Thursday, 2017-05-11

« Wednesday, 2017-05-10 Index Friday, 2017-05-12 »
*** chlong has quit IRC00:08
*** cpuga has joined #openstack-lbaas00:12
*** cpuga_ has joined #openstack-lbaas00:22
*** cpuga has quit IRC00:24
*** cpuga_ has quit IRC02:02
*** cpuga has joined #openstack-lbaas02:02
*** JudeC has joined #openstack-lbaas02:13
*** yamamoto has joined #openstack-lbaas02:29
*** catintheroof has joined #openstack-lbaas02:43
rm_workjohnsom: is that good to go now?03:07
rm_workI can review03:07
johnsomYeah, I think so03:07
rm_workkk03:12
*** cody-somerville has joined #openstack-lbaas03:18
*** cody-somerville has quit IRC03:18
*** cody-somerville has joined #openstack-lbaas03:18
*** catintheroof has quit IRC03:35
*** fnaval has joined #openstack-lbaas04:20
*** fnaval has quit IRC04:21
*** fnaval has joined #openstack-lbaas04:22
openstackgerritElancheran S proposed openstack/python-octaviaclient master: Add member commands to client  https://review.openstack.org/46303504:23
openstackgerritMerged openstack/octavia master: Reintroducing local certificate manager, as it is useful for testing  https://review.openstack.org/46361504:30
openstackgerritMerged openstack/octavia master: Trivial fix typos in documents  https://review.openstack.org/46343004:45
*** blogan has quit IRC04:49
*** blogan has joined #openstack-lbaas04:50
*** pcaruana has joined #openstack-lbaas04:52
*** pcaruana has quit IRC05:03
*** cody-somerville has quit IRC05:11
*** csomerville has joined #openstack-lbaas05:12
openstackgerritJude Cross proposed openstack/python-octaviaclient master: Add member commands to client  https://review.openstack.org/46303505:13
openstackgerritDong Jun proposed openstack/octavia master: Remove invalid url link  https://review.openstack.org/46390705:19
*** fnaval has quit IRC05:33
*** JudeC has quit IRC05:42
*** gcheresh has joined #openstack-lbaas05:52
*** belharar has joined #openstack-lbaas06:03
*** cpuga has quit IRC06:22
*** voelzmo has joined #openstack-lbaas06:27
*** rcernin has joined #openstack-lbaas06:33
*** pcaruana has joined #openstack-lbaas06:47
*** tesseract has joined #openstack-lbaas07:12
*** aojea has joined #openstack-lbaas07:19
*** todin has joined #openstack-lbaas08:43
todinhi, where could I find an installation how-to for octavia?08:44
todinthere must be more than https://docs.openstack.org/mitaka/networking-guide/config-lbaas.html08:44
*** yamamoto has quit IRC10:05
*** yamamoto has joined #openstack-lbaas10:32
openstackgerritNir Magnezi proposed openstack/octavia master: Switch to use stable remote_client  https://review.openstack.org/44500310:53
nmagnezi_todin, are you deploying manually?10:56
todinnmagnezi_: the first step is to depoly it manually, to understand it, but then I want to use puppet.10:57
nmagnezi_todin, if it is for leaning proposes, i'd suggest you look at the octavia devstack plugin to understand what it does10:58
nmagnezi_learning*10:58
nmagnezi_rm_work, somehow i missed 'Octavia - Load Balancing for OpenStack' and looks like it contains a lot of info.. :-)11:02
nmagnezi_rm_work, docs for loadbalancers migration to Octavia would be fantastic to have.. seriously.11:02
nmagnezi_:)11:02
todinnmagnezi_: I am already playing with the devstack plugin, but I thought a tutrial would be simpler as reversenginiering of the config11:06
nmagnezi_todin, I worked with beagles (tripleO core) to add a support for Octavia deployment there. beagles submitted multiple patches against it. some of the steps are documentd here https://review.openstack.org/#/c/447496/11:10
nmagnezi_todin, I don't know about any official Octavia doc for a manual step by step installation, but maybe one of the cores or the PTL would11:11
todinnmagnezi_: okay, thanks, the cores and PTL are in this channel?11:16
*** yamamoto has quit IRC11:29
nmagnezi_todin, yes https://review.openstack.org/#/admin/groups/370,members11:34
*** kong has quit IRC11:53
*** kong has joined #openstack-lbaas11:53
*** yamamoto has joined #openstack-lbaas12:29
*** chlong has joined #openstack-lbaas12:33
*** yamamoto has quit IRC12:39
*** catintheroof has joined #openstack-lbaas12:42
*** yamamoto has joined #openstack-lbaas12:45
*** yamamoto has quit IRC12:47
*** yamamoto has joined #openstack-lbaas13:04
*** gcheresh has quit IRC13:20
*** amotoki has joined #openstack-lbaas13:36
*** catintheroof has quit IRC13:39
*** belharar has quit IRC13:44
*** amotoki has quit IRC13:49
*** blogan_ has joined #openstack-lbaas13:54
*** blogan has quit IRC13:55
*** amotoki_away is now known as amotoki13:58
*** cpuga has joined #openstack-lbaas13:59
*** fnaval has joined #openstack-lbaas13:59
*** cpuga has quit IRC14:00
*** cpuga has joined #openstack-lbaas14:01
johnsomtodin: Take a look at the quickstart guide in or docs.  https://doc.openstack.org/developer/octavia14:12
johnsomAlso: https://github.com/openstack/puppet-octavia14:14
*** amotoki_ has joined #openstack-lbaas14:21
nmagnezi_johnsom, morning :)14:24
johnsomMorning14:26
johnsomI think I should put some of these links on our wiki...  It is getting hard to keep track of them all14:26
*** amotoki__ has joined #openstack-lbaas14:29
*** amotoki_ has quit IRC14:31
*** amotoki__ has quit IRC14:34
*** amotoki_ has joined #openstack-lbaas14:41
*** amotoki_ has quit IRC14:41
*** amotoki_ has joined #openstack-lbaas14:47
*** amotoki_ has quit IRC14:47
*** amotoki_ has joined #openstack-lbaas14:51
*** amotoki_ has quit IRC14:51
*** amotoki_ has joined #openstack-lbaas14:54
*** amotoki__ has joined #openstack-lbaas14:55
*** amotoki__ has quit IRC14:55
*** fnaval has quit IRC14:58
*** amotoki_ has quit IRC14:58
*** amotoki__ has joined #openstack-lbaas14:58
xgermanWe are in the hall of shame for the wsgi thing15:06
xgermanI thought we did that?!15:07
*** voelzmo has quit IRC15:08
*** fnaval has joined #openstack-lbaas15:09
johnsomWe did do it, then they changed the goal.  Why are we in the wall of shame????15:13
johnsomIt's not due yet15:13
*** pcaruana has quit IRC15:18
xgermanThey keep a rolling total15:19
*** rcernin has quit IRC15:19
xgermanNow, we have the microbe15:20
xgermanMicro version debate again15:20
johnsomhttps://governance.openstack.org/tc/goals/pike/deploy-api-in-wsgi.html#octavia15:21
johnsomYeah, microversions.  A mess IMO15:22
johnsomI think they need to slow down on the goals until they have a firm grasp on the amount of work, the need, and the right way to implement so we don't end up having to change the criteria mid-way through the release.15:24
johnsomI bet half of the "completion" criteria on that page is for the OLD criteria15:25
xgermanOh, they are making new goals...15:27
johnsomI know, I'm just saying, I think they need to slow it down.  Especially after the mess that happened in Pike15:28
xgermanhttps://usercontent.irccloud-cdn.com/file/jqipM7XX/IMG_2364.JPG15:28
xgermanHere are there new ideas15:28
johnsomYeah, I have commented on that etherpad.15:29
*** amotoki_ has joined #openstack-lbaas15:30
xgermanGr8t - those days I am just disappointing that nobody invites me to parties...15:30
xgermanAnd then litters my twitter feed with food and drinks15:30
johnsomOuch15:30
johnsomI didn't get any invites this time either15:31
*** amotoki__ has quit IRC15:32
xgermanYeah. There was an RDO party (looking at Nir) and a Suse party15:32
johnsomHa15:32
johnsomThey could have at least invited you to Suse15:32
xgermanYep.15:32
xgermanVery disappointed15:32
xgermanAuthentication goal didn't make it15:39
openstackgerritMichael Johnson proposed openstack/octavia master: Update devstack to properly configure nova-lxd img  https://review.openstack.org/45323715:40
*** amotoki_ has quit IRC15:41
*** blogan__ has joined #openstack-lbaas15:46
*** amotoki_ has joined #openstack-lbaas15:47
*** blogan_ has quit IRC15:48
*** catintheroof has joined #openstack-lbaas15:54
*** aojea has quit IRC15:58
*** blogan__ is now known as blogan16:07
*** yamamoto has quit IRC16:13
*** yamamoto has joined #openstack-lbaas16:14
*** catintheroof has quit IRC16:17
*** amotoki_ has quit IRC16:20
*** aojea has joined #openstack-lbaas16:28
*** fnaval has quit IRC16:30
*** fnaval has joined #openstack-lbaas16:40
*** amotoki_ has joined #openstack-lbaas16:49
*** aojea has quit IRC16:57
openstackgerritMichael Johnson proposed openstack/octavia master: Add v2 pool API section  https://review.openstack.org/45827216:57
johnsomCaught a typo with the id column16:58
johnsomJoy, launchpad is down17:01
*** catintheroof has joined #openstack-lbaas17:06
*** tesseract has quit IRC17:10
*** csomerville has quit IRC17:14
*** csomerville has joined #openstack-lbaas17:15
*** voelzmo has joined #openstack-lbaas17:19
*** amotoki__ has joined #openstack-lbaas17:21
*** amotoki_ has quit IRC17:22
*** amotoki__ has quit IRC17:24
*** catintheroof has quit IRC17:25
-openstackstatus- NOTICE: The Launchpad/UbuntuOne SSO OpenID provider is offline, preventing logins to review.openstack.org, wiki.openstack.org, et cetera; ETA for fix is unknown17:26
*** ChanServ changes topic to "The Launchpad/UbuntuOne SSO OpenID provider is offline, preventing logins to review.openstack.org, wiki.openstack.org, et cetera; ETA for fix is unknown"17:26
rm_workjohnsom: nova-lxd config? woo :P17:31
johnsomJust rebased to see if it was still broken17:31
*** voelzmo has quit IRC17:32
rm_workdo we know if the stuff Lubosz was working on merged?17:33
rm_workin the other projects17:33
rm_worknova?17:33
johnsomI think it was ubuntu launchpad lxd stuff.  No, I don't know if it is available now.17:34
johnsomI was just kind of poking it to see where we stand.  At some point I would like the lxd gate to let us know when it's fixed, but yeah, more important things at the moment.17:34
johnsomSpeaking of: https://review.openstack.org/44608217:35
johnsomIf you can log in.  Launchpad is down17:35
rm_workyeah i'm logged in17:38
rm_worklet me look17:38
*** amotoki_ has joined #openstack-lbaas17:39
rm_workjohnsom: couple questions17:43
openstackgerritMichael Johnson proposed openstack/octavia master: Add v2 member API section  https://review.openstack.org/46405017:43
johnsomWIP ^^^17:43
rm_workposted17:43
rm_workwoo, getting there... that'd be most of the core17:43
johnsomOk, it is mostly the "cookie-cutter" for tempest plugins17:43
rm_worktalked to a guy from CERN about his networking needs, looks like some of the work I just split off from FLIP and merged is stuff he needs too :P17:44
rm_workso that's good17:44
rm_workI had a feeling I wasn't alone17:44
johnsomYeah, the placeholders and tests are there so the gates pass until we actually get started on it17:44
rm_workk17:44
rm_workmostly it's about the .. template stuff17:45
johnsomYeah, it's by design.  I guess the copyright is valid, will fix that17:46
rm_workso ".. Change below this line" should still be there? >_>17:47
johnsomYeah, it's part of that tags block.  Everyone has it....17:47
rm_workblegh k17:47
johnsomhttps://raw.githubusercontent.com/openstack/neutron/master/README.rst17:47
rm_workbecause neutron is a shining star of quality :P17:48
johnsomI think it is a marker if they bulk update that block again17:48
rm_workk17:48
openstackgerritMichael Johnson proposed openstack/octavia-tempest-plugin master: Initial setup of the octavia-tempest-plugin repo  https://review.openstack.org/44608217:48
rm_workk17:49
rm_workooo, cupcakes17:49
johnsomHahahaha17:49
*** amotoki_ has quit IRC17:50
rm_workwhen you have time, I'm curious about your comments on https://review.openstack.org/#/c/463851/17:53
rm_worknot necessarily a complete code review17:53
rm_workbut just the high level strategy17:54
rm_workyou said a few words before, but17:54
johnsomYeah, I back-burnered it to make progress on api-ref17:54
rm_workyeah actually...17:54
rm_workthat's fine17:54
rm_workdo api-ref first :P17:54
rm_worki just have a burning curiosity17:54
openstackgerritAdam Harwell proposed openstack/octavia master: WIP: Floating IP Network Driver (spans L3s)  https://review.openstack.org/43561217:54
rm_workrebase17:55
johnsomIf it's really something you want comments on I can look at it this afternoon, but I've set a goal to at least get member and health monitor done today, which is really a stretch anyway17:55
johnsom2.5 cups of coffee to get started and some good pizza leftovers waiting for lunch.  I am in write docs mode.17:56
*** JudeC has joined #openstack-lbaas17:57
*** amotoki_ has joined #openstack-lbaas17:57
*** voelzmo has joined #openstack-lbaas18:01
*** voelzmo has quit IRC18:03
*** amotoki_ has quit IRC18:04
*** voelzmo has joined #openstack-lbaas18:05
*** voelzmo has quit IRC18:07
*** voelzmo has joined #openstack-lbaas18:08
rm_workoh yeah it's not really a big deal18:10
rm_worki haven't seen Sam anyway, he may be out18:10
rm_workI wanted to give him an update and ask for his thoughts too18:10
*** voelzmo has quit IRC18:12
*** voelzmo has joined #openstack-lbaas18:12
*** voelzmo has quit IRC18:13
*** atoth has joined #openstack-lbaas18:15
rm_workjohnsom: lol you got owned by reqs18:30
rm_workhttps://review.openstack.org/#/c/446082/218:30
rm_worki don't even know how18:31
rm_workthere's one req18:31
rm_worklol pbr18:31
rm_workI can fix it for you18:31
openstackgerritAdam Harwell proposed openstack/octavia-tempest-plugin master: Initial setup of the octavia-tempest-plugin repo  https://review.openstack.org/44608218:32
johnsomThanks18:32
*** voelzmo has joined #openstack-lbaas18:35
johnsomI guess it's been sitting there too long...18:36
rm_workyeah i didn't realize that repo existed18:39
*** ChanServ changes topic to "Welcome to LBaaS / Octavia"18:39
-openstackstatus- NOTICE: OpenID authentication through LP/UO SSO is working again18:39
openstackgerritMerged openstack/octavia master: Remove invalid url link  https://review.openstack.org/46390718:51
*** atoth has quit IRC18:51
*** voelzmo has quit IRC18:52
*** gcheresh has joined #openstack-lbaas19:01
rm_workalright need to do filtering still, and start on a spec for providers...19:02
*** gcheresh has quit IRC19:02
rm_workI am kinda hoping xgerman can start the providers spec as he seems to have a better idea about all the various things that requires <_<19:02
rm_workjohnsom: i'm considering in the client, creating a section "loadbalancer single", with "loadbalancer single create" and "loadbalancer single delete"19:04
rm_workmight make sense for cascade-delete19:04
*** gcheresh has joined #openstack-lbaas19:04
rm_workand would let us set up all the ridiculous options necessary for single-create19:04
johnsomIf you could pick up stats too it would be great, I'm not going to get to it before I leave for vacation19:05
rm_workugh k19:05
rm_worki'm not sure19:05
rm_worki have another conf next week19:05
rm_workand a big internal demo to do all of the prep work for19:05
johnsomrm_work cascade delete is already a --cascade on the delete cli19:05
rm_worklike i was thinking the week after for a while19:05
rm_workOH right, k19:05
rm_workthen maybe just for the create19:05
johnsomSo on create would it just have a ton of params?  How would that work???19:06
rm_work*i was thinking the week after next to start on ANY of this stuff19:06
rm_workthat's why i was trying to be super productive this week19:06
rm_worki mean19:06
rm_workyes :/19:06
rm_workbut i can imagine people might want it19:06
rm_workthe simplest form is actually not THAT long19:07
rm_workas i said, i'm just *considering* adding that19:07
johnsomOk.  I don't envy you on that as I think the details are going to get ugly (i.e. multiple members under a specific pool)19:07
rm_worklol yes19:07
rm_workmultiple pools/members? uuuugh19:07
rm_worki'm toying with it19:07
*** aojea has joined #openstack-lbaas19:07
rm_workfiltering/stats/providers need to come first19:07
johnsomWell, it's not useful if you can't add multiple members19:07
rm_workright19:08
johnsomYeah, that all sounds good to me.  I am mostly heads down on API-REF19:08
rm_workjust saying it'll suck19:08
johnsomThen install guide, then HA guide I think.  Those two will add value.19:09
johnsomPlus we still need to move over to the uWSGI goal..  sigh19:09
rm_workT_T19:15
*** voelzmo has joined #openstack-lbaas19:44
*** JudeC has quit IRC19:51
rm_workah http://logs.openstack.org/37/453237/5/check/gate-octavia-v1-dsvm-scenario-lxd-ubuntu-xenial-nv/d79855c/logs/devstacklog.txt.gz#_2017-05-11_16_18_43_24319:55
rm_workweird, image is missing19:55
rm_workahhh duh19:55
rm_worki can fix that19:55
*** ssmith has joined #openstack-lbaas19:55
johnsomssmith Amphora is a term we use for the "load balancing element" in Octavia.  Currently it is implemented as a service VM, but in the future container or bare metal.19:56
ssmithOK, I'm using neutron lbaas19:57
johnsomOk, but you could be using the Octavia driver behind neutron-lbaas.  Are you using the old haproxy namespace driver or Octavia driver for neutron-lbaas?19:57
johnsom[service_providers]19:58
johnsomservice_provider = LOADBALANCERV2:Octavia:neutron_lbaas.drivers.octavia.driver.OctaviaDriver:default19:58
johnsomin /etc/neutron/neutron_lbaas.conf19:58
ssmithhaproxy20:01
johnsomAh, ok.  Bear with us, most of us don't run that driver.20:01
*** JudeC has joined #openstack-lbaas20:02
johnsomrm_work Do you remember where the namespace driver drops the certs?20:03
*** voelzmo has quit IRC20:03
johnsomI know it doesn't use encrypted storage like octavia20:03
ssmithI think /var/lib/neutron-lbaas/cert20:06
johnsomor under /var/lib/lbaas20:08
johnsomAre the certs there that you expect?20:08
ssmithchecking20:08
johnsomYou can look at them with "openssl x509 -in <filename> -noout -text20:08
ssmithstill checking. BTW, https LB immediately comes back with ERR_CONNECTION_REFUSED20:14
ssmithDo the cert containers need the intermediary bundle with them too?20:14
ssmithok nothing there in those directories20:15
*** kobis has joined #openstack-lbaas20:15
rm_worknothing?20:17
rm_workno haproxy configs or anything?20:17
*** tomtomtom has joined #openstack-lbaas20:17
ssmith[certificates]20:17
ssmithbarbican_auth = barbican_acl_auth20:17
ssmithcert_manager_type = barbican20:17
ssmithstorage_path = /var/lib/neutron-lbaas/certficates/20:17
ssmithneutron:neutron is the owner of the directory20:20
openstackgerritAdam Harwell proposed openstack/octavia master: Update devstack to properly configure nova-lxd img  https://review.openstack.org/45323720:24
rm_workssmith: storage_path isn't actually used for your setup, so it's not relevant20:32
rm_workjust FYI20:32
rm_workit shouldn't harm anything either20:32
*** gcheresh has quit IRC20:33
ssmithSo there fact that the dir is empty isn't an idication of anything wrong?20:34
*** sanfern_mobile has joined #openstack-lbaas20:36
rm_workthat dir isn't even a real thing20:38
rm_work*on* the amphora (you'll need to ssh to it on the mgmt port) there should be stuff in /var/lib/octavia/20:38
*** sanfern_mobile has quit IRC20:40
*** sanfern_mobile has joined #openstack-lbaas20:41
rm_worksanfern_mobile: o/20:41
rm_workssmith: just to verify... you said you're not using octavia, you're using neutron-lbaas?20:42
rm_workbut, what driver do you use for neutron-lbaas?20:42
tomtomtom@rm_work we are using the NS Haproxy driver I can get more specific in a moment, grabbing the exact line out of config20:44
tomtomtomspecific driver line is: neutron_lbaas.drivers.haproxy.namespace_driver.HaproxyNSDriver20:45
rm_workOK then you don't have an amphora20:46
rm_workso ignore all of that20:46
rm_workI am not sure exactly how that driver works with certs20:47
tomtomtom@rm_work, so the neutron-agents container has the lbaas_agent.ini file with the driver, but the neutron-server container is where the service_auth section lives in neutron.conf, should the neutron.conf's be the same across the board?20:47
tomtomtomright now they are very different20:47
rm_worki have NO idea how the HaproxyNSDriver works20:47
tomtomtomok np20:48
*** cpuga has quit IRC20:48
rm_workalmost everyone who worked on it is no longer involved :(20:48
rm_workper one of our panels at the summit, we are going to be moving it out into a community supported repo20:48
rm_workand while we can assist, we can't dedicate much to it in terms of resources20:49
tomtomtomis it advisable in any way to try the octavia driver? in openstack newton or is that an entire overhaul of neutron?20:49
ssmithIt was installed with a playbook in the OpenStack Ansible repo, right?20:50
tomtomtomyes20:50
rm_workthe octavia driver is a much better option, IMO, but it is admittedly much more effort to set up20:51
rm_workhaproxyNS is not a good option for anything production <_<20:51
tomtomtomout entire system was built using openstack-ansible stable/newton20:51
tomtomtom:( ouch20:51
rm_workyeah we have an openstack-ansible role, but i think it was only just merged, so probably not until pike :(20:51
rm_workyou could probably backport it without much issue though...20:52
tomtomtomyeah I see the os_octavia role in the git clone, and I just did an implementation of barbican with the included os_barbican role.  I figure the same kind of work I did in barbican would also be necessary for octavia20:53
rm_workif you use containers, we're also in kolla and kolla-ansible20:54
johnsomhttps://docs.openstack.org/developer/openstack-ansible-os_octavia/20:54
johnsomFYI20:54
ssmithOK, tom, I'm voting to re-install as Octavia20:58
xgerman+1 - we had people install Newton Octavia with os_ansible21:00
rm_workFYI we're running octavia from master, against a Liberty cloud21:00
rm_workit's pretty much release independant21:01
xgermanYeah, that works, too :-)21:01
rm_workI'm not sure if anything in the apis we use have changed since ... like, forever21:01
rm_workwe just use the stable nova/neutron/glance APIs21:01
*** aojea has quit IRC21:05
*** sanfern has joined #openstack-lbaas21:07
tomtomtomok so is it safe to use the os_octavia role that came with stable/newton?21:09
tomtomtomfrom the openstack-ansible git repo21:09
ssmithSo is Octavia still under dev or is it ready of production?21:10
rm_workcan it be both?21:14
rm_workwe are using the 1.0 tag for the Pike release because we think it is ready as of Pike21:14
rm_workbut we're definitely still under heavy development21:14
ssmithlol, yes, ready for production but still having features added21:14
rm_workthere are a few things that need to be finished/merged for us to be at 100%, unfortunately :(21:15
ssmithDoes it work and does it work with terminated_https?21:15
rm_workwe're missing filtering capability and the stats calls from the neutron lbaasv2 api21:15
rm_workyes21:15
ssmithWell then you're going to get 1.0 tested in a Prod env21:16
rm_work:P21:16
*** sanfern_mobile has quit IRC21:16
rm_workI'm working on the same now, as are a couple of other args21:16
rm_work*orgs21:16
rm_workhopefully we'll have a few good production deployments to show by the time Pike comes out21:16
*** sanfern has quit IRC21:26
openstackgerritAdam Harwell proposed openstack/octavia master: Update devstack to properly configure nova-lxd img  https://review.openstack.org/45323721:37
rm_workit's weird that the lxd job doesn't have the same logs that the octavia stuff does21:38
rm_workit's not doing the trace logging21:38
johnsomTrace is turned on with -x21:39
johnsomBut, wait, we don't have that on in our gates.  It is WAY to chatty21:39
openstackgerritAdam Harwell proposed openstack/octavia master: Update devstack to properly configure nova-lxd img  https://review.openstack.org/45323721:39
rm_workjohnsom: so compare21:39
rm_workuhh one sec there's links coming21:40
rm_workthis is the normal gate (ignore the fact that it's building tgz, the test was wrong, i fixed it just now): http://logs.openstack.org/37/453237/6/check/gate-octavia-v1-dsvm-scenario-ubuntu-xenial/ca1edba/logs/devstacklog.txt.gz#_2017-05-11_20_39_29_21421:41
rm_workit's showing like21:41
rm_work2017-05-11 20:39:29.219 | ++ /opt/stack/new/octavia/devstack/plugin.sh:build_octavia_worker_image:87 :   /opt/stack/new/octavia/diskimage-create/diskimage-create.sh -o /opt/stack/new/octavia/diskimage-create/amphora-x64-haproxy.tgz '' '' '' -t tgz21:41
rm_work(I think those empty quotes are what broke it BTW, removed them in this next patch)21:42
rm_workcompare to the lxd gate run: http://logs.openstack.org/37/453237/6/check/gate-octavia-v1-dsvm-scenario-lxd-ubuntu-xenial-nv/bc48b02/logs/devstacklog.txt.gz#_2017-05-11_20_54_27_08221:43
rm_workall of that is missing21:43
*** cpuga has joined #openstack-lbaas21:44
rm_workbbiab changing location21:49
*** catintheroof has joined #openstack-lbaas21:53
johnsomIt doesn't look like that lxd gate is running devstack plugin.sh in echo mode21:54
johnsomMy guess is they aren't setting Debug=true in the local.conf21:54
openstackgerritMichael Johnson proposed openstack/octavia master: Add v2 member API section  https://review.openstack.org/46405022:26
openstackgerritMichael Johnson proposed openstack/octavia master: Add v2 member API section  https://review.openstack.org/46405022:30
*** nmagnezi_ has quit IRC22:31
*** nmagnezi has joined #openstack-lbaas22:31
rm_workOH I think he overrides the local.conf22:43
rm_workhttps://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/octavia.yaml#L244-L24722:44
rm_workor is that not what is happening22:44
rm_workI don't know, but I could add it there?22:44
rm_worklet's try https://review.openstack.org/46409022:45
rm_workjohnsom: ^^22:45
johnsomrm_work: https://github.com/openstack/octavia/tree/master/octavia/tests/contrib/hooks22:46
rm_workah22:46
johnsomhttps://github.com/openstack/octavia/blob/master/octavia/tests/contrib/gate_hook.sh#L2022:46
rm_workso not what I just did22:46
johnsomYeah, better place for that22:47
rm_workbut why isn't it on22:47
johnsomAh, yeah, the hooks stuff is new and the best way to do that stuff22:47
rm_workit *is* a scenario...22:48
johnsomThere is a gate problem with adding/creating local.conf in too many places or something22:48
rm_work:/22:48
rm_workso maybe in the gate22:49
rm_workwe should remove the lxd conf settings from the gate22:49
rm_workand add it to the hook22:49
rm_work?22:49
johnsomYeah, just move all that local conf stuff to one of those hook files and you should be good22:49
rm_workk22:49
rm_workhmm i don't know what var i pull "lxd" out of though22:51
rm_workthat gets... passed in?22:51
rm_workwell anyway I fixed that issue22:53
rm_workhttp://logs.openstack.org/37/453237/8/check/gate-octavia-v1-dsvm-scenario-lxd-ubuntu-xenial-nv/873344d/logs/devstacklog.txt.gz#_2017-05-11_22_14_32_93622:53
rm_worknow it's getting perms issues on some stuff during image build22:53
rm_workoh that's during upload?22:53
rm_workI think22:53
rm_workerr22:53
rm_workfff having debug would be good22:54
rm_worki'm still unclear on why debug isn't being set though22:55
rm_workoh, no this is *octavia* debug22:55
rm_workumm22:55
rm_workyeah so that conf-hook thing is fine22:57
rm_workit's something with the devstack localrc22:57
*** harlowja has quit IRC22:57
*** harlowja has joined #openstack-lbaas22:57
rm_worknormal scenario: http://logs.openstack.org/37/453237/8/check/gate-octavia-v1-dsvm-scenario-ubuntu-xenial/918d052/logs/local.conf.txt.gz22:58
rm_worklxd scenario: http://logs.openstack.org/37/453237/6/check/gate-octavia-v1-dsvm-scenario-lxd-ubuntu-xenial-nv/bc48b02/logs/local.conf.txt.gz22:58
rm_workI am not seeing why it doesn't do the plugin.sh in echo mode22:58
rm_workblegh22:59
rm_worki'll figure it out22:59
johnsomI am starting the health monitor api-ref, so I'm not going to dig too much into that unless you need me too...22:59
rm_worknah i'll do it later22:59
rm_worki'm gonna grab dinner22:59
rm_worko/22:59
rm_workjust frustrating22:59
rm_workat least it's closer by one step22:59
rm_work<_<22:59
*** kbyrne has quit IRC23:17
*** kbyrne has joined #openstack-lbaas23:19
johnsomrm_work I thought we fixed these: https://bugs.launchpad.net/octavia/+bug/169026823:58
openstackLaunchpad bug 1690268 in octavia "Octavia v2 API health monitor never goes "ONLINE"" [High,Triaged]23:58
johnsomOr at least I commented on it....23:59
*** sanfern has joined #openstack-lbaas23:59
johnsomI just saw it while getting example output for the docs23:59
« Wednesday, 2017-05-10 Index Friday, 2017-05-12 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!