03:18 <anteaya> mikal had told me he expected to be here for this meeting
09:03:19 <obondarev> looks so
09:03:29 <anteaya> perhaps he will be here before the end
09:03:32 <anteaya> so onward
09:03:42 <anteaya> #topic the state of the Neutron spec (obondarev)
09:03:48 <anteaya> anything new?
09:03:53 <obondarev> nope
09:04:01 <anteaya> next then
09:04:09 <obondarev> ok
09:04:16 <anteaya> #topic the state of implementation (obondarev)
09:04:30 <obondarev> no changes on db patch which is expected as jlibosva is unavailable till April
09:04:39 <obondarev> as we agreed, the patch is ready for testing
09:04:46 <anteaya> great
09:04:49 <obondarev> I put a comment on the patch with the steps which allowed me to migrate data for FlatDHCP model
09:04:52 <gus> (hi)
09:05:36 <obondarev> nova-net proxy patch is ready for testing as well, except floating ips and security groups
09:05:53 <obondarev> #link https://review.openstack.org/#/c/150490/
09:05:56 <gus> obondarev: I saw that, that was a nice amount of detail.
09:06:11 <obondarev> so in fact the two patches may be tested together: migrating data first and enabling proxy mode on nova-api and nova-compute
09:06:20 <anteaya> okay
09:06:26 <anteaya> so nice work
09:06:29 <obondarev> I post a comment on the proxy patch with the steps for testing as well
09:06:36 <anteaya> great
09:06:39 <obondarev> next plans are to test more cases - add/remove fixed ips for vms
09:06:45 <anteaya> okay
09:06:56 <obondarev> the more important is to get feedback
09:07:31 <obondarev> however I didn't ask dansmith to review is since we retarget it for Liberty
09:07:40 <anteaya> obondarev: yes
09:07:44 <obondarev> and nova folks may probably have more important reviews at this stage of the cycle
09:07:57 <anteaya> well I had a long discussion with dansmith last week
09:08:05 <obondarev> so I hope he'll take a look when has time for it
09:08:24 <obondarev> anteaya: interesting
09:08:33 <anteaya> and though you still want to pursue this direction, he indicated to me that he doesn't support it
09:08:42 <anteaya> and I had talked to mikal afterwards
09:08:48 <anteaya> who I had hoped would be here
09:09:12 <anteaya> but we have to proceed with what we have
09:09:23 <anteaya> and you have done a lot of work and have something to test
09:09:24 <anteaya> well done
09:09:29 <obondarev> anteaya: I'd ask Dan to post the comment with his thoughts and concerns on the review
09:09:58 <obondarev> as now I'm unsure what exactly he doesn't like
09:10:08 <anteaya> so do you want ot get some feedback first or keep working on it
09:10:15 <anteaya> the whole approach
09:10:27 <obondarev> yep
09:10:29 <anteaya> was what I heard him say
09:10:40 <anteaya> so since we don't have mikal
09:10:44 <anteaya> I leave the choice to you
09:10:49 <anteaya> what do you want ot do now?
09:11:05 <obondarev> well, I remember Dan was concerned that the approach will not work for complex cases
09:11:12 <obondarev> like allocate_for_instance
09:11:15 <anteaya> oh you heard that
09:11:18 <anteaya> oh good
09:11:38 <obondarev> now the patch allows is
09:11:42 <obondarev> it*
09:12:34 <obondarev> so..
09:13:00 <obondarev> I'd really like to know what problems do folks see with the approach
09:13:18 <anteaya> well it sounds like you have made a decision
09:13:22 <anteaya> go get some feedback
09:13:47 <anteaya> and I just got an email from mikal, he is at a conference and won't be attending this week
09:13:52 <anteaya> but it was nice of him to email
09:13:53 <obondarev> yes, because I'm probably missing something
09:14:12 <anteaya> okay so you have a way forward for next week
09:14:20 <anteaya> anything else under this topic?
09:14:33 <obondarev> sorry, which way do you mean?
09:14:42 <obondarev> ask Dan for feedback?
09:14:45 <anteaya> getting feedback on your work thus far
09:14:47 <anteaya> yes
09:14:56 <obondarev> anteaya: ok, thanks
09:14:59 <anteaya> great
09:15:06 <anteaya> okay to move to the next topic?
09:15:13 <obondarev> yep
09:15:14 <anteaya> #topic documentation
09:15:23 <anteaya> I have nothing new here
09:15:44 <anteaya> the patch hasn't changed
09:15:54 <anteaya> next
09:15:58 <anteaya> #topic testing
09:16:17 <anteaya> so can we talk about testing the patches we have?
09:16:37 <obondarev> we can
09:16:39 <anteaya> you have been successful with a flat network setup obondarev?
09:16:58 <obondarev> flatdhcp actually
09:17:16 <anteaya> awesome
09:17:28 <anteaya> so I think we should get some addition testers
09:17:30 <obondarev> had to disable floating ips migration though
09:17:40 <obondarev> as it's not ready
09:17:42 <anteaya> gus: are you in any position to be able to test the current patches?
09:17:47 <obondarev> in both patches
09:18:27 <gus> I can run through a basic devstack scenario, but it's probably a repeat of what obondarev has already performed.
09:18:31 <anteaya> obondarev: is the fact you had to disable floating ips migration in your notes on the patches?
09:18:42 <gus> I don't have any larger/different setups that I can test unfortunately.
09:18:43 <anteaya> gus: I see value in that
09:19:07 <anteaya> I'm fine if we can confirm obondarev's findings
09:19:13 <obondarev> anteaya: yes it is, in the comment on the db migration patch
09:19:41 <gus> anteaya: sure, I'll give it a go.
09:19:56 <anteaya> obondarev: the proxy patch doesn't mention the db migration patch in the commit message
09:20:00 <gus> (It won't be until next week however - I'm out helping high school kids with robots for the rest of this week)
09:20:00 <anteaya> do you think it should?
09:20:09 <anteaya> gus: thank you
09:20:20 <anteaya> gus: oh cool, that sounds like a lot of fun
09:20:27 <anteaya> gus: how are they doing with them/
09:20:29 <anteaya> ?
09:20:29 <obondarev> anteaya: actually no, it is not dependent
09:21:19 <anteaya> obondarev: I didn't say dependent, but if you don't have a link to the db migration patch in the notes on the proxy patch, how should a reviewer or potential tester find it?
09:21:34 <anteaya> let's make it easy for people
09:21:47 <obondarev> anteaya: so testing proxy may be independent from testing db migration
09:22:04 <anteaya> you just said you tested them together
09:22:17 <obondarev> yeah, but that's not necessary
09:22:18 <anteaya> so testing them together is a scenario, is it not?
09:22:29 <anteaya> I am fine with it not being necessary
09:23:00 <anteaya> but I would like to make it possible for people who would like to, to find the db migration patch from the proxy patch
09:23:04 <obondarev> ok, I can add a note to the commit message that it can be tested together with db migration
09:23:11 <anteaya> wonderful thank you
09:23:20 <obondarev> anteaya: sure, np
09:23:41 <anteaya> anything more on testing this week?
09:24:05 <anteaya> moving on
09:24:11 <anteaya> #topic open discussion
09:24:25 <anteaya> so sdague and others are at the ops meetup
09:24:35 <anteaya> apparently there are about 180 people there
09:24:39 <anteaya> they had a good first day
09:24:49 <anteaya> the migration didn't come up yet
09:25:04 <anteaya> but about 75% of the people in attendance are already running neutron
09:25:17 <anteaya> #link https://etherpad.openstack.org/p/PHL-ops-tags
09:25:42 <anteaya> this is based upon someone asking the question what service are they running and then guesstimating the show of hands
09:25:57 <obondarev> would be nice to know how many of those who're using nova net is still interested in neutron migration
09:26:02 <anteaya> I found the etherpad informative
09:26:18 <anteaya> agreed
09:26:46 <anteaya> though we can't take the folks who show up to the meetup in the states as definitive of what our users need
09:27:00 <obondarev> agree
09:27:10 <anteaya> as we know some folks in some countries can't get to the states
09:27:24 <anteaya> but yes, as much info as possible would be good
09:27:45 <anteaya> I trust sdague will do his best to gather as much info as he can from the group on the topic
09:27:58 <anteaya> I wouldn't have any ability to contact him now anyway
09:28:21 <anteaya> I could email but couldn't guarentee he would read it before he presents
09:28:31 <anteaya> anything more for open discussion?
09:28:56 <anteaya> I'll go with no
09:29:04 <anteaya> thanks for all your hard work obondarev
09:29:09 <anteaya> it is awesome
09:29:16 <anteaya> and thanks gus
09:29:17 <obondarev> anteaya: thanks
09:29:21 <anteaya> for just being you
09:29:26 <anteaya> obondarev: :)
09:29:31 <anteaya> happy roboting
09:29:37 <anteaya> see you next week!
09:29:40 <gus> yes, thanks obondarev
09:29:46 * gus waves
09:29:48 <anteaya> #endmeeting