15:01:54 <haleyb> #startmeeting neutron_dvr
15:01:54 <openstack> Meeting started Wed Oct  5 15:01:54 2016 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:56 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:58 <openstack> The meeting name has been set to 'neutron_dvr'
15:02:03 <haleyb> #chair Swami
15:02:03 <openstack> Current chairs: Swami haleyb
15:02:19 <Swami> haleyb: I might have a hard stop at 8.30 since i need to go out for Dr's appt.
15:03:06 <haleyb> Swami: np, thanks, promise not to assign you any "homework" after that :)
15:03:16 <haleyb> #topic Announcements
15:04:23 <haleyb> Just one general announcement - if you have any release critical issues raise them today, RC is done come Friday I believe
15:04:54 <Swami> haleyb: thanks
15:05:36 <haleyb> #topic Bugs
15:05:45 <Swami> haleyb: thanks
15:06:14 <Swami> This week we have one bug that has been tagged for dvr. I think it is an old bug but recently been tagged for DVR.
15:06:37 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1629539
15:06:38 <openstack> Launchpad bug 1629539 in neutron "Broken distributed virtual router w/ lbaas v1" [Undecided,Incomplete]
15:06:59 <Swami> The bug is related to supporting LBaaS v1 in Mitaka for DVR.
15:07:15 <haleyb> that is almost too large of a bug description
15:07:35 <Swami> It needs triaging. With respect to LBaaS v1 there should not be any issue, but we should look into it.
15:08:09 <Swami> The person who filed bug states that the namespaces are in place and the iptable rules are in place, but he is not able to get to through the VIP.
15:09:09 <haleyb> Swami: right, and it's mitaka so it will be harder for us to reproduce (plain mitaka, stable, ?)
15:09:27 <Swami> haleyb: yep
15:09:50 <Swami> haleyb: I am not even sure if LBaaS v1 is supported in Mitaka or not.
15:10:07 <haleyb> Swami: fwiw that submitter does use the openstack list a bit, and a lot of issues are solved with configuration changes
15:10:28 <Swami> haleyb: good to know.
15:10:45 <Swami> The next one in the list is
15:10:48 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1593354
15:10:49 <openstack> Launchpad bug 1593354 in neutron "SNAT HA failed because of missing nat rule in snat namespace iptable" [Undecided,New]
15:11:21 <Swami> jschwarz: Did you get a chance to validate this mitaka.
15:11:30 <jschwarz> Swami, nope :<
15:11:54 <Swami> jschwarz: no problem, can you confirm this week if possible.
15:12:02 <jschwarz> Swami, will try
15:12:13 <Swami> jschwarz: ok thanks.
15:12:14 <jschwarz> Swami, can't promise as my work week seems to be pretty tied by now :<
15:12:43 <Swami> jschwarz: ok, I do have a mitaka setup and will try to test it.
15:12:53 <jschwarz> Swami, excellent, thanks
15:13:17 <Swami> The next in the list
15:13:19 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1612192
15:13:20 <openstack> Launchpad bug 1612192 in neutron "L3 DVR: Unable to complete operation on subnet" [High,Confirmed]
15:14:13 <Swami> haleyb: Did we see anymore of this bug in the gate this week.
15:14:51 <haleyb> Swami: doing a logstash refresh now... nope
15:15:36 <Swami> haleyb: thanks
15:16:00 <Swami> The follow on bug is also gate related one.
15:16:03 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1612804
15:16:04 <openstack> Launchpad bug 1612804 in neutron "test_shelve_instance fails with sshtimeout" [High,Confirmed]
15:17:10 <Swami> Let us keep monitoring it and see if there is real DVR issue there.
15:17:12 <haleyb> i see some, but will have to do more filterint
15:17:51 <Swami> Ok, let us move on with the next one.
15:17:53 <Swami> #link https://bugs.launchpad.net/neutron/+bug/140345
15:17:55 <openstack> Launchpad bug 140345 in Coccinella "Errors getting messages in Tcl chatroom" [Medium,Fix released]
15:18:17 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/1403455 maybe :)
15:18:19 <openstack> Launchpad bug 1403455 in neutron "neutron-netns-cleanup doesn't clean up all L3 agent spawned processes" [High,Triaged] - Assigned to Manjeet Singh Bhatia (manjeet-s-bhatia)
15:18:34 <Swami> haleyb: thanks
15:19:28 <Swami> There is no existing patch yet for this bug and it has been tagged for RC3. Is it going to happen.
15:19:43 <haleyb> i don't know if manjeet is online?  not sure of nick
15:19:50 <Swami> haleyb: have you seen any patch on this bug, but I do see that there were lot of discussions.
15:20:21 <haleyb> it won't make rc3 as it's not critical
15:20:28 <Swami> haleyb: ok
15:20:34 <Swami> The next in the list is
15:20:36 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1606741
15:20:38 <openstack> Launchpad bug 1606741 in neutron "Metadata service for instances is unavailable when the l3-agent on the compute host is dvr_snat mode" [High,In progress]
15:21:26 <Swami> The existing patch is still under merge conflict. #link https://review.openstack.org/#/c/352686/
15:21:40 <Swami> I do see comments in there for alternate approach.
15:21:47 <haleyb> the person working on that un-assigned the bug, and hasn't updated the patch recently
15:22:22 <Swami> haleyb: ok if we don't see any response we can take it up.
15:23:08 <haleyb> Ok, last comment from assaf was looking for options
15:23:25 <Swami> haleyb: yes did see that.
15:23:45 <Swami> There is one more bug related to Metadata which kind skipped from our radar.
15:23:48 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1526855
15:23:49 <openstack> Launchpad bug 1526855 in neutron "VMs fail to get metadata in large scale environments" [Medium,Confirmed] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:24:17 <Swami> This is VM instance not able to reach the metadata proxy within the given retry count.
15:24:39 <Swami> This is only seen with DVR and I think it is because of the Dynamic nature of the router creation in DVR.
15:24:49 <jschwarz> discon
15:24:52 <jschwarz> oops
15:25:20 <Swami> This is only also seen for the new routers that are created for the VM instances that has a subnet on the router.
15:25:55 <Swami> We need to figure out is there a way to speed up this process. Again it might be tricky or thats how it works.
15:26:13 <Swami> jschwarz: no problem, recently it has been disconnecting quiet often.
15:26:31 <Swami> I am currently investigating this issue.
15:26:51 <Swami> The next one is
15:26:52 <haleyb> Swami: or make sure the ML2 logic exists for this? i.e. port not plugged before router ready
15:27:08 * haleyb can't remember the name at the moment
15:27:30 <Swami> haleyb: will check it out.
15:27:35 <Swami> The next one is
15:27:38 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1506567
15:27:39 <openstack> Launchpad bug 1506567 in neutron "No information from Neutron Metering agent" [High,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:27:59 <Swami> I need some review on this patch #link https://review.openstack.org/377108
15:28:12 <haleyb> Swami: yes, i owe you another review on that
15:28:17 <Swami> haleyb: I have addressed all your original comments.
15:28:45 <Swami> The next one is
15:28:48 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1571676
15:28:49 <openstack> Launchpad bug 1571676 in neutron "After binding a floating IP to VM, the static route can't work in DVR." [Undecided,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:29:06 <Swami> This patch also needs final review. #link https://review.openstack.org/#/c/308068/
15:29:41 <Swami> That's all I had for bugs.
15:29:52 <jschwarz> nothing from me
15:29:53 <Swami> There are two more patches related to the RFE.
15:30:05 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1577488
15:30:11 <openstack> Launchpad bug 1577488 in neutron "[RFE]"Fast exit" for compute node egress flows when using DVR" [Wishlist,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:30:22 <Swami> #link https://review.openstack.org/#/c/283757/ - patch 1
15:30:35 <Swami> #link https://review.openstack.org/#/c/355062/ - patch 2
15:31:02 <Swami> As we wind up on the Newton RC release this week, I hope we can pay attention to this RFE patch in the coming week to push it in.
15:31:25 <Swami> haleyb: That's all I had for this week.
15:31:38 <haleyb> Swami: yes, i will get back to reviewing all your patches
15:32:04 <Swami> haleyb: thanks
15:33:31 <haleyb> #topic Gate Failures
15:33:45 <Swami> haleyb: jschwarz: I need to leave now. If you have any questions ping me in the IRC channel when I am available.
15:33:58 <haleyb> I don't think DVR has been a gate issue as of late
15:34:00 <jschwarz> Swami, bye
15:34:31 <haleyb> Swami: see you later, thanks
15:35:01 <haleyb> jschwarz: on the topic of gate things, any progress on the HA job front?
15:36:00 <jschwarz> haleyb, I'm not actively working on it - anilvenkata is
15:36:13 <jschwarz> haleyb, I don't see him around so I don't have an answer for this
15:36:40 <jschwarz> haleyb, I did see a talk between anil, amuller and clarkb about 3-nodes setups for gate jobs
15:36:44 <haleyb> jschwarz: yes, i didn't see him figured you might know
15:37:08 <jschwarz> to say the least, clarkb was not very happy about this prospect seeing as how 2-nodes setups are fragile as it is with the subnetpools
15:37:43 <jschwarz> so this might be difficult to achieve... maybe we need to emphesize that it'll be used only in experimental jobs right now
15:38:11 <haleyb> then someone has to fix subnetpools :)
15:38:17 <jschwarz> XD
15:39:03 <haleyb> let's move on then
15:39:17 <haleyb> #topic Stable backports
15:39:47 <haleyb> I have been tracking most of the backports, does anyone have one I missed?
15:39:53 <jschwarz> haleyb, Anil submitted https://review.openstack.org/#/c/382191/ which might be relevant to us
15:40:07 <haleyb> jschwarz: i guess that's directed at you since it's just us and crickets
15:40:20 <jschwarz> lol
15:40:26 <jschwarz> I like the sound of crickets in the evening
15:41:01 <haleyb> there was a stable/mitaka as well, both have some typos in the release notes, and seem to fail jenkins
15:41:21 <jschwarz> aye, I see that now
15:41:57 <jschwarz> also i see that Anil has quite a few of backports: https://review.openstack.org/#/q/owner:%22venkata+anil%22+status:open
15:42:14 <jschwarz> most notably the famous l2pop patch
15:42:52 <haleyb> i'll make sure i'm a reviewer so i get notified
15:43:13 <jschwarz> I'm sure anil will appreciate :)
15:44:13 <haleyb> #topic Open discussion
15:44:29 <jschwarz> haleyb, so how about that presidential debate?
15:44:33 <jschwarz> ;-)
15:45:17 <haleyb> jschwarz: should have been called the not-so-presidential debate ?
15:45:40 <jschwarz> haleyb, it's a bit amusing to watch from the outside, so I'm not complaining
15:46:06 <haleyb> it just shouldn't be amusing :(
15:46:08 <haleyb> anyways...
15:46:43 <jschwarz> nothing from me today
15:46:56 <jschwarz> except that Israeli holidays suck so I'm barely working from now till the summit
15:46:57 <jschwarz> :<
15:47:20 * haleyb wonders how barely working could be bad
15:47:41 * jschwarz is a dedicated worker
15:48:06 <jschwarz> some would say, workaholic
15:48:38 <haleyb> i work to live, used to live to work but too old for that now :)
15:48:53 <jschwarz> reasonable :)
15:48:54 <haleyb> someone has to ruin the golf course
15:50:01 <haleyb> ok, we're done it seems
15:50:06 <haleyb> #endmeeting