| *** BjoernT has quit IRC | 00:09 | |
| *** madhu_ak has quit IRC | 01:02 | |
| *** purandar_k has quit IRC | 01:07 | |
| *** Purandar has joined #openstack-lbaas | 01:08 | |
| *** kevo has quit IRC | 01:09 | |
| reedip | isnt health monitor just for monitorig | 01:13 |
|---|---|---|
| reedip | i mean pool and member have a base-derived membership but health-monitor and pool have somewhat of a Friend class( of C++) type relationship? | 01:14 |
| *** Purandar has quit IRC | 01:21 | |
| *** alejandrito has joined #openstack-lbaas | 01:26 | |
| fnaval | reedip: the only difference that i see between members vs health monitors is that only one health monitor can be associated to a pool vs more than one member can be associated to a pool | 01:31 |
| fnaval | but API/object design wise, I don't think there should be any difference between a hm and a member | 01:32 |
| reedip | hi fnaval, I was going through your bug :) and thats why came here | 01:32 |
| fnaval | from what I can see in the API, in order to move a member to another pool, we would have to remove the member from the pool then add the member to another pool. 2 steps vs. 1 step(update a member's pool_id to point to the new pool) | 01:32 |
| fnaval | =) | 01:32 |
| fnaval | not really a bug - but a design enhancement request | 01:32 |
| reedip | thats a bug :P | 01:33 |
| reedip | depends on viewpoint | 01:33 |
| fnaval | kk | 01:33 |
| fnaval | well, i think this bug that i just wrote up might be more critical: https://bugs.launchpad.net/neutron/+bug/1571909 | 01:33 |
| openstack | Launchpad bug 1571909 in neutron "Neutron-LBaaS v2: Deleting pool that has members changes state of other load balancers" [Undecided,New] | 01:33 |
| fnaval | i haven't had any re-verify it though | 01:33 |
| fnaval | anyone* | 01:33 |
| reedip | anyways, I think you may have a point in simplifying the update of members from user point of view | 01:34 |
| reedip | lets see what the core LBs have to say | 01:34 |
| reedip | blogan, sbalukoff : any comments^^ | 01:35 |
| fnaval | thanks for looking into it reedip | 01:35 |
| reedip | naah I didnt do anything much ... | 01:35 |
| fnaval | thanks for acknowledging it then! =) | 01:36 |
| reedip | :) | 01:36 |
| *** ducttape_ has joined #openstack-lbaas | 01:37 | |
| *** Purandar has joined #openstack-lbaas | 01:49 | |
| *** woodster_ has quit IRC | 02:08 | |
| *** ducttape_ has quit IRC | 02:19 | |
| *** alejandrito has quit IRC | 02:41 | |
| *** fnaval has quit IRC | 02:44 | |
| *** fnaval has joined #openstack-lbaas | 02:45 | |
| *** Alex_Stef has quit IRC | 03:01 | |
| *** jorgem| has quit IRC | 03:17 | |
| *** jorgem has joined #openstack-lbaas | 03:17 | |
| *** bana_k has joined #openstack-lbaas | 03:18 | |
| *** amotoki has quit IRC | 03:20 | |
| *** neelashah has joined #openstack-lbaas | 03:29 | |
| *** armax has quit IRC | 03:30 | |
| *** intr1nsic has quit IRC | 03:41 | |
| *** links has joined #openstack-lbaas | 03:41 | |
| *** amotoki has joined #openstack-lbaas | 03:41 | |
| *** neelashah has quit IRC | 04:01 | |
| *** amotoki has quit IRC | 04:03 | |
| *** intr1nsic has joined #openstack-lbaas | 04:07 | |
| *** fnaval has quit IRC | 04:18 | |
| *** ducttape_ has joined #openstack-lbaas | 04:20 | |
| kong | hi, guys, I always encounter this bug when I am testing on devstack (vm creation is easy to fail because of insufficient resource) | 04:21 |
| kong | then the loadbalancer is in ERROR in neutron, PENDING_CREATE in octavia | 04:21 |
| kong | https://bugs.launchpad.net/octavia/+bug/1567463 | 04:22 |
| openstack | Launchpad bug 1567463 in octavia "Trying to create a load balancer while no avaliable compute node ends with a non-deletable loadbalancer" [Undecided,Confirmed] | 04:22 |
| *** kevo has joined #openstack-lbaas | 04:23 | |
| *** ducttape_ has quit IRC | 04:25 | |
| *** intr1nsic has quit IRC | 04:27 | |
| *** fnaval has joined #openstack-lbaas | 04:29 | |
| *** yamamoto has quit IRC | 04:31 | |
| *** amotoki has joined #openstack-lbaas | 04:33 | |
| *** intr1nsic has joined #openstack-lbaas | 04:33 | |
| blogan | kong: thats bc there's a timeout value neutron-lbaas has to poll octavia, once that timeout is hit it throws it into ERROR state | 04:52 |
| blogan | reedip: that's scary | 04:52 |
| blogan | oh i mean fnaval | 04:53 |
| blogan | i read the author wrong | 04:53 |
| kong | blogan: yes, I know that periodic polling. after ERROR state, users can not delete that lb | 04:54 |
| kong | the lb status in octavia db is pending_creating forever | 04:54 |
| kong | pending_create | 04:54 |
| blogan | fnaval, reedip: as for simplifying the update of members by updating the pool_id of a member, i think just deleting the member and recreating another pool is fine for now, unless htere's a huge calling for the 1 call vs 2 | 04:54 |
| blogan | kong: ah i would hope the octavia one would also eventually error out and then an ERROR'ed LB can be deleted | 04:55 |
| blogan | however brings up the issues of the PENDING_UPDATE not allowing a delete | 04:55 |
| blogan | which honestly, a delete of a load balancer would be nice to be able to do at any time | 04:56 |
| reedip | blogan: no, its not a big issue, but what I feel fnaval wishes is to have a similar call flow or execution flow from user perspective | 04:56 |
| blogan | reedip: oh its a good point, but since in the URI the member is tied directly to a pool, updating the pool_id to another pool's id while still keeping the pool_id in the URL is going ot be a bit odd | 04:57 |
| kong | blogan: do you think it makes sense if we provide a worker that do the check work periodically? or we just leave that thing to scripts operators should provide :-) | 04:57 |
| *** cody-somerville has quit IRC | 04:57 | |
| blogan | kong: check the status you mean? | 04:57 |
| blogan | kong: and throw it in ERROR if stalled? | 04:58 |
| kong | blogan: yes, timeout machanism | 04:58 |
| kong | so lb has a chance to be error | 04:58 |
| kong | so can be deleted | 04:58 |
| reedip | blogan: user can perform an update at the CLI level/Horizon , and internally the CLI/Horizon can call delete-member and add-member but yes the UID of the member would then be different | 04:58 |
| blogan | kong: that honestly should just be part of any create/update/delete workflow | 04:58 |
| blogan | in octavia | 04:58 |
| kong | is that what we suppose to solve in Newton? | 04:59 |
| blogan | reedip: yeah, i'll admit the members being children of the pools in the URI is just odd, and it was probably my doing for whatever insane reason i had at the time | 04:59 |
| blogan | i still prefer showing parent/child relationships in the URI but it's not consistent with the rest of the API in this case | 05:00 |
| blogan | kong: that would be a good thing to solve! | 05:00 |
| blogan | kong: stability and reliability need to be improved immensely, this would help | 05:00 |
| kong | blogan: yeah, especially for production usage like us :-) | 05:01 |
| blogan | kong: us too, soon :) | 05:01 |
| reedip | I actually like kong's idea of the timeout but I think something like this is already implemented, isnt it blogan? | 05:01 |
| kong | not in octavia side, reedip | 05:02 |
| blogan | reedip: i could have sworn i reported that as a bug and it did get created | 05:02 |
| blogan | now i must go check | 05:02 |
| *** cody-somerville has joined #openstack-lbaas | 05:03 | |
| blogan | oh my bug was to immediately throw the LB into ERROR whenever the nova instance goes ERROR | 05:04 |
| blogan | kong: are the nova VMs staying in build? | 05:04 |
| kong | blogan: actually, my vm stuck in spawning for some reason | 05:05 |
| blogan | kong: yeah we probably need a timeout on waiting for the nova instance to go active | 05:06 |
| blogan | actually i thought we had that | 05:06 |
| blogan | now i must go check again :) | 05:06 |
| kong | blogan: ok, thanks | 05:06 |
| kong | guys, I need to head off | 05:07 |
| * kong going home :-) | 05:07 | |
| blogan | kong: what are your config options for amp_active_retries? | 05:08 |
| blogan | and amp_active_wait_sec | 05:08 |
| blogan | under the [controller_worker] section | 05:08 |
| blogan | unless you already went home | 05:08 |
| blogan | in which case you can answer later :) | 05:08 |
| *** rcernin has joined #openstack-lbaas | 05:10 | |
| *** yamamoto has joined #openstack-lbaas | 05:14 | |
| *** chlong has quit IRC | 05:16 | |
| *** SteF has joined #openstack-lbaas | 05:17 | |
| *** SteF is now known as Alex_Staf | 05:17 | |
| *** cody-somerville has quit IRC | 05:43 | |
| *** anilvenkata has joined #openstack-lbaas | 05:55 | |
| *** Purandar has quit IRC | 06:04 | |
| *** ducttape_ has joined #openstack-lbaas | 06:21 | |
| *** numan_ has joined #openstack-lbaas | 06:24 | |
| *** ducttape_ has quit IRC | 06:25 | |
| *** amotoki has quit IRC | 06:27 | |
| *** fawadkhaliq has joined #openstack-lbaas | 06:49 | |
| *** nmagnezi has joined #openstack-lbaas | 06:55 | |
| *** amotoki has joined #openstack-lbaas | 07:06 | |
| *** fawadkhaliq has quit IRC | 07:12 | |
| openstackgerrit | Christian Zunker proposed openstack/neutron-lbaas: Prevent LBaaS agent from filling up logs. https://review.openstack.org/305284 | 07:14 |
| *** pcaruana has joined #openstack-lbaas | 07:16 | |
| *** amotoki_ has joined #openstack-lbaas | 07:20 | |
| *** amotoki has quit IRC | 07:22 | |
| *** pcaruana has quit IRC | 07:23 | |
| *** kobis has joined #openstack-lbaas | 07:23 | |
| *** tesseract has joined #openstack-lbaas | 07:24 | |
| *** tesseract is now known as Guest22945 | 07:24 | |
| kong | blogan: hi, amp_active_retries=100 and amp_active_wait_sec=1 | 07:46 |
| *** mariusv has quit IRC | 07:48 | |
| kong | blogan: the root cause of the inconsistency is becuase communication between neutron and octavia-api, octavia-api and octavia-worker are both asynchronous, if there is something happened in octavia-worker, we must figure out a way to make the state consistent | 07:53 |
| *** amotoki_ has quit IRC | 08:06 | |
| *** ducttape_ has joined #openstack-lbaas | 08:22 | |
| *** ducttape_ has quit IRC | 08:26 | |
| *** bana_k has quit IRC | 08:37 | |
| *** eezhova has joined #openstack-lbaas | 08:46 | |
| *** amotoki has joined #openstack-lbaas | 09:17 | |
| *** bana_k has joined #openstack-lbaas | 09:37 | |
| *** amotoki has quit IRC | 09:43 | |
| *** amotoki has joined #openstack-lbaas | 10:00 | |
| openstackgerrit | Kobi Samoray proposed openstack/octavia: (WIP) Implement string validation health monitor https://review.openstack.org/307688 | 10:08 |
| *** amotoki has quit IRC | 10:14 | |
| *** bana_k has quit IRC | 10:21 | |
| *** ducttape_ has joined #openstack-lbaas | 10:23 | |
| *** ducttape_ has quit IRC | 10:27 | |
| *** bana_k has joined #openstack-lbaas | 10:52 | |
| *** rtheis has joined #openstack-lbaas | 11:10 | |
| *** yamamoto has quit IRC | 11:30 | |
| *** doug-fish has joined #openstack-lbaas | 11:30 | |
| *** Guest22945 is now known as tesseract | 11:54 | |
| *** tesseract is now known as Guest73397 | 11:54 | |
| *** matt-borland has joined #openstack-lbaas | 12:00 | |
| *** yamamoto has joined #openstack-lbaas | 12:09 | |
| *** ducttape_ has joined #openstack-lbaas | 12:14 | |
| *** yamamoto has quit IRC | 12:24 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/307763 | 12:27 |
| *** ducttape_ has quit IRC | 12:31 | |
| *** yamamoto has joined #openstack-lbaas | 12:34 | |
| *** yamamoto has quit IRC | 12:39 | |
| *** yamamoto has joined #openstack-lbaas | 12:44 | |
| *** yamamoto has quit IRC | 12:45 | |
| *** yamamoto has joined #openstack-lbaas | 12:54 | |
| *** yamamoto has quit IRC | 13:03 | |
| *** yamamoto has joined #openstack-lbaas | 13:05 | |
| *** ducttape_ has joined #openstack-lbaas | 13:11 | |
| *** yamamoto has quit IRC | 13:14 | |
| *** ducttape_ has quit IRC | 13:16 | |
| *** yamamoto has joined #openstack-lbaas | 13:16 | |
| *** reedip_ has joined #openstack-lbaas | 13:17 | |
| *** neelashah has joined #openstack-lbaas | 13:25 | |
| *** BjoernT has joined #openstack-lbaas | 13:26 | |
| *** eranra has joined #openstack-lbaas | 13:30 | |
| *** piet has joined #openstack-lbaas | 13:34 | |
| *** yamamoto has quit IRC | 13:41 | |
| *** yamamoto has joined #openstack-lbaas | 13:43 | |
| *** fnaval has quit IRC | 13:48 | |
| *** piet has quit IRC | 13:53 | |
| openstackgerrit | Kobi Samoray proposed openstack/neutron-lbaas: Add string checking param to LBaaSv2 healthmonitor https://review.openstack.org/258560 | 13:59 |
| *** links has quit IRC | 14:03 | |
| openstackgerrit | Kobi Samoray proposed openstack/octavia: (WIP) Implement string validation health monitor https://review.openstack.org/307688 | 14:05 |
| *** fnaval has joined #openstack-lbaas | 14:05 | |
| *** piet has joined #openstack-lbaas | 14:06 | |
| *** woodster_ has joined #openstack-lbaas | 14:19 | |
| *** bana_k has quit IRC | 14:19 | |
| *** piet has quit IRC | 14:21 | |
| *** numan_ has quit IRC | 14:26 | |
| *** TrevorV|Home has joined #openstack-lbaas | 14:27 | |
| *** ducttape_ has joined #openstack-lbaas | 14:29 | |
| *** piet has joined #openstack-lbaas | 14:29 | |
| *** ducttape_ has quit IRC | 14:31 | |
| *** ducttape_ has joined #openstack-lbaas | 14:31 | |
| *** ducttape_ has quit IRC | 14:32 | |
| *** Purandar has joined #openstack-lbaas | 14:32 | |
| *** ducttape_ has joined #openstack-lbaas | 14:33 | |
| *** Purandar has quit IRC | 14:36 | |
| *** amotoki has joined #openstack-lbaas | 14:43 | |
| *** fnaval has quit IRC | 14:49 | |
| *** fnaval has joined #openstack-lbaas | 14:50 | |
| *** armax has joined #openstack-lbaas | 14:58 | |
| *** Purandar has joined #openstack-lbaas | 14:58 | |
| *** amotoki has quit IRC | 15:02 | |
| *** nmagnezi has quit IRC | 15:05 | |
| *** numan_ has joined #openstack-lbaas | 15:08 | |
| Frito | http://www.duckychannel.com.tw/page-en/Ducky-One-TKL/ | 15:13 |
| Frito | mischan | 15:14 |
| *** kobis has quit IRC | 15:15 | |
| *** kevo has quit IRC | 15:19 | |
| *** armax has quit IRC | 15:27 | |
| *** ducttape_ has quit IRC | 15:37 | |
| *** eranra has quit IRC | 15:38 | |
| *** eezhova has quit IRC | 15:42 | |
| TrevorV|Home | Hey johnsom xgerman how much have you guys worked in neutron lbaas? | 15:43 |
| xgerman | very little | 15:43 |
| johnsom | Some, not nearly as much as Octavia, but I'm coming up to speed. Posted a patch last night even | 15:44 |
| xgerman | blogan almost wrote it single handedly | 15:44 |
| TrevorV|Home | Yeah, xgerman but I can't get ahold of him today | 15:44 |
| *** ducttape_ has joined #openstack-lbaas | 15:44 | |
| TrevorV|Home | I'm having stupid issues right now. | 15:44 |
| xgerman | ok, anything we can look at? | 15:44 |
| TrevorV|Home | I'm trying to add rise/fall thresholds to healthmonitors in neutron lbaas, and for some reason tests throw a 400 when I add that field to the request body. | 15:45 |
| TrevorV|Home | Here is my review:D | 15:45 |
| *** links has joined #openstack-lbaas | 15:45 | |
| TrevorV|Home | https://review.openstack.org/#/c/306084 | 15:45 |
| * xgerman is looking | 15:45 | |
| TrevorV|Home | I compared it against the single-create review for "parity" so-to-speak, and I didn't find anything out of place from what I can tell. | 15:46 |
| xgerman | mmmh, it looks a bit like you are adding some new endpoints | 15:48 |
| xgerman | nah, that looks good | 15:53 |
| fnaval | speaking of neutron-lbaas reviews, reviews on some new tests requested please: https://review.openstack.org/#/q/owner:%22Franklin+Naval+%253Cfranklin.naval%2540gmail.com%253E%22++status:open | 15:54 |
| fnaval | thanks in advance | 15:54 |
| TrevorV|Home | xgerman yeah, I was just adding functionality and making it driver-enabled, so to speak. Like I did with the lb-graph. | 15:55 |
| TrevorV|Home | So now I'm lost as to what isn't there | 15:56 |
| johnsom | Yeah, that is neat. It's like it is triggering the notimplemented exception | 15:56 |
| xgerman | it will for all drivers which are not the NOOP | 15:56 |
| *** links has quit IRC | 15:57 | |
| johnsom | Yeah, I know some of the tests cycle through the drivers, but it's not clear from this output that it is the issue | 15:58 |
| TrevorV|Home | xgerman that raises it because it does a check on the driver. There is a method | 15:58 |
| *** doug-fish has quit IRC | 16:00 | |
| *** Guest73397 has quit IRC | 16:00 | |
| *** doug-fish has joined #openstack-lbaas | 16:00 | |
| johnsom | TrevorV|Home Ok, so the scenario tests don't surprise me, they use some internal models that need updating. | 16:02 |
| johnsom | For the unit tests, this file will need to be updated: https://github.com/openstack/neutron-lbaas/blob/master/neutron_lbaas/tests/unit/drivers/common/test_agent_driver_base.py | 16:03 |
| xgerman | you can look at the cascading delete patch for all the files which need to be touched | 16:04 |
| *** doug-fish has quit IRC | 16:09 | |
| *** bana_k has joined #openstack-lbaas | 16:14 | |
| TrevorV|Home | johnsom I don't understand what would need updated in that file though | 16:14 |
| TrevorV|Home | xgerman the single create stuff works, and I simulated changes accordingly for this patch, so I'm confused. | 16:14 |
| *** bana_k has quit IRC | 16:15 | |
| *** numan_ has quit IRC | 16:16 | |
| johnsom | TrevorV|Home http://logs.openstack.org/84/306084/2/check/gate-neutron-lbaas-python27/ec7beec/console.html#_2016-04-19_08_19_05_538 | 16:17 |
| johnsom | http://logs.openstack.org/84/306084/2/check/gate-neutron-lbaas-python27/ec7beec/console.html#_2016-04-19_08_19_19_377 | 16:17 |
| johnsom | etc | 16:17 |
| johnsom | Since driver base itself will be "not implemented" | 16:17 |
| *** neelashah has quit IRC | 16:17 | |
| TrevorV|Home | johnsom that's capturing an error thrown in test_db_loadbalancerv2 though | 16:18 |
| *** yamamoto has quit IRC | 16:19 | |
| TrevorV|Home | johnsom it might be bad practice, but I tend to pick a single test to work with until I get it passing correctly, and then I check for the next, etc etc. | 16:20 |
| TrevorV|Home | So the test I've been working on is: test_create_health_monitor_with_fall_threshold | 16:20 |
| TrevorV|Home | When debugging, the 400 response message says a healthmonitor has no argument fall_threshold | 16:21 |
| TrevorV|Home | But given that I've added that via an extension, I can't figure out where to debug why that fails. It never makes it to the plugin file, which is the first point of contact in neutron lbaas, from what I can reason | 16:21 |
| *** kevo has joined #openstack-lbaas | 16:26 | |
| *** doug-fish has joined #openstack-lbaas | 16:27 | |
| *** piet has quit IRC | 16:29 | |
| *** neelashah has joined #openstack-lbaas | 16:30 | |
| *** neelashah has quit IRC | 16:30 | |
| *** piet has joined #openstack-lbaas | 16:31 | |
| *** neelashah has joined #openstack-lbaas | 16:31 | |
| *** fawadkhaliq has joined #openstack-lbaas | 16:45 | |
| *** anilvenkata has quit IRC | 16:48 | |
| *** kevo has quit IRC | 16:50 | |
| *** kevo has joined #openstack-lbaas | 16:54 | |
| TrevorV|Home | johnsom I got ahold of blogan and he helped me figure out where in neutron it begins, and when its taking the hm body its failing to accept the rise_threshold field. | 16:55 |
| johnsom | Ok, cool | 16:57 |
| *** kevo has quit IRC | 16:57 | |
| fnaval | dougwig: I added a comment to https://bugs.launchpad.net/neutron/+bug/1571907 | 16:57 |
| openstack | Launchpad bug 1571907 in neutron "Neutron-LBaaS v2: Invalid tenant id accepted on "add member to pool"" [Undecided,Won't fix] | 16:57 |
| fnaval | I think the bug should be reopened. | 16:58 |
| TrevorV|Home | johnsom no its not cool lulz, because I can't figure out what needs to change for neutron to actually load the extension so it checks v2 AND my extension for fields for the health monitor | 17:00 |
| TrevorV|Home | Its obviously not a neutron specific change, it has to be something I've done in neutron lbaas, but I've followed everything super closely | 17:01 |
| *** rcernin has quit IRC | 17:04 | |
| *** madhu_ak has joined #openstack-lbaas | 17:08 | |
| *** yamamoto has joined #openstack-lbaas | 17:20 | |
| *** yamamoto has quit IRC | 17:28 | |
| *** neelashah has quit IRC | 17:33 | |
| *** neelashah has joined #openstack-lbaas | 17:35 | |
| *** kevo has joined #openstack-lbaas | 17:36 | |
| TrevorV|Home | johnsom I figured it out.... | 17:39 |
| TrevorV|Home | In my extension, the "EXTENDED_ATTRIBUTES_2_0"... "healthmonitor" needed to be pluralized... | 17:39 |
| xgerman | typos!! | 17:40 |
| johnsom | Oh brother. | 17:40 |
| TrevorV|Home | Yeah. I think I hate that the most about extension in Neutron LBaaS. I'm not even sure why its relevant. I'm sure someone will say "because the magic needs it"... which we all love. | 17:42 |
| xgerman | yeah, they seem like a hassle | 17:43 |
| *** minwang2 has joined #openstack-lbaas | 17:48 | |
| TrevorV|Home | Its alright, now I'm just updating tests. | 17:56 |
| TrevorV|Home | Then it'll be ready for review. | 17:56 |
| TrevorV|Home | Oh, johnsom are we having a meeting tomorrow? | 17:56 |
| johnsom | Yes, probably a short one. No meeting next week | 17:56 |
| *** crc32 has joined #openstack-lbaas | 17:57 | |
| TrevorV|Home | johnsom there are a TON of meetings next week, you mean :) | 17:59 |
| TrevorV|Home | Alright cool, I'll just have a review or two that I'll need some eyes on, that's about all I got. | 17:59 |
| TrevorV|Home | I have a serious backlog on reviewing other reviews though, so I think I might take the productivity hit this afternoon and bust through a bunch. | 18:00 |
| *** doug-fis_ has joined #openstack-lbaas | 18:00 | |
| TrevorV|Home | Anything you guys need eyes on? johnsom xgerman ? | 18:00 |
| a2hill | https://review.openstack.org/#/c/303220/ | 18:01 |
| *** piet has quit IRC | 18:01 | |
| *** doug-fi__ has joined #openstack-lbaas | 18:01 | |
| johnsom | https://review.openstack.org/#/c/300292/ | 18:02 |
| *** neelashah1 has joined #openstack-lbaas | 18:02 | |
| *** doug-fish has quit IRC | 18:04 | |
| *** doug-fis_ has quit IRC | 18:04 | |
| TrevorV|Home | Alright, I got both of your reviews in tabs, I'll knock those out this afternoon. a2hill looks like Adam already +2'd you | 18:04 |
| TrevorV|Home | brb grabbin lunch | 18:04 |
| *** neelashah has quit IRC | 18:05 | |
| *** bana_k has joined #openstack-lbaas | 18:27 | |
| *** anilvenkata has joined #openstack-lbaas | 18:31 | |
| *** pck_ is now known as pck | 18:35 | |
| *** doug-fi__ has quit IRC | 18:36 | |
| madhu_ak | dougwig, mestery could you approve this backport patch: https://review.openstack.org/#/c/260688/ | 18:38 |
| madhu_ak | Thanks dougwig | 18:45 |
| *** bana_k has quit IRC | 18:50 | |
| *** anilvenkata has quit IRC | 18:51 | |
| *** cody-somerville has joined #openstack-lbaas | 18:52 | |
| openstackgerrit | Trevor Vardeman proposed openstack/neutron-lbaas: Add Health Monitor Thresholds https://review.openstack.org/306084 | 18:58 |
| *** TrevorV|Home has quit IRC | 18:59 | |
| *** fawadkhaliq has quit IRC | 19:00 | |
| *** neelashah1 has quit IRC | 19:00 | |
| *** TrevorV|Home has joined #openstack-lbaas | 19:01 | |
| fnaval | madhu_ak: if you have time please take a look at some of the neutron-lbaas test reviews: https://review.openstack.org/#/q/owner:%22Franklin+Naval+%253Cfranklin.naval%2540gmail.com%253E%22++status:open | 19:01 |
| *** doug-fis_ has joined #openstack-lbaas | 19:04 | |
| *** neelashah has joined #openstack-lbaas | 19:06 | |
| *** doug-fis_ has quit IRC | 19:09 | |
| *** ChrisBenson has quit IRC | 19:12 | |
| *** doug-fish has joined #openstack-lbaas | 19:14 | |
| fnaval | also minwang2 ^^ | 19:16 |
| *** cody-somerville has quit IRC | 19:20 | |
| madhu_ak | yeah fnaval. Will review them. Sorry I could not do them yesterday | 19:26 |
| fnaval | np thanks madhu_ak | 19:27 |
| fnaval | should the protocol and port of a listener match the protocol and port of a pool? why could they be different? | 19:44 |
| *** johnsom_laptop has joined #openstack-lbaas | 19:46 | |
| fnaval | directed to anyone who would know ^^ | 19:46 |
| madhu_ak | dougwig when I try to run experimental job for octavia tempest tests (as in this patch: https://review.openstack.org/#/c/306053/), it is complaining about the $PROJECTS variable while setting up devstack. Log: http://logs.openstack.org/53/306053/2/experimental/gate-tempest-dsvm-octavia/4e55e16/logs/devstacklog.txt.gz#_2016-04-14_23_04_36_563. Should the export $PROJECTS in gate hooks be moved to octavia.yaml under project config? | 19:55 |
| *** piet has joined #openstack-lbaas | 20:02 | |
| *** neelashah has quit IRC | 20:05 | |
| *** johnsom_laptop has quit IRC | 20:05 | |
| *** eezhova has joined #openstack-lbaas | 20:07 | |
| blogan | johnsom: do you happen to know the haproxy key to get the total connections? is is stot or cumconns? | 20:09 |
| johnsom | I can look real quick | 20:09 |
| blogan | johnsom: i think our code is using stot, but Frito is telling me thats total sessions | 20:10 |
| Frito | hey | 20:10 |
| blogan | and i have no idea which one is right, i could never find good documentation on the stats haproxy returns | 20:11 |
| johnsom | stot [LFBS]: cumulative number of connections | 20:11 |
| Frito | yea, I think we're mapping stot from the "show status" command to the column "total_connections" in the db. I'm wondering if that should be "CumConns" from the "show info" one | 20:11 |
| johnsom | So, yeah, that is probably wrong | 20:11 |
| johnsom | Probably should be: scur [LFBS]: current sessions | 20:12 |
| blogan | johnsom: for total_connections? | 20:12 |
| johnsom | Oh, yeah, never mind, looking at that wrong line in our code. So, yeah, we are using stot for "totconns" and scur for conns | 20:13 |
| johnsom | So, it's right. | 20:13 |
| blogan | yeah accoring to the docs it does say cumulative number of connections | 20:14 |
| Frito | If you don't mind where are you seeing that. I don't see "stot" on https://cbonte.github.io/haproxy-dconv/configuration-1.5.html . I found it on http://feurix.org/projects/hatop/readme/ | 20:14 |
| johnsom | http://cbonte.github.io/haproxy-dconv/configuration-1.5.html#9.1 | 20:14 |
| blogan | Frito: i don't see a CumConns | 20:14 |
| *** fawadkhaliq has joined #openstack-lbaas | 20:14 | |
| johnsom | yeah, I don't either | 20:14 |
| blogan | i'mlookig here | 20:15 |
| blogan | http://www.haproxy.org/download/1.5/doc/configuration.txt | 20:15 |
| Frito | blogan: if you have a HAProxy instance running and run "show info" you'll see CumConns. It's not in our code at all. I'm wondering if that's the one we should be using. | 20:15 |
| blogan | search for stot | 20:15 |
| johnsom | No, stot is right | 20:15 |
| Frito | ctrl+f stot gives me 0 of 0 :-( | 20:15 |
| johnsom | Frito http://cbonte.github.io/haproxy-dconv/configuration-1.5.html#9.1 | 20:16 |
| blogan | yeah use johnsom's lol | 20:16 |
| Frito | Ha, got ya | 20:16 |
| johnsom | It's #7 on the list | 20:16 |
| blogan | it at least has anchors | 20:16 |
| blogan | does show info show the csv that we are parsing from the stats socket in code? | 20:16 |
| Frito | No | 20:17 |
| johnsom | No | 20:17 |
| Frito | It shows the other key value list | 20:17 |
| Frito | the show info call is coded but it's only invoked by a unit test | 20:17 |
| johnsom | It's a per-process "status" with show info. | 20:17 |
| johnsom | So, if you are running in multiproc it's only partial | 20:17 |
| *** neelashah has joined #openstack-lbaas | 20:17 | |
| blogan | and we run multiproc :( | 20:18 |
| a2hill | Theres ways to get around that sorta | 20:18 |
| johnsom | We use show stat | 20:19 |
| a2hill | blogan: the things bounds is going to work on | 20:19 |
| *** fawadkhaliq has quit IRC | 20:22 | |
| *** fawadkhaliq has joined #openstack-lbaas | 20:22 | |
| Frito | thx blogan / johnsom. /me is still trying to get acclimated. | 20:23 |
| johnsom | NP | 20:23 |
| johnsom | Frito New to the LBaaS team? | 20:23 |
| Frito | Yea, I'm on loan for a few months to help out w/ some things here at RAX | 20:24 |
| johnsom | Cool. Welcome | 20:24 |
| Frito | :-) Thanks | 20:24 |
| Frito | I'll try not to do too many dumb things ;-) | 20:24 |
| johnsom | If you have Octavia questions I can probably help. | 20:24 |
| Frito | Hehe. Actually my next one was if there's a reason we're not including the uptime as part of the heartbeat. | 20:25 |
| Frito | I think that's something else we needed to be able to bill correctly. | 20:25 |
| johnsom | You have been talking to Trevor | 20:25 |
| Frito | I have off and on. | 20:26 |
| johnsom | I think blogan had some thoughts there too. | 20:26 |
| blogan | i think we need to just have a "created_at" column in the db on the load balancer, and that tracks the uptime | 20:27 |
| TrevorV|Home | Frito, johnsom is poking fun at me :D | 20:27 |
| TrevorV|Home | blogan, I think that's already there right? | 20:27 |
| johnsom | My take is the "uptime" you would need to do such billing exclusion would need to come from outside of the amphora. Failovers and such make the haproxy uptime not very useful | 20:27 |
| Frito | TrevorV|Home: Be nice. I might help him ;-) | 20:27 |
| *** doug-fish has quit IRC | 20:27 | |
| TrevorV|Home | ha ha ha, I'm bein nice!! | 20:27 |
| blogan | that and uptime is more a metric for passive resources being used, so even if there is a downtime, passive resources are still being used | 20:28 |
| johnsom | Yeah, personally I would lean towards blogan's comment. | 20:28 |
| Frito | Okay. So not uptime but just a static "created_at" for when the record is created. | 20:28 |
| Frito | ? | 20:28 |
| TrevorV|Home | Yeah | 20:29 |
| johnsom | Frito We had a discussion about this at one of the Octavia IRC meetings: http://eavesdrop.openstack.org/meetings/octavia/2016/octavia.2016-04-06-20.00.log.html | 20:29 |
| TrevorV|Home | That's what blogan clarified for me | 20:29 |
| johnsom | Might be worth a read | 20:29 |
| blogan | Frito: yeah and then at some point there will have to be something that does calculate the uptime | 20:29 |
| blogan | based on that date | 20:29 |
| blogan | and pushes it to something | 20:29 |
| blogan | :) | 20:29 |
| TrevorV|Home | Which is a "now - created" sort of thing | 20:29 |
| blogan | something calculates and something pushes | 20:29 |
| blogan | and something load balances | 20:29 |
| * johnsom waves his hands in the air | 20:29 | |
| Frito | Ha. Okay. Thanks for the meeting notes. I'll look over those as well. | 20:29 |
| TrevorV|Home | blogan, the bit that's doing that calculation and the other bit that's doing the pushing, those are already marked in tasks | 20:29 |
| blogan | okie | 20:30 |
| *** armax has joined #openstack-lbaas | 20:32 | |
| *** eezhova has quit IRC | 20:32 | |
| *** ducttape_ has quit IRC | 20:36 | |
| *** doug-fish has joined #openstack-lbaas | 20:36 | |
| *** doug-fish has quit IRC | 20:36 | |
| *** doug-fish has joined #openstack-lbaas | 20:36 | |
| *** ducttape_ has joined #openstack-lbaas | 20:36 | |
| *** ducttape_ has quit IRC | 20:38 | |
| *** ducttape_ has joined #openstack-lbaas | 20:39 | |
| *** armax_ has joined #openstack-lbaas | 20:40 | |
| *** armax has quit IRC | 20:42 | |
| *** armax_ is now known as armax | 20:42 | |
| *** ducttape_ has quit IRC | 20:52 | |
| *** ducttape_ has joined #openstack-lbaas | 20:52 | |
| *** rtheis has quit IRC | 21:07 | |
| *** BjoernT has quit IRC | 21:12 | |
| *** fawadkhaliq has quit IRC | 21:23 | |
| *** fawadkhaliq has joined #openstack-lbaas | 21:24 | |
| *** neelashah has quit IRC | 21:24 | |
| *** ducttape_ has quit IRC | 21:24 | |
| *** cody-somerville has joined #openstack-lbaas | 21:25 | |
| *** ducttape_ has joined #openstack-lbaas | 21:25 | |
| *** johnsom_laptop has joined #openstack-lbaas | 21:26 | |
| *** matt-borland has quit IRC | 21:26 | |
| *** ducttape_ has quit IRC | 21:27 | |
| *** fawadkhaliq has quit IRC | 21:28 | |
| *** TrevorV|Home has quit IRC | 21:36 | |
| *** kevo has quit IRC | 21:39 | |
| *** fawadkhaliq has joined #openstack-lbaas | 22:14 | |
| *** intr1nsic has quit IRC | 22:20 | |
| *** apuimedo has joined #openstack-lbaas | 22:21 | |
| *** piet has quit IRC | 22:23 | |
| *** intr1nsic has joined #openstack-lbaas | 22:26 | |
| *** piet has joined #openstack-lbaas | 22:26 | |
| *** kevo has joined #openstack-lbaas | 22:38 | |
| *** johnsom_ has joined #openstack-lbaas | 23:00 | |
| *** johnsom_laptop has quit IRC | 23:00 | |
| *** crc32 has quit IRC | 23:11 | |
| *** fawadkhaliq has quit IRC | 23:19 | |
| *** apuimedo is now known as celebdor | 23:19 | |
| *** fawadkhaliq has joined #openstack-lbaas | 23:19 | |
| *** ducttape_ has joined #openstack-lbaas | 23:22 | |
| *** johnsom_ has quit IRC | 23:23 | |
| *** fawadkhaliq has quit IRC | 23:24 | |
| *** ducttape_ has quit IRC | 23:25 | |
| *** ducttape_ has joined #openstack-lbaas | 23:28 | |
| *** neelashah has joined #openstack-lbaas | 23:42 | |
| *** HenryG has quit IRC | 23:45 | |
| *** woodster_ has quit IRC | 23:48 | |
| *** HenryG has joined #openstack-lbaas | 23:48 | |
| *** ducttape_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!