Tuesday, 2015-03-24

*** ijw has quit IRC00:01
*** achanda has quit IRC00:03
rloog'night everyone. I'm off; will be back on Thursday.00:03
*** rloo has quit IRC00:03
NobodyCamnight rloo00:05
NobodyCamJayF: :(00:05
NobodyCamuggh00:05
JayFNobodyCam: talking to them now, doesn't seem like we're likely to be bitten by the exact race he was talking about00:06
JayFNobodyCam: he said it impacts /dev/disk/by-uuid/* links more than anything else00:06
JayFi.e. it says it's done but that link isn't there yet00:06
*** Marga_ has quit IRC00:08
*** Marga_ has joined #openstack-ironic00:08
*** Marga__ has joined #openstack-ironic00:10
*** Marga_ has quit IRC00:10
TheJuliaBadCub_Hiding: gray goose would work00:11
mrdaWhy are you North Americans away tomorrow?00:12
JayFI'm not away tomorrow00:12
JayFafaik there aren't any holidays00:13
TheJuliajroll: so I guess the concern I would have is if its a concern, then initalizaiton of other tests should also likely be a concern00:13
*** igordcard_ has quit IRC00:13
TheJuliabut I'm not going to think about it anymore today because I want to actually sleep tonight00:13
egonSleeping is overrated.00:13
NobodyCammrda: just BadCub_Hiding as he is going to pick up his car00:13
TheJuliait is, sadly, required00:13
mrdaoh00:14
* TheJulia actually has to take meds to fall asleep00:14
NobodyCameveryone else will be here00:14
egon:(00:14
* NobodyCam goes back to other things00:15
*** Marga__ has quit IRC00:28
*** naohirot has joined #openstack-ironic00:30
*** jcoufal has joined #openstack-ironic00:38
*** coolsvap|afk is now known as coolsvap00:50
openstackgerritJay Faulkner proposed openstack/ironic: Ensure configdrive isn't mounted in CoreOS ramdisks  https://review.openstack.org/16706301:00
openstackgerritJay Faulkner proposed openstack/ironic: Ensure configdrive isn't mounted in CoreOS ramdisks  https://review.openstack.org/16706301:02
*** mtanino has joined #openstack-ironic01:09
*** jerryz has joined #openstack-ironic01:19
*** mtanino has quit IRC01:32
*** korekhov has quit IRC01:59
*** r-daneel has quit IRC02:02
*** korekhov has joined #openstack-ironic02:08
*** ijw_ has quit IRC02:19
*** jcoufal has quit IRC02:29
*** korekhov has quit IRC02:29
*** korekhov has joined #openstack-ironic02:30
*** spandhe has quit IRC02:32
*** kkoski has joined #openstack-ironic02:34
*** korekhov has quit IRC02:34
*** achanda has joined #openstack-ironic02:37
*** killer_prince has quit IRC02:39
*** absubram has joined #openstack-ironic02:40
*** hj-hp has quit IRC02:41
*** ramineni has joined #openstack-ironic02:42
*** killer_prince has joined #openstack-ironic02:46
*** killer_prince is now known as lazy_prince02:46
*** achanda has quit IRC02:53
*** achanda has joined #openstack-ironic02:56
naohirotBadCub_Hiding: Hi, I updated https://wiki.openstack.org/wiki/Ironic/Drivers#Drivers03:01
naohirotBadCub_Hiding: thanks :)03:01
*** alex_xu has quit IRC03:12
*** kkoski has quit IRC03:16
*** achanda has quit IRC03:24
*** achanda has joined #openstack-ironic03:32
*** Nisha has joined #openstack-ironic03:34
*** achanda has quit IRC03:41
*** achanda has joined #openstack-ironic03:43
*** spandhe has joined #openstack-ironic03:52
*** achanda has quit IRC03:53
*** achanda has joined #openstack-ironic03:59
*** Marga_ has joined #openstack-ironic04:06
*** Marga_ has quit IRC04:06
*** Marga_ has joined #openstack-ironic04:07
*** spandhe_ has joined #openstack-ironic04:12
*** zhidong has joined #openstack-ironic04:13
*** spandhe has quit IRC04:14
*** spandhe_ is now known as spandhe04:14
openstackgerritAnusha Ramineni proposed stackforge/proliantutils: ILO: (PART-1) Test cases for RIS Operations  https://review.openstack.org/16677904:28
openstackgerritAnusha Ramineni proposed stackforge/proliantutils: ILO: (PART-1) Test cases for RIS Operations  https://review.openstack.org/16677904:31
ramineniNisha: ^^04:32
*** coolsvap is now known as coolsvap|afk04:44
*** ChuckC has joined #openstack-ironic04:51
*** sandhya has joined #openstack-ironic05:04
openstackgerritNisha Agarwal proposed stackforge/proliantutils: Adding functionality required by ilo-inspection  https://review.openstack.org/16414205:08
Nisharamineni ^^^^05:09
openstackgerritShivanand Tendulker proposed openstack/ironic: grub2 bootloader support for uefi boot mode  https://review.openstack.org/16619205:09
Nishastendulker ^^^^05:09
*** Marga_ has quit IRC05:10
*** harlowja is now known as harlowja_away05:22
*** Nisha_away has joined #openstack-ironic05:28
*** Nisha has quit IRC05:28
*** chenglch has joined #openstack-ironic05:33
*** lazy_prince has quit IRC05:37
*** rameshg87 has joined #openstack-ironic05:38
rameshg87good morning ironic05:38
*** killer_prince has joined #openstack-ironic05:41
*** killer_prince is now known as lazy_prince05:41
Nisha_awayrameshg87, https://review.openstack.org/16414205:43
rameshg87Nisha_away: ack05:45
rameshg87Nisha_away: i hope you are not really away ;-)05:46
Nisha_awayrameshg87, hehe05:46
openstackgerritPSPradhan proposed stackforge/proliantutils: Update RIS library  https://review.openstack.org/16329005:47
mrdaMorning rameshg8705:59
*** ukalifon1 has joined #openstack-ironic06:00
rameshg87good afternoon (or rather evening) mrda :)06:00
mrda:)06:03
openstackgerritPSPradhan proposed stackforge/proliantutils: Update RIS library  https://review.openstack.org/16329006:04
*** achanda has quit IRC06:15
*** Nisha_away is now known as Nisha06:17
* Nisha nisha is back06:18
*** ramineni is now known as ramineni-away06:18
*** ramineni-away is now known as ramineni06:19
*** yog__ has joined #openstack-ironic06:36
*** pcaruana has quit IRC06:41
*** chlong has quit IRC06:42
*** Marga_ has joined #openstack-ironic06:52
*** dlpartain has joined #openstack-ironic07:00
openstackgerritAnusha Ramineni proposed stackforge/proliantutils: Refactor code in RIS  https://review.openstack.org/16711707:01
*** chlong has joined #openstack-ironic07:03
*** dlpartain has left #openstack-ironic07:09
*** killer_prince has joined #openstack-ironic07:13
openstackgerritYuiko Takada proposed stackforge/ironic-discoverd: Add supporting generate config files  https://review.openstack.org/16328107:13
*** lazy_prince has quit IRC07:14
*** killer_prince is now known as lazy_prince07:14
*** lazy_prince has quit IRC07:18
*** killer_prince has joined #openstack-ironic07:20
*** killer_prince is now known as lazy_prince07:20
openstackgerritYuiko Takada proposed stackforge/ironic-discoverd: Add supporting generate config files  https://review.openstack.org/16328107:22
*** lazy_prince has quit IRC07:25
*** lazy_prince has joined #openstack-ironic07:30
*** Marga_ has quit IRC07:33
*** Marga_ has joined #openstack-ironic07:33
openstackgerritZhenguo Niu proposed openstack/ironic: Add attempt of max retries to power state sync logs  https://review.openstack.org/16712207:37
*** pas-ha has joined #openstack-ironic07:41
*** yog__ has quit IRC07:41
*** yog__ has joined #openstack-ironic07:41
openstackgerritGhe Rivero proposed openstack/ironic-python-agent: Use oslo_log lib  https://review.openstack.org/16279007:44
openstackgerritGhe Rivero proposed openstack/ironic-python-agent: Sync from oslo.incubator  https://review.openstack.org/16279107:44
HaomengGheRivero: ping07:48
GheRiveropong07:48
HaomengGheRivero: I have a question about i18n lib07:48
GheRiveroHaomeng: sure. tell me07:49
HaomengGheRivero: do you know if the new i18n lib from oslo, which support lazy translation or not?07:49
GheRiveroyes it does07:49
HaomengGheRivero: means if it can handle the 'accept-language' header parameters in the request07:49
*** Marga_ has quit IRC07:49
HaomengGheRivero: that means we can do nothing in our ironic code, i18n lib can hanndle the lazy translation, right?07:50
GheRiveroshpuld be possible07:50
GheRiverodoc is not very verbose about that, but I guess it will be possible07:51
GheRiverohttp://docs.openstack.org/developer/oslo.i18n/usage.html#lazy-translation07:51
HaomengGheRivero: ok, let me check the doc, thank you:)07:52
HaomengGheRivero: :)07:52
HaomengGheRivero: but I can not find the enable_lazy() method call in our ironic code07:53
HaomengGheRivero: so should we add enable_lazy() call to enable it?07:53
GheRiveroI was just checking that. No we don't call it :/ but we should07:53
HaomengGheRivero: agree with you07:54
HaomengGheRivero: but not sure if "trans_msg = oslo_i18n.translate(msg, desired_locale=my_locale)" should be add also07:54
HaomengGheRivero: maybe it should be in the restapi handling logic07:54
GheRiveroHaomeng: agree07:55
HaomengGheRivero: :)07:56
HaomengGheRivero: GheRivero: how do you think if this is feature or not, if yes, can not add in such k3 release07:56
HaomengGheRivero: maybe a lot of code should be changed to support lazy translate07:56
*** Nisha has quit IRC07:56
*** pas-ha has quit IRC07:56
GheRiveroHaomeng: I think is not too much code, but some people will consider it a feature :(07:57
*** hyakuhei has joined #openstack-ironic07:57
*** Nisha has joined #openstack-ironic07:57
GheRiveroHaomeng: we can propose a patch and wait to someone too complain... or not07:58
*** pas-ha has joined #openstack-ironic07:58
HaomengGheRivero: yes, but so far, I have no idea which code should be changed, let me check with other projects07:58
openstackgerritGhe Rivero proposed openstack/ironic: Use oslo_log lib  https://review.openstack.org/15760207:58
openstackgerritGhe Rivero proposed openstack/ironic: Sync with oslo.incubator  https://review.openstack.org/16594007:58
HaomengGheRivero: depens on our ironic existing code07:58
HaomengGheRivero: this is nova code - https://github.com/openstack/nova/blob/f4e3effb72e9bba2dd4c5471447e45e4ee9b529f/nova/api/openstack/wsgi.py#L116808:00
HaomengGheRivero: but we have no such wsgi python code in our ironic api08:02
GheRiveroHaomeng: too bad. We must look were to put it in our rest code then08:03
HaomengGheRivero: yes, so I think this is not a bug level change08:04
HaomengGheRivero: https://github.com/openstack/ironic/blob/b4f1db88683f5bb9ff0af7159e34c3117aa0beff/ironic/cmd/api.py#L4808:05
HaomengGheRivero: we just create a simple wsgi server -  wsgi = simple_server.make_server08:05
HaomengGheRivero: any idea? :)08:08
GheRiveroHaomeng: not now. But I'll check on this later and be sure to find one08:10
HaomengGheRivero: strange, nova code did not call enable_lazy() also08:10
HaomengGheRivero: ok, thank you08:10
HaomengGheRivero: no rush:)08:10
HaomengGheRivero: keystone called - https://github.com/openstack/keystone/blob/ce701f27710c1000d5d7babd3a7bd3fe7e1af083/keystone/server/wsgi.py#L2508:10
*** chlong has quit IRC08:12
HaomengGheRivero: another question here08:12
GheRiveroHaomeng: truth be told, I give up worrying about i18n a long time ago. It's mostly error prone and almost impossible to find any error/solution in google/internet if is not in English08:13
HaomengGheRivero: I understand the wsgi should have common code in oslo, but not sure, do you know? if yes, we can run with same common code to support lazy translation08:13
HaomengGheRivero: yes, i18n is our gap:)08:13
HaomengGheRivero: :)08:15
GheRiveroHaomeng: I remember a couple of cycles ago there was something about a common wsgi, but lost track of it, so I guess they didn't make it08:15
HaomengGheRivero: ok, got it:)08:15
HaomengGheRivero: so dont worry:)08:15
HaomengGheRivero: :)08:15
*** ifarkas has joined #openstack-ironic08:21
*** ndipanov has joined #openstack-ironic08:28
*** lazy_prince has quit IRC08:28
*** killer_prince has joined #openstack-ironic08:28
*** killer_prince is now known as lazy_prince08:28
*** dtantsur|afk is now known as dtantsur08:33
dtantsurMorning!08:33
Nishadtantsur, morning!08:34
dtantsuro/08:35
*** Nisha_away has joined #openstack-ironic08:36
*** Nisha has quit IRC08:36
*** hyakuhei has quit IRC08:38
openstackgerritAnusha Ramineni proposed stackforge/proliantutils: Fix exceptions  https://review.openstack.org/16713908:45
*** killer_prince has joined #openstack-ironic08:47
*** lazy_prince has quit IRC08:47
*** killer_prince is now known as lazy_prince08:47
*** andreykurilin_ has joined #openstack-ironic08:53
*** Marga_ has joined #openstack-ironic08:53
*** Nisha has joined #openstack-ironic08:55
*** Nisha_away has quit IRC08:55
*** andreykurilin_ has quit IRC08:59
*** zhidong has quit IRC09:00
*** andreykurilin_ has joined #openstack-ironic09:00
*** jistr has joined #openstack-ironic09:04
*** romcheg has joined #openstack-ironic09:06
*** lucasagomes has joined #openstack-ironic09:11
*** derekh has joined #openstack-ironic09:15
*** egon has quit IRC09:19
*** lazy_prince has quit IRC09:20
*** tiagogomes_ has joined #openstack-ironic09:21
*** Marga_ has quit IRC09:21
*** ukalifon1 has quit IRC09:21
*** killer_prince has joined #openstack-ironic09:22
*** killer_prince is now known as lazy_prince09:22
*** egon has joined #openstack-ironic09:24
*** hyakuhei has joined #openstack-ironic09:39
*** killer_prince has joined #openstack-ironic09:47
*** lazy_prince has quit IRC09:48
*** killer_prince is now known as lazy_prince09:48
*** ukalifon has joined #openstack-ironic09:48
*** naohirot has quit IRC09:50
*** achanda has joined #openstack-ironic09:51
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Add config drive documentation and script  https://review.openstack.org/16620109:53
openstackgerritAnusha Ramineni proposed stackforge/proliantutils: Fallback to RIBCL  https://review.openstack.org/16715109:54
raminenirameshg87: ^^^09:54
*** achanda has quit IRC09:56
*** lazy_prince has quit IRC09:57
*** killer_prince has joined #openstack-ironic09:58
*** killer_prince is now known as lazy_prince09:58
*** andreykurilin_ has quit IRC09:59
*** andreykurilin_ has joined #openstack-ironic09:59
rameshg87ramineni: ack10:04
*** chenglch has quit IRC10:07
*** alex_xu has joined #openstack-ironic10:07
devanandamorning, all10:08
rameshg87devananda: morning10:09
rameshg87devananda: back to france again ? :)10:09
devanandarameshg87: yup10:10
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Add root device hints documentation  https://review.openstack.org/16680110:11
lucasagomesdevananda, morning, that's very early :)10:11
devanandalucasagomes: nope. it's 11am :)10:11
lucasagomesoh france10:11
lucasagomesjust read it now10:11
devanandagenerating the state diagram still doesn't work for me. dont suppose anyone else has tried tools/states_to_dot.py ?10:12
* lucasagomes haven't10:14
devanandaoh. it's misisng a quotation mark :-/10:14
lucasagomesdevananda, btw, once you have a time mind taking a look at the comments at https://review.openstack.org/166201 ? There's some discussion whether the script to generate the configdrive should live under tools (now tools/contrib) or should be in the ironic-client10:15
*** pelix has joined #openstack-ironic10:15
lucasagomesoh :/ it would be good to have unittests for such tools10:15
devanandaclient10:15
openstackgerritDevananda van der Veen proposed openstack/ironic: Fix formatting error in states_to_dot  https://review.openstack.org/16715910:16
*** andreykurilin__ has joined #openstack-ironic10:17
*** andreykurilin_ has quit IRC10:17
*** vdrok_afk is now known as vdrok10:23
vdrokmorning ironic10:23
*** Nisha has quit IRC10:23
*** spandhe has quit IRC10:26
openstackgerritMerged stackforge/ironic-discoverd: More generic option for tuning adding ports  https://review.openstack.org/16690810:30
*** andreykurilin_ has joined #openstack-ironic10:37
*** andreykurilin__ has quit IRC10:37
lintanhi devananda10:41
devanandajroll: how close to accurate is this? http://www.websequencediagrams.com/cgi-bin/cdraw?lz=Tm92YS0-QVBJOiBzZXQgaW5zdGFuY2VfaW5mbwoADg9wcm92aXNpb25fc3RhdGUKQVBJLT5Db25kdWN0b3I6IGRvX25vZGVfZGVwbG95KCkKABMJABoNdXBkYXRlIHB4ZSwgdGZ0cCBjb25maWdzACMMTmV1dHJvbgAkCURIQ1BCT09UABcNb2RlOiBwb3dlciBvbgpOb2RlAC4LREhDUCByZXF1ZXN0CgBIBwAsCG5leHQtc2VydmVyID0gAIE5CQA8BwCBRwthdHRlbXB0cwCBIwVib290AG4Sc2VuZCBhZ2VudCBrZXJuZWwsIHJhbWRpc2ssICYAgVQHAIEUCACBLAVydW5zACoHA10:43
devanandaargh. sorry - didn't realize the link was *that* long :(10:44
*** andreykurilin_ has quit IRC10:45
*** hyakuhei has quit IRC10:47
*** yuanying has quit IRC10:51
lintandevananda, do we have a plan to replace ironic's object code by using oslo.versionedobjets in L?11:00
devanandalintan: it has'nt been discussed yet11:00
devanandalintan: but I believe that is the direction we want to go, once oslo.versionedobjects is ready for us to migrate to11:01
*** sandhya has quit IRC11:03
lintandevananda, OK, it will be great if Ironic can make it11:04
devanandalintan: are you volunteering to work on that? :)11:05
*** ramineni has quit IRC11:05
lintandevananda, Yes, for sure.11:05
devanandalintan: great!11:05
devanandalintan: feel free to add it here with any details you'd like -- https://etherpad.openstack.org/p/ironic-liberty-priorities11:06
devanandathat is where we're sketching ideas about what to do in Liberty11:06
lintanOK, I get it11:06
*** dtantsur is now known as dtantsur|bbl11:18
* rameshg87 goes home11:18
*** rameshg87 has left #openstack-ironic11:18
*** jamielennox is now known as jamielennox|away11:45
* lucasagomes spent 1 hour to find out that it was a cache problem, the new image was in glance but ironic was referring to an old image that only existed in the cache :( :( :(11:56
lucasagomesnot sure, should we check if deploy_{kernel,ramdisk} still exist in glance ? And only them get it from the cache?11:57
lucasagomes:/11:57
ShrewsGood morning Ironicers12:03
*** Haomeng|2 has joined #openstack-ironic12:05
TheJuliagood morning12:06
*** Haomeng has quit IRC12:08
lucasagomesTheJulia, Shrews good morning12:08
*** chlong has joined #openstack-ironic12:12
*** chlong has quit IRC12:15
*** lucasagomes is now known as lucas-hungry12:15
*** kkoski has joined #openstack-ironic12:17
*** kkoski has quit IRC12:21
*** jcoufal has joined #openstack-ironic12:22
*** athomas has joined #openstack-ironic12:23
*** dtantsur|bbl is now known as dtantsur12:26
*** thiagop has quit IRC12:28
*** jistr is now known as jistr|class12:30
*** chlong has joined #openstack-ironic12:32
*** chlong has quit IRC12:35
*** chlong has joined #openstack-ironic12:36
*** ChuckC has quit IRC12:37
*** chlong has quit IRC12:39
*** Marga_ has joined #openstack-ironic12:42
*** dlpartain has joined #openstack-ironic12:43
*** dlpartain has left #openstack-ironic12:43
*** dprince has joined #openstack-ironic12:44
*** chlong has joined #openstack-ironic12:56
*** kkoski has joined #openstack-ironic12:56
*** subscope has quit IRC12:58
*** ChuckC has joined #openstack-ironic13:01
devanandalucas-hungry: did glance image have new uuid, or was it new file but same image uuid?13:08
devanandaShrews, TheJulia, g'morning13:09
openstackgerritZhenguo Niu proposed openstack/ironic: Add a return after saving node power state  https://review.openstack.org/16720813:19
*** pas-ha has quit IRC13:20
*** Marga_ has quit IRC13:21
*** rameshg87 has joined #openstack-ironic13:27
lucas-hungrydevananda, glance had a new UUID the old UUID was deleted from glance13:29
lucas-hungrybut it was still registered in the node as pxe_deploy_ramdisk/kernel13:29
*** lucas-hungry is now known as lucasagomes13:29
lucasagomessince the old uuid was in the cache13:29
lucasagomesit was using it13:29
devanandalucas-hungry: gotcha. but node instance_info was updated?13:29
lucasagomesthe pxe_deploy options are in the driver_info13:30
devanandaer, right13:30
lucasagomesso it wasn't updated, after I deleted the cache13:30
lucasagomesit started failing because the image wasn't found13:30
lucasagomesso I updated the references and it works13:30
lucasagomescache invalidation is hard I know heh13:30
devanandaseems like a bug, heh13:30
lucasagomesit was my bad, but took me some time to figure that out13:30
lucasagomesyeah13:30
lucasagomesprobably, if it's a glance UUID we should check that out13:31
lucasagomesI mean, if pxe_deploy options point to a glance uuid (now it's possible to point to any http url, or file in the fs) we maybe should check out if that UUID is still available in glance before getting it from the cache13:32
*** ChuckC has quit IRC13:32
devanandaTheJulia: random question - wdyt of bifrost's ironic-deploy using dynamically generated inventory (ie, fetch from ironic, rather than from csv) ?13:32
*** pas-ha has joined #openstack-ironic13:35
TheJuliadevananda: I like the idea, although I'm not sure the use case would prove to be very useful in that case.  If I'm managing a fleet of machines, I could see myself wanting to provision a known set of them, deploy to a known set of them, but not all unprovisioned or all machines in inventory.  Naturally, with the right requirements and use cases though, it could be very useful13:37
NobodyCamgood morning Ironic-y folk13:37
TheJuliaGood morning NobodyCam13:37
devanandaTheJulia: indeed. name matching comes to mind :)13:38
NobodyCammorning TheJulia devananda lucasagomes13:38
jrolldevananda: flowchart lgtm, just missing create port and "plug vifs"13:38
devanandag'morning, NobodyCam13:38
lucasagomesNobodyCam, good morning :)13:38
NobodyCam:)13:38
lucasagomesjroll, good morning13:38
jrollmorning devananda NobodyCam TheJulia and lucasagomes :)13:38
devanandajroll: meh. good enough then13:38
NobodyCammorning jroll13:38
TheJuliadevananda: true... search="row5cabinet4"13:38
TheJulia(if that was in the name)13:39
jrollTheJulia: so yeah... parallel tests. since dhcp_factory.DHCPFactory._dhcp_provider is a global-ish, singleton-ish sort of thing... that's going to get racy13:40
jrollwould love someone else to sanity check me on that though13:40
jrollhttps://review.openstack.org/#/c/166903/4/ironic/tests/dhcp/test_factory.py13:40
*** romcheg has quit IRC13:40
TheJuliajroll: that makes sense, I think.  I need more coffee to spend cycles on it.13:41
*** lazy_prince is now known as killer_prince13:41
jrollindeed :)13:41
NobodyCamrameshg87: happen to be around?13:43
TheJuliadevananda: realistically, it would have to be a dynamic inventory addition... and I just thought of two modules to add.13:44
* TheJulia adds another card to her wall13:44
*** r-daneel has joined #openstack-ironic13:44
rameshg87NobodyCam: hello13:45
rameshg87morning13:45
NobodyCammorning rameshg87 :) I just had a question you */might/* know the answer to13:46
*** mtanino has joined #openstack-ironic13:46
rameshg87NobodyCam: sure ..13:46
NobodyCamdo you if anyone is activly workong on https://bugs.launchpad.net/ironic/+bug/1432498 ?13:47
openstackLaunchpad bug 1432498 in Ironic "Whole disk image feature is broken in UEFI environment for pxe drivers" [High,Confirmed]13:47
rameshg87NobodyCam: stendulker from my team had found this is happening13:47
NobodyCamrameshg87: ya I was looking at https://launchpad.net/ironic/+milestone/kilo-rc113:48
NobodyCamand that one of the bugs that didn't have a assignee13:48
rameshg87NobodyCam: he found the whole disk image support had broken both partition-image and whole-disk-image13:48
rameshg87NobodyCam: but he has fixed the partition-image one13:48
rameshg87NobodyCam: but he is still investigating, and havn't found out the right fix for whole-disk-image thing13:48
NobodyCamcan I tag him as the assignee for the bug?13:49
rameshg87NobodyCam: yeah i think you can ..13:49
rameshg87NobodyCam: if he isn't taking it, one of us from the ilo driver team will try to take it up from him13:49
NobodyCam:) I'll do that and dbl check with him when I see him online :)13:50
TheJuliadevananda: thinking about it, we would need to store enough metadata in ironic to build configdrives, else we would have to build them in advance.... many possibilities and issues.13:50
rameshg87NobodyCam: sure, thanks13:51
NobodyCam:) thank you rameshg87 :)13:51
rameshg87and NobodyCam thanks for the udevsettle thing, it worked :) (i hope you saw that)13:52
*** jistr|class is now known as jistr13:53
jrollI like this page (the full book is great too) https://github.com/csabapalfi/release-it/blob/master/2-stability_antipatterns.md13:53
NobodyCamrameshg87: awesome... thou I saw something from JayF late lastnight about that13:53
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Add local boot with partition images documentation  https://review.openstack.org/16678813:53
NobodyCamrameshg87: see the bottum of http://eavesdrop.openstack.org/irclogs/%23openstack-ironic/%23openstack-ironic.2015-03-23.log13:55
*** achanda has joined #openstack-ironic13:55
NobodyCambottom even :-p13:56
* rameshg87 checks13:56
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Switch to oslo.config  https://review.openstack.org/16328113:57
rameshg87NobodyCam: oh13:57
rameshg87NobodyCam: so introduce back sleep again :)13:58
rameshg87NobodyCam: for the kernel events to get in ?13:58
rameshg87NobodyCam: but i tried 2-3 times on that hardware in which it failed earlier - it was going through13:58
rameshg87NobodyCam: yeah may be it depends upon how fast the kernel recieves those events (may be it happened 3-4 times properly on my system)13:59
NobodyCamrameshg87: ya13:59
jrolljay followed up on that and said we probably won't hit that race13:59
rameshg87jroll: was it to us ? :)13:59
NobodyCambut maybe best to check with JayF if you see him, as he was still looking in to things lastnight13:59
rameshg87NobodyCam: okay13:59
jrollrameshg87: yes14:00
NobodyCamjroll: ++ very true he did say that14:00
jroll00:06:31            JayF | NobodyCam: talking to them now, doesn't seem like we're likely to be bitten by the exact race he was talking about14:00
jroll00:06:46            JayF | NobodyCam: he said it impacts /dev/disk/by-uuid/* links more than anything else14:00
jroll00:06:54            JayF | i.e. it says it's done but that link isn't there yet14:00
NobodyCamoh that was in todays log14:00
NobodyCamlol14:00
*** achanda has quit IRC14:00
NobodyCamjroll: no word on WhiteBoard?14:03
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Switch to oslo.config  https://review.openstack.org/16328114:04
rameshg87anyone has some opinion on this - http://lists.openstack.org/pipermail/openstack-dev/2015-March/059738.html ?14:07
rameshg87i think we never discussed actively on what needs to be done - except in one of the irc meetings14:08
rameshg87it didn't get anywhere at that time, and we were running after kilo items14:08
rameshg87since kilo is almost settled - i think we have got to think of what can be done for this - perhaps worth taking in summit if it goes that far ..14:09
jrollNobodyCam: I haven't heard anything, no14:09
NobodyCamI'm checking now14:09
lucasagomesrameshg87, chassis capabilities?14:09
rameshg87lucasagomes: yeah14:09
lucasagomesrameshg87, afaict, we all agree that chassis can offer some capabilities and that's because it was first introduced in our API14:09
lucasagomeswe just haven't used it yet14:09
lucasagomesI mean, coded14:10
rameshg87lucasagomes: yeah so can we use it ?14:10
rameshg87lucasagomes: i know many hardware have so many things that can be done through chassis14:10
lucasagomesyeah, well I think so14:10
lucasagomesI don't see why we wouldn't use it14:10
rameshg87lucasagomes: if people are ready to do it - why no allow chassis operations atleast through a driver-vendor-passthru14:10
lucasagomesit's already in the API and won't be deleted on v1/14:10
lucasagomesso14:10
lucasagomesrameshg87, but we do, no?14:11
lucasagomesI mean I don't anything against anyone using it14:11
rameshg87lucasagomes: okay14:11
lucasagomesI even remember dtantsur suggesting to use it when someone wanted to use a script that was doing some chassis operations14:11
rameshg87lucasagomes: i assume we don't have anyone tried to use chassis in that way, right ?14:11
lucasagomessuggested*14:11
rameshg87lucasagomes: yes14:12
lucasagomesrameshg87, yes, that's what I think as well14:12
rameshg87lucasagomes: okay14:12
lucasagomeswe just didn't have someone to code anything yet14:12
rameshg87lucasagomes: have you looked at https://review.openstack.org/#/c/158577/ ?14:12
lucasagomesrameshg87, I haven't, but that was what dtantsur suggested to use driver passthru on chassis no?14:13
rameshg87lucasagomes: as per the spec, they have come up with a script (which uses a sort-of-pluggable-api)14:13
dtantsurmorning to everyone are missed today :)14:13
NobodyCammorning dtantsur :)14:13
lucasagomesdtantsur, morning14:13
rameshg87lucasagomes: yeah, may be a switch to driver-level-vendor-passthru might be better idea ?14:14
dtantsurlucasagomes, I think it was me, because I pretty much hate adding functionality as scripts (especially with my Red Hat on :)14:14
rameshg87dtantsur: o/14:14
jrollwe have a driver vendor passthru :)14:14
lucasagomesdtantsur, +114:14
lucasagomesrameshg87, +114:15
lucasagomesand being on driver level passthru we can even add tests to it14:15
jrolldtantsur: you mean your red fedora?14:15
lucasagomesunittests*14:15
rameshg87lucasagomes: yeah14:15
dtantsurjroll, definitely :)14:15
* lucasagomes looks at his fedora on the wall :)14:15
lucasagomeswith all that being said about scripts, the config drive script (https://review.openstack.org/#/c/166201/) do you guys think it would be better to be part of the ironic CLI?14:17
* dtantsur shouts YES14:17
lucasagomesok... I will put a patch up for that to see how it looks like14:19
dtantsurlucasagomes, I think we already have a copy-paste of the same thingy in rdo-manager, right? I'd wish we could just rely on the client...14:19
lucasagomesdtantsur, do we?14:19
dtantsurnot sure14:19
lucasagomesidk about rdo-manager. We use nova there14:19
rameshg87lucasagomes: if it was in cli, the node-set-provision-state could accept the directory containing config drive files, no ?14:19
rameshg87:)14:19
lucasagomesso we don't generate it standalone like that14:19
dtantsurah maybe14:19
*** BadCub_Hiding is now known as BadCub14:19
dtantsurrameshg87, that's the idea14:19
lucasagomesrameshg87, yeah that's the idea14:19
BadCubMornin folks14:20
NobodyCamgood mornign BadCub14:20
dtantsur:)14:20
rameshg87okay14:20
dtantsurBadCub, o/14:20
lucasagomesBadCub, good morning14:20
lucasagomesdtantsur, rameshg87 you guys think that using the same --configdrive option but pointing to a dir would do it?14:20
lucasagomesor would be better to have a separated one?14:20
lucasagomes--generate-configdrive <dir>14:20
lucasagomessomething like that14:20
* lucasagomes votes for a separated one14:20
rameshg87lucasagomes: --configdrive-dir14:21
rameshg87yeah14:21
rameshg87a separate one14:21
lucasagomescool14:21
lucasagomesokie14:21
dtantsurlucasagomes, I don't have a strong opinion on this14:21
NobodyCambrb14:24
*** achanda has joined #openstack-ironic14:32
* NobodyCam is back14:32
rameshg87i save the state of my devstack vm and restore it back when i want to work again - sometimes all the services are terribly slow14:35
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Switch to oslo.config  https://review.openstack.org/16328114:35
rameshg87anyone faced the issue ? any suggestions to try ?14:35
rameshg87do you guys save the state of your devstack vm while going off-work ?14:36
rameshg87*if you run stack on your laptop*14:36
dtantsurrameshg87, I did some time ago. Usually slowness disappears after some time.14:36
rameshg87dtantsur: exactly14:36
lucasagomesrameshg87, I have some local env but I shutdown my vms14:36
devanandajroll: the ascii for that graphic: https://github.com/devananda/talks/blob/master/ascii/agent-flow.txt14:37
rameshg87dtantsur: for the first one hour or so, it's terribly slow14:37
dtantsurifarkas, lucasagomes, may I ask you both to review this change to discoverd: https://review.openstack.org/#/c/163281/ ? It's pretty big and dangerous14:37
lucasagomesthen later I just did a ./rejoin-stack (or something likt that)14:37
lucasagomesyou gotta set script /dev/null first too14:37
jrolldevananda: neat14:37
ifarkasdtantsur, sure thing14:37
devanandalucasagomes: any chance you'll happen to be in London on Sunday? :)14:37
dtantsurrameshg87, an hour? hmm... I think it's less for me. I however never run stuff on the laptop directly, usually in the vm14:37
BadCubg'mornin devananda14:37
lucasagomesdtantsur, sure will take a look14:38
rameshg87dtantsur: okay14:38
devanandamorning, BadCub14:38
lucasagomesdevananda, :( not really r you going to be there?14:38
devanandalucasagomes: yah. in Bristol mon-tue, but going to spend sunday hanging out in london14:38
pshigemorning14:38
dtantsuro/14:39
NobodyCammorning pshige14:39
BadCubg'mornin pshige14:39
lucasagomesdevananda, oh sounds pretty good. I would need to catch a flight to go there tho :/14:39
devanandaugh :(14:40
lucasagomesyeah different islands :(14:40
jrollexpense it, it's a business meeting right?14:40
jroll:)14:40
*** pas-ha has quit IRC14:40
dtantsurlol14:40
lucasagomeshah14:40
lucasagomeslondon is a great city anyway, I should visit it again. I don't even remember the last time I've been there, maybe 201214:42
*** achanda has quit IRC14:42
*** stendulker has joined #openstack-ironic14:42
dtantsurlucasagomes, you should visit Brno at least once this year :)14:43
*** mjturek1 has quit IRC14:43
lucasagomesdtantsur, +1 man, it's been a year already14:43
lucasagomesmarch 17 I remember because of the st patricks14:43
dtantsurkind of, yeah. I just started at RH :)14:43
lucasagomesdtantsur, nice! yeah, brno is great! zebras + beers :)14:44
dtantsur++14:44
lucasagomeszebras == ribs in czech14:44
dtantsuractually žebras (if IIRC understands this)14:44
dtantsur* IRC14:45
lucasagomesyeah, well I don't know how to put that thing hah14:45
* BadCub heads off to get ready to go get the beloved Bumble Bee from it's long vacation from home14:45
lucasagomesdtantsur, btw all the config options for discoverd are on that conf.py file right?14:46
dtantsurlucasagomes, except for edeploy plugin14:46
lucasagomescause that was the reason why we didn't port ironic to olso_config yet, since the config options are all around the code (and it's good cause it's self contained on each module)14:46
lucasagomeswe didn;t want to create one file where all the references should live14:46
lucasagomesdtantsur, hmm I see14:46
dtantsurlucasagomes, well, discoverd is MUCH smaller and has much less plugins, so we can afford having all options in our place14:47
lucasagomesyeah14:47
*** mjturek1 has joined #openstack-ironic14:48
jrolllucasagomes: wait, does oslo_config require all the options to be declared in one place?14:48
dtantsurjroll, not necessary, but it requires a way to list them all14:48
lucasagomesjroll, it requires you to return a list of all options14:48
dtantsurjroll, discoverd has always had it's options in one place14:48
jrolloh. ugh.14:48
lucasagomesdifferently from the old config generation that would look at the code14:49
jrollright14:49
lucasagomesjroll, ceilometer is using oslo_config take a look14:49
lucasagomesjroll, https://github.com/openstack/ceilometer/blob/master/ceilometer/opts.py14:49
lucasagomesx.x14:49
dtantsurjroll, better take a look at discoverd patch :D14:49
lucasagomesyeah or that ^14:49
* dtantsur is good at shamelessly make people review code14:49
jrolloh my god14:49
* jroll has a sad14:50
jrolldtantsur: you clearly aren't very good or you would have provided a link :)14:50
lucasagomesjroll, one benefit of oslo_config is that you can have different config files for different services14:50
NobodyCamlol14:50
lucasagomessay ironic-api and ironic-conductor for e.g14:50
* dtantsur has failed :(14:50
* BadCub is laughing14:51
*** EmilienM|PTO is now known as EmilienM14:51
dtantsurjroll, https://review.openstack.org/#/c/163281/14:51
jrolllucasagomes: meh14:51
* jroll /ignore dtantsur14:51
lucasagomesjroll, :) yeah just saying heh14:51
lucasagomesthat's why we don't use in ironic14:51
lucasagomesmaintaining a hard-coded list of opts is a bit too much14:52
*** pas-ha has joined #openstack-ironic14:52
dtantsurespecially with a lot of drivers14:52
jrollwell, it's more of a hardcoded list of modules14:53
jrollbut yeah.14:53
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Switch to oslo.config  https://review.openstack.org/16328114:57
dtantsurlucasagomes, ifarkas, quick fix, sorry ^^^14:57
jlvillalGood morning Ironic.  I saw far too many of you online at 6:30am Pacific Time when I woke up ;)14:57
jlvillallucasagomes: Gblame? :)14:58
dtantsurjlvillal, morning :)14:58
jlvillaldtantsur: dobré odpoledne15:00
dtantsur:)15:00
lucasagomesjlvillal, awesome Gwrite for git add everytime now15:00
*** ChuckC has joined #openstack-ironic15:00
NobodyCammorning jlvillal15:01
jlvillallucasagomes: Cool.  I should use that more (Gwrite).  I mostly just use Gblame with the going back in time feature.15:01
jlvillalNobodyCam: Good morning.  You were one of the early risers I was talking about :)15:01
BadCubg'mornin jlvillal15:01
NobodyCamjlvillal: lol yeppers15:02
jlvillalBadCub: :)  Thanks15:02
lucasagomesjlvillal, yeah Gblame is very good too, I just played a bit with it. Didn't need to go back in history yet15:03
lucasagomesbut I'm sure I will use it soon15:03
NobodyCamanyone happen to know why Rakesh abandoned https://review.openstack.org/#/c/151268 ?15:03
jlvillallucasagomes: You can use '~' or '^' for going back in history15:03
jlvillallucasagomes: and 'o' is nice to actually see the git commit of the line you are on.15:04
BadCubokay, now I am really leaving. Have a good day folks. :-)15:04
jlvillalBadCub: Bye.  Are you outside the US today?15:04
NobodyCamjlvillal: nope15:05
lucasagomesdtantsur, why you need to import conf on the main.py?15:05
lucasagomesjlvillal, +115:05
NobodyCamBadCub's car has been in the shop for like 6 keeks15:05
NobodyCamweeks even15:05
dtantsurlucasagomes, otherwise the options are now know on startup15:06
dtantsurlucasagomes, e.g. trying to check CONF.discoverd.debug fails15:06
lucasagomesoh15:06
jlvillalNobodyCam: BadCub :Ouch!  That is a long time.15:06
lucasagomesgotcha15:06
*** achanda has joined #openstack-ironic15:06
NobodyCam:-p ya15:06
*** achanda has quit IRC15:06
*** hj-hp has joined #openstack-ironic15:07
jrollNobodyCam: what car?15:07
NobodyCamthe focus, Ford had to replace the motor15:08
jrolloh yeah, didn't a tune blow it or something?15:08
NobodyCamone of the connectiong rods gave out :(15:09
NobodyCam(while getting a new exaust)15:10
jrolloh huh15:10
* jroll has a bad memory then15:10
NobodyCamit will be nice to rid our selfs of the rental thing15:11
NobodyCamlol then we just have to get my car down from WA15:13
jrollheh15:13
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Update from global requirements  https://review.openstack.org/16726615:14
*** ijw has joined #openstack-ironic15:19
openstackgerritDmitry Tantsur proposed openstack/python-ironicclient: Clean openstack-common module list  https://review.openstack.org/16279915:19
*** ijw has quit IRC15:20
*** Marga_ has joined #openstack-ironic15:22
NobodyCamoh TheJulia have you seen: https://github.com/dominis/ansible-shell15:22
*** romcheg has joined #openstack-ironic15:22
*** ijw has joined #openstack-ironic15:22
jrollwhoa15:23
jrollneat.15:23
*** viktors has quit IRC15:23
jrollI like it15:23
NobodyCam:)15:23
*** ijw has quit IRC15:23
*** ijw has joined #openstack-ironic15:24
*** ijw has quit IRC15:25
*** ijw has joined #openstack-ironic15:26
*** yog__ has quit IRC15:29
rameshg87folks any idea if we can use logs.openstack.org for posting temporary results15:34
rameshg87i mean for a third-party-ci15:34
* rameshg87 should rather check in infra15:34
NobodyCamya I'd infra is the best place to ask that one :)15:35
NobodyCami'd THINK infra ...15:35
NobodyCam:-p15:35
*** BertieFulton has joined #openstack-ironic15:35
rameshg87okay15:38
NobodyCambrb15:38
openstackgerritJay Faulkner proposed openstack/ironic: Ensure configdrive isn't mounted in CoreOS ramdisks  https://review.openstack.org/16706315:40
*** absubram has quit IRC15:40
* JayF fixed tests and ^ is RFR15:42
*** anderbubble has joined #openstack-ironic15:42
dtantsurjroll, JayF, mind having a look at https://review.openstack.org/#/c/164440/ ? touches agent deploy15:44
JayFI will right meow15:44
dtantsurmeow-meow :)15:45
JayFI was thinking we landed support in the agent for partition images?15:45
JayFOr am I just conflating that with the iscsi+agent support for partition images15:45
dtantsurI think it was bumped, at least the Ironic part...15:45
JayFah, okay15:45
dtantsurthe spec has landed https://review.openstack.org/#/c/137363/15:46
JayFdtantsur: how do we know if an image is whole disk or not?15:46
dtantsurJayF, there's some capability/property for that IIRC15:46
JayFIt seems like that attribute is already set somehow in driver_internal_info15:46
dtantsuraha the patch got bumped: https://review.openstack.org/#/c/162008/15:47
JayFI'm mainly curious about how the driver_internal_info value gets set15:48
JayFmaking sure we aren't adding a dependency on a formerly optional property to an image or something similar15:48
* dtantsur does not remember...15:48
jrollI assume it looks at glance metadata?15:50
jrollor actually iirc15:50
jrollit's the lack of kernel/ramdisk on the glance image15:50
JayF+1, assuming confidence in where that data comes from15:50
*** hj-hp has quit IRC15:51
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Add root device hints documentation  https://review.openstack.org/16680115:51
*** bradjones has quit IRC15:51
dtantsurjroll, leaving https://review.openstack.org/#/c/164440/ for you to approve?15:51
lucasagomesjroll, +1 yeah I would assume that as well15:51
*** bradjones has joined #openstack-ironic15:51
*** bradjones has quit IRC15:51
*** bradjones has joined #openstack-ironic15:51
JayFdtantsur: fwiw jroll is the only one of the j's who can approve a thing in Ironic :P15:53
jrolldtantsur: not sure why you need my approval >.>15:53
jrollbut I'll look15:53
dtantsurJayF, aha, but you're a core on -specs, aren't you?15:54
* jroll +A15:54
JayFdtantsur: I'm core on specs/IPA but not Ironic15:54
JayFdtantsur: JoshNang is core on IPA but not specs/ironic15:54
dtantsuroh15:55
JayFUgh, my ability to find etherpads16:01
JayFis greatly diminished with the death of the whiteboard16:01
jrollrecovery is in progress16:01
jrollplease hold16:02
JayFI've listened to this same 2 minute elevator music song16:02
JayFfor two days16:02
JayF:P16:02
jrollwell16:02
jrollfungi found it16:02
lucasagomeshttps://wiki.openstack.org/wiki/Ironic#Etherpads (we should really add the etherpads there)16:02
jrolljust needs to restore16:02
NobodyCamya fungi is working on it16:02
*** EmilienM is now known as EmilienM|PTO16:02
JayFBadCub / Others https://review.openstack.org/#/c/167063/ + https://bugs.launchpad.net/ironic/+bug/1433812 need to get on the review day etherpad16:02
openstackLaunchpad bug 1433812 in Ironic "ironic-python-agent silently fails to write a configdrive if a previous drive is found" [High,In progress] - Assigned to Jay Faulkner (jason-oldos)16:03
JayFcan I just add it or should someone be prodded?16:03
jrolljfdi16:03
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Add option to delete ports after introspection  https://review.openstack.org/16691516:03
jrollman, I really want to put a module called requirements.txt on pypi16:03
jrollI always forget the -416:03
NobodyCamjroll: +++ jdi!16:03
jroll-r16:03
jrolland make it just download a script that executes pip install -r requirements.txt16:04
*** romcheg has quit IRC16:05
*** Marga_ has quit IRC16:05
*** jistr has quit IRC16:05
*** romcheg has joined #openstack-ironic16:06
lucasagomesjroll, create an alias?16:06
jrolllucasagomes: nah, this way everyone else gets the benefits :)16:06
lucasagomesheh fair enuff16:06
jrollalso. way too many servers for an alias. :P16:06
NobodyCamJayF: +2'd16:08
tiagogomes_hi, can someone clarify what are the network requirements for deploying with Ironic?  I have installed Ironic in a VM, and the VM is bridged with the host network, which correspods to the office network. I am trying to deploy to a baremetal server that is on the same office network, but I think that the baremetal server is trying to DHCP from the office DHCP server, instead of Neutron's DHCP server.  It is mandatory t16:11
tiagogomes_hat the VM and the baremetal server are in a network without an existing DHCP server?16:11
*** penick has joined #openstack-ironic16:11
jrolltiagogomes_: in general, if there's more than one DHCP server on a network, you're going to have problems16:12
NobodyCamya16:12
NobodyCamyou could exclude the nodes mac from the existing dhcp server but thats not the path to start down16:13
tiagogomes_ok, then Neutron doesn't create any virtual network or vlan with the baremetal server?16:14
JayFYou'd have to have neutron configured to control a physical switch for that to even be possible (not many folks do)16:16
*** chlong has quit IRC16:16
JayFbut support for that didn't land in Ironic for K, I believe16:16
rameshg87folks 1 question about our basic tempest test16:16
rameshg87https://github.com/openstack/tempest/blob/master/tempest/scenario/test_baremetal_basic_ops.py#L129-L13016:17
rameshg87i think we should make it adding floating ip and testing connectivity with that should really be a configurable option16:17
rameshg87wdyt ?16:17
NobodyCamrameshg87: that seems like a basic operation, what is the use case behind making it optional16:18
rameshg87NobodyCam: can we do configuration of floating ips on a provider flat network ?16:19
* rameshg87 might start asking stupid questions 16:20
rameshg87NobodyCam: in our ci, i have the hardware lying around in a flat network16:20
*** stendulker has quit IRC16:20
rameshg87NobodyCam: i just configured devstack for a single provider network, stacked and then ran the test_baremetal_server_ops test16:20
tiagogomes_JayF do you have any suggestion of how I can try the deployment then?16:22
rameshg87NobodyCam: i think i should run with an l3 agent if i want to have floating ips16:22
JayFtiagogomes_: you can use the agent deployment method with DHCP provider of none16:22
rameshg87NobodyCam: that's really a neutron thing - i would like to make it optional :(16:22
JayFtiagogomes_: and configure your DHCP server to boot the deploy ramdisk16:22
rameshg87NobodyCam: does that make sense ?16:22
* JayF bbl16:23
tiagogomes_JayF I can't do that, I don't want to mess with the office DHCP server16:23
NobodyCamrameshg87: is this for something like 3rd party testing?16:23
rameshg87NobodyCam: yes16:23
JayFtiagogomes_: You can't run Ironic without being able to control the DHCP server yourself or without letting Ironic control it16:23
JayFtiagogomes_: the only exception is the iLo drivers that use Virtual Media to transmit the deploy ramdisk, but that only works with some HP hardware16:24
NobodyCamack then i would okay with it. because in that case your just looking for the drivers results16:24
rameshg87NobodyCam: yes16:24
NobodyCamrameshg87: ++16:24
NobodyCammakes sense to me16:24
rameshg87NobodyCam: if i just verify the connectivity with my private ip, the test runs fine16:24
rameshg87NobodyCam: just like we do after rebuild - https://github.com/openstack/tempest/blob/master/tempest/scenario/test_baremetal_basic_ops.py#L14816:25
NobodyCamoh brb... bbt16:26
jlvillalI just saw this: https://bugs.launchpad.net/bugs/143594816:27
openstackLaunchpad bug 1435948 in Ironic "Incorrect assertEqual statements in unit tests" [Low,Triaged]16:27
lucasagomesdevananda, ping re deploy kernel/ramdisk on the flavor16:27
lucasagomesdevananda, https://github.com/openstack/nova/blob/master/nova/virt/ironic/patcher.py#L158-L160 before we were thinking about removing this functionality, but now, after the ipa-as-default-ramdisk I think it makes sense to keep it16:27
jlvillalkkoski: Are you going to work on https://bugs.launchpad.net/bugs/1435948 ?  If not, I could work on it.16:28
lucasagomessince we can use the same ramdisk for multiple drivers, it would make sense to be able to specify the deploy_{kernel, ramdisk} in the flavor16:28
lucasagomesand we normalized the parameter names too, so you don't need to specify the "pxe_" namespace anymore16:28
*** absubram has joined #openstack-ironic16:29
lucasagomescan pass deploy_{kernel, ramdisk}=<id> for both agent and pxe drivers16:29
jrollwhy not just specify on the node?16:30
lucasagomesjroll, you could do both16:30
lucasagomesthe driver respects it16:30
jrollyeah16:30
lucasagomesI mean, if it's in the flavor the driver would write it to the node16:30
lucasagomesif it's not in the flavor the driver just don't do anything16:31
lucasagomesjroll, makes sense?16:31
jrollyeah, just think it's odd to have two ways to configure the same thing16:31
jrollslash confusing16:31
lucasagomesright16:31
NobodyCamjlvillal: thats been open for a while I beleieve16:31
jrollupdate node.driver_info, why isn't it using my new ramdisk16:32
lucasagomesyeah I see that, that's why I'm asking. If it's confusing we should remove it then16:32
jrolloh it's on the flavor16:32
lucasagomesI just found it a bit useful to have in the flavor16:32
lucasagomessince it's a "global" way to set a deploy ramdisk16:32
jrollidk, it would be confusing to me16:32
jrollI guess, idk, that's what automation is for :)16:32
rameshg87dtantsur: please have a look at this bug: https://bugs.launchpad.net/ironic/+bug/143595916:32
openstackLaunchpad bug 1435959 in Ironic "rebuild fails with ilo* drivers" [Undecided,New]16:32
lucasagomesjroll, right16:32
lucasagomesok better remove from flavor then16:32
rameshg87dtantsur: please prioritize as high and for rc1. a critical bug in ilo driver :(16:33
lucasagomeswe should have removed it this cycle16:33
rameshg87fix rolling right-away16:33
lucasagomesthat was kept to keep compability with icehouse16:33
jrollwell16:33
dtantsurrameshg87, I can bump it, but please do the triaging: I can't confirm bug in ilo16:33
dtantsurrameshg87, "triaging" meaning "finding the cause" here16:33
jrolllucasagomes: I think I tried to, but we didn't deprecate it properly. that may have been juno though.16:33
*** ukalifon has quit IRC16:34
rameshg87dtantsur: i have tested the fix and patch rolling in a minute16:34
lucasagomesjroll, heh yeah16:34
lucasagomesnp we remove it from nova in L16:34
dtantsurrameshg87, please update the bug description too, so that I can mark it as triaged16:34
jrolllucasagomes: we still could remove it this cycle16:34
lucasagomesjroll, will nova accept a patch for that?16:34
rameshg87updating16:34
lucasagomesif so I can put a patch up removing it16:34
lucasagomeswell I put a patch up and then they can merge it this cycle or the next16:34
jrolllucasagomes: idk, if they don't then the patch waits til L16:35
lucasagomesdoesn't matter really16:35
jrollyeah16:35
lucasagomescool ok will do that16:35
lucasagomesjroll, thanks :016:35
lucasagomes:)16:35
jrollnp :)16:35
jlvillalNobodyCam: The email I got said "NEW", maybe that is the state?16:36
rameshg87dtantsur: done16:36
dtantsurthnx16:36
TheJuliaJayF: fyi, I've been using a fairly recent coreos ipa image from yesterday evening and have my confgdrives generally seem to be writing out without any issues, although I'm still passing the coreos.configdrive=0 setting by default.16:41
openstackgerritRamakrishnan G proposed openstack/ironic: Attach ilo_boot_iso only if node is active  https://review.openstack.org/16731716:42
openstackgerritRamakrishnan G proposed openstack/ironic: Attach ilo_boot_iso only if node is active  https://review.openstack.org/16731716:43
NobodyCamjlvillal: oh ya that one is new16:44
NobodyCam:-p16:44
jlvillalNobodyCam: :)16:45
*** ijw has quit IRC16:46
NobodyCamTheJulia: JayF also put up this https://review.openstack.org/16706316:46
TheJuliaI spotted that earlier this morning :)16:47
TheJuliaor was it last night16:47
TheJuliaheh16:47
NobodyCam:)16:48
rameshg87folks please have a look at https://review.openstack.org/167317 when you get some time - a small fix but a pretty important one :)16:49
rameshg87good night ironic, see you tomorrow16:49
NobodyCamnight rameshg8716:49
TheJuliagoodnight16:49
*** rameshg87 has quit IRC16:49
*** spandhe has joined #openstack-ironic16:49
*** bandicot has joined #openstack-ironic16:50
* TheJulia crosses her fingers because she just turned on cleaning16:50
NobodyCamTheJulia: woo ... /me holds breath16:50
*** vdrok is now known as vdrok_afk16:51
JoshNang:O16:51
*** hyakuhei has joined #openstack-ironic16:52
* TheJulia taps her foot waiting for the node to finish initial deployment JUST so she can delete the node16:52
*** Nisha has joined #openstack-ironic16:57
TheJuliawell, so far so good16:57
jlvillalNobodyCam: Darn, fixing the tests found one failed test :(17:00
TheJuliaAn ipa.debug option would rock :)17:00
kkoskijvillal: Are you talking about https://bugs.launchpad.net/ironic/+bug/1435948?17:00
openstackLaunchpad bug 1435948 in Ironic "Incorrect assertEqual statements in unit tests" [Low,Triaged] - Assigned to John L. Villalovos (happycamp)17:00
jlvillalkkoski: Yep17:00
kkoskikk17:00
jlvillalkkoski: I commented in the bug about failure.  Investigating now.17:00
kkoskiI was just about to take a look at fixing it, but it sounds like you've got it handled17:01
dtantsurg'night folks, see you tomorrow17:01
*** dtantsur is now known as dtantsur|afk17:01
NobodyCamnight dtantsur|afk17:01
jlvillalkkoski: I pinged you earlier :)17:01
kkoskijlvillal: Yeah sorry, I reported it, then grabbed some lunch :P17:01
JoshNangTheJulia: agreed. after fighting with devstack and ipa last week, i have a couple things i want to fix17:02
openstackgerritJohn L. Villalovos proposed openstack/ironic: WIP: Fix broken unittests usage of sort()  https://review.openstack.org/16732517:02
jlvillalkkoski: ^^^  but failing one test.  In case you wanted to see patch so far17:02
TheJuliawow, I think my test VM's disk is only 10G and it is taking a really long time :(17:02
*** bandicot has quit IRC17:03
*** korekhov has joined #openstack-ironic17:05
JoshNangyeahhh it's not quick at all :/17:05
NobodyCamTheJulia: https://www.youtube.com/watch?v=uWvhS2nvqJA17:05
*** bandicot has joined #openstack-ironic17:06
TheJuliaNobodyCam: does it include a blowtorch?17:07
NobodyCamlol nope just a sldge hammer17:07
TheJuliamelting hard disks sound more fun17:07
*** kevinbenton has quit IRC17:07
*** Marga_ has joined #openstack-ironic17:08
NobodyCamTheJulia: but that may not get rid of all the data? :-p http://www.extremetech.com/computing/140975-scientists-accidentally-create-quantum-dots-out-of-lava17:09
*** jmccrory has joined #openstack-ironic17:10
NobodyCamseems we still need a few years of development to be able to read the data but...17:11
NobodyCamlol17:11
TheJuliaFine then.  First we melt, then we drop the ingots into a volcano... or launch them into the daystar.17:11
NobodyCamhehhehe17:12
kkoskijlvillal: I'm taking a look now17:16
jlvillalkkoski: Cool :)17:16
*** romcheg has quit IRC17:17
*** harlowja_away is now known as harlowja17:18
TheJuliaCleaning is now enabled by default in bifrost17:22
jroll\o/17:24
lucasagomesTheJulia, hi, bifrost uses IPA with pxe_ drivers?17:25
lucasagomesand ipxe too right?17:25
jrollboooooooooo pxe17:25
jroll:P17:25
lucasagomesjroll, heh no swift man! :P17:25
jrolldon't need swift for IPA :)17:25
lucasagomesit should work without any other openstack component, I think that's the reason17:25
lucasagomesjrist, with agent?17:26
jrollonly if you use glance17:26
lucasagomesoh17:26
TheJulialucasagomes: ipxe, agent_ssh for testing and agent_ipmitool for 'normal' use17:26
*** ijw has joined #openstack-ironic17:26
lucasagomesyeah true17:26
lucasagomesTheJulia, oh gotcha17:26
lucasagomescool17:26
lucasagomesjroll, ^ so it's agent_ see :D17:26
*** mjturek1 has quit IRC17:26
jrollyep I know :)17:26
jrollso uh, who wants to paste our whiteboard back in and format it?17:26
jrollhttp://paste.openstack.org/show/196032/17:26
jlvillalkkoski: That one test seems totally broken17:27
*** hyakuhei has quit IRC17:27
jlvillalkkoski: Maybe lucasgomes would know.17:27
TheJuliaI also put a playbook that is essentially a single playbook test, very simple and basic, needs work, but it kind of just works for testing in VM mode.  A little later, I'm going to fire it off on a machine with 8GB of ram and see how it goes. :)17:27
*** achanda has joined #openstack-ironic17:28
NobodyCamLast WhiteBOard text restored17:28
NobodyCamhistory does appear to be lost17:28
TheJuliaIn any event, sufficent coffee has been ingested, time to look at the issue from yesterday17:28
NobodyCambut thanks to fungi in -infra we have the text from the last rev back17:29
JoshNangTheJulia: yay for cleaning!17:29
NobodyCam*may require some reformatting17:29
lucasagomesTheJulia, :) good stuff17:30
jlvillallucasagomes: https://review.openstack.org/167325    gets this error: http://paste.openstack.org/show/196091/17:30
jlvillallucasagomes: I noticed you worked on that code.  Not sure if you have any ideas.17:30
jrolljlvillal: that's ramineni's code17:31
lucasagomesjlvillal, oh... hmmm .sort() is wrong indeed17:31
lucasagomesjlvillal, I would thing you should do sorted(list)17:31
lucasagomeslike17:32
lucasagomesself.assertEqual(sorted(list1), sorted(list2))17:32
* JayF cleaned up a section of the etherpad17:32
jlvillallucasagomes: Yes that would also work.  I was using the unittest builtin method to compare lists via six.17:32
lucasagomesjlvillal, right, not sure what is best17:33
jlvillallucasagomes: But the test is broken.  I'll keep investigating to see if I can figure out why.17:33
lucasagomesjlvillal, fair enuff, try sorted()17:33
jlvillallucasagomes: Also fails: http://paste.openstack.org/show/196094/17:35
lucasagomesright it doesn't fail on the sort tho17:36
lucasagomesit fails because there's one uuid which is not being passed17:36
lucasagomes u'1be26c0b-03f2-4d2e-ae87-c02d7f33c781',17:36
jlvillallucasagomes: Yeah17:36
lucasagomesthat uuid is not present on the first dict17:36
lucasagomesso I assume it's something with the test itself17:36
lucasagomesor code that might be wrong17:36
jlvillallucasagomes: Thanks.  I'll keep looking :)17:36
lucasagomesjlvillal, cool! thank you for fixing it :)17:37
jlvillallucasagomes: It isn't fixed yet ;)17:37
yuriyzhello Ironic jlvillal I commented this already https://review.openstack.org/#/c/167325/1/ironic/tests/db/test_ports.py17:37
lucasagomeswill be soon heh :D17:37
lucasagomesyuriyz, morning17:37
lucasagomesor evening :)17:37
yuriyzmaybe evening :)17:37
jlvillalyuriyz: Thanks.  Either Gerrit is slow sending emails or our mail system has a problem.  So I hadn't seen the email notification.17:38
NobodyCamhi yuriyz :)17:38
jlvillalyuriyz: Thanks!17:38
yuriyzmorning NobodyCam17:38
NobodyCam:)17:38
*** BertieFulton has quit IRC17:41
NobodyCambrb17:43
kkoskiyuriyz: So the test was forgetting about the existing port, and was assuming that it would get back only the 5 ports that it just created?17:43
*** derekh has quit IRC17:45
yuriyzkkoski yes you are right we create one extra before in setUp()17:45
kkoskicool, thanks17:45
kkoskijlvillal: does that take care of the broken test for you?17:45
jlvillalkkoski: Yes.17:46
openstackgerritMerged openstack/python-ironicclient: Use oslo.i18n lib  https://review.openstack.org/16279717:46
kkoskijlvillal: Are you just going to submit a fix then?17:47
kkoskiI'm still trying to set up a dev environment, so that would be fastest :P17:47
jlvillaljlvillal: Yes I will push new patch soon.17:48
jlvillalkkoski: Working on commit message17:48
*** zz_jgrimm is now known as jgrimm17:48
openstackgerritGhe Rivero proposed openstack/python-ironicclient: Clean openstack-common module list  https://review.openstack.org/16279917:49
openstackgerritGhe Rivero proposed openstack/python-ironicclient: Sync oslo.incubator  https://review.openstack.org/16280617:49
*** ijw has quit IRC17:50
openstackgerritJohn L. Villalovos proposed openstack/ironic: Fix broken unittests usage of sort()  https://review.openstack.org/16732517:50
jlvillalkkoski: Pushed ^^^^17:52
NobodyCamjroll: :( re: https://review.openstack.org/#/c/165430/17:53
jrollyep.17:53
kkoskijlvillal: thanks!17:53
NobodyCamTheJulia: hows that clenning doing? still going?17:54
jlvillalkkoski: Thank you for spotting the issue! :)17:54
JayFjroll: or any other Ironic core around; this needs +2A to resolve #1433812 enough for Kilo (even though we should fix it "better" long term) https://review.openstack.org/#/c/167063/17:55
*** tiagogomes_ has quit IRC17:56
jroll+A17:57
lucasagomesjroll, JayF I was too late :( but it would be good to add to ipxe template and elilo template as well17:58
* lucasagomes puts a follow up patch17:58
TheJuliaNobodyCam: oh, it completed fairly quickly17:58
jrolllucasagomes: oh ha, sorry17:58
NobodyCamTheJulia: nice :)17:58
JayFlucasagomes: want me to write the follow up or are you already on it?17:59
lucasagomesJayF, I would start, but if you have it handy there17:59
lucasagomesI don't mind, I know we have too many templates and it sucks17:59
lucasagomesI hope we can make it better in the future17:59
lucasagomes(specially if we start using only 1 ramdisk for deployments)18:00
*** ijw has joined #openstack-ironic18:00
openstackgerritMerged openstack/ironic: Raise exception for Agent Deploy driver when using partition images  https://review.openstack.org/16444018:02
* lucasagomes notices there's no unittests for ipxe and elilo templates (y)18:02
lucasagomessigh18:02
*** Haomeng has joined #openstack-ironic18:02
JayFlucasagomes: heh, that means I'm /extra/ willing to do it18:02
JayFlucasagomes: actually, let me do it18:02
JayFlucasagomes: I can't +2 a thing, and you can18:02
lucasagomesJayF, cool, yeah go ahead18:02
JayF...should I add the tests too then?18:02
JayF:(18:03
lucasagomes:)18:03
lucasagomesplease?18:03
*** bandicot has quit IRC18:03
lucasagomesand I will have to go now :(18:03
NobodyCamnight lucasagomes18:03
NobodyCamJayF: awesome!18:04
lucasagomesI will take a look at the patch later, JayF add me as a reviewer please18:04
lucasagomeshave a good night everyone18:04
JayFI will, goodnight!18:04
*** jcoufal has quit IRC18:04
*** jamielennox|away is now known as jamielennox18:04
*** lucasagomes is now known as lucas-dinner18:04
*** Haomeng|2 has quit IRC18:05
NobodyCamTheJulia: have you seen jroll's comment on 166903?18:05
TheJuliaNobodyCam: yes, working on it right now18:08
*** jmccrory has quit IRC18:09
NobodyCamwoo hoo :)18:09
*** jmccrory has joined #openstack-ironic18:09
clif_hadam_g: JayF said you might know why I can't attach to a devstack's screen instance18:17
adam_gclif_h, hmm. not sure what happens when you try?18:17
JayFman, you don't have to point a big giant arrow at me18:17
adam_g:)18:17
JayFno need to tell adam_g I'm driving the bus :)18:17
clif_hlol18:17
adam_gimag18:18
adam_ger18:18
clif_hscreen -r gives '18:18
clif_h'cannot open terminal - please check'18:18
clif_hthe terminal in question only has root permissions18:18
clif_hand screen -r as root says no session to reattach to18:18
JayFclif_h: part of me wants to suggest you open the perms on that terminal, but I think that'd be gross :)18:18
clif_hidc its just a devstacked machine18:18
adam_gclif_h, huh. no ive never seen that18:19
NobodyCambrb18:19
clif_hchmod did it18:20
*** Marga_ has quit IRC18:24
*** achanda has quit IRC18:28
*** ifarkas has quit IRC18:38
openstackgerritJosh Gachnang proposed openstack/ironic: Allow instance_uuid to be removed during cleaning  https://review.openstack.org/16703318:39
*** mgoddard1 has joined #openstack-ironic18:43
*** Nisha has quit IRC18:45
*** mgoddard has quit IRC18:45
*** hyakuhei has joined #openstack-ironic18:48
*** hyakuhei has quit IRC18:48
*** mgoddard1 has quit IRC18:50
*** achanda has joined #openstack-ironic18:57
*** pas-ha has quit IRC18:57
*** achanda has quit IRC19:00
*** achanda has joined #openstack-ironic19:00
*** pelix has quit IRC19:00
*** jistr has joined #openstack-ironic19:02
*** mgoddard has joined #openstack-ironic19:02
*** ijw has quit IRC19:08
NobodyCamJoshNang: should we test that other operations do fail in cleaning as we expect them to? on ^^^^19:08
JoshNangNobodyCam: oh, duh. very much yes.19:09
NobodyCam:)19:09
*** jcoufal has joined #openstack-ironic19:11
openstackgerritJohn L. Villalovos proposed openstack/ironic: Fix broken unittests usage of sort()  https://review.openstack.org/16732519:13
openstackgerritMerged openstack/ironic: Fix formatting error in states_to_dot  https://review.openstack.org/16715919:18
openstackgerritMerged openstack/ironic: Ensure configdrive isn't mounted in CoreOS ramdisks  https://review.openstack.org/16706319:18
*** Marga_ has joined #openstack-ironic19:21
NobodyCamjlvillal: instresting use of six in test_chassis19:23
NobodyCamcan I ask why not just self.assertEqual(len(x), len(y))19:24
jlvillalNobodyCam: The unittest framework has a function to compare lists irregardless of order.  But they changed the name from Py2 vs Py3.  Thus six.19:24
jlvillalNobodyCam: Not trying to compare length.  Trying to compare contents19:24
NobodyCamahh so assertCountEqual is a missleading name19:26
jlvillalNobodyCam: I would not disagree.  But that is what unittest in Python3 chose as the name :(19:27
NobodyCamlol ack :)19:27
jlvillalNobodyCam: a and b have the same elements in the same number, regardless of their order19:28
openstackgerritJosh Gachnang proposed openstack/ironic: Allow instance_uuid to be removed during cleaning  https://review.openstack.org/16703319:30
*** jistr has quit IRC19:31
*** spandhe has quit IRC19:38
*** morganfainberg is now known as needslesscoffee19:39
*** spandhe has joined #openstack-ironic19:40
*** mgoddard has quit IRC19:41
*** achanda has quit IRC19:42
*** mgoddard has joined #openstack-ironic19:47
*** needslesscoffee is now known as morganfainberg19:50
*** jerryz has quit IRC19:52
*** jcoufal has quit IRC19:58
*** mgoddard has quit IRC20:04
*** ijw has joined #openstack-ironic20:18
*** mgoddard has joined #openstack-ironic20:24
*** hj-hp has joined #openstack-ironic20:24
*** romcheg has joined #openstack-ironic20:25
*** ijw_ has joined #openstack-ironic20:30
*** ijw has quit IRC20:33
*** jmccrory1 has joined #openstack-ironic20:34
*** jmccrory has quit IRC20:34
*** achanda has joined #openstack-ironic20:35
*** jmccrory has joined #openstack-ironic20:39
*** jmccrory1 has quit IRC20:42
mrdaMorning Ironic20:44
NobodyCammorning mrda20:44
mrdao/20:44
NobodyCamok I have a dumb git question. can I branch from a branch or better to create new branch just rebase on top of the orginial branch?20:45
TheJuliaNobodyCam: you can branch from a branch20:45
NobodyCam:)20:46
NobodyCamTy20:46
TheJuliaif your wanting to rebase, might not be a bad idea to create a new branch and do the rebase in that branch so you can go back if you have to20:47
mrda...especially if it's a nasty rebase20:48
TheJulia+120:48
NobodyCamshouldn't be but I'll do that... just sounds safer20:49
TheJuliajust remember the branch name you used20:49
*** mgoddard has quit IRC20:52
NobodyCambrb20:55
*** spandhe has quit IRC21:04
*** spandhe has joined #openstack-ironic21:06
*** openstackgerrit has quit IRC21:07
*** openstackgerrit has joined #openstack-ironic21:07
*** trown is now known as trown|outttypeww21:09
* BadCub is back21:09
TheJuliawb21:10
BadCubty :-)21:13
openstackgerritMerged openstack/ironic: Fix broken unittests usage of sort()  https://review.openstack.org/16732521:16
*** ndipanov has quit IRC21:22
*** andreykurilin_ has joined #openstack-ironic21:31
*** harlowja is now known as harlowja_away21:32
*** jmccrory has quit IRC21:34
*** andreykurilin__ has joined #openstack-ironic21:36
*** andreykurilin_ has quit IRC21:36
*** dprince has quit IRC21:38
*** kkoski has quit IRC21:40
*** mdbooth has quit IRC21:48
*** mdbooth has joined #openstack-ironic21:49
*** jmccrory has joined #openstack-ironic21:53
*** jmccrory has quit IRC21:53
*** ijw has joined #openstack-ironic21:54
*** ijw_ has quit IRC21:57
*** Marga_ has quit IRC21:59
*** lucas-dinner has quit IRC22:04
*** achanda_ has joined #openstack-ironic22:04
*** spandhe has quit IRC22:05
*** harlowja_away is now known as harlowja22:05
*** hj-hp has quit IRC22:06
*** hj-hp has joined #openstack-ironic22:07
*** achanda has quit IRC22:08
*** spandhe has joined #openstack-ironic22:08
TheJuliajroll: so your suggestion re: 166903 seems to work out better in the long run, it found more discrepencies so \o/ and :(22:11
NobodyCamuggh but still \o/22:12
NobodyCamya for US finding them and not a user !!!!22:13
jrollTheJulia: awesome22:13
* TheJulia goes and finds dinner22:15
NobodyCam:) enjoy the dinnering22:16
*** Marga_ has joined #openstack-ironic22:18
*** andreykurilin__ has quit IRC22:23
*** andreykurilin_ has joined #openstack-ironic22:23
jlvillalJayF: Question on: https://review.openstack.org/#/c/162790/5/ironic_python_agent/cmd/agent.py  On line 108 on the right hand side.  Do you know what a call to CONF() does?22:23
*** jcoufal has joined #openstack-ironic22:23
openstackgerritMichael Krotscheck proposed openstack/ironic: Added vagrant VM for developer use  https://review.openstack.org/16741522:25
JayFThere's something about how we load conf22:25
JayFbecause we inject configs there for metrics that can be sent from conductor22:25
*** kkoski has joined #openstack-ironic22:26
JayFjroll: ^ do you remember the details on that?22:26
*** jrist has quit IRC22:28
*** anderbubble has quit IRC22:28
*** jrist has joined #openstack-ironic22:29
*** Marga_ has quit IRC22:30
*** athomas has quit IRC22:31
NobodyCamugg22:33
NobodyCambrb22:33
jrollJayF: jlvillal: https://github.com/openstack/oslo.config/blob/master/oslo_config/cfg.py#L1802-185822:33
*** kkoski has quit IRC22:33
jlvillaljroll: Thanks!22:33
jrolljlvillal: that said, I don't remember offhand what it does :)22:34
jlvillaljroll: JayF: Though makes me wonder in what order should https://review.openstack.org/#/c/162790/5/ironic_python_agent/cmd/agent.py happen.  Before CONF() or after it?22:34
jlvillalOr maybe it doesn't matter.22:35
*** romcheg has quit IRC22:35
*** Marga_ has joined #openstack-ironic22:36
*** kkoski has joined #openstack-ironic22:37
jrolljlvillal: seems right at a glance22:39
jrolljlvillal: the only way to know is to run the code22:39
jlvillal:)22:39
openstackgerritJulia Kreger proposed openstack/ironic: Correct update_dhcp_opts methods  https://review.openstack.org/16690322:39
TheJuliaand now I really go to dinner22:40
openstackgerritChris Krelle proposed openstack/ironic: Check temp dir is usable for ipmitool driver  https://review.openstack.org/16038322:46
*** absubram has quit IRC22:53
*** kkoski has quit IRC22:53
*** kkoski has joined #openstack-ironic22:54
openstackgerritChris Krelle proposed openstack/ironic: Check temp dir is usable for ipmitool driver  https://review.openstack.org/16038322:58
*** kkoski has quit IRC22:59
*** kkoski has joined #openstack-ironic23:00
*** penick has quit IRC23:02
*** penick has joined #openstack-ironic23:07
*** andreykurilin_ has quit IRC23:07
*** andreykurilin__ has joined #openstack-ironic23:07
*** hj-hp has quit IRC23:12
*** chlong has joined #openstack-ironic23:20
*** penick has quit IRC23:22
*** jmccrory has joined #openstack-ironic23:27
*** andreykurilin__ has quit IRC23:34
*** chlong has quit IRC23:35
*** chlong has joined #openstack-ironic23:35
openstackgerritJay Faulkner proposed openstack/ironic: Ensure configdrive isn't mounted for ipxe/elilo  https://review.openstack.org/16744923:35
*** ChuckC has quit IRC23:35
*** ijw_ has joined #openstack-ironic23:37
*** ChuckC has joined #openstack-ironic23:39
openstackgerritChris Krelle proposed openstack/ironic: Add check for usable tftp directory to PXE driver  https://review.openstack.org/16745123:39
*** ijw has quit IRC23:40
JayFI'm like 99% sure I did something wrong on https://review.openstack.org/16744923:43
JayFI put a comment on it about where23:43
JayFbut someone with knowledge of these things giving me a review would be super helpful23:43
*** kkoski has quit IRC23:46
openstackgerritChris Krelle proposed openstack/ironic: Add check for usable tftp directory to PXE driver  https://review.openstack.org/16745123:49
*** penick has joined #openstack-ironic23:52
*** Haomeng|2 has joined #openstack-ironic23:56
*** Haomeng has quit IRC23:59

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