| *** yamamoto_ has joined #openstack-lbaas | 00:14 | |
| johnsom | Well, fun. So, the dbus option, which would work well for us, is only in the 1.3.x version (distros are shipping 1.2.x). So, short of doing the whole snmp thing which I really don't like the overhead for such a small use, I think I am going to punt and do process checking and call it good. | 00:22 |
|---|---|---|
| *** leitan has quit IRC | 00:38 | |
| *** cody-somerville has quit IRC | 00:45 | |
| *** leitan has joined #openstack-lbaas | 00:47 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 00:47 |
| *** KeithMnemonic1 has joined #openstack-lbaas | 00:48 | |
| *** KeithMnemonic2 has quit IRC | 00:51 | |
| *** leitan_ has joined #openstack-lbaas | 00:52 | |
| *** leitan has quit IRC | 00:55 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 01:03 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 01:13 |
| leitan_ | johnsom: lets call it a working v1 version | 01:23 |
| johnsom | Ha, yeah. | 01:23 |
| johnsom | BTW, if you are happy with https://review.openstack.org/470051 can you vote? | 01:24 |
| * leitan_ checking | 01:24 | |
| johnsom | It helps the other cores to know that the reporter tested it as working | 01:24 |
| leitan_ | johnsom: voted +1 | 01:26 |
| johnsom | Thanks | 01:27 |
| johnsom | I'm working on the logic for the keepalive check. Should have something posted soon | 01:27 |
| leitan_ | johnsom: great, i cant test it if im still here, or tomorrow | 01:30 |
| *** fnaval has quit IRC | 01:42 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 01:43 |
| *** sanfern has joined #openstack-lbaas | 01:48 | |
| *** fnaval has joined #openstack-lbaas | 02:15 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Check keepalived health in the amphora https://review.openstack.org/470081 | 02:17 |
| *** leitan_ has quit IRC | 02:17 | |
| johnsom | Well, that should do the trick. I still need to update the tests though, so marking WIP. | 02:17 |
| johnsom | leitan If you have time to test while I am updating tests, https://review.openstack.org/470081 | 02:18 |
| *** links has joined #openstack-lbaas | 02:23 | |
| *** kbyrne has quit IRC | 02:24 | |
| *** kbyrne has joined #openstack-lbaas | 02:26 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 02:29 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-lbaas master: Updated from global requirements https://review.openstack.org/467471 | 02:30 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia master: Updated from global requirements https://review.openstack.org/470038 | 02:32 |
| *** fnaval_ has joined #openstack-lbaas | 02:33 | |
| *** fnaval has quit IRC | 02:36 | |
| *** kobis has joined #openstack-lbaas | 02:39 | |
| *** kobis has quit IRC | 02:47 | |
| *** sanfern has quit IRC | 02:59 | |
| *** cody-somerville has joined #openstack-lbaas | 03:23 | |
| *** armax has quit IRC | 03:41 | |
| *** gans819 has joined #openstack-lbaas | 03:42 | |
| *** armax has joined #openstack-lbaas | 03:45 | |
| *** armax has quit IRC | 04:00 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 04:08 |
| *** sanfern has joined #openstack-lbaas | 04:14 | |
| *** gcheresh has joined #openstack-lbaas | 04:23 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Check keepalived health in the amphora https://review.openstack.org/470081 | 04:32 |
| *** gcheresh has quit IRC | 04:32 | |
| *** yamamoto_ has quit IRC | 04:46 | |
| *** gcheresh has joined #openstack-lbaas | 05:08 | |
| *** rcernin has joined #openstack-lbaas | 05:26 | |
| *** yamamoto_ has joined #openstack-lbaas | 05:32 | |
| *** csomerville has joined #openstack-lbaas | 05:51 | |
| *** cody-somerville has quit IRC | 05:54 | |
| openstackgerrit | Reedip proposed openstack/octavia master: Extend api to accept qos_policy_id https://review.openstack.org/458308 | 06:02 |
| openstackgerrit | Merged openstack/neutron-lbaas master: Updated from global requirements https://review.openstack.org/467471 | 06:57 |
| *** tesseract has joined #openstack-lbaas | 07:04 | |
| *** aojea has joined #openstack-lbaas | 07:12 | |
| *** gans85 has joined #openstack-lbaas | 07:19 | |
| *** gans819 has quit IRC | 07:21 | |
| *** sanfern has quit IRC | 07:21 | |
| *** sanfern has joined #openstack-lbaas | 07:22 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7rule commands to client https://review.openstack.org/462808 | 08:13 |
| *** aojea has quit IRC | 08:39 | |
| *** aojea has joined #openstack-lbaas | 08:40 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add healthmonitor commands to client https://review.openstack.org/463830 | 08:47 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Update help text for all commands https://review.openstack.org/466441 | 08:56 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7policy commands to client https://review.openstack.org/462431 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add Listener commands to client https://review.openstack.org/461231 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Update help text for all commands https://review.openstack.org/466441 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add member commands to client https://review.openstack.org/463035 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add healthmonitor commands to client https://review.openstack.org/463830 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add pool commands to client https://review.openstack.org/461992 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add l7rule commands to client https://review.openstack.org/462808 | 09:02 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Update help text for all commands https://review.openstack.org/466441 | 09:07 |
| *** JudeC has quit IRC | 09:18 | |
| *** cpuga has quit IRC | 09:21 | |
| *** gans819 has joined #openstack-lbaas | 09:27 | |
| *** gans85 has quit IRC | 09:30 | |
| *** gans819 has quit IRC | 09:31 | |
| *** aojea has quit IRC | 09:33 | |
| *** aojea has joined #openstack-lbaas | 09:33 | |
| *** cpuga has joined #openstack-lbaas | 09:35 | |
| *** gcheresh_ has joined #openstack-lbaas | 09:47 | |
| *** gcheresh has quit IRC | 09:47 | |
| *** cpuga has quit IRC | 09:55 | |
| *** lunarlamp has quit IRC | 10:08 | |
| *** gcheresh_ has quit IRC | 10:11 | |
| *** mariusv has joined #openstack-lbaas | 10:11 | |
| *** yamamoto_ has quit IRC | 10:22 | |
| *** mariusv has quit IRC | 10:44 | |
| *** mariusv has joined #openstack-lbaas | 10:46 | |
| *** sanfern has quit IRC | 10:47 | |
| *** cpuga has joined #openstack-lbaas | 10:55 | |
| *** cpuga has quit IRC | 11:00 | |
| *** fnaval_ has quit IRC | 11:06 | |
| *** yamamoto has joined #openstack-lbaas | 11:47 | |
| *** catintheroof has joined #openstack-lbaas | 12:08 | |
| *** aojea has quit IRC | 12:15 | |
| *** aojea has joined #openstack-lbaas | 12:16 | |
| *** aojea has quit IRC | 12:20 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/octavia master: Updated from global requirements https://review.openstack.org/470038 | 12:35 |
| *** yamamoto has quit IRC | 12:43 | |
| *** yamamoto has joined #openstack-lbaas | 12:50 | |
| *** cpuga has joined #openstack-lbaas | 12:57 | |
| *** gcheresh_ has joined #openstack-lbaas | 12:59 | |
| *** yamamoto has quit IRC | 13:00 | |
| *** cpuga has quit IRC | 13:01 | |
| *** aojea has joined #openstack-lbaas | 13:17 | |
| *** aojea has quit IRC | 13:22 | |
| *** aojea has joined #openstack-lbaas | 13:35 | |
| *** fnaval has joined #openstack-lbaas | 13:35 | |
| *** links has quit IRC | 13:35 | |
| *** fnaval_ has joined #openstack-lbaas | 13:39 | |
| *** fnaval has quit IRC | 13:42 | |
| *** yamamoto has joined #openstack-lbaas | 14:01 | |
| *** yamamoto has quit IRC | 14:06 | |
| *** yamamoto has joined #openstack-lbaas | 14:07 | |
| *** csomerville has quit IRC | 14:25 | |
| *** chandankumar is now known as chandan|sleeping | 14:28 | |
| *** cody-somerville has joined #openstack-lbaas | 14:31 | |
| *** Mahesh_ has joined #openstack-lbaas | 14:37 | |
| Mahesh_ | hi | 14:38 |
| Mahesh_ | i have configured LBaas V1 in mitakka | 14:39 |
| Mahesh_ | i when i created a new LB pool shows error status | 14:39 |
| *** catintheroof has quit IRC | 14:45 | |
| Mahesh_ | hi | 14:46 |
| *** Mahesh_ has left #openstack-lbaas | 14:47 | |
| johnsom | Mahesh_ Hi, unfortunately LBaaS v1 has been end-of-life for a number of years and we don't even have the code for it any more. I am not of much help with v1. | 14:48 |
| *** aojea has quit IRC | 14:56 | |
| *** MaheshS has joined #openstack-lbaas | 15:02 | |
| MaheshS | i have configure LBaas V1 in mitakk | 15:02 |
| MaheshS | when i lunch a new Loadbalancer pool the status is error | 15:03 |
| MaheshS | ERROR neutron_lbaas.services.loadbalancer.agent.agent_manager AttributeError: 'HaproxyNSDriver' object has no attribute 'create_pool' | 15:03 |
| *** MaheshS has quit IRC | 15:04 | |
| johnsom | MaheshS I really recommend you do not use LBaaS v1. Mitaka was the last release that even has the code. It was deleted after that. | 15:04 |
| johnsom | It has been deprecated for years | 15:05 |
| *** rcernin has quit IRC | 15:05 | |
| *** cpuga has joined #openstack-lbaas | 15:10 | |
| *** cpuga has quit IRC | 15:14 | |
| *** armax has joined #openstack-lbaas | 15:15 | |
| *** sanfern has joined #openstack-lbaas | 15:27 | |
| *** cody-somerville has quit IRC | 15:34 | |
| *** gcheresh has joined #openstack-lbaas | 15:57 | |
| *** gcheresh_ has quit IRC | 15:57 | |
| *** tesseract has quit IRC | 16:00 | |
| *** cody-somerville has joined #openstack-lbaas | 16:02 | |
| *** cody-somerville has quit IRC | 16:02 | |
| *** cody-somerville has joined #openstack-lbaas | 16:02 | |
| *** krypto has joined #openstack-lbaas | 16:14 | |
| *** yamamoto has quit IRC | 16:14 | |
| *** yamamoto has joined #openstack-lbaas | 16:15 | |
| *** pcaruana has quit IRC | 16:16 | |
| johnsom | Hmmm, our fields in the OSC are pretty names i.e. "Admin State Up" instead of "admin_state_up" | 16:19 |
| *** yamamoto has quit IRC | 16:19 | |
| johnsom | I wonder if that will be a problem for specifying columns and stuff | 16:20 |
| johnsom | https://www.irccloud.com/pastebin/gg9lhN4U/ | 16:21 |
| johnsom | Hmmm | 16:22 |
| *** sanfern has quit IRC | 16:23 | |
| *** sanfern has joined #openstack-lbaas | 16:23 | |
| *** cpuga has joined #openstack-lbaas | 16:26 | |
| *** fnaval_ has quit IRC | 16:26 | |
| johnsom | Both neutron and glance show the API column names | 16:27 |
| johnsom | Yeah, I can get "Name" to work, but no combination of "Admin State Up" work | 16:29 |
| *** chandan|sleeping is now known as chandankumar | 16:30 | |
| *** krypto has quit IRC | 16:34 | |
| *** fnaval has joined #openstack-lbaas | 16:35 | |
| johnsom | https://bugs.launchpad.net/octavia/+bug/1695323 | 16:43 |
| openstack | Launchpad bug 1695323 in octavia "Octavia v2 API lb create admin_state_down shows ONLINE op_status" [High,Triaged] | 16:43 |
| johnsom | Octavia v2 API lb create admin_state_up = False shows ONLINE op_status (fixed the title) | 16:44 |
| *** blogan_ has joined #openstack-lbaas | 16:54 | |
| *** blogan has quit IRC | 16:56 | |
| johnsom | This on is bad... | 17:00 |
| johnsom | https://bugs.launchpad.net/octavia/+bug/1695331 | 17:00 |
| openstack | Launchpad bug 1695331 in octavia "Octavia v2 API Requesting a VIP address is not working" [Critical,Triaged] | 17:00 |
| *** cody-somerville has quit IRC | 17:06 | |
| *** cody-somerville has joined #openstack-lbaas | 17:06 | |
| *** yamamoto has joined #openstack-lbaas | 17:16 | |
| *** yamamoto has quit IRC | 17:21 | |
| rm_work | umm | 17:23 |
| *** sshank has joined #openstack-lbaas | 17:26 | |
| *** JudeC has joined #openstack-lbaas | 17:26 | |
| *** marcin12345 has joined #openstack-lbaas | 17:26 | |
| johnsom | Ok, that is my last set of comments on the OSC loadbalancer commands patch | 17:27 |
| johnsom | JudeC I am done reviewing the OSC loadbalancer patch | 17:27 |
| *** harlowja has quit IRC | 17:28 | |
| JudeC | Awesome, I was reading through your comments on it. Thanks for taking the time to review all of that. | 17:28 |
| johnsom | Of course, sorry for putting on the rubber glove, but that is kind of the role.... grin | 17:29 |
| JudeC | I don't mind at all. | 17:32 |
| *** fnaval has quit IRC | 17:32 | |
| rm_work | johnsom: is port 1025 what we use for VRRP? | 17:32 |
| rm_work | noticed it's also open on our Security Groups | 17:32 |
| johnsom | Umm, let me look again. | 17:33 |
| johnsom | Ugh, I had this all captured on a wiki at a previous life. It's either for that or the haproxy sync, which increments for each listener since they are separate processes | 17:35 |
| johnsom | HAPROXY_BASE_PEER_PORT = 1025 | 17:36 |
| johnsom | It's for haproxy peer sync. It starts at 1025 and increments, though not necessarily in order due to deletes, etc. | 17:36 |
| johnsom | VRRP is it's own protocol I think | 17:37 |
| johnsom | JudeC Opps, found one more. updated_at is missing from the output | 17:38 |
| johnsom | Added another comment | 17:38 |
| *** fnaval has joined #openstack-lbaas | 17:43 | |
| rm_work | johnsom: ah ok | 17:44 |
| rm_work | johnsom: it's interesting that it still opens that port on the SG when in SINGLE mode | 17:45 |
| johnsom | Yeah, well, technically we don't need allowed address pairs on the SINGLE topo ports either, but ... | 17:45 |
| rm_work | lol | 17:46 |
| rm_work | well, MY driver does | 17:46 |
| rm_work | since it operates with FLIPs | 17:46 |
| rm_work | and here those require AAP from the FLIP :) | 17:46 |
| rm_work | also the OSC flip associate command is totally broken | 17:47 |
| rm_work | have to use neutron-client to associate them T_T | 17:47 |
| johnsom | Ha, yeah, lots of things with OSC are um, grade D but kind of functional | 17:47 |
| johnsom | So, yeah, if you are motivated to not open the peer ports, go for it | 17:47 |
| rm_work | http://paste.openstack.org/raw/611325/ | 17:48 |
| rm_work | it's via "server add floating ip" | 17:48 |
| rm_work | not "floating ip associate" | 17:48 |
| rm_work | which first of all is wtf | 17:48 |
| rm_work | and second, it seems to *guess* which port to attach to | 17:48 |
| rm_work | which it gets wrong. | 17:48 |
| rm_work | IDKWTF | 17:49 |
| johnsom | Nice! | 17:49 |
| johnsom | It should just bind to ALL of them, right? hahahaha | 17:49 |
| johnsom | plus, ip-addresses are always unique in a project... Oye.... | 17:50 |
| johnsom | Wow | 17:50 |
| rm_work | who thought this up | 17:50 |
| rm_work | someone who doesn't understand how floating ips work, i guess | 17:51 |
| *** cody-somerville has quit IRC | 17:51 | |
| johnsom | Maybe that was a nova networking thing???? | 17:51 |
| rm_work | in OSC?? | 17:52 |
| rm_work | lol | 17:52 |
| johnsom | Hey, grasping at straws | 17:52 |
| rm_work | it DOES associate a flip... | 17:52 |
| rm_work | just not at all correctly | 17:52 |
| rm_work | it seems to pick the first fixed-ip on a box and use that port/ip | 17:52 |
| rm_work | and ... maybe the first FLIP it finds with that IP? | 17:53 |
| rm_work | fortunately we don't have overlapping FLIP subnets | 17:53 |
| *** JudeC has quit IRC | 17:54 | |
| johnsom | rm_work You were asking about accounts and authorization, did lubosz ever finish hooking up the RBAC stuff? Are we checking that beyond just admin/not-admin yet? | 17:58 |
| rm_work | not sure | 17:58 |
| rm_work | i just noticed that when i try to request things with project_id | 17:59 |
| rm_work | it doesn't actually work | 17:59 |
| rm_work | but I CAN delete by ID if i know the ID | 17:59 |
| rm_work | so i don't know whether it's that my account is not coming up as admin | 17:59 |
| rm_work | and that we don't check properly on delete | 17:59 |
| johnsom | https://bugs.launchpad.net/octavia/+bug/1662985 | 17:59 |
| openstack | Launchpad bug 1662985 in octavia "Octavia needs to validate the requester has permission to create/update objects" [High,Triaged] | 17:59 |
| *** csomerville has joined #openstack-lbaas | 17:59 | |
| rm_work | or that it is, and we broke project_id searches | 18:00 |
| johnsom | https://bugs.launchpad.net/octavia/+bug/1690481 | 18:00 |
| openstack | Launchpad bug 1690481 in octavia "Octavia v2 API needs to enforce RBAC (policy) rules" [High,New] | 18:00 |
| johnsom | Those two are still open | 18:00 |
| rm_work | whelp | 18:02 |
| *** kobis has joined #openstack-lbaas | 18:05 | |
| *** cody-somerville has joined #openstack-lbaas | 18:11 | |
| *** csomerville has quit IRC | 18:14 | |
| rm_work | that is really bad | 18:16 |
| rm_work | soo err | 18:16 |
| rm_work | i wonder if at least users can't delete other users' stuff without being admin? >_> | 18:16 |
| *** cody-somerville has quit IRC | 18:17 | |
| *** yamamoto has joined #openstack-lbaas | 18:17 | |
| *** cody-somerville has joined #openstack-lbaas | 18:18 | |
| *** cody-somerville has quit IRC | 18:18 | |
| *** cody-somerville has joined #openstack-lbaas | 18:18 | |
| *** blogan_ is now known as blogan | 18:18 | |
| *** csomerville has joined #openstack-lbaas | 18:20 | |
| *** kobis has quit IRC | 18:22 | |
| *** leitan has joined #openstack-lbaas | 18:23 | |
| *** yamamoto has quit IRC | 18:23 | |
| *** cody-somerville has quit IRC | 18:23 | |
| *** sshank has quit IRC | 18:29 | |
| *** leitan_ has joined #openstack-lbaas | 18:33 | |
| *** leitan has quit IRC | 18:33 | |
| *** cody-somerville has joined #openstack-lbaas | 18:33 | |
| *** csomerville has quit IRC | 18:35 | |
| *** catintheroof has joined #openstack-lbaas | 18:38 | |
| *** sanfern has quit IRC | 18:39 | |
| *** cpuga has quit IRC | 18:49 | |
| *** cpuga has joined #openstack-lbaas | 18:50 | |
| *** leitan_ has quit IRC | 18:52 | |
| *** leitan has joined #openstack-lbaas | 18:54 | |
| *** leitan has quit IRC | 18:59 | |
| johnsom | Alight lunch time then I will finish up reviewing OSC listener | 19:00 |
| johnsom | OSC stuff is looking pretty good so far, just a few things to address | 19:01 |
| *** catintheroof has quit IRC | 19:02 | |
| openstackgerrit | Carlos Puga proposed openstack/octavia master: Spec detailing Octavia service flavors support https://review.openstack.org/392485 | 19:11 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Fix HM flows to set Pool back to active https://review.openstack.org/466454 | 19:14 |
| rm_work | ^^ rebase | 19:14 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Floating IP Network Driver (spans L3s) https://review.openstack.org/435612 | 19:14 |
| *** yamamoto has joined #openstack-lbaas | 19:19 | |
| *** yamamoto has quit IRC | 19:24 | |
| *** cpuga has quit IRC | 19:31 | |
| *** blogan_ has joined #openstack-lbaas | 19:41 | |
| *** blogan has quit IRC | 19:42 | |
| *** blogan__ has joined #openstack-lbaas | 19:51 | |
| *** blogan_ has quit IRC | 19:52 | |
| *** blogan__ is now known as blogan | 19:55 | |
| *** JudeC has joined #openstack-lbaas | 20:03 | |
| *** harlowja has joined #openstack-lbaas | 20:04 | |
| *** rcernin has joined #openstack-lbaas | 20:12 | |
| *** blogan is now known as covfefe_ | 20:14 | |
| *** yamamoto has joined #openstack-lbaas | 20:20 | |
| *** cpuga has joined #openstack-lbaas | 20:22 | |
| *** covfefe_ is now known as blogan | 20:24 | |
| *** blogan is now known as stazis | 20:25 | |
| *** yamamoto has quit IRC | 20:26 | |
| *** stazis is now known as carlos_diablo_ga | 20:26 | |
| *** carlos_diablo_ga is now known as blogan | 20:27 | |
| *** blogan is now known as MAGAnaminous | 20:32 | |
| *** gcheresh has quit IRC | 20:35 | |
| *** MAGAnaminous is now known as donald-j-trump | 20:40 | |
| *** donald-j-trump is now known as blogan | 20:46 | |
| johnsom | rm_work https://youtu.be/SrDSqODtEFM | 20:56 |
| johnsom | +2 on OSC pool code? You didn't run it.... | 20:57 |
| johnsom | descriptor 'upper' requires a 'str' object but received a 'unicode' | 20:57 |
| *** sshank has joined #openstack-lbaas | 20:57 | |
| rm_work | johnsom: i did literally every object create/update/delete | 21:01 |
| rm_work | johnsom: but i just tested from the very end of the chain | 21:01 |
| rm_work | so maybe it was fixed later | 21:01 |
| rm_work | i didn't check them out each individually | 21:02 |
| johnsom | I am running from the end of chain | 21:02 |
| rm_work | err | 21:02 |
| rm_work | weird? | 21:02 |
| rm_work | it worked for me | 21:02 |
| rm_work | i was able to create fully working LBs | 21:02 |
| rm_work | with every object type | 21:02 |
| johnsom | It looks like lb is required though listener is an option | 21:02 |
| *** blogan_ has joined #openstack-lbaas | 21:07 | |
| *** blogan has quit IRC | 21:08 | |
| *** blogan_ is now known as blogan | 21:13 | |
| rm_work | on pools? | 21:17 |
| johnsom | Yeah | 21:17 |
| rm_work | yeah probably i made it using LB | 21:17 |
| rm_work | was that what caused the error? using listener without LB? | 21:17 |
| rm_work | i may not have tested that specifically | 21:17 |
| johnsom | I tried using both, haven't narrowed the trigger for the error yet | 21:17 |
| rm_work | hmm | 21:18 |
| rm_work | i didn't get any breaks :/ | 21:18 |
| johnsom | https://www.irccloud.com/pastebin/GwUsN1DV/ | 21:19 |
| johnsom | https://review.openstack.org/#/c/461992/8/octaviaclient/osc/v2/utils.py : 166 | 21:20 |
| *** yamamoto has joined #openstack-lbaas | 21:22 | |
| rm_work | err which python are oyu | 21:22 |
| rm_work | running with | 21:22 |
| rm_work | ah 2.7 derp | 21:22 |
| rm_work | it's only stamped all over that log :P | 21:23 |
| rm_work | i might be on 3.5 | 21:23 |
| johnsom | Grin | 21:23 |
| johnsom | Yeah, my 2.7 vm was available for testing... | 21:24 |
| rm_work | i can poke at it again on 2.7 for a final pass | 21:24 |
| rm_work | but it takes *a while* to do everything | 21:24 |
| rm_work | >_> | 21:24 |
| rm_work | @JudeC: let me know when you've finished addressing all of Michael's comments | 21:25 |
| johnsom | Yeah, I'm still reviewing listener, barely started pool just because I was trying to test listener | 21:25 |
| rm_work | there was a certain amount of "let's iterate" in my head too, to be fair | 21:25 |
| rm_work | see the commit at the end of the chain | 21:25 |
| rm_work | i basically told Jude just to do one sweep after all of those to make the strings and such consistent and line up/refactor some things | 21:26 |
| johnsom | Yeah. I just want to make sure we have a good start. It's not like much of what I am finding is huge work, just little stuff | 21:26 |
| johnsom | But important stuff. Like the column names, that is a "it doesn't work right" issue. Probably easy to fix. | 21:27 |
| rm_work | yeah | 21:27 |
| JudeC | Good :) I will get started on it soon. Still dealing with some production issues from some other stuff I am working on :( | 21:27 |
| rm_work | i'm busy doing kubernetes template rendering to make our octavia deployment work >_> | 21:27 |
| *** yamamoto has quit IRC | 21:27 | |
| johnsom | Hmmm | 21:27 |
| johnsom | Fun... | 21:27 |
| johnsom | Well, it gives me time to make progress on reviews | 21:28 |
| rm_work | yeah we left you a *pile* | 21:28 |
| johnsom | At least it wasn't steaming.... | 21:33 |
| JudeC | :P | 21:42 |
| *** blogan_ has joined #openstack-lbaas | 21:53 | |
| *** blogan has quit IRC | 21:54 | |
| *** blogan__ has joined #openstack-lbaas | 22:01 | |
| *** sshank has quit IRC | 22:02 | |
| *** blogan_ has quit IRC | 22:02 | |
| *** tongl has joined #openstack-lbaas | 22:11 | |
| tongl | hi, quick question. In LBaaS/Octativa when adding a member to a pool, does it require the pool member's subnet is the same as loadbalancer's subnet? | 22:14 |
| johnsom | tongl No. In most cases they are not the same | 22:14 |
| tongl | Great, thanks <johnsom> | 22:18 |
| *** yamamoto has joined #openstack-lbaas | 22:24 | |
| *** yamamoto has quit IRC | 22:29 | |
| tongl | In Octavia reference implementation, why do we need two neutron ports for loadbalancer? can we just use one port for one-arm loadbalancer? | 22:32 |
| johnsom | You can do a one-armed solution by using the same subnet for the VIP and the members. This includes if the members are routable from that subnet. However, note, if you do a port list via neutron you will see two ports for the VIP, but one of them is a "fake" port just for the allowed-address-pairs. It's not a "true" port and does not get bound to the instance. | 22:33 |
| *** ptoohill- has joined #openstack-lbaas | 22:34 | |
| *** rcernin has quit IRC | 22:39 | |
| *** cody-somerville has quit IRC | 22:40 | |
| *** ptoohill has quit IRC | 22:40 | |
| *** cody-somerville has joined #openstack-lbaas | 22:40 | |
| *** cody-somerville has quit IRC | 22:40 | |
| *** cody-somerville has joined #openstack-lbaas | 22:40 | |
| tongl | For example, if we create a loadbalancer on the private subnet with name lb1: neutron lbaas-loadbalancer-create --name lb1 private-subnet, loadbalancer has vip_address 10.0.0.10. Is the port with IP 10.0.0.10 "fake" port? | 22:41 |
| johnsom | tongl Yes, neutron port with prefix "loadbalancer-" if you do a neutron port show on it will have status "DOWN" and vif_type "unbound". The actual neutron port that is attached will be "ACTIVE" and a vif_type like "ovs". neutron just tracks secondary IPs on ports (allowed-address-pairs) by creating "fake" ports | 22:47 |
| johnsom | The real port will also have a "allowed_address_pairs" field that is populated with the fake port details | 22:47 |
| *** cpuga has quit IRC | 22:49 | |
| *** cpuga has joined #openstack-lbaas | 22:49 | |
| *** cpuga has quit IRC | 22:54 | |
| *** ianychoi has quit IRC | 22:59 | |
| rm_work | johnsom: this is super weird, when you add members they go into ERROR status immediately, then eventually to ONLINE via HM | 23:00 |
| rm_work | i thought they should just start in *OFFLINE* then go to ONLINE | 23:00 |
| rm_work | or like... NO_MONITOR | 23:01 |
| *** ianychoi has joined #openstack-lbaas | 23:01 | |
| johnsom | Well, oper_status is measured, so maybe the status reporting does error for no value or we pre-populate with error. | 23:02 |
| johnsom | I can look if you want | 23:02 |
| rm_work | hmm | 23:02 |
| rm_work | i'm about to look, just commented in case you knew something already | 23:03 |
| johnsom | No, not off my head | 23:04 |
| rm_work | looks like when you add one member, ALL members briefly enter ERROR status | 23:04 |
| rm_work | maybe something about how HAProxy reports? | 23:04 |
| johnsom | Now that is strange, but yeah, could be. We pull that status straight for HAProxy. The timing would have to be just right though..... | 23:05 |
| johnsom | So, the initial create state of the member is here: https://github.com/openstack/octavia/blob/master/octavia/db/prepare.py#L168 | 23:07 |
| johnsom | Yeah, my money is when we do the config reload, the status in haproxy reports something other than UP or doesn't list them on the listener | 23:15 |
| johnsom | sorry, listener->backend | 23:15 |
| johnsom | You could put debugging in here: https://github.com/openstack/octavia/blob/master/octavia/amphorae/backends/utils/haproxy_query.py#L123 | 23:16 |
| johnsom | To see what it does | 23:16 |
| johnsom | hot patch an amp agent | 23:16 |
| johnsom | JudeC FYI, I am done reviewing load balancer, listener, and pools now. Starting members. | 23:19 |
| JudeC | you are a beast | 23:21 |
| * johnsom Hopes said in the best way possible.... Ha | 23:23 | |
| *** yamamoto has joined #openstack-lbaas | 23:26 | |
| johnsom | This code is epic-ly cleaner than the old neutron client. This makes me happy | 23:28 |
| JudeC | Well that makes me feel really good haha | 23:29 |
| JudeC | yeah beast in a good way | 23:29 |
| *** yamamoto has quit IRC | 23:31 | |
| johnsom | I am pretty sure my work on the neutron client for LBaaSv2 was a form of LBaaS team hazing | 23:39 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!