Thursday, 2018-04-05

*** yamamoto has joined #openstack-lbaas00:34
*** yamamoto has quit IRC00:39
*** kevinbenton_ has joined #openstack-lbaas00:46
*** kevinbenton has quit IRC00:51
*** sbalukoff has quit IRC00:51
*** ltomasbo has quit IRC00:51
*** sbalukoff has joined #openstack-lbaas00:58
*** voelzmo has joined #openstack-lbaas01:11
*** Swami has quit IRC01:15
*** yamamoto has joined #openstack-lbaas01:35
openstackgerritMerged openstack/python-octaviaclient master: Add release notes link to README
*** yamamoto has quit IRC01:41
*** harlowja has quit IRC01:42
openstackgerritJacky Hu proposed openstack/python-octaviaclient master: Update the outdated links
*** voelzmo has quit IRC01:45
*** yamamoto has joined #openstack-lbaas02:37
*** yamamoto has quit IRC02:43
openstackgerritMerged openstack/octavia master: Correctly validate member subnet_id in batches
*** fnaval has quit IRC03:08
*** fnaval has joined #openstack-lbaas03:16
*** harlowja has joined #openstack-lbaas03:37
*** yamamoto has joined #openstack-lbaas03:39
*** voelzmo has joined #openstack-lbaas03:42
*** yamamoto has quit IRC03:45
*** yamamoto has joined #openstack-lbaas03:52
*** voelzmo has quit IRC04:07
*** links has joined #openstack-lbaas04:23
openstackgerritNguyen Hai proposed openstack/python-octaviaclient master: Minor changes in docs
*** harlowja has quit IRC05:22
*** Swami has joined #openstack-lbaas06:07
SwamiI am seeing an Lbaas error 'Unknow Octavia Error' while trying to create a loadbalancer. I don't see any errors or warning in the octavia worker log, but the neutron server throws an error while the Octavia driver in 'neutron-lbaas' is trying to post a Request Body. See the error message in the link provided.
SwamiIs there way to figure out why it is sending a ServiceNotAvailable Exception06:13
*** pcaruana has joined #openstack-lbaas06:28
*** AlexeyAbashkin has joined #openstack-lbaas06:38
*** Swami has quit IRC06:44
*** velizarx has joined #openstack-lbaas07:01
*** imacdonn has quit IRC07:04
*** imacdonn has joined #openstack-lbaas07:05
*** tesseract has joined #openstack-lbaas07:15
*** velizarx has quit IRC07:18
*** rcernin has quit IRC07:30
*** velizarx has joined #openstack-lbaas07:57
*** logan- has quit IRC08:11
*** oanson has quit IRC08:12
*** irenab has quit IRC08:12
*** oanson has joined #openstack-lbaas08:13
*** logan- has joined #openstack-lbaas08:13
*** irenab has joined #openstack-lbaas08:14
*** xgerman_ has quit IRC08:25
*** xgerman_ has joined #openstack-lbaas08:25
*** ArchiFleKs has joined #openstack-lbaas08:30
ArchiFleKshi, I'm using octavia, and I often run into an error when deleting a LoadBalancer, it tries to delete the security group and failed : Attempt 1 to remove security group 850a146a-8216-4bca-b11e-b617129a3b22 failed and the LB never gets deleted and fall into error08:31
*** strigazi has quit IRC08:35
*** korean101 has joined #openstack-lbaas08:38
korean101HI guys08:38
korean101This issues was fixed? (
openstackLaunchpad bug 1583694 in neutron "[RFE] DVR support for Allowed_address_pair port that are bound to multiple ACTIVE VM ports" [Wishlist,Fix released] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)08:38
korean101when I use DVR, Can I set floating ip to<load_balancer_vip_port_id>?08:39
*** links has quit IRC08:40
korean101openstack floating ip set --port <load_balancer_vip_port_id> <floating_ip_id> (When I use DVR, Is this command possible?08:40
*** links has joined #openstack-lbaas08:50
*** salmankhan has joined #openstack-lbaas09:10
*** strigazi has joined #openstack-lbaas09:12
*** Alex_Staf has joined #openstack-lbaas09:13
*** velizarx has quit IRC09:14
*** velizarx has joined #openstack-lbaas09:15
*** velizarx has quit IRC09:27
*** velizarx has joined #openstack-lbaas09:29
*** salmankhan has quit IRC09:56
*** velizarx has quit IRC09:59
*** velizarx has joined #openstack-lbaas10:01
*** salmankhan has joined #openstack-lbaas10:14
*** Alex_Staf has quit IRC10:56
*** atoth has joined #openstack-lbaas11:49
*** samccann has joined #openstack-lbaas11:59
*** velizarx has quit IRC12:15
*** velizarx has joined #openstack-lbaas12:37
*** salmankhan has quit IRC12:38
*** links has quit IRC12:38
*** links has joined #openstack-lbaas12:39
*** salmankhan has joined #openstack-lbaas13:06
*** samccann has quit IRC13:10
*** salmankhan has quit IRC13:11
*** samccann has joined #openstack-lbaas13:11
*** Alex_Staf has joined #openstack-lbaas13:19
*** salmankhan has joined #openstack-lbaas13:21
*** links has quit IRC13:23
*** Alex_Staf has quit IRC13:39
*** fnaval has quit IRC13:50
*** fnaval has joined #openstack-lbaas13:51
*** fnaval has quit IRC13:55
*** voelzmo has joined #openstack-lbaas13:59
xgerman_ArchiFleKs: I am seeing the same thing in Pike14:00
*** fnaval has joined #openstack-lbaas14:02
*** KeithMnemonic has joined #openstack-lbaas14:15
*** Alex_Staf has joined #openstack-lbaas14:26
*** pcaruana has quit IRC14:41
*** salmankhan has quit IRC14:47
*** links has joined #openstack-lbaas14:54
*** salmankhan has joined #openstack-lbaas14:55
*** voelzmo has quit IRC14:56
*** salmankhan has quit IRC14:56
*** salmankhan has joined #openstack-lbaas14:56
*** Swami has joined #openstack-lbaas15:03
johnsomInteresting that both of those are OSA deployed15:03
johnsomSwami That looks like the URL the neutron-lbaas driver is connecting to the octavia API has a mis-configured load balancer.15:05
*** links has quit IRC15:06
*** voelzmo has joined #openstack-lbaas15:06
Swamijohnsom: let me check the configuration again.15:09
johnsomI'm digging into the code more too.  I just don't have that string in our code, so it is coming from somewhere else....15:09
johnsomYeah, if there is nothing in the octavia api logs, the only thing I can think is the call isn't getting to the API because something in between has a problem.15:14
johnsomkorean101 It is my understanding that was fixed in Pike, but I have also heard there is another bug:
openstackLaunchpad bug 1753434 in neutron "Unbound ports floating ip not working with address scopes in DVR HA " [Medium,Confirmed] - Assigned to Miguel Lavalle (minsel)15:16
Swamijohnsom: Ok I will check the octavia api logs and also check the driver configured for any mismatch.15:29
johnsomOk, sounds good15:30
*** openstackgerrit has quit IRC15:34
*** sapd has quit IRC15:36
*** sapd has joined #openstack-lbaas15:36
*** pcaruana has joined #openstack-lbaas15:47
*** Swami has quit IRC15:54
*** velizarx has quit IRC15:56
*** tesseract has quit IRC15:59
*** voelzmo has quit IRC16:02
*** AlexeyAbashkin has quit IRC16:10
*** pcaruana has quit IRC16:16
rm_workArchiFleKs: I have a patch up that is a big giant hammer solution to your problem16:40
rm_workif that merges, it would be useful for you... but i wish we could find a root cause for some of these16:41
johnsomWe have solved a bunch of these, it's just we didn't backport all of them and people are running old code.17:09
rm_workyeah, I wonder what version ArchiFleKs is on17:14
johnsomI know german's is pike17:15
johnsomI just cut a new pike stable today, so maybe that will help folks, we have backported some of these fixes17:16
*** salmankhan has quit IRC17:18
*** harlowja has joined #openstack-lbaas17:21
*** harlowja has quit IRC17:25
*** SumitNaiksatam has joined #openstack-lbaas17:34
*** AlexeyAbashkin has joined #openstack-lbaas17:34
*** AlexeyAbashkin has quit IRC17:38
cgoncalvesjohnsom: thanks!17:45
johnsomNP - Thanks for reminding me17:45
*** samccann has quit IRC17:57
xgerman_just in time18:04
*** pcaruana has joined #openstack-lbaas18:08
*** voelzmo has joined #openstack-lbaas18:13
*** sshank has joined #openstack-lbaas18:20
*** harlowja has joined #openstack-lbaas18:29
*** pcaruana has quit IRC18:47
*** openstackgerrit has joined #openstack-lbaas19:01
openstackgerritGerman Eichberger proposed openstack/neutron-lbaas master: Gate API test for the lbaasv2-proxy plugin
*** sshank has quit IRC19:04
*** voelzmo has quit IRC19:06
*** voelzmo has joined #openstack-lbaas19:41
*** yamamoto has quit IRC19:45
*** Alex_Staf has quit IRC20:25
*** velizarx has joined #openstack-lbaas20:31
*** Alex_Staf has joined #openstack-lbaas20:34
*** velizarx has quit IRC20:40
*** yamamoto has joined #openstack-lbaas20:46
*** sshank has joined #openstack-lbaas20:46
*** yamamoto has quit IRC20:52
*** Alex_Staf has quit IRC21:19
*** Swami has joined #openstack-lbaas21:28
Swamijohnsom: The driver configured is this service_provider = LOADBALANCERV2:Octavia:neutron_lbaas.drivers.octavia.driver.OctaviaDriver:default -- Seems ok to me.21:33
johnsomYes, that looks right21:33
SwamiAnd also this service_provider = LOADBALANCERV2:Haproxy:neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver21:33
johnsomYeah, that is the legacy (old) namespace driver.21:34
SwamiYes both are configured. Having both configured should not cause any problems, right?21:34
johnsommultiple providers is fine21:34
johnsomYeah, the user can select the provider at LB creation time21:34
johnsomSwami You can look at the gate jobs for example config files:
Swamijohnsom: ok will compare it.21:36
johnsomservice_provider = LOADBALANCERV2:Octavia:neutron_lbaas.drivers.octavia.driver.OctaviaDriver:default21:36
johnsomSwami based on your description this morning it seems like this setting:21:37
johnsombase_url =
johnsomin one of the neutron config files might be pointing to the wrong place, or a load balancer that is not configured right.21:38
SwamiYes I do see [octavia] base_url = http://xyz:9876 configured.21:38
johnsomThis should point to the IP/port of one of the octavia API processes21:38
johnsomYeah, take a look at that, something seems odd there.21:39
SwamiSure will check. Thanks21:39
johnsomIt should return:21:39
SwamiDoes the base_url need to have 'loadbalancer' key or it is optional.21:46
johnsomIt's up to how you deployed the API and the version of OpenStack you are running.  In Pike we had the WSGI goal that moved us from using ports to using a path.21:47
johnsomI'm pretty sure for you, the port is still in use21:47
Swamiwe do have IP/port pointing to the API process. This is what we have now. [octavia]21:48
Swamibase_url = https://qe007-cp1-vip-OCT-API-mgmt:987621:48
Swamirequest_poll_timeout = 300021:48
johnsomYeah, if you curl that you should get the version document similar to what I pasted21:48
*** yamamoto has joined #openstack-lbaas21:48
SwamiBut you mentioned in Pike we move to WSGI goad that moved from using ports to using path.21:48
SwamiWe are in Pike, so will IP/port still work21:49
johnsomEven with Pike you can still deploy with the old port method.21:49
johnsomYeah, it's just not default anymore21:49
SwamiThen I will check the mapping of the port to the service.21:49
*** rcernin has joined #openstack-lbaas21:49
*** yamamoto has quit IRC21:53
*** yamamoto has joined #openstack-lbaas22:49
*** yamamoto has quit IRC22:55
*** threestrands has joined #openstack-lbaas23:02
*** threestrands has quit IRC23:02
*** threestrands has joined #openstack-lbaas23:02
*** fnaval has quit IRC23:11
*** SumitNaiksatam has quit IRC23:26
openstackgerritAdam Harwell proposed openstack/octavia master: Add debug timing logging to amphora health update
rm_workjohnsom: so yeah, if you didn't notice, i have a bunch of random little changes i posted that are really tiny/simple and hopefully we can get merged :P23:46
johnsomI saw them, but have been distracted with other things, like trying to adapt the tempest test to your network....23:46
*** sshank has quit IRC23:47
*** yamamoto has joined #openstack-lbaas23:51
rm_workaugh i have 9 patches ready to merge (IMO) that I need to track T_T23:53
rm_worklol yes, and thanks for that :P23:53
rm_workthere are other providers with no tenant networks I am pretty sure, not JUST me :P23:53
johnsomBleh, well, after a day of futzing with this I blame you.... grin23:54
*** yamamoto has quit IRC23:57

Generated by 2.15.3 by Marius Gedminas - find it at!