15:00:15 <carl_baldwin> #startmeeting neutron_l3
15:00:15 <openstack> Meeting started Thu Sep 10 15:00:15 2015 UTC and is due to finish in 60 minutes.  The chair is carl_baldwin. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:19 <openstack> The meeting name has been set to 'neutron_l3'
15:00:22 <carl_baldwin> #topic Announcements
15:00:30 <haleyb> hi
15:00:38 <carl_baldwin> #link https://wiki.openstack.org/wiki/Meetings/Neutron-L3-Subteam#Agenda
15:00:54 <neiljerram> hi
15:01:00 <vikram> hi
15:01:23 <carl_baldwin> For announcements, Liberty-3 has been cut and Neutron has a small number of FFEs.  Otherwise, we’re in bug smashing mode.
15:01:31 <pavel_bondar> hi
15:01:58 <carl_baldwin> The QA mid-cycle is next week.  I’ll be attending.  I know regXboi will be too.
15:02:05 <carl_baldwin> Any other announcements?
15:02:46 <carl_baldwin> #topic Bugs
15:02:53 <mlavalle> hi
15:03:01 <carl_baldwin> mlavalle: hi
15:03:06 <mlavalle> so, https://bugs.launchpad.net/neutron/+bug/1461172
15:03:07 <openstack> Launchpad bug 1461172 in neutron "neutron.tests.functional.agent.test_l3_agent.MetadataL3AgentTestCase.test_access_to_metadata_proxy times out intermittently" [Critical,Fix committed] - Assigned to Armando Migliaccio (armando-migliaccio)
15:03:19 <carl_baldwin> We had a critical come in last night.  I just added it to the agenda.
15:03:29 <mlavalle> armax proposed and merged a fix for it
15:03:46 <carl_baldwin> Very nice armax.
15:03:46 <mlavalle> an so far we haven't seen any new ocurrences
15:04:09 <mlavalle> just checked a couple of minutes ago and the count is 0
15:04:20 <carl_baldwin> mlavalle: Great.
15:04:44 <carl_baldwin> I just realized that I probably didn’t add the new bug to the place on the wiki where we’re all probably looking.
15:04:54 <mlavalle> regXboi removed a couple of high importance bugs yesterday from the agenda
15:05:00 <carl_baldwin> #link https://bugs.launchpad.net/neutron/+bug/1494114
15:05:02 <openstack> Launchpad bug 1494114 in neutron "Bad router request: Router already has a port on subnet" [Critical,Confirmed] - Assigned to Carl Baldwin (carl-baldwin)
15:05:15 <pavel_bondar> and I have #link https://bugs.launchpad.net/neutron/+bug/1494351
15:05:15 <openstack> Launchpad bug 1494351 in neutron "Observed StaleDataError in gate-neutron-dsvm-api tests if reference IPAM driver is used" [Undecided,New] - Assigned to Pavel Bondar (pasha117)
15:05:29 <pavel_bondar> just filed it today, probably too late
15:05:52 <mlavalle> so we are making good progress with the high importance list as well
15:06:13 <mlavalle> the one reminaing is https://bugs.launchpad.net/neutron/+bug/1365473
15:06:15 <openstack> Launchpad bug 1365473 in neutron "Unable to create a router that's both HA and distributed" [High,In progress] - Assigned to Adolfo Duarte (adolfo-duarte)
15:06:38 <mlavalle> I checked yesterday and it's been worked on. I'll keep tracking it
15:07:03 <mlavalle> I'll start working the new added by carl_baldwin and pavel_bondar
15:07:29 <carl_baldwin> 1365473 seems to be spawning other bugs...
15:08:50 <carl_baldwin> pavel_bondar: I marked yours High.  Do you have a handle on it?  Or, do you need a hand?
15:09:16 <pavel_bondar> I think I need a hand
15:10:25 <pavel_bondar> carl_baldwin: have no yet much experience on issues like that: "UPDATE statement on table 'ipamavailabilityranges' expected to update 1 row(s); 0 were matched."
15:10:33 <pavel_bondar> so it may take a while to investigate
15:11:14 <carl_baldwin> pavel_bondar: Since it doesn’t affect this release, I think its priority is inherently a little lower at this time.  Will you be around the neutron channel after the meeting?  We could look at it.
15:11:48 <pavel_bondar> carl_baldwin: ok, if it is not time critical I can handle that
15:12:09 <pavel_bondar> carl_baldwin: will check on neutron channel after meeteing
15:12:57 <carl_baldwin> mlavalle_: Do you have time to look in to https://bugs.launchpad.net/neutron/+bug/1494114 ?  I can help with that too if you need.
15:12:58 <openstack> Launchpad bug 1494114 in neutron "Bad router request: Router already has a port on subnet" [Critical,Confirmed] - Assigned to Carl Baldwin (carl-baldwin)
15:13:16 * mlavalle is sorry. got disconnected
15:13:41 <carl_baldwin> mlavalle: No worries.  You back yet?
15:13:47 <mlavalle> yeah
15:13:50 <mlavalle> :-)
15:15:34 * carl_baldwin waits to let mlavalle catch up.
15:16:46 <mlavalle> carl_baldwin: I'll look at https://bugs.launchpad.net/neutron/+bug/1494114
15:16:47 <openstack> Launchpad bug 1494114 in neutron "Bad router request: Router already has a port on subnet" [Critical,Confirmed] - Assigned to Carl Baldwin (carl-baldwin)
15:16:57 <mlavalle> if I need help, i'll yell
15:17:24 <mlavalle> and I'll add it to the correct section of the wiki
15:17:27 <carl_baldwin> mlavalle: ok, I’ll be around (most of the time)
15:17:43 <carl_baldwin> mlavalle: Thanks!  Anything else on bugs?
15:17:50 <mlavalle> not from me
15:18:48 <carl_baldwin> #topic BGP
15:18:56 <carl_baldwin> tidwellr: vikram: hi
15:18:57 <tidwellr> hi
15:19:35 <tidwellr> we've got progress, I was able to run things end-to-end late last week
15:19:43 <vikram> hi
15:19:43 <carl_baldwin> tidwellr: awesome!
15:19:54 <tidwellr> discovering issues along the way of course
15:19:55 <vikram> I am also testing
15:19:57 <juno> will not dynamic routing be integrated in liberty?
15:20:13 <tidwellr> juno: no, it won't make it
15:20:24 <juno> ok
15:20:27 <tidwellr> juno: we're aiming for M-1 now
15:20:51 <vikram> tidwellr: Did you tested it with quagga?
15:20:55 <juno> tidwellr: what M associate 1?
15:21:01 <vikram> tidwellr: I am the actual peering?
15:21:13 <vikram> juno: Mitaka-1
15:21:26 <tidwellr> vikram: yes, had some fits and starts
15:21:34 <tidwellr> vikram, they were mostly ryu issues
15:21:43 <vikram> tidewellr: awesome:)
15:22:06 <vikram> tidewellr: please record the steps and logs.. we got to document it for Test Ref
15:22:40 <vikram> test devref or so..
15:23:07 <tidwellr> given the state of address scopes, I'm a little lost on how to proceed on advertising tenant networks via BGP
15:23:10 <carl_baldwin> Sounds like good progress.
15:23:59 <tidwellr> that can be an offline discussion, I'm really focused on getting route advertisements flowing even if they're wrong
15:24:22 <carl_baldwin> tidwellr: ping me on that.  Address scopes will be disabled for Liberty but is very near completion.
15:24:35 <tidwellr> carl_baldwin: ack
15:25:12 <carl_baldwin> I’m hoping to get some of the refactoring and general enhancements merged soon.  Then full address scopes functionality should be a matter of cherry-picking a couple of patches.
15:25:47 <carl_baldwin> I talked to mestery about continuing to merge some of the patches that I have lined up which are small general fixes and prepare the way for the real implementation.
15:26:23 <tidwellr> ok, so early in Mitaka we should see it merged?
15:26:45 <carl_baldwin> So, haleyb and ZZelle can continue to merge those and then merged the final implementation soon after Mitaka opens.
15:26:59 <carl_baldwin> s/merged/merge/
15:27:04 <carl_baldwin> tidwellr: yes
15:27:09 <tidwellr> awesome
15:27:24 <haleyb> carl_baldwin: ack, was wondering about those ip_lib changes
15:27:37 <vikram> carl_baldwin: Thanks
15:27:48 <tidwellr> those patches will merge before BGP does, so no worries then
15:28:04 <carl_baldwin> haleyb: If you’re comfortable with the changes going in at this stage then you can +2/+A as appropriate.
15:28:35 <carl_baldwin> Most of them are pretty innocuous, I think.  But, I’m biases since I’m the author.
15:28:37 <carl_baldwin> :)
15:29:39 <carl_baldwin> tidwellr: You may need to modify your local code to reenable address scopes.  I’ll post a patch to do that as soon as Mitaka opens.
15:30:00 <carl_baldwin> tidwellr: Essentially, revert this patch locally:  https://review.openstack.org/#/c/221502/
15:30:16 <tidwellr> right, that's just a one-liner
15:32:12 <tidwellr> nothing else from me
15:32:16 <carl_baldwin> tidwellr: Thanks
15:32:20 <carl_baldwin> #topic DNS
15:32:27 <mlavalle> hi again
15:32:29 <carl_baldwin> mlavalle: Still connected?  :)
15:32:36 <mlavalle> yeah :-)
15:32:58 <mlavalle> so the functionality for floating ips in https://review.openstack.org/#/c/212213/ is complete
15:33:17 <mlavalle> today I am working on the allmbic migration script
15:33:46 <mlavalle> once I have that, I will be able to deploy the patchset to my devstack and test the functionality end to end
15:33:56 <mlavalle> that will happen today and tomorrow
15:34:16 <mlavalle> so by next week I am hoping I will have this functionality debugged
15:34:20 <carl_baldwin> mlavalle: Excellent.
15:34:39 <carl_baldwin> mlavalle: Did you have any thoughts on connecting VMs directly to a external network?
15:35:03 <mlavalle> yes, that is the next stage....
15:35:28 <mlavalle> I wanted to have to have the fip functionality ready, to secure the presentation / demo in Tokyo
15:35:37 <carl_baldwin> mlavalle: I feel like I’ve thought this through before but my memories are fading.  Could we discuss it in channel sometime?  Maybe after we get the critical bugs squashed?
15:35:48 <carl_baldwin> mlavalle: That sounds like a good plan.
15:36:24 <mlavalle> onc fips are good, i'll loop back to the VM's connected to external networks
15:36:46 <carl_baldwin> mlavalle: Also, as we’re beginning to route internal addresses to the external networks without NAT, we should think about how this DNS work will need to adapt.
15:37:06 <mlavalle> will do.....
15:37:25 <carl_baldwin> We’ve got some time and don’t need to hold up the floating IP stuff you’re wrapping up.
15:37:45 <mlavalle> the other aspect is that we are going to need designate + neutron in the gate to merge all this
15:38:10 <mlavalle> Kiall will start or started yesterday an email thread to get help form the infra guys
15:38:17 <carl_baldwin> mlavalle: That is a good point.  I was lurking yesterday (or the day before) in the designate meeting.
15:38:28 <mlavalle> he will be cooying carl_baldwin in that email
15:38:42 <carl_baldwin> mlavalle: I will look for it.  Will it be on the ML?
15:38:51 <mlavalle> yes, ML thread
15:39:08 <carl_baldwin> Great.
15:39:20 <mlavalle> and that's it this week for DNS update
15:39:48 <carl_baldwin> mlavalle: Thanks.
15:40:03 <carl_baldwin> #topic Address Scopes
15:40:19 <carl_baldwin> I think I’ve already given my update.  But, I’ll summarize.
15:40:52 <carl_baldwin> The API extension will be reenabled as soon as Mitaka opens with this:  https://review.openstack.org/#/c/222250/
15:41:06 <carl_baldwin> If you need to play with it, you’ll want to cherry-pick that patch locally.
15:41:36 <carl_baldwin> I just didn’t want people playing with the API and get a false sense for how it will really behave without the complete implementation.
15:42:08 <carl_baldwin> That’s about it from me.
15:42:33 <carl_baldwin> #topic Open Discussion
15:42:42 <carl_baldwin> Have I missed any topics that we should discuss today?
15:43:22 <mlavalle> I don't think so
15:44:13 <carl_baldwin> ok  We’ll get some time back.
15:44:22 * carl_baldwin goes to tackle the critical bugs.
15:44:29 <carl_baldwin> #endmeeting