Wednesday, 2018-03-14

*** gyee has quit IRC00:20
*** rbudden has joined #openstack-ironic00:33
*** liuzz has joined #openstack-ironic00:48
*** phuongnh has joined #openstack-ironic01:02
*** r-daneel has quit IRC01:02
*** itlinux has quit IRC01:12
*** harlowja has quit IRC01:20
*** yangyapeng has joined #openstack-ironic01:22
*** hshiina has joined #openstack-ironic01:24
*** yangyapeng has quit IRC01:26
*** yangyapeng has joined #openstack-ironic01:41
*** yufei has joined #openstack-ironic01:46
*** tiendc has joined #openstack-ironic01:55
openstackgerritzenghui.shi proposed openstack/ironic-specs master: Hardware interface for BIOS configuration  https://review.openstack.org/49648101:55
*** yangyapeng has quit IRC02:25
*** r-daneel has joined #openstack-ironic02:26
*** AlexeyAbashkin has joined #openstack-ironic02:36
*** AlexeyAbashkin has quit IRC02:40
*** fragatina has quit IRC02:46
*** fragatina has joined #openstack-ironic02:49
*** fragatina has quit IRC02:51
*** OctopusZhang has joined #openstack-ironic02:57
*** yufei has quit IRC02:59
*** OctopusZhang is now known as yufei02:59
*** zshi has quit IRC03:17
*** zshi has joined #openstack-ironic03:17
*** hshiina2 has joined #openstack-ironic03:31
*** hshiina has quit IRC03:31
*** AlexeyAbashkin has joined #openstack-ironic03:36
*** AlexeyAbashkin has quit IRC03:40
*** rbudden has quit IRC03:44
*** khyr0n has joined #openstack-ironic04:02
*** fragatina has joined #openstack-ironic04:08
*** namnh has joined #openstack-ironic04:09
*** fragatina has quit IRC04:09
*** fragatina has joined #openstack-ironic04:10
*** harlowja has joined #openstack-ironic04:12
*** ianychoi_ has joined #openstack-ironic04:12
*** ianychoi has quit IRC04:15
*** links has joined #openstack-ironic04:22
*** moshele has joined #openstack-ironic04:28
*** diga has joined #openstack-ironic04:29
openstackgerritOpenStack Proposal Bot proposed openstack/networking-generic-switch master: Updated from global requirements  https://review.openstack.org/55152104:30
*** moshele has quit IRC04:36
*** karthi has joined #openstack-ironic04:39
*** harlowja has quit IRC04:40
*** zshi has quit IRC04:51
*** zshi has joined #openstack-ironic04:51
*** karthi has quit IRC05:04
*** karthi has joined #openstack-ironic05:05
*** karthi has quit IRC05:09
*** dpawar has joined #openstack-ironic05:17
*** karthi has joined #openstack-ironic05:22
openstackgerritOpenStack Proposal Bot proposed openstack/bifrost master: Updated from global requirements  https://review.openstack.org/55274105:23
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-python-agent master: Updated from global requirements  https://review.openstack.org/55275605:31
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-tempest-plugin master: Updated from global requirements  https://review.openstack.org/55148905:31
openstackgerritOpenStack Proposal Bot proposed openstack/networking-baremetal master: Updated from global requirements  https://review.openstack.org/55151905:40
openstackgerritOpenStack Proposal Bot proposed openstack/networking-generic-switch master: Updated from global requirements  https://review.openstack.org/55152105:41
*** masber has quit IRC05:53
openstackgerritoctopuszhang proposed openstack/networking-generic-switch master: Use save_config provided by netmiko  https://review.openstack.org/54823805:54
*** karthi has quit IRC06:03
*** e0ne has joined #openstack-ironic06:04
*** karthi has joined #openstack-ironic06:11
*** karthi has quit IRC06:15
*** karthi has joined #openstack-ironic06:18
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-ui master: Imported Translations from Zanata  https://review.openstack.org/55279806:24
*** moshele has joined #openstack-ironic06:33
*** dims has quit IRC06:41
*** dims has joined #openstack-ironic06:43
*** ianychoi__ has joined #openstack-ironic06:43
*** ianychoi_ has quit IRC06:46
*** dims has quit IRC06:48
*** dims has joined #openstack-ironic06:49
*** ianychoi__ is now known as ianychoi06:51
*** mjura has joined #openstack-ironic06:55
*** pcaruana has joined #openstack-ironic07:04
openstackgerritMerged openstack/ironic-python-agent master: change python-libguestfs to python-guestfs for ubuntu  https://review.openstack.org/55183307:08
*** karthi has quit IRC07:15
*** karthi has joined #openstack-ironic07:15
openstackgerritoctopuszhang proposed openstack/networking-generic-switch master: Use save_config provided by netmiko  https://review.openstack.org/54823807:16
*** marios has joined #openstack-ironic07:22
openstackgerritDebayan Ray proposed openstack/sushy master: Make SessionService attribute optional  https://review.openstack.org/55281707:30
*** masber has joined #openstack-ironic07:32
*** OctopusZhang__ has joined #openstack-ironic07:32
*** yufei has quit IRC07:32
*** OctopusZhang_ has joined #openstack-ironic07:35
*** karthi has quit IRC07:38
*** karthi has joined #openstack-ironic07:38
*** OctopusZhang__ has quit IRC07:39
*** dpawar has quit IRC07:41
*** dpawar has joined #openstack-ironic07:48
*** AlexeyAbashkin has joined #openstack-ironic07:57
*** stendulker has joined #openstack-ironic08:05
*** diga has quit IRC08:08
*** karthi has quit IRC08:10
*** tesseract has joined #openstack-ironic08:14
*** karthi has joined #openstack-ironic08:16
*** arnewiebalck_ has quit IRC08:19
*** hoonetorg has quit IRC08:30
*** karthi has quit IRC08:33
*** mjura has quit IRC08:35
*** e0ne has quit IRC08:37
*** mjura has joined #openstack-ironic08:37
*** hoonetorg has joined #openstack-ironic08:44
*** khyr0n_ has joined #openstack-ironic08:45
*** athomas has joined #openstack-ironic08:47
*** khyr0n has quit IRC08:47
*** karthi has joined #openstack-ironic09:03
*** MattMan has quit IRC09:11
*** lucas-afk is now known as lucasagomes09:11
*** MattMan has joined #openstack-ironic09:12
*** mgoddard_ has joined #openstack-ironic09:24
*** karthi has quit IRC09:24
*** karthi has joined #openstack-ironic09:24
*** derekh has joined #openstack-ironic09:34
*** jcoufal has joined #openstack-ironic09:38
*** karthi has quit IRC09:43
*** jcoufal_ has joined #openstack-ironic09:56
*** jcoufal has quit IRC09:59
*** namnh has quit IRC10:00
*** OctopusZhang_ has quit IRC10:02
*** hshiina2 has quit IRC10:11
*** rcernin has quit IRC10:13
*** karthi has joined #openstack-ironic10:14
openstackgerritAndreas Jaeger proposed openstack/networking-baremetal master: Avoid tox_install.sh  https://review.openstack.org/55058710:18
*** karthi has quit IRC10:22
*** karthi has joined #openstack-ironic10:23
*** priteau has joined #openstack-ironic10:26
*** phuongnh has quit IRC10:28
*** karthi has quit IRC10:29
*** karthi has joined #openstack-ironic10:32
openstackgerritIlya Etingof proposed openstack/sushy-tools master: documentation on simulators added  https://review.openstack.org/52344510:32
*** mikal_ has joined #openstack-ironic10:52
*** mikal has quit IRC10:53
*** ianychoi has quit IRC11:07
*** ianychoi has joined #openstack-ironic11:13
*** stendulker has quit IRC11:15
*** serlex has joined #openstack-ironic11:16
openstackgerritIlya Etingof proposed openstack/sushy-tools master: Abstract away the virtualization management interface  https://review.openstack.org/52332111:17
openstackgerritIlya Etingof proposed openstack/sushy-tools master: documentation on simulators added  https://review.openstack.org/52344511:18
openstackgerritIlya Etingof proposed openstack/sushy-tools master: OpenStack cloud driver added  https://review.openstack.org/52387911:21
*** dtantsur|afk is now known as dtantsur11:26
dtantsurmorning/afternoon ironic11:26
*** karthi has quit IRC11:28
*** jcoufal_ is now known as jcoufal11:31
etingofdtantsur, o/11:35
*** karthi has joined #openstack-ironic11:36
*** sambetts|afk is now known as sambetts11:36
sambettso/11:36
dtantsuro/11:37
dtantsurwanna some nova fun? https://bugs.launchpad.net/nova/+bug/175560211:37
openstackLaunchpad bug 1755602 in OpenStack Compute (nova) "Ironic computes may not be discovered when node count is less than compute count" [Medium,In progress] - Assigned to Dan Smith (danms)11:37
dtantsurtl;dr cells attack again \o/11:37
*** karthi has quit IRC11:38
sambettsle sigh.11:38
*** tiendc has quit IRC11:52
*** e0ne has joined #openstack-ironic12:01
*** ajya has joined #openstack-ironic12:02
*** lucasagomes is now known as lucas-hungry12:03
*** vaatiesther has joined #openstack-ironic12:06
vaatiestheretingof12:07
*** trown|outtypewww is now known as trown12:08
*** trown is now known as trown|ruck12:08
*** dprince has joined #openstack-ironic12:10
openstackgerritSam Betts proposed openstack/ironic-lib master: [WIP] Use udevadm instead of hexdump to get mbr disk id  https://review.openstack.org/55289012:13
*** vaatiesther has quit IRC12:14
*** Sasha_ has joined #openstack-ironic12:21
*** racedo has joined #openstack-ironic12:22
*** hw_wutianwei has joined #openstack-ironic12:29
Sasha_etingof Hi, Ilya! I'mSasha. I want to become an intern in Outreachy program in redfish. How I can start?12:32
dtantsurlucas-hungry: hey! could you please move https://launchpad.net/sushy-tools to ironic-drivers team?12:36
dtantsurSasha_: hi! I'm not Ilya, but I see that he has a bug open for taking: https://bugs.launchpad.net/sushy-tools/+bug/175113412:37
openstackLaunchpad bug 1751134 in Sushy Tools "[RFE] Add unit tests" [Undecided,New]12:37
etingofSasha_, hey, welcome! ;) You should first pass the https://www.outreachy.org/apply/eligibility/ test, then follow the project contribution guidelines -- https://www.outreachy.org/2018-may-august/communities/openstack/#extend-redfish-hardware-provisioning-tooling-for-p12:39
dtantsuretingof: also check if there is some work to be done on the ironic side (e.g. for inspection)12:40
dtantsuroh, and someone can introduce redfish support into the standalone CI job12:40
dtantsurI was planning to do it.. but.. free time... :)12:40
*** derekh has quit IRC12:41
Sasha_dtansur: Thank you, I've looked at sushy project,but didn't look at sushy-tools.12:45
*** karthi has joined #openstack-ironic12:45
Sasha_etingof, I passed the test and read about redfish project itself.12:46
etingofdtantsur, well, that probably depends on how strong the intern(s) turn out to be12:46
etingofSasha_, now you need to pick a bug at LP tagged 'outreachy' and work on it. also you should officially apply for the project internship at Outreachy web site12:48
* dtantsur thought it starts in May..12:50
Sasha_etingof, OK, sure12:51
*** r-daneel has quit IRC12:53
*** SerenaFeng has joined #openstack-ironic12:58
openstackgerritHarald Jensås proposed openstack/ironic master: Set dhcp option boot-file-url if IPv6  https://review.openstack.org/55289712:58
*** rloo has joined #openstack-ironic13:02
Sasha_etingof, shoud I assign this bag myself or i'll be assigned automatically, when I upload code to review?13:03
*** mjturek has joined #openstack-ironic13:05
*** lucas-hungry is now known as lucasagomes13:07
lucasagomesdtantsur, sure!13:07
lucasagomesdtantsur, done :-)13:08
etingofSasha_, you should assign the bug on yourself13:09
Sasha_etingof, thanks13:10
*** rbudden has joined #openstack-ironic13:10
*** Goneri has joined #openstack-ironic13:13
*** jcoufal_ has joined #openstack-ironic13:20
*** rbudden has quit IRC13:20
*** etingof has quit IRC13:21
*** jcoufal has quit IRC13:22
*** e0ne has quit IRC13:25
radezbfournie: ping, could I get a bit of you time today? I'm still not able to get ironic to deploy on the overcloud, I've tried a few different ways with and with out containers and seem not to be able to get that pxe container to start up13:26
dtantsurthanks lucasagomes13:26
radezI wasn't sure if the ironic-inspector container  was supposed to be running too?13:27
*** karthi has quit IRC13:31
dtantsurradez: ironic-inspector is not needed for deployment13:31
radezdtantsur: thx13:34
jrollmorning ironic13:34
radezdtantsur: is neutron the one to handle the dhcp offers for the node? I don't have tftp up but I'm also not getting a dhcp offer for the node13:34
bfournieradez: are you seeing a dhcp discover from the node in the qdhcp netns13:36
bfournie?13:36
radezyup13:36
bfournieradez: and no offer?13:36
radezI see the bootp packets on the ctlplane interface on the control node but no offer is returned13:36
radezotoh, would there be a diffenet interface to listen on otther than the ctlplane's interface?13:37
*** dpawar has quit IRC13:38
radezI didn't look for the ip from the provisioning network that I'm creating13:38
bfournieradez: you should run tcpdump in the qdhcp netns13:38
radezah, k lemme do that...13:39
radezsec, I'm recreating my overcloud giddy up for the new ironic node right now13:39
*** derekh has joined #openstack-ironic13:40
bfournieradez: sudo ip netns exec qdhcp-<id> tcpdump …13:41
*** SerenaFeng has quit IRC13:41
radezbfournie: thx, not quite there yet, but I'd would have had to refresh my brain on that syntax :)13:42
bfournieradez: I can get on your system and look too, but it won't be for about an hour, we had 2 feet of snow yesterday and I need to dig out to get to office13:42
bfournieradez: also, to get netns id use "sudo ip netns"13:43
radezhehe, np, I think I may have borked my undercloud, I did all the configuring for the overcloud on the undercloud and I'm trying to clean it out now13:43
radezI'll let you know when I get to the tcpdump13:43
bfournieradez: ok thanks13:44
radezactually I think I'll just abandon the UC and redelpoy it next time I need to redeploy the overcloud13:44
*** ajya has quit IRC13:47
*** crushil has joined #openstack-ironic13:50
*** rloo has quit IRC13:51
*** rloo has joined #openstack-ironic13:52
*** baha has joined #openstack-ironic13:52
*** bfournie has quit IRC13:52
*** jcoufal has joined #openstack-ironic13:54
*** jcoufal_ has quit IRC13:57
*** racedo has quit IRC13:58
*** milan_ has joined #openstack-ironic14:00
rpiosoGood morning, ironicers.14:00
*** rbudden has joined #openstack-ironic14:12
*** SerenaFeng has joined #openstack-ironic14:13
*** karthi has joined #openstack-ironic14:14
*** trown|ruck is now known as trown|brb14:18
*** trown|brb is now known as trown|ruck14:20
radezdtantsur: ping, iconic on the overcloud... IronicNetwork defaults to ctlplane, but cltplane isn't attached to ovs for the packets to get to the qdhcp14:22
dtantsurradez: let's wait for bfournie please14:22
* dtantsur is overloaded now14:22
radezdtantsur: np14:22
openstackgerritSam Betts proposed openstack/ironic-lib master: [WIP] Use udevadm instead of hexdump to get mbr disk id  https://review.openstack.org/55289014:23
*** moshele has quit IRC14:30
TheJuliaGood morning14:44
dtantsurmorning TheJulia14:45
TheJuliadtantsur: you spoke of this free time thing? I thought it was a myth ;)14:45
dtantsurTheJulia: I keep hearing about it. maybe it's real. or maybe it's kind of a modern religion?14:46
*** e0ne has joined #openstack-ironic14:58
openstackgerritShivanand Tendulker proposed openstack/ironic master: Fix ``agent`` deploy interface to call ``boot.prepare_instance``  https://review.openstack.org/49905015:03
*** SerenaFeng has quit IRC15:06
rlooHappy Pi Day everyone, including dtantsur, TheJulia, radez, jroll. May we all have lots of pie!15:11
*** itlinux has joined #openstack-ironic15:12
dtantsurhey rloo15:12
dtantsurI'm having a lot of tripleo debugging, does it count?15:12
radezcheers rloo :)15:12
rloodtantsur: undercloud pi and overcloud pi? :)15:13
dtantsurmistral and tempest :)15:13
* rloo gives dtantsur a fresh apple pie with ice cream15:14
dtantsurdo we have human-friendly docs documenting https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L2358-L2433 ?15:18
*** links has quit IRC15:18
dtantsurTheJulia, jlvillal ^^^15:18
openstackgerritJulia Kreger proposed openstack/ironic-specs master: Add Rocky Priorities  https://review.openstack.org/55017415:26
TheJuliadtantsur: afaik, no15:27
dtantsurokay, writing15:27
TheJuliadtantsur: thanks15:27
*** moshele has joined #openstack-ironic15:34
NobodyCamGood Morning Ironic'ers15:35
NobodyCamHappy PiDay15:35
NobodyCamrloo: :) and a happy Pi day to you15:36
rlooNobodyCam: good morning, and thx! :)15:37
NobodyCam:)15:37
*** e0ne has quit IRC15:43
openstackgerritRuby Loo proposed openstack/python-ironicclient master: [doc] Add 'openstack create' command to command reference  https://review.openstack.org/55293215:45
dtantsurmorning NobodyCam15:46
NobodyCamGood morning dtantsur :)15:46
openstackgerritDmitry Tantsur proposed openstack/ironic-tempest-plugin master: Add example configuration for running our tests with nova  https://review.openstack.org/55293415:49
dtantsurTheJulia ^^^15:49
dtantsurideally we need to verify it, but it's at least some start15:49
TheJuliadtantsur: Yeah, I feel like it is missing explicit instructions on how to execute it15:52
TheJuliadtantsur: I guess the case is someone trying to run it against a deployment as a validation tool?15:52
*** etingof has joined #openstack-ironic15:53
*** bfournie has joined #openstack-ironic15:55
openstackgerritIlya Etingof proposed openstack/sushy-tools master: added system boot mode support  https://review.openstack.org/52610115:56
*** e0ne has joined #openstack-ironic15:58
*** SerenaFeng has joined #openstack-ironic16:01
openstackgerritIlya Etingof proposed openstack/sushy-tools master: added system boot mode support  https://review.openstack.org/52610116:03
*** mjura has quit IRC16:04
*** diga has joined #openstack-ironic16:04
rloosambetts, vdrok: is this a problem still? https://bugs.launchpad.net/nova/+bug/159983616:08
openstackLaunchpad bug 1599836 in OpenStack Compute (nova) "Booting Ironic instance, neutron port remains in DOWN state" [Medium,Confirmed]16:08
dtantsurTheJulia: for the former, we need to update https://docs.openstack.org/ironic-tempest-plugin/latest/usage.html#running16:09
dtantsurfor the latter - yes16:09
TheJuliaack, thanks16:09
*** masber has quit IRC16:10
*** dprince has quit IRC16:13
TheJuliaAnyone curious in doing the project onboarding session in Vancouver16:15
TheJulia?16:15
*** r-daneel has joined #openstack-ironic16:15
dtantsurI'm curious in one in Berlin :)16:17
sambettsrloo: if the environment hasn't loaded the networking-baremetal driver then they'll still see that16:19
TheJuliaI expected that :)16:19
rloosambetts: :-( we don't have that documented anywhere.16:19
rloosambetts: or do we? i will look...16:19
openstackgerritIlya Etingof proposed openstack/sushy-tools master: OpenStack cloud driver added  https://review.openstack.org/52387916:20
*** exodusftw has quit IRC16:22
*** diga has quit IRC16:22
TheJuliarloo: I _feel_ like we changed the docs towards the end of queens16:22
rlooTheJulia: yeah, i see it mentioned for flat network interface.16:23
rlooTheJulia: and we have a general 'recommend' to use it16:24
*** vkmc has joined #openstack-ironic16:24
*** exodusftw has joined #openstack-ironic16:25
rloosambetts: so should the bug be closed, since the solution is to use the networking-baremetal plugin?16:25
TheJuliarloo: I feel like the bug is fix released16:25
TheJuliarloo: and likely worth noting in the docs that there is the dependency16:25
TheJulias/docs/comments16:25
*** karthi has quit IRC16:27
openstackgerritIlya Etingof proposed openstack/sushy-tools master: added system boot mode support  https://review.openstack.org/52610116:28
*** masber has joined #openstack-ironic16:29
*** exodusftw has quit IRC16:30
*** AlexeyAbashkin has quit IRC16:30
TheJulialucasagomes: pas-ha: w/r/t ironic-staging-drivers... is there any desire to also just move that over to storyboard?16:30
* dtantsur votes yes16:30
lucasagomesTheJulia, I guess so, I think we should follow the rest of the ironic projects16:31
dtantsuretingof: see my question on https://bugs.launchpad.net/sushy-tools/+bug/175113316:31
openstackLaunchpad bug 1751133 in Sushy Tools "[RFE] Implement integration tests" [Undecided,Incomplete]16:31
lucasagomeseven tho it's not part of the umbrealla16:31
TheJuliaYeah, It kind of makes sense for to provide project grouping level visibility, maybe we would create two boards, official/unofficial16:31
* TheJulia is unsure at the moment16:31
*** dprince has joined #openstack-ironic16:32
*** SerenaFe_ has joined #openstack-ironic16:33
*** exodusftw has joined #openstack-ironic16:33
*** karthi has joined #openstack-ironic16:35
*** SerenaFeng has quit IRC16:35
etingofdtantsur, yes, responded16:37
dtantsuretingof: so, s/integration/functional/ ?16:38
dtantsurTheJulia: would be handy, since we have more unofficial projects, like vendor libs16:38
TheJuliaanyone want admin rights on my storyboard instance to play around? I can fire off import of a few other things so we can play with boards16:39
TheJuliadtantsur: rloo ^^^16:39
rlooTheJulia: nah, let's just pull the switch and use it16:40
*** hjensas has quit IRC16:40
* TheJulia likes rloo's view of this16:40
jrollTheJulia: this action item at the bottom, do you know if that's referring to deploy steps or something else? https://etherpad.openstack.org/p/ironic-rocky-ptg-beyond-the-qcow216:40
etingofdtantsur, reworded16:40
*** crushil has quit IRC16:41
TheJuliajroll: looking16:41
jrollspecifically "Further decompose what the actual most usable parts are"16:42
TheJuliajroll: I believe that is referring to deploy steps16:42
jrollthought so, thanks for confirming16:42
*** hjensas has joined #openstack-ironic16:42
dtantsurthanks etingof16:43
dtantsurjroll: hey, have you seen https://bugs.launchpad.net/nova/+bug/1755602 already?16:43
openstackLaunchpad bug 1755602 in OpenStack Compute (nova) "Ironic computes may not be discovered when node count is less than compute count" [Medium,In progress] - Assigned to Dan Smith (danms)16:43
rloosambetts: ok, i 'closed' (made invalid) that nova bug.16:44
openstackgerritPavlo Shchelokovskyy proposed openstack/networking-generic-switch master: Avoid tox-install.sh  https://review.openstack.org/55297416:44
*** karthi has quit IRC16:44
*** masber has quit IRC16:44
*** mjturek has quit IRC16:45
TheJuliajroll: yeah, we've had specs propose things like generic tftp configuration writing interfaces, and if we further decomposed things that could open the gate for crazy things like deploying lightbulbs, switches, raspberry pi's.  But the thought is truly further down the road16:45
*** mjturek has joined #openstack-ironic16:45
dtantsurpas-ha: hey, have you checked that doc/requirements.txt will be correctly updated by the bot?16:45
pas-hadunno for sure, this is what's happening / happened in other projects already16:46
*** milan_ has quit IRC16:46
jrollTheJulia: totally, just thought it was in an odd spot16:46
jrolldtantsur: yes, saw your link this morning16:46
TheJuliajroll: yeah, downside of thinking like 5 steps away is I might end up typing things like that16:47
pas-hadtantsur: http://codesearch.openstack.org/?q=reno&i=nope&files=doc%2Frequirements.txt&repos=16:47
jlvillaldtantsur, No docs that I know of for the ironic_configure_tempest function besides the comments.16:48
pas-hadtantsur: and here https://review.openstack.org/#/q/file:doc/requirements.txt+owner:%22OpenStack+Proposal+Bot%2216:49
pas-haso it seems to work16:49
dtantsurpas-ha: TIL!16:49
dtantsurwe need the same for other projects then16:49
dtantsurjlvillal: https://review.openstack.org/55293416:50
patchbotpatch 552934 - ironic-tempest-plugin - Add example configuration for running our tests wi...16:50
pas-hayep, would shave some time on gate tests, and local ones as well16:50
*** masber has joined #openstack-ironic16:51
jlvillaldtantsur, thanks16:51
*** Sasha_ has quit IRC16:51
dtantsurjlvillal: actually, gimme a minute, I need to clarify one more thing16:51
*** SerenaFe_ has quit IRC16:52
openstackgerritDmitry Tantsur proposed openstack/ironic-tempest-plugin master: Add example configuration for running our tests with nova  https://review.openstack.org/55293416:53
dtantsurjlvillal: now ^^^16:53
*** trown|ruck is now known as trown|lunch16:55
*** marios has quit IRC16:56
openstackgerritIlya Etingof proposed openstack/sushy-tools master: documentation on simulators added  https://review.openstack.org/52344516:56
pas-hadtantsur: btw there's a similar patch to networking-baremetal, tests are passing https://review.openstack.org/#/c/55058716:56
patchbotpatch 550587 - networking-baremetal - Avoid tox_install.sh16:56
dtantsurcool16:56
TheJuliahttps://review.openstack.org/#/c/550587 seems like a bad idea in terms of integration to me16:59
patchbotpatch 550587 - networking-baremetal - Avoid tox_install.sh16:59
TheJuliabecause our master could end up trailing with queens neutron for some time until a release is made, which ultimately means cross project gating and CI feedback would have release related delays17:00
*** sapd_ has joined #openstack-ironic17:01
jrollthat's a good point17:01
openstackgerritMerged openstack/ironic-python-agent master: Updated from global requirements  https://review.openstack.org/55275617:01
openstackgerritMerged openstack/ironic-lib master: Updated from global requirements  https://review.openstack.org/55275517:01
TheJuliaIf we think we fight breaking changes now, we'll never be able to release on time if breaking changes we don't understand appear in pypy suddenly17:02
jrollyeah, agree17:02
dtantsurTheJulia: won't upper-constraints protect us?17:03
jrolldtantsur: the problem is u-c won't be updated until late in the cycle when neutron makes a release17:03
TheJuliaonly if everything is gating together and no new constraints change gets in that breaks integration17:03
jrollor even until the final release17:03
dtantsuroh, yeah..17:03
TheJuliaso then we're talking about T* release supporting only Rocky neutron17:04
dtantsurbut I'm not sure we're the ones to make this decision, to be honest17:04
*** sapd has quit IRC17:05
TheJuliaI feel like this move is a side effect of other desires for pypi packaging of services, and that there are appropriate places for it, and there are places where it is a bad idea17:05
jrollI'm sure the same thing will come up for other neutron plugins, they should have similar concerns17:05
dtantsuryep17:05
jrollsambetts: ^^ you aware of the larger effort going on?17:05
dtantsurin any case, what we used to do will no longer work17:05
jrolldtantsur: why not?17:06
dtantsurjroll: tox_install.sh seems to be broken by it17:06
dtantsurI recall it was only usable for dependencies that are not in g-r/u-c17:06
* TheJulia suspects openstack just shot itself in the foot17:06
clarkbno no17:07
sambettsI remember there being a conversation about it, I believe zuulv3 handles the intergration thing by reading the requirements and working out the siblings of the projects17:07
dtantsurTheJulia: isn't it what openstack is about?17:07
*** crushil has joined #openstack-ironic17:07
dtantsur:)17:07
clarkbthe test setups are built to handle it17:07
sambettsthis ^17:07
jrollO_o17:07
clarkbif you use the tox siblings (and I believe devstack has similar handling for it) require projects on jobs will get installed from source17:07
jrolldoes zuul just do a sed as needed on those?17:07
clarkbbut otherwise you can accurately represent your dependencies in the dependency listing tools17:07
clarkbjroll: its more involved than that for tox in particular because virtualenvs17:08
TheJuliaclarkb: do you have a good example you can point us to?17:08
sambettsjroll: the zuul jobs install them into the tox env manually, and that satifies the pip install17:08
clarkbjroll: it actually does an install of everything in vanilla mode then goes back over and overwrites using source installs as instructed17:08
jrollclarkb: I see, thanks17:08
jrollTheJulia: +1, I'm also curious how it knows which ones to install from source17:09
clarkbit uses the required-projects list as specified in the zuul job17:09
clarkbso anything you list as a required source project will get installed from source17:09
dtantsurmagic and ponies..17:09
sambettshttp://logs.openstack.org/90/552890/2/check/openstack-tox-py27/d4f98f4/job-output.txt.gz#_2018-03-14_14_31_50_74502317:09
TheJuliaI guess there is actually value in this in that it would help enforce n-1 compatibility from the consumption standpoint...17:09
TheJuliadtantsur: lots of magic17:09
sambettsthat is an example ^17:09
TheJuliadtantsur: possibly my little pony levels17:10
clarkbdevstack is a little trickier because I think it will install test requirements too for reasons and also installs services from source17:10
clarkbso I'm not entirely sure how it will pan out there yet17:10
clarkbbut it already has tooling for editing constraints to handle source installs of the python-*client libs17:11
jrollso I guess I worry a bit about people running tests locally, and that varying from what we do in CI17:11
clarkbso I expect that will just get applied to neutron now too17:11
TheJuliajroll: it might just need to be something that there is visibiity/knowledge to... :\17:12
jrollTheJulia: that really sucks from a dev standpoint, honestly17:12
TheJulia3rd party ci could also be another major issue....17:12
clarkbyes there has always been tension on that front17:12
TheJuliaif they have their own initiators17:12
jrollpart of the reason we have the PTI is so tooling and humans both have a standard way to run tests17:12
jroll(AIUI)17:12
clarkbI don't think we've ever found a great answer to it. FWIW openstack has for a long long time explicitly not tested libs against master by default and instead tests their releases17:13
clarkbthen the libs are expected to in their own tests check compatibility17:13
clarkbwhich is how we've dealt with this particular issue. So CI is doing a very specific set of testing to check for that but should also be run in the local mode too17:13
TheJuliaexcept plugins shouldn't be treated like libraries17:13
clarkbplugins can only ever be installed from source?17:15
clarkbor rather don't have specific releases?17:15
sambettsplugins should be written against a stable API from the main project, and should declare their support for a particlar version, tracking master is impossible because they can break you at any time17:15
jrollsambetts: so we should expect plugins to be a cycle behind?17:16
sambettsjroll: not nessesarely, we start working on the next release as soon as the beta release for the next release of neutron is made17:17
TheJuliaand still possibly break or be broken during release week17:17
jrollsambetts: so x.y.z.b1 and such are on pypi?17:18
TheJuliasambetts: that implies that we track every piece of work and change going into the other project, or that they have CI testing that provides them feedback that htey are not breaking the plugin17:18
*** pcaruana has quit IRC17:18
openstackgerritPavlo Shchelokovskyy proposed openstack/networking-generic-switch master: Avoid tox-install.sh  https://review.openstack.org/55297417:18
clarkbTheJulia: ci feedback that they are not breaking plugins (or generally any other consumable api) is I think the intent17:18
clarkbTheJulia: and is generally how we structure things17:18
TheJuliaclarkb: except then we have to get projects to take such tests...  I've already stated this intent with horizon at-least17:19
sambettsTheJulia: the exact opposite actually, pinning to a particular neutron version means we actively have to change the version and therefore find out things we need to fix, instead of being silently broken by changes17:19
sambettsor running ci on every neutron patch17:19
TheJuliaCI on every neutron patch would be ideal, because that would provide visibility that they are making a breaking change and from there teams can negotiate and work together17:20
clarkbsambetts: yes neutron if it is going to be reconsumable and explicitly on pypi for this purpose should have testing to ensure they don't break things all the time17:20
*** e0ne has quit IRC17:20
TheJuliaI will quite literally table flip if our CI is dead for all plugins the week of release.17:20
clarkbthis doesn't necessarily mean integration testing with every plugin but maybe a representative set along with unittests and functional tests that exercise the apis17:20
jrollTheJulia: the cycle-trailing model may be a reasonable solution17:22
*** moshele has quit IRC17:22
TheJuliaclarkb: that is a good point, an "intergration unit test" would catch that soft of break quickly17:22
TheJuliajroll: to prevent mass table flipping, likely17:22
jrollyes :)17:22
sambettsjroll: its the only model that we've found that works for networking-cisco, trying to track every change in neutron is impossible, so we wait for the churn to stop and then fix what we need to17:23
*** moshele has joined #openstack-ironic17:23
* TheJulia wonders if anyone has a video or picture of an office of flipped tables17:23
jrollor if it just is an independent release model, just release when ready17:23
TheJuliajroll: I feel that doing this would eventually mutate to that for all of openstack17:24
jrollTheJulia: https://media.giphy.com/media/1hiVNxD34TpC0/giphy.gif17:24
jrollTheJulia: the independent model? I would approve of that for all of openstack :)17:25
*** moshele has quit IRC17:25
TheJuliathere just wouldn't be able to be a single coordinated release, but there could be explicit integration mappings that could be considered a release.... I guess....17:25
clarkbin places where there are explicit external contracts (largely libraries, think oslo and python-*client) the solution to this exact problem is constructing tests on the library side (in this case neutron) to do its best to prevent merging changes that would break consumers of those apis17:25
clarkbthis way you don't have to track every single change, the api is kept stable and it works on the next release17:26
openstackgerritIlya Etingof proposed openstack/virtualbmc master: Migrate CLI to cliff  https://review.openstack.org/54859417:26
clarkband when gaps are found and something breaks part of fixing the break is adding sufficient testing to prevent it from breaking in the future17:26
TheJuliaclarkb: so who is expected to put that sort of testing in to place to prevent such breaking changes?17:27
sambettsTheJulia: its the neutron driver API, so neutron are17:28
jrollclarkb: agree, we've just never really done this for plugins, and I expect we'll have some "learning experiences"17:28
clarkbTheJulia: aiui the library side has typically done the initial setup so for new oslo project they add a job or two for it and some unittests17:28
clarkbTheJulia: then when gaps are found its usually combo of people that were broken and the library adding tests to prevent it in the future17:28
TheJuliaI guess this means that unit tests will need to be such that all integration is not mocked out to be emulated in the unit testing and actually extends across the barrier17:31
jrollI was thinking more that neutron is the "library" in this case17:31
sambettsthis ^17:31
jrollit has a public python API for plugins, that API should be tested17:31
jrolljust like a library17:31
sambetts+100017:31
clarkbjroll: yup17:31
*** serlex has quit IRC17:32
jrollwe can't 100% prevent breakages without every plugin adding its integration tests to neutron's gate, but good unit tests can get close17:32
TheJuliaexcept individual consumers need to actually test that17:32
clarkband if we aren't going to make that assertion having neutron in global requirements is probably not a great idea (granted we don't have that ssertion from many external deps either but we should be able to do better here)17:32
TheJuliait can't be entirely contained to neutron's library17:32
TheJulianor can the library consuming mock out everything that it would normally talk to across that api17:33
jrollTheJulia: yes, plugins will need integration tests, but we can't run integration tests for hundreds of plugins on every neutron change17:33
jrollcould do a nightly job or something, maybe, dunno17:33
TheJuliajroll: agree, perhaps a bot that posts the gif you pasted into irc channels of impacted projects17:34
* TheJulia is attempting to be silly17:34
sambettshonestly from experience this is basically how we've been operating netowkring-cisco anyway17:35
sambettswe leave neutron master UTs non-voting, and test against the latest stable until the first neutron beta comes out17:36
sambettsthen we add a new CI job with neutron pinned to that version17:36
sambettswhich starts of non=voting17:36
sambettsthen once we've fixed everything we make it voting17:36
TheJuliaYeah, and just thinking about it all down the road this will greatly reduce the ability to have a whole integrated release because it is going to cascade into other projects17:37
* TheJulia forsees release model changes17:38
clarkbfwiw oslo seems to do a reasonable job without having needed to drastically change how we release things17:39
sambettsit does anyway, we've only just nailed down this process and its actually made it so that we can release sooner with the rest of OpenStack, before we spent so many dev cycles fixing stuff we ended up trailing a whole release17:39
clarkbyes things break but not all the often and we tend to catch them and fix them17:39
TheJuliaclarkb: except olso is a central point and thus can't really be used as a good example of impact17:39
clarkbnot sure I understand? oslo is effectively neutron in this case17:39
clarkblots of things consume oslo and they don't break every release17:40
clarkblots of thing want to consume neutron, we just have to be careful but should be able to achieve not breaking them constantly either17:40
sambettsfor what its worth (all except neutron moving stuff into neutron lib) the neutron ml2 driver API has been stable since mitaka17:40
TheJuliaso neutron will effectively commit to never break or change it's api moving forward?17:40
*** khyr0n_ has quit IRC17:40
clarkbTheJulia: I can't speak for them on that. What I am saying is if they want to be listed in global requirements then they need to do something along those lines17:40
sambettsnetworking-cisco (with some shims for the neutron-lib stuff) supports mitaka to queens for all our ml2 drivers17:40
clarkbotherwise they shouldn't go in global requirements imo as they aren't fit to be consumed that way17:40
sambettsTheJulia: or if they are going to break they should give a decent deprecation warning17:41
TheJuliaclarkb: ahh, I see17:41
TheJuliasambetts: yeah, except for a minimally maintained stable thing, sudden visibility at g-r bump is going to be frustrating.... then again, it might be the only way17:41
sambettsTheJulia: the deprecation warning should have been around a decent amount of time before a g-r bump that breaks that thing17:42
sambettsand they cover all over the UTs17:43
TheJuliasambetts: but that implies someone actually sees the warnings on a stable project17:43
TheJuliaand also implies that they raise awareness17:43
TheJuliaunless we have some sort of automated deprecation hunting thing17:43
sambettsrelease notes is a thing17:43
TheJuliastill requires someone to read and grok it17:43
TheJuliaand then act17:44
sambettsthats no different to our stance on the REST api right17:45
TheJuliasambetts: what do you mean?17:46
sambettsactually ignore that, I was thinking of something else17:46
*** moshele has joined #openstack-ironic17:46
TheJuliaheh17:46
*** dtantsur is now known as dtantsur|afk17:47
dtantsur|afko/17:47
sambettso/17:47
TheJuliaI just have the recent neutron->neutron-lib code changes in mind where the deprecation note was burried deep in the code, no warning, etc17:47
TheJuliagoodnight dtantsur|afk17:47
TheJuliagranted, I believe the note went in before release notes were even a thing17:48
sambettsTheJulia: most of the deprecation with the neutron-lib stuff output big warnings in the UTs for at least a cycle17:48
TheJulia"most" :)17:48
sambettsyeah... they weren't so good at that at first, but then again neutron-lib is them moving to a stable API for drivers from an undefined API17:49
TheJuliaregardless, I guess all we can really do is wait and see17:49
*** itlinux has quit IRC17:50
*** etingof has quit IRC17:50
sambetts+117:51
*** e0ne has joined #openstack-ironic17:53
*** moshele has quit IRC17:54
*** links has joined #openstack-ironic17:57
*** itlinux has joined #openstack-ironic17:58
TheJuliaOh, well, we can also drink beer in preparation17:58
sambettshaha yup17:59
jrollclarkb: I guess I assumed neutron would only be updated in g-r only after the final release, that's my primary concern, only updating and learning of breakages every 6 months. if we can target the milestone releases in g-r it seems much less impactful17:59
TheJulialots.. and lots... of beer17:59
clarkbjroll: constraints auto updates whenever things get released (still has to be reviewed but those tend to mostly just be formality)18:00
TheJuliaThat is true, all teams with integration points might need to budget for that hit.. which I guess means for us try and wrap the cycle a week earlier than before...18:01
jrollclarkb: yep, though I don't think it uses pre-releases?18:01
*** lucasagomes is now known as lucas-afk18:02
clarkbhrm ya it wont see those18:02
*** derekh has quit IRC18:04
openstackgerritMerged openstack/ironic-inspector master: Updated from global requirements  https://review.openstack.org/55275418:04
openstackgerritMerged openstack/ironic master: Updated from global requirements  https://review.openstack.org/55275318:05
*** gyee has joined #openstack-ironic18:05
*** mgoddard_ has quit IRC18:07
*** AlexeyAbashkin has joined #openstack-ironic18:07
*** crushil has quit IRC18:09
*** harlowja has joined #openstack-ironic18:11
*** AlexeyAbashkin has quit IRC18:11
*** fragatina has quit IRC18:15
*** fragatina has joined #openstack-ironic18:16
*** e0ne has quit IRC18:16
*** trown|lunch is now known as trown|ruck18:38
*** athomas_ has joined #openstack-ironic18:44
rlooi want to approve this RFE; need another core to ack: https://bugs.launchpad.net/virtualbmc/+bug/175291118:44
openstackLaunchpad bug 1752911 in VirtualBMC "[RFE] Rebase VirtualBMC to cliff" [Wishlist,In progress] - Assigned to Ilya Etingof (etingof)18:44
jrollrloo: +1'd18:45
rloojroll: thx!18:45
jrollnp :)18:45
*** mgoddard_ has joined #openstack-ironic18:48
*** e0ne has joined #openstack-ironic18:54
*** itlinux_ has joined #openstack-ironic18:57
*** itlinux has quit IRC18:57
*** fragatina has quit IRC19:01
*** crushil has joined #openstack-ironic19:04
*** tesseract has quit IRC19:20
*** athomas_ has quit IRC19:22
*** ajya has joined #openstack-ironic19:33
*** links has quit IRC19:34
*** moshele has joined #openstack-ironic19:35
*** ajya has quit IRC19:35
*** e0ne has quit IRC19:36
*** moshele has quit IRC19:40
*** itlinux_ has quit IRC19:40
*** moshele has joined #openstack-ironic19:41
*** jcoufal has quit IRC19:44
*** e0ne has joined #openstack-ironic19:48
*** mgoddard_ has quit IRC19:48
*** rbudden has quit IRC19:49
*** itlinux has joined #openstack-ironic19:52
*** crushil has quit IRC19:53
*** sambetts is now known as sambetts|afk19:53
openstackgerritJulia Kreger proposed openstack/bifrost master: Define ironic_db_password default  https://review.openstack.org/55303620:04
openstackgerritJulia Kreger proposed openstack/bifrost master: Update to fedora-27 for testing  https://review.openstack.org/54711920:06
rlooTheJulia: backport to pike. is it OK? https://review.openstack.org/#/c/552515/20:21
patchbotpatch 552515 - ironic (stable/pike) - Introduce keystoneauth adapters for clients20:21
rlooTheJulia: author sez it is to address https://bugs.launchpad.net/ironic/+bug/175516420:21
openstackLaunchpad bug 1755164 in Ironic "iLO driver generates public tempurls for vmedia" [Medium,Triaged]20:21
rlooTheJulia: and someone else said it was to address a security issue (which didn't make sense to me)20:21
TheJuliauhh.. interesting20:23
*** mgoddard_ has joined #openstack-ironic20:25
*** etingof has joined #openstack-ironic20:25
jrollrloo: TheJulia: I would prefer they fix that directly on stable, rather than backporting the keystone auth thing20:26
TheJuliaYeah...20:26
jrollthat backport isn't to address a security issue, but rather that it's a dependency for the patch in master that addresses the security issue20:26
TheJulia++20:26
*** e0ne has quit IRC20:27
rloojroll: ok, please comment :) why does a fix in master need a patch in stable.  i am guessing they want to backport the fix...20:27
jrollrloo: suspect the fix doesn't apply at all without that backport20:28
* jroll comments20:28
rloothx jroll. TheJulia, did we make jroll a stable-maint-person yet?20:28
*** hw_wutianwei has quit IRC20:31
TheJuliaI commented/20:31
TheJuliaas well20:31
TheJuliaAnd I think we should give jroll some time to breath ;)20:32
*** hw_wutianwei has joined #openstack-ironic20:32
rlooTheJulia: :)20:33
*** AlexeyAbashkin has joined #openstack-ironic20:36
jroll:P20:37
*** ajya has joined #openstack-ironic20:37
*** ElCoyote_ has quit IRC20:39
*** fragatina has joined #openstack-ironic20:39
*** AlexeyAbashkin has quit IRC20:40
*** rbudden has joined #openstack-ironic20:46
*** dprince has quit IRC20:52
*** moshele has quit IRC20:56
*** trown|ruck is now known as trown|outtypewww21:03
*** rcernin has joined #openstack-ironic21:08
rloojroll: i wrote a sep spec for the deploy step framework cuz i didn't want to lose the bigger questions in the original spec :)21:11
jrollrloo: yeah, I saw that after commenting, fine with me :)21:11
*** Goneri has quit IRC21:12
openstackgerritAndreas Jaeger proposed openstack/networking-generic-switch master: Avoid tox-install.sh  https://review.openstack.org/55307721:15
*** AJaeger has joined #openstack-ironic21:33
*** AJaeger has left #openstack-ironic21:34
*** itlinux has quit IRC21:40
*** ricardoas has joined #openstack-ironic21:55
*** baha has quit IRC21:59
*** ajya has quit IRC22:01
*** mgoddard_ has quit IRC22:14
*** priteau has quit IRC22:15
*** e0ne has joined #openstack-ironic22:21
*** e0ne has quit IRC22:22
*** fragatina has quit IRC22:23
*** fragatina has joined #openstack-ironic22:24
*** fragatina has quit IRC22:29
*** fragatina has joined #openstack-ironic22:36
*** fragatina has quit IRC22:36
*** bfournie has quit IRC22:37
*** fragatina has joined #openstack-ironic22:38
*** fragatina has quit IRC22:40
*** marlinc has quit IRC22:44
*** fragatina has joined #openstack-ironic22:46
*** masber has quit IRC22:57
*** masber has joined #openstack-ironic23:00
*** rbudden has quit IRC23:12
*** hw_wutianwei has quit IRC23:24
*** hw_wutianwei has joined #openstack-ironic23:24
*** marlinc has joined #openstack-ironic23:29
*** rloo has quit IRC23:43
*** itlinux has joined #openstack-ironic23:47
openstackgerritRuby Loo proposed openstack/ironic master: Add descriptions for config option choices  https://review.openstack.org/55310123:49
*** dprince has joined #openstack-ironic23:51
*** hamzy has quit IRC23:56
*** hamzy has joined #openstack-ironic23:58

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!