Monday, 2018-05-14

*** masber has joined #openstack-ironic00:03
*** masber has quit IRC00:03
*** masber has joined #openstack-ironic00:03
*** masber has quit IRC00:04
*** masber has joined #openstack-ironic00:04
*** masber has quit IRC00:04
*** masber has joined #openstack-ironic00:06
*** masber has quit IRC00:06
*** masber has joined #openstack-ironic00:08
*** etingof has quit IRC00:32
*** etingof has joined #openstack-ironic00:45
*** etingof has quit IRC00:53
*** hshiina has joined #openstack-ironic00:59
*** etingof has joined #openstack-ironic01:07
*** zhangfei has joined #openstack-ironic01:08
*** phuongnh has joined #openstack-ironic01:14
*** dprince has joined #openstack-ironic01:27
*** trungnv has joined #openstack-ironic01:33
*** tiendc has joined #openstack-ironic01:39
*** jesusaur has quit IRC01:45
*** baoli has joined #openstack-ironic01:45
*** jesusaur has joined #openstack-ironic01:53
*** zhangfei has quit IRC01:54
*** baoli has quit IRC01:59
*** baoli has joined #openstack-ironic02:00
*** baoli has quit IRC02:04
*** zhangfei has joined #openstack-ironic02:07
*** dprince has quit IRC02:10
*** baoli has joined #openstack-ironic02:17
*** baoli has quit IRC02:28
*** baoli has joined #openstack-ironic02:29
openstackgerritjiapei proposed openstack/ironic master: Add documentatin for XClarity Driver  https://review.openstack.org/55996002:32
*** baoli has quit IRC02:34
*** ijw has joined #openstack-ironic02:38
*** ijw__ has quit IRC02:41
*** ijw_ has joined #openstack-ironic02:42
*** etingof has quit IRC02:42
*** ijw has quit IRC02:45
*** etingof has joined #openstack-ironic02:56
openstackgerritmelissaml proposed openstack/python-ironic-inspector-client master: Trivial: Update pypi url to new url  https://review.openstack.org/56332703:07
*** etingof has quit IRC03:22
openstackgerritDao Cong Tien proposed openstack/ironic-python-agent master: Rescue bug: tinyipa fails to acquire IP in multitenant env  https://review.openstack.org/56627903:24
*** etingof has joined #openstack-ironic03:35
*** etingof has quit IRC03:41
*** etingof has joined #openstack-ironic03:54
*** threestrands has joined #openstack-ironic03:59
*** liuzz_ has joined #openstack-ironic03:59
*** threestrands_ has joined #openstack-ironic04:00
*** threestrands_ has quit IRC04:01
*** threestrands_ has joined #openstack-ironic04:02
*** liuzz has quit IRC04:02
*** threestrands has quit IRC04:04
*** kevinbenton has joined #openstack-ironic04:05
*** hoonetorg has quit IRC04:21
*** hoonetorg has joined #openstack-ironic04:39
*** threestrands_ has quit IRC04:39
*** links has joined #openstack-ironic04:41
*** rbartal has joined #openstack-ironic04:41
*** fragatina has quit IRC04:57
*** zhangfei has quit IRC04:58
*** threestrands has joined #openstack-ironic05:04
*** fragatina has joined #openstack-ironic05:07
*** threestrands has quit IRC05:07
*** zhangfei has joined #openstack-ironic05:10
*** toabctl has joined #openstack-ironic05:12
openstackgerritMerged openstack/ironic master: Switch to the fake-hardware hardware type for API tests  https://review.openstack.org/56451905:16
*** fragatina has quit IRC05:22
*** fragatina has joined #openstack-ironic05:22
*** hshiina2 has joined #openstack-ironic05:25
*** hshiina has quit IRC05:29
*** fragatina has quit IRC05:43
*** marios has joined #openstack-ironic05:46
openstackgerritzenghui.shi proposed openstack/ironic master: Indicate the continuation of cleaning steps in BIOS caching exception msg.  https://review.openstack.org/56819305:50
*** zshi has quit IRC05:52
*** zshi has joined #openstack-ironic05:53
*** mjura has joined #openstack-ironic06:01
*** hshiina2 is now known as hshiina06:21
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic master: Mark xclarity password as secret  https://review.openstack.org/56763706:26
*** pcaruana has joined #openstack-ironic06:31
*** jtomasek has joined #openstack-ironic06:32
*** tesseract has joined #openstack-ironic07:10
*** ajya has joined #openstack-ironic07:12
*** jaganathan has joined #openstack-ironic07:17
*** mgoddard has joined #openstack-ironic07:27
*** jaganathan has quit IRC07:31
*** jaganathan has joined #openstack-ironic07:31
*** trungnv has quit IRC07:35
*** fragatina has joined #openstack-ironic07:35
*** trungnv has joined #openstack-ironic07:35
*** rcernin has quit IRC07:39
*** AlexeyAbashkin has joined #openstack-ironic07:46
*** racedo has joined #openstack-ironic07:46
*** fragatina has quit IRC07:54
*** kevinbenton has quit IRC07:56
*** tiendc has quit IRC07:57
*** tiendc has joined #openstack-ironic07:58
*** lucas-afk is now known as lucasagomes07:59
*** athomas has joined #openstack-ironic08:06
*** erolg has joined #openstack-ironic08:12
*** dougsz has joined #openstack-ironic08:12
*** jaganathan_ has joined #openstack-ironic08:14
*** jaganathan has quit IRC08:14
*** dtantsur|afk is now known as dtantsur08:17
dtantsurmorning ironic, happy Monday08:18
*** jaganathan_ has quit IRC08:18
*** racedo has quit IRC08:26
*** derekh has joined #openstack-ironic08:29
*** racedo has joined #openstack-ironic08:29
*** jaganathan_ has joined #openstack-ironic08:30
etingofdtantsur, o/ though Fridays tend to be happier somehow08:31
*** alex_xu has quit IRC08:32
dtantsur:)08:32
*** alex_xu has joined #openstack-ironic08:33
*** arnewiebalck has joined #openstack-ironic08:39
*** MattMan has quit IRC08:41
*** MattMan has joined #openstack-ironic08:41
*** rbartal has quit IRC09:00
*** milan_ has joined #openstack-ironic09:04
*** masber has quit IRC09:06
*** dtantsur is now known as dtantsur|brb09:08
*** rbartal has joined #openstack-ironic09:13
openstackgerritIlya Etingof proposed openstack/ironic master: Adds more `ipmitool` errors as retryable  https://review.openstack.org/56721709:31
openstackgerritDao Cong Tien proposed openstack/ironic master: Update CI jobs for rescue mode  https://review.openstack.org/52870409:31
openstackgerritDao Cong Tien proposed openstack/ironic master: DNM: check rescue in multi-tenancy env  https://review.openstack.org/56628609:32
*** mgoddard has quit IRC09:33
*** stendulker has joined #openstack-ironic09:37
*** zhangfei has quit IRC09:40
*** zhangfei has joined #openstack-ironic09:41
etingofdtantsur|brb, there is still a chance to make Monday better by merging these immortal patches ;-) -- https://review.openstack.org/#/c/523879/09:46
patchbotpatch 523879 - sushy-tools - OpenStack cloud driver added09:46
*** diga has joined #openstack-ironic09:53
*** tiendc has quit IRC09:54
*** sambetts|afk is now known as sambetts10:02
sambettsMorning o/10:02
*** mgoddard has joined #openstack-ironic10:04
*** serlex has joined #openstack-ironic10:30
etingofsambetts, \o10:30
*** rbartal has quit IRC10:34
hjensasmorning o/10:35
*** racedo has quit IRC10:36
*** racedo has joined #openstack-ironic10:37
*** dtantsur|brb is now known as dtantsur10:38
openstackgerritHironori Shiina proposed openstack/ironic master: Fix tenant DeprecationWarning from oslo_context  https://review.openstack.org/56825710:48
*** rbartal has joined #openstack-ironic10:49
*** dprince has joined #openstack-ironic10:49
sambettshey etingof, hjensas10:50
*** e0ne has joined #openstack-ironic10:53
*** dprince has quit IRC10:55
*** dprince has joined #openstack-ironic10:56
*** arnewiebalck has quit IRC10:56
*** stendulker_ has joined #openstack-ironic11:02
*** links has quit IRC11:05
*** stendulker has quit IRC11:05
*** stendulker_ has quit IRC11:06
*** dprince has quit IRC11:11
*** zhangfei has quit IRC11:13
*** rbartal has quit IRC11:14
*** links has joined #openstack-ironic11:17
*** phuongnh has quit IRC11:18
*** needsleep is now known as TheJulia11:25
TheJuliaGood morning11:25
sambettso/ TheJulia11:25
*** rbartal has joined #openstack-ironic11:25
dtantsurmorning TheJulia11:31
*** dougsz has quit IRC11:34
*** mgoddard has quit IRC11:38
*** diga has quit IRC11:40
* TheJulia needs lots of coffee11:47
*** dprince has joined #openstack-ironic11:49
*** livelace has joined #openstack-ironic11:50
*** links has quit IRC11:52
*** links has joined #openstack-ironic11:52
*** rh-jelabarre has joined #openstack-ironic11:53
openstackgerrityolanda.robla proposed openstack/ironic master: Add Node BIOS support - REST API  https://review.openstack.org/51257911:53
*** hshiina is now known as hshiina|afk11:55
*** jcoufal has joined #openstack-ironic11:57
*** weshay is now known as weshay_interview11:59
*** dprince has quit IRC12:00
*** mgoddard has joined #openstack-ironic12:11
*** gyankum has joined #openstack-ironic12:11
*** dougsz has joined #openstack-ironic12:12
openstackgerritKaifeng Wang proposed openstack/ironic master: Power fault recovery: db and rpc implementation  https://review.openstack.org/55570812:12
openstackgerritMerged openstack/ironic master: Fix tenant DeprecationWarning from oslo_context  https://review.openstack.org/56795312:16
openstackgerritMerged openstack/ironic master: Fix tenant DeprecationWarning from oslo_context  https://review.openstack.org/56825712:16
openstackgerritKaifeng Wang proposed openstack/ironic master: Power fault recovery: API implementation  https://review.openstack.org/55601512:19
*** jaganathan_ has quit IRC12:19
*** jaganathan_ has joined #openstack-ironic12:19
TheJuliaAnyone know what Kaifeng's IRC handle is?12:20
* dtantsur no idea12:21
*** trown|outtypewww is now known as trown12:25
*** dprince has joined #openstack-ironic12:29
TheJuliahmm12:32
TheJuliaokay12:32
TheJuliaIf anyone knows, please let me know12:32
etingofTheJulia, may be you can ask via gerrit code review...12:35
TheJuliaI've sworn I've seen them on IRC at some point, of course I may just be loosing the remaining bits of my mind12:36
dtantsuryep, they've been here at least several times12:38
dtantsurthe nick was more or less straightforward12:38
TheJuliaThat is what I was thinking12:38
* dtantsur looks at ironic unit tests and wants to cry12:38
* dtantsur runs to the woods to live with other owls12:38
TheJuliadtantsur: it is too early in the week to cry12:38
dtantsurit's just the right time for somebody looking at how deeply we rely on the 'fake' classic driver there12:39
dtantsurlike, I need to rewrite maybe a few hundreds of tests12:39
* dtantsur cries and runs, cries and runs12:39
* TheJulia slides booze over to dtantsur to hopefully improve things a little bit12:40
*** gyankum has quit IRC12:42
*** bfournie has joined #openstack-ironic12:43
dtantsurTheJulia, sambetts, should we move on with https://review.openstack.org/#/c/305864/ ? it started more than 2 years ago, which is a bit too much for such a trivial change :D12:57
patchbotpatch 305864 - ironic-specs - Boot management for in-band inspection12:57
sambettsdtantsur: +10013:00
TheJuliadtantsur: yes, although wasn't there a reason on holding on that....13:00
*** rloo has joined #openstack-ironic13:00
dtantsurTheJulia: I *think* 1. we kept arguing about the exact design, 2. there are/were some dreams of reworking inspector completely13:01
sambettswe talked about moving inspection stuff into ironic, but I don't see that happening for a while13:01
*** jcoufal_ has joined #openstack-ironic13:01
sambettsdtantsur++13:01
*** weshay_interview is now known as weshay13:01
TheJuliaWe were also beginning to wade into drivercomp then13:02
sambettsI'm trying to find the gerrit patch with the conversation about changing the ironic config option13:02
sambettsfor that work13:02
dtantsurI think the latest version reflects what we last discussed13:03
*** mjturek has joined #openstack-ironic13:03
*** jcoufal has quit IRC13:03
sambettsdtantsur: yeah looks right13:04
openstackgerrityolanda.robla proposed openstack/ironic master: Add Node BIOS support - REST API  https://review.openstack.org/51257913:05
sambettsdtantsur: merged13:07
sambettsthanks TheJulia13:07
TheJuliawoot13:07
dtantsurthat was fast, thanks :)13:07
TheJuliaI concur, i re-read it and was like "oh, this is way simpler now, I'm onboard!"13:07
sambettsunfortunatly some of these things just get lost in the backlog :(13:07
dtantsurnow I can update https://review.openstack.org/#/c/316801/ :)13:08
patchbotpatch 316801 - ironic-inspector - Add manage_boot parameter to introspection API13:08
dtantsurthough I wonder if it's faster to do it from scratch... >_<13:08
TheJuliadtantsur: Well, to be fair, you could have updated it before....13:08
dtantsurright, I was swamped by other priority things, and so decided to wait for the resolution13:09
*** rbudden has joined #openstack-ironic13:11
* TheJulia nods :)13:13
*** Goneri has joined #openstack-ironic13:15
dtantsurOMG merge conflicts in EVERY SINGLE FILE13:15
*** ElCoyote_ has joined #openstack-ironic13:16
openstackgerritMerged openstack/ironic-specs master: Boot management for in-band inspection  https://review.openstack.org/30586413:18
rlooHappy Monday morning ironickers, including dtantsur, TheJulia, sambetts13:22
rlooTheJulia: it is (or was) kaifeng13:22
openstackgerritMerged openstack/ironic master: Add more parameter explanation when create a node  https://review.openstack.org/56245213:23
TheJuliarloo: thanks13:23
* TheJulia wonders if owls like laphroaig....13:24
sambettsrloo o/13:28
*** baoli has joined #openstack-ironic13:30
*** superdan is now known as dansmith13:33
*** baoli has quit IRC13:34
*** baoli has joined #openstack-ironic13:34
openstackgerritMerged openstack/ironic-ui stable/queens: Imported Translations from Zanata  https://review.openstack.org/56440213:34
dtantsurTheJulia: a lot! :)13:35
dtantsurmorning rloo13:35
rloo:)13:35
rlooTheJulia: I was just reviewing this, but did I miss something? The RFE isn't approved yet: https://review.openstack.org/#/c/55675813:38
patchbotpatch 556758 - ironic - Power fault recovery: applying maintenance_type13:38
dtantsurI thought it has a spec13:39
openstackgerritMerged openstack/ironic-ui master: Imported Translations from Zanata  https://review.openstack.org/56440313:39
dtantsurrloo: http://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/power-fault-recovery.html we probably forgot to update the RFE13:39
TheJuliayeah, spec landed, tag just never got updated13:40
openstackgerritMerged openstack/ironic-ui master: Updated from global requirements  https://review.openstack.org/55149013:40
rlooOUCH. Cuz... we have a usecase where we want both maintenance reasons: 1. from ironic/internally (power failure, clean failure... although the maintenance reason itself should indicate why it failed), and 2. manual via API.13:41
rlooI thought there was another RFE about it but honestly, i am having a hard time navigating via storyboard13:42
TheJuliaI'm not sure I grok the disconnect, although I've not looked at the api patches yet13:42
dtantsurrloo: there was a huge RFE that never got any progress13:42
rlooTheJulia: this particular solution only allows for 1 maintenace-reason at any given time.13:42
dtantsurso we ended up with a smaller RFE that just fixes that awful problem of nodes silently becoming non-operable13:43
dtantsurrloo: I think if you want 2 maintenance_reasons, you need a new RFE13:43
dtantsurit was not possible previously as well, so this RFE is clearly a way forward13:43
rloodtantsur: true, except that for 2 maintenance reasons, one for manual and one for internal-ironic, the internal node fields would be diff than adding this new 'maintenance_type' field.13:44
rloodtantsur: which means problems upgrading if we were ever to support two diff maintenance_reasons13:44
dtantsurI'm not quite convinced we should13:44
TheJuliaUhhh.. The reason tag indicates when a failure has occured and we have last_error13:45
rlooTheJulia: the 'maintenance_reason' tag? or do you mean some other 'reason' tag?13:45
TheJuliamaintenance_reason has two failure cases which are set upon those failures being encountered in the code base, at that point we also update last_error13:46
dtantsurif we allow two reasons, we should allow three and more. and this brings us back to that huge specific fauls spec that nobody reviewed13:46
TheJuliaso we have then stored the error, and taken the node out of service13:46
dtantsurwhich makes me think that we should finish this and then see if we have somebody to work on more13:46
TheJuliaiteration++13:46
rlooI don't think this solution lends itself to iterating w/o issues regarding upgrades13:47
openstackgerritDmitry Tantsur proposed openstack/ironic-inspector master: [WIP] Add manage_boot parameter to introspection API  https://review.openstack.org/31680113:48
rloosec, /me will actually read the spec/problem first...13:48
TheJuliaWell, it may never need iteration, it is the needful behavior.13:48
dtantsurrloo: I don't think it's fair to block a ready-to-land work based on something that is not obvious will even happen13:48
dtantsuror do you mean that you're going to start working on it around today? ;)13:48
rloodtantsur: heh, this story didn't get updated/tagged and there's no info there that there even was a spec. that isn't fair either.13:49
rloodtantsur: can you at least give me a few minutes to wrap my head around it?13:49
rloodtantsur: it was fair for me to put a -2 when i did, and i am reaching out/asking about it nwo.13:49
dtantsurI don't think we -2 patches because some core hasn't read the spec, but sure :)13:50
*** jaypipes_ is now known as jaypipes13:51
rloodtantsur: the -2 was cuz https://storyboard.openstack.org/#!/story/1596107 is not tagged with rfe-approved. i had no idea even that there was a spec cuz it wasn't obvious to me that a spec had been approved.13:51
* TheJulia hates the "rfe" process13:51
dtantsurokay, I'm fixing this one13:51
TheJuliaI already fixed the tag13:51
rlooTheJulia: then please propose something else.13:51
dtantsurokay, so now it's approved according to our process13:51
TheJuliaThis community doesn't like change, I can't change that at this point13:52
dtantsurnote that I don't question your right to -2 patches13:52
dtantsurwe should just be fair to the contributor who is not present on why we do it13:52
TheJuliaI just don't like the way we use it to block value adding work that is in a ready state to move forward13:52
dtantsurs/be/should be/13:52
*** rbartal has quit IRC13:53
*** jgwentworth is now known as melwitt13:57
TheJuliaI am with ruby, she had every right to block it until her concerns are understood in thinking that something hasn't been approved to be a thing. Design/overall effort changing is inevitable with every feature, and the smaller features are naturally going to get more traction and land faster than epics or epics of epics.13:58
*** beekneemech is now known as bnemec13:59
*** baha has joined #openstack-ironic14:00
jrollmorning14:02
*** links has quit IRC14:03
dtantsurTheJulia: I don't disagree with that, my point goes to the further discussion re "Should we delay the much needed work for the sake of potential yet-to-be-implemented-maybe-some-day bigger work"14:04
dtantsurmorning jroll14:04
jroll\o14:04
jrollsome perspective14:04
jrollwe have an internal patch that allows for both an ironic-set reason and a user-set reason14:05
jrollit is from the before times, so we don't (yet) have full context on the use case14:05
*** r-daneel has joined #openstack-ironic14:05
*** baoli has quit IRC14:05
*** jiapei has joined #openstack-ironic14:06
*** baoli has joined #openstack-ironic14:06
jrollso I suspect ruby is wondering if what we're doing covers the needs of large operators, and if the incremental step we're doing now will clobber any further progress14:06
*** rbartal has joined #openstack-ironic14:06
rloonot 'clobber', but make it a bit more difficult to move forward in a coherent fashion.14:07
jrollsure14:08
dtantsurwell, the input from fields I have in my hands is that Ironic is doing a horrible job with the power faults14:08
dtantsurand the patch we have in our hands solves that in a way14:08
dtantsurI wish your downstream patch was proposed upstream (long ago, ideally)14:08
jrollme too, trust me :)14:09
rloodtantsur: the downstream patch would not have directly addressed this particular use case. If I understand, you want *some field* to encode the reason so you can search on it.14:09
jrollbut I also think we should give cores more than 30 minutes of irc discussion to think through the effects of landing a patch :)14:09
dtantsurrloo: not really14:09
rloocuz otherwise, you coudl have just prefixed the mainteance string with 'CLEAN FAILURE: ...'14:10
dtantsurjroll: will till end of the cycle work? ;)14:10
rloojroll: to be honest, the thinking is during the spec review, which i missed.14:10
dtantsurrloo: right, the goal here is to be able to retry the failed power sync14:10
rloojroll: but having said that, i believe that i am usually open during the coding phase, to rethink things, so I hope we are too.14:10
jrollrloo: sure, but we also can't say "this matches the spec, cores may no longer question it"14:10
jrollyes14:11
dtantsurwhich requires knowing whether maintenance was caused by an operator or by ironic14:11
dtantsurjroll: oh, no, we cannot. that's certainly not my intention here.14:11
jrolldtantsur: I know you're joking about end of cycle, but I hope you get what I mean. I want to speed up progress as much as anyone here14:11
rloodtantsur: right, but you REALLY want, some encoded reason. Not 'operator' vs 'user'. but CLEAN_FAILURE, RESCUE_FAILURE, ... MANUAL14:11
jrolldtantsur: yes, I'm exaggerating for effect :)14:11
dtantsurrloo: this is a side effect, I guess. I personally only care about the maintenance going away if the BMC behaves again/network returned.14:12
*** baoli has quit IRC14:12
dtantsur* ironic-induced maintenance only - hence the database change14:12
sambettsrecoverable mainteiance vs unrecoverable maintenace14:13
dtantsurright, something like that14:13
sambettsman I can't spell today..14:13
dtantsurMonday14:14
rloodtantsur: so i think it is for our internal/coding/speed purposes, that we have a new separate maintenance_type field. Cuz like i mentioned above, we could prepend the mainteancne_reason strings with eg 'CLEANFAILURE', 'RESCUEFAILURE', etc and use those as indicators of ironic-induced maintenance.14:14
*** baoli has joined #openstack-ironic14:14
sambettsI mean this personally feels like a boolean14:14
dtantsurrloo: it is slow and not overly i18n-friendly14:14
dtantsurI'm fine with a boolean, but I'm not sure it solves rloo's concerns14:14
sambettsmaintenance: true/false, maintenance_reason: ..., ironic_induced: true/false14:15
rlooyou mean maintenance_type being a boolean?14:15
rlooi don't think so. you need the type, so that your (our) periodic task to try to sync power, can know which nodes are in maint due to power failure.14:15
rloo(note, I haven't actually looked at the code to address this; i just read the spec)14:16
* dtantsur W-1 https://review.openstack.org/#/c/555708/ until we agree14:16
patchbotpatch 555708 - ironic - Power fault recovery: db and rpc implementation14:16
dtantsurah, correct14:16
dtantsurwe cannot retry a cleaning failure14:16
* dtantsur cannot think today14:16
rloook, tell you what. let me think about it, will get back to you later today. and i promise, i'll review this patch (series)  in return :)14:17
dtantsuranother approach would be to untangle maintenance and faults (guarding by an API version)14:17
rloo(although i really, really, hate to promise anything)14:17
rloodtantsur: do you know offhand, the 'faults' big spec?14:17
rloocuz i honestly haven't read it14:17
*** rpioso|afk is now known as rpioso14:18
* jroll can find it14:18
rpiosoGood morning14:18
dtantsurrloo: it proposed a whole new generic mechanism for faults14:18
dtantsurwith a new API family, new drivers API, etc14:18
rloodtantsur: and this one fits into the big picture?14:18
jrollrloo: https://review.openstack.org/#/c/334113/14:18
patchbotpatch 334113 - ironic-specs - Add spec for specific faults support14:18
dtantsurrloo: not quite. I'm just not convinced we should do specific faults at all14:19
TheJulia\o/ discussion /o\ downstream meeting consuming brain14:19
rloothx jroll.14:19
jrollnp14:19
rloodtantsur: ah, so you're meh on 334113?14:19
rloodtantsur: or, er, you require convincing :)14:19
jrollI'm meh on 334113 simply because I think it's eating the whole pie in one bite14:20
rloodtantsur: i'll take a look to see if it convinces me ...14:20
jrolland we don't have two rackspace employees dedicated to implementing it anymore14:20
*** cjloader has joined #openstack-ironic14:20
*** cjloader has quit IRC14:20
*** zhangfei has joined #openstack-ironic14:20
rlooi like it when people dream big14:20
* jroll grumbles "Oh, So It's Cancelled?"14:20
*** cjloader has joined #openstack-ironic14:21
TheJuliadreaming big and then breaking it into small consumable bites is always the best14:21
TheJuliaKind of like huge pizzas14:21
* TheJulia now wants izza14:21
rloo++14:21
TheJuliapizza14:21
rloo++ to both!14:21
* rloo removes ++ to pizza. too early for me to eat that.14:22
dtantsurrloo: I'm with jroll on it14:22
openstackgerritSam Betts proposed openstack/ironic master: Stop double json encoding the error message responses  https://review.openstack.org/56783714:22
* jroll has a fridge full of pulled pork for the week \o/14:23
* sambetts so jelly 14:23
TheJuliammmmm pulled pork14:23
TheJuliaI've not made pulled pork in ages14:24
jrollhttps://imgur.com/3rsJroz.jpg for those not already hungry14:25
* rloo votes for a pizza & pulled pork & porridge & plum pudding ironic meeting today14:25
TheJuliaporridge and plum pudding?14:25
rloothe two other p* foods that i could think of in 1 sec. although popcorn comes to mind now.14:26
olivierb-Hello Ironicers/Bifrosters. I posted the following on openstack-dib IRC but may be this is more related to ironic/bifrost:14:26
jrolldon't forget potato salad14:27
olivierb-just another question on gpt generated dib image which is then provisioned to a machine using bifrost/ironic. When I launch fdisk on the disk where the image is written, I get: GPT PMBR size mismatch (125829119 != 209715199) will be corrected by w(rite).(message is repeated twice). However when I launch a KVM/libvirt instance using the qcow2 directly I do not get any error/warning. Does this ring a bell to someone ??? Should I14:27
olivierb- worry about this errror ? My gut feeling tells me I should ;-)14:27
* TheJulia makes a sad face at popcorn14:27
TheJuliaolivierb-: that seems familar, I'm on a meeting right now and can't fully contex switch14:29
rloojroll: do you know if that faults stuff (the solution advocated in the spec) was discussed with operators? I seem to recall that there had been some discussion, but I don't know if it was generic 'yeah we need to deal with faults' and/or whether it was this particular proposal14:30
rloojroll: also wondering -- is there a usecase for there being more than one fault occurring at any given time?14:30
olivierb-TheJulia no pb, no urgency there14:30
jrollrloo: I think we had some ops in a summit session or soemthing about it, not sure when/where/who though. it was driven by our operational needs with onmetal14:31
jrollI'd ask penick if he remembers discussing the specific proposal14:31
rlooand my other thought (for now): after putting in place whatever 'fault' mechanism. would it make sense to allow an operator to put that node into that fault -- to get the/wahtever self-healing to do the work?14:31
TheJuliaolivierb-: so, your going to get the geometry mismatch because the drive written to is larger14:31
*** zhangfei has quit IRC14:31
TheJuliaolivierb-: that being said, I thought we had code to catch and address that in IPA14:31
jrollrloo: not sure about more than one fault, if it's in maintenance nothing else should come up14:31
rloojroll: yeah, that's what I would think. although the spec talks about multiple faults.14:32
jrollrloo: I'm also not sure about operator putting the node into a fault mode. maybe it makes sense. I could see that being useful for a user to say "hey this disk died" and the operator being able to mark that14:32
olivierb-TheJulia checking with parted I get Warning: Not all of the space available to /dev/vda appears to be used, you can fix the GPT to use all of the space (an extra 83886080 blocks) or continue  so that may be exactly what you suggested14:33
jrollrloo: I wasn't super involved with that spec and I've blocked most of it from my memory14:33
TheJuliaoh you know what, I bet it is in ironic-lib....14:33
rloojroll: totally understandable :)14:33
*** chihhsin has quit IRC14:34
TheJuliaolivierb-: are you writing a configuration drive out?14:34
*** gyankum has joined #openstack-ironic14:34
*** gyankum has quit IRC14:34
olivierb-TheJulia sorry what do you mean exactly by that ?14:35
*** chihhsin has joined #openstack-ironic14:35
openstackgerritSam Betts proposed openstack/python-ironicclient master: Stop double json decoding API error messages  https://review.openstack.org/56783614:35
TheJuliaolivierb-: when you boot up, do you have a config-2 partition?14:35
olivierb-nope, I have only two as follows:14:36
TheJuliaolivierb-: are you deploying to baremetal using a playbook?14:36
olivierb-Number  Start   End     Size    File system  Name    Flags14:36
olivierb- 1      1049kB  9437kB  8389kB               "BSP"   bios_grub14:36
olivierb- 2      9437kB  60.0GB  60.0GB               "root"  lvm14:36
olivierb-and my disk is 100Gb wide14:36
olivierb-however when I try to create a new partition I also get:14:37
olivierb-(parted) mkpart14:37
olivierb-Partition name?  []? test14:37
olivierb-File system type?  [ext2]? ext414:37
olivierb-Start? 60Gb14:37
olivierb-End? 100%14:37
olivierb-Warning: You requested a partition from 60.0GB to 107GB (sectors 117187500..209715199).14:37
olivierb-The closest location we can manage is 60.0GB to 64.4GB (sectors 117204992..125829086).14:37
olivierb-Is this still acceptable to you?14:37
olivierb-Yes/No?14:37
sambettsolivierb-: please use paste.openstack.org for outputs like that14:38
olivierb-sambetts , oops sorry about this14:38
TheJuliaso we have code to correct that, but only when we are writinga configuration drive out to the disk14:39
olivierb-TheJulia could you please point me to this code so that I have a look ?14:39
*** rajinir has joined #openstack-ironic14:40
olivierb-http://paste.openstack.org/show/720926/, sorry again for the former copy/paste in IRC14:40
TheJuliait looks like we would need to make a method in ironic-lib a first class citizen instead of being private, and calling that from IPA in general if it is GPT14:40
TheJuliahttps://github.com/openstack/ironic-python-agent/blob/0912a24b384518e402394e63fc1ad94579d38b76/ironic_python_agent/extensions/standby.py#L472-L479 <-- IPA deciding what to do and to call ironic-lib14:40
TheJuliahttps://github.com/openstack/ironic-lib/blob/300e171546ed6832106596861bd3ff50fa061dce/ironic_lib/disk_utils.py#L761-L76214:41
TheJuliaolivierb-: seems like it shouldn't be too hard of a bug to fix, if your up for it14:42
TheJuliaWe would need to cut an ironic-lib release to make the method public though14:42
olivierb-thanks for the pointer, I'll try to understand the code better and shoot for a fix14:42
TheJuliabut that is not too big of a deal14:42
TheJuliaolivierb-: it will be two, one for ironic-lib and one for ironic-python-agent. The IPA fix likely won't work until we merge/release ironic-lib, For what it is worth14:43
*** zhangfei has joined #openstack-ironic14:44
olivierb-got it thanks TheJulia14:44
TheJuliaEveryone updating whiteboard? :)14:46
*** rbudden has quit IRC14:48
*** stendulker has joined #openstack-ironic14:53
*** rbudden has joined #openstack-ironic14:53
*** hshiina has joined #openstack-ironic14:55
*** marlinc has quit IRC14:59
*** mjura has quit IRC14:59
*** ijw_ has quit IRC14:59
*** ijw has joined #openstack-ironic14:59
TheJulia#startmeeting ironic15:00
openstackMeeting started Mon May 14 15:00:03 2018 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.15:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:00
TheJuliao/15:00
*** openstack changes topic to " (Meeting topic: ironic)"15:00
openstackThe meeting name has been set to 'ironic'15:00
mrainerio/15:00
TheJuliaGood morning everyone!15:00
vdroko/15:00
rpiosoo/15:00
jroll\o15:00
rlooo/15:00
hshiinao/15:00
jiapeio/15:00
TheJuliaIt is time for our weekly ironic meeting! (Our project name is never going to get old....) :)15:00
stendulkero/15:01
TheJuliaOur agenda is on the wiki.15:01
TheJulia#link https://wiki.openstack.org/wiki/Meetings/Ironic15:01
mjturek\o15:01
*** gcb has joined #openstack-ironic15:01
jrollnice and light :)15:01
TheJuliaindeed!15:01
TheJulia#topic Announcements / Reminders15:01
*** openstack changes topic to "Announcements / Reminders (Meeting topic: ironic)"15:01
dtantsuro/15:02
TheJulia#info Our meeting is cancelled next week due to the Summit.15:02
hjensaso/15:02
*** pcaruana has quit IRC15:02
TheJuliaThat is all I have right now for announcements/reminders. Does anyone else have anything they would like to raise?15:03
rloothe rocky-2 milestone is coming up (sort of) June 7th week15:03
rloothat is nova-spec freeze15:03
rlooare there any from us, that we need to help ?15:04
TheJulia#info Rocky-2 milestone approaching - June 7th15:04
TheJuliarloo: We should follow up on rescue I believe. I think that is the only thing we need to worry about right now.15:05
rlooTheJulia: ok. (or any other non-ironic project specs. i guess i shouldn't just focus on nova.)15:05
hshiinabp for rescue was approved for rocky15:05
hshiinain nova15:05
rloohshiina: awesome!15:06
rlooso we're good for now :)15:06
*** milan_ has quit IRC15:06
TheJuliaAnyway, if there is nothing else, we can proceed to reviewing action items.15:06
TheJuliahshiina: excellent to hear15:06
TheJulia#topic Reviewing action items from last week15:07
*** openstack changes topic to "Reviewing action items from last week (Meeting topic: ironic)"15:07
*** S4ren has joined #openstack-ironic15:07
TheJulia#link http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-05-07-15.00.txt15:07
*** marlinc has joined #openstack-ironic15:08
TheJuliaI was super busy last week and fell down on the action items some last week. I did clean-up the whiteboard some, I forgot to send out an email w/r/t cancelling the meeting next week, and I didn't get a chance to look through what other projects are doing re prioritization in storyboard.15:08
rlooCongrats to TheJulia for having the most (and all) action items!15:08
TheJuliai know! I should throw a party!15:08
TheJulia#action TheJulia to send meeting re-meeting cancellation for next week.15:09
TheJuliaRe: priorities in storyboard, I think I'll try and get with folks in vancouver and wrap my brain around it then.15:09
TheJuliaOne last item, the status update email discussion is ongoing on the list, I know etingof replied if memory serves, but I've not had a chance to read it.15:10
TheJuliaWith that being said, if there are no objections, we can move on.15:10
TheJulia#topic Review subteam status reports15:11
*** openstack changes topic to "Review subteam status reports (Meeting topic: ironic)"15:12
TheJulia#link https://etherpad.openstack.org/p/IronicWhiteBoard15:12
TheJuliaStarting around line 8915:12
TheJuliaerr15:12
TheJuliaLine 131!15:12
TheJuliaerr15:12
TheJuliaLine 14515:12
TheJuliaClearly I need lots of coffee15:13
TheJuliaand a vacation.15:13
jroll+100015:13
openstackgerritDmitry Tantsur proposed openstack/sushy-tools master: Clean up templates from items that cannot be implemented for libvirt  https://review.openstack.org/53826815:13
jrolloh my, bios is close \o/15:13
openstackgerritDmitry Tantsur proposed openstack/sushy-tools master: Support EthernetInterface resource in the emulator  https://review.openstack.org/53833815:14
dtantsuryep :)15:14
TheJuliaVery very close \o/15:14
rpiosoAwesome15:14
jrollclassic driver removal also nearing the finish line15:15
TheJuliaI'm going to put a few specs that relate to items that are outstanding on this list on the priorities for this week15:15
jrollsounds like a good idea to me15:15
TheJuliaThat way we get some eyes on them15:16
rloodtantsur: wrt the classic driver removal. are you OK if I delete the stuff that has already been done? hard to see what is left to do15:16
dtantsurrloo: yes please15:16
dtantsurI did not update the list, because IIRC TheJulia planned to move it to storyboard15:16
dtantsuror maybe I should do it? :)15:16
TheJuliadtantsur: you know it better and I've been crazy busy recently15:16
dtantsuractually, it's a good question: should we track it on the whiteboard or storyboard?15:16
rloodtantsur: in the classic driver removal, there is a question @ L244ish15:17
rloodtantsur: kill it :)15:17
TheJuliaThe preference we seem to have as a community is to track efforts on etherpads, changing to just using storyboard for that might feel a bit painful15:17
*** derekh has quit IRC15:17
dtantsurrloo: I think I even did it already :)15:17
jrollI'd prefer storyboard, with whiteboard as the "what got done last week / needs help this week"15:18
rloodtantsur: please update the whiteboard then :D15:18
jrollbut open to whatever15:18
TheJuliaI remember dtantsur lightly singed the wiki15:18
TheJuliaThere may have been some smoldering afterwards :)15:18
rloo++ if we are using storyboard, we might want to use it for what it was meant/good at.15:18
rloodtantsur: 'fake' -> 'fake_hardware' hardware type? :)15:19
TheJuliaWhich is the effort tracking, I've personally been trying to break things up and use multiple tasks which is the way it was intended, but yeah...  One step at a time15:19
dtantsurrloo: yep, unfortunately this is MUCH harder than just sed15:20
TheJuliaIs everyone good to move on to priorities?15:20
*** derekh has joined #openstack-ironic15:20
*** derekh has quit IRC15:20
dtantsuryep15:20
rloo+ good15:20
jiapei++15:20
*** zhangfei has quit IRC15:21
TheJulia#topic Priorities for the next week15:21
*** openstack changes topic to "Priorities for the next week (Meeting topic: ironic)"15:21
TheJulia#link https://etherpad.openstack.org/p/IronicWhiteBoard15:21
*** derekh has joined #openstack-ironic15:21
TheJuliaLine 9215:21
rloowhat about adding this to priorities: https://review.openstack.org/#/c/50706715:21
patchbotpatch 507067 - ironic - [devstack] Switch ironic to uWSGI15:21
TheJuliarloo: I actually have that on the list I was about to paste in....15:21
rlooTheJulia: :)15:21
*** gcb has quit IRC15:22
TheJuliadtantsur: is the hw-types topic unblocked?15:23
TheJuliadtantsur: or are there more api-jobs that need changing?15:23
olivierb-TheJulia & all, I made the changes in ironic-lib and ironic-python-agent code in my own repos ? what is the (best) way to take this into account ? Just git clone/checkout properly on bifrost master or change some of the requirements.txt files. Sorry for the freshman developper questions ;-)15:23
dtantsuryep, but no patches open right this second15:23
dtantsurI mean, yes, unblocked15:23
rlooolivierb-: we're in the weekly ironic meeting now (you're welcome to join). Shoudl be done in 30-ish minutes.15:24
olivierb-will head back to home/gym and will connect later to read your advices, cheers15:24
TheJulia30-ish minutes if my cat does not find the router, and swipe it from the edge of the earth into the universe.... (so far during the meeting five things have been swiped onto the floor....)15:25
dtantsurENEEDATREAT15:25
rpiosoTheJulia: Sounds like your cat doesn't like meetings :-)15:26
TheJuliapretty much....15:26
jrollgood cat15:26
TheJuliaAnyway, I've added the power fault stuff to the list since it is in solid shape as far as I've seen thus far and does improve things, I know there was discussion earlier, so we can remove it or not. I've also placed some specs at the beginning of the list that could use some eyeballs. The virtual media/dhcp-less boot deploy interface nokia is interested in needs a minor revision, but I also feel it is in good shape.15:27
* TheJulia feels bad for not having read the buffer from when she was on a call.15:27
jrollseems like a decent list, though far too long to complete :)15:28
TheJuliaWell, no meeting next week :)15:28
jrollprobably not much review work either :P15:28
rlooTheJulia is giving us homework. heh!15:30
TheJulialol15:30
TheJuliaBut we can't call it homework ;)15:30
rloos/homework/ironi-ng/15:31
TheJulialol15:31
TheJuliaI like that!15:31
* rloo thinks we're ready to move on15:31
rloo:)15:32
TheJuliaYup15:32
* TheJulia lost connectivity for a minute :\15:33
TheJuliaAnd not at the paws of the cat15:33
TheJuliaSo since everyone seems good with the list and we have no specific RFEs or discusison topics, lets jump to Open Discussion15:34
TheJulia#topic Open Discussion15:34
*** openstack changes topic to "Open Discussion (Meeting topic: ironic)"15:34
jrollI have a quick thing here15:35
jrollthis discussion happened a while back:15:35
jroll#link http://lists.openstack.org/pipermail/openstack-dev/2018-April/129216.html15:35
jrolland again today15:35
jroll#link http://lists.openstack.org/pipermail/openstack-dev/2018-May/130423.html15:35
TheJuliaolivierb-: Best way for you to test your changes is to craft a custom IPA image or patch it. If your booting form pxe, then it is easy with taking apart the ramdisk and patching it15:35
openstackgerritMerged openstack/ironic-inspector master: Raise KeyboardInterrupt on SIGTERM - Workaround  https://review.openstack.org/56333515:35
jrolltl;dr people want to get pycrypto out of global-requirements in rocky, some projects are blocking that, including pyghmi15:35
jrollis anyone available to switch to the cryptography library in pyghmi and coordinate that with jarrod?15:36
*** srart has quit IRC15:36
TheJuliaI had someone reach out to me first thing and ask for jarrod's contact info to discuss exactly that15:36
jrollthat was me, see the april link for his plan :)15:37
TheJuliaif it was you this morning, my brain is a blur15:37
jrolloh, not this morning, no15:37
jrollso maybe we'll wait to see if this person and/or jarrod makes progress soon15:38
jrollI just want to keep the requirements team unblocked :)15:38
TheJuliaThe person that asked me seems to not be around at the moment15:39
TheJulia#action TheJulia to follow-up w/r/t pycrypto in pyghmi - to determine if we have someone to take that on.15:39
dtantsurTheJulia: it was etingof?15:39
jrollthanks TheJulia15:39
TheJuliadtantsur: well, I wasn't going to name him because I hadn't gotten confirmation back yet, but I'll follow-up15:39
rlooaction item, action item, action item, ... :)15:40
TheJuliaSpeaking of following up things, there was the ceph tempurl thing last week, who is picking that patch up?15:40
rloo(for etingof I mean, cuz he isn't here)15:40
* TheJulia is asking now so it does not become yet another action item15:40
TheJuliarloo: Excellent idea!15:40
jrollTheJulia: it's on my list15:40
*** erolg has quit IRC15:40
TheJuliajroll: awesome! Thanks!15:41
rloothx jroll!15:41
jrollnp15:41
mgoddardolivierb-, TheJulia: in kayobe we use environment variables to configure the IPA repo URL & version to use. In bifrost these could be set in dib_env_vars. I'm not sure how you could use a custom ironic-lib, perhaps via requirements.txt like you say. Take a look at the ironic-agent DIB element: https://github.com/openstack/diskimage-builder/tree/master/diskimage_builder/elements/ironic-agent15:42
TheJuliaI'd really just rip the cpio file apart and rebuild it, but I'm a little old fashioned15:42
mgoddardrloo: I was wondering what the status of deploy steps is. Are you planning to implement it?15:42
rloomgoddard: yes, still planning on it. unless someone else wants to do it.15:43
rloomgoddard: hoping to get around to it next month.15:43
mgoddardok great. I'm up for helping on the coding and/or reviewing side - whatever works for you15:44
rloothx mgoddard!15:44
TheJuliaAwesome, well does anyone have anything else to bring up/discuss today?15:45
mgoddardzshi: how is the BIOS work going? Are there any tempest tests and documentation in progress?15:45
mgoddardWould be good to see tempest tests before merging the API15:46
*** ajya has quit IRC15:46
*** ajya2 has joined #openstack-ironic15:46
mgoddardyolanda_: ^^15:46
mgoddardwe can take this offline if everyone else is done here15:47
TheJuliamgoddard: tempest api read tests, I guess could exist easily... otherwise we don't explicitly test cleaning in the gate, although maybe this is a good route to get there15:47
TheJuliaredfish support for bios could help us get there I guess15:48
TheJuliaJust thinking outloud15:48
mgoddardTheJulia: yeah, the fake driver was built in a way that much of the BIOS code could be tested via tempest15:48
mgoddardi.e., just writes any requested options to the DB15:49
TheJuliaTrue, and there are two sides to it15:49
TheJuliaWell, if there is nothing else, I guess we can all get ten minutes back :)15:50
TheJuliawell, more like 9 minutes now15:50
TheJulia:)15:50
jroll\o/15:50
dtantsur:)15:51
TheJuliaAwesome!15:51
*** srart has joined #openstack-ironic15:51
TheJuliaThanks everyone!15:51
jrollthanks!15:51
dtantsurthanks15:51
TheJulia#endmeeting15:51
*** openstack changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/developer/ironic/ | Bugs: https://bugs.launchpad.net/ironic"15:51
*** marios has quit IRC15:51
openstackMeeting ended Mon May 14 15:51:27 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:51
openstackMinutes:        http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-05-14-15.00.html15:51
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-05-14-15.00.txt15:51
openstackLog:            http://eavesdrop.openstack.org/meetings/ironic/2018/ironic.2018-05-14-15.00.log.html15:51
* dtantsur can boast grabbing ID 1000 in storyboard for his pet project :D https://storyboard.openstack.org/#!/project/100015:51
TheJuliadtantsur: nice!15:52
*** stendulker has quit IRC15:59
*** hshiina has quit IRC16:00
*** trown is now known as trown|lunch16:01
*** mgoddard has quit IRC16:03
*** ajya2 has quit IRC16:06
*** ajya has joined #openstack-ironic16:06
*** ijw has quit IRC16:08
*** lucasagomes is now known as lucas-afk16:10
*** gyee has joined #openstack-ironic16:13
*** baha has quit IRC16:17
TheJuliaRunning into town to get some things to fix a water leak, I should be back in an hour or so16:19
*** pcaruana has joined #openstack-ironic16:20
etingofTheJulia, dtantsur rloo sorry folks, I got a minor hiccup right before the meeting16:30
rlooetingof: no worries. Are you OK to work on the pycrypto/pyghmi thing?16:30
* sambetts ----> home 16:36
sambettsnight all o/16:36
*** sambetts is now known as sambetts|afk16:36
etingofrloo, sure, I am delighted ;)16:40
rpiososambetts|afk: Good night16:40
rlooetingof: thank you so much. And I am so glad we can make you feel delighted! :)16:41
*** dprince has quit IRC16:44
*** rbartal has quit IRC16:49
*** baha has joined #openstack-ironic16:55
*** dougsz has quit IRC16:57
*** AlexeyAbashkin has quit IRC16:57
*** derekh has quit IRC17:00
*** trown|lunch is now known as trown17:00
*** serlex has quit IRC17:05
*** dprince has joined #openstack-ironic17:06
*** hemna_ has joined #openstack-ironic17:07
*** hemna_ has quit IRC17:07
*** hemna_ has joined #openstack-ironic17:07
*** e0ne has quit IRC17:10
*** tesseract has quit IRC17:11
*** dtantsur is now known as dtantsur|afk17:14
dtantsur|afksee you17:14
jrollnight dtantsur|afk :)17:16
openstackgerritIlya Etingof proposed openstack/ironic master: SNMPv3 security features added to the `snmp` driver  https://review.openstack.org/54799917:17
*** efried has quit IRC17:19
*** efried has joined #openstack-ironic17:19
*** baoli has quit IRC17:21
*** etingof is now known as etingof|afk17:21
*** baoli has joined #openstack-ironic17:26
*** fragatina has joined #openstack-ironic17:45
*** fragatina has quit IRC17:46
*** fragatina has joined #openstack-ironic17:47
*** fragatina has quit IRC17:47
*** fragatina has joined #openstack-ironic17:48
*** S4ren has quit IRC17:48
*** r-daneel_ has joined #openstack-ironic18:00
*** r-daneel has quit IRC18:01
*** r-daneel_ is now known as r-daneel18:01
TheJuliaetingof|afk: awesome, thanks! Have a wonderful evening18:02
*** anupn has joined #openstack-ironic18:08
*** harlowja has joined #openstack-ironic18:12
*** e0ne has joined #openstack-ironic18:18
*** ajya has quit IRC18:23
olivierb-TheJulia, like you I like to work directly on gzipped,cpioed initramfs, however I just discovered that what is inside the ipa ramdisk are .pyo files not .py :-( would it be an issue replacing the .pyo by its patched .py version without anything else ?18:30
*** hamzy has quit IRC18:30
TheJuliaolivierb-: doh :(18:30
TheJuliaolivierb-: maybe use a coreos ramdisk?18:30
TheJuliawe don't strip the objects out of those18:30
*** hamzy has joined #openstack-ironic18:33
anupnoliverb-: you can get the .py files by setting export PYOPTIMIZE_TINYIPA=false18:37
openstackgerritDmitry Tantsur proposed openstack/python-ironic-inspector-client master: Support passing manage_boot argument in Python API  https://review.openstack.org/33947918:37
anupnolivierb-:  you can get the .py files by setting export PYOPTIMIZE_TINYIPA=false18:37
olivierb-anupn thx however I am not rebuillding the ramdisk, i am decompressing it and un-archiving it to patch it and reconstruct18:39
*** mgoddard has joined #openstack-ironic18:40
*** etingof|afk is now known as etingof18:40
*** Guest16323 is now known as mgagne18:40
*** mgagne has joined #openstack-ironic18:40
*** mgoddard has quit IRC18:48
openstackgerritJulia Kreger proposed openstack/ironic master: Fix E501 errors  https://review.openstack.org/56702818:54
*** trown is now known as trown|brb18:54
*** r-daneel has quit IRC18:59
*** r-daneel_ has joined #openstack-ironic18:59
*** r-daneel_ is now known as r-daneel19:01
*** mrunge has quit IRC19:21
*** Goneri has quit IRC19:21
openstackgerritchandra shekar proposed openstack/ironic-specs master: Added new spec for L3 based Ironic deployment  https://review.openstack.org/54393619:22
*** dprince has quit IRC19:23
*** mgoddard has joined #openstack-ironic19:25
olivierb-are there some checksum mechanism implemented because after regenerating the ipa.initramfs the boot process is stuck in booting the kernel ? I just gunzipped + cpio extract, cpio archive + gzip without changing any contents and same issue19:28
olivierb-(just to try without any changeds)19:29
*** trown|brb is now known as trown19:31
olivierb-may be TheJulia knows ^^^19:36
olivierb-https://gist.github.com/obourdon/3c1208853c03316b041a48e37ec9df65 is what I used without being able to get a successfull boot19:38
*** dprince has joined #openstack-ironic19:39
TheJuliaolivierb-: add "--format='newc'" to your cpio command19:40
*** mrunge has joined #openstack-ironic19:41
olivierb-TheJulia thx just rebuilt ipa and restarted provisioning to see19:49
olivierb-boot in progress19:50
TheJuliano problem, it is slightly obscure knowledge19:50
TheJulia:)19:50
olivierb-horray, working now thanks19:50
olivierb-final question for today, are the console logs of 2nd IPA pass (writing qcow2 image to HDD) retrieved+stored somewhere on bifrost master node ?19:58
openstackgerritJulia Kreger proposed openstack/python-ironicclient master: Wire in header microversion into negotiation  https://review.openstack.org/55802719:58
*** athomas has quit IRC20:00
*** dprince has quit IRC20:09
*** efried has quit IRC20:17
*** mgoddard has quit IRC20:26
*** racedo has quit IRC20:27
openstackgerritIlya Etingof proposed openstack/sushy-tools master: OpenStack cloud driver added  https://review.openstack.org/52387920:32
*** baoli has quit IRC20:37
*** baoli has joined #openstack-ironic20:37
*** baoli has quit IRC20:41
*** pcaruana has quit IRC20:43
*** jcoufal_ has quit IRC20:43
*** livelace has quit IRC20:54
*** Sukhdev has joined #openstack-ironic20:55
*** racedo has joined #openstack-ironic20:55
TheJuliaolivierb-: console, no. ironic ships its logging data up to the conductor... /var/lib/ironic... I think20:56
TheJuliaolivierb-: did your patch work?20:57
olivierb-nope the call to the new code is made but then the machine fails to restart, I think there is an uncaught exception but need to debug it further. will do this tomorrow early morning, time to go to sleep now ;-) thanks for everything TheJulia20:59
*** trown is now known as trown|outtypewww21:00
*** hemna_ has quit IRC21:04
openstackgerritMerged openstack/python-ironicclient master: Switch to none auth for standalone mode  https://review.openstack.org/35906121:06
*** r-daneel has quit IRC21:07
*** r-daneel has joined #openstack-ironic21:07
*** e0ne has quit IRC21:16
*** bnemec has quit IRC21:17
*** bnemec has joined #openstack-ironic21:18
*** baha has quit IRC21:31
*** mjturek has quit IRC21:40
*** cjloader has quit IRC21:50
*** cjloader_ has joined #openstack-ironic21:50
*** rcernin has joined #openstack-ironic21:53
*** cjloader has joined #openstack-ironic21:55
*** cjloader_ has quit IRC21:55
*** harlowja has quit IRC21:56
*** cjloader has quit IRC21:59
*** rh-jelabarre has quit IRC22:03
*** bfournie has quit IRC22:06
*** jistr has quit IRC22:07
*** etingof has quit IRC22:08
*** jistr has joined #openstack-ironic22:10
*** threestrands has joined #openstack-ironic22:16
*** etingof has joined #openstack-ironic22:21
*** hemna_ has joined #openstack-ironic22:29
*** bfournie has joined #openstack-ironic22:46
*** r-daneel has quit IRC22:54
*** ElCoyote_ has quit IRC22:59
*** baoli has joined #openstack-ironic23:00
*** baoli has quit IRC23:04
*** rajinir has quit IRC23:19
*** rpioso is now known as rpioso|afk23:26
*** oanson has quit IRC23:37
*** oanson has joined #openstack-ironic23:48
*** srart has quit IRC23:52

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