| rm_work | LGTM based on initial testing | 00:01 |
|---|---|---|
| rm_work | I think we merge the feature and then bugfix if we need to | 00:01 |
| johnsom | +1 | 00:01 |
| johnsom | I think it's decent. | 00:01 |
| rm_work | yep | 00:02 |
| rm_work | xgerman_? :P | 00:02 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: SSL Health Monitors didn't actually ... check very much https://review.openstack.org/475944 | 00:04 |
| rm_work | ^^ threw SSL on the end since it had a merge conflict and i think we'll just go down the line | 00:04 |
| rm_work | assuming we can get this stuff all just merged in today :) | 00:04 |
| rm_work | (might be a stretch of an assumption, and relies on xgerman_ still being around lol) | 00:05 |
| johnsom | That would be great! | 00:06 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Stop provider and flavor API stubs failing calls https://review.openstack.org/476290 | 00:10 |
| johnsom | Ok, I feel slightly dirty with that, but I think it's best thing to keep us moving forward. | 00:10 |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Updating for octavia-dashboard https://review.openstack.org/446178 | 00:15 |
| *** Aju has quit IRC | 00:20 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Stop provider and flavor API stubs failing calls https://review.openstack.org/476290 | 00:20 |
| xgerman_ | Hi | 00:22 |
| johnsom | xgerman_ I think rm_work was hoping you were ready to +2 this chain up to https://review.openstack.org/#/c/476271 | 00:23 |
| johnsom | The rest of the RBAC stuffs | 00:23 |
| rm_work | lulz... that is super weird johnsom, but i agree i guess -- i liked it when we were just clear(ish) that "these aren't implemented yet" as a 400 | 00:25 |
| rm_work | but it's better to just ... not do anything for now maybe | 00:25 |
| *** afranc has joined #openstack-lbaas | 00:27 | |
| johnsom | Well, I guess I could do some type of validation and return not implemented exceptions. | 00:29 |
| johnsom | It would be better than TypeError: 'provider' is an invalid keyword argument for LoadBalancer | 00:29 |
| johnsom | Since, octavia is good | 00:29 |
| johnsom | If not fake at the moment | 00:29 |
| johnsom | Yeah, that is less sketchy | 00:30 |
| johnsom | I will re-spin | 00:30 |
| *** sanfern has quit IRC | 00:31 | |
| johnsom | After I fix this nova exception eating that had driven me nuts for over a year | 00:31 |
| rm_work | rofl | 00:41 |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Updating for octavia-dashboard https://review.openstack.org/446178 | 00:44 |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Updating for octavia-dashboard https://review.openstack.org/446178 | 00:49 |
| rm_work | xgerman_: guessing no on that for today? | 01:38 |
| rm_work | johnsom: you didn't actually +2 the HM change at the end :P | 01:38 |
| rm_work | (after I updated it) | 01:39 |
| johnsom | Yeah, I need to look at that again | 01:39 |
| *** chlong has joined #openstack-lbaas | 01:44 | |
| rm_work | just added the migration, and copied an error-catch thing we had already in the v1 api that takes care of that weird message | 01:44 |
| rm_work | (i didn't just type it from scratch) | 01:44 |
| johnsom | Yeah, sorry, I had to run a surprise errand, just got back | 01:45 |
| rm_work | yeah no worries | 01:45 |
| rm_work | probably it's not going to merge today since xgerman_ is maybe not around and/or maybe actually wants to spend some time reviewing/testing | 01:45 |
| rm_work | :P | 01:46 |
| *** rstarmer has quit IRC | 01:59 | |
| xgerman_ | rm_work I thought I +2'd Seth | 02:01 |
| xgerman_ | Steh | 02:01 |
| xgerman_ | Autocorrect | 02:01 |
| *** rstarmer has joined #openstack-lbaas | 02:01 | |
| xgerman_ | +A | 02:02 |
| *** chlong has quit IRC | 02:04 | |
| *** rstarmer has quit IRC | 02:12 | |
| JudeC | johnsom: Well in working on this I realized limiting the query doesn't really matter inside the pagination helper as there seems to be two calls to the db. | 02:22 |
| *** gongysh has joined #openstack-lbaas | 02:24 | |
| JudeC | So as an optimization i can make it so the pagination helper only looks for the ID of the item to be returned and then the _convert_db_to_type does the heavy lifting from there. | 02:24 |
| JudeC | so up until _convert_db_to_type is run I can just return the ID from the DB and everything works fine. | 02:25 |
| johnsom | but these are lists, ID won't work right? That is like only one row | 02:27 |
| reedip | johnsom , xgerman_ got your message | 02:32 |
| reedip | will check out the qos patch again and push a PS out | 02:32 |
| JudeC | Yeah, it seems to just be unoptimized. the pagination helper will order things for us and is actually returning irrelevant stuff if it returns anything other than the ID. This is because on each controller we run "result = self._convert_db_to_type(load_balancers, [lb_types.LoadBalancerResponse])" which then calls from_data_model(). | 02:35 |
| JudeC | which then populates all of the data, so essentially we are requesting all of the same data twice. | 02:36 |
| JudeC | unless i am grossly misunderstanding something, but I am running the API right now in devstack with the query just returning the ID and it seems to be working perfectly. | 02:36 |
| *** chlong has joined #openstack-lbaas | 02:37 | |
| *** rstarmer has joined #openstack-lbaas | 02:39 | |
| JudeC | Here I will commit these changes because I feel like I am doing a crappy job of explaining what I am doing. | 02:40 |
| korean101 | rm_work: OMG | 02:45 |
| korean101 | rm_work: lbaas-loadbalancer-create works fiine! | 02:46 |
| korean101 | rm_work: OH really thanks | 02:46 |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 02:48 |
| *** JudeC has quit IRC | 02:48 | |
| johnsom | reedip Thanks, I want to get that into Pike and wasn't sure the status | 02:50 |
| *** JudeC has joined #openstack-lbaas | 02:50 | |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 02:53 |
| JudeC | Added comments to explain as well. | 02:56 |
| *** yamamoto has joined #openstack-lbaas | 02:56 | |
| *** JudeC has quit IRC | 02:57 | |
| *** amotoki_away is now known as amotoki | 03:06 | |
| *** gongysh has quit IRC | 03:39 | |
| *** gongysh has joined #openstack-lbaas | 03:42 | |
| *** gongysh has quit IRC | 03:53 | |
| openstackgerrit | Vu Cong Tuan proposed openstack/neutron-lbaas master: Remove support for py34 https://review.openstack.org/476331 | 03:59 |
| openstackgerrit | Vu Cong Tuan proposed openstack/neutron-lbaas master: Replace the usage of 'manager' with 'os_primary' https://review.openstack.org/476333 | 04:00 |
| *** chlong has quit IRC | 04:02 | |
| *** chlong has joined #openstack-lbaas | 04:16 | |
| *** links has joined #openstack-lbaas | 04:21 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Octavia Feature Classification https://review.openstack.org/451177 | 04:23 |
| openstackgerrit | Vu Cong Tuan proposed openstack/neutron-lbaas master: Replace the usage of 'manager' with 'os_primary' https://review.openstack.org/476333 | 04:25 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Provide better ComputeBuildExcpetion messages https://review.openstack.org/476348 | 04:53 |
| johnsom | Ok, that should make nova faults much more clear to the operator | 04:55 |
| *** yuanying_ has joined #openstack-lbaas | 05:03 | |
| *** yuanying has quit IRC | 05:05 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Stop provider and flavor API stubs failing calls https://review.openstack.org/476290 | 05:20 |
| *** gcheresh has joined #openstack-lbaas | 05:27 | |
| *** kobis has joined #openstack-lbaas | 05:45 | |
| *** gongysh has joined #openstack-lbaas | 05:47 | |
| *** csomerville has quit IRC | 06:02 | |
| *** cody-somerville has joined #openstack-lbaas | 06:03 | |
| *** cody-somerville has quit IRC | 06:03 | |
| *** cody-somerville has joined #openstack-lbaas | 06:03 | |
| *** yuanying_ has quit IRC | 06:19 | |
| *** sanfern has joined #openstack-lbaas | 06:20 | |
| *** yuanying has joined #openstack-lbaas | 06:20 | |
| *** rcernin has joined #openstack-lbaas | 06:30 | |
| *** rcernin has quit IRC | 06:37 | |
| *** kobis has quit IRC | 06:43 | |
| *** rcernin has joined #openstack-lbaas | 06:50 | |
| *** JudeC has joined #openstack-lbaas | 06:54 | |
| *** gcheresh has quit IRC | 06:54 | |
| *** tesseract has joined #openstack-lbaas | 07:02 | |
| *** pcaruana has joined #openstack-lbaas | 07:05 | |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 07:12 |
| openstackgerrit | ZhaoBo proposed openstack/octavia master: Extend api to accept qos_policy_id https://review.openstack.org/458308 | 07:22 |
| korean101 | anyone knows this? (http://paste.openstack.org/show/IKRY7ac1qbE3AolfGFxt/) | 07:40 |
| korean101 | not octavia's problem? | 07:41 |
| *** kobis has joined #openstack-lbaas | 07:57 | |
| openstackgerrit | Nir Magnezi proposed openstack/octavia-dashboard master: Optimize the link address https://review.openstack.org/455298 | 08:02 |
| *** yamamoto has quit IRC | 08:03 | |
| *** yuanying_ has joined #openstack-lbaas | 08:09 | |
| *** yuanying has quit IRC | 08:13 | |
| *** gcheresh has joined #openstack-lbaas | 08:17 | |
| openstackgerrit | Hengqing Hu proposed openstack/octavia-dashboard master: Updating for octavia-dashboard https://review.openstack.org/446178 | 08:20 |
| openstackgerrit | Nir Magnezi proposed openstack/octavia-dashboard master: Optimize the link address https://review.openstack.org/455298 | 08:38 |
| *** dmk0202 has joined #openstack-lbaas | 08:40 | |
| *** mdavidson has quit IRC | 08:41 | |
| *** sanfern has quit IRC | 08:42 | |
| *** mdavidson has joined #openstack-lbaas | 08:50 | |
| *** yamamoto has joined #openstack-lbaas | 09:04 | |
| *** yamamoto has quit IRC | 09:09 | |
| *** yamamoto has joined #openstack-lbaas | 09:09 | |
| *** JudeC has quit IRC | 09:25 | |
| *** yamamoto has quit IRC | 09:30 | |
| *** yamamoto has joined #openstack-lbaas | 09:36 | |
| *** dmk0202 has quit IRC | 09:40 | |
| *** yamamoto has quit IRC | 09:46 | |
| *** wasmum has quit IRC | 10:04 | |
| *** wasmum has joined #openstack-lbaas | 10:06 | |
| *** wasmum has quit IRC | 10:11 | |
| *** yamamoto has joined #openstack-lbaas | 10:21 | |
| *** wasmum has joined #openstack-lbaas | 10:22 | |
| *** yamamoto has quit IRC | 10:28 | |
| *** yamamoto has joined #openstack-lbaas | 10:34 | |
| *** yamamoto has quit IRC | 10:39 | |
| *** yamamoto has joined #openstack-lbaas | 10:40 | |
| *** gongysh has quit IRC | 10:47 | |
| *** yamamoto has quit IRC | 11:11 | |
| *** yamamoto has joined #openstack-lbaas | 11:33 | |
| *** yamamoto has quit IRC | 11:39 | |
| *** sanfern has joined #openstack-lbaas | 12:09 | |
| *** sanfern has quit IRC | 12:19 | |
| *** yamamoto has joined #openstack-lbaas | 12:19 | |
| *** yamamoto has quit IRC | 12:19 | |
| *** chlong has quit IRC | 12:26 | |
| *** gongysh has joined #openstack-lbaas | 12:27 | |
| *** kobis has quit IRC | 12:34 | |
| *** kobis has joined #openstack-lbaas | 12:35 | |
| *** yamamoto has joined #openstack-lbaas | 12:36 | |
| *** kobis has quit IRC | 12:40 | |
| *** yamamoto has quit IRC | 12:51 | |
| *** sanfern has joined #openstack-lbaas | 13:00 | |
| *** kobis has joined #openstack-lbaas | 13:08 | |
| *** sanfern has quit IRC | 13:14 | |
| *** yamamoto has joined #openstack-lbaas | 13:15 | |
| *** sanfern has joined #openstack-lbaas | 13:17 | |
| *** catintheroof has joined #openstack-lbaas | 13:20 | |
| *** chlong has joined #openstack-lbaas | 13:39 | |
| *** cpuga has joined #openstack-lbaas | 13:44 | |
| *** cpuga has quit IRC | 13:45 | |
| *** cpuga has joined #openstack-lbaas | 13:46 | |
| *** kobis has quit IRC | 13:50 | |
| *** wasmum has quit IRC | 13:50 | |
| *** wasmum has joined #openstack-lbaas | 13:53 | |
| *** kobis has joined #openstack-lbaas | 13:56 | |
| *** pksingh has joined #openstack-lbaas | 14:29 | |
| pksingh | johnsom: Hello | 14:31 |
| pksingh | johnsom: i tried to analyse the issue https://bugs.launchpad.net/octavia/+bug/1695331, and posted my comments there | 14:32 |
| openstack | Launchpad bug 1695331 in octavia "Octavia v2 API Requesting a VIP address is not working" [Critical,In progress] - Assigned to Pradeep Kumar Singh (pradeep-singh-u) | 14:32 |
| pksingh | johnsom: please have a look when you have some time , thanks | 14:32 |
| *** links has quit IRC | 14:45 | |
| *** kobis has quit IRC | 14:55 | |
| *** gcheresh has quit IRC | 15:06 | |
| *** isotope has joined #openstack-lbaas | 15:09 | |
| *** rcernin has quit IRC | 15:09 | |
| *** KeithMnemonic has quit IRC | 15:14 | |
| *** fnaval has quit IRC | 15:14 | |
| *** fnaval has joined #openstack-lbaas | 15:29 | |
| *** isotope has quit IRC | 15:37 | |
| *** pcaruana has quit IRC | 15:42 | |
| *** isotope has joined #openstack-lbaas | 15:49 | |
| *** isotope has quit IRC | 15:49 | |
| *** isotope has joined #openstack-lbaas | 15:56 | |
| rm_work | korean101: congrats :) | 16:02 |
| rm_work | korean101: that is something wrong in neutron-lbaas... i'm not sure what's going on there | 16:02 |
| rm_work | somehow the stats object coming across from the octavia streamer isn't what it's expecting? | 16:03 |
| rm_work | korean101: what version of octavia are you running, and what version of neutron-lbaas? | 16:03 |
| johnsom | Yeah, I wonder if that addition of the "error" column caused a version dependency | 16:03 |
| *** gongysh has quit IRC | 16:05 | |
| *** rstarmer has quit IRC | 16:06 | |
| *** yamamoto has quit IRC | 16:08 | |
| rm_work | it might have | 16:08 |
| rm_work | we might need to backport a bugfix to have it ignore unknown columns or something | 16:09 |
| *** yamamoto has joined #openstack-lbaas | 16:10 | |
| *** amotoki is now known as amotoki_away | 16:15 | |
| *** yamamoto has quit IRC | 16:15 | |
| rm_work | xgerman_: you +2'd the end of the chain but nothing in front :P | 16:21 |
| xgerman_ | ha | 16:25 |
| xgerman_ | you guys need to give me better instructions | 16:25 |
| rm_work | lol | 16:29 |
| johnsom | eandersson Padding your stackalytics? The refactor patch was already workflowed.... | 16:35 |
| isotope | johnsom I'm trying to work around bug #1592612 | 16:39 |
| openstack | bug 1592612 in octavia "LBaaS TLS is not working with non-admin tenant" [High,Confirmed] https://launchpad.net/bugs/1592612 | 16:39 |
| xgerman_ | those RBAC patches are pretty repetitive | 16:39 |
| isotope | I tried both options that you recommeneded and still have the same error | 16:39 |
| johnsom | xgerman_ that last refactor patch cleans some of that up | 16:40 |
| rm_work | xgerman_: yes, that's why i refactored it all at the end | 16:40 |
| isotope | Do I need to do both the policy file change and acls? | 16:40 |
| johnsom | isotope No, either one should work. | 16:40 |
| rm_work | isotope: should just be one | 16:40 |
| johnsom | isotope Which lbaas driver are you using? Octavia driver? | 16:40 |
| isotope | the device driver is neutron_lbaas.drivers.haproxy.namespace_driver.HaproxyNSDriver | 16:42 |
| openstackgerrit | Merged openstack/octavia-dashboard master: Exclude node_modules from pep8 checking https://review.openstack.org/475800 | 16:42 |
| johnsom | isotope Oh, ok the old legacy driver | 16:43 |
| isotope | Ah, should I switch to octavia? | 16:43 |
| *** sanfern has quit IRC | 16:43 | |
| isotope | I'm running openstack newton | 16:43 |
| johnsom | Hmm, ok, so with that driver, the user should be your neutron service account that you use for the "admin_id" | 16:44 |
| isotope | yeah I beleive that is neutron, which I tried with the same result | 16:44 |
| johnsom | isotope Yeah, that old driver doesn't scale very well and has no HA | 16:44 |
| eandersson | johnsom, lol | 16:44 |
| eandersson | having more people review code and said they raid it never hurts | 16:45 |
| eandersson | and adding yourself so that you get notifications in case gate fails (although that can be done by just adding yourself as a watcher or something I guess) | 16:45 |
| eandersson | *read | 16:45 |
| isotope | so right now I have the policy file edited with "secret:get": "rule:all_users" | 16:45 |
| johnsom | eandersson Yeah, but make sure they are good reviews and not just +1's... grin. Part of the PTL job to make sure folks are doing quality reviews. | 16:46 |
| johnsom | isotope I think there is more than just secret required on the barbican side. There is probably a container one too. Let me see if I can look at their policy file | 16:47 |
| isotope | Ah | 16:47 |
| johnsom | isotope If you are just testing, you could do "admin": "role:admin or user:neutron" | 16:51 |
| isotope | okay I'll try that now | 16:52 |
| *** sanfern has joined #openstack-lbaas | 16:52 | |
| *** eandersson has quit IRC | 16:54 | |
| *** sshank has joined #openstack-lbaas | 16:54 | |
| johnsom | For specific policies, I think it is consumers:post, consumers:delete, container:get, and secret:get, though I might be wrong. | 16:56 |
| rm_work | I think that is correct | 16:57 |
| isotope | okay I have all suggested policy file changes in place, testing now | 17:01 |
| isotope | same error | 17:02 |
| *** cpuga has quit IRC | 17:05 | |
| johnsom | What is the error you are seeing in the neutron svc log? | 17:09 |
| isotope | neutron.api.v2.resource CertManagerError: Could not process TLS container http://X.X.X.X:9311/v1/containers/5b3dbc0e-32e3-492e-9e09-34918800c0a0, The request you have made requires authentication. | 17:10 |
| johnsom | isotope Is the [service_auth] section setup correctly in the neutron_lbaas.conf? | 17:11 |
| isotope | thats from the neutron-server.log | 17:11 |
| johnsom | That is the other thing that could be tried. Set that to use the admin account | 17:13 |
| isotope | doing that now | 17:14 |
| *** eandersson has joined #openstack-lbaas | 17:16 | |
| eandersson | isantosp: What version of Barbican? | 17:16 |
| isotope | barbican 4.0.1 | 17:18 |
| *** sshank has quit IRC | 17:21 | |
| *** sshank has joined #openstack-lbaas | 17:23 | |
| *** sshank has quit IRC | 17:23 | |
| *** sshank has joined #openstack-lbaas | 17:23 | |
| sanfern | korean101, I am also seeing same issue in my test, Octavia DB has status ACTIVE for LB's but not in neutron DB | 17:25 |
| rm_work | johnsom: https://review.openstack.org/#/c/476290/3 | 17:26 |
| rm_work | sanfern: so the event streamer is going between octavia and neutron-lbaas, but it's erroring out because of a bad field in loadbalancer_stats ? | 17:27 |
| sanfern | rm_work, yes | 17:28 |
| rm_work | hmmmmmmmm | 17:29 |
| rm_work | what versions are you running | 17:29 |
| rm_work | can you see what field is causing this error? | 17:29 |
| rm_work | i don't know how much debug logging we'd do around that... | 17:29 |
| *** yamamoto has joined #openstack-lbaas | 17:30 | |
| sanfern | rm_work, not able to find field | 17:30 |
| rm_work | hmmm | 17:30 |
| rm_work | johnsom: when did we change that table? | 17:30 |
| rm_work | you said we added something? | 17:30 |
| *** fnaval_ has joined #openstack-lbaas | 17:30 | |
| *** fnaval_ has quit IRC | 17:30 | |
| *** fnaval_ has joined #openstack-lbaas | 17:31 | |
| johnsom | So, yeah, someone recently added "request_errors" to the stats in octavia, but neutron DB models don't have that | 17:32 |
| johnsom | https://github.com/openstack/neutron-lbaas/blob/master/neutron_lbaas/db/loadbalancer/models.py#L53 | 17:32 |
| *** fnaval has quit IRC | 17:32 | |
| johnsom | u'info_payload': {u'bytes_out': 212, u'request_errors': 0, u'total_connections': 1, u'active_connections': 0, u'bytes_in': 205}} | 17:32 |
| rm_work | yeah | 17:33 |
| rm_work | hmm | 17:33 |
| rm_work | ok | 17:33 |
| johnsom | So, I would kind of vote to strip it from the event before we send it to oslo messaging | 17:33 |
| rm_work | ahhh | 17:33 |
| rm_work | rather than patch+backport neutron-lbaas? | 17:33 |
| johnsom | because I don't feel the need to add that to all of n-lbaas | 17:33 |
| rm_work | k | 17:33 |
| rm_work | fine by me | 17:33 |
| rm_work | you doing it or should I? | 17:33 |
| johnsom | Yeah, I can do it today | 17:34 |
| rm_work | or does someone else here want to take it? | 17:34 |
| rm_work | we don't have to do everything :P lol | 17:34 |
| openstackgerrit | Merged openstack/neutron-lbaas master: Remove support for py34 https://review.openstack.org/476331 | 17:34 |
| *** reedip_ has joined #openstack-lbaas | 17:34 | |
| *** yamamoto has quit IRC | 17:35 | |
| openstackgerrit | Merged openstack/neutron-lbaas master: Updated from global requirements https://review.openstack.org/475562 | 17:36 |
| openstackgerrit | Merged openstack/neutron-lbaas master: Fixes module not found error in NetScaler LBaaSV2 driver https://review.openstack.org/473988 | 17:36 |
| isotope | okay so in debug mode I see this attempt right before it fails | 17:36 |
| isotope | Creating consumer registration for container http://X.X.X.X:9311/v1/containers/5b3dbc0e-32e3-492e-9e09-34918800c0a0 as lbaas: lbaas://RegionOne | 17:36 |
| johnsom | Hmm, so it's trying to use a user called lbaas???? | 17:37 |
| isotope | yeah that looks jacked up | 17:37 |
| johnsom | Yeah, I would search your /etc/neutron to see where that is coming from. It has to be in one of those configs | 17:38 |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to pools v2 API https://review.openstack.org/475868 | 17:43 |
| johnsom | Yeah, isotope I suspect it's the service_auth section I mentioned earlier. I just checked and it's default is "admin" not "lbaas" | 17:43 |
| *** SumitNaiksatam has joined #openstack-lbaas | 17:43 | |
| *** cpuga has joined #openstack-lbaas | 17:49 | |
| *** cpuga has quit IRC | 17:52 | |
| *** cpuga has joined #openstack-lbaas | 17:52 | |
| *** cpuga has quit IRC | 17:56 | |
| *** tesseract has quit IRC | 17:59 | |
| *** sshank has quit IRC | 18:00 | |
| *** cpuga has joined #openstack-lbaas | 18:00 | |
| *** cpuga_ has joined #openstack-lbaas | 18:01 | |
| *** cody-somerville has quit IRC | 18:02 | |
| *** cody-somerville has joined #openstack-lbaas | 18:02 | |
| *** cody-somerville has quit IRC | 18:02 | |
| *** cody-somerville has joined #openstack-lbaas | 18:02 | |
| *** cpuga has quit IRC | 18:04 | |
| rm_work | johnsom: ah https://review.openstack.org/#/c/476290/ has a conflict now anyway :) | 18:05 |
| rm_work | so you can fix that when you fix the comma | 18:05 |
| *** JudeC has joined #openstack-lbaas | 18:07 | |
| *** sshank has joined #openstack-lbaas | 18:08 | |
| *** csomerville has joined #openstack-lbaas | 18:10 | |
| *** cpuga_ has quit IRC | 18:13 | |
| *** cody-somerville has quit IRC | 18:13 | |
| *** sshank has quit IRC | 18:22 | |
| *** sshank has joined #openstack-lbaas | 18:22 | |
| *** sshank has quit IRC | 18:23 | |
| *** sshank has joined #openstack-lbaas | 18:23 | |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to members v2 API https://review.openstack.org/475920 | 18:32 |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to health monitors v2 API https://review.openstack.org/475934 | 18:32 |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to L7 policies v2 API https://review.openstack.org/475945 | 18:32 |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to l7rules v2 API https://review.openstack.org/475953 | 18:34 |
| openstackgerrit | Merged openstack/octavia master: Add RBAC enforcement to quotas v2 API https://review.openstack.org/475980 | 18:34 |
| openstackgerrit | Merged openstack/octavia master: Refactor the RBAC auth enforcement a bit https://review.openstack.org/476271 | 18:34 |
| openstackgerrit | Merged openstack/octavia master: SSL Health Monitors didn't actually ... check very much https://review.openstack.org/475944 | 18:35 |
| rm_work | johnsom: \o/ | 18:35 |
| *** sshank has quit IRC | 18:35 | |
| *** sshank has joined #openstack-lbaas | 18:35 | |
| johnsom | For the RBAC win! | 18:37 |
| rm_work | also for the SSL-healthcheck win | 18:40 |
| *** sshank has quit IRC | 18:46 | |
| *** gcheresh has joined #openstack-lbaas | 19:06 | |
| *** pksingh has quit IRC | 19:06 | |
| *** fnaval_ has quit IRC | 19:09 | |
| *** fnaval has joined #openstack-lbaas | 19:10 | |
| *** cpuga has joined #openstack-lbaas | 19:11 | |
| xgerman_ | rm_work once you give me clear instrcutions… | 19:11 |
| rm_work | :P | 19:11 |
| *** cpuga has quit IRC | 19:12 | |
| *** cpuga has joined #openstack-lbaas | 19:12 | |
| *** cpuga has quit IRC | 19:13 | |
| *** cpuga has joined #openstack-lbaas | 19:14 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Stop provider and flavor API stubs failing calls https://review.openstack.org/476290 | 19:16 |
| johnsom | Ok, with that, I am going to run to lunch and pick up flowers (anniversary today). Back in a bit. | 19:17 |
| rm_work | ;) | 19:18 |
| *** isotope has quit IRC | 19:35 | |
| *** reedip_ has quit IRC | 19:36 | |
| *** sshank has joined #openstack-lbaas | 19:36 | |
| *** eandersson has quit IRC | 19:38 | |
| *** gcheresh has quit IRC | 19:47 | |
| *** gcheresh has joined #openstack-lbaas | 19:48 | |
| *** blogan has joined #openstack-lbaas | 19:50 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Allow operators to disable v1 or v2.0 api endpoints https://review.openstack.org/476690 | 20:11 |
| rm_work | johnsom: ^^ | 20:12 |
| openstackgerrit | Merged openstack/octavia master: Provide better ComputeBuildExcpetion messages https://review.openstack.org/476348 | 20:12 |
| openstackgerrit | Santhosh Fernandes proposed openstack/octavia master: Remove request_errors from lb_stats https://review.openstack.org/476691 | 20:12 |
| *** gcheresh has quit IRC | 20:15 | |
| *** gcheresh has joined #openstack-lbaas | 20:15 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Allow operators to disable v1 or v2.0 api endpoints https://review.openstack.org/476690 | 20:25 |
| johnsom | I think we should add a warning log message on API startup about the V1 API | 20:28 |
| *** isotope has joined #openstack-lbaas | 20:28 | |
| johnsom | Especially if both are enabled | 20:29 |
| *** cpuga has quit IRC | 20:37 | |
| rm_work | lol | 20:39 |
| rm_work | well | 20:39 |
| rm_work | they're both enabled by default | 20:39 |
| rm_work | so you'd have to explicitly disable both in config | 20:39 |
| johnsom | No, just a warning log message | 20:40 |
| rm_work | yeah | 20:40 |
| rm_work | but i mean | 20:40 |
| rm_work | I guess? >_> | 20:40 |
| johnsom | A safety message that in case someone screws up | 20:40 |
| rm_work | alright | 20:40 |
| rm_work | will do | 20:40 |
| johnsom | Basically, "The v1 API is enabled. Make sure the API is not publicly accessible" or something like that | 20:41 |
| *** cpuga has joined #openstack-lbaas | 20:42 | |
| rm_work | oh | 20:47 |
| rm_work | yeah one sec | 20:47 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Allow operators to disable v1 or v2.0 api endpoints https://review.openstack.org/476690 | 20:50 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Allow operators to disallow creation of TLS Termination listeners https://review.openstack.org/476701 | 20:50 |
| rm_work | johnsom: ^^ fixed and one more | 20:50 |
| *** cpuga has quit IRC | 20:50 | |
| *** cpuga has joined #openstack-lbaas | 20:51 | |
| johnsom | "is this intentional?" lol | 20:51 |
| xgerman_ | yeah, what’s the difference to not configuring BBQ | 20:52 |
| xgerman_ | for #2 | 20:52 |
| johnsom | Probably a much nicer error message | 20:54 |
| rm_work | ^^ this | 20:54 |
| rm_work | right now it just explodes | 20:55 |
| johnsom | Plus, they could use the, gasp, local cert store | 20:55 |
| johnsom | I guess we could re-use the setting and add a none option, but... | 20:57 |
| rm_work | well, open to thoughts | 20:57 |
| rm_work | this was just the first thing i thought of to do it | 20:58 |
| *** gcheresh has quit IRC | 21:00 | |
| rm_work | johnsom: looks like the request_errors thing was NOT the issue | 21:01 |
| johnsom | Really? | 21:05 |
| johnsom | How so? | 21:06 |
| rm_work | i mean it just wasn't | 21:06 |
| rm_work | saw sanfern's logs | 21:06 |
| rm_work | with his patch | 21:06 |
| rm_work | request_errors isn't in the data that it's receiving anymore | 21:07 |
| rm_work | and it still has the same model mapping error | 21:07 |
| rm_work | brb | 21:12 |
| johnsom | sanfern What version of neutron-lbaas is this again? | 21:14 |
| rm_work | he said he was off for the night, but I think he's on newton n-lbaas and ocata octavia | 21:15 |
| johnsom | Ah, ok. Sigh | 21:16 |
| johnsom | https://github.com/openstack/neutron-lbaas/blob/stable/newton/neutron_lbaas/drivers/driver_base.py#L66 | 21:16 |
| johnsom | Geez, that is even a mess in octata | 21:17 |
| *** catintheroof has quit IRC | 21:18 | |
| *** sshank has quit IRC | 21:18 | |
| johnsom | Fixed here: https://review.openstack.org/#/c/449573/ | 21:19 |
| johnsom | Though I'm still suspicious of the error message not mapping | 21:19 |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 21:24 |
| johnsom | JudeC So what is the status on that. I was trying to look at it this morning. I couldn't find the second DB query you mentioned | 21:25 |
| JudeC | Yeah I tore into it all last night trying to figure it out... | 21:26 |
| JudeC | Talked to rm_work about it like 20 minutes ago | 21:26 |
| johnsom | Ok | 21:26 |
| JudeC | He was saying it is due to sqlalchemy's lazy loading? | 21:26 |
| johnsom | Did you folks make sense of that stuff? | 21:26 |
| JudeC | So I restrict the query to only return the ID but due to lazy loading it still has access to everything in the table I guess which is kinda beyond me. | 21:27 |
| johnsom | sqlalchemy has demons (like evil spirits) | 21:28 |
| johnsom | With a dash of voodoo thrown in | 21:28 |
| johnsom | So this version is just a rebase? | 21:29 |
| JudeC | I use sqlalchemy's load_only method which should defer column selection to only the ones specified in the model passed into load_only but like you said voodoo happens behind the scenes with sqlalhemy and it still populates the object with all of the data when to_data_model() is executed. | 21:30 |
| JudeC | yeah just a rebase | 21:30 |
| johnsom | Yeah, I think the key there is deferred | 21:30 |
| johnsom | So, just so I don't feel like I am too crazy. There is only actually one query call right? | 21:31 |
| JudeC | I believe so yes. | 21:31 |
| JudeC | I thought there had to be another one when I was looking at it last night but that was before Adam explained the lazy loading thing to me. | 21:32 |
| JudeC | rm_work* | 21:32 |
| johnsom | Ok, so I think it might be a good idea to remove that load_only change. Let me take a minute and see if there is a different way to do taht | 21:33 |
| JudeC | Alright, I will admit my backround with ORMs comes from ruby sequel and some django so I am not the strongest when dealing with database stuff like this. Takes me quite a bit or research to fully understand what I am doing -.-. | 21:34 |
| JudeC | (getting better though) | 21:35 |
| johnsom | Yeah, I have used some other ORMs in the past. sqlalchemy is pretty sucky IMO. The docs are bad too | 21:35 |
| JudeC | That is the main thing for me. If the docs are good I am usually golden.. haha | 21:36 |
| JudeC | Same can be said for most developers though... | 21:36 |
| JudeC | I will push the change up to remove the load only thing right now. one sec | 21:37 |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 21:40 |
| *** armax has joined #openstack-lbaas | 21:41 | |
| johnsom | JudeC Maybe with_entities | 21:42 |
| *** fnaval has quit IRC | 21:43 | |
| JudeC | I tried that, I can dig into it again. The problem I ran into was when I applied with_entities it no longer had access to from_data_model() | 21:44 |
| JudeC | I will need to double check that. | 21:44 |
| johnsom | Ok | 21:44 |
| JudeC | I believe that was the issue with that though. | 21:44 |
| JudeC | hmm, let me spin my wheels at that again though and see if I can get some traction. | 21:45 |
| johnsom | Well, I am just about at the point where I say good enough and let someone down the line worry about optimizing it. | 21:50 |
| johnsom | I like to do due-diligence on stuff like this, but there is a line | 21:50 |
| *** blogan has quit IRC | 21:56 | |
| *** fnaval has joined #openstack-lbaas | 21:56 | |
| *** chlong has quit IRC | 21:57 | |
| JudeC | if it weren't for the lazy_loading this would be so easy... | 21:58 |
| JudeC | T_T | 21:59 |
| johnsom | I lean towards, good enough | 21:59 |
| johnsom | I am going to load it up, do one more set of tests and vote | 22:03 |
| *** armax_ has joined #openstack-lbaas | 22:05 | |
| *** eandersson has joined #openstack-lbaas | 22:05 | |
| *** armax has quit IRC | 22:05 | |
| *** armax_ is now known as armax | 22:05 | |
| johnsom | JudeC Do you have listener stats API in process? | 22:09 |
| JudeC | Not yet, I limited the stats to just loadbalancer for now. | 22:09 |
| johnsom | Ok, NP | 22:10 |
| johnsom | I will rebase and add RBAC to the status and LB stats patch | 22:10 |
| JudeC | wont be hard to implement though. | 22:10 |
| johnsom | Right | 22:10 |
| *** cpuga has quit IRC | 22:11 | |
| *** fnaval has quit IRC | 22:17 | |
| *** ajo has quit IRC | 22:21 | |
| *** nmagnezi has quit IRC | 22:22 | |
| *** ajo has joined #openstack-lbaas | 22:22 | |
| *** ajo has quit IRC | 22:28 | |
| rm_work | johnsom: OH so we just need to backport He Qing's patch | 22:30 |
| *** blogan has joined #openstack-lbaas | 22:31 | |
| rm_work | can you do that? probably all the way to newton? | 22:31 |
| rm_work | before they freeze it | 22:31 |
| *** nmagnezi has joined #openstack-lbaas | 22:32 | |
| johnsom | Yeah, I just think we need to test it as I think it is still going to fail on the error field | 22:35 |
| *** nmagnezi has quit IRC | 22:36 | |
| rm_work | yeah but | 22:36 |
| rm_work | we still should just backport that | 22:36 |
| rm_work | as we'd need to backport the other fix separately anyway | 22:37 |
| rm_work | right? | 22:37 |
| *** ajo has joined #openstack-lbaas | 22:38 | |
| *** nmagnezi has joined #openstack-lbaas | 22:42 | |
| johnsom | Yeah | 22:43 |
| johnsom | Well, the hack I recommended just needs to go back to ocata on Octavia side and is fine | 22:44 |
| johnsom | So, independent in that way | 22:44 |
| *** nmagnezi has quit IRC | 22:47 | |
| *** nmagnezi has joined #openstack-lbaas | 22:47 | |
| *** ajo has quit IRC | 22:48 | |
| rm_work | k | 22:50 |
| *** nmagnezi has quit IRC | 22:51 | |
| *** ajo has joined #openstack-lbaas | 22:53 | |
| rm_work | johnsom: I was debating setting v1 from "CURRENT" to "DEPRECATED" | 22:57 |
| rm_work | when I moved v2 from EXPERIMENTAL to CURRENT | 22:58 |
| *** ajo has quit IRC | 22:58 | |
| johnsom | I have been holding off on making the deprecated calls until we have the providers stuff. | 23:00 |
| rm_work | yeah | 23:01 |
| rm_work | that's why i didn't | 23:01 |
| rm_work | so I guess they're just both CURRENT | 23:01 |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 23:02 |
| JudeC | stupid me ^ (doing to many things at once) | 23:02 |
| rm_work | lol | 23:03 |
| rm_work | just need to rebase the other two patches now :P | 23:03 |
| johnsom | Yeah, I will get to it. Just testing the filtering stuff to sign off | 23:04 |
| JudeC | Accidentally had a pep8 error bleh. | 23:04 |
| rm_work | oh were you going to rebase for him? :P | 23:04 |
| johnsom | Yeah, I was going to rebase and add the RBAC stuff | 23:04 |
| johnsom | Hmm, shouldn't this work? | 23:07 |
| johnsom | https://www.irccloud.com/pastebin/rJHNfYRf/ | 23:07 |
| JudeC | yes that for sure should work.... hmm | 23:08 |
| JudeC | very weird, it is working for me right now... | 23:09 |
| johnsom | Yeah, I am an idiot | 23:09 |
| johnsom | didn't escape the second & | 23:09 |
| *** chlong has joined #openstack-lbaas | 23:10 | |
| JudeC | Hey youre not alone, I spent like 20 minutes trying to figure out why something wasn't working and my curl request had a , instead of a & | 23:10 |
| johnsom | Yeah, I'm in that afternoon fog | 23:11 |
| JudeC | my fog lasts until 10pm and then I am actually able to focus properly. | 23:11 |
| johnsom | Ok, I think filtering is in good shape. I opened a bug for that conditional stuff, but that is low priority | 23:11 |
| johnsom | Yeah, after coffee in the morning I am good until about 3, then it's fog until about 5pm ha | 23:12 |
| johnsom | The 5pm part doesn't sit well with the spousal unit | 23:12 |
| johnsom | rm_work https://review.openstack.org/#/c/469275 | 23:13 |
| johnsom | Whenever you are happy with it | 23:13 |
| rm_work | k | 23:13 |
| rm_work | yeah i want to glance but i think it was ok | 23:13 |
| JudeC | Yeah my girlfriend unit is due for an upgrade to a spousal one, but she shares in that view. :P | 23:13 |
| johnsom | Hahaha | 23:14 |
| johnsom | Today is the end of our 4 year upgrade warranty | 23:14 |
| *** ajo has joined #openstack-lbaas | 23:15 | |
| JudeC | Its your anniversary? | 23:16 |
| johnsom | Yes, ran out for flowers over lunch | 23:16 |
| JudeC | ah man get out of here. | 23:16 |
| johnsom | Nah, we both are working today. | 23:16 |
| johnsom | Doing dinner tonight though | 23:16 |
| JudeC | Awesome! well have fun :) | 23:17 |
| *** ajo has quit IRC | 23:20 | |
| JudeC | brb, if anyone needs me. | 23:21 |
| *** ajo has joined #openstack-lbaas | 23:21 | |
| rm_work | JudeC: one quick thing (maybe two) | 23:21 |
| rm_work | commented | 23:21 |
| rm_work | johnsom: you may have to +2 again | 23:22 |
| johnsom | NP | 23:22 |
| JudeC | ah yes looking at your comments now. | 23:22 |
| johnsom | rm_work You led him astray in this status tree patch.... _custom_actions doesn't work the way you think | 23:26 |
| rm_work | damn | 23:26 |
| johnsom | I remember you proposed that before | 23:26 |
| rm_work | I said i wasn't sure | 23:26 |
| rm_work | but that it looked interesting | 23:26 |
| rm_work | and kinda like maybe magic? | 23:26 |
| rm_work | no idea how it actually works or what it does | 23:26 |
| johnsom | Yeah, it OVERRIDES the method | 23:26 |
| rm_work | interesting | 23:27 |
| rm_work | is that .. bad? | 23:27 |
| johnsom | So now we have like three GETs with two of them having the same signature | 23:27 |
| rm_work | lol | 23:27 |
| rm_work | wait so it does what now? | 23:27 |
| johnsom | Maps a HTTP method to a python method | 23:28 |
| rm_work | ah | 23:28 |
| johnsom | So if you didn't want to use the builtin get or get_one | 23:28 |
| rm_work | so it overrides the normal GET /loadbalancers | 23:28 |
| rm_work | doesn't add a method "statuses" with type "GET" | 23:28 |
| johnsom | But in this case GET id/status and GET id both have the same method signature | 23:28 |
| rm_work | errr | 23:29 |
| *** nmagnezi has joined #openstack-lbaas | 23:29 | |
| johnsom | It doesn't grab the path, no | 23:29 |
| rm_work | lol so then what happens | 23:29 |
| johnsom | pecan gets strange | 23:29 |
| JudeC | interesting, dang it that seemed like such an awesome thing too. :/ | 23:29 |
| rm_work | yeah | 23:29 |
| rm_work | so it ... has to use the pattern that /members uses on pool? | 23:30 |
| johnsom | I can't remember exactly, but when the Intel folks tried it, it ended poorly | 23:30 |
| rm_work | rofl | 23:30 |
| rm_work | it seemed really neat if it worked the way I guessed | 23:30 |
| rm_work | ah well | 23:30 |
| johnsom | Yeah, that _lookup stuff | 23:30 |
| rm_work | yep | 23:30 |
| johnsom | _custom_actions maps method names to the list of valid HTTP verbs for those custom actions. | 23:30 |
| johnsom | Well, let me post the rebase at least | 23:31 |
| rm_work | lol k | 23:31 |
| rm_work | sorry JudeC, you've got some fixin' to do | 23:31 |
| *** gongysh has joined #openstack-lbaas | 23:31 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Add status tree to V2 API https://review.openstack.org/474390 | 23:31 |
| rm_work | hopefully you're fixin' for a fixin' | 23:31 |
| JudeC | I know... T_T | 23:32 |
| johnsom | If I didn't have a hard stop today I would do it, but.... | 23:32 |
| rm_work | lol np | 23:32 |
| rm_work | johnsom: what did you think of the disablers | 23:32 |
| rm_work | :thumbsup: :thumbsdown: ? | 23:32 |
| johnsom | I think it is a fine idea. I just wanted to kick the tires. I can do that now | 23:33 |
| rm_work | ok np i was just looking for the "yeah that seems like an ok approach" | 23:33 |
| rm_work | vs "plz no" | 23:33 |
| *** nmagnezi has quit IRC | 23:33 | |
| *** gongysh has quit IRC | 23:33 | |
| johnsom | I think we need to clean up the API validators, they are all over the place. | 23:33 |
| rm_work | yes | 23:34 |
| johnsom | I also think these enable/disable things should be grouped in the config file. I was going to look at that again to make sure they are close | 23:34 |
| rm_work | i know | 23:34 |
| rm_work | i was considering it | 23:34 |
| johnsom | [api_features]? | 23:35 |
| openstackgerrit | Jude Cross proposed openstack/octavia master: Add filtering and field selection to API https://review.openstack.org/469275 | 23:35 |
| rm_work | hmmm | 23:35 |
| rm_work | yeah maybe | 23:35 |
| rm_work | i was considering both that and moving all the validations into like a common self._run_api_validations() thing | 23:35 |
| johnsom | Yeah, it just seems odd to drop those all in [DEFAULT] | 23:36 |
| johnsom | Not that there isn't already API "stuff" in the DEFAULT section | 23:36 |
| *** gongysh has joined #openstack-lbaas | 23:37 | |
| *** gongysh has quit IRC | 23:38 | |
| *** ajo has quit IRC | 23:38 | |
| johnsom | Ugh, with this wsgi stuff those don't log until the first API request | 23:41 |
| *** ajo has joined #openstack-lbaas | 23:43 | |
| *** nmagnezi has joined #openstack-lbaas | 23:45 | |
| *** armax has quit IRC | 23:46 | |
| *** ajo has quit IRC | 23:48 | |
| *** nmagnezi has quit IRC | 23:49 | |
| johnsom | Ok, commented. | 23:52 |
| *** SumitNaiksatam has quit IRC | 23:53 | |
| *** isotope has quit IRC | 23:53 | |
| *** ajo has joined #openstack-lbaas | 23:55 | |
| *** nmagnezi has joined #openstack-lbaas | 23:55 | |
| *** nmagnezi has quit IRC | 23:59 | |
| *** ajo has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!