| *** hoangcx has quit IRC | 00:14 | |
| *** felipemonteiro_ has joined #openstack-meeting-5 | 00:40 | |
| *** felipemonteiro_ has quit IRC | 00:46 | |
| *** yamahata has quit IRC | 01:19 | |
| *** slaweq has joined #openstack-meeting-5 | 01:20 | |
| *** slaweq has quit IRC | 01:25 | |
| *** markvoelker_ has joined #openstack-meeting-5 | 01:52 | |
| *** markvoelker has quit IRC | 01:53 | |
| *** slaweq has joined #openstack-meeting-5 | 03:21 | |
| *** slaweq has quit IRC | 03:26 | |
| *** yamahata has joined #openstack-meeting-5 | 04:11 | |
| *** slaweq has joined #openstack-meeting-5 | 05:22 | |
| *** slaweq has quit IRC | 05:27 | |
| *** ricolin has joined #openstack-meeting-5 | 06:24 | |
| *** slaweq has joined #openstack-meeting-5 | 06:36 | |
| *** slaweq_ has joined #openstack-meeting-5 | 06:39 | |
| *** slaweq has quit IRC | 06:41 | |
| *** dims has quit IRC | 06:54 | |
| *** dims has joined #openstack-meeting-5 | 06:56 | |
| *** dims has quit IRC | 07:01 | |
| *** dims has joined #openstack-meeting-5 | 07:02 | |
| *** yamahata has quit IRC | 07:49 | |
| *** radeks has joined #openstack-meeting-5 | 08:25 | |
| *** derekh has joined #openstack-meeting-5 | 08:40 | |
| *** ricolin has quit IRC | 09:25 | |
| *** tuanla____ has joined #openstack-meeting-5 | 09:34 | |
| *** ricolin has joined #openstack-meeting-5 | 09:56 | |
| *** ricolin has quit IRC | 09:59 | |
| *** ricolin has joined #openstack-meeting-5 | 10:17 | |
| *** ricolin has quit IRC | 10:34 | |
| *** ricolin has joined #openstack-meeting-5 | 10:48 | |
| *** ricolin has quit IRC | 11:17 | |
| *** mugsie has quit IRC | 11:42 | |
| *** mugsie has joined #openstack-meeting-5 | 11:42 | |
| *** mugsie has quit IRC | 11:42 | |
| *** mugsie has joined #openstack-meeting-5 | 11:42 | |
| *** tuanla____ has quit IRC | 12:02 | |
| *** yamamoto_ has quit IRC | 12:21 | |
| *** yamamoto has joined #openstack-meeting-5 | 12:27 | |
| *** ricolin has joined #openstack-meeting-5 | 12:38 | |
| *** mjturek has joined #openstack-meeting-5 | 12:48 | |
| *** ricolin has quit IRC | 12:59 | |
| *** markvoelker_ has quit IRC | 13:16 | |
| *** markvoelker has joined #openstack-meeting-5 | 13:18 | |
| *** hoangcx has joined #openstack-meeting-5 | 13:36 | |
| *** hoangcx_ has joined #openstack-meeting-5 | 13:37 | |
| *** hoangcx has quit IRC | 13:41 | |
| *** hongbin_ has joined #openstack-meeting-5 | 13:46 | |
| *** mlavalle has joined #openstack-meeting-5 | 13:53 | |
| *** lujinluo has joined #openstack-meeting-5 | 13:55 | |
| *** TuanVu has joined #openstack-meeting-5 | 13:58 | |
| *** annp_ has joined #openstack-meeting-5 | 13:58 | |
| lujinluo | #startmeeting neutron_upgrades | 14:00 |
|---|---|---|
| openstack | Meeting started Thu Apr 19 14:00:31 2018 UTC and is due to finish in 60 minutes. The chair is lujinluo. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:00 |
| *** openstack changes topic to " (Meeting topic: neutron_upgrades)" | 14:00 | |
| openstack | The meeting name has been set to 'neutron_upgrades' | 14:00 |
| mlavalle | o/ | 14:00 |
| lujinluo | o/ | 14:00 |
| *** ihar has joined #openstack-meeting-5 | 14:01 | |
| ihar | o/ | 14:01 |
| lujinluo | First thing first. mlavalle has some concerns about this patch, https://review.openstack.org/#/c/556667/ | 14:01 |
| lujinluo | not sure if ihar gets time to check the comments already | 14:01 |
| TuanVu | Hi mlavalle | 14:02 |
| TuanVu | Hi Ihar | 14:02 |
| mlavalle | admitedly my concern my be unwarranted | 14:02 |
| TuanVu | Hi Luo-san | 14:02 |
| mlavalle | hey TuanVu | 14:02 |
| lujinluo | hey TuanVu ! | 14:02 |
| ihar | mlavalle: I saw point on more comments needed. agreed. | 14:02 |
| lujinluo | but i agree that, we may need more comments to help future readers | 14:03 |
| ihar | I won't be able to get to this patch till Monday though so that's a bummer. | 14:03 |
| lujinluo | ok, i will try to put some comments but i need your review/double check just in case i misunderstand anything | 14:04 |
| ihar | I see there is another comment there not just about comments though | 14:04 |
| *** hungpv_ has joined #openstack-meeting-5 | 14:05 | |
| ihar | mlavalle: I didn't have a chance to think about it. how critical is the weak ref comment? is there a chance for getting this in with the caveat I will follow up on Monday? | 14:05 |
| mlavalle | ihar: yeah, as long as we answer to ourselves the question soon and feel comfortable that we are not get in trouble, I am fine | 14:06 |
| mlavalle | so maybe let's revise the patch with comments, get it in | 14:07 |
| ihar | ok. I will put it in my list to follow up. | 14:07 |
| mlavalle | and then work next week on the weakreference concern | 14:07 |
| mlavalle | does that work? | 14:07 |
| ihar | if lujinluo is able to handle the comments part, sure | 14:07 |
| mlavalle | do you want to do that lujinluo? | 14:08 |
| lujinluo | i can help with the comments but maybe not weakreference | 14:08 |
| mlavalle | I will help. it is a good exercise for you and me lujinluo. We better get very familiar with this piece of coe :-) | 14:08 |
| ihar | yeah not weak ref. I will need to take a look into it but seems from the comment it's a legit thing. | 14:08 |
| ihar | yes you better are, it's as you said esoteric | 14:09 |
| lujinluo | mlavalle: exactly ;) | 14:09 |
| mlavalle | cool! | 14:09 |
| *** yamahata has joined #openstack-meeting-5 | 14:10 | |
| lujinluo | #actionitem mlavalle and lujinluo help with the comments in auto expire fk patch | 14:10 |
| ihar | lujinluo: it's #action btw | 14:10 |
| lujinluo | thanks, i was wondering why the magic did not happen | 14:10 |
| mlavalle | LOL | 14:10 |
| lujinluo | #action mlavalle and lujinluo help with the comments in auto expire fk patch | 14:10 |
| lujinluo | ok, anyway.. next one is the auto detect old and new engine facade https://review.openstack.org/#/c/553617/17 | 14:11 |
| lujinluo | i understand that we need the former patch before get this one in, but it is worth to mention the status | 14:12 |
| ihar | yeah, it's good to go, just need to nudge the base one | 14:12 |
| *** skazi has quit IRC | 14:13 | |
| lujinluo | \o/ | 14:13 |
| lujinluo | #topic OVO | 14:13 |
| *** openstack changes topic to "OVO (Meeting topic: neutron_upgrades)" | 14:13 | |
| lujinluo | https://review.openstack.org/#/q/status:open+project:openstack/neutron+branch:master+topic:bp/adopt-oslo-versioned-objects-for-db | 14:13 |
| lujinluo | first one is new https://review.openstack.org/#/c/562489/ | 14:14 |
| *** hoangcx_ is now known as hoangcx | 14:14 | |
| ihar | CI so green, I can't stop being amazed | 14:14 |
| lujinluo | i think TuanVu is trying to tackle joins here | 14:14 |
| TuanVu | yes | 14:14 |
| TuanVu | if there's any concern, please feel free to let me know | 14:15 |
| lujinluo | wow, yeah, all green. so good for our eyes | 14:15 |
| lujinluo | ihar: i actually have a concern of the way we deal with joins | 14:15 |
| ihar | ok this patch moved the code but it didn't solve the issue that we pass sqlalchemy queries around | 14:15 |
| lujinluo | previously we just moved all the methods to object | 14:15 |
| ihar | which should not be the way we interact with ovo objects | 14:15 |
| lujinluo | yes, i think eventually we need a way to deal with joins | 14:16 |
| lujinluo | i mean, we can rewrite all the joins with get_object and filter in python, but do you prefer if we can have something like join_object()? ihar | 14:17 |
| ihar | in this particular case, why not adding a new extra filter called 'service_type' and handle it accordingly in ovo? | 14:17 |
| lujinluo | no, i mean. if we do not limit to this particular case | 14:18 |
| lujinluo | because most of the remaining work is about joins | 14:18 |
| ihar | lujinluo: we already have get_object[s] as a common interface. it would be nice if we can stick to it, just expand it for the Subnet class to support a new extra filter for service_type. Then you could just Subnet.get_objects(context, network_id=nid, service_type=service_type) and not care about implementation complexity | 14:18 |
| ihar | in general, I would recommend expanding standard interfaces like get_object[s] to support additional filters (of course it requires some if/else branching in their implementations depending on which filters are passed). | 14:20 |
| ihar | we may add custom getters like get_subnets_by_service_type(...) if it seems hard to achieve right now | 14:20 |
| ihar | but we should never have any part of sqlalchemy interfaces / types be exposed (so no queries passed around as arguments, or models returned from ovo methods) | 14:21 |
| mlavalle | is the corllary that at this point you see the OVO interface as stable and the rest of the work should happen underneath? | 14:21 |
| ihar | mlavalle: more or less, I think what we have is a good interface that we should try not to expand with custom methods without clear reason. | 14:21 |
| lujinluo | ok, so you would prefer that we manipulate with get_object(s) | 14:22 |
| mlavalle | ihar: that's good insight. thanks | 14:22 |
| ihar | thing is, if some other plugin wants to use ovo objects and needs to filter by service_type, what should they do? | 14:22 |
| ihar | supposedly we are moving base interface into neutron-lib and, as per late boden's patch, provide a registry to get objects from | 14:22 |
| mlavalle | yeap | 14:23 |
| ihar | but all the external plugins know about objects received from the registry is that they all have standard methods like get_object[s] or update. there is no way for them to learn that there is get_subnets_by_service_type too. | 14:23 |
| ihar | if we implement it the right away with standard interfaces, plugins will be able to transparently request the needed filters without caring about db table structures. | 14:24 |
| mlavalle | exactly | 14:24 |
| lujinluo | yeah, so we have to add extra filters to get_object() not extra methods | 14:24 |
| ihar | lujinluo: yes I would prefer get_object[s], and I strongly disagree with passing queries as arguments | 14:24 |
| lujinluo | agree | 14:24 |
| ihar | lujinluo: yes, that would be ideal. unless of course we have a good reason for a custom method. (like it's too hard to do it; then we may get back to it later) | 14:25 |
| ihar | I left a comment along these lines | 14:27 |
| lujinluo | i see. i was testing a local patch with 3 tables joins | 14:27 |
| TuanVu | thanks for the comment, Ihar | 14:28 |
| lujinluo | so with that being said, i should revise corresponding get_object(s) | 14:28 |
| ihar | now, I don't disagree that making joins work through standard interfaces is quite an undertaking. | 14:28 |
| *** skazi has joined #openstack-meeting-5 | 14:28 | |
| ihar | but we should try it. maybe once we actually try it, we learn something that would defeat all I just said. | 14:28 |
| ihar | but unless we try on a patch or two, we don't know | 14:28 |
| lujinluo | got it | 14:29 |
| ihar | TuanVu: does it make sense? | 14:29 |
| TuanVu | yes, I agree | 14:29 |
| mlavalle | in other words, let's force ourselvers to find a good reason to change the current API | 14:29 |
| lujinluo | ++ | 14:30 |
| lujinluo | ok, next one https://review.openstack.org/#/c/507772/ | 14:31 |
| ihar | I see rebase helped with InvalidRequestError, that's good. | 14:31 |
| ihar | still a failure with query count though. | 14:32 |
| TuanVu | as I mentioned on Gerrit, there's still problem with "duplicated queries" for "networksecuritybindings" and "externalnetworks" | 14:32 |
| TuanVu | yes | 14:32 |
| ihar | I don't have a good answer except that I can look into it on Monday | 14:33 |
| TuanVu | thank you in advance, Ihar :) | 14:33 |
| lujinluo | i think i have the same issue with port. so i need to double check tmr or later too | 14:34 |
| lujinluo | next is segment https://review.openstack.org/#/c/559414/ | 14:34 |
| lujinluo | hongbin has addressed all the comments from ihar | 14:34 |
| hongbin_ | o/ | 14:35 |
| lujinluo | wow, hi hongbin_ i did not expect you would be here | 14:35 |
| mlavalle | he is everywhere | 14:35 |
| hongbin_ | ^^ | 14:36 |
| lujinluo | LOL | 14:36 |
| mlavalle | omnipresent | 14:36 |
| ihar | ack, +2 | 14:36 |
| ihar | man, this green ci is amazing | 14:36 |
| lujinluo | but i think we need to wait for the fk patch before +w, right? | 14:36 |
| ihar | we can +w whenever we like as long as it's depends-on (which it is) | 14:37 |
| ihar | zuul will not merge it until all dependencies are met | 14:37 |
| lujinluo | understood! | 14:37 |
| lujinluo | next one https://review.openstack.org/#/c/561834/ | 14:37 |
| lujinluo | i think it needs some more work wrt joins | 14:38 |
| TuanVu | yes | 14:38 |
| ihar | yes it's the same issues as discussed above | 14:38 |
| TuanVu | as above discussion | 14:38 |
| lujinluo | ok, the next two patches are mine.. and i apologize for not being able to work on them last week | 14:39 |
| lujinluo | i will respin them before our next meeting | 14:39 |
| ihar | ok. afair we wanted to base port stuff on mlavalle's multiple port bindings rfe implementation | 14:40 |
| mlavalle | yeap | 14:40 |
| lujinluo | yes, but it seems mlavalle 's patch is in merge conflict | 14:41 |
| ihar | so it requires quite some work I suspect | 14:41 |
| mlavalle | lujinluo: I'll fix it | 14:41 |
| lujinluo | thanks, i will rebase once it is ready | 14:41 |
| mlavalle | :-) | 14:41 |
| lujinluo | ok, that's all for ovo | 14:41 |
| ihar | mlavalle: please do! that rfe is one of things I promised to get to done before I move to other places. :) | 14:41 |
| lujinluo | #topic open discussion | 14:41 |
| *** openstack changes topic to "open discussion (Meeting topic: neutron_upgrades)" | 14:41 | |
| mlavalle | ihar: thanks. I think we can merge it soon | 14:42 |
| lujinluo | ihar: ohh, you just reminded us of some heart-breaking stuff | 14:42 |
| lujinluo | tha you are moving to other places ;( | 14:42 |
| lujinluo | that* | 14:42 |
| ihar | to the topic of open discussion, I have nothing specific to discuss | 14:42 |
| mlavalle | I just one to mention one thing | 14:43 |
| lujinluo | yes, please | 14:43 |
| mlavalle | This past Monday I added a topic to the Neutron weekly meeting on OVO backlog | 14:43 |
| mlavalle | #link https://wiki.openstack.org/wiki/Network/Meetings#Blueprints | 14:43 |
| lujinluo | thank you! | 14:44 |
| mlavalle | I mentioned it during the meeting and I will continue doing so every week | 14:44 |
| TuanVu | thank you very much, mlavalle | 14:44 |
| mlavalle | so people can see the backlog and help if they have bandwidth | 14:44 |
| lujinluo | is it why we are having hongbin_ here ? lol | 14:44 |
| hungpv_ | 😂 | 14:44 |
| mlavalle | no, as I said, hongbin_ is omnipresent | 14:44 |
| mlavalle | \it doesn't have to do with any of my efforts | 14:45 |
| lujinluo | LOL, i see. but this is really appreciated anyway | 14:45 |
| TuanVu | yeah | 14:45 |
| TuanVu | it would be great if there's more people working on remaining OVO parts | 14:45 |
| ihar | yeah btw one thing to mention in this venue | 14:46 |
| ihar | I already mentioned it passing above but to make sure everyone is on the same page | 14:46 |
| ihar | we merged boden's patch https://review.openstack.org/#/c/553838/ that adds OVO registry | 14:46 |
| *** markvoelker has quit IRC | 14:47 | |
| ihar | tl;dr projects (incl. neutron) will register their objects and then other subprojects will be able to get classes from this registry by name. | 14:47 |
| ihar | so we get rid of module import dependency between two | 14:47 |
| lujinluo | nice | 14:48 |
| ihar | it's new thing, not released yet, but I believe boden will follow up with adoption once it is | 14:48 |
| TuanVu | thanks for this info, Ihar | 14:48 |
| *** markvoelker has joined #openstack-meeting-5 | 14:48 | |
| lujinluo | need to take closer look to follow | 14:49 |
| mlavalle | we will release neutron-lib next week | 14:49 |
| ihar | this is partially the reason why I am more concerned about custom methods now because we are going big league now, officially exposing ovo as part of neutron interfaces. | 14:49 |
| mlavalle | it will be included in that release | 14:49 |
| ihar | and plugins may start catching up to this interface more now | 14:49 |
| ihar | (so far we had just a bunch that use ovo in any capacity) | 14:50 |
| lujinluo | yeah, i will keep that in mind too. need to consider more | 14:50 |
| ihar | ok that's all I have | 14:50 |
| lujinluo | thanks! | 14:51 |
| mlavalle | o/ | 14:51 |
| lujinluo | dose anyone have anything else? | 14:51 |
| TuanVu | thank you, Ihar :) | 14:51 |
| lujinluo | if not, we can call it a day! | 14:51 |
| ihar | it seems like we can indeed | 14:51 |
| hungpv_ | Good night you ak | 14:51 |
| ihar | o/ | 14:51 |
| lujinluo | #endmeeting | 14:51 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/" | 14:52 | |
| openstack | Meeting ended Thu Apr 19 14:51:59 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:52 |
| ihar | have a nice rest! | 14:52 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/neutron_upgrades/2018/neutron_upgrades.2018-04-19-14.00.html | 14:52 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/neutron_upgrades/2018/neutron_upgrades.2018-04-19-14.00.txt | 14:52 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/neutron_upgrades/2018/neutron_upgrades.2018-04-19-14.00.log.html | 14:52 |
| TuanVu | None from me | 14:52 |
| lujinluo | yeah, i finally got the command right! | 14:52 |
| TuanVu | see you later, mlavalle, Ihar and Luo-san | 14:52 |
| TuanVu | haha, congrats, Luo :D | 14:53 |
| ihar | lujinluo: you get better and better at it every meeting. I should have passed the batton to you all along. | 14:53 |
| ihar | ok ciao everyone, have a good evening (or any other daytime you may have) | 14:53 |
| *** ihar has left #openstack-meeting-5 | 14:53 | |
| mlavalle | TuanVu, lujinluo: good night! | 14:54 |
| lujinluo | mlavalle: thank you! have a good day! | 14:54 |
| *** lujinluo has quit IRC | 14:55 | |
| *** mlavalle has left #openstack-meeting-5 | 15:00 | |
| *** TuanVu has quit IRC | 15:06 | |
| *** yamamoto has quit IRC | 15:13 | |
| *** yamamoto has joined #openstack-meeting-5 | 15:14 | |
| *** yamamoto has quit IRC | 15:19 | |
| *** annp_ has quit IRC | 15:23 | |
| *** yamahata has quit IRC | 15:23 | |
| *** ricolin has joined #openstack-meeting-5 | 15:25 | |
| *** yamamoto has joined #openstack-meeting-5 | 15:34 | |
| *** yamahata has joined #openstack-meeting-5 | 15:56 | |
| *** slaweq_ has quit IRC | 15:59 | |
| *** slaweq has joined #openstack-meeting-5 | 15:59 | |
| *** slaweq has quit IRC | 16:00 | |
| *** slaweq has joined #openstack-meeting-5 | 16:00 | |
| *** felipemonteiro has joined #openstack-meeting-5 | 16:05 | |
| *** slaweq has quit IRC | 16:05 | |
| *** radeks has quit IRC | 16:10 | |
| *** hungpv_ has quit IRC | 16:58 | |
| *** derekh has quit IRC | 17:01 | |
| *** yamamoto has quit IRC | 17:20 | |
| *** felipemonteiro_ has joined #openstack-meeting-5 | 17:55 | |
| *** felipemonteiro has quit IRC | 17:59 | |
| *** slaweq has joined #openstack-meeting-5 | 18:01 | |
| *** slaweq has quit IRC | 18:06 | |
| *** felipemonteiro__ has joined #openstack-meeting-5 | 18:08 | |
| *** felipemonteiro_ has quit IRC | 18:08 | |
| *** ricolin has quit IRC | 18:16 | |
| *** pabelanger has quit IRC | 18:17 | |
| *** pabelanger has joined #openstack-meeting-5 | 18:17 | |
| *** yamamoto has joined #openstack-meeting-5 | 18:20 | |
| *** yamamoto has quit IRC | 18:30 | |
| *** diman has joined #openstack-meeting-5 | 18:37 | |
| *** diman has quit IRC | 18:42 | |
| *** felipemonteiro_ has joined #openstack-meeting-5 | 19:09 | |
| *** felipemonteiro__ has quit IRC | 19:13 | |
| *** slaweq has joined #openstack-meeting-5 | 19:30 | |
| *** slaweq has quit IRC | 19:30 | |
| *** slaweq has joined #openstack-meeting-5 | 19:54 | |
| *** slaweq has quit IRC | 20:54 | |
| *** felipemonteiro_ has quit IRC | 20:54 | |
| *** slaweq has joined #openstack-meeting-5 | 20:54 | |
| *** slaweq has quit IRC | 20:59 | |
| *** yamamoto has joined #openstack-meeting-5 | 21:49 | |
| *** diman has joined #openstack-meeting-5 | 22:38 | |
| *** diman has quit IRC | 22:42 | |
| *** slaweq has joined #openstack-meeting-5 | 22:55 | |
| *** hongbin_ has quit IRC | 22:57 | |
| *** slaweq has quit IRC | 23:00 | |
| *** hoangcx_ has joined #openstack-meeting-5 | 23:16 | |
| *** hoangcx has quit IRC | 23:18 | |
| *** mjturek has quit IRC | 23:20 | |
| *** hoangcx_ is now known as hoangcx | 23:22 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!