| *** vivek-ebay has quit IRC | 00:04 | |
| *** vivek-ebay has joined #openstack-lbaas | 00:30 | |
| *** vivek-ebay has quit IRC | 00:52 | |
| *** mestery has quit IRC | 02:12 | |
| *** mestery has joined #openstack-lbaas | 02:14 | |
| *** woodster_ has joined #openstack-lbaas | 03:26 | |
| *** amotoki has joined #openstack-lbaas | 04:27 | |
| *** vivek-ebay has joined #openstack-lbaas | 05:47 | |
| *** woodster_ has quit IRC | 06:01 | |
| *** vivek-ebay has quit IRC | 06:37 | |
| *** ksamoray has joined #openstack-lbaas | 06:49 | |
| *** vivek-ebay has joined #openstack-lbaas | 08:08 | |
| *** vivek-ebay has quit IRC | 08:13 | |
| *** jschwarz has joined #openstack-lbaas | 08:22 | |
| *** ksamoray has quit IRC | 10:04 | |
| *** amotoki has quit IRC | 10:48 | |
| *** vivek-ebay has joined #openstack-lbaas | 10:58 | |
| *** vivek-ebay has quit IRC | 11:03 | |
| *** ksamoray has joined #openstack-lbaas | 11:12 | |
| *** mageshgv has joined #openstack-lbaas | 11:43 | |
| *** mageshgv has quit IRC | 11:44 | |
| *** ksamoray has quit IRC | 12:15 | |
| *** ksamoray has joined #openstack-lbaas | 12:17 | |
| *** mikedillion has joined #openstack-lbaas | 13:26 | |
| *** markmcclain has joined #openstack-lbaas | 13:37 | |
| *** ksamoray has quit IRC | 13:55 | |
| *** markmcclain has quit IRC | 14:06 | |
| *** markmcclain has joined #openstack-lbaas | 14:14 | |
| *** ajmiller has joined #openstack-lbaas | 14:32 | |
| *** markmcclain has quit IRC | 14:53 | |
| ptoohill | This: https://review.openstack.org/#/c/115548/ never made it to the feature branch. I think | 15:02 |
|---|---|---|
| ptoohill | Looks like ill be able to take care of that. | 15:17 |
| ptoohill | I do have something else to bring up though; The handling/creation/association of floating IPs. Anyone have any thoughts about this? | 15:19 |
| ptoohill | Will start a thread about this soon. | 15:19 |
| *** sbalukoff has quit IRC | 15:19 | |
| *** woodster_ has joined #openstack-lbaas | 15:40 | |
| *** mlavalle has joined #openstack-lbaas | 15:42 | |
| *** fnaval has joined #openstack-lbaas | 16:19 | |
| *** VijayB has joined #openstack-lbaas | 16:30 | |
| *** jschwarz has quit IRC | 16:35 | |
| *** crc32 has joined #openstack-lbaas | 16:36 | |
| *** sbfox has joined #openstack-lbaas | 16:50 | |
| *** crc32 has quit IRC | 17:06 | |
| *** vivek-ebay has joined #openstack-lbaas | 17:09 | |
| *** vivek-ebay has quit IRC | 17:16 | |
| *** Youcef has joined #openstack-lbaas | 17:29 | |
| TrevorV | blogan, you online? | 17:31 |
| ptoohill | Sent a thread out regarding FLIPs, please respond when you have time. | 17:41 |
| *** sbfox has quit IRC | 17:51 | |
| *** sbfox has joined #openstack-lbaas | 17:53 | |
| *** vivek-ebay has joined #openstack-lbaas | 17:57 | |
| *** sbalukoff has joined #openstack-lbaas | 18:02 | |
| *** mikedillion has quit IRC | 18:29 | |
| *** mikedillion has joined #openstack-lbaas | 18:30 | |
| rm_work | So there was a thread on OpenStack-dev ML about "not accessing keystone from neutron" | 18:32 |
| rm_work | trying to figure out what's up with that... I think they said "for anything but auth", so probably fine | 18:32 |
| rm_work | but I wonder if there is an accepted way to do that besides "include keystoneclient in my code and make a connection" | 18:33 |
| ptoohill | So, they don't want us to use keystone for anything but auth in neutron? | 18:36 |
| ptoohill | - _- | 18:37 |
| rm_work | i think that makes sense | 18:40 |
| ptoohill | Then yea, how would you get access to trusts in that case. | 18:40 |
| rm_work | i would qualify trusts as "auth" :P | 18:41 |
| ptoohill | lel, works for me ;) | 18:41 |
| *** crc32 has joined #openstack-lbaas | 18:52 | |
| *** crc32 has quit IRC | 18:52 | |
| *** vivek-ebay has quit IRC | 18:58 | |
| *** vivek-ebay has joined #openstack-lbaas | 18:58 | |
| *** sbfox has quit IRC | 19:01 | |
| TrevorV | ptoohill, https://github.com/stackforge/octavia | 19:25 |
| ptoohill | Thank you | 19:32 |
| *** VijayB has quit IRC | 19:39 | |
| *** vivek-ebay has quit IRC | 19:49 | |
| *** sbfox has joined #openstack-lbaas | 19:52 | |
| rm_work | dougwig: what do you think of barbican_utils.TLSInfo.get() ? | 19:57 |
| rm_work | dougwig: or rather possibly barbican_utils.TLSInfo.(de)register()? | 19:57 |
| dougwig | fine with me. i just hate redundancy in the namespace/name, given the 80 column limit. | 19:57 |
| rm_work | barbican_utils.Cerificate is a little weird | 19:57 |
| rm_work | because "Certificate" wouldn't imply "Private Key" | 19:57 |
| rm_work | which is included | 19:57 |
| rm_work | what do you think of "get( )" versus "get_and_register( )" versus "register( )" | 19:59 |
| rm_work | for the sake of clarity | 19:59 |
| rm_work | get() would be what someone might assume to look for, granted that they have a barbican container ID and need to "get" the data | 20:00 |
| rm_work | but they might not realize it's a registration action, and that it requires that they "deregister" when they're done with it (LB deletion, etc) | 20:00 |
| rm_work | versus "register" which is concise but doesn't necessarily make it clear that the return value will be the info the dev wants | 20:01 |
| rm_work | versus "get_and_register" which is kinda just... clear but quite verbose and also not the function name someone would assume immediately | 20:01 |
| rm_work | dougwig / sbalukoff | 20:01 |
| rm_work | naming problems >_< | 20:02 |
| dougwig | indeed, since get implies read-only, i might go with something like "def get(..., register=False)". keep the expected naming, make them do a smidge extra to involve a write. | 20:03 |
| rm_work | hrm | 20:04 |
| rm_work | well | 20:04 |
| rm_work | on one hand, I really wanted to make it *difficult* to NOT register | 20:04 |
| rm_work | but on the other hand... is it actually more harmful to have a bunch of false registrations because people weren't paying attention? | 20:05 |
| rm_work | then people learn to not take the system seriously | 20:05 |
| *** VijayB_ has joined #openstack-lbaas | 20:06 | |
| *** VijayB_ has quit IRC | 20:07 | |
| *** ajmiller has quit IRC | 20:22 | |
| sbalukoff | rm_work: I like to think of these things, also, in terms of RESTful semantics: A GET being read only can be repeated as many times as you like without causing damage (same with a PUT, in theory). In this case, a 'get_and_register" action sounds like it wouldn't be a call you'd want repeated, so I'd actually go for the 'register' name, even though it's not exactly standard. | 20:46 |
| sbalukoff | (And this would translate into a 'POST' that browsers know they aren't supposed to repeat unless the user specifically forces this.) | 20:49 |
| sbalukoff | Not that anyone is actually using a browser here to make these API calls, but the semantics translate over well. | 20:49 |
| *** vivek-ebay has joined #openstack-lbaas | 20:50 | |
| rm_work | sbalukoff: well, in this case the register action in Barbican is idempotent, so it doesn't really matter :P | 20:50 |
| *** vivek-ebay has quit IRC | 20:54 | |
| sbalukoff | rm_work: If its idempotent, then I don't have a problem with it using GET or PUT-like semantics. | 20:56 |
| ptoohill | [openstack-dev] [Neutron][LBaaS][Octavia] Floating IP management | 21:40 |
| ptoohill | Not sure if i should link the mail archive or not, but this is fairly important to us, so if you guys could chime in with any thoughts/input I would really appreciate it. | 21:41 |
| ptoohill | Also, don't think I should have used the Octavia tag, sorry about that. | 21:45 |
| *** VijayB_ has joined #openstack-lbaas | 21:47 | |
| *** vivek-ebay has joined #openstack-lbaas | 21:49 | |
| sbalukoff | ptoohill: Yeah, this is not trivial given how Neutron works. :P ... I really need to make the time to map out all our front-end topology options, as far as Neutron specifically is concerned. | 22:07 |
| rm_work | ptoohill: yeah the Octavia tag may mess up people's filtering | 22:14 |
| rm_work | hopefully still shows up to Neutron people | 22:14 |
| *** vivek-ebay has quit IRC | 22:52 | |
| *** vivek-ebay has joined #openstack-lbaas | 22:52 | |
| *** vivek-ebay has quit IRC | 22:53 | |
| *** vivek-ebay has joined #openstack-lbaas | 22:56 | |
| *** markmcclain has joined #openstack-lbaas | 22:58 | |
| *** mikedillion has quit IRC | 23:11 | |
| *** VijayB_ has quit IRC | 23:12 | |
| *** VijayB_ has joined #openstack-lbaas | 23:14 | |
| *** markmcclain has quit IRC | 23:32 | |
| *** markmcclain has joined #openstack-lbaas | 23:35 | |
| *** markmcclain1 has joined #openstack-lbaas | 23:35 | |
| rm_work | whelp | 23:36 |
| rm_work | first draft going up soon <_< | 23:37 |
| rm_work | (for https://review.openstack.org/#/c/123492/ and I guess technically it's like... the 20th draft, but whatever) | 23:37 |
| *** markmcclain has quit IRC | 23:39 | |
| rm_work | oh hmmmm originally Evgeny had this as a stub | 23:47 |
| rm_work | I am totally putting code in here... | 23:47 |
| rm_work | uhhh | 23:47 |
| rm_work | I guess I'll remove the line that says this is a stub >_> | 23:47 |
| rm_work | uhhh and yeah this is a rebase nightmare I am going to wait to submit until tomorrow when I can have blogan double-check that this is all right | 23:55 |
| rm_work | I'm apparently #6 in the chain | 23:55 |
| rm_work | funtimes <_< | 23:55 |
| *** sbfox has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!