| *** spatel has joined #openstack-lbaas | 00:21 | |
| *** Vorrtex has quit IRC | 00:36 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Allow multiple VIPs per LB https://review.opendev.org/660239 | 00:43 |
|---|---|---|
| *** rcernin has joined #openstack-lbaas | 00:45 | |
| *** rcernin has quit IRC | 00:45 | |
| *** rcernin has joined #openstack-lbaas | 00:45 | |
| *** spatel has quit IRC | 01:01 | |
| *** rcernin has quit IRC | 01:21 | |
| *** rcernin has joined #openstack-lbaas | 01:21 | |
| *** goldyfruit has joined #openstack-lbaas | 01:50 | |
| *** hongbin has joined #openstack-lbaas | 01:55 | |
| *** yamamoto has joined #openstack-lbaas | 02:50 | |
| *** yamamoto has quit IRC | 04:00 | |
| *** yamamoto has joined #openstack-lbaas | 04:06 | |
| *** ramishra has joined #openstack-lbaas | 04:11 | |
| *** hongbin has quit IRC | 04:11 | |
| *** schaney__ has quit IRC | 04:46 | |
| *** pcaruana has joined #openstack-lbaas | 04:50 | |
| *** gcheresh has joined #openstack-lbaas | 05:07 | |
| *** gcheresh has quit IRC | 05:19 | |
| *** gcheresh has joined #openstack-lbaas | 05:20 | |
| openstackgerrit | Merged openstack/octavia master: Create Amphora V2 provider driver https://review.opendev.org/659689 | 05:21 |
| *** luksky has joined #openstack-lbaas | 05:34 | |
| *** luksky has quit IRC | 05:53 | |
| *** ccamposr has joined #openstack-lbaas | 06:00 | |
| *** gthiemon1e is now known as gthiemonge | 06:31 | |
| openstackgerrit | Adit Sarfaty proposed openstack/neutron-lbaas stable/stein: Support URL query params in healthmonitor url_path https://review.opendev.org/660930 | 06:33 |
| *** tesseract has joined #openstack-lbaas | 07:12 | |
| *** rcernin has quit IRC | 07:14 | |
| *** rpittau|afk is now known as rpittau | 07:19 | |
| *** happyhemant has joined #openstack-lbaas | 08:19 | |
| *** luksky has joined #openstack-lbaas | 08:20 | |
| *** gcheresh has quit IRC | 08:35 | |
| *** gcheresh has joined #openstack-lbaas | 08:38 | |
| *** luksky has quit IRC | 09:28 | |
| *** ramishra has quit IRC | 09:30 | |
| *** luksky has joined #openstack-lbaas | 10:00 | |
| *** jlaffaye_ has quit IRC | 10:05 | |
| *** ramishra has joined #openstack-lbaas | 10:11 | |
| *** sapd1_x has joined #openstack-lbaas | 10:22 | |
| *** ivve has joined #openstack-lbaas | 10:27 | |
| *** vishalmanchanda has joined #openstack-lbaas | 10:28 | |
| *** yamamoto has quit IRC | 10:37 | |
| *** happyhemant has quit IRC | 10:38 | |
| *** yamamoto has joined #openstack-lbaas | 11:01 | |
| *** takamatsu has quit IRC | 11:01 | |
| *** takamatsu has joined #openstack-lbaas | 11:02 | |
| *** yamamoto has quit IRC | 11:06 | |
| *** yamamoto has joined #openstack-lbaas | 11:13 | |
| *** ramishra_ has joined #openstack-lbaas | 12:03 | |
| *** ramishra has quit IRC | 12:05 | |
| openstackgerrit | Merged openstack/octavia-lib master: Add project_id to all of the data model objects https://review.opendev.org/663434 | 12:06 |
| *** trident has quit IRC | 12:21 | |
| *** trident has joined #openstack-lbaas | 12:26 | |
| *** Emine has joined #openstack-lbaas | 12:32 | |
| *** yamamoto has quit IRC | 12:44 | |
| *** yamamoto has joined #openstack-lbaas | 12:51 | |
| *** pcaruana has quit IRC | 12:52 | |
| *** yamamoto has quit IRC | 13:00 | |
| *** lucashxu has joined #openstack-lbaas | 13:11 | |
| *** boden has joined #openstack-lbaas | 13:11 | |
| openstackgerrit | Ann Taraday proposed openstack/octavia master: [Jobboard] Importable flow functions https://review.opendev.org/659538 | 13:11 |
| openstackgerrit | Ann Taraday proposed openstack/octavia master: Use retry for AmphoraComputeConnectivityWait https://review.opendev.org/662791 | 13:15 |
| ataraday | johnsom, Hi! Could you please take a look onto https://review.opendev.org/#/c/662791 and https://review.opendev.org/659538 ? | 13:17 |
| openstackgerrit | Ann Taraday proposed openstack/octavia master: [WIP] Jobboard based controller https://review.opendev.org/647406 | 13:25 |
| *** vishalmanchanda has quit IRC | 13:28 | |
| *** mloza has joined #openstack-lbaas | 13:29 | |
| *** lucashxu has quit IRC | 13:34 | |
| *** yamamoto has joined #openstack-lbaas | 13:39 | |
| *** ramishra_ is now known as ramishra | 13:47 | |
| *** goldyfruit has quit IRC | 13:51 | |
| *** goldyfruit has joined #openstack-lbaas | 13:51 | |
| *** Vorrtex has joined #openstack-lbaas | 13:51 | |
| *** ricolin has joined #openstack-lbaas | 13:57 | |
| *** Vorrtex has quit IRC | 14:03 | |
| *** yamamoto has quit IRC | 14:10 | |
| *** gcheresh has quit IRC | 14:39 | |
| *** goldyfruit_ has joined #openstack-lbaas | 14:42 | |
| *** goldyfruit has quit IRC | 14:42 | |
| *** rpittau is now known as rpittau|brb | 14:45 | |
| *** yamamoto has joined #openstack-lbaas | 14:47 | |
| *** goldyfruit_ has quit IRC | 14:50 | |
| *** goldyfruit has joined #openstack-lbaas | 14:50 | |
| *** luksky has quit IRC | 14:53 | |
| boden | hi johnsom ... I left you a comment on https://review.opendev.org/#/c/663459. but if you want me to help resolve the failures I can take a shot at it; just let me know | 14:53 |
| *** luksky has joined #openstack-lbaas | 14:55 | |
| *** yamamoto has quit IRC | 14:55 | |
| johnsom | boden They are a known issue with our gates and pulling in octavia-lib. It's installing the pypi version and not the depends-on version. | 14:57 |
| boden | johnsom ah | 14:58 |
| johnsom | boden Oh, and derp, I failed to add the depends-on, so there is that.... | 14:58 |
| boden | johnsom just let me know if I can help with something | 14:58 |
| johnsom | boden I think rm_work and I were pushing to do an octavia-lib release soonish that would make this go away | 14:58 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Add project_id to all of the provider objects https://review.opendev.org/663459 | 14:59 |
| boden | ack | 14:59 |
| johnsom | boden If you pull those down local, you should be able to move forward with the nsx driver. | 14:59 |
| boden | johnsom yeah; hope to get to that today, or tomorrow | 15:00 |
| johnsom | boden I also wanted to see some API gate output to see if I did the documentation update correctly. I'm out of VMs to spin one up with tempest at the moment (too many parallel tasks at the moment). | 15:00 |
| johnsom | boden Ok cool, it should be fully functional as posted. I may just need to tweak the docs update. | 15:01 |
| boden | johnsom I also wanted to mention that I couldn't find https://docs.openstack.org/octavia/latest/contributor/guides/providers.html. from the main octiva docs... either its not added to index or maybe I'm just not seeing the link :) | 15:01 |
| *** sapd1_x has quit IRC | 15:02 | |
| johnsom | It is linked on the "contributors" page, Contributor Reference section, second from the top | 15:02 |
| *** rpittau|brb is now known as rpittau | 15:26 | |
| *** rpittau is now known as rpittau|afk | 15:46 | |
| *** ccamposr has quit IRC | 15:56 | |
| *** luksky has quit IRC | 16:00 | |
| *** tesseract has quit IRC | 16:03 | |
| *** Emine has quit IRC | 16:50 | |
| *** ricolin has quit IRC | 16:54 | |
| *** yamamoto has joined #openstack-lbaas | 18:53 | |
| *** yamamoto has quit IRC | 18:58 | |
| boden | heya johnsom. if you get another min to chat, pls let me know | 19:05 |
| johnsom | boden o/ | 19:06 |
| boden | johnsom so I was digging into if/why the vmware-nsx code really needed the populated "related objects".. for example why is a populated LB needed for a Listener create/update | 19:06 |
| boden | it turns out that it is used.. for example here https://opendev.org/x/vmware-nsx/src/branch/master/vmware_nsx/services/lbaas/nsx_v3/implementation/listener_mgr.py#L43 | 19:07 |
| johnsom | Ok, And you don't need it? grin | 19:07 |
| boden | is called for Listener, but as you can see it needs some details about the LB to realize the operation | 19:07 |
| boden | I guess I'm not sure how to proceed with these cases... what do you suggest? | 19:08 |
| boden | do we just use the octavia client or something? | 19:10 |
| johnsom | Hmmm, so you must not have a concept of a "VIP" or "load balancer", it's really tied around the listener. | 19:10 |
| johnsom | Yeah, that is one option, is use the client (openstacksdk really) to look up that information. You would need a keystone token though. Do you already have access to the right token? | 19:11 |
| boden | johnsom would be in the context no? | 19:11 |
| johnsom | The other thing we can do is add a query interface to the octavia-lib | 19:12 |
| johnsom | We don't pass a context to the drivers, so I'm not sure where that comes from. | 19:12 |
| boden | hmmm in that case I'm not sure | 19:13 |
| boden | would have to dig | 19:13 |
| johnsom | I'm not opposed to creating a "get" set of interfaces for the drivers | 19:13 |
| boden | johnsom you mean as getters on the ProviderDriver ?? | 19:14 |
| johnsom | Extending this: https://github.com/openstack/octavia-lib/blob/master/octavia_lib/api/drivers/driver_lib.py | 19:14 |
| johnsom | Such that it has methods for looking up the other objects | 19:15 |
| boden | johnsom got it... do you want me to take a stab at that? it'll probably take some help via code reviews since I haven't done much in this code before | 19:18 |
| johnsom | Hmm, so looking at the code, it looks like it is getting the context from pecan, so if that works, you probably do have everything you need to use the SDK option. | 19:21 |
| johnsom | https://github.com/openstack/vmware-nsx/blob/master/vmware_nsx/services/lbaas/octavia/octavia_driver.py#L382 | 19:22 |
| johnsom | Frankly that keeps the security cleaner overall. | 19:23 |
| johnsom | With that you can use the sdk get methods such as https://docs.openstack.org/openstacksdk/latest/user/proxies/load_balancer_v2.html#openstack.load_balancer.v2._proxy.Proxy.get_load_balancer | 19:24 |
| johnsom | There are examples of using the SDK here: https://github.com/openstack/octavia-dashboard/blob/master/octavia_dashboard/api/rest/lbaasv2.py | 19:25 |
| johnsom | This is what our dashboard plugin uses | 19:25 |
| boden | johnsom thank you for all that info... let me go off and do some hacking on it and see where I can get | 19:26 |
| johnsom | Ok, ping me if you have questions. This should work well actually. | 19:27 |
| boden | johnsom will do.. certainly appreciate your help here! | 19:27 |
| cgoncalves | lxkong, python-octaviaclient releases for queens, rocky and stein out of the door. enjoy :) | 19:44 |
| *** luksky has joined #openstack-lbaas | 19:49 | |
| *** henriqueof has joined #openstack-lbaas | 20:57 | |
| *** KeithMnemonic has joined #openstack-lbaas | 21:17 | |
| *** goldyfruit has quit IRC | 21:17 | |
| *** goldyfruit has joined #openstack-lbaas | 21:18 | |
| *** goldyfruit has quit IRC | 21:26 | |
| *** goldyfruit has joined #openstack-lbaas | 21:27 | |
| *** goldyfruit_ has joined #openstack-lbaas | 21:29 | |
| *** goldyfruit has quit IRC | 21:29 | |
| *** boden has quit IRC | 21:46 | |
| *** goldyfruit_ has quit IRC | 21:52 | |
| *** rouk has quit IRC | 21:54 | |
| *** goldyfruit has joined #openstack-lbaas | 21:58 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Limit cryptsetup key RAM usage https://review.opendev.org/663784 | 22:10 |
| lxkong | cgoncalves: thanks | 22:14 |
| *** goldyfruit has quit IRC | 22:19 | |
| *** gthiemonge has quit IRC | 22:22 | |
| *** gthiemonge has joined #openstack-lbaas | 22:22 | |
| *** rcernin has joined #openstack-lbaas | 22:23 | |
| *** rcernin has quit IRC | 22:23 | |
| *** rcernin has joined #openstack-lbaas | 22:24 | |
| *** rcernin has quit IRC | 22:26 | |
| *** rcernin has joined #openstack-lbaas | 22:28 | |
| *** goldyfruit has joined #openstack-lbaas | 22:38 | |
| *** yamamoto has joined #openstack-lbaas | 22:47 | |
| *** luksky has quit IRC | 22:51 | |
| *** goldyfruit has quit IRC | 23:02 | |
| *** gthiemonge has quit IRC | 23:20 | |
| *** gthiemonge has joined #openstack-lbaas | 23:20 | |
| *** goldyfruit has joined #openstack-lbaas | 23:34 | |
| rm_work | johnsom: is there any actual performance drawback to just blindly enabling `net.ipv4.ip_forward=1` and `net.ipv6.conf.all.forwarding=1` on all amps? | 23:41 |
| *** yamamoto has quit IRC | 23:41 | |
| rm_work | right now we do it based on the ... first vip's address type | 23:41 |
| *** yamamoto has joined #openstack-lbaas | 23:41 | |
| rm_work | but does it even matter? can't we even just do it in one of the elements? | 23:42 |
| rm_work | if there's no ipv4 or ipv6 vip it just won't matter, right? | 23:42 |
| rm_work | this seems overly complicated | 23:42 |
| rm_work | in fact, we do a modprobe on ip_vs and set `net.ipv4.vs.conntrack=1` blindly anyway | 23:43 |
| rm_work | I think I am going to move all of that to the amp element <_< | 23:43 |
| rm_work | we already do a ton of sysctl write values in 20-haproxy-tune-kernel | 23:45 |
| johnsom | We should not need forwarding | 23:47 |
| *** rcernin has quit IRC | 23:55 | |
| *** rcernin has joined #openstack-lbaas | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!