13:59:53 <sc68cal> #startmeeting neutron_ipv6
13:59:54 <openstack> Meeting started Tue Jul 22 13:59:53 2014 UTC and is due to finish in 60 minutes.  The chair is sc68cal. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:59:55 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:59:58 <openstack> The meeting name has been set to 'neutron_ipv6'
14:00:29 <xuhanp> hello
14:00:45 <HenryG> Hi xuhanp
14:00:51 <baoli_> Hi
14:00:57 <dane_leblanc> hello
14:02:02 <sc68cal> I think our agenda will be fairly quick today, since the J-2 milestone is this week
14:02:46 <xuhanp> sc68cal, any luck about the dhcpv6 stateful blueprint?  I know the approve date has been passed.
14:03:47 <sc68cal> xuhanp: your spec hasn't been -2'd
14:04:18 <xuhanp> sc68cal, good to know. So still possible to be approved?
14:04:24 <HenryG> The only way to get a BP approved now is to send an email to openstack-dev and ask for a Spec Freeze Exception
14:04:50 <xuhanp> HenryG, I didn't know that. I can certainly do that.
14:05:32 <sc68cal> I agree- this should be an exception, since the code is also in gerrit and ready for review
14:05:59 <xuhanp> by the way, sc68cal, we got the reply from dnsmasq developer about stateless dhcp. I will update the code soon
14:06:38 <HenryG> xuhanp: It's important to check the Neutron Meeting weekly. It's recorded in eavesdrop.
14:06:57 <xuhanp> HenryG, will do. Thanks!
14:07:03 <baoli_> xuhanp, can you let us know that's about the stateless dhcp?
14:07:13 <baoli_> s/that/what
14:07:48 <xuhanp> baoli_, we found problem to let dnsmasq only takes care of optional information instead of sending out address in stateless mode.
14:08:17 <xuhanp> We didn't know how to configure dnsmasq in the past and asked dnsmasq developers by email.
14:08:20 <baoli_> xuhanp, I see, thanks.
14:12:29 <sc68cal> xuhanp: did they reply on the mailing list or directly to you?
14:13:01 <xuhanp> sc68cal, they replied the guy in my team who sent out the email at the beginning.
14:13:26 <xuhanp> We thought they won't reply so I asked your help. But it turned out they replied.
14:13:38 <sc68cal> xuhanp: ah - that's good :)
14:13:45 <xuhanp> :-)
14:15:06 <HenryG> Is the code review for stateless on gerrit yet?
14:15:31 <xuhanp> I am working on the unit test. But should be on in these two days.
14:15:52 <HenryG> xuhanp: No problem, I thought maybe I missed it.
14:16:02 <sc68cal> xuhanp: https://review.openstack.org/106299 ?
14:16:22 <xuhanp> sc68cal, yes.
14:16:50 <HenryG> That's the stateful one.
14:16:50 <sc68cal> cool
14:16:51 <xuhanp> current it only covers stateful part. The stateless part is coming
14:17:07 <xuhanp> or do you think they should come in two patches?
14:17:28 <sc68cal> xuhanp: depens on the size of the patches
14:17:42 <HenryG> It's usually better to have smaller change sizes if possible.
14:17:58 <HenryG> So best to split them up if it makes sense.
14:18:06 <sc68cal> ^ +1
14:18:07 <xuhanp> OK.
14:18:10 <xuhanp> will do.
14:18:39 <HenryG> You can make stateless depend on stateful if you rely on code there.
14:19:00 <xuhanp> yep. That's the plan
14:19:06 <HenryG> cool
14:21:12 <sc68cal> HenryG: how goes the bug for testing the radvd work
14:21:50 <HenryG> So first, for the record, the RADVD code is upstream. Yay!
14:22:26 <sc68cal> #link https://bugs.launchpad.net/neutron/+bug/1345341
14:22:28 <uvirtbot> Launchpad bug 1345341 in neutron "radvd needs functional tests" [High,New]
14:22:47 <sc68cal> HenryG: baoli_: Congrats!
14:23:42 <HenryG> Yes, so the functional testing is remaining. I have the very basics ready.
14:24:29 <sc68cal> ok - to be clear, these tests are for the neutron test tree, not tempest
14:24:46 <baoli_> we also need to have devstack support for radvd to be approved
14:24:57 * sc68cal gets confused about functional testing vs tempest vs neutron unit tests
14:25:22 <HenryG> sc68cal: Yes. But I need to get marun to confirm that I am on the right track, and specifically what functionality should be tested there.
14:25:40 <sc68cal> HenryG: Cool.
14:25:44 <baoli_> https://review.openstack.org/#/c/107801/
14:25:53 <sc68cal> baoli_: agree -we also need python-neutronclient changes to be merged
14:26:16 <HenryG> sc68cal: It is a bit confusing. It is neither unit testing nor tempest tests. It is functional testing. :)
14:26:44 <HenryG> The twilight zone.
14:27:15 <sc68cal> https://review.openstack.org/#/c/75871/
14:28:11 <HenryG> Looks like that ^^ needs a rebase.
14:28:17 <baoli_> sc68cal, I am actually using the subnet cli for the two modes without using the patch
14:28:24 <sc68cal> HenryG: again? <grumbles>
14:28:54 <xuhanp> sc68cal, will rebase soon
14:29:29 <sc68cal> baoli_: ah - some witchcraft :)
14:29:43 <baoli_> sc68cal, I means it works without requiring the python-neutronclietn change
14:30:01 <baoli_> you just need to append the two modes at the end of the neutron subnet-create command line
14:30:25 <sc68cal> baoli_: agree - but without xuhanp 's patch it won't show up in --help strings
14:30:39 <baoli_> sc68cal, that's right
14:32:12 <sc68cal> are there any other topics to discuss today?
14:35:54 <sc68cal> Ok everyone - until next week! Keep up the fantastic work, we're getting tons of stuff merged for Juno
14:36:25 <sc68cal> #endmeeting