| *** amotoki is now known as amotoki_away | 00:17 | |
| *** amotoki_away is now known as amotoki | 00:18 | |
| *** sshank has quit IRC | 00:19 | |
| *** amotoki is now known as amotoki_away | 00:27 | |
| *** sanfern has quit IRC | 00:34 | |
| *** sanfern has joined #openstack-lbaas | 00:36 | |
| *** junbo has quit IRC | 00:45 | |
| rm_work | johnsom: AH so | 00:48 |
|---|---|---|
| rm_work | johnsom: if a failover fails, the LB is still ACTIVE, then you try to make a change, the change will fail due to the Amp being unreachable, then your LB is stuck in ERROR | 00:49 |
| rm_work | retry the failover, it works, LB is back up, but still in ERROR | 00:49 |
| *** junbo has joined #openstack-lbaas | 00:49 | |
| johnsom | Ummm, that doesn't sound like the upstream logic at all | 00:50 |
| johnsom | If failover fails it should always go to ERROR from the capstone task | 00:51 |
| rm_work | erm | 00:56 |
| rm_work | i'm not seeing that | 00:56 |
| rm_work | like failover BROKE in the middle of the run | 00:56 |
| rm_work | with an exception | 00:56 |
| rm_work | there | 00:57 |
| rm_work | *there's: AmphoraToErrorOnRevertTask | 00:57 |
| rm_work | but | 00:57 |
| rm_work | I see nothing that would send the LB to error | 00:57 |
| rm_work | nor anything that would send it back to ACTIVE once a failover completes | 00:57 |
| rm_work | (the same behavior could probably be replicated just by trying to do an action DURING a failover) | 00:58 |
| rm_work | johnsom: can you show me where the failover flow would make the *LB* immutable? | 00:58 |
| johnsom | Oh, you are right! it is not acting on the LB... Oye | 01:03 |
| rm_work | T_T | 01:03 |
| johnsom | SoB | 01:04 |
| johnsom | That thing has been through a few iterations. | 01:04 |
| rm_work | I'm going to find ALL the bugs. In like a week. because production | 01:05 |
| rm_work | <_< | 01:05 |
| * rm_work cries | 01:05 | |
| johnsom | So, we need to somehow detect if this is a spares or in-use and lock the LB | 01:05 |
| rm_work | we know this | 01:05 |
| rm_work | # If this is an unallocated amp (spares pool), we're done | 01:05 |
| rm_work | if status != constants.AMPHORA_ALLOCATED: | 01:05 |
| rm_work | I'm already futzing with my version of the failover | 01:06 |
| rm_work | I can try to fix them both | 01:06 |
| rm_work | just, what do we do | 01:06 |
| rm_work | mark the LB ... PENDING_UPDATE? | 01:06 |
| rm_work | and then back | 01:06 |
| rm_work | ? | 01:06 |
| rm_work | do we need to lock it? | 01:06 |
| johnsom | Yeah, just like an PUT call | 01:06 |
| johnsom | Yeah, I think we should definitely lock it | 01:06 |
| rm_work | like, actually do a lock attempt? | 01:07 |
| rm_work | i mean, lock yes, but *run the lock-get code* | 01:07 |
| rm_work | I think no | 01:07 |
| rm_work | because we need to just ... override whatever | 01:07 |
| rm_work | but then we'll set it to ACTIVE at the end? | 01:07 |
| rm_work | do we need to save the existing status? | 01:07 |
| rm_work | IDK | 01:07 |
| johnsom | So, we need to basically call this: | 01:09 |
| johnsom | self._test_lb_and_listener_statuses( | 01:09 |
| johnsom | context.session, lb_id=db_pool.load_balancer_id, | 01:09 |
| johnsom | listener_ids=self._get_affected_listener_ids(db_pool)) | 01:09 |
| johnsom | We are only acting on provisioning status, operating is measured and not a factor here. | 01:09 |
| rm_work | right but | 01:09 |
| johnsom | Then octavia.controller.worker.tasks.database_tasks.MarkLBAndListenersActiveInDB | 01:09 |
| rm_work | but | 01:09 |
| rm_work | what if the amp crashes DURING an update | 01:10 |
| rm_work | the LB is in PENDING_UPDATE | 01:10 |
| rm_work | we send a new config | 01:10 |
| rm_work | something about that crashes the amp agent | 01:10 |
| rm_work | amp goes down | 01:10 |
| rm_work | we try to lock | 01:10 |
| rm_work | fail | 01:10 |
| johnsom | Wait, we are talking failover right? | 01:10 |
| rm_work | yes | 01:10 |
| johnsom | Any user action is locked with test_lb_and_listener_statuses | 01:10 |
| rm_work | yes | 01:11 |
| rm_work | but i'm saying | 01:11 |
| johnsom | So, no other controller or API can touch it. | 01:11 |
| rm_work | that also CHECKS immutability | 01:11 |
| rm_work | what if it's ALREADY locked | 01:11 |
| rm_work | when we need to failover | 01:11 |
| johnsom | gotcha | 01:11 |
| rm_work | so, OK -- we ignore the check, and just straight up set the status | 01:11 |
| rm_work | so, at the end, we reset to ACTIVE? | 01:12 |
| rm_work | what happens to the status it was in? | 01:12 |
| rm_work | what if we failover during a PENDING_DELETE? (is that possible?) | 01:12 |
| johnsom | I think we should wait for the locking flow to release the lock. Hmmm | 01:12 |
| johnsom | No, delete should set HM busy to block failovers | 01:12 |
| johnsom | Yeah, LB delete marks HM busy | 01:13 |
| johnsom | Can't start a failover | 01:13 |
| johnsom | The locking flow *should* fail and go to ERROR | 01:13 |
| johnsom | Right? | 01:14 |
| rm_work | err | 01:15 |
| rm_work | so a failover IMO is #1 priority | 01:15 |
| johnsom | I mean ignoring the lock is *bad* IMO. It means two controller processes could be *TRYING* to act on the same entity | 01:15 |
| rm_work | any operation is invalid (besides DELETE?) if we have no amp | 01:15 |
| rm_work | i think the first thing we do is just SET the status of the LB, period | 01:16 |
| rm_work | though it'd be nice to know what the status WAS | 01:16 |
| johnsom | Yeah, I don't think so. If it was ERROR the LB is in unkown state land and should stay in ERROR. The user then has the choice of calling up the operator or deleting and re-creating | 01:17 |
| rm_work | if something is stuck in PENDING_UPDATE, it could still be serving traffic, and if we just don't failover because it's in that state, we have downtime | 01:17 |
| rm_work | hmm k | 01:17 |
| rm_work | but ERROR is actually fine | 01:17 |
| rm_work | since it's not immutable :P | 01:17 |
| rm_work | so we'd get a lock | 01:17 |
| johnsom | We should never get "stuck" That is fail on our part | 01:17 |
| johnsom | Yeah, but failing over ERROR is just going farther into "unknown state" land | 01:18 |
| rm_work | ok but | 01:18 |
| rm_work | that means we should... check the state first? | 01:18 |
| johnsom | ERROR is the "Oh Sh1t" state | 01:18 |
| rm_work | right but | 01:18 |
| rm_work | test_lb_and_listener_statuses | 01:18 |
| rm_work | will pass and give us a lock | 01:18 |
| rm_work | right? | 01:18 |
| rm_work | on ERROR | 01:18 |
| johnsom | No, I don't think so | 01:18 |
| rm_work | ERROR isn't immutable? | 01:19 |
| rm_work | I thought | 01:19 |
| johnsom | I am using that as an example, I don't think we can just use that code, but it should be similar | 01:19 |
| rm_work | hmmmmmmm | 01:20 |
| rm_work | ok | 01:20 |
| rm_work | we need to map this out I think | 01:20 |
| rm_work | i don't know whether "spec" style writeup or just ... code is better | 01:21 |
| rm_work | to start the conversation | 01:21 |
| johnsom | https://github.com/openstack/octavia/blob/master/octavia/db/repositories.py#L641 | 01:21 |
| johnsom | Yeah, in the back of my mind a state diagram would be nice, but, see above, hate graphviz | 01:22 |
| rm_work | MUTABLE_STATES includes ERROR, right? | 01:22 |
| johnsom | So, delete is special case and bypassed ERROR | 01:22 |
| rm_work | also: that was a weird way to do that | 01:22 |
| johnsom | Yeah, agreed that code is ummm | 01:23 |
| rm_work | and the docstring is wrong | 01:23 |
| rm_work | actively wrong | 01:23 |
| rm_work | lol | 01:23 |
| *** amotoki_away is now known as amotoki | 01:23 | |
| rm_work | oh hmm | 01:24 |
| rm_work | I see | 01:24 |
| johnsom | So, it's date night here. I'm already getting an eye. | 01:24 |
| rm_work | ERROR *is* immutable except for a delete | 01:24 |
| rm_work | alright | 01:24 |
| rm_work | later :P | 01:24 |
| rm_work | i was wrong about ERROR immutability | 01:24 |
| rm_work | so fine | 01:24 |
| rm_work | alright | 01:24 |
| johnsom | Yeah, it would be bad to allow update on ERROR | 01:24 |
| johnsom | I will try to check in later and see if you need to bounce something off mee | 01:26 |
| rm_work | i have it fixed kinda in my flow, going to port that to the main flow and put it up and call it a day | 01:41 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Floating IP Network Driver (spans L3s) https://review.openstack.org/435612 | 01:43 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Failover should cause LB to be Immutable https://review.openstack.org/479109 | 01:56 |
| rm_work | ^^ there | 01:56 |
| *** sanfern has quit IRC | 02:25 | |
| *** csomerville has quit IRC | 02:33 | |
| *** csomerville has joined #openstack-lbaas | 02:33 | |
| *** cody-somerville has joined #openstack-lbaas | 02:35 | |
| *** csomerville has quit IRC | 02:38 | |
| *** aojea has joined #openstack-lbaas | 02:50 | |
| *** aojea has quit IRC | 02:54 | |
| *** cody-somerville has quit IRC | 02:58 | |
| *** yamamoto has joined #openstack-lbaas | 03:05 | |
| *** kobis has joined #openstack-lbaas | 03:19 | |
| *** gans has joined #openstack-lbaas | 03:25 | |
| *** armax has joined #openstack-lbaas | 03:31 | |
| *** kobis has quit IRC | 03:35 | |
| *** sanfern has joined #openstack-lbaas | 03:38 | |
| *** armax has quit IRC | 03:48 | |
| openstackgerrit | Santhosh Fernandes proposed openstack/octavia master: This patch will emit provisioning status to be sync with neutron db https://review.openstack.org/478385 | 03:49 |
| *** gans has quit IRC | 04:06 | |
| *** diltram has quit IRC | 04:18 | |
| *** diltram has joined #openstack-lbaas | 04:23 | |
| *** yuanying has quit IRC | 04:24 | |
| *** yuanying has joined #openstack-lbaas | 04:24 | |
| *** diltram has quit IRC | 04:37 | |
| *** diltram has joined #openstack-lbaas | 04:40 | |
| *** aojea has joined #openstack-lbaas | 05:27 | |
| *** yuanying_ has joined #openstack-lbaas | 05:32 | |
| *** yuanying has quit IRC | 05:34 | |
| *** aojea has quit IRC | 05:42 | |
| *** rcernin has joined #openstack-lbaas | 05:45 | |
| *** aojea has joined #openstack-lbaas | 06:00 | |
| *** aojea has quit IRC | 06:01 | |
| *** links has quit IRC | 06:07 | |
| *** links has joined #openstack-lbaas | 06:21 | |
| *** diltram has quit IRC | 06:22 | |
| *** diltram has joined #openstack-lbaas | 06:29 | |
| *** sticker has quit IRC | 06:31 | |
| *** diltram has quit IRC | 06:35 | |
| *** diltram has joined #openstack-lbaas | 06:42 | |
| *** pcaruana has joined #openstack-lbaas | 06:45 | |
| *** tesseract has joined #openstack-lbaas | 07:14 | |
| *** kobis has joined #openstack-lbaas | 07:17 | |
| *** aojea has joined #openstack-lbaas | 07:25 | |
| *** kobis has quit IRC | 07:30 | |
| *** gtrxcb has quit IRC | 07:38 | |
| openstackgerrit | cheng proposed openstack/octavia master: Fix url_path valid check https://review.openstack.org/479220 | 09:17 |
| *** yamamoto has quit IRC | 09:35 | |
| *** yamamoto has joined #openstack-lbaas | 09:36 | |
| *** yamamoto has quit IRC | 09:38 | |
| *** yamamoto has joined #openstack-lbaas | 09:42 | |
| *** yamamoto has quit IRC | 09:42 | |
| *** links has quit IRC | 09:44 | |
| *** yamamoto has joined #openstack-lbaas | 09:46 | |
| *** yamamoto has quit IRC | 09:51 | |
| *** links has joined #openstack-lbaas | 10:01 | |
| *** yamamoto has joined #openstack-lbaas | 10:10 | |
| *** yamamoto has quit IRC | 10:12 | |
| *** yamamoto has joined #openstack-lbaas | 10:13 | |
| *** yamamoto has quit IRC | 10:16 | |
| *** yamamoto has joined #openstack-lbaas | 10:16 | |
| *** yamamoto has quit IRC | 10:18 | |
| *** diltram has quit IRC | 10:33 | |
| *** diltram has joined #openstack-lbaas | 10:37 | |
| *** diltram has quit IRC | 10:42 | |
| *** diltram has joined #openstack-lbaas | 10:47 | |
| *** yamamoto has joined #openstack-lbaas | 10:52 | |
| *** diltram has quit IRC | 10:55 | |
| *** sanfern has quit IRC | 10:57 | |
| *** diltram has joined #openstack-lbaas | 11:02 | |
| *** amotoki is now known as amotoki_away | 11:11 | |
| *** rtjure has joined #openstack-lbaas | 11:32 | |
| *** yamamoto has quit IRC | 11:34 | |
| *** diltram has quit IRC | 11:35 | |
| *** yamamoto has joined #openstack-lbaas | 11:36 | |
| *** diltram has joined #openstack-lbaas | 11:46 | |
| *** diltram has quit IRC | 11:51 | |
| *** diltram has joined #openstack-lbaas | 11:56 | |
| *** yamamoto has quit IRC | 11:58 | |
| *** diltram has quit IRC | 12:06 | |
| *** diltram has joined #openstack-lbaas | 12:13 | |
| *** amotoki_away is now known as amotoki | 12:19 | |
| *** yamamoto has joined #openstack-lbaas | 12:26 | |
| *** yamamoto has quit IRC | 12:55 | |
| *** yamamoto has joined #openstack-lbaas | 12:56 | |
| *** yamamoto has quit IRC | 13:01 | |
| *** diltram has quit IRC | 13:04 | |
| *** diltram has joined #openstack-lbaas | 13:08 | |
| *** links has quit IRC | 13:20 | |
| *** sanfern has joined #openstack-lbaas | 13:39 | |
| *** yamamoto has joined #openstack-lbaas | 13:58 | |
| *** reedip_ has joined #openstack-lbaas | 13:59 | |
| *** amotoki is now known as amotoki_away | 14:04 | |
| *** sanfern has quit IRC | 14:08 | |
| *** sanfern has joined #openstack-lbaas | 14:08 | |
| *** yamamoto has quit IRC | 14:10 | |
| johnsom | o/ | 14:12 |
| johnsom | Looks like devstack was broken last night (not our code). I will kick off a recheck to see if it is fixed now | 14:13 |
| johnsom | https://www.irccloud.com/pastebin/ZdNrTlRN/ | 14:14 |
| *** reedip_ has quit IRC | 14:18 | |
| sanfern | johnsom, yes my patch failed :( | 14:21 |
| johnsom | Yeah, a whole bunch of them did | 14:22 |
| *** aojea has quit IRC | 14:34 | |
| xgerman_ | o/ | 14:53 |
| *** rcernin has quit IRC | 15:11 | |
| johnsom | It looks like devstack is fixed | 15:11 |
| *** aojea has joined #openstack-lbaas | 15:47 | |
| xgerman_ | are healthmonitors top level nowerdays as well? | 15:49 |
| *** aojea has quit IRC | 15:51 | |
| *** diltram has quit IRC | 15:53 | |
| johnsom | We have a handy API reference for that: https://developer.openstack.org/api-ref/load-balancer/v2/index.html#health-monitor | 15:53 |
| johnsom | Grin | 15:53 |
| johnsom | Object wise they are still under pools, but the API supports the concept of them being independent | 15:54 |
| johnsom | It is actually a wonky-ness in the LBaaSv2 API if you ask me | 15:55 |
| *** diltram has joined #openstack-lbaas | 15:59 | |
| *** mdavidson has quit IRC | 16:03 | |
| *** cody-somerville has joined #openstack-lbaas | 16:03 | |
| *** cody-somerville has joined #openstack-lbaas | 16:03 | |
| *** csomerville has joined #openstack-lbaas | 16:07 | |
| *** cody-somerville has quit IRC | 16:10 | |
| *** tesseract has quit IRC | 16:11 | |
| openstackgerrit | QingchuanHao proposed openstack/neutron-lbaas master: Fix object's dict variable can be changed unexpectedly https://review.openstack.org/479364 | 16:23 |
| openstackgerrit | QingchuanHao proposed openstack/neutron-lbaas master: Fix object's dict variable can be changed unexpectedly https://review.openstack.org/479364 | 16:24 |
| rm_work | johnsom: i've started to actually be ok with things being top-level | 16:57 |
| rm_work | which makes me feel dumb previously | 16:57 |
| rm_work | but i think maybe it was not explained well | 16:57 |
| rm_work | like if i get a unique ID for an object | 16:59 |
| rm_work | why should i have to track a bunch of other IDs | 16:59 |
| rm_work | just to look it up | 16:59 |
| rm_work | i know what i want | 16:59 |
| rm_work | it's the pool+member thing we dealt with in the client | 17:00 |
| rm_work | "why do i suddenly need a poolID to look up a member I have a guid for" | 17:00 |
| johnsom | Yeah, my comment is we are not consistent with it, which is a bit annoying | 17:01 |
| *** aojea has joined #openstack-lbaas | 17:04 | |
| *** sshank has joined #openstack-lbaas | 17:14 | |
| xgerman_ | healthmonitor on top makes sense since our idea was to share them | 17:15 |
| xgerman_ | same with pools | 17:15 |
| xgerman_ | listeners | 17:15 |
| xgerman_ | on the same token we could do members - what a rabbit hole | 17:15 |
| rm_work | i was always in favor of | 17:15 |
| rm_work | everything is both | 17:15 |
| rm_work | it's not like it's hard | 17:15 |
| johnsom | Ugh, this new theme for the docs is making me.... Grumpy | 17:17 |
| *** reedip_ has joined #openstack-lbaas | 17:35 | |
| rm_work | johnsom: oh, it's not good? | 17:40 |
| rm_work | i glanced at it in the patch for it and it looked a little better I thought | 17:41 |
| johnsom | It has worts IMO. Like wanted a Contents on each page (I have already disabled this) and then sub-section titles being bold where the section titles are not. Some of the font and layout is odd too | 17:42 |
| rm_work | johnsom: uhhh did this test not get updated with the rest of the code, but everything still passed? lolol https://review.openstack.org/#/c/478859/ | 17:42 |
| johnsom | We will see what I come up with | 17:42 |
| *** pcaruana has quit IRC | 17:42 | |
| johnsom | Evidently | 17:43 |
| rm_work | ugh i thought that test-and-set would raise | 17:45 |
| *** sanfern has quit IRC | 18:03 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Failover should cause LB to be Immutable https://review.openstack.org/479109 | 18:26 |
| rm_work | johnsom: ^^ ok there you go, that is what I meant to have happen | 18:26 |
| johnsom | Ok | 18:26 |
| *** reedip_ has quit IRC | 18:29 | |
| *** tomtomtom has joined #openstack-lbaas | 18:55 | |
| tomtomtom | hi, anyone know what the "handler" would be from the octavia-api.log in this message? 2017-06-30 18:49:58.483 19126 INFO octavia.api.v1.controllers.load_balancer [req-770b6ae0-44da-4c60-b0a5-3766f2c22ea4 - 8826fa87d1544309bc4f36e650e37eb2 - default default] Sending created Load Balancer 34e270c0-d483-46a9-9079-beae82ee99d2 to the handler | 18:55 |
| tomtomtom | It's configured in openstack newton with neutron | 18:56 |
| tomtomtom | thanks in advance for any help | 18:56 |
| johnsom | tomtomtom It is the handler that will take action on the request. It is set in the configuration file: api_handler = queue_producer | 18:57 |
| johnsom | queue_producer is a bad name for octavia handler | 18:57 |
| johnsom | It puts the request on an oslo messaging queue which one of the octavia controller worker processes will pick up. | 18:58 |
| johnsom | This will be changing, likely in queens as we add support for other providers (vendor drivers). | 18:58 |
| tomtomtom | ok so somewhere in the creation of a load balancer I end up getting an error, would the next logical place to look be the octavia-worker log? | 18:58 |
| johnsom | Yes, o-cw log | 18:59 |
| tomtomtom | thank you | 19:00 |
| johnsom | Sure, come back if you need help understanding the error | 19:00 |
| *** aojea has quit IRC | 19:09 | |
| tomtomtom | Yeah it's a failed to create compute instance, and it looks like it's complaining about this specifically: 2017-06-30 18:50:00.919 250 ERROR octavia.compute.drivers.nova_driver [-] Error building nova virtual machine.: GlanceNoTaggedImages: No Glance images are tagged with octavia-amphora-image tag. | 19:13 |
| tomtomtom | So I probably need a glance image with that tag it can use to build the compute instance. | 19:13 |
| johnsom | Yes, you need an amphora image setup | 19:14 |
| tomtomtom | Ok I will check the docs on that then. Thank you for your help. | 19:14 |
| johnsom | https://docs.openstack.org/developer/octavia/guides/dev-quick-start.html#create-amphora-image | 19:14 |
| johnsom | diskimage-create -s 2 | 19:15 |
| tomtomtom | awesome! thanks | 19:18 |
| *** m-greene- has joined #openstack-lbaas | 19:21 | |
| *** m-greene has quit IRC | 19:22 | |
| openstackgerrit | Carlos Puga proposed openstack/octavia master: Spec detailing Octavia service flavors support https://review.openstack.org/392485 | 19:55 |
| *** aojea has joined #openstack-lbaas | 20:01 | |
| *** cpuga has joined #openstack-lbaas | 20:01 | |
| *** blogan has quit IRC | 21:05 | |
| rm_work | xgerman_: https://review.openstack.org/#/c/478693/ | 21:23 |
| rm_work | johnsom: once that merges you can make a good release right? | 21:23 |
| xgerman_ | k | 21:23 |
| johnsom | rm_work yeah, what about that other patch in there? | 21:23 |
| johnsom | https://review.openstack.org/#/c/478724/ | 21:24 |
| johnsom | Oh, I guess it's just tests so | 21:24 |
| rm_work | though that LGTM too | 21:26 |
| johnsom | I down from over 100 errors in our docs and docstrings to <20... So getting close. | 21:27 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Allow PUT to /pools/<id>/members to batch update members https://review.openstack.org/477034 | 21:28 |
| rm_work | johnsom: \o/ yay | 21:28 |
| rm_work | eventually I would like to get people to look at that | 21:28 |
| rm_work | ideally before feature freeze | 21:28 |
| rm_work | but not a SUPER hurry | 21:28 |
| rm_work | obviously we've got some bugs that are higher priority :P | 21:28 |
| johnsom | Bonus here is we will be doing syntax checking on docstrings going forward | 21:30 |
| *** aojea has quit IRC | 21:31 | |
| *** cpuga has quit IRC | 21:32 | |
| *** diltram has quit IRC | 21:38 | |
| *** diltram has joined #openstack-lbaas | 21:43 | |
| *** gtrxcb has joined #openstack-lbaas | 21:45 | |
| rm_work | i thought we already did | 21:53 |
| rm_work | T_T | 21:53 |
| johnsom | I can assure you, my morning has demonstrated that we are not | 21:53 |
| rm_work | lol. | 22:02 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update Octavia docs for documentation migration https://review.openstack.org/478947 | 22:03 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update Octavia docs for documentation migration https://review.openstack.org/478947 | 22:12 |
| johnsom | Ok this is odd: | 22:18 |
| johnsom | http://logs.openstack.org/93/478693/1/gate/gate-python-octaviaclient-pep8-ubuntu-xenial/642b8c6/console.html#_2017-06-30_21_27_01_002190 | 22:18 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update Octavia docs for documentation migration https://review.openstack.org/478947 | 22:20 |
| *** sshank has quit IRC | 22:21 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update Octavia docs for documentation migration https://review.openstack.org/478947 | 22:27 |
| openstackgerrit | Michael Johnson proposed openstack/python-octaviaclient master: Remove rogue calls to str.upper in l7rules https://review.openstack.org/478693 | 22:29 |
| openstackgerrit | Michael Johnson proposed openstack/python-octaviaclient master: Remove rogue calls to str.upper in l7rules https://review.openstack.org/478693 | 22:57 |
| openstackgerrit | Michael Johnson proposed openstack/python-octaviaclient master: Remove rogue calls to str.upper in l7rules https://review.openstack.org/478693 | 23:09 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Update Octavia docs for documentation migration https://review.openstack.org/478947 | 23:12 |
| openstackgerrit | Michael Johnson proposed openstack/python-octaviaclient master: Remove rogue calls to str.upper in l7rules https://review.openstack.org/478693 | 23:14 |
| johnsom | Not my day | 23:14 |
| *** armax has joined #openstack-lbaas | 23:22 | |
| *** armax has quit IRC | 23:41 | |
| rm_work | T_T | 23:45 |
| rm_work | johnsom: figure most of it out? | 23:45 |
| johnsom | Yes | 23:46 |
| johnsom | There is trickery for dealing with upper constraints. So I added it to your patch | 23:46 |
| rm_work | ah | 23:47 |
| rm_work | lol | 23:47 |
| rm_work | that makes | 23:47 |
| rm_work | ok | 23:47 |
| rm_work | i am not sure i get it but ok | 23:47 |
| rm_work | :P | 23:47 |
| johnsom | Yeah | 23:47 |
| rm_work | need xgerman_ on https://review.openstack.org/478693 | 23:48 |
| rm_work | i'm heading out a little early | 23:48 |
| rm_work | but PM me if you need any reviews tonight/weekend | 23:48 |
| johnsom | Ok, I think I am good at the moment. Would like feedback on the new docs layout when you are bored and have a few: http://docs-draft.openstack.org/47/478947/7/check/gate-octavia-docs-ubuntu-xenial/606bb7e//doc/build/html/ | 23:51 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!