15:01:11 <haleyb> #startmeeting neutron_dvr
15:01:12 <openstack> Meeting started Wed Jun  8 15:01:11 2016 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:13 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:16 <openstack> The meeting name has been set to 'neutron_dvr'
15:01:25 <haleyb> #chair Swami
15:01:26 <openstack> Current chairs: Swami haleyb
15:01:36 <Swami> hi
15:01:44 <Swami> you are not alone.
15:02:21 <haleyb> hah, i'm hoping obondarev is here today as well
15:02:41 <obondarev> haleyb: yep
15:02:46 <obondarev> hi folks
15:02:51 <Swami> obondarev: are you in US
15:02:58 <obondarev> Swami: nope
15:03:08 <Swami> obondarev: I saw you working all day yesterday our time.
15:03:22 <obondarev> Swami: not really :)
15:03:25 <Swami> obondarev: I thought you migth be visiting.
15:03:34 <Swami> s/migth/might
15:04:22 <haleyb> #topic Announcements
15:04:35 <obondarev> Swami: just checking reviews sometimes in the evening
15:04:58 <Swami> obondarev: no problem
15:05:10 <haleyb> N-1 is out, N-2 will sneak-up
15:05:46 <haleyb> want to move onto bugs though
15:05:49 <haleyb> #topic Bugs
15:06:03 <Swami> hi
15:06:28 <Swami> We had one bug this week.
15:06:35 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1590041
15:06:35 <openstack> Launchpad bug 1590041 in neutron "DVR: regression with router rescheduling" [High,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:06:46 <Swami> This is a regression bug.
15:07:24 <Swami> It seems there is one condition in the agent gateway check that is being missed since we initialize the object.
15:07:59 <Swami> the checks are made to the object instead of the namespace since it used to be intialized and created at the same time.
15:08:06 <Swami> I have pushed in a patch to address it.
15:08:20 <Swami> #link https://review.openstack.org/#/c/326729/2
15:09:08 <Swami> obondarev has also pushed in a patch to revert the original one.
15:09:12 <Swami> #link https://review.openstack.org/#/c/326574/
15:10:15 <Swami> I had a chat with obondarev about the revert, and he feels since it affects the stable branch, we can revert and push the fix back.
15:10:29 <Swami> haleyb: what do you think?
15:12:07 <haleyb> I guess I would feel more comfortable with the revert, but did your new patch fix the problem obondarev was seeing?
15:12:36 <obondarev> haleyb: there might be other problems as I stated in the review
15:12:52 <obondarev> we should test the case where gw is removed during agent is down
15:12:56 <Swami> obondarev: haleyb: I don't expect additional problems.
15:13:51 <Swami> obondarev: I will test it out. I have pushed in one more version just now to address the TRACE that you were seeing. But will test the gw removed scenario during agent is down.
15:14:28 <Swami> haleyb: I think with the addition of the 'exists' method it should be doing the right thing.
15:15:29 <haleyb> Swami: for now i would just revert, since the exists() change overlaps with your other change we've been iterating on imo
15:15:46 <Swami> haleyb: can you also review it.
15:16:15 <Swami> haleyb: I am ok with reverting.
15:16:29 <haleyb> Swami: i just pulled-up PS3
15:17:28 <Swami> haleyb:thanks take a look at it and let me know.
15:17:50 <Swami> haleyb: This patch should be my highest priority for today.
15:18:08 <Swami> obondarev: if possible can you also validate the PS3.
15:18:18 <obondarev> Swami: I will
15:18:41 <Swami> obondarev: thanks
15:20:16 <Swami> I only had one new bug this week.
15:20:50 <Swami> With respect to the old ones, the live-migration patch that is still underreview with nova.
15:21:09 <Swami> haleyb: The tempest test for validating the live migration has been fixed.
15:21:28 <Swami> #link https://review.openstack.org/#/c/286855/
15:22:03 <Swami> So I have pinged mriedem to take a look at it.
15:22:19 <haleyb> Swami: yes, what about the experimental jobs?  should that multinode live migration be passing?
15:22:55 <Swami> haleyb: I am not sure about that test that is failing, but the neutron-multinode and dvr-multinode are passing.
15:23:29 <Swami> haleyb: The original concern was that the neutron-multinode and dvr-multinode was failing.
15:23:47 <Swami> haleyb: we need to check what is the difference between these tests.
15:23:48 <haleyb> Swami: ok, we'll have to wait for matt to take a look, then we can remove the "DNM"
15:24:48 <Swami> There was one more bug that we discussed last week related to live migration that was filed, where the pings to floatingip fails.
15:24:55 <Swami> I don't have the bug number handy.
15:25:23 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1585165
15:25:23 <openstack> Launchpad bug 1585165 in neutron "floating ip not reachable after vm migration" [High,New] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:26:33 <Swami> This is a valid bug, and it seems that we are not cleaning up the old-node fip rules, after the migration happens.
15:27:12 <Swami> rossella_s: mentioned that she is going to push in a patch after my discussions with her.
15:28:20 <haleyb> do we need to re-assign the bug :)
15:28:36 <Swami> haleyb: yes I just tagged the bug to l3-dvr-backlog.
15:30:01 <Swami> haleyb: probably need to assign it to rossella_s if she is still working on it. But for now live it as such and if we don't see any patch from rossella_s I will push one.
15:31:09 <Swami> That's all I had for bugs today.
15:31:48 <haleyb> Swami: thanks, let me know if you need help with the migration bug
15:31:54 <Swami> Just a note on the RFE patches.
15:32:02 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1583694
15:32:02 <openstack> Launchpad bug 1583694 in neutron "[RFE] DVR support for Allowed_address_pair port that are bound to multiple ACTIVE VM ports" [Wishlist,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:32:21 <Swami> #link https://review.openstack.org/#/c/320669/ - server side patch.
15:32:55 <Swami> armax had a -2 on this patch since I was dealing with config options which he did not like and aslo you have mentioned in your first review as well.
15:33:09 <Swami> I have removed the config option right now on this patch.
15:33:29 <Swami> #link https://review.openstack.org/#/c/323618/ - agent side patch.
15:33:53 <Swami> Before I spend more cycles on these patches, can you check and let me know if we are heading on the right direction.
15:34:48 <haleyb> Swami: yes, will look
15:35:03 <Swami> haleyb: thanks
15:37:15 <haleyb> Swami: i'd like to also pick a time (next week?) to go through the old bugs with patches, as we've been accumulating some
15:37:30 <Swami> haleyb: sure.
15:38:40 <haleyb> let's skip to stable
15:38:46 <haleyb> #topic Stable backports
15:39:36 <haleyb> Swami had some backports up that ihar recently re-did to fix commit id issues, starting at https://review.openstack.org/#/c/327013/
15:40:03 <haleyb> https://review.openstack.org/#/c/327017/ too
15:42:00 <haleyb> Swami: we just need to get the series put on the wiki since the review id's changed, but think we can move forward with them
15:42:23 <haleyb> swami told me he is having irc issues right now
15:43:00 <Swami> hi
15:43:02 <Swami> back in
15:43:12 <Swami> using another client, something went wrong.
15:43:30 <haleyb> Swami: i had just mentioned the backports ihar re-did
15:43:41 <Swami> haleyb: yes.
15:43:55 <Swami> haleyb: thanks
15:44:30 <haleyb> https://review.openstack.org/#/c/327013/ https://review.openstack.org/#/c/327017/ https://review.openstack.org/#/c/327018/ and there were others
15:45:18 <Swami> haleyb:yes you are right.
15:45:27 <haleyb> These are all stable/liberty targetted
15:45:40 <Swami> haleyb: yes
15:46:30 <Swami> There was one other patch for /stable/mitaka which still had some merge conflicts.
15:46:33 <Swami> #link https://review.openstack.org/#/c/296850/
15:46:52 <Swami> I will look into it after I deal with the high priority ones.
15:48:04 <haleyb> Swami: ok, I'll work through the liberty ones first
15:48:16 <Swami> haleyb: yes that works.
15:49:03 <haleyb> anything else on stable?
15:49:21 <Swami> haleyb: nothing that i knew of.
15:49:28 <haleyb> #topic Open discussion
15:49:51 <haleyb> anyone have something to discuss?
15:50:27 <Swami> haleyb: obondarev: carl_baldwin: will be going on vacation next couple of weeks. I may not be able to join the DVR meeting for next two weeks.
15:50:53 <Swami> will be working couple of days on each week, but in a different timezone.
15:50:57 <haleyb> Swami: but who's going to fix all the bugs? :)
15:51:11 <obondarev> Swami: have a good vacation :)
15:51:15 <haleyb> Swami: safe travels
15:51:24 <Swami> haleyb: I mentioned that I will still be working, but will have limited time.
15:51:38 <Swami> haleyb: obondarev: thanks
15:52:46 <haleyb> ok, we get 8 minutes back
15:53:01 <haleyb> #endmeeting