| *** clev is now known as clev-away | 00:07 | |
| *** Aish has joined #openstack-lbaas | 00:16 | |
| *** Aish has left #openstack-lbaas | 00:16 | |
| *** madhu_ak has quit IRC | 00:16 | |
| *** minwang2 has quit IRC | 00:23 | |
| *** tiny-hands has joined #openstack-lbaas | 00:31 | |
| blogan_ | kev0: do you have neutron running? | 01:10 |
|---|---|---|
| *** diogogmt has joined #openstack-lbaas | 01:11 | |
| *** hummel_ has joined #openstack-lbaas | 01:15 | |
| *** haigang has joined #openstack-lbaas | 01:16 | |
| hummel_ | hi guys.. | 01:16 |
| hummel_ | wondering if someone would be available to answer some questions in relation to LB on openstack | 01:17 |
| *** kev0 has quit IRC | 01:17 | |
| hummel_ | ? | 01:19 |
| openstackgerrit | Brandon Logan proposed openstack/octavia: Setup base neutron network driver https://review.openstack.org/209210 | 01:27 |
| blogan_ | hummel_: i dont much time so hopefully its quick! | 01:28 |
| blogan_ | xgerman: ^^ | 01:28 |
| hummel_ | two questions : | 01:29 |
| hummel_ | 1. can the LB listen to multiple ports on the one VIP ( i dont think so, from what ive read) | 01:29 |
| hummel_ | 2. Can you insert 'X-Forwarded-Proto' headers ? | 01:30 |
| blogan_ | hummel_: #1: not for lbaasv1, lbaasv2 it can | 01:32 |
| blogan_ | hummel_: #2 I believe that depends on the driver being used, and I do not knwo which ones do that right now | 01:33 |
| blogan_ | hummel_: it is something we will undoubtedly do with octavia | 01:34 |
| blogan_ | hummel_: but i don't believe it is in yet, though i could be wrong | 01:34 |
| hummel_ | thanks blogan_ i'm pretty new to this, when you say lbaas v1 and v2, what is v2 commonly referred to as? | 01:36 |
| johnsom | We do not have X-Forwarded-Proto at this time | 01:36 |
| johnsom | v1 is the old LBaaS API version, where v2 is the new LBaaS API version. | 01:37 |
| hummel_ | thanks johnsom | 01:37 |
| blogan_ | hummel_: it is commonly referred to as lbaasv2? | 01:37 |
| blogan_ | i didnt mean the ? | 01:37 |
| blogan_ | hummel_: its experimental for kilo | 01:37 |
| hummel_ | ok so basically v2 isn't production to the public as yet is what you are saying? | 01:38 |
| blogan_ | hummel_: so relatively new, but v1 will be deprecated at some point | 01:38 |
| hummel_ | any links for doco on v2 ? | 01:38 |
| johnsom | http://developer.openstack.org/api-ref-networking-v2-ext.html#lbaas-v2.0 | 01:38 |
| blogan_ | too fast johnsom | 01:39 |
| johnsom | Kind distraction from my sqlalchemy nightmares | 01:39 |
| blogan_ | lol | 01:39 |
| hummel_ | thanks, had that up already, haha | 01:39 |
| blogan_ | i dont think sqlalchemy fantasies/dreams exist | 01:39 |
| *** bana_k has quit IRC | 01:40 | |
| hummel_ | cant do the multiple port on 1 VIP via the CLI ? | 01:41 |
| johnsom | Yes. The CLI command for lbaasv2 are prefixed lbaas- | 01:43 |
| blogan_ | hummel_: http://docs.openstack.org/cli-reference/content/neutronclient_commands.html | 01:44 |
| blogan_ | find the lbaas- comamnds | 01:44 |
| blogan_ | loadbalancer is basically a v1 vip | 01:45 |
| blogan_ | so it goes loadbalancer -> many listeners -> 1 pool -> many members | 01:45 |
| hummel_ | yup got ya, so with the CLI it's a v1 VIp, so i wouldn't be able to assign multiple ports to the one VIp which is what i'm trying to achieve? | 01:45 |
| blogan_ | hummel_: not with the v1 calls, any lb- | 01:45 |
| hummel_ | but neutron-lbaas yes? | 01:46 |
| blogan_ | yes | 01:46 |
| blogan_ | you would just create 2 listeners on the same load balancer | 01:46 |
| blogan_ | or more listeners | 01:46 |
| blogan_ | but obviously can't have 2 listeners on the same load balancer listening on the same port | 01:47 |
| blogan_ | okay i gotta go now | 01:47 |
| hummel_ | yeah so example would be like this | 01:47 |
| hummel_ | VIP : 10.10.10.10 | 01:47 |
| hummel_ | Ports : 80 & 443 | 01:47 |
| blogan_ | yep, thats the main use case | 01:47 |
| hummel_ | So VIP listens on both ports and then directs that traffic to two servers sitting behind it | 01:47 |
| hummel_ | easy in the F5 world :) | 01:47 |
| blogan_ | easy in every load balancing world it owuld seem, other than v1 lbaas | 01:48 |
| hummel_ | stupid question time, how does one tell if i'm on v1 lbaas or v2 ? (i dont own the environment) just use it.. | 01:48 |
| *** woodster_ has quit IRC | 01:49 | |
| blogan_ | now v2 does not allow a pool to be on 2 different listeners right now, so you'd have to create 2 identical pools for 80 and 443 to go to the same backend (assuming this is the common ssl termination case) | 01:49 |
| *** crc32 has joined #openstack-lbaas | 01:49 | |
| blogan_ | hummel_: depends, how are you interacting with it? | 01:50 |
| hummel_ | i can interact via the web GUI or via CLI | 01:50 |
| crc32 | blogan_: You having issues with vpn? | 01:50 |
| blogan_ | and what parts of the system can you access? | 01:50 |
| blogan_ | there isn't a web GUI for v2 yet, so thats easy, thats v1 | 01:50 |
| johnsom | cli v1 commands start with lb-, v2 commands start with lbaas- | 01:50 |
| blogan_ | for the CLI, lb- or lbaas- will work, both will not work | 01:51 |
| blogan_ | mutually exclusive | 01:51 |
| blogan_ | crc32: i was this afternoon, i quit trying to ge ton | 01:51 |
| hummel_ | ok so easy way is to test using the CLI | 01:51 |
| hummel_ | let me switch networks, i'll be back and try it out | 01:51 |
| blogan_ | hummel_: yep, unless you have access ot hte neutron.conf | 01:51 |
| blogan_ | oh | 01:51 |
| hummel_ | nah dont have that luxury.. | 01:51 |
| blogan_ | you could make a curl request to to the neutron endpoint to look at /extensions | 01:52 |
| blogan_ | that'll show lbaas or lbaasv2 | 01:52 |
| *** crc32 has quit IRC | 01:52 | |
| blogan_ | okay im going for real! | 01:52 |
| blogan_ | talk to yall later | 01:52 |
| blogan_ | good luck hummel_ | 01:52 |
| blogan_ | im sure johnsom can answer your questions as well, if not ML is always good or tomorrow | 01:52 |
| *** hummel__ has joined #openstack-lbaas | 01:53 | |
| hummel__ | ok im back | 01:53 |
| blogan_ | or just idle in here and ill get it back | 01:53 |
| blogan_ | lol ive been taling to a ghost | 01:53 |
| blogan_ | anyway, i gotta go | 01:53 |
| johnsom | Yeah, I about to sign off for the day too | 01:53 |
| hummel__ | ill fire up my linux VM and see what i can/cant do | 01:53 |
| blogan_ | any questions? johnsom cna answer, ML works, or just idle in here and someone will eventually see it | 01:53 |
| *** bana_k has joined #openstack-lbaas | 01:53 | |
| *** blogan_ has quit IRC | 01:54 | |
| *** hummel_ has quit IRC | 01:56 | |
| *** ajmiller has joined #openstack-lbaas | 01:59 | |
| *** fnaval_ has quit IRC | 02:00 | |
| *** johnsom_ has joined #openstack-lbaas | 02:02 | |
| *** ajmiller_ has joined #openstack-lbaas | 02:02 | |
| *** crc32 has joined #openstack-lbaas | 02:04 | |
| *** fnaval has joined #openstack-lbaas | 02:05 | |
| *** ajmiller has quit IRC | 02:05 | |
| *** johnsom has quit IRC | 02:06 | |
| *** crc32 has quit IRC | 02:06 | |
| *** crc32 has joined #openstack-lbaas | 02:06 | |
| *** fnaval has quit IRC | 02:08 | |
| *** fnaval has joined #openstack-lbaas | 02:09 | |
| crc32 | guess I'll only be able to vpn in from my linux box. The cisco client on the mac side is too restrictive. Its like taking over all routing so connections on my LAN would be forced over the VPN. | 02:14 |
| *** vivek-ebay has quit IRC | 02:22 | |
| *** vivek-ebay has joined #openstack-lbaas | 02:28 | |
| *** vivek-ebay has quit IRC | 02:28 | |
| *** tiny-hands has quit IRC | 02:29 | |
| *** tiny-hands has joined #openstack-lbaas | 02:35 | |
| *** sbalukoff has quit IRC | 02:46 | |
| *** bana_k has quit IRC | 02:58 | |
| *** crc32 has quit IRC | 02:59 | |
| *** diogogmt has quit IRC | 03:01 | |
| *** crc32 has joined #openstack-lbaas | 03:09 | |
| *** hummel__ has quit IRC | 03:13 | |
| rm_you | crc32: weird, mine doesn't do that | 03:17 |
| johnsom_ | rm_you - Are you a sqlalchemy expert? | 03:18 |
| crc32 | its on this new labtop It uses the direct cisco client. The one that looks like a white circle with green loops | 03:23 |
| *** ganeshna has joined #openstack-lbaas | 03:25 | |
| johnsom_ | Here is the challenge: Execute this "date_sub(now(), INTERVAL 10 SECOND)" in sqlalchemy such that the times are all calculated in the database engine. (no python calculations) | 03:26 |
| rm_you | johnsom_: not an EXPERT, but i know some stuff | 03:27 |
| rm_you | hmm | 03:27 |
| rm_you | ah | 03:27 |
| rm_you | so | 03:27 |
| rm_you | that's specific to the DB impl | 03:27 |
| rm_you | so not encouraged, I think | 03:27 |
| rm_you | there's like... sqlalchemy.mysql.NOW or something as a constant | 03:27 |
| johnsom_ | I got as far as: func.date_sub( | 03:27 |
| johnsom_ | func.now(), text('interval ' + str(timeout) + ' second') | 03:27 |
| rm_you | let me recheck | 03:27 |
| johnsom_ | Which works, but the "text" is bad and breaks sqllite | 03:28 |
| rm_you | oh nm | 03:28 |
| rm_you | it's func.now() | 03:28 |
| rm_you | from sqlalchemy.sql import func | 03:28 |
| rm_you | func.now() forwards it through to the DB | 03:28 |
| rm_you | so uh | 03:29 |
| johnsom_ | breaking sqllite means the tests fail | 03:29 |
| rm_you | if you do func.now() it should give you the DB's "NOW" which gives you a DateTime object, and there's no reason you can do the MATH locally, i think | 03:29 |
| johnsom_ | There is a post on stackoverflow saying you can do this: | 03:30 |
| rm_you | why do you need to do the math inside SQL? | 03:30 |
| johnsom_ | func.dateadd(func.now(), bindparam('tomorrow', timedelta(days=1), Interval())) | 03:30 |
| johnsom_ | But I can't seem to get it to work | 03:30 |
| rm_you | ah | 03:30 |
| rm_you | yeah that looks somewhat accurate | 03:31 |
| johnsom_ | Time skew | 03:31 |
| johnsom_ | If the math is done on the controller, but the DB is different, we fail over | 03:31 |
| johnsom_ | It will also be a problem with multiple controllers. So, I'm trying to push that calc down to the DB | 03:31 |
| rm_you | yeah i see the post you're talking about | 03:31 |
| johnsom_ | four hours of my life gone to this stinking issue... ugh | 03:32 |
| rm_you | what does the traceback say when it breaks | 03:33 |
| johnsom_ | Let me run it again so I don't give you the wrong output by mistake | 03:34 |
| *** vivek-ebay has joined #openstack-lbaas | 03:37 | |
| johnsom_ | https://gist.github.com/johnsom/f1f5766d6ee9eaad4ddd | 03:37 |
| johnsom_ | I put the code snippet in the comment | 03:39 |
| *** tiny-hands has quit IRC | 03:42 | |
| *** vivek-eb_ has joined #openstack-lbaas | 03:43 | |
| *** vivek-ebay has quit IRC | 03:43 | |
| *** crc32 has quit IRC | 03:49 | |
| xgerman | johnsom_ at that point I would say ML | 03:49 |
| johnsom_ | Yeah, might be there | 03:50 |
| *** KunalGan_ has quit IRC | 04:08 | |
| *** minwang2 has joined #openstack-lbaas | 04:17 | |
| *** minwang2 has quit IRC | 04:28 | |
| *** crc32 has joined #openstack-lbaas | 04:34 | |
| *** vivek-ebay has joined #openstack-lbaas | 04:35 | |
| *** vivek-eb_ has quit IRC | 04:38 | |
| *** ganeshna_ has joined #openstack-lbaas | 04:51 | |
| *** ganeshna has quit IRC | 04:54 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia: Implement UDP heartbeat sender and receiver https://review.openstack.org/201882 | 04:57 |
| rm_work | johnsom_: hmm, looking | 04:59 |
| johnsom_ | Ugly hack | 04:59 |
| johnsom_ | still WIP | 04:59 |
| rm_work | got distracted doing another coding exercise (writing a basic AVL Tree to make sure I still can) | 05:00 |
| rm_work | the things I do during my free time... | 05:00 |
| johnsom_ | Basically I figure out what database it is and feed it the native bit. | 05:00 |
| rm_work | ugh | 05:01 |
| johnsom_ | Signing off for the night. Feel free to update if you come up with an idea. | 05:01 |
| rm_work | i might have a very basic sqlalchemy testbed/DB i can play with | 05:01 |
| rm_work | I feel like I never actually write my own code, I just end up helping everyone else solve their issues, lol | 05:01 |
| rm_work | at least that has been my life recently | 05:01 |
| johnsom_ | Sorry for using the phone-a-friend | 05:02 |
| *** ajmiller_ has quit IRC | 05:02 | |
| rm_work | heh np | 05:11 |
| rm_work | I will play with it and see what I can find | 05:11 |
| *** haigang has quit IRC | 05:14 | |
| *** kiran-r has joined #openstack-lbaas | 05:18 | |
| *** mlavalle has quit IRC | 05:35 | |
| *** vivek-ebay has quit IRC | 05:37 | |
| *** amotoki has quit IRC | 05:39 | |
| *** numan has joined #openstack-lbaas | 05:55 | |
| *** kiran-r has quit IRC | 05:59 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Octavia driver correctly reports statuses https://review.openstack.org/213597 | 06:02 |
| *** bana_k has joined #openstack-lbaas | 06:05 | |
| *** amotoki has joined #openstack-lbaas | 06:09 | |
| openstackgerrit | Kawaguchi Kentaro proposed openstack/neutron-lbaas: add lbaasv2 api validation check for delay and timeout values https://review.openstack.org/209420 | 06:33 |
| *** crc32 has quit IRC | 06:36 | |
| *** haigang has joined #openstack-lbaas | 06:37 | |
| bana_k | anyone there ? | 06:54 |
| *** ganeshna_ has quit IRC | 06:57 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/214992 | 07:19 |
| *** jschwarz has joined #openstack-lbaas | 07:23 | |
| *** bana_k has quit IRC | 07:35 | |
| *** ganeshna has joined #openstack-lbaas | 07:50 | |
| rm_work | johnsom_: awesome, well this ate about 4 hours of my time as well, but it might be worth it since I think I have an answer for you, even if it's one you won't like at first :P | 07:53 |
| rm_work | johnsom_: basically, I think you're worrying about a problem that doesn't exist, re: timezone issues | 07:53 |
| rm_work | and skew shouldn't be an issue either, assuming we use ntp to keep servers in sync, which is pretty much a no-brainer | 07:54 |
| rm_work | worst case one might be off by 0.001 or something | 07:54 |
| rm_work | and hopefully our timeout thresholds aren't cutting it anywhere near that close | 07:55 |
| openstackgerrit | Merged openstack/octavia: Updating cert_parser for cert loading https://review.openstack.org/208035 | 08:10 |
| rm_work | I really wanted to merge https://review.openstack.org/#/c/202829/18 but I just couldn't bring myself to do it. T_T | 08:27 |
| openstackgerrit | Merged openstack/octavia: Correct shebang in create_certificates.sh https://review.openstack.org/214633 | 08:33 |
| *** apuimedo has joined #openstack-lbaas | 08:46 | |
| openstackgerrit | Bertrand Lallau proposed openstack/octavia: Correct messaging config set in devstack https://review.openstack.org/215023 | 09:00 |
| *** sbalukoff has joined #openstack-lbaas | 09:09 | |
| *** apuimedo is now known as apuimedo|away | 09:10 | |
| openstackgerrit | venkata anil proposed openstack/neutron-lbaas: Update status of members from Haproxy statistics https://review.openstack.org/215027 | 09:12 |
| openstackgerrit | Bharath M proposed openstack/octavia: Add Housekeeping to manage spare amphora and DB cleanup https://review.openstack.org/202829 | 10:04 |
| *** tiny-hands has joined #openstack-lbaas | 10:06 | |
| *** tiny-hands has quit IRC | 10:08 | |
| *** amotoki has quit IRC | 10:08 | |
| *** bharath_ has joined #openstack-lbaas | 10:13 | |
| *** ganeshna has quit IRC | 10:34 | |
| *** apuimedo|away is now known as apuimedo | 11:04 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements https://review.openstack.org/215088 | 11:22 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements https://review.openstack.org/214992 | 11:24 |
| *** vjay10 has joined #openstack-lbaas | 11:38 | |
| *** woodster_ has joined #openstack-lbaas | 11:48 | |
| *** vjay10 has quit IRC | 11:57 | |
| *** bharath_ has quit IRC | 12:13 | |
| *** haigang has quit IRC | 12:18 | |
| *** chlong has joined #openstack-lbaas | 12:46 | |
| *** chlong has quit IRC | 13:00 | |
| *** chlong has joined #openstack-lbaas | 13:04 | |
| *** tiny-hands has joined #openstack-lbaas | 13:05 | |
| *** clev-away is now known as clev | 13:36 | |
| *** diogogmt has joined #openstack-lbaas | 13:39 | |
| *** eezhova has quit IRC | 13:59 | |
| *** eezhova has joined #openstack-lbaas | 13:59 | |
| *** diogogmt has quit IRC | 14:04 | |
| *** starry has joined #openstack-lbaas | 14:12 | |
| *** ajmiller has joined #openstack-lbaas | 14:13 | |
| *** kiran-r has joined #openstack-lbaas | 14:14 | |
| *** kiran-r has quit IRC | 14:15 | |
| *** diogogmt has joined #openstack-lbaas | 14:17 | |
| *** ganeshna has joined #openstack-lbaas | 14:25 | |
| *** starry has quit IRC | 14:28 | |
| *** albertom-afk is now known as albertom | 14:31 | |
| *** diogogmt has quit IRC | 14:34 | |
| *** vjay10 has joined #openstack-lbaas | 14:35 | |
| *** vjay10 has quit IRC | 14:44 | |
| openstackgerrit | Bertrand Lallau proposed openstack/octavia: Correct rabbit messaging config set in devstack https://review.openstack.org/215023 | 14:48 |
| *** fnaval has quit IRC | 14:51 | |
| *** minwang2 has joined #openstack-lbaas | 14:53 | |
| *** mlavalle has joined #openstack-lbaas | 14:55 | |
| *** fnaval has joined #openstack-lbaas | 15:01 | |
| *** ganeshna has quit IRC | 15:03 | |
| *** ganeshna_ has joined #openstack-lbaas | 15:03 | |
| johnsom_ | rm_work I hit this because the health check was off and failing over amps when it shouldn't. That's when I decided we should address this now. | 15:04 |
| *** alejandrito has joined #openstack-lbaas | 15:08 | |
| *** ganeshna_ has quit IRC | 15:10 | |
| openstackgerrit | Merged openstack/octavia: Updated from global requirements https://review.openstack.org/214992 | 15:11 |
| *** diogogmt has joined #openstack-lbaas | 15:16 | |
| *** kfox1111 has quit IRC | 15:19 | |
| *** diogogmt has quit IRC | 15:19 | |
| *** diogogmt has joined #openstack-lbaas | 15:22 | |
| *** TrevorV has joined #openstack-lbaas | 15:37 | |
| *** TrevorV has quit IRC | 15:41 | |
| *** TrevorV has joined #openstack-lbaas | 15:43 | |
| *** jorgem has joined #openstack-lbaas | 15:44 | |
| *** bana_k has joined #openstack-lbaas | 16:00 | |
| *** bana_k has quit IRC | 16:04 | |
| *** crc32 has joined #openstack-lbaas | 16:05 | |
| *** minwang2 has quit IRC | 16:16 | |
| *** jorgem has quit IRC | 16:17 | |
| *** kiran-r has joined #openstack-lbaas | 16:20 | |
| *** jorgem has joined #openstack-lbaas | 16:22 | |
| *** vivek-ebay has joined #openstack-lbaas | 16:27 | |
| *** _kiran_ has joined #openstack-lbaas | 16:30 | |
| *** kiran-r has quit IRC | 16:31 | |
| *** numan has quit IRC | 16:35 | |
| *** numan has joined #openstack-lbaas | 16:36 | |
| blogan | johnsom_: you said you were getting off around the same time iw as lst night, but i see you stuck it out another 3 hours | 16:37 |
| jschwarz | blogan, hi, can you please have a look at https://review.openstack.org/174384 ? :) | 16:39 |
| blogan | jschwarz: sure | 16:39 |
| jschwarz | thanks :) | 16:39 |
| *** numan has quit IRC | 16:39 | |
| *** clev is now known as clev-away | 16:45 | |
| *** vjay10 has joined #openstack-lbaas | 16:46 | |
| johnsom_ | blogan Actually I came back to see if Adam had worked his magic | 16:50 |
| *** minwang2 has joined #openstack-lbaas | 16:53 | |
| *** jschwarz has quit IRC | 17:05 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Adding amphora failover flows https://review.openstack.org/202336 | 17:06 |
| *** bana_k has joined #openstack-lbaas | 17:08 | |
| *** jorgem has quit IRC | 17:09 | |
| *** KunalGandhi has joined #openstack-lbaas | 17:11 | |
| bana_k | in ssh driver we have for amp in listener.load_balancer.amphorae for update | 17:15 |
| *** Bana_ has joined #openstack-lbaas | 17:19 | |
| *** bana_k has quit IRC | 17:20 | |
| *** bana_k has joined #openstack-lbaas | 17:20 | |
| *** vjay10 has quit IRC | 17:24 | |
| *** madhu_ak has joined #openstack-lbaas | 17:26 | |
| *** Bana_ has quit IRC | 17:30 | |
| openstackgerrit | Sherif Abdelwahab proposed openstack/octavia: Amphora Flows and Service Drivers for Active Standby https://review.openstack.org/206252 | 17:40 |
| *** jorgem has joined #openstack-lbaas | 17:41 | |
| bana_k | sorry did someone answer my question I got disconnected for a while :/ | 17:45 |
| *** clev-away is now known as clev | 17:48 | |
| *** bharath has joined #openstack-lbaas | 17:48 | |
| *** SumitNaiksatam has joined #openstack-lbaas | 17:48 | |
| johnsom_ | bana_k I don't think we got the full question. Can you repeat it? | 17:49 |
| *** johnsom_ is now known as johnsom | 17:49 | |
| *** _kiran_ has quit IRC | 17:50 | |
| TrevorV | bana_k, I only say you stating something about the ssh_driver: | 17:54 |
| TrevorV | in ssh driver we have for amp in listener.load_balancer.amphorae for update | 17:54 |
| TrevorV | That's what you had typed | 17:54 |
| bana_k | yes in amphorae ssh driver we are updating the amp in loop | 17:54 |
| bana_k | and I was wondering for amp in listener.load_balancer.amphorae | 17:55 |
| bana_k | about this | 17:55 |
| bana_k | is this mean there are more than one amphorae per loadbalancer/listener ? | 17:55 |
| blogan | bana_k: lb to amp is 1:M | 17:56 |
| blogan | to support HA | 17:56 |
| openstackgerrit | Brandon Logan proposed openstack/octavia: Setup base neutron network driver https://review.openstack.org/209210 | 17:57 |
| bana_k | oh for HA eh. | 17:57 |
| bana_k | thanks | 17:57 |
| blogan | xgerman, rm_work: fixedt hat typo | 17:57 |
| TrevorV | bana_k, did you figure out the information about the failed deletes? | 17:58 |
| bana_k | sorry not yet. I had some other things on my head. | 17:58 |
| bana_k | will do it today | 17:58 |
| bana_k | n ll confirm | 17:58 |
| TrevorV | No pressure, was curious | 17:58 |
| bana_k | ok. | 17:58 |
| TrevorV | Prioritize appropriately, I'm not worried about it, just didn't want you to be hung up on something because of it | 17:58 |
| bana_k | blogan : if thats the case is it scalable? | 17:59 |
| blogan | bana_k: in what way? | 17:59 |
| bana_k | Trevorv : got it. | 17:59 |
| bana_k | blogan : if its 1:M what happens in the case we have lots of LB with 1:M | 18:00 |
| bana_k | n we have frequent update requests | 18:00 |
| blogan | bana_k: well 1) make sure your nova has the capacity 2) if that isn't adequate, we'll also support just a single node topology where its 1:1 | 18:01 |
| blogan | basically what we ahve right now | 18:01 |
| blogan | bana_k: frequent update requests? | 18:01 |
| bana_k | to ssh driver | 18:01 |
| blogan | bana_k: updates to what? all lbs? | 18:01 |
| bana_k | yes | 18:01 |
| *** abdelwas has joined #openstack-lbaas | 18:02 | |
| blogan | bana_k: well load balancers usually don't get updated frequently, they're mostly just a set an forget thing | 18:02 |
| blogan | from our experience with our current cloud load balancing | 18:02 |
| bana_k | blogan : ok, makes sense | 18:02 |
| bana_k | thanks for the clarification | 18:02 |
| blogan | bana_k: if you have scalability concerns by all means share them | 18:06 |
| *** jorgem has quit IRC | 18:06 | |
| blogan | bana_k: might be somethign we dont realize | 18:06 |
| bana_k | sure :) | 18:06 |
| johnsom | bedis Can you point us to a description of the meaning of all of the stats socket "status" values? | 18:10 |
| johnsom | Example "OPEN" | 18:11 |
| *** minwang2 has quit IRC | 18:12 | |
| rm_work | blogan: did you want to look at https://review.openstack.org/#/c/202829/19 ? | 18:18 |
| rm_work | blogan: or can I +A | 18:18 |
| rm_work | johnsom: How far off were the times for you? Mine were seriously spot on | 18:19 |
| rm_work | johnsom: and if we're using ntp as we should be, I can't imagine it would be a real problem | 18:19 |
| johnsom | It was far enough off to cause early failover | 18:19 |
| rm_work | at least, it seems the other option is bad enough that IMO it is worse | 18:19 |
| johnsom | I posted some comments on the patchset | 18:19 |
| rm_work | we'd have to essentially manually support every type of SQL dialect | 18:19 |
| *** madhu_ak has quit IRC | 18:22 | |
| johnsom | If I remember correctly, all of the mainstream DB's support the mysql syntax, but I'm not going to code it that way without testing. | 18:23 |
| *** madhu_ak has joined #openstack-lbaas | 18:24 | |
| blogan | rm_work: let me look at it real quick | 18:24 |
| rm_work | johnsom: neither of Postgres or SQLite support anything remotely similar to the MySQL syntax, or to each other | 18:25 |
| rm_work | and what I remember of MS' T-SQL and Oracle's PL/SQL is that they are also totally different for this | 18:25 |
| rm_work | Postgres doesn't have ANY of the time functions I mentioned from my research, and neither does SQLite | 18:26 |
| johnsom | Postgresql does support interval. SQLite - not what I consider mainstream db | 18:26 |
| rm_work | yeah, interval | 18:26 |
| johnsom | Oracle also does interval | 18:26 |
| rm_work | but date_add, addtime, datetimediff | 18:26 |
| rm_work | none of those exist | 18:26 |
| rm_work | and that is the base function | 18:26 |
| johnsom | So does SQL server | 18:26 |
| rm_work | func.date_add() doesn't do anything magic, it just becomes the text "date_add( )" | 18:27 |
| rm_work | right, INTERVAL is not the issue at all | 18:27 |
| johnsom | SQL server is dateadd | 18:27 |
| rm_work | dateadd != date_add | 18:28 |
| johnsom | I thought sqlalchemy translated that func.date_add for the dialect | 18:28 |
| rm_work | nope | 18:28 |
| rm_work | not at all | 18:28 |
| rm_work | func.* is a magic function | 18:28 |
| rm_work | it literally just takes the function name you put and makes it into text in the SQL query | 18:28 |
| johnsom | Hmm, I thought it was one of just a few exceptions where it actually translates it | 18:28 |
| rm_work | unless it is one of a few very specific exceptions | 18:28 |
| rm_work | date_add is not one of them | 18:28 |
| rm_work | they are: | 18:28 |
| rm_work | coalesce / max / min / sum / now / concat / char_length / random / count / current_date / current_time / current_timestamp / current_user / localtime / localtimestamp / session_user / sysdate / user | 18:30 |
| johnsom | My impression of sqlalchemy is getting even lower that it was before.... | 18:31 |
| rm_work | https://github.com/zzzeek/sqlalchemy/blob/master/lib/sqlalchemy/sql/functions.py#L537 | 18:32 |
| rm_work | starting there and going down | 18:32 |
| rm_work | oh, I think "next_value" is also one | 18:32 |
| *** kiran-r has joined #openstack-lbaas | 18:33 | |
| *** abdelwas_ has joined #openstack-lbaas | 18:38 | |
| abdelwas_ | Hi blogan johnsom xgerman rm_you, This is the patch for the Active/Standby https://review.openstack.org/#/c/206252/14 | 18:38 |
| abdelwas_ | Would it be possible to start reviewing this patch. It is indeed missing unit and functional tests, but reviewing it in parallel while I develop tests will expedite the whole thing | 18:39 |
| *** madhu_ak has quit IRC | 18:40 | |
| *** abdelwas has quit IRC | 18:40 | |
| openstackgerrit | Merged openstack/octavia: Add Housekeeping to manage spare amphora and DB cleanup https://review.openstack.org/202829 | 18:45 |
| *** madhu_ak has joined #openstack-lbaas | 18:46 | |
| *** jorgem has joined #openstack-lbaas | 18:49 | |
| *** vivek-ebay has quit IRC | 18:49 | |
| *** vivek-ebay has joined #openstack-lbaas | 18:49 | |
| *** davidlenwell has quit IRC | 18:50 | |
| *** davidlenwell has joined #openstack-lbaas | 18:51 | |
| *** kiran-r has quit IRC | 18:53 | |
| *** davidlenwell has quit IRC | 18:58 | |
| *** davidlenwell has joined #openstack-lbaas | 18:59 | |
| openstackgerrit | Merged openstack/octavia: Setup base neutron network driver https://review.openstack.org/209210 | 19:04 |
| *** Aish has joined #openstack-lbaas | 19:05 | |
| blogan | abdelwas_: it is not possible to review it, you're request has been denied | 19:06 |
| blogan | your | 19:06 |
| blogan | i kid | 19:06 |
| rm_work | yeah i will look at it after i get in | 19:08 |
| *** jorgem has quit IRC | 19:10 | |
| *** madhu_ak has quit IRC | 19:22 | |
| *** woodster_ has quit IRC | 19:27 | |
| xgerman | thanks guys - really appreciated | 19:28 |
| *** clev is now known as clev-away | 19:28 | |
| *** davidlenwell has quit IRC | 19:28 | |
| *** sbalukoff has quit IRC | 19:28 | |
| *** clev-away is now known as clev | 19:30 | |
| *** davidlenwell has joined #openstack-lbaas | 19:31 | |
| *** woodster_ has joined #openstack-lbaas | 19:31 | |
| *** madhu_ak has joined #openstack-lbaas | 19:31 | |
| *** rm_work is now known as rm_work|away | 19:32 | |
| abdelwas_ | blogan: since you are kidding, I will raise the request bar to a merge :) | 19:34 |
| abdelwas_ | rm_work thank you | 19:34 |
| *** jorgem has joined #openstack-lbaas | 19:35 | |
| *** bana_k has quit IRC | 19:38 | |
| *** sbalukoff has joined #openstack-lbaas | 19:42 | |
| *** alejandrito has quit IRC | 19:47 | |
| *** alejandrito has joined #openstack-lbaas | 19:49 | |
| *** ajmiller has quit IRC | 19:53 | |
| openstackgerrit | Phillip Toohill proposed openstack/octavia: Fixes missed refactoring of barbican client https://review.openstack.org/215314 | 20:08 |
| *** alejandrito has quit IRC | 20:09 | |
| *** alejandrito has joined #openstack-lbaas | 20:13 | |
| openstackgerrit | Brandon Logan proposed openstack/octavia: Check for port ownership on vip deallocation https://review.openstack.org/215322 | 20:39 |
| blogan | johnsom: that fixes the lb delete error i was running nito and i think yall were | 20:41 |
| johnsom | Ok, will check it out later today | 20:41 |
| openstackgerrit | Phillip Toohill proposed openstack/octavia: Fixes missed refactoring of barbican client https://review.openstack.org/215314 | 20:42 |
| johnsom | spelunking in haproxy code to get our status messages right.... The docs don't exactly cover the codes well.... | 20:43 |
| blogan | johnsom: that sounds awful | 20:43 |
| xgerman | yeah, it took me some trial and error to figure out what Up, etc. means | 20:57 |
| blogan | xgerman: duh, Up is the opposite of Down | 21:01 |
| xgerman | you would think... | 21:01 |
| blogan | are you telling me Up is Down? Left is Right? | 21:01 |
| xgerman | It’s not that sinister... | 21:02 |
| blogan | i think bedis created a bouncer just for this channel and never uses it | 21:02 |
| johnsom | It's things list "OPEN" and "FULL" that throw me off.... | 21:02 |
| johnsom | OPEN means the listener has connection slots, FULL means the listener is at maxconn BTW | 21:03 |
| xgerman | also it’s important to name things properly | 21:03 |
| *** rm_work|away is now known as rm_work | 21:13 | |
| *** tiny-hands has left #openstack-lbaas | 21:21 | |
| *** Aish has quit IRC | 21:24 | |
| *** TrevorV has quit IRC | 21:33 | |
| *** clev is now known as clev-away | 21:33 | |
| *** Aish has joined #openstack-lbaas | 21:33 | |
| *** clev-away is now known as clev | 21:34 | |
| openstackgerrit | Trevor Vardeman proposed openstack/octavia: Adding amphora failover flows https://review.openstack.org/202336 | 21:35 |
| *** abdelwas_ has quit IRC | 21:40 | |
| *** alejandrito has quit IRC | 21:49 | |
| *** alejandrito has joined #openstack-lbaas | 21:52 | |
| *** diogogmt_ has joined #openstack-lbaas | 22:00 | |
| *** diogogmt has quit IRC | 22:00 | |
| *** diogogmt_ is now known as diogogmt | 22:00 | |
| openstackgerrit | min wang proposed openstack/octavia: Add cert tracking and rotating in Housekeeping https://review.openstack.org/215359 | 22:06 |
| openstackgerrit | Merged openstack/octavia: Fixes missed refactoring of barbican client https://review.openstack.org/215314 | 22:20 |
| *** chlong has quit IRC | 22:25 | |
| openstackgerrit | Merged openstack/octavia: Check for port ownership on vip deallocation https://review.openstack.org/215322 | 22:34 |
| *** Aish has quit IRC | 22:35 | |
| *** jorgem has quit IRC | 22:37 | |
| openstackgerrit | Brandon Logan proposed openstack/octavia: test_plug_vip not testing properly in py34 https://review.openstack.org/212868 | 22:43 |
| rm_work | xgerman / johnsom https://review.openstack.org/#/c/212868/3 | 22:46 |
| rm_work | minor py34 issue | 22:46 |
| johnsom | Ok | 22:46 |
| xgerman | +A | 22:47 |
| openstackgerrit | Merged openstack/octavia: test_plug_vip not testing properly in py34 https://review.openstack.org/212868 | 22:57 |
| *** crc32 has quit IRC | 22:57 | |
| rm_work | we're really knocking out a lot of the open patchsets :) | 23:00 |
| rm_work | looking nice to see the list of CRs dwindling | 23:00 |
| rm_work | anyway, bbl | 23:00 |
| *** rm_work is now known as rm_work|away | 23:01 | |
| *** alejandrito has quit IRC | 23:06 | |
| *** tiny-hands has joined #openstack-lbaas | 23:07 | |
| *** diogogmt has quit IRC | 23:12 | |
| *** diogogmt has joined #openstack-lbaas | 23:14 | |
| openstackgerrit | min wang proposed openstack/octavia: Add cert tracking and rotating in Housekeeping https://review.openstack.org/215359 | 23:34 |
| *** SumitNaiksatam has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!