Thursday, 2016-12-22

johnsomI am so curious why you need to add to the command line...00:01
*** ducttape_ has joined #openstack-lbaas00:01
rm_worknet.ifnames=0 biosdevnames=000:04
rm_workgotta do it :/00:04
rm_workalso didn't we do away with the ubuntu specific elements?00:04
rm_worki guess not? haproxy-octavia / haproxy-octavia-ubuntu00:04
johnsomNot all of them yet.00:04
rm_workseems like they exist for everything still00:04
johnsomThe directories are there for some just for backward compatibility (though I would probably just nuke them myself.00:05
johnsomMost are just elements-deps00:05
johnsomYou can't just find the interface you need by mac or something?  That is how we do it upstream.  As it stands now, the amp doesn't care what the interfaces are named00:07
rm_workit's cloud-init00:08
rm_workif cloud-init doesn't see an eth0, it freaks out00:08
rm_worki think it might be something with this cloud00:08
rm_workbut even stuff booted with octavia...00:08
johnsomYeah, might be worth investigating a bit more (happy to look at console logs) as cloud-init doesn't get a eth0 on ubuntu 16.x, it's ens*  The kernel renames it before cloud-init starts00:10
johnsomIf you really want to get into cloud-init's brains, you can also mount the config drive:00:12
johnsomblkid -t LABEL="config-2" -odevice00:12
johnsomwill give you the device00:12
johnsomjust mount it like a cdrom.00:12
rm_workright, this cloud is non-ubuntu00:13
johnsomHowever, I will warn you, it can take a while to fish through all the stuff that ends up in there00:13
rm_workand they disable the new-naming on all their images00:13
johnsomSo it should still be an eth*.00:14
rm_workwhich is why i think their cloud-init is set up that way00:14
johnsomYeah, I'm just saying, I don't think cloud-init should care00:14
rm_worki think they force their own cloud-init on every VM00:14
johnsomLike Rackspace does00:15
rm_workheh yeah00:15
rm_worki'll figure out a way around this00:15
rm_worki may have to write my own non-netns driver00:15
rm_workand amp-agent00:15
rm_workmaybe can just rip a lot of the code from the liberty versions00:16
rm_workthere's no custom networks here00:16
rm_workjust one main one00:16
johnsomAre they maybe not supporting config drive?  Maybe they are an EC2 shop only?00:16
johnsomI can't remember if I set it up to only accept config drive configs or not...00:17
rm_workconfig-drive def works00:18
rm_workthe agent config is in place and it starts up00:18
johnsomOh, ok00:18
rm_workjust for whatever reason, the cloud-init scripts that run DEMAND an eth000:19
rm_workor else no networking will come up00:19
johnsomMaybe it's just another one of those "we have an ancient version" issues like git00:20
*** amoralej|off has quit IRC00:28
*** ajo has quit IRC00:29
*** jschwarz has quit IRC00:29
*** fnaval has joined #openstack-lbaas00:39
*** ducttape_ has quit IRC00:40
*** diltram_ has joined #openstack-lbaas00:41
*** diltram has quit IRC00:45
*** fnaval_ has quit IRC00:45
*** reedip has quit IRC00:45
*** fnaval has quit IRC00:48
*** reedip has joined #openstack-lbaas00:55
*** ajo has joined #openstack-lbaas00:56
johnsomWell, tomorrow I get to decide which code at fault for the quota scenario test failure.  It think the quota code is getting run without a project ID...  I.e. chicken/egg problem00:56
*** ankur-gupta-f1 has quit IRC00:57
*** xgerman has quit IRC00:57
*** amoralej has joined #openstack-lbaas01:01
*** ankur-gupta-f1 has joined #openstack-lbaas01:07
*** xgerman has joined #openstack-lbaas01:07
openstackgerritMichael Johnson proposed openstack/octavia: Add quota support to Octavia
*** amoralej has quit IRC01:19
*** yuanying has joined #openstack-lbaas01:20
*** ajo has quit IRC01:20
*** ducttape_ has joined #openstack-lbaas01:27
*** ajo has joined #openstack-lbaas01:32
*** jschwarz has joined #openstack-lbaas01:32
*** diogogmt has joined #openstack-lbaas01:35
*** amoralej has joined #openstack-lbaas01:38
*** catintheroof has quit IRC01:42
*** bana_k has quit IRC02:43
*** fnaval has joined #openstack-lbaas02:47
*** yuanying has quit IRC02:49
*** yuanying has joined #openstack-lbaas02:50
*** yuanying has quit IRC02:54
*** ducttape_ has quit IRC03:05
*** ducttape_ has joined #openstack-lbaas03:19
*** ducttape_ has quit IRC03:29
*** amotoki has joined #openstack-lbaas03:36
*** links has joined #openstack-lbaas03:42
*** yamamoto_ has joined #openstack-lbaas03:46
*** yuanying has joined #openstack-lbaas03:50
*** yuanying has quit IRC03:50
*** yuanying_ has joined #openstack-lbaas03:50
*** diogogmt has quit IRC03:54
*** fnaval has quit IRC04:03
*** ducttape_ has joined #openstack-lbaas04:29
*** ducttape_ has quit IRC04:34
*** yamamoto_ has quit IRC04:36
*** yamamoto_ has joined #openstack-lbaas04:43
*** bana_k has joined #openstack-lbaas04:52
*** yamamoto_ has quit IRC04:53
*** yamamoto_ has joined #openstack-lbaas04:57
*** yamamoto_ has quit IRC04:57
*** bana_k has quit IRC05:30
*** ducttape_ has joined #openstack-lbaas05:31
*** ducttape_ has quit IRC05:35
*** yamamoto_ has joined #openstack-lbaas05:39
*** amotoki has quit IRC05:47
*** pcaruana has joined #openstack-lbaas06:17
*** bana_k has joined #openstack-lbaas06:23
*** anilvenkata has joined #openstack-lbaas06:30
*** armax has joined #openstack-lbaas06:42
*** openstackgerrit has quit IRC06:48
*** yatin_away is now known as yatin06:54
*** tesseract has joined #openstack-lbaas06:58
*** tesseract has quit IRC06:58
*** tesseract has joined #openstack-lbaas06:59
*** tesseract is now known as Guest4733706:59
*** amotoki has joined #openstack-lbaas07:08
*** jsheeren has joined #openstack-lbaas07:15
*** kobis has joined #openstack-lbaas07:16
*** openstackgerrit has joined #openstack-lbaas07:20
openstackgerritJingLiu proposed openstack/neutron-lbaas: Replace method vars().get to getattr
*** armax has quit IRC07:24
*** amotoki_ has joined #openstack-lbaas07:26
*** amotoki has quit IRC07:28
*** ducttape_ has joined #openstack-lbaas07:32
*** ducttape_ has quit IRC07:37
*** ihrachys has joined #openstack-lbaas07:47
*** openstackgerrit has quit IRC07:48
*** ihrachys has quit IRC08:07
*** amotoki_ has quit IRC08:30
*** ducttape_ has joined #openstack-lbaas08:33
*** ducttape_ has quit IRC08:38
*** bana_k has quit IRC08:46
*** amotoki has joined #openstack-lbaas09:02
*** amotoki has quit IRC09:15
*** nmagnezi has joined #openstack-lbaas09:28
*** openstackgerrit has joined #openstack-lbaas09:39
openstackgerritNir Magnezi proposed openstack/octavia: WIP - Fix the amphora-agent support for RH based Linux flavors
*** amotoki has joined #openstack-lbaas09:51
*** amotoki has quit IRC10:20
*** gongysh has joined #openstack-lbaas10:39
*** ihrachys has joined #openstack-lbaas10:57
*** ducttape_ has joined #openstack-lbaas11:05
*** ducttape_ has quit IRC11:26
*** ducttape_ has joined #openstack-lbaas11:35
*** yamamoto_ has quit IRC11:46
*** ducttape_ has quit IRC11:52
*** gongysh has quit IRC12:08
*** anilvenkata has quit IRC12:17
*** yamamoto has joined #openstack-lbaas12:37
*** ducttape_ has joined #openstack-lbaas12:52
*** ducttape_ has quit IRC12:57
*** links has quit IRC13:18
*** chlong has joined #openstack-lbaas13:23
*** amoralej is now known as amoralej|lunch13:28
*** kobis has quit IRC13:41
*** armax has joined #openstack-lbaas14:03
*** kobis has joined #openstack-lbaas14:12
*** yatin has quit IRC14:16
*** ducttape_ has joined #openstack-lbaas14:19
*** kobis has quit IRC14:26
*** yatin has joined #openstack-lbaas14:29
*** kobis has joined #openstack-lbaas14:29
*** ducttape_ has quit IRC14:32
*** amoralej|lunch is now known as amoralej14:38
*** m-greene_ has joined #openstack-lbaas14:39
*** m-greene has quit IRC14:40
*** m-greene_ is now known as m-greene14:40
*** kobis has quit IRC14:45
*** nmagnezi has quit IRC14:51
*** ducttape_ has joined #openstack-lbaas14:56
*** m-greene has quit IRC15:08
*** jsheeren has quit IRC15:10
*** ducttape_ has quit IRC15:19
*** yamamoto has quit IRC15:25
*** ducttape_ has joined #openstack-lbaas15:26
*** ducttape_ has quit IRC15:41
*** yamamoto has joined #openstack-lbaas15:48
*** yamamoto has quit IRC15:48
*** yamamoto has joined #openstack-lbaas15:48
*** madhu_ak has joined #openstack-lbaas15:54
*** Guest47337 has quit IRC16:03
*** kobis has joined #openstack-lbaas16:07
openstackgerritMichael Johnson proposed openstack/octavia: Add quota support to Octavia
*** ducttape_ has joined #openstack-lbaas16:15
openstackgerritMichael Johnson proposed openstack/octavia: Fix missing NovaServerGroupDelete
*** yamamoto has quit IRC16:20
*** bana_k has joined #openstack-lbaas16:24
*** kobis has quit IRC16:28
*** ducttape_ has quit IRC16:34
*** kobis has joined #openstack-lbaas16:34
johnsomxgerman FYI, I am also working on getting scenarios running local so I can clean up the scenario testing for quotas.16:49
johnsomSo, if you have an ah-ha moments, let me know.16:49
xgermanI think the v2 scenarios ran but gave an error when accessing Octavia V2…16:49
xgermannah, just tox -e16:50
johnsomYeah, I'm up to the point where it only runs 2 tests16:50
xgermanmmh, how does it decide what to skip16:50
xgermanI hardcoded Octavia to be present in my tests16:51
johnsomWell, I think I need a tempest.conf16:51
xgermanyou can add stuff programmatically + I ran that inside devstack16:51
*** fnaval has joined #openstack-lbaas16:52
*** kobis has quit IRC16:52
*** fnaval has quit IRC16:52
*** nmagnezi_ has joined #openstack-lbaas16:54
*** fnaval has joined #openstack-lbaas16:55
*** fnaval has quit IRC16:55
*** fnaval has joined #openstack-lbaas16:57
*** fnaval_ has joined #openstack-lbaas16:59
*** fnaval__ has joined #openstack-lbaas17:00
*** fnaval has quit IRC17:02
*** fnaval has joined #openstack-lbaas17:02
*** fnaval_ has quit IRC17:04
*** fnaval__ has quit IRC17:05
*** ducttape_ has joined #openstack-lbaas17:05
*** m-greene has joined #openstack-lbaas17:07
*** ducttape_ has quit IRC17:18
*** yamamoto has joined #openstack-lbaas17:20
*** mugsie_ has quit IRC17:22
*** yamamoto has quit IRC17:30
*** mugsie_ has joined #openstack-lbaas17:49
*** ihrachys has quit IRC17:56
*** kobis has joined #openstack-lbaas18:02
*** m-greene has quit IRC18:04
*** m-greene has joined #openstack-lbaas18:09
*** ducttape_ has joined #openstack-lbaas18:18
*** ducttape_ has quit IRC18:23
*** Matias is now known as Urahara18:24
*** kobis has quit IRC18:37
*** kobis has joined #openstack-lbaas18:45
*** kobis has quit IRC18:52
*** kobis has joined #openstack-lbaas18:55
*** amoralej is now known as amoralej|off19:01
openstackgerritOpenStack Proposal Bot proposed openstack/neutron-lbaas: Updated from global requirements
*** ducttape_ has joined #openstack-lbaas19:06
openstackgerritOpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements
*** kobis has quit IRC19:07
*** ducttape_ has quit IRC19:23
*** nmagnezi_ has quit IRC19:30
*** yatin has quit IRC19:38
*** rcernin has joined #openstack-lbaas19:38
*** yatin has joined #openstack-lbaas19:50
johnsomxgerman FYI, that global requirements has a major version jump in it, so not what I would call low-impact....19:52
xgermanoh, taskflow19:53
xgermanneed to look more closely next time19:53
rm_workugh maybe the octavia scenarios would be useful for me ...20:17
rm_workwhat problems were you having johnsom?20:17
*** ducttape_ has joined #openstack-lbaas20:24
openstackgerritGerman Eichberger proposed openstack/octavia: Add tempesttest for the N-LBaaS API
xgerman^^ hours of pep8 appeasing…20:27
rm_workwonder if n-lbaas disabled some hacking rules that we didn't20:29
xgermanvery likely20:29
*** ducttape_ has quit IRC20:29
rm_workwait what even IS a Nova Server Group?20:30
rm_workwe can delete it before we delete the VM??20:30
xgermanit controls anti-affinity20:31
johnsomrm_work I am working on the quotas patch.  Mostly trying to figure out the right was to test under noauth since we need project-id for quotas.  Right now we have hacked it into the requests, but that isn't the right answer20:34
rm_workhmm, dunno20:38
rm_worki mean20:38
rm_workif that works... :/20:39
xgermanif we work under noauth we can’t really enforce quotas. Wouldn’t hose be enforced by n-lbaas instead?20:41
johnsomWe are nlbaas now..  that is the point.  I just think setting up a noauth "project" would make our tests better for both quota and policy20:45
johnsomDo it once and not on each test20:45
xgermanwe couldn’t find another letter - n always reminds me of neutron20:47
johnsomrm_work BTW, on your question about NovaServerGroupDelete, this was just missing from the cascade delete flow where it is present in the main delete flow.  Probably a refactor in order on the cascade delete flow so we don't have this kind of issue, but low priority....20:51
johnsomxgerman I was not coining a new term "nlbaas", it still means neutron-lbaas to me, but what I was trying to say is with the merge we have to deal with our own quotas going forward.20:52
xgermanwe need to enforce our own quotas… and if they come though neutrpn lbaas thsi will handle it20:53
johnsomLooking at other projects they all seem to handle this differently.  One going as far as to create a noauth/test keystone middleware replacement for testing.20:53
johnsomWell, I expect for the pass through proxy with neutron lbaas we would remove the quota check for our calls in neutron.  We don't want to be worried with two databases with quota info20:55
xgermanisn’t quota a neutron function and not lbaas?>20:56
johnsomI just hit the problem that now more scenario tests are running quota check code, so we really need to get the project_id stuff figured out in a reasonable way for the tests.20:56
johnsomYes, that is why I said we would remove them in neutron.20:56
johnsomI.e pass them through as if the objects had no quota.20:57
johnsomThe fun part is going to be the quota setting calls in neutron.  Probably some custom code for that part.20:57
xgermanyeah, also they will have policies is neutron20:58
xgerman‘so I would just make the internal endpoint quota free and the external one quota20:58
xgermanalso did we model if somebody creates LB’s with Neutron LBaaS and then goes to Octavia with the new API…20:59
xgermannot that he gets twice the loadbalancers that way21:00
johnsomSince neutron-lbaas just passes the calls straight through, it's all one quota and all one code, so works the same21:00
xgermanso if somebody sets an LBaaS quota in neutron we carry that over? and if they are different last-one wins?21:01
xgermanor we just ignore neutron and make people configure octavia21:01
johnsomThere is no octavia driver when using the pass through proxy21:01
johnsompass through proxy is API request to neutron gets passed straight through to octavia API.21:02
xgermanyep, but quotas are configured in neutron.conf?21:02
xgermanso we ignore those?21:03
xgermanand the admin needs to establish them in Ocatvia21:03
xgermansame with policy files21:03
johnsomYeah, in the future the quota for LBaaS objects in neutron-conf would be -121:03
xgermanok, got it21:04
rm_workhmm so if i go through neutron-lbaas it creates a VIP for me... going direct to octavia, it seems we don't do that? vip is *required*?21:08
johnsomIf one isn't passed in, it should create it21:09
johnsomrm_work Are you making a single call create?21:10
rm_workalso, VIP is definitely mandatory :/21:11
johnsomAh, I was confusing.  I was thinking port21:13
johnsomGoing through neutron-lbaas, but default, creates the VIP port for us and passes in a port ID for the vip.21:14
xgermanI thought we can create our own VIP in Octavia… blogan worked on some flows21:15
johnsomYeah, he worked on it to work as I described above.  Before we always created the VIP ports ourselves21:15
xgermanbut we still do that if none is submitted?21:16
johnsomRight if the v1 api just gets a subnet, it will create the port21:16
rm_worksubnetID is ...21:18
rm_workthe subnet of the FIP's IP?21:18
johnsomYeah, subnet ID you want your VIP on21:18
xgermanbut not FIP21:18
xgermanFIP is on the router forwarding to VIP21:18
xgermanaka in some net which is connected to the router21:19
johnsomYeah, none of this manages FIPs.  (FIP== the nat IP)21:19
bloganxgerman: johnsom answered correctly21:21
xgermanI would have expected nothing less from him ;-)21:21
johnsomI just had to go deep on that stuff to fix the failover stuff21:24
*** ducttape_ has joined #openstack-lbaas21:25
*** ducttape_ has quit IRC21:30
rm_workurg wait so21:34
rm_workI create a FIP21:34
rm_workit's on the FIP network21:34
rm_workit has an IP21:34
xgermanand then you associate that FIP with the VIP21:35
rm_workwhat is a VIP21:35
rm_workI thought the FIP was the VIP21:35
rm_workhold on21:35
xgermanwell, we spin the LB up in the mgmt net and the user’s subnet21:36
rm_workuhh so21:36
xgermanit will need a VIP in the user’s subnet21:36
rm_workto create a LB21:36
xgermanand then if the subnet isn’t connected to the Internet21:36
xgermanyou need some FIP on the ext-net which get associated with the VIP21:36
rm_worki'm following that (stephen's thing)21:36
rm_workfirst command... i need to pass in an IP and a subnet21:37
rm_workwhat are those from?21:37
xgermanthi is the VIP21:37
rm_workok, what type of object is the VIP?21:37
xgermanit’s just an IP21:37
rm_workbut I don't have an IP anywhere yet21:37
xgermanwell, he is passing on IP, port and net21:38
rm_workwhat command do I run to create an IP?21:38
rm_workwhat type of openstack object is it21:38
xgermanso I think if you only pass in subnet we create a port21:38
rm_worki can't just pick an IP address out of thin air21:38
rm_workit has to be assigned to me somehow21:38
rm_workis it a port?21:39
rm_workdo I just need to create a neutron port on some subnet?21:39
xgermanthat might be sufficeint - neutron assigns ips often via dhcp21:40
xgermanbut I always had Octavia deal with that ;-)21:40
rm_workyeah but ... i can't, right?21:40
rm_workthis is mandatory to create the highest level object21:40
rm_workI have to pass in at LEAST "ip_address" and "subnet" for VIP21:40
xgermansubnet should be all you need21:40
rm_workit requires more21:42
rm_workif i just provide subnet it flips out, AFAICT21:42
rm_workok yeah so creating a port and passing that info seems right21:45
xgermanwe have code to create a port AFAIK21:45
rm_workthen ...21:47
rm_workwhy is VIP a required attribute on create of a LB21:47
xgermanwell, that code won’t work unless you pass aport in… so umh21:48
xgermanlikely needs rafactor21:49
rm_workyeah this seems un-like the neutron API :P21:49
johnsomHang on here.21:49
johnsomSo, you need either a port or a subnet.  IP is (should be) optional.21:49
rm_workwhen i didn't provide one, it got an exception and bailed21:49
rm_worki can try to reproduce really quick21:50
johnsomIf you pass in a port_id, we use it for the VIP.  If  not we create one for you21:50
xgermanjohnsom you need both21:50
xgermanthere seems to be a bug in the first of21:50
johnsomThat is odd, we shouldn't need the IP.21:50
xgermanwe require a port-id but the message says it shoudl be eitehr - or21:50
xgermanand the code looks like it, too21:50
rm_workah that's right21:51
rm_workno it is21:51
rm_workit's not A and not B21:51
rm_workmeaning, one has to exist21:51
johnsomThat logic is correct21:51
rm_worki read it wrong first but second reading it made sense21:51
johnsomRight, port_id or subnet_id are required21:51
xgermanoh, ok21:51
xgermanthen all you would need is a subnet21:51
johnsomBut I don't think we require an IP specified.  We can allocate that from the subnet if one isn't given, so if that fails it is a bug21:52
xgermanwe shouldn;’t even require a port21:52
johnsomWe don't require a port21:52
xgermanrm_work says we do21:52
johnsomWe require a port_id OR a subnet.21:52
rm_workoh hmm... one sec, might have worked that time without the ip21:52
rm_worki meant it was requiring subnet and IP21:53
rm_worktrying to repro21:53
xgermanthe way it should work is that you cna pass in a subnet and it should be happy21:53
rm_workyeah maybe it is working now... maybe something was wrong with the subnet i was trying befor21:55
xgermandon’t scare us like that :-)_21:56
johnsomHa, I was going to try it, but I hit my quota code...  so, yeah.21:56
xgermanat least something works21:57
rm_work2016-12-22 14:56:59.797 4238 ERROR oslo_messaging.rpc.server     return data_models.Vip(ip_address=fixed_ip.ip_address,21:57
rm_work2016-12-22 14:56:59.797 4238 ERROR oslo_messaging.rpc.server AttributeError: 'NoneType' object has no attribute 'ip_address'21:57
rm_worki think that's maybe because the subnet could be out of IPs?21:57
* rm_work guesses wildly21:57
rm_workso it might just be a bad subnet21:57
rm_workyeah looks like.22:00
rm_workhmm, trying to use the management subnet as the vip subnet and it breaks network connectivity when it plugs the VIP, it seems22:01
rm_worki know that's not ideal but i don't have access to multiple working subnets :/22:01
johnsomYeah, I just tested, it works without an IP22:02
xgermanToday’s reading:
johnsomInteresting.  Oslo added a consistent hash library recently too.  I think it is very basic though22:05
johnsomYeah, 3-4 projects are all using some form of consistent hashing22:07
xgermanthis is awesome. This algo is in haproxy 1.722:08
xgermannow we need to add that to the pool ;-)22:11
rm_workugh ssl handshake failure between amp and controller WHY22:12
*** Kiall has joined #openstack-lbaas22:13
johnsomrm_work you need to sacrifice something to OpenStack22:14
rm_workah because i'm dumb, that's why22:14
rm_workmixed config sources22:15
xgermanI guess he already sacrificed sanity :-)22:15
rm_workso close...22:15
rm_workOH SHIT22:17
rm_workit went active :322:17
* rm_work is feeling super ridiculous after struggling with getting everything working for two weeks22:17
*** madhu_ak has quit IRC22:24
*** chlong has quit IRC22:34
*** chlong has joined #openstack-lbaas22:41
openstackgerritMichael Johnson proposed openstack/octavia: Add quota support to Octavia
*** ducttape_ has joined #openstack-lbaas22:54
*** m-greene has quit IRC22:59
*** Urahara is now known as Matias23:15
rm_workwhy is "weight" required on members?23:22
rm_workshouldn't it just default to "1" or something?23:22
rm_workit doesn't even make sense for ROUND_ROBIT23:22
rm_workoh wait, did we make ROUND_ROBIN and WEIGHTED_ROUND_ROBIN separate or is it just always weighted23:23
*** ducttape_ has quit IRC23:25
johnsomweight 0 means it doesn't get new connections23:29
johnsomweight is not required, it defaults to 123:30
johnsomFYI, the gates are broken to some degree.  OSC client errors doing keystone things.23:31
johnsomA bunch of projects are impacted23:32
rm_workjohnsom: according to that, weight is required23:37
rm_workis it just wrong in the docs?23:37
johnsomrm_work Yes, I think so:
rm_workyeah, seems so, was able to create one without specifying23:38
rm_worki thought the docs were generated23:38
rm_workah also subnet ISN'T required23:39
rm_workis that a deviation from n-lbaas?23:39
johnsomThat is interesting.  I but it pukes without it though23:39
rm_workhmm, no, n-lbaas api doc says subnet_id isn't required... wtf23:40
rm_workit's a client bug23:40
rm_workneutron lbaas-member-create requires subnet23:40
rm_workbut the API doesn't actually23:40
rm_workthat explains my confusion23:41
johnsomI think it should be required in the api23:41
johnsomNot sure, would have to dig through the code.23:42
rm_worki kept trying to USE it, finding it required... being confused... then later looking it up and thinking "i thought it WAS required, must have been mistaken"23:42
rm_workand then trying again and it's required in the client23:42
rm_workit shouldn't be required IMO23:42
rm_workif you know you can get out via the VIP subnet23:42
rm_workyou shouldn't need to specify one for the member23:42
rm_workespecially given the way this is set up in some clouds, it'd be very confusing23:43
rm_workas we have a bunch of subnets connected via routing, but not directly pluggable23:43
rm_workso like, my member might very well be on abcd-efg subnet, but i'd need to have the amp bind on subnet qwer-tyuip23:44
rm_workand there's no way i'd know that as an end user23:44
rm_workin fact, i'm probably going to have to figure out a way to have octavia select a default vip subnet, and make it not require a VIP object23:44
rm_workbecause the user won't have any way of knowing what networks they can get a VIP on23:45
rm_workeverything is just routed together23:45
rm_workugh, network setups always throw a wrench in everything because everyone does it totally differently in their clouds23:45
rm_workjohnsom: unless, do we do anything to the routes that would prevent traffic flow from going out the VIP interface to reach a member?23:49
johnsomrm_work No, I don't think we do anymore23:51
rm_workthen subnet_id should absolutely be optional23:51
johnsomI get your point23:51
rm_workand I need to figure out how to make this VIP thing work23:52
*** ducttape_ has joined #openstack-lbaas23:59

Generated by 2.14.0 by Marius Gedminas - find it at!