15:01:55 <rossella_s> #startmeeting neutron_upgrades
15:01:56 <openstack> Meeting started Mon Feb  1 15:01:55 2016 UTC and is due to finish in 60 minutes.  The chair is rossella_s. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:57 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:02:00 <openstack> The meeting name has been set to 'neutron_upgrades'
15:02:02 <dguitarbite> hello
15:02:05 <rossella_s> hi all! thanks for joining
15:02:39 <rossella_s> Ihar is traveling today so I will replace him
15:02:53 <rossella_s> #topic Announcements
15:03:00 <mhickey> rossella_s: go for it! :)
15:03:53 <rossella_s> as you might have read in the dev mailing list there will be an upgrade code sprint in Brno, March 14-16
15:04:05 <rossella_s> I hope to see you all there!
15:04:16 <mhickey> rossella_s: tentative
15:04:24 * pc_m lurking
15:04:54 <korzen> rossella_s: I'm planning to be there, needs to get approval for travelling
15:04:59 <rossella_s> mhickey, right, I should have stated it more clearly
15:05:11 <rossella_s> korzen, go for it
15:05:42 <mhickey> I need approval too
15:06:11 <rossella_s> I guess it's time to start asking for it
15:06:13 <rossella_s> let's move on
15:06:16 <rossella_s> #topic partial grenade
15:06:59 <rossella_s> sc68cal seems to be not here
15:08:21 <rossella_s> I don't have any fresh info, only what was shared in the dev mailing list...there are still test failures when ssh-ing into an instance, might be another MTU issue
15:08:45 <rossella_s> #topic versioned objects
15:09:46 <rossella_s> I pushed an update for the port object patch https://review.openstack.org/#/c/253641/ mhickey helped me in fixing the UT failures
15:10:33 <roaet> Hi. Re OVO: electrocucaracha and saisriki are selecting two other OVO objects to convert. They will update the wiki asap.
15:10:39 <rossella_s> now we have the allowed_address_pairs extension that is included in the port object, to make sure we can manage extension the right way
15:11:39 <rossella_s> roaet, we discussed that earlier, we want to start porting port and network. After we finish the port for those objects we can start moving to other ones
15:11:42 <korzen> https://review.openstack.org/#/c/264273 subnet OVO thanks for the review rossella_s :) I have replied and pushed new patch, still working on UT and final touches
15:11:54 <rossella_s> korzen, thanks a lot!
15:12:19 <roaet> rossella_s: ok so don't start on the other objects. Got it. I'll have them look at those patches and help where they can.
15:12:48 <korzen> roaet, I think that SubnetPool OVO can be taken
15:12:56 <korzen> I did not start working on it
15:12:57 <rossella_s> roaet, it would be great if electrocucaracha and saisriki join our meeting
15:13:13 <rossella_s> korzen, would you mind updating the backlog then?
15:13:13 <mhickey> rossella_s: extrac dhcp opt https://review.openstack.org/#/c/273072/, patch on the way to fix gate and ut issues.
15:13:47 <korzen> rossella_s, OK I will add it to the backlog
15:13:56 <roaet> There appear to be many objects in work. korzen I will tell them to check out SubnetPool.
15:13:57 <rossella_s> mhickey, how are the UT for the other patches going? can you update us since you were so brave to volunteer to fix them?
15:14:12 <mhickey> rossella_sok...
15:14:40 <mhickey> allowed address pairs (https://review.openstack.org/#/c/268274/): has now passed
15:15:16 <rossella_s> roaet, there are only 2 objects: port and network...of course the objects that compose them need to be ported too. Please tell them that we can coordinate better if they join this meeting.
15:15:19 <mhickey> port (https://review.openstack.org/#/c/253641): I will need to take a look at. still an issue
15:15:36 <mhickey> dhcp extra: details above ^^^
15:15:49 <rossella_s> thanks mhickey !
15:15:52 <mhickey> Thats it for the moment
15:16:04 <mhickey> no problem! :)
15:16:11 <rossella_s> dguitarbite, do you have any update?
15:16:41 <dguitarbite> rossella_s: Not much done as of now.
15:16:45 <mhickey> rossella_s: one last thing...
15:17:32 <rossella_s> dguitarbite, I can't see your name in the list under Object implementation in the team page, can you update it if you are working on some object please?
15:17:34 <dguitarbite> rossella_s: I have some code written but not good enough to discuss it for the time being. I should send a PR by day after if time permits.
15:17:45 <dguitarbite> rossella_s: yes, Ill do that.
15:18:01 <rossella_s> dguitarbite, thanks
15:18:09 <rossella_s> mhickey, go on please :)
15:18:59 <mhickey> I just tried to get the ut's working on port and allowed address pairs patches. I will tackle the good review comments from korzen and others shortly.
15:19:44 <rossella_s> mhickey, good job!
15:19:52 <rossella_s> dguitarbite, which port extension are you working on?
15:20:16 <dguitarbite> Security Groups and port security both of them. I thought its one task here.
15:20:28 <dguitarbite> rossella_s: I can give away port security I have not touched anything in there yet
15:20:37 <rossella_s> dguitarbite, nope it's two separate extensions
15:21:10 <rossella_s> dguitarbite, ok, then please update the team page accordingly
15:21:18 <rossella_s> dguitarbite, thanks!
15:21:25 <dguitarbite> rossella_s: Ok, Ill update it. No sweat
15:22:18 <rossella_s> anything else regarding objects?
15:22:42 <korzen> I've got 2 opens
15:23:03 <saisriki> @rossella_s, myself and electrocucaracha just joined
15:23:13 <korzen> 1) support for multiple primary keys
15:23:19 <rossella_s> saisriki, electrocucaracha welcome!
15:23:30 <rossella_s> korzen, good point
15:23:38 <korzen> 2) Add IP address type decorator for sqlalchemy, nova style
15:23:48 <rossella_s> korzen, very good points
15:24:03 <korzen> someone interested?
15:24:18 <rossella_s> korzen, for the first one if I good remember you have a way to handle that in the network ovo, we just need to make it general
15:25:29 <korzen> rossella_s yea, not so much general I think
15:25:46 <saisriki> korzen: I can take a look at IP address type decorator for sqlalchemy
15:26:03 <rossella_s> thanks saisriki !
15:26:24 <korzen> rossella_s, do you think it can be merged in Mitaka? the IP address decorator?
15:26:40 <korzen> it would require the migration script
15:27:57 <rossella_s> korzen, we can make it but we need to work on it and finish it up quickly
15:28:28 <korzen> saisriki thanks, the details are in my commit: https://review.openstack.org/#/c/264273/10/neutron/objects/subnet.py@151
15:29:06 <korzen> it would be beneficial to have it ready in Mitaka, the IP address and IP network CIDR
15:29:20 <saisriki> I guess, the idea for me and electrocucaracha was to look at OVO migration of SubnetPools
15:29:47 <saisriki> I don't think I can work on two items at the same time!!\
15:30:20 <rossella_s> saisriki, feel free to pick the item that it's most convenient for you. We have plenty of work to distribute
15:30:37 <saisriki> rossella_s: thank you
15:31:02 <korzen> saisriki, I will take a look in near future in the IP address decorator so feel free to start the SubnetPool OVO first
15:31:15 <rossella_s> korzen, thanks for stepping up
15:31:16 <saisriki> korzen: ok, thanks!!\
15:32:09 <rossella_s> anything else?
15:32:44 <rossella_s> 3
15:32:46 <rossella_s> 2
15:32:47 <rossella_s> 1
15:32:48 <electrocucarach_> me
15:32:58 <electrocucarach_> sorry, I have the patch for the documentation
15:33:14 <rossella_s> electrocucarach_, can you paste the link?
15:33:15 <electrocucarach_> should I create a bug and attach that patch to that bug?
15:33:19 <rossella_s> please
15:34:11 <electrocucarach_> well, this is the commit that I did in my own repo... but should I create the bug in launchpad?
15:34:12 <electrocucarach_> https://github.com/electrocucaracha/neutron/commit/a1e0ad83539cb912b3b51dbacedfe7c417d9d7a9
15:34:55 <rossella_s> electrocucarach_, yes please create the bug
15:34:58 <electrocucarach_> It's for autogenerate the db schemas in the documentation
15:35:20 <rossella_s> electrocucarach_, that's a very nice improvement, thanks for that!
15:35:47 <electrocucarach_> well, that was the idea of roaet
15:35:57 <rossella_s> thanks roaet too
15:36:26 <rossella_s> anything else?
15:36:37 <mhickey> no
15:36:40 <electrocucarach_> no
15:36:55 <rossella_s> so let's move on
15:37:06 <rossella_s> #topic open discussion
15:37:40 <rossella_s> there's one thing to discuss if we want to have a specif tag to track the ovo work
15:37:47 <rossella_s> I think this was asked by pc_m
15:38:19 <rossella_s> s/tag/rfe sorry
15:38:22 <pc_m> yeah, could just be a bug # so that items can be tracked, or use a tag on the commits.
15:38:38 <mhickey> pc_m: good idea
15:38:47 <korzen> for me it is a good step, lets inform others what we are working on
15:39:09 <rossella_s> it seems we all agree here
15:39:09 <roaet> Agree a tag is a good idea
15:39:23 <pc_m> cool
15:39:40 <rossella_s> that was easy. pc_m thanks for your input!
15:39:50 <pc_m> sure
15:40:07 <pc_m> helps those lurking to see what is going on :)
15:40:37 <rossella_s> pc_m, I perfectly agree...it actually helps people involved too to keep track of what's going on
15:40:38 <korzen> the question is are we going to have 1 RFE
15:40:47 <korzen> or RFE per OVO
15:40:53 <mhickey> welcome pc_m
15:41:08 <rossella_s> korzen, I think one RFE...we can then use the Partial-bug tag in the commit
15:41:10 <pc_m> mhickey: hi
15:41:12 <mhickey> korzen, I think ihrachys suggested 1?
15:41:31 <pc_m> Could just use a topic on the commits
15:41:39 <korzen> rossella_s mhickey, ok fine for me
15:42:06 <pc_m> Or a bug so that there is a place to see the commits. The topic is nice for searches. (could do both, if desired).
15:42:51 <rossella_s> pc_m, a topic requires a blueprint
15:43:02 <pc_m> ah
15:43:24 <pc_m> I thought you could create one w/o.
15:44:20 <pc_m> Maybe it's a manually process, if not using a BP.
15:44:24 <rossella_s> pc_m, well the topic is set by the branch name you use
15:44:43 <rossella_s> pc_m, for convention we use to set the topic to bp/bp_name
15:44:57 <pc_m> rossella_s: I used to do it for bugs too.
15:45:25 <rossella_s> we can agree that we set a specific topic for ovo but there's no way to reinforce it...I think it's better to use an RFE then
15:46:03 <pc_m> it would most like need to be a team convention.
15:46:33 <pc_m> likely
15:47:02 <rossella_s> yep...so I prefer using "stardard" way to group patches, either RFE or blueprint...
15:47:23 <pc_m> rossella_s: sure. np. just a thought
15:48:12 <rossella_s> pc_m, mine it's just a thought too
15:49:04 <rossella_s> anything else we want to discuss?
15:49:28 <mhickey> ok, from me
15:49:56 <korzen> about https://review.openstack.org/#/c/273517 OVO common enum classes for IP version and IPv6 modes
15:50:10 <mhickey> korzen: yes
15:50:38 <korzen> like I said in the review, we should sync with Ihar on that one, he requested the strict versioning
15:50:55 <mhickey> korzen: sure, I agree.
15:51:17 <rossella_s> korzen, agreed, let's wait for his feedback
15:51:31 <korzen> I'm ok with both approaches: the integer or enum field
15:51:39 <mhickey> korzen: I am just caught between whether 2 int values are worth it.
15:52:18 <mhickey> korzen: thanks! :)
15:52:26 <rossella_s> anything else?
15:53:08 <rossella_s> 3
15:53:10 <rossella_s> 2
15:53:12 <rossella_s> 1
15:53:17 <mhickey> bye, thanks rossella_s
15:53:31 <rossella_s> thanks all for attending!
15:53:34 <korzen> bye, thx all :)
15:53:47 <rossella_s> I am very happy regarding the progress so let's keep going like this!
15:53:48 <saisriki> thank you
15:53:52 <rossella_s> #endmeeting