| sbalukoff | blogan: Also, some of our tests are dumb and use FAKE_UUID_1 for everything. | 00:00 |
|---|---|---|
| sbalukoff | Hence the reason I'm prepending the model name. | 00:00 |
| blogan | sbalukoff: i know, if i ever got like you know, free time, i'd test it out to see if it'd make all that simpler, but i'd also be afraid it'd cause more intermittent bugs | 00:00 |
| rm_work | well | 00:00 |
| rm_work | UUID != id() | 00:00 |
| rm_work | even if they have the same UUID their id() will be different, since it's a reference to the object in memory | 00:00 |
| sbalukoff | Aah! See-- we don't want to reference the object in memory, the way we're building the data model graph. | 00:01 |
| blogan | lets leave it as is now! | 00:01 |
| rm_work | so as long as nothing regenerates the objects, id() would work | 00:01 |
| sbalukoff | blogan: Agreed. | 00:01 |
| *** manishg has quit IRC | 00:01 | |
| rm_work | but i feel like they may end up regenerating so yeah | 00:01 |
| rm_work | >_> | 00:01 |
| rm_work | i am just curious if it works :P | 00:01 |
| * blogan put the bait out, and rm_work took it | 00:02 | |
| blogan | you could say i am the master of baiting people... | 00:02 |
| sbalukoff | Just to verify: Listener's can't share the same tls_containers? | 00:02 |
| * blogan throws smoke bomb and disappears | 00:02 | |
| blogan | sbalukoff: ah good point, the primary key would be a composite key | 00:03 |
| sbalukoff | blogan: So.... listeners *CAN* share tls_containers? | 00:03 |
| blogan | sbalukoff: i dont see why not, rm_work would have to verify that though | 00:04 |
| sbalukoff | TrevorV: If that's the case, we're going to have to go with the line I gave above-- if it's a combination key, we should create the unique key using both the listener_id and tls_container_id | 00:04 |
| sbalukoff | blogan: It is a valid configuration if the key/cert is a wildcard cert. | 00:04 |
| sbalukoff | blogan: That is to say, you will occasionally see that in production in the real world. I don't know whether Octavia is built to handle that correctly right now. | 00:05 |
| rm_work | they can | 00:05 |
| rm_work | but it doesn't matter? | 00:05 |
| sbalukoff | It does. | 00:06 |
| rm_work | because the only field that matters on the object is the container_id | 00:06 |
| rm_work | soooo if it gets the same one but from a different listener... | 00:06 |
| sbalukoff | the unique key there needs to be unique through *all* connected objects in the graph. | 00:06 |
| rm_work | who cares | 00:06 |
| TrevorV | Yeah sbalukoff good call, I'll update | 00:06 |
| blogan | rm_work: yeah but if two listeners have that same tls_container_id, those are 2 different entries in the db | 00:06 |
| blogan | PRIMARY KEY (`listener_id`,`tls_container_id`), | 00:06 |
| rm_work | yeah but the way it's stored it should still be fine IIRC? | 00:06 |
| blogan | thats the sni table | 00:06 |
| sbalukoff | rm_work: It doesn't hurt to use both IDs to create the unique key. | 00:07 |
| fnaval | madhu_ak: you can continue the progress made on the original TLS scenario test, here: https://review.openstack.org/#/c/164828/ | 00:07 |
| sbalukoff | TrevorV: I'm realizing you'll probably run into another error: | 00:07 |
| sbalukoff | Oh wait! | 00:08 |
| madhu_ak | sounds good. Stll question remains for me. Are we going to rely on tempest files in our tree now? ;) | 00:08 |
| sbalukoff | I'm wrong. | 00:08 |
| madhu_ak | fnaval, ^^ | 00:08 |
| sbalukoff | Also, I think rm_work is right. However, still-- it doesn't hurt to use the combination of both ids. | 00:08 |
| rm_work | yeah a composite key would be "fully correct" | 00:08 |
| sbalukoff | So, yes, please do that, TrevorV. | 00:08 |
| rm_work | but functionally meh :P | 00:08 |
| rm_work | he's doing it | 00:08 |
| TrevorV | It also doesn't hurt me to do that in the review, just have to add some chars and we gucci | 00:09 |
| sbalukoff | Yep, let's go with the composite key. | 00:09 |
| sbalukoff | Sweet. | 00:09 |
| fnaval | madhu_ak: for Neutron-LBaaS? | 00:09 |
| fnaval | or Octavia? | 00:09 |
| fnaval | madhu_ak: if we get the scenario test for Octavia merged then we can use that as a reference to refactor neutron-lbaas tempest tests | 00:10 |
| *** diogogmt has quit IRC | 00:10 | |
| madhu_ak | are we going to implement TLS scenario tests for lbaas or octavia? if in lbaas, it is going to be the same way like it is written? | 00:10 |
| fnaval | I think the more tests the better | 00:10 |
| rm_work | yes please | 00:10 |
| rm_work | all the tests | 00:10 |
| sbalukoff | madhu_ak: I think we want them, yes! | 00:11 |
| fnaval | if we can have that TLS test for both octavia and lbaas | 00:11 |
| fnaval | then that would be good | 00:11 |
| rm_work | though, we aren't on a strict deadline for merging the testing since it literally does not matter whether we get it in any particular release | 00:11 |
| rm_work | so I'm not prioritizing reviewing it right now, sorry :( | 00:11 |
| rm_work | after this feature freeze on Monday maybe | 00:11 |
| sbalukoff | Yeah, I apologize, but I'm going to be working on neutron-lbaas L7 mostly in the next few days. :/ | 00:11 |
| sbalukoff | Or critical bugs. | 00:12 |
| TrevorV | sbalukoff same for me, with single-create.... | 00:12 |
| rm_work | yes | 00:12 |
| sbalukoff | Yep. | 00:12 |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Use unique SNI identifier when building data model https://review.openstack.org/284464 | 00:12 |
| TrevorV | change is goin up | 00:12 |
| TrevorV | there tis | 00:12 |
| sbalukoff | Ok, looking now... | 00:12 |
| TrevorV | Alrighty | 00:12 |
| TrevorV | I'll wait a sec in case you see something else | 00:12 |
| madhu_ak | okay, my question is, if we work on TLS scenario tests (this one: https://review.openstack.org/#/c/164828/), then if this tempest refactoring (https://review.openstack.org/#/c/273817/) merges sooner, then thats going to be a problem? | 00:13 |
| sbalukoff | Ok, that's +2'ed from my perspective. | 00:13 |
| *** klindgren_ has joined #openstack-lbaas | 00:13 | |
| rm_work | yeah i could +2 | 00:13 |
| fnaval | madhu_ak: there might have been some changes in the way that the calls are being made to barbican; but the TLS test itself is similar in both; the only difference is how we call octavia/neutron-lbaas that's different. but you don't have to worrry about that since it's all being done in the base class/clients | 00:13 |
| rm_work | I almost wish there was an underscore or something between the ids, but it literally does not matter | 00:14 |
| rm_work | it's just for tracking uniqueness | 00:14 |
| madhu_ak | okay, will work on https://review.openstack.org/#/c/164828/ then | 00:14 |
| sbalukoff | rm_work: Yep. | 00:14 |
| fnaval | madhu_ak: for your https://review.openstack.org/#/c/164828/ most of the changes will be in base.py | 00:14 |
| fnaval | the refactor will occur but should not affect the test itself. | 00:14 |
| *** klindgren has quit IRC | 00:15 | |
| fnaval | madhu_ak: but i understand what you're getting at, it's be good to have that base class ready so that merge conflicts and other bad stuff can be avoided. | 00:16 |
| madhu_ak | hmm, gotcha. | 00:16 |
| fnaval | for example, I'd like to see these TLS methods inside of base or a helper class: https://review.openstack.org/#/c/164828/17/neutron_lbaas/tests/tempest/v2/scenario/test_load_balancer_tls.py | 00:17 |
| fnaval | maybe that can be done now and not even have to rely on base.py | 00:17 |
| fnaval | create a helper class, such as: tls_connection_helper | 00:17 |
| sbalukoff | rm_work or blogan: If you got a minute, this is a critical bugfix: https://review.openstack.org/#/c/284008 | 00:17 |
| *** yamamoto_ has joined #openstack-lbaas | 00:17 | |
| fnaval | that way we can reuse in more TLS type tests | 00:17 |
| madhu_ak | yeah, without relyaing on base.py is the good idea | 00:18 |
| *** TrevorV has quit IRC | 00:19 | |
| *** crc32 has quit IRC | 00:19 | |
| blogan | sbalukoff: reviewing | 00:20 |
| sbalukoff | Danke! | 00:20 |
| *** bana_k has joined #openstack-lbaas | 00:21 | |
| rm_work | trevorv says bye | 00:21 |
| rm_work | looking sbalukoff | 00:21 |
| sbalukoff | Cool beans! | 00:22 |
| blogan | sbalukoff: one possible bad suggestion for a future patch, _get_affected_listener_ids consolidated into the base class as well | 00:23 |
| sbalukoff | blogan: That and the _test_and_set_listener_lb_statuses | 00:23 |
| sbalukoff | Yeah, there's definitely some optimization we can do there. | 00:24 |
| blogan | yep i was just typing that | 00:24 |
| fnaval | thanks madhu_ak - ping me anytime for any assistance | 00:24 |
| blogan | with the to_data_model method becoming this huge recursive method, i worry about making generic methods that work for all types bc it can get hard to read and maintain | 00:25 |
| madhu_ak | fnaval, sure | 00:25 |
| xgerman | blogan — write your own ORM | 00:26 |
| blogan | xgerman: thats what sbalukoff almost did | 00:27 |
| blogan | well he and i | 00:27 |
| sbalukoff | Heh! | 00:27 |
| sbalukoff | Basically, yes. | 00:27 |
| rm_work | i enjoy the new static method | 00:27 |
| blogan | actually we more or less wrote the mapper | 00:27 |
| * blogan glares at rm_work | 00:27 | |
| xgerman | if there would only some oslo arm? | 00:27 |
| xgerman | orm | 00:27 |
| blogan | remember when staticmethods broke that one thing? | 00:27 |
| rm_work | yeah that one time in that one place | 00:28 |
| sbalukoff | Haha | 00:28 |
| blogan | i've got that review bookmarked | 00:28 |
| rm_work | heh | 00:28 |
| sbalukoff | Seriously? I'm not sure which review you're talking about. | 00:28 |
| * blogan looks it up excitedly | 00:29 | |
| blogan | https://review.openstack.org/#/c/277544/ | 00:30 |
| blogan | yeah! | 00:30 |
| blogan | sbalukoff: my argument against staticmethods is to just use a classmethod if you want it namespaced to the class, otherwise just use a module method | 00:31 |
| blogan | no need for a staticmethod | 00:31 |
| openstackgerrit | James Arendt proposed openstack/neutron-lbaas: Neutron LBaaS leaking admin_password into logs https://review.openstack.org/284483 | 00:31 |
| sbalukoff | Aah. | 00:31 |
| *** Purandar has quit IRC | 00:46 | |
| *** manishg has joined #openstack-lbaas | 00:48 | |
| *** Aish has left #openstack-lbaas | 00:57 | |
| *** ajmiller has quit IRC | 01:02 | |
| *** manishg has quit IRC | 01:03 | |
| *** manishg has joined #openstack-lbaas | 01:03 | |
| *** paco20151113 has joined #openstack-lbaas | 01:04 | |
| openstackgerrit | Merged openstack/octavia: Fix LB/Listener status updates for HM/Member https://review.openstack.org/284008 | 01:05 |
| *** paco20151113 has quit IRC | 01:06 | |
| *** minwang2 has quit IRC | 01:08 | |
| *** bana_k has quit IRC | 01:08 | |
| *** ducttape_ has joined #openstack-lbaas | 01:20 | |
| *** ducttape_ has quit IRC | 01:30 | |
| *** ducttape_ has joined #openstack-lbaas | 02:03 | |
| *** Purandar has joined #openstack-lbaas | 02:03 | |
| *** manishg_ has joined #openstack-lbaas | 02:04 | |
| *** diogogmt has joined #openstack-lbaas | 02:05 | |
| *** manishg has quit IRC | 02:05 | |
| *** kevo has quit IRC | 02:08 | |
| openstackgerrit | Merged openstack/octavia: Implements: blueprint anti-affinity server group https://review.openstack.org/272344 | 02:08 |
| openstackgerrit | Merged openstack/octavia: Use unique SNI identifier when building data model https://review.openstack.org/284464 | 02:10 |
| openstackgerrit | Merged openstack/octavia: Stop using non-existent method of Mock https://review.openstack.org/280766 | 02:10 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Adding debug logging to amphora driver https://review.openstack.org/284498 | 02:14 |
| *** Bjoern_ has joined #openstack-lbaas | 02:27 | |
| *** armax has joined #openstack-lbaas | 02:28 | |
| *** amotoki has joined #openstack-lbaas | 02:29 | |
| *** ducttape_ has quit IRC | 02:43 | |
| *** yamamoto_ has quit IRC | 02:43 | |
| *** ducttape_ has joined #openstack-lbaas | 02:43 | |
| *** links has joined #openstack-lbaas | 02:47 | |
| *** manishg has joined #openstack-lbaas | 02:54 | |
| *** manishg_ has quit IRC | 02:55 | |
| *** ajmiller has joined #openstack-lbaas | 02:57 | |
| *** manishg has quit IRC | 02:59 | |
| *** ducttape_ has quit IRC | 03:06 | |
| *** ducttape_ has joined #openstack-lbaas | 03:12 | |
| *** yuanying has quit IRC | 03:16 | |
| *** amotoki has quit IRC | 03:19 | |
| *** diogogmt has quit IRC | 03:23 | |
| *** diogogmt has joined #openstack-lbaas | 03:26 | |
| *** ducttape_ has quit IRC | 03:26 | |
| *** amotoki has joined #openstack-lbaas | 03:30 | |
| *** manishg has joined #openstack-lbaas | 03:30 | |
| *** links has quit IRC | 03:34 | |
| *** manishg has quit IRC | 03:35 | |
| *** amotoki has quit IRC | 03:35 | |
| *** ajmiller has quit IRC | 03:38 | |
| *** yamamoto_ has joined #openstack-lbaas | 03:38 | |
| *** amotoki has joined #openstack-lbaas | 03:40 | |
| *** ducttape_ has joined #openstack-lbaas | 03:41 | |
| *** diogogmt has quit IRC | 03:43 | |
| *** neelashah has joined #openstack-lbaas | 03:45 | |
| *** chlong_ has quit IRC | 03:47 | |
| *** chlong_ has joined #openstack-lbaas | 03:49 | |
| *** Bjoern_ has quit IRC | 03:51 | |
| *** ducttape_ has quit IRC | 03:53 | |
| *** johnsom_ has joined #openstack-lbaas | 03:55 | |
| *** ducttape_ has joined #openstack-lbaas | 03:56 | |
| *** woodster_ has quit IRC | 03:56 | |
| *** manishg has joined #openstack-lbaas | 04:00 | |
| *** amotoki has quit IRC | 04:01 | |
| *** manishg has quit IRC | 04:05 | |
| *** yuanying has joined #openstack-lbaas | 04:07 | |
| *** amotoki has joined #openstack-lbaas | 04:13 | |
| openstackgerrit | Paco Peng proposed openstack/octavia: Fixed make sure to get IPv4 value by awk in sample local.sh file https://review.openstack.org/284527 | 04:14 |
| openstackgerrit | Paco Peng proposed openstack/octavia: Fixed make sure to get IPv4 value by awk in sample local.sh file https://review.openstack.org/283929 | 04:19 |
| *** ducttape_ has quit IRC | 04:24 | |
| *** links has joined #openstack-lbaas | 04:29 | |
| *** ducttape_ has joined #openstack-lbaas | 04:29 | |
| openstackgerrit | Paco Peng proposed openstack/octavia: Fixed make sure to get IPv4 value by awk in sample local.sh file https://review.openstack.org/283929 | 04:31 |
| *** ducttape_ has quit IRC | 04:37 | |
| *** neelashah has quit IRC | 04:38 | |
| openstackgerrit | Michael Johnson proposed openstack/neutron-lbaas: Temp scenario test logging to debug gate issue https://review.openstack.org/284533 | 04:43 |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Adding debug logging to amphora driver https://review.openstack.org/284498 | 04:48 |
| *** armax has quit IRC | 04:51 | |
| *** manishg has joined #openstack-lbaas | 05:01 | |
| *** manishg_ has joined #openstack-lbaas | 05:03 | |
| *** manishg has quit IRC | 05:04 | |
| johnsom_ | Great, gates are broken again.... | 05:20 |
| johnsom_ | sbalukoff be extra happy right now | 05:20 |
| sbalukoff | Heh! | 05:20 |
| sbalukoff | Because all the L7 stuff is through? | 05:20 |
| sbalukoff | (As well as a good dozen follow-up patches?) | 05:20 |
| sbalukoff | Oi... yeah, that's not good. | 05:21 |
| johnsom_ | Yeah. I put in some logging patches to try to run the session persistence to ground and both the lbaas and octavia patches just errored out | 05:22 |
| johnsom_ | Pretty sure my LOG.debug didn't break them... | 05:22 |
| sbalukoff | yeah, time to dig in and figure out what did. :/ | 05:23 |
| sbalukoff | ERROR! Specified hosts options do not match any hosts | 05:25 |
| sbalukoff | I don't think that's normal... | 05:25 |
| johnsom_ | Yeah, not normal | 05:26 |
| johnsom_ | The really odd thing is the devstack scripts seem to have completed | 05:26 |
| sbalukoff | yeah, I saw that too. | 05:27 |
| sbalukoff | Octavia even gets started correctly, though we're not even getting one API command... | 05:33 |
| sbalukoff | So we're failing before we can even start a tempest run, but apparently after the build complete. | 05:33 |
| sbalukoff | completes. | 05:33 |
| sbalukoff | Which makes sense, given that it takes 15-20 minutes to fail. | 05:33 |
| johnsom_ | Yeah, so strange | 05:36 |
| *** manishg_ has quit IRC | 05:37 | |
| sbalukoff | I'm seeing a shit-ton of other failures happening on other projects in the check queue. | 05:38 |
| sbalukoff | So I doubt it's only us. | 05:38 |
| johnsom_ | Well, I think it is time to call it a night and let infra figure this out | 05:38 |
| johnsom_ | Yeah, It hit lbaas and octavia at the same time, so not likely us | 05:38 |
| sbalukoff | Ok. | 05:39 |
| sbalukoff | Well, comcast shut me down last night... | 05:39 |
| johnsom_ | Have a good one and keep the grin that your stuff got in last night and didn't get tied up in this mess | 05:39 |
| sbalukoff | I think I still might have enough energy to pump out another revision on the neutron-lbaas l7 patch... | 05:39 |
| sbalukoff | So, I'll probably be on for a while yet. | 05:39 |
| sbalukoff | Have a good one! | 05:39 |
| johnsom_ | Ok, good luck. Catch you tomorrow | 05:40 |
| sbalukoff | Yep! | 05:40 |
| *** johnsom_ has quit IRC | 05:40 | |
| *** numans has joined #openstack-lbaas | 05:46 | |
| *** yuanying has quit IRC | 05:47 | |
| *** numan_ has joined #openstack-lbaas | 05:50 | |
| *** numan_ has quit IRC | 05:51 | |
| *** numan_ has joined #openstack-lbaas | 05:53 | |
| *** numans has quit IRC | 05:54 | |
| *** numan_ has quit IRC | 05:56 | |
| *** numans has joined #openstack-lbaas | 05:56 | |
| *** doug-fish has joined #openstack-lbaas | 06:00 | |
| *** numans has quit IRC | 06:01 | |
| *** bana_k has joined #openstack-lbaas | 06:03 | |
| *** manishg has joined #openstack-lbaas | 06:08 | |
| *** manishg has quit IRC | 06:12 | |
| *** numans has joined #openstack-lbaas | 06:13 | |
| *** Purandar has quit IRC | 06:27 | |
| *** bana_k has quit IRC | 06:42 | |
| *** minwang2 has joined #openstack-lbaas | 06:45 | |
| *** fawadkhaliq has joined #openstack-lbaas | 06:50 | |
| *** prabampm has joined #openstack-lbaas | 06:50 | |
| *** fnaval has quit IRC | 06:54 | |
| *** doug-fish has quit IRC | 06:54 | |
| *** yuanying has joined #openstack-lbaas | 06:54 | |
| *** fnaval has joined #openstack-lbaas | 07:02 | |
| *** manishg has joined #openstack-lbaas | 07:09 | |
| *** manishg has quit IRC | 07:14 | |
| *** kobis has joined #openstack-lbaas | 07:16 | |
| *** kobis has quit IRC | 07:18 | |
| *** minwang2 has quit IRC | 07:29 | |
| *** nmagnezi has joined #openstack-lbaas | 07:46 | |
| *** kobis has joined #openstack-lbaas | 07:49 | |
| *** fawadkhaliq has quit IRC | 07:51 | |
| *** yuanying has quit IRC | 08:01 | |
| *** kevo has joined #openstack-lbaas | 08:03 | |
| *** manishg has joined #openstack-lbaas | 08:10 | |
| *** yuanying has joined #openstack-lbaas | 08:11 | |
| *** evgenyf has joined #openstack-lbaas | 08:14 | |
| *** manishg has quit IRC | 08:14 | |
| *** yuanying has quit IRC | 08:16 | |
| *** yuanying has joined #openstack-lbaas | 08:18 | |
| *** Oku_OS has quit IRC | 08:18 | |
| *** yuanying has quit IRC | 08:26 | |
| *** johnsom_ has joined #openstack-lbaas | 08:28 | |
| *** johnsom has quit IRC | 08:30 | |
| *** yuanying has joined #openstack-lbaas | 08:32 | |
| *** kevo has quit IRC | 08:37 | |
| *** yuanying has quit IRC | 08:38 | |
| *** yuanying has joined #openstack-lbaas | 08:41 | |
| *** yuanying has quit IRC | 08:47 | |
| *** ihrachys has joined #openstack-lbaas | 08:51 | |
| *** jschwarz has joined #openstack-lbaas | 08:52 | |
| *** amotoki has quit IRC | 08:53 | |
| openstackgerrit | Stephen Balukoff proposed openstack/neutron-lbaas: Fix shared-pools lbaas_listeners migration https://review.openstack.org/283802 | 08:56 |
| *** amotoki has joined #openstack-lbaas | 09:10 | |
| *** Oku_OS has joined #openstack-lbaas | 09:10 | |
| *** manishg has joined #openstack-lbaas | 09:11 | |
| *** manishg has quit IRC | 09:15 | |
| *** fawadkhaliq has joined #openstack-lbaas | 09:17 | |
| *** [1]evgenyf has joined #openstack-lbaas | 09:19 | |
| *** evgenyf has quit IRC | 09:21 | |
| *** [1]evgenyf is now known as evgenyf | 09:21 | |
| *** fawadkhaliq has quit IRC | 09:29 | |
| *** paco20151113 has joined #openstack-lbaas | 09:53 | |
| sbalukoff | Hey paco20151113! | 10:02 |
| sbalukoff | It's just me on right now... and I'm about to head to bed anyway. | 10:02 |
| sbalukoff | but it's good to see you here! | 10:02 |
| *** manishg has joined #openstack-lbaas | 10:11 | |
| *** manishg has quit IRC | 10:16 | |
| *** pcaruana has joined #openstack-lbaas | 10:42 | |
| *** manishg has joined #openstack-lbaas | 11:12 | |
| *** manishg has quit IRC | 11:16 | |
| *** fawadkhaliq has joined #openstack-lbaas | 11:27 | |
| *** yamamoto_ has quit IRC | 12:07 | |
| *** manishg has joined #openstack-lbaas | 12:13 | |
| *** manishg has quit IRC | 12:18 | |
| *** rtheis has joined #openstack-lbaas | 12:42 | |
| *** fawadkhaliq has quit IRC | 12:42 | |
| *** fawadkhaliq has joined #openstack-lbaas | 12:43 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/octavia: WIP: glance: support relying on tags to extract image id https://review.openstack.org/284690 | 12:44 |
| *** diogogmt has joined #openstack-lbaas | 12:46 | |
| *** fawadkhaliq has quit IRC | 12:47 | |
| *** woodster_ has joined #openstack-lbaas | 12:52 | |
| *** pcaruana has quit IRC | 12:55 | |
| *** yamamoto has joined #openstack-lbaas | 12:55 | |
| *** yamamoto_ has joined #openstack-lbaas | 13:01 | |
| *** yamamoto has quit IRC | 13:01 | |
| *** pcaruana has joined #openstack-lbaas | 13:09 | |
| *** diogogmt has quit IRC | 13:10 | |
| *** neelashah has joined #openstack-lbaas | 13:10 | |
| *** manishg has joined #openstack-lbaas | 13:14 | |
| *** ducttape_ has joined #openstack-lbaas | 13:17 | |
| *** manishg has quit IRC | 13:18 | |
| *** links has quit IRC | 13:26 | |
| *** ducttape_ has quit IRC | 13:36 | |
| *** rtheis has quit IRC | 13:36 | |
| *** rtheis has joined #openstack-lbaas | 13:42 | |
| *** rtheis has quit IRC | 13:44 | |
| *** rtheis has joined #openstack-lbaas | 13:45 | |
| *** armax has joined #openstack-lbaas | 13:47 | |
| *** fawadkhaliq has joined #openstack-lbaas | 13:53 | |
| openstackgerrit | Kobi Samoray proposed openstack/octavia: [WIP] Implement XFF header support for Octavia https://review.openstack.org/257901 | 14:04 |
| openstackgerrit | Kobi Samoray proposed openstack/octavia: [WIP] Implement custom header support for Octavia https://review.openstack.org/257901 | 14:05 |
| *** links has joined #openstack-lbaas | 14:07 | |
| *** evgenyf has quit IRC | 14:08 | |
| rm_work | and today I *woke up* at 6:30am, wtf | 14:10 |
| *** yamamoto_ has quit IRC | 14:14 | |
| *** manishg has joined #openstack-lbaas | 14:14 | |
| amotoki | hi LBaaS team, are there any prioritized neutronclient reviews? | 14:14 |
| amotoki | I plan to push a new release of neutronclient next week. | 14:15 |
| *** manishg has quit IRC | 14:18 | |
| *** fawadkhaliq has quit IRC | 14:20 | |
| rm_work | amotoki: yeah there is one | 14:24 |
| rm_work | let me find it | 14:24 |
| rm_work | amotoki: it looks like the only one I'm aware of is https://review.openstack.org/#/c/217276/ which you have already commented on | 14:25 |
| amotoki | rm_work: thanks. | 14:26 |
| amotoki | rm_work: there are some lbaas related patches. thanks for clarification :) | 14:26 |
| rm_work | We would like to get that in the next release, so hopefully we can finish it before Monday... | 14:26 |
| rm_work | There may be others, but that is the one that is most important | 14:27 |
| amotoki | rm_work: even if we missed it in a next release, we can release more (unless we keep backward-compat) | 14:27 |
| rm_work | ok | 14:27 |
| rm_work | we would like to ship it along with the L7 feature in Mitaka | 14:28 |
| rm_work | but clients are not in the same release cycle maybe? | 14:28 |
| amotoki | agree and nice feature | 14:28 |
| amotoki | client is released independently, but there is a dependency freeze for client | 14:28 |
| amotoki | it happens at M-3. | 14:28 |
| rm_work | ok | 14:29 |
| amotoki | http://releases.openstack.org/mitaka/schedule.html | 14:29 |
| rm_work | Monday | 14:29 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/octavia: WIP: glance: support relying on tags to extract image id https://review.openstack.org/284690 | 14:34 |
| ihrachys | sbalukoff: nmagnezi: dougwig: ^ | 14:34 |
| ihrachys | that's the implementation part, need to write some tests too | 14:34 |
| ihrachys | tested locally, seems to work | 14:34 |
| *** amotoki has quit IRC | 14:37 | |
| *** jschwarz_ has joined #openstack-lbaas | 14:37 | |
| rm_work | taking a look ihrachys | 14:37 |
| ihrachys | rm_work: thanks | 14:38 |
| *** jschwarz has quit IRC | 14:38 | |
| *** yamamoto has joined #openstack-lbaas | 14:39 | |
| rm_work | ihrachys: posted some comments | 14:40 |
| ihrachys | looking | 14:40 |
| *** localloop127 has joined #openstack-lbaas | 14:40 | |
| rm_work | to clarify: I know _() handles translation | 14:40 |
| ihrachys | oh octavia already has session handling code nice! | 14:41 |
| rm_work | but not how it's treated by default compared to _LI/_LE/_LW | 14:41 |
| rm_work | yeah | 14:41 |
| ihrachys | the patch is doomed to be a one-liner ;) | 14:41 |
| rm_work | heh | 14:41 |
| ihrachys | rm_work: ack on i18n. I copy pasted it from another exception in the file. I will take a look more closely what to use there. | 14:41 |
| rm_work | hmmm | 14:42 |
| rm_work | you are right, expanding the review in that file i see a lot of those | 14:42 |
| rm_work | some of which I know for sure are my fault | 14:43 |
| ihrachys | yeah, but maybe there are some new guidelines and we should use _LW there | 14:43 |
| rm_work | which makes me wonder wtf i was doing | 14:43 |
| ihrachys | :) | 14:43 |
| rm_work | since I seriously don't know what that behavior will be | 14:43 |
| ihrachys | I am new to the code base, so I can't tell wtf you were doing ;) | 14:43 |
| rm_work | that was more of a question really, I wonder if default behavior is _() == _LI() or something | 14:44 |
| ihrachys | I don't think so | 14:44 |
| ihrachys | _ is the 'main' domain, at least in neutron | 14:44 |
| rm_work | hmm | 14:44 |
| ihrachys | and then you have additional domains per level | 14:44 |
| rm_work | maybe it's fine | 14:44 |
| ihrachys | prolly _LE domain is more appropriate | 14:44 |
| ihrachys | since it's an error | 14:44 |
| *** armax has quit IRC | 14:44 | |
| ihrachys | so if translators want to translate errors only, they get it in their .po file | 14:45 |
| rm_work | yeah it's been a while since i dealt with i18n | 14:45 |
| rm_work | (since when i was working on very small projects) | 14:45 |
| ihrachys | sometimes I wonder whether we actually have enough translations to care ;) | 14:45 |
| rm_work | now we have whole translation teams | 14:45 |
| rm_work | I just know it is good practice :P | 14:46 |
| *** TrevorV has joined #openstack-lbaas | 14:50 | |
| rm_work | morning thomasem | 14:52 |
| rm_work | err | 14:52 |
| rm_work | TrevorV: | 14:52 |
| thomasem | Hello anyway! | 14:52 |
| TrevorV | rm_work go to bed | 14:52 |
| rm_work | :P | 14:52 |
| rm_work | TrevorV: woke up at 6:30am because wut | 14:53 |
| TrevorV | The question isn't when did you wake up, its when did you fall asleep | 14:53 |
| thomasem | Adrenaline to begin the hangover, possibly? | 14:53 |
| thomasem | :D | 14:53 |
| rm_work | 3 hours of sleep per night for about two weeks has finally caught up to me | 14:53 |
| rm_work | crashed out at like midnight | 14:53 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/octavia: WIP: glance: support relying on tags to extract image id https://review.openstack.org/284690 | 14:53 |
| rm_work | my body has no idea wtf is happening | 14:54 |
| TrevorV | That's a common state for you in general I think rm_work | 14:56 |
| rm_work | ihrachys: see this is where I say "I wish those were staticmethods on BaseComputeTask and blogan says "dear god no, they are perfect as they are" :P | 14:56 |
| rm_work | as much as I'd like to make blogan suffer, I think it's fine | 14:56 |
| *** woodster_ has quit IRC | 14:56 | |
| ihrachys | I hate when people use classes where a module is just fine :P | 14:57 |
| rm_work | heh | 14:57 |
| rm_work | I just don't like having anything floating around in my modules | 14:57 |
| rm_work | that isn't a class | 14:57 |
| rm_work | probably a Java carry-over | 14:57 |
| ihrachys | prolly :) | 14:58 |
| rm_work | did you actually have a launchpad BP? | 14:58 |
| ihrachys | rm_work: nope, I need to post one | 14:58 |
| rm_work | the boilerplate in your spec file is still there for that | 14:59 |
| *** diogogmt has joined #openstack-lbaas | 14:59 | |
| rm_work | k | 14:59 |
| rm_work | I ... don't know if we would have really required any of that | 14:59 |
| rm_work | but if you've already done it, may as well finish | 14:59 |
| ihrachys | rm_work: yesterday I was told we need a spec | 14:59 |
| rm_work | heh | 14:59 |
| ihrachys | rm_work: and I immediately started to feel like home | 14:59 |
| ihrachys | :) | 14:59 |
| rm_work | well, *I* would have +2'd without one, but maybe people are starting to ask for that | 14:59 |
| ihrachys | neutron attitude! | 14:59 |
| rm_work | yeah probably dougwig told you that | 14:59 |
| *** manishg has joined #openstack-lbaas | 15:00 | |
| ihrachys | rm_work: honestly, I didn't think the code will be THAT tiny | 15:00 |
| rm_work | the spec is 10x longer than the code | 15:00 |
| rm_work | lol | 15:00 |
| ihrachys | ah missed the reno entry | 15:01 |
| rm_work | ok, 4-5 times | 15:01 |
| ihrachys | wait, I will write some tests and it will be 'just' 2 times :) | 15:01 |
| rm_work | and i do not need to read this spec to understand what this is for, which means ... the spec is totally redundant, but whatever | 15:02 |
| rm_work | we have a few different opinions on specs in this team | 15:02 |
| ihrachys | as long as you don't have an lbaas stadium... | 15:02 |
| ihrachys | oh wait... :) | 15:02 |
| *** ducttape_ has joined #openstack-lbaas | 15:03 | |
| *** yamamoto has quit IRC | 15:04 | |
| dougwig | uh oh, what did i do? | 15:04 |
| dougwig | i think sbalukoff wanted a spec. i'm fine with code. | 15:04 |
| rm_work | ihrachys: one more comment | 15:06 |
| rm_work | ah so it's sbalukoff's fault :P | 15:06 |
| ihrachys | rm_work: shoot | 15:06 |
| rm_work | posted | 15:06 |
| ihrachys | oh right sorry | 15:06 |
| rm_work | :P | 15:06 |
| *** yamamoto has joined #openstack-lbaas | 15:06 | |
| ihrachys | oh you are absolutely right | 15:06 |
| ihrachys | I am a dirty coder :P | 15:07 |
| rm_work | heh | 15:07 |
| *** yamamoto has quit IRC | 15:07 | |
| TrevorV | rm_work did you get a chance to look at single-create at all yesterday/this morning? | 15:09 |
| TrevorV | A lot of changes merged last night that I'm now rebasing on top of, but it shouldn't have affected my review | 15:09 |
| TrevorV | However I am about to push up the rebased patch | 15:10 |
| TrevorV | Just sayin | 15:10 |
| rm_work | ah thought you were still working on bits | 15:11 |
| rm_work | i can look today | 15:11 |
| *** links has quit IRC | 15:12 | |
| ihrachys | dougwig: agreed on warning instead of error. ideally, glance would have unique tags | 15:13 |
| *** numans has quit IRC | 15:13 | |
| *** jschwarz__ has joined #openstack-lbaas | 15:13 | |
| *** jschwarz_ has quit IRC | 15:16 | |
| rm_work | yeah I guess theoretically worst case is you have two, one that is already currently building and one that's new, so you might just get a slight delay on propogation of the new image | 15:18 |
| rm_work | which ... is way better than a user error, yes | 15:18 |
| *** Purandar has joined #openstack-lbaas | 15:19 | |
| rm_work | or you might just get the new one and be good :P | 15:19 |
| *** kobis has quit IRC | 15:19 | |
| *** ajmiller has joined #openstack-lbaas | 15:20 | |
| *** woodster_ has joined #openstack-lbaas | 15:25 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Get Me A Load Balancer Controller https://review.openstack.org/257013 | 15:26 |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Get me a Load Balancer API https://review.openstack.org/256974 | 15:26 |
| *** rtheis has quit IRC | 15:28 | |
| *** rtheis has joined #openstack-lbaas | 15:29 | |
| *** rtheis has quit IRC | 15:31 | |
| *** rtheis has joined #openstack-lbaas | 15:32 | |
| *** armax has joined #openstack-lbaas | 15:33 | |
| *** rtheis has quit IRC | 15:35 | |
| *** rtheis has joined #openstack-lbaas | 15:37 | |
| *** Purandar has quit IRC | 15:38 | |
| *** bana_k has joined #openstack-lbaas | 15:41 | |
| *** paco20151113 has quit IRC | 15:50 | |
| *** manishg has quit IRC | 15:53 | |
| *** nmagnezi has quit IRC | 16:00 | |
| *** jschwarz__ is now known as jschwarz | 16:04 | |
| *** yamamoto has joined #openstack-lbaas | 16:08 | |
| *** bana_k has quit IRC | 16:08 | |
| *** diogogmt has quit IRC | 16:08 | |
| *** yamamoto has quit IRC | 16:13 | |
| *** Purandar has joined #openstack-lbaas | 16:13 | |
| *** fnaval has quit IRC | 16:16 | |
| *** kobis has joined #openstack-lbaas | 16:22 | |
| *** Purandar has quit IRC | 16:24 | |
| *** rtheis_ has joined #openstack-lbaas | 16:25 | |
| *** rtheis has quit IRC | 16:25 | |
| *** fnaval has joined #openstack-lbaas | 16:28 | |
| johnsom_ | ihrachys lol, that was way simpler than I thought it would be | 16:29 |
| ihrachys | johnsom_: indeed | 16:29 |
| *** johnsom_ is now known as johnsom | 16:30 | |
| * ihrachys is spinning on unit tests, stay tight while I respin | 16:30 | |
| *** Bjoern_ has joined #openstack-lbaas | 16:35 | |
| *** Bjoern_ is now known as Bjoern_zZzZzZzZ | 16:35 | |
| *** Purandar has joined #openstack-lbaas | 16:35 | |
| TrevorV | xgerman you online yet? | 16:35 |
| xgerman | yep | 16:36 |
| TrevorV | I have a question... | 16:36 |
| xgerman | sure | 16:36 |
| TrevorV | In the "rest_api_driver" why do we pass in "vip | 16:36 |
| TrevorV | to the update method? | 16:36 |
| TrevorV | In fact, why do we pass in vip to a lot of these methods? | 16:36 |
| TrevorV | Its never really used. | 16:36 |
| johnsom | Yeah, I saw that recently too. It looks like just to log | 16:36 |
| TrevorV | johnsom but the logging line says | 16:37 |
| johnsom | Oh, maybe that was listener-update | 16:37 |
| TrevorV | "updating listener %s, vip %s" but the vip isn't actually being updated... | 16:37 |
| *** Bjoern_zZzZzZzZ is now known as Bjoern_ | 16:37 | |
| xgerman | oh, the spec called for avip | 16:38 |
| TrevorV | Oh. Does another driver use vip informations or something? | 16:38 |
| TrevorV | Maybe that's why | 16:38 |
| TrevorV | (though we don't have another driver right now, right? :P ) | 16:39 |
| xgerman | I had the same concern those days and blogan/dougwig said we needed the VIP | 16:39 |
| xgerman | maybe the A10 VM driver will need it | 16:39 |
| TrevorV | Gotcha, maybe it will. | 16:40 |
| TrevorV | Its totally unused in the rest driver though, and looks a little silly :P But it also wouldn't be priority to me to remove it if it wasn't supposed to be there anymore, ya know? | 16:40 |
| TrevorV | I'm more confused about my single-create in octavia | 16:40 |
| TrevorV | It tosses an error when updating the listeners in the rest driver | 16:41 |
| xgerman | well, it’s definitely for othger rivers | 16:41 |
| TrevorV | I'm going to enable debugging and figure it out | 16:41 |
| dougwig | the lb has the ip, so in effect that is the vip. what is the vip you're passing in? | 16:41 |
| TrevorV | dougwig its the vip from the LB | 16:43 |
| dougwig | can't i just reference the lb from the listener? | 16:43 |
| TrevorV | Yes, yes you can. | 16:43 |
| TrevorV | If all the proper references are made on the object, yes. | 16:43 |
| TrevorV | (otherwise we gots bugs) | 16:43 |
| xgerman | yeah, it was in the spec for some reason but we don’t need it for the rest driver | 16:44 |
| dougwig | then i have no idea. might indeed make the logs easier to parse. | 16:44 |
| blogan | i'm sure its something we don't need anymore, probably had a different idea in the beginning | 16:44 |
| blogan | i suppose if listener's were ever able to exist on more than one lb, the vip might become useful | 16:45 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/octavia: glance: support relying on tags to extract image id https://review.openstack.org/284690 | 16:48 |
| ihrachys | dougwig: sbalukoff: johnsom: ajo: claiming it done (full disclosure: haven't tested devstack changes locally just yet) ^ | 16:48 |
| ajo | ihrachys, :) | 16:49 |
| ajo | btw,is gerrit down? | 16:49 |
| ajo | ihrachys, ^ | 16:50 |
| johnsom | Yep | 16:50 |
| ajo | I'm geting "503 / Service unavailable" | 16:50 |
| ajo | :/ | 16:50 |
| ihrachys | yeah, prolly down | 16:50 |
| * ajo is sad | 16:50 | |
| ihrachys | #sadpanda | 16:50 |
| ajo | #sadgerrit | 16:50 |
| ajo | eh, back up! | 16:50 |
| ajo | that was quick | 16:50 |
| ihrachys | yeah, prolly restarted httpd | 16:50 |
| *** minwang2 has joined #openstack-lbaas | 16:52 | |
| *** markvan has quit IRC | 16:53 | |
| *** markvan has joined #openstack-lbaas | 16:54 | |
| *** amotoki has joined #openstack-lbaas | 16:54 | |
| *** minwang2 has quit IRC | 16:55 | |
| *** manishg has joined #openstack-lbaas | 16:59 | |
| *** kevo has joined #openstack-lbaas | 16:59 | |
| dougwig | 503 means httpd was up, so it was probably the java monster that got restarted. | 17:00 |
| *** minwang2 has joined #openstack-lbaas | 17:01 | |
| johnsom | In my half awake state all that pops into my mind is "Me hungry" when I hear "java monster" | 17:02 |
| *** nmagnezi has joined #openstack-lbaas | 17:04 | |
| *** jwarendt has quit IRC | 17:07 | |
| *** yamamoto has joined #openstack-lbaas | 17:09 | |
| *** numans has joined #openstack-lbaas | 17:11 | |
| *** yamamoto has quit IRC | 17:14 | |
| dougwig | that's all the heap hears, too. | 17:18 |
| openstackgerrit | German Eichberger proposed openstack/octavia: Cascading LB delete https://review.openstack.org/282587 | 17:18 |
| openstackgerrit | German Eichberger proposed openstack/octavia: Improved LoadBalancer Delete https://review.openstack.org/282113 | 17:18 |
| *** minwang2 has quit IRC | 17:26 | |
| *** markvan has quit IRC | 17:28 | |
| *** markvan has joined #openstack-lbaas | 17:30 | |
| *** rtheis_ is now known as rtheis | 17:31 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Get Me A Load Balancer Controller https://review.openstack.org/257013 | 17:45 |
| *** _ducttape_ has joined #openstack-lbaas | 17:46 | |
| *** ducttape_ has quit IRC | 17:49 | |
| *** kevo has quit IRC | 17:50 | |
| *** minwang2 has joined #openstack-lbaas | 17:53 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 17:53 | |
| *** amotoki has quit IRC | 17:56 | |
| *** localloop127 has quit IRC | 17:58 | |
| *** madhu_ak has quit IRC | 17:59 | |
| *** madhu_ak has joined #openstack-lbaas | 18:05 | |
| *** neelashah has quit IRC | 18:06 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Get Me A Load Balancer Controller https://review.openstack.org/257013 | 18:10 |
| *** yamamoto has joined #openstack-lbaas | 18:10 | |
| *** jschwarz has quit IRC | 18:11 | |
| *** Bjoern_ is now known as Bjoern_zZzZzZzZ | 18:12 | |
| *** jwarendt has joined #openstack-lbaas | 18:12 | |
| *** yamamoto has quit IRC | 18:15 | |
| *** Bjoern_zZzZzZzZ is now known as Bjoern_ | 18:17 | |
| *** neelashah has joined #openstack-lbaas | 18:18 | |
| *** neelashah has quit IRC | 18:19 | |
| *** neelashah has joined #openstack-lbaas | 18:20 | |
| *** evgenyf has joined #openstack-lbaas | 18:21 | |
| *** numans has quit IRC | 18:22 | |
| *** numans has joined #openstack-lbaas | 18:22 | |
| *** Aish has joined #openstack-lbaas | 18:22 | |
| *** ihrachys has quit IRC | 18:24 | |
| openstackgerrit | German Eichberger proposed openstack/octavia: Cascading LB delete https://review.openstack.org/282587 | 18:32 |
| xgerman | ^^ those patches need some love... | 18:32 |
| TrevorV | xgerman did you remove WIP yet? | 18:34 |
| TrevorV | I was bout to take a gander | 18:34 |
| xgerman | yep | 18:34 |
| xgerman | just now | 18:34 |
| TrevorV | kk, I'll go take a look | 18:34 |
| xgerman | Now I will focus on the LBaaS V2 side of the house | 18:34 |
| *** piet has joined #openstack-lbaas | 18:41 | |
| *** Purandar has quit IRC | 18:41 | |
| *** piet has quit IRC | 18:46 | |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: Adds Cascade Delete for LoadBalancers to Octavia Driver https://review.openstack.org/284340 | 18:48 |
| madhu_ak | johnsom, dougwig, could you review https://review.openstack.org/#/c/284875/ | 18:53 |
| *** localloop127 has joined #openstack-lbaas | 18:54 | |
| madhu_ak | fnaval ^^ | 18:56 |
| *** evgenyf has quit IRC | 18:56 | |
| fnaval | madhu_ak: yep looking at it; though, I'll need to familiarize myself on how that all works so I can't give input right away | 18:56 |
| madhu_ak | thanks | 18:56 |
| *** piet has joined #openstack-lbaas | 18:58 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Get Me A Load Balancer Controller https://review.openstack.org/257013 | 18:59 |
| *** localloop127 has quit IRC | 18:59 | |
| *** piet has quit IRC | 19:02 | |
| *** piet has joined #openstack-lbaas | 19:04 | |
| *** bana_k has joined #openstack-lbaas | 19:04 | |
| *** kevo has joined #openstack-lbaas | 19:08 | |
| *** numans has quit IRC | 19:10 | |
| *** yamamoto has joined #openstack-lbaas | 19:11 | |
| *** _ducttape_ has quit IRC | 19:12 | |
| *** bana_k has quit IRC | 19:14 | |
| *** yamamoto has quit IRC | 19:17 | |
| *** fawadkhaliq has joined #openstack-lbaas | 19:22 | |
| *** evgenyf has joined #openstack-lbaas | 19:22 | |
| *** markvan has quit IRC | 19:25 | |
| sbalukoff | Hey folks. | 19:27 |
| *** ducttape_ has joined #openstack-lbaas | 19:27 | |
| TrevorV | My patch is failing because it can't find a vip that its trying to deallocate... is this a possible bug recently merged? | 19:27 |
| *** markvan has joined #openstack-lbaas | 19:27 | |
| *** Purandar has joined #openstack-lbaas | 19:28 | |
| *** neelashah has quit IRC | 19:31 | |
| *** Aish has quit IRC | 19:33 | |
| *** Aish has joined #openstack-lbaas | 19:34 | |
| TrevorV | Sorry, I think its something I've added. pdb isn't very nice to me... | 19:36 |
| sbalukoff | Is the gate still busted? | 19:37 |
| *** manishg has quit IRC | 19:41 | |
| *** manishg has joined #openstack-lbaas | 19:43 | |
| *** neelashah has joined #openstack-lbaas | 19:45 | |
| *** manishg has quit IRC | 19:50 | |
| *** manishg has joined #openstack-lbaas | 19:50 | |
| *** localloop127 has joined #openstack-lbaas | 19:55 | |
| *** manishg has quit IRC | 19:56 | |
| *** madhu_ak has quit IRC | 19:57 | |
| *** localloop127 has quit IRC | 20:00 | |
| *** evgenyf has quit IRC | 20:01 | |
| *** manishg has joined #openstack-lbaas | 20:04 | |
| *** manishg has quit IRC | 20:04 | |
| *** minwang2 has quit IRC | 20:08 | |
| *** ducttape_ has quit IRC | 20:09 | |
| xgerman | sbalukoff, hi | 20:10 |
| *** madhu_ak has joined #openstack-lbaas | 20:11 | |
| *** piet has quit IRC | 20:11 | |
| *** fawadkhaliq has quit IRC | 20:13 | |
| *** bana_k has joined #openstack-lbaas | 20:13 | |
| *** yamamoto has joined #openstack-lbaas | 20:13 | |
| *** fawadkhaliq has joined #openstack-lbaas | 20:13 | |
| *** piet has joined #openstack-lbaas | 20:14 | |
| *** piet has quit IRC | 20:14 | |
| *** bana_k has quit IRC | 20:15 | |
| *** bana_k has joined #openstack-lbaas | 20:15 | |
| *** minwang2 has joined #openstack-lbaas | 20:15 | |
| *** ducttape_ has joined #openstack-lbaas | 20:16 | |
| TrevorV | xgerman hey man, where are the amphora agent logs so I don't have to kill the proc and set up pdb? | 20:17 |
| *** yamamoto has quit IRC | 20:17 | |
| xgerman | var log upstart | 20:18 |
| xgerman | need to be root to open | 20:18 |
| TrevorV | Do we have the root password or something somwhere? | 20:18 |
| TrevorV | somewhere***? | 20:18 |
| xgerman | sudo always works for me | 20:22 |
| xgerman | but I am running as ubuntu | 20:22 |
| TrevorV | xgerman I meant for logging in, but it just told me to ssh as ubuntu | 20:22 |
| TrevorV | all good, I got in :D | 20:22 |
| xgerman | cool | 20:22 |
| xgerman | ah, yeah, ssh... | 20:22 |
| TrevorV | Is there an easy way to verify the haproxy file manually? | 20:24 |
| TrevorV | Like, just so I can see what it complains about? | 20:24 |
| xgerman | not sure… you can run haproxy with some command line parameter to check syntax | 20:25 |
| xgerman | the agent does that when you upload a file | 20:26 |
| xgerman | in other news lbaas v2 keeps listeners and pools around after a cascade delete | 20:27 |
| xgerman | https://www.irccloud.com/pastebin/P8iTcPsj/ | 20:27 |
| xgerman | blogan ideas — short of deleting stuff in a for loop? | 20:30 |
| xgerman | you would think a data model could cascade by itself | 20:31 |
| blogan | xgerman: it keeps them int he db? | 20:31 |
| xgerman | yep | 20:31 |
| xgerman | our models don’t cascade | 20:31 |
| blogan | just pool to members do | 20:32 |
| blogan | xgerman: https://github.com/openstack/neutron-lbaas/blob/master/neutron_lbaas/db/loadbalancer/models.py#L190 | 20:33 |
| xgerman | yeah, | 20:33 |
| xgerman | and I think we need to make pools cascade on LB, etc. | 20:33 |
| *** piet has joined #openstack-lbaas | 20:34 | |
| xgerman | but I am no sqlalchemy expert | 20:34 |
| *** bana_k has quit IRC | 20:34 | |
| blogan | well here's the dilemma, we want this to only work for deleting a lb | 20:35 |
| blogan | if you put those cascade directives on all the relationships, it will work for lb, but also cause it to happen on deleting a listener as well | 20:35 |
| blogan | and with shared pools, that'll take some investigation to figure otu if it'll delete the pool when it no longer has any references to anything else | 20:36 |
| blogan | which might be what delete-orphan does, but not sure | 20:36 |
| xgerman | yeah, it’s a mess | 20:39 |
| blogan | so looping might be the answer for nwo | 20:39 |
| xgerman | or we back out sbalukoff’s pool patch ;-) | 20:40 |
| sbalukoff | xgerman: I will destroy you. | 20:40 |
| blogan | well the issue would still be there | 20:40 |
| sbalukoff | This is neutron-lbaas? | 20:40 |
| xgerman | yep | 20:40 |
| sbalukoff | *sigh* | 20:40 |
| blogan | yeah but i don't think we do cascade in octavia either | 20:40 |
| sbalukoff | Yep, my patch doesn't alter or fix stuff like that that's "broken" from before. | 20:40 |
| blogan | neutron's api doesn't do cascade either | 20:41 |
| blogan | i dont think its broken, its just being consistent with neutron | 20:41 |
| sbalukoff | Heh! | 20:41 |
| blogan | don't allow a user to delete a parent if it has children | 20:41 |
| *** bana_k has joined #openstack-lbaas | 20:41 | |
| blogan | (except for pools lol) | 20:41 |
| sbalukoff | Apparently. | 20:41 |
| xgerman | I would expect SNI to be the same | 20:41 |
| blogan | im not sure | 20:42 |
| xgerman | actually - no — | 20:42 |
| xgerman | so unless listener delete does it himself we have a bug | 20:42 |
| blogan | side question, about ipv6, does ipv6 have a broadcast address like ipv4? | 20:42 |
| blogan | an ipv6 network/subnet | 20:42 |
| xgerman | I think they only have prefixes | 20:43 |
| blogan | what do you mean? | 20:44 |
| xgerman | I guess I know too little to make sense | 20:45 |
| blogan | or i know too little to understand | 20:47 |
| blogan | im sure thats more likely | 20:47 |
| xgerman | ok, filed: https://bugs.launchpad.net/neutron/+bug/1549989 | 20:49 |
| openstack | Launchpad bug 1549989 in neutron "LBaaS V2 listener delete leaves SNI container behind" [Low,New] | 20:49 |
| xgerman | so I need to loop and delete pools, listeners, health monitors (and l7 if that becomes available) | 20:49 |
| xgerman | grrr | 20:50 |
| blogan | that, or we put the cascades and change how the code detects that parent ahs a child for the calls that don't allow cascade | 20:51 |
| *** piet has quit IRC | 20:51 | |
| *** ducttape_ has quit IRC | 20:52 | |
| *** ducttape_ has joined #openstack-lbaas | 20:53 | |
| xgerman | well, we already do that for LB | 20:54 |
| xgerman | that would be a good topic for dougwig to weigh in | 20:54 |
| xgerman | we want to avoid “orchestration" | 20:54 |
| dougwig | eh? | 20:54 |
| dougwig | we're not going to do the delete the same as create? make it a new driver call else fail? | 20:55 |
| *** rtheis has quit IRC | 20:55 | |
| xgerman | we do | 20:56 |
| xgerman | but labs_v2 doesn’t delete stuff | 20:56 |
| *** localloop127 has joined #openstack-lbaas | 20:56 | |
| xgerman | so cascade_delete on octavia blows everything away but labs_v2 retain pools, listeners, etc. | 20:56 |
| xgerman | so we can either change the data model to cascade or add a new cascade_delete method which loops and deletes stuff in the DB | 20:57 |
| *** localloop127 has quit IRC | 21:01 | |
| *** bana_k has quit IRC | 21:05 | |
| *** markvan has quit IRC | 21:06 | |
| *** rtheis has joined #openstack-lbaas | 21:06 | |
| TrevorV | sbalukoff do you know what the haproxy config should reflect for l7 changes? | 21:07 |
| sbalukoff | TrevorV: You should see indented lines that start with 'acl' | 21:07 |
| sbalukoff | But! | 21:07 |
| sbalukoff | That will only happen if the l7policies have rules. | 21:07 |
| *** markvan has joined #openstack-lbaas | 21:08 | |
| TrevorV | Well I'm not seeing that on my single create in the controller worker... dammit | 21:08 |
| *** piet has joined #openstack-lbaas | 21:08 | |
| TrevorV | sbalukoff in which section? | 21:09 |
| sbalukoff | In the the frontend section. | 21:09 |
| TrevorV | Should I see 2 backends if my listener has a default pool and my l7 policy has a redirect pool (which is a different one than the default) | 21:10 |
| TrevorV | ? | 21:10 |
| sbalukoff | What do your l7policy and l7rule tables look like? | 21:10 |
| sbalukoff | TrevorV: Yes, but only if the acl lines show up as well. | 21:10 |
| TrevorV | I'm not sure I follow what you said there, but I'm just trying to get an idea of what its supposed to look at so I can figure out which part I've missed. | 21:11 |
| TrevorV | I also don't know what you've asked me just now. What is the information the DB holds for my policy/rule? | 21:11 |
| TrevorV | sbalukoff ^^ | 21:12 |
| *** ducttape_ has quit IRC | 21:13 | |
| *** yamamoto has joined #openstack-lbaas | 21:14 | |
| TrevorV | sbalukoff I think its because I haven't added anything in the jinja config files in my review... I'm guessing we need a tweak in here to get it to pick up ALL the config options... | 21:17 |
| TrevorV | Still verifying though | 21:17 |
| TrevorV | Actually it looks like it "should" have done it. | 21:17 |
| TrevorV | Now I gotta see why it didn't. | 21:17 |
| TrevorV | _transform_listener should have done it it looks like. | 21:18 |
| sbalukoff | TrevorV: Yes, I was asking about the DB tables.... | 21:18 |
| sbalukoff | TrevorV: Are you operating on the data model or the SQLAlchemy model? | 21:19 |
| TrevorV | Sorry, words are tough for me sometimes. I know the DB has values in it, the problem might be another SQLAlchemy funky thing. I'll try to diagnose right now. | 21:19 |
| TrevorV | Its in the controller worker, so... SQLAlchemy model? I think? | 21:19 |
| sbalukoff | Because... if it's the data model... the data model is mostly static. And you kind of end up having to update relationships yourself. | 21:19 |
| *** yamamoto has quit IRC | 21:19 | |
| sbalukoff | no, controller worker operates on the data model. | 21:19 |
| TrevorV | Oh really? | 21:19 |
| sbalukoff | Ok.... I'm in the middle of troubleshooting something urgent. | 21:19 |
| TrevorV | No no please keep going | 21:20 |
| sbalukoff | Almost finished with it (I think). | 21:20 |
| *** fawadkhaliq has quit IRC | 21:20 | |
| sbalukoff | I can potentially help you with this afterward. | 21:20 |
| TrevorV | I'm mostly just opening dialogue so I can look at what I'm thinking and see if it makes sense | 21:20 |
| TrevorV | sorry :( | 21:20 |
| TrevorV | I should really not do that. | 21:20 |
| TrevorV | My bad | 21:20 |
| sbalukoff | No worries. | 21:20 |
| *** neelashah has quit IRC | 21:21 | |
| *** rtheis has quit IRC | 21:26 | |
| *** rtheis has joined #openstack-lbaas | 21:26 | |
| *** piet has quit IRC | 21:32 | |
| *** ducttape_ has joined #openstack-lbaas | 21:35 | |
| xgerman | blogan I am completely in the weeds with that LBaaS V2 driver | 21:36 |
| xgerman | so we are raising a NotImplemented in the delete method but the DB work gets done in LoadBalancerManager instead of BaseLoadBalancerManager | 21:37 |
| xgerman | so I think best would have been to have a property cascade on the driver | 21:37 |
| xgerman | will probably add that | 21:38 |
| *** Purandar has quit IRC | 21:39 | |
| *** neelashah has joined #openstack-lbaas | 21:47 | |
| openstackgerrit | Madhusudhan Kandadai proposed openstack/octavia: Add pre_test_hook to run gate jobs https://review.openstack.org/284946 | 21:50 |
| blogan | xgerman: you mean have the LoadBalancerManager do db calls? | 21:50 |
| xgerman | apparently everything gets wrapped | 21:51 |
| *** piet has joined #openstack-lbaas | 21:51 | |
| xgerman | I think I have something n which might work... | 21:52 |
| *** localloop127 has joined #openstack-lbaas | 21:57 | |
| *** localloop127 has quit IRC | 22:02 | |
| madhu_ak | dougwig resubmitted project-config patch: https://review.openstack.org/#/c/284875 with pre_test_hook patch: https://review.openstack.org/284946. Please review when you get time. Thanks | 22:03 |
| *** manishg has joined #openstack-lbaas | 22:08 | |
| *** manishg has quit IRC | 22:12 | |
| *** manishg has joined #openstack-lbaas | 22:13 | |
| *** yamamoto has joined #openstack-lbaas | 22:15 | |
| *** nmagnezi has quit IRC | 22:15 | |
| *** piet has quit IRC | 22:18 | |
| *** fawadkhaliq has joined #openstack-lbaas | 22:19 | |
| *** yamamoto has quit IRC | 22:20 | |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: [WIP] Adds Cascade option for load balancer drivers https://review.openstack.org/268237 | 22:26 |
| xgerman | ^^v blogan suggestions welcome | 22:26 |
| openstackgerrit | Stephen Balukoff proposed openstack/neutron-lbaas: Fix shared-pools lbaas_listeners migration https://review.openstack.org/283802 | 22:28 |
| openstackgerrit | Madhusudhan Kandadai proposed openstack/octavia: Add pre_test_hook to run gate jobs https://review.openstack.org/284946 | 22:28 |
| openstackgerrit | Phillip Toohill proposed openstack/octavia: Adding network driver for a vip on a single amp https://review.openstack.org/197858 | 22:42 |
| *** armax has quit IRC | 22:43 | |
| *** armax has joined #openstack-lbaas | 22:44 | |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: [WIP] Adds Cascade option for load balancer drivers https://review.openstack.org/268237 | 22:51 |
| *** neelashah has quit IRC | 22:52 | |
| openstackgerrit | Madhusudhan Kandadai proposed openstack/neutron-lbaas: [WIP] Neutron LBaaS: TLS Barbican Scenario Test https://review.openstack.org/164828 | 22:54 |
| *** armax has quit IRC | 22:57 | |
| *** localloop127 has joined #openstack-lbaas | 22:58 | |
| *** localloop127 has quit IRC | 23:03 | |
| TrevorV | johnsom xgerman rm_work blogan if you guys get a chance tonight, can you pick apart the single-create reviews? I finished a fully-configured manual test just now, all the way down to the haproxy config, and it all worked. | 23:07 |
| TrevorV | Maybe you guys can help me find any breaks? | 23:07 |
| TrevorV | Thanks! | 23:07 |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: Adds Cascade Delete for LoadBalancers to Octavia Driver https://review.openstack.org/284340 | 23:07 |
| openstackgerrit | German Eichberger proposed openstack/neutron-lbaas: [WIP] Adds Cascade option for load balancer drivers https://review.openstack.org/268237 | 23:07 |
| TrevorV | I'll be offline for now though, be back online later tonight to check for updates. (I'd love to see this merged by Monday!!) | 23:08 |
| *** TrevorV has quit IRC | 23:08 | |
| xgerman | I am under the weather so not sure how much I will get done | 23:08 |
| sbalukoff | TrevorV: Oh, did you figure out the haproxy rendering problem? | 23:09 |
| rm_work | missed him by a few seconds | 23:10 |
| *** frankie_ has joined #openstack-lbaas | 23:10 | |
| rm_work | looks like he did though | 23:10 |
| xgerman | yeah, cascade delete probably will bomb the gates | 23:11 |
| *** yamamoto_ has joined #openstack-lbaas | 23:16 | |
| *** fnaval has quit IRC | 23:19 | |
| *** yamamoto_ has quit IRC | 23:22 | |
| openstackgerrit | Phillip Toohill proposed openstack/octavia: Adding network driver for a vip on a single amp https://review.openstack.org/197858 | 23:23 |
| *** ducttape_ has quit IRC | 23:24 | |
| *** Purandar has joined #openstack-lbaas | 23:26 | |
| *** yuanying has joined #openstack-lbaas | 23:33 | |
| *** yuanying has quit IRC | 23:34 | |
| *** yuanying has joined #openstack-lbaas | 23:34 | |
| *** rtheis has quit IRC | 23:34 | |
| *** Bjoern_ has quit IRC | 23:36 | |
| *** Aish has quit IRC | 23:45 | |
| *** Aish has joined #openstack-lbaas | 23:45 | |
| *** SumitNaiksatam has quit IRC | 23:46 | |
| *** fawadkhaliq has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!