Wednesday, 2018-01-24

*** hshiina has joined #openstack-ironic00:00
*** zhurong has joined #openstack-ironic00:01
*** vsaienk0 has joined #openstack-ironic00:02
*** penick has joined #openstack-ironic00:07
*** rbudden has quit IRC00:12
*** vsaienk0 has quit IRC00:12
*** vsaienk0 has joined #openstack-ironic00:13
*** akhilaki_ has joined #openstack-ironic00:16
*** akhilaki has joined #openstack-ironic00:16
*** zhurong has quit IRC00:20
*** oanson has quit IRC00:22
*** oanson has joined #openstack-ironic00:24
*** yangyapeng has quit IRC00:25
*** yangyapeng has joined #openstack-ironic00:25
*** yangyapeng has quit IRC00:30
*** penick has quit IRC00:30
*** sasha2 has quit IRC00:32
*** penick has joined #openstack-ironic00:32
*** rbudden has joined #openstack-ironic00:42
*** liuzz has joined #openstack-ironic00:42
*** vsaienk0 has quit IRC00:48
openstackgerritOpenStack Proposal Bot proposed openstack/bifrost master: Updated from global requirements  https://review.openstack.org/53702800:49
*** vsaienk0 has joined #openstack-ironic00:55
openstackgerritOpenStack Proposal Bot proposed openstack/ironic master: Updated from global requirements  https://review.openstack.org/53705100:55
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-inspector master: Updated from global requirements  https://review.openstack.org/53705200:55
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-lib master: Updated from global requirements  https://review.openstack.org/53705300:55
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-python-agent master: Updated from global requirements  https://review.openstack.org/53705400:56
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-ui master: Updated from global requirements  https://review.openstack.org/53493500:56
*** penick has quit IRC00:58
*** r-daneel has quit IRC01:01
openstackgerritOpenStack Proposal Bot proposed openstack/networking-baremetal master: Updated from global requirements  https://review.openstack.org/53707801:05
openstackgerritOpenStack Proposal Bot proposed openstack/networking-generic-switch master: Updated from global requirements  https://review.openstack.org/53708001:06
*** phuongnh has joined #openstack-ironic01:11
*** chas has joined #openstack-ironic01:11
*** tiendc has joined #openstack-ironic01:12
*** akhilaki_ has quit IRC01:14
*** crushil has joined #openstack-ironic01:15
*** akhilaki has quit IRC01:16
*** chas has quit IRC01:16
*** yangyapeng has joined #openstack-ironic01:17
*** yangyapeng has quit IRC01:26
*** yangyapeng has joined #openstack-ironic01:26
*** early has quit IRC01:27
*** rbudden has quit IRC01:27
*** cjloader has joined #openstack-ironic01:29
*** early has joined #openstack-ironic01:30
*** Goneri has quit IRC01:30
*** oanson has quit IRC01:38
openstackgerritOpenStack Proposal Bot proposed openstack/virtualbmc master: Updated from global requirements  https://review.openstack.org/53720001:41
*** oanson has joined #openstack-ironic01:41
*** hieulq_ has quit IRC01:42
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-ui master: Updated from global requirements  https://review.openstack.org/53493501:56
openstackgerritDao Cong Tien proposed openstack/ironic master: DNM: test rescue mode  https://review.openstack.org/53318701:57
openstackgerritDao Cong Tien proposed openstack/ironic master: DNM: standalone test rescue mode  https://review.openstack.org/53475101:58
*** hieulq_ has joined #openstack-ironic01:59
openstackgerritDao Cong Tien proposed openstack/ironic master: DNM: standalone test rescue mode  https://review.openstack.org/53475102:01
openstackgerritJim Rollenhagen proposed openstack/python-ironicclient master: Traits support  https://review.openstack.org/53262202:06
*** gcb has joined #openstack-ironic02:09
*** vsaienk0 has quit IRC02:09
openstackgerritMerged openstack/networking-generic-switch master: Reraise NGS exceptions seen during netmiko config  https://review.openstack.org/53053202:11
*** ricardoas has joined #openstack-ironic02:12
*** ricardoas-away has quit IRC02:15
*** rloo has quit IRC02:18
*** cjloader has quit IRC02:18
openstackgerritOpenStack Proposal Bot proposed openstack/python-ironic-inspector-client master: Updated from global requirements  https://review.openstack.org/53428502:19
*** harlowja has quit IRC02:20
*** vsaienk0 has joined #openstack-ironic02:21
openstackgerritMichael Turek proposed openstack/ironic-python-agent master: Use lshw in place of dmidecode for the default hardware manager  https://review.openstack.org/50179902:25
openstackgerritOpenStack Proposal Bot proposed openstack/sushy-tools master: Updated from global requirements  https://review.openstack.org/53405902:25
*** yolanda has quit IRC02:30
*** nicodemos has quit IRC02:39
*** ricardoas has quit IRC02:41
*** nicodemos has joined #openstack-ironic02:43
*** ricardoas has joined #openstack-ironic02:46
openstackgerritDao Cong Tien proposed openstack/python-ironicclient master: Add support for RESCUE and UNRESCUE provision states  https://review.openstack.org/40834102:49
openstackgerritDao Cong Tien proposed openstack/python-ironicclient master: Add rescue_interface to node and driver  https://review.openstack.org/51730202:49
openstackgerrityuan liang proposed openstack/ironic-python-agent master: Execute error in _detect_cna_card  https://review.openstack.org/53469002:54
openstackgerrityuan liang proposed openstack/ironic-python-agent master: Execute error in _detect_cna_card  https://review.openstack.org/53469002:57
*** chas has joined #openstack-ironic02:59
*** chas has quit IRC03:04
*** cjloader_ has joined #openstack-ironic03:08
*** crushil has quit IRC03:09
*** vsaienk0 has quit IRC03:11
*** tuanla____ has joined #openstack-ironic03:14
tuanla____Morning ironic03:14
*** crushil_ has joined #openstack-ironic03:15
*** crushil__ has joined #openstack-ironic03:24
*** cjloader_ has quit IRC03:27
*** crushil_ has quit IRC03:28
*** fragatina has quit IRC03:32
*** jlabarre has quit IRC03:42
*** hshiina has quit IRC03:42
*** vsaienk0 has joined #openstack-ironic03:51
*** hshiina has joined #openstack-ironic03:54
*** nicodemos has quit IRC03:54
*** nicodemos has joined #openstack-ironic03:54
*** Goneri has joined #openstack-ironic03:58
*** chas has joined #openstack-ironic03:59
*** vsaienk0 has quit IRC04:00
*** tiendc has quit IRC04:00
*** tuanla____ has quit IRC04:00
*** tiendc has joined #openstack-ironic04:01
*** tuanla____ has joined #openstack-ironic04:01
*** crushil__ has quit IRC04:03
*** chas has quit IRC04:03
*** jaianshu has joined #openstack-ironic04:11
*** nicodemos has quit IRC04:13
*** ricardoas has quit IRC04:13
*** Goneri has quit IRC04:14
*** nicodemos has joined #openstack-ironic04:16
*** ricardoas has joined #openstack-ironic04:16
*** akhilaki_ has joined #openstack-ironic04:21
*** akhilaki has joined #openstack-ironic04:21
*** nicodemos has quit IRC04:33
*** tuanla____ has quit IRC04:33
*** tuanla____ has joined #openstack-ironic04:34
*** ricardoas has quit IRC04:35
*** nicodemos has joined #openstack-ironic04:38
*** vsaienk0 has joined #openstack-ironic04:40
*** ricardoas has joined #openstack-ironic04:41
*** masber has joined #openstack-ironic04:42
*** dpawar has joined #openstack-ironic04:49
*** chas has joined #openstack-ironic04:50
*** chas has quit IRC04:54
*** akhilaki_ has quit IRC04:54
*** akhilaki has quit IRC04:54
*** ijw has quit IRC05:09
*** vsaienk0 has quit IRC05:11
*** harlowja has joined #openstack-ironic05:19
*** links has joined #openstack-ironic05:32
*** vsaienk0 has joined #openstack-ironic05:36
*** vsaienk0 has quit IRC05:58
*** rcernin_ has joined #openstack-ironic06:03
*** rcernin has quit IRC06:03
*** vsaienk0 has joined #openstack-ironic06:06
*** fragatina has joined #openstack-ironic06:13
*** fragatina has quit IRC06:14
*** fragatina has joined #openstack-ironic06:14
*** rwsu has joined #openstack-ironic06:21
*** gcb has quit IRC06:22
*** ijw has joined #openstack-ironic06:26
*** ijw has quit IRC06:33
*** marios has joined #openstack-ironic06:37
*** threestrands has joined #openstack-ironic06:39
openstackgerritShivanand Tendulker proposed openstack/ironic master: Agent rescue implementation  https://review.openstack.org/40043706:40
openstackgerritShivanand Tendulker proposed openstack/ironic master: Add API methods for [un]rescue  https://review.openstack.org/35083106:40
openstackgerritShivanand Tendulker proposed openstack/ironic master: Add rescue interface field to node-related notifications  https://review.openstack.org/50288906:40
openstackgerritShivanand Tendulker proposed openstack/ironic master: Update description for config params of 'rescue' interface  https://review.openstack.org/51791906:40
openstackgerritShivanand Tendulker proposed openstack/ironic master: Add documentation for rescue interface  https://review.openstack.org/41960606:40
*** vsaienk0 has quit IRC06:41
*** vsaienk0 has joined #openstack-ironic06:42
*** hjensas has quit IRC06:46
*** threestrands has quit IRC06:49
openstackgerritTapio Tallgren proposed openstack/bifrost master: Rewrite iptables rules with Ansible module  https://review.openstack.org/53691106:52
*** makowals has joined #openstack-ironic06:53
openstackgerritGao Fei proposed openstack/ironic master: Replace curly quotes with straight quotes  https://review.openstack.org/53727707:04
*** mjura has joined #openstack-ironic07:08
*** pcaruana has joined #openstack-ironic07:10
*** rcernin has joined #openstack-ironic07:12
*** rcernin_ has quit IRC07:12
openstackgerritKaifeng Wang proposed openstack/python-ironicclient master: Can not set portgroup mode as a number  https://review.openstack.org/53728307:17
*** pcaruana has quit IRC07:21
*** pcaruana has joined #openstack-ironic07:22
openstackgerrityuan liang proposed openstack/ironic-python-agent master: Execute error in _detect_cna_card  https://review.openstack.org/53469007:25
*** vsaienk0 has quit IRC07:27
*** vsaienk0 has joined #openstack-ironic07:29
*** e0ne has joined #openstack-ironic07:30
*** jaganathan has quit IRC07:31
*** stendulker has joined #openstack-ironic07:36
*** chas has joined #openstack-ironic07:36
*** rwsu has quit IRC07:41
*** jaganathan has joined #openstack-ironic07:44
openstackgerritHarald Jensås proposed openstack/ironic master: Flat networks use node.uuid when binding ports.  https://review.openstack.org/53604007:45
*** jtomasek has quit IRC07:48
*** jtomasek has joined #openstack-ironic07:49
*** hjensas has joined #openstack-ironic07:52
*** trown|outtypewww has quit IRC07:53
*** trown has joined #openstack-ironic07:53
openstackgerritHarald Jensås proposed openstack/ironic master: Devstack - use neutron segments (routed provider networks)  https://review.openstack.org/53127507:55
openstackgerritHarald Jensås proposed openstack/ironic master: Wait for ironic-neutron-agent to report state  https://review.openstack.org/53163707:55
*** e0ne has quit IRC08:01
*** pcaruana has quit IRC08:01
*** e0ne has joined #openstack-ironic08:01
*** skazi has quit IRC08:02
openstackgerritTapio Tallgren proposed openstack/bifrost master: Rewrite iptables rules with Ansible module  https://review.openstack.org/53691108:07
*** skazi has joined #openstack-ironic08:12
*** AlexeyAbashkin has joined #openstack-ironic08:23
*** tesseract has joined #openstack-ironic08:27
*** vsaienk0 has quit IRC08:30
*** vsaienk0 has joined #openstack-ironic08:35
*** pcaruana has joined #openstack-ironic08:39
*** rcernin has quit IRC08:44
pas-hamorning ironic ^)08:44
*** pmannidi has quit IRC08:45
*** tesseract has quit IRC08:54
pas-hadtantsur|afk: unit tests on inspector/master are failing.08:54
pas-hadtantsur|afk: reason - fresh bump of u-c of 'construct' lib. latest (2.8.22) version has backward-incompatible changes to exceptions presented (FieldError -> FormatFieldError and StreamError).08:56
pas-hadtantsur|afk: what do we do? file a bug for construct hoping they fix it in next release and block 2.8.22 in g-r, or update our code and update g-r to be >= 2.8.22?08:57
pas-haseems only inspector is actually dependent on it.. http://codesearch.openstack.org/?q=construct&i=nope&files=%5Erequirements.txt&repos=08:58
pas-haso I'd go with option 2 (fix code and update g-r to require at latest 2.8.22)08:59
*** jaianshu_ has joined #openstack-ironic09:03
*** tesseract has joined #openstack-ironic09:03
*** jaianshu has quit IRC09:05
openstackgerritya.wang proposed openstack/ironic master: Add timeout for all HTTP requests via the requests library.  https://review.openstack.org/53731109:07
*** harlowja has quit IRC09:07
*** etingof has quit IRC09:14
*** milan_ has joined #openstack-ironic09:15
pas-hadtantsur|afk: the breaking commit in 'construct' is https://github.com/construct/construct/commit/812753a79ad8590a563c9881c5ffd8c2125e04e209:25
*** chas has quit IRC09:35
*** derekh has joined #openstack-ironic09:39
*** yangyapeng has quit IRC09:50
*** absubram has joined #openstack-ironic09:50
*** etingof has joined #openstack-ironic09:50
*** yangyapeng has joined #openstack-ironic09:51
*** amarao has joined #openstack-ironic09:52
*** absubram has quit IRC09:54
*** yangyapeng has quit IRC09:55
*** absubram has joined #openstack-ironic09:56
*** tuanla____ has quit IRC09:58
*** dtantsur|afk is now known as dtantsur10:19
dtantsurmorning ironic10:19
dtantsurpas-ha: I think we should at least start with the former10:19
dtantsurbecause then the stable branches are also broken, and we cannot bump requirements there10:20
sambetts|afkstable branches should be protected by upper-constraints10:20
*** sambetts|afk is now known as sambetts10:20
dtantsurpas-ha: in any case, could you please talk to the requirements/release team?10:20
dtantsursambetts: only in our CI, not downstream10:20
dtantsurwe hope that downstreams follow u-c, but who knows?10:21
sambettsthey should do buts thats another story :/ looking at the construct thing this change was only in a patch release 2.8.21 to 2.8.22 so should not be breaking from their side10:24
sambettsso yeah I think they've just made a bad release and we should update g-r to block it10:24
dtantsurpas-ha: could you please talk to upstream then?10:26
sambettswow... https://github.com/construct/construct/issues/428 apprently the dev for construct doesn't believe in backwards compatiblity or sem-ver...10:28
*** jaianshu__ has joined #openstack-ironic10:29
dtantsuroh damn, how did we end up using it?..10:29
dtantsurwe really need some rating system for dependencies. "1 star - does not work; 2 stars - works, but the author is immature and have no clue how real world works"10:30
dtantsuretc10:30
sambettsnot sure.. but basically the attitude seems to be pin and you won't get broken10:30
*** jaianshu_ has quit IRC10:32
*** lucas-afk is now known as lucasagomes10:33
dtantsursambetts: we need to plan getting away from this crap10:33
*** rcernin has joined #openstack-ironic10:34
*** yangyapeng has joined #openstack-ironic10:39
sambettsyeah... may be able to use the native python struct lib ... just looking10:39
*** jaganathan has quit IRC10:39
*** phuongnh has quit IRC10:40
dtantsureven if it's less convenient, at least it won't get broken at any moment with a resolution "lol your problems dudes" :)10:42
dtantsursambetts: lemme know if you think it's doable to rewrite the same code in python or a saner library10:43
*** jaganathan has joined #openstack-ironic10:51
*** AlexeyAbashkin has quit IRC10:53
*** AlexeyAbashkin has joined #openstack-ironic10:54
*** yangyapeng has quit IRC10:58
*** liuzz has quit IRC10:59
*** yangyapeng has joined #openstack-ironic10:59
openstackgerritparesh sao proposed openstack/ironic master: Implementation for UEFI iSCSI boot for ILO  https://review.openstack.org/46828811:05
*** jaganathan has quit IRC11:08
*** absubram has quit IRC11:11
openstackgerritDmitry Tantsur proposed openstack/ironic master: Switch the CI to hardware types  https://review.openstack.org/53687511:11
openstackgerritparesh sao proposed openstack/ironic master: Implementation for UEFI iSCSI boot for ILO  https://review.openstack.org/46828811:13
openstackgerritDmitry Tantsur proposed openstack/ironic master: Switch the CI to hardware types  https://review.openstack.org/53687511:15
*** gmonteiro has joined #openstack-ironic11:17
*** racedo has joined #openstack-ironic11:17
openstackgerritDmitry Tantsur proposed openstack/ironic master: Clean up CI playbooks  https://review.openstack.org/53589611:17
gmonteiroGood morning Ironic11:17
openstackgerritDmitry Tantsur proposed openstack/ironic master: Deprecate classic drivers  https://review.openstack.org/53692811:18
*** dtantsur is now known as dtantsur|brb11:19
*** sapd_ has quit IRC11:19
*** sapd_ has joined #openstack-ironic11:19
*** jaganathan has joined #openstack-ironic11:21
openstackgerritDmitry Tantsur proposed openstack/ironic master: Migrate python-oneviewclient validations to oneview hardware type  https://review.openstack.org/52459911:32
openstackgerritDmitry Tantsur proposed openstack/ironic master: Add a timeout for powering on/off a node on oneview  https://review.openstack.org/52462411:32
*** jkilpatr has quit IRC11:33
openstackgerritHarald Jensås proposed openstack/ironic master: Flat networks use node.uuid when binding ports.  https://review.openstack.org/53604011:36
openstackgerritHarald Jensås proposed openstack/ironic master: Devstack - use neutron segments (routed provider networks)  https://review.openstack.org/53127511:36
openstackgerritHarald Jensås proposed openstack/ironic master: Wait for ironic-neutron-agent to report state  https://review.openstack.org/53163711:36
openstackgerritHarald Jensås proposed openstack/ironic master: Add documentation for baremetal mech  https://review.openstack.org/53735311:36
openstackgerritHarald Jensås proposed openstack/ironic master: [WIP] Add baremetal mechanism driver to gate jobs  https://review.openstack.org/53735411:36
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Add baremetal neutron agent  https://review.openstack.org/45623511:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: start_flag = True, only first time, or conf change  https://review.openstack.org/53370711:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Switch from MechanismDriver to SimpleAgentMechanismDriverBase  https://review.openstack.org/52183811:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Use reporting_interval option from neutron  https://review.openstack.org/53679211:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Devstack - Add ironic-neutron-agent  https://review.openstack.org/53011711:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Add support to bind type vlan networks  https://review.openstack.org/53234911:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Add dsvm job  https://review.openstack.org/53040911:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Docs - Add multi-tenant devstack - n-g-s vlans  https://review.openstack.org/53445911:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Make the agent distributed using hashring and notifications  https://review.openstack.org/52470911:38
openstackgerritMerged openstack/ironic master: Add RPC object for traits  https://review.openstack.org/53226811:38
openstackgerritMerged openstack/networking-generic-switch master: Updated from global requirements  https://review.openstack.org/53708011:38
openstackgerritHarald Jensås proposed openstack/ironic master: Flat networks use node.uuid when binding ports.  https://review.openstack.org/53604011:43
openstackgerritHarald Jensås proposed openstack/ironic master: Add documentation for baremetal mech  https://review.openstack.org/53735311:43
openstackgerritHarald Jensås proposed openstack/ironic master: Devstack - use neutron segments (routed provider networks)  https://review.openstack.org/53127511:43
openstackgerritHarald Jensås proposed openstack/ironic master: Wait for ironic-neutron-agent to report state  https://review.openstack.org/53163711:43
openstackgerritHarald Jensås proposed openstack/ironic master: [WIP] Add baremetal mechanism driver to gate jobs  https://review.openstack.org/53735411:43
*** stendulker has quit IRC11:45
openstackgerritMerged openstack/python-ironicclient master: Updated from global requirements  https://review.openstack.org/53716311:51
openstackgerritMerged openstack/python-ironic-inspector-client master: Updated from global requirements  https://review.openstack.org/53428511:51
*** jkilpatr has joined #openstack-ironic11:55
*** jaianshu__ has quit IRC11:56
openstackgerritMerged openstack/virtualbmc master: Updated from global requirements  https://review.openstack.org/53720011:59
openstackgerritMerged openstack/sushy-tools master: Updated from global requirements  https://review.openstack.org/53405912:06
openstackgerritMerged openstack/ironic-ui master: Imported Translations from Zanata  https://review.openstack.org/53243612:06
*** zhurong has joined #openstack-ironic12:10
*** vsaienk0 has quit IRC12:15
openstackgerritGustavo Diniz Monteiro proposed openstack/ironic master: Remove python-oneviewclient from oneview hardware type.  https://review.openstack.org/52472912:47
*** vsaienk0 has joined #openstack-ironic12:48
*** gmonteiro has quit IRC12:53
*** dpawar has quit IRC13:01
jrollmorning ironic13:05
jrollI have a thing this morning, I won't be around until 1700 utc or later, jfyi. happy to help with anything that needs some work once I'm back13:05
pas-hadtantsur|brb: sambetts filed an issue with construct any way https://github.com/construct/construct/issues/45313:09
pas-haso, should we add <2.8.22 to g-r for now?13:10
*** trown is now known as trown|ruck13:15
pas-hathe construct author proposed to pull 2.8.22 out from PyPI, and move this change to a next major 2.9 release13:18
pas-haI think we could add <2.9 and pin ===2.8.21 then (until we move out of consuming this lib)13:18
*** tiendc has quit IRC13:21
*** hjensas has quit IRC13:23
pas-hadtantsur|brb: sambetts patch to requirements https://review.openstack.org/53737813:25
patchbotpatch 537378 - requirements - Block construct >= 2.9 and pin to 2.8.2113:25
*** MattMan has quit IRC13:26
*** MattMan has joined #openstack-ironic13:26
*** SerenaFeng has joined #openstack-ironic13:33
*** makowals has quit IRC13:34
*** makowals has joined #openstack-ironic13:35
*** makowals has quit IRC13:43
*** makowals has joined #openstack-ironic13:45
*** rloo has joined #openstack-ironic13:46
*** AlexeyAbashkin has quit IRC13:47
*** rcernin has quit IRC13:47
openstackgerritHarald Jensås proposed openstack/ironic master: [WIP] Add baremetal mechanism driver to gate jobs  https://review.openstack.org/53735413:48
*** hjensas has joined #openstack-ironic13:49
*** hjensas has quit IRC13:49
*** hjensas has joined #openstack-ironic13:49
*** jjohnson2 has joined #openstack-ironic13:50
openstackgerritMark Goddard proposed openstack/ironic master: Add traits field to node notifications  https://review.openstack.org/53697913:52
openstackgerritMark Goddard proposed openstack/ironic master: Add documentation for node traits  https://review.openstack.org/53698013:52
openstackgerritMark Goddard proposed openstack/ironic master: Fix nits found in node traits  https://review.openstack.org/53738613:52
rloogood morning ironickers, mgoddard, dtantsur|brb, sambetts, pas-ha, TheJulia, and everyone else13:54
*** dtantsur|brb is now known as dtantsur13:54
dtantsurmorning rloo13:54
rloosambetts, dtantsur|brb: wrt construct -- would be worth opening a bug so we don't forget the reason why we want to get off it. that link was great (or not great)13:54
dtantsurpas-ha: great that they collaborate, at least. and thanks for tracking it. I think we should update g-r to exclude 2.8.22 and cap 2.9 for now13:55
mgoddardmorning rloo & co13:56
*** sasha2 has joined #openstack-ironic13:58
TheJuliagood morning13:58
dtantsurmorning TheJulia, mgoddard13:59
*** weshay|rover is now known as weshay|mtg14:00
*** AlexeyAbashkin has joined #openstack-ironic14:00
*** Goneri has joined #openstack-ironic14:03
TheJuliarloo: replied to your comments on python-ironicclient version list14:06
rlooTheJulia: thx14:06
dtantsurlet's try to merge these two patches today, if possible14:06
dtantsurI'll get some food and will be available for review/discussion14:06
*** krtaylor has quit IRC14:07
rlooTheJulia: also, let me know about https://review.openstack.org/#/c/531029/, while i look at your comments in the other one14:08
patchbotpatch 531029 - python-ironicclient - Facilitate latest Rest API use14:08
*** bfournie has quit IRC14:08
*** bfournie has joined #openstack-ironic14:08
*** krtaylor has joined #openstack-ironic14:09
*** SerenaFeng has quit IRC14:10
rlooTheJulia, dtantsur: I guess it is worth discussing. I am sorry about bikeshedding, but we're exposing 'negotiate' so I think it should be clear what that means and/or whether we want to expose that term to our users.14:12
rlooTheJulia, dtantsur: I don't actually see any negotiation, the client decides based on what the server supports and what the user indicated.14:12
*** bfournie has quit IRC14:13
*** links has quit IRC14:15
rlooTheJulia, dtantsur: how does novaclient (or any other OpenStack client) handle this? Are there any that do?14:16
*** MattMan has quit IRC14:16
*** MattMan has joined #openstack-ironic14:16
*** hshiina has quit IRC14:17
*** zhurong has quit IRC14:18
*** cdearborn has joined #openstack-ironic14:18
TheJuliarloo: most just default to latest avialable14:18
TheJuliawe're the only client that put an explicit pin lock in, at least the last time I looked through clients14:19
rlooTheJulia: latest available supported by server? even if the client request is a lower version?14:19
TheJuliarloo: I'm not sure what you mean by "I don't actually see any negotiation"? The bulk of the logic was already there14:20
TheJuliarloo: it is an override at that point to what the server can support14:20
*** gmonteiro has joined #openstack-ironic14:21
TheJuliarloo: posted replies to latest14:22
dtantsurTheJulia, rloo, I feel like "negotiation" is a more or less established word for it14:22
rlooTheJulia: i didn't test your patches, just looked at them. So is the flow something like 1. user makes request, client issues request to server with specified (if any) version, server response is error -- cannot handle version or request, (negotiation started) client determines a better version, resends request?14:22
dtantsurrloo: the essence is: the client has some supported versions, the server has some supported versions, they negotiate on something that fits both the best14:23
TheJuliarloo: incorrect for latest or a list, we explicitly go ask the server up front in those cases14:23
rlooTheJulia, dtantsur: apologies, but I don't understand and it isn't documented -- we really should have had a spec for this.14:23
TheJuliabut only when there is a request.14:23
TheJuliafrom there, then the rest of the negotiation logic applies14:23
dtantsurrloo: this is an attempt of formalizing it: https://review.openstack.org/53281414:23
patchbotpatch 532814 - api-wg - Add guideline on exposing microversions in SDKs14:23
rloodtantsur: thx, let me read it14:24
dtantsursuggestions like "let us define the word 'negotiate' formally" are welcome :)14:24
TheJuliarloo: with that being said, also trying to get parts of that done as I have time/braincells in small chunks. The minimum viable we need, if nova will ever allow it to land, is a list of valid versions.14:24
rlooTheJulia, dtantsur: I understand that there is a need for this, i just want to make sure it is clear what the need/requirements are, and that these patches address them. but i cannot do that by just looking at the patches themselves. I am missing context. And I know, the deadline is tomorrow :-(14:25
dtantsurso, the API-SIG spec may provide some answers to "why" and "how", I hope14:26
*** crushil has joined #openstack-ironic14:27
openstackgerritMerged openstack/ironic master: Migrate oneview hardware type to use python-hpOneView  https://review.openstack.org/52431014:29
*** baha has joined #openstack-ironic14:31
*** cjloader has joined #openstack-ironic14:31
*** rbudden has joined #openstack-ironic14:31
*** jlabarre has joined #openstack-ironic14:31
TheJuliarloo: greatest reason for this, have a facility to allow nova-compute to be upgraded in grenade tests....14:31
*** yolanda has joined #openstack-ironic14:32
rlooTheJulia: :) I really blame it on the grenade infrastructure anyway!14:32
*** rloo has quit IRC14:32
dtantsura few years pass, and maybe we'll be finally allowed to use ansible playbooks instead...14:33
*** rloo has joined #openstack-ironic14:33
*** AlexeyAbashkin has quit IRC14:33
*** makowals has quit IRC14:33
*** cjloader has quit IRC14:34
*** lucasagomes is now known as lucas-hungry14:35
TheJuliamaybe, but even then we need high order execution control, and there will still always be concerns over their users peception of order where as our users already know, the only reason problems crop up is becaues of singular hard pins and functionality is tied to the microvesion14:36
*** makowals has joined #openstack-ironic14:36
dtantsursmall update: I've just realized I won't make it for the next meeting - my train arrives after it starts14:37
*** gmonteiro has left #openstack-ironic14:40
TheJuliaso either \o/ or /o\14:40
*** gmonteiro has joined #openstack-ironic14:40
*** gmonteiro has left #openstack-ironic14:40
*** dpawar has joined #openstack-ironic14:41
*** crushil has quit IRC14:42
*** bfournie has joined #openstack-ironic14:43
*** serlex has joined #openstack-ironic14:46
* dtantsur looks at zuul backlog in despair14:48
dtantsurthe day before feature freeze, isn't it?14:48
rloodtantsur: yup :-(14:49
*** jcoufal has joined #openstack-ironic14:50
rloodtantsur: ok, read the spec. i don't understand something, well, maybe more than one thing, but it has me stuck. L165 in https://review.openstack.org/#/c/532814/5/guidelines/sdk-exposing-microversions.rst14:50
patchbotpatch 532814 - api-wg - Add guideline on exposing microversions in SDKs14:50
*** jcoufal has quit IRC14:51
rloodtantsur: if 1.2 was specified, and it is supported, why would there be an exception?14:51
dtantsurrloo: the exception will be if the version is not supported14:51
rloodtantsur: or rather, if 1.2 was specified and it was not supported, why would the exit msg be 'at least 1.2 is required'?14:51
rloodtantsur: maybe that is just a typo thing. all you know is that 1.2 is not supported, right?14:52
dtantsurrloo: we also assume that versions don't have gaps. so if 1.2 is not supported, 1.3 etc are not either14:52
dtantsur"at least" means 1.3 would work too14:52
dtantsurbut we don't have 1.2, 1.3 or any newer, only e.g. 1.114:53
rloodtantsur: how do you know that though? that you don't have 1.3 or any newer?14:53
rloodtantsur: you just know that 1.2 isn't supported14:53
rloodtantsur: what if the min of the server is 1.514:53
dtantsurrloo: fair enough, this may be an unfortunate wording14:53
dtantsurbut you get the idea: 1.2 is not there14:54
rloodtantsur: ok, so the only thing we can say is that 1.2 is not supported, right?14:54
dtantsurrloo: yep. I'll update it in the next update of that spec14:54
edleafedtantsur: I think it's safe to assume (and document) that versions should always be increasing14:54
rlooedleafe: that's true. but all you (I) know in that example, is that 1.2 is not supported.14:55
edleafeIOW, you should never have "we support 1.1, 1.3, 1.4..." and thus have 1.2 fail14:55
rlooedleafe: oh, that's interesting, cuz i was wondering about that.14:56
edleaferloo: what I'm saying is you also know that any 1.x version greater than 1.2 is also not supported14:56
rlooedleafe: in the eg style of our requirements, where we can skip one or two or... :)14:56
openstackgerritDmitry Tantsur proposed openstack/ironic master: Switch the CI to hardware types  https://review.openstack.org/53687514:57
rlooedleafe: well, unless the server's min is 1.4, then 1.2 isn't supported but 1.4 is14:57
edleafeNow you can have 1.2 not supported, but support 2.0, 2.1, 3.0, etc14:57
edleaferloo: ah, I was assuming that we already knew the min14:57
rlooedleafe: i am assuming nothing. except that i can read :)14:57
edleaferloo: I can't always assume that for myself :)14:57
rlooedleafe: :D14:58
*** jcoufal has joined #openstack-ironic14:58
openstackgerritDmitry Tantsur proposed openstack/ironic master: Clean up CI playbooks  https://review.openstack.org/53589614:58
openstackgerritDmitry Tantsur proposed openstack/ironic master: Deprecate classic drivers  https://review.openstack.org/53692814:58
rloodtantsur: ok, now i'm at the 'list of versions', L21214:59
*** AlexeyAbashkin has joined #openstack-ironic15:00
rloodtantsur: if negotiated is 1.42 it barks. if it isn't 1.42 -- what version will be used, 1.0?15:00
*** crushil has joined #openstack-ironic15:00
dtantsurrloo: yes, meow executes with 1.0 then15:00
*** dpawar has quit IRC15:00
rloodtantsur: oooohhhhh.15:01
* dtantsur likes discussing microversions in terms of cute animals15:01
rloodtantsur: and if at L208, 1.0 is not supported but 1.42 is, would there be an exception?15:01
dtantsurrloo: maybe it will be easier if instead of 1.0 and 1.42 you think in terms of "pike" and "queens"?15:01
*** mjturek has joined #openstack-ironic15:01
dtantsurrloo: no, no exception, because 1.42 can be picked15:01
rloodtantsur: it is ok, numbers or releases, but lets use versions since we aren't going actually support named versions right (please don't say yes)15:02
dtantsurlol15:02
dtantsurno, we aren't15:02
rloodtantsur: ok, so L213 will always be true and L216 will never be executed.15:02
dtantsurrloo: if 1.42 is present - yes15:02
rloodtantsur: so the list thing is used instead of the hardcoded, to get exactly one version. i mean, the hardcoded could be done, calling code could just loop through each version when calling catsclient.v1.get_client()15:04
rloodtantsur: sorry, not to get exactly one version. to specify the only versions to be used.15:04
*** dpawar has joined #openstack-ironic15:04
rloodtantsur: the list one is going to be hairy i think. to implement correctly. for all methods.15:05
dtantsurrloo: yeah, you can emulate the list of versions by looping through them from the highest to the lowest15:05
dtantsurrloo: not particularly. keep in mind that the negotiation is only done once15:05
rloodtantsur: if your list has 10 versions, 1.2, 1.3, 1.4, ..., 1.18, 1.20, you get a client that can use all those versions. if .negotiated_api_version == 1.20, cat.bark(), else: cat.meow() uses which version? 1.18?15:07
*** etingof has quit IRC15:07
TheJuliarloo: supporting named versions in the api construct would be really mean for any 3rd party re-implementation of the api.....15:07
dtantsurrloo: you don't know without checking. the highest of these that the server supports.15:08
rlooTheJulia: i do NOT want to support that :)15:08
TheJuliadtantsur: actually, the negotiation, in theory, can be re-performed... I think15:09
dtantsurrloo: why? this is the most sane semantic IMO15:09
TheJuliaThe code path is at least friendly to it if memory serves15:09
dtantsurand this is exactly what nova needs of us15:10
rloodtantsur: it seems to me that once a version is decided or negotiated (from list of versions), we're just going to use that version under the hood for all future calls. So in L213 to 217 -- client is going to use whatever negotiated version it has already decided.15:10
dtantsurrloo: yep15:10
dtantsurthis is common for all approaches, except for specifying a version per method15:10
*** yangyapeng has quit IRC15:11
rloodtantsur: ok, then that is really a fast way of looping via the hard-coded way of getting a client.15:11
dtantsurrloo: essentially15:11
*** yangyapeng has joined #openstack-ironic15:12
rloodtantsur: so we'd have to modify the nova code to say if version 1.foo is supported, use configdrive. if version 1.bar is supported, go on vacation. etc.15:12
TheJuliarloo: exactly15:12
dtantsurrloo: right, this is the goal of this work15:12
rloodtantsur: so we don't really need to land the 'list of versions' patch in this client.15:13
dtantsurrloo: why?15:13
rloodtantsur: but we do need to land the 'beautiful client, what version have you decided to use'15:13
TheJuliarloo: actually, we do, because we need to define possible versions to use, since we can't constrain without a max_version and latest15:13
rloodtantsur: cuz we can loop on the hardcoded version thing, to get the actual versoin.15:14
dtantsurrloo: yes, and we can use python-requests and through ironicclient away15:14
dtantsurbut we try to help people doing things that they'll have to do every time15:14
rloodtantsur: oh, i am not saying not to do it. just that it isn't urgent do to it for tomorrow.15:14
dtantsurrloo: if we don't want to switch nova now - yes15:15
dtantsurbut then the first patch is not urgent either..15:15
rloodtantsur: we could still switch to nova, as long as we support the hardcoded approach. in nova code, we can get client by trying the versions of interest.15:15
rloodtantsur: and provide a means to know what version the client is actually going to use15:16
dtantsurrloo: yep, and we can do it already, there is a patch by jroll. but it's ugly and error-prone, so we want to hide it inside of ironicclient15:16
*** yangyapeng has quit IRC15:17
*** lucas-hungry is now known as lucas-brb15:17
rloodtantsur: are you talking about the patch that jroll submitted this week?15:17
dtantsuryep15:17
*** e0ne has quit IRC15:18
*** e0ne has joined #openstack-ironic15:19
rloodtantsur: you mean we'd replace the _find_client_with_version() code in https://review.openstack.org/#/c/535786/5/nova/virt/ironic/client_wrapper.py?15:19
patchbotpatch 535786 - nova - Ironic: negotiate microversion to allow downgrade ...15:19
*** e0ne has quit IRC15:20
TheJuliaso, actually you can reach all the way into the http_client object directly and pull out os_ironic_api_version that has been re-negotiated, but I don't think there is any way to know that "yes, this has negotiated" at present15:21
*** dpawar has quit IRC15:21
*** dpawar has joined #openstack-ironic15:22
TheJuliaI'll put it this way, the less code in nova, the better, since they really, to my perception, were downright hostile about us landing negotiation logic to make things better and allow us to just upgrade nova with ironic half-upgraded15:22
rlooTheJulia: if I understand it, we'd change _find_client_with_version (or not have it), by setting kwargs['os_ironic_api_version'] = ['1.34', '1.35'].15:23
TheJuliacorrect, and we can then just microversion guard the logic with the explicit version needed15:23
rlooTheJulia: i read Dan's response, and it makes sense.15:23
openstackgerritDmitry Tantsur proposed openstack/ironic-inspector master: Switch the CI to hardware types  https://review.openstack.org/53741515:23
*** rpioso|afk is now known as rpioso15:24
-openstackstatus- NOTICE: gerrit has been suffering from a full disk, some mails may have been lost in the last couple of hours. we will now restart gerrit to address ongoing slowness, too15:24
rlooTheJulia: wrt microversion guard, that's what jroll's patch does. we'd have to do that for any future stuff. which is ok with me.15:24
rpiosoGood morning, ironicers.15:24
TheJuliarloo: was there one posted to the patch? because I only read the responses in irc15:24
rlooTheJulia: yup.15:24
rlooTheJulia: I didn't read/look at irc discussion15:25
*** dpawar has quit IRC15:26
*** hjensas has quit IRC15:27
TheJuliarloo: and we already have to touch the nova client code every time, but which is cleaner, a new version to a list, or exposed retry logic..... I think doing it in the library makes far more sense for a consumer. Granted, I have strong feelings in this regard because we were using the hard pin in shade and that contributed to my work on it stalling because I was in a catch 22 kind of situation with the15:27
TheJuliastatic pin.15:27
TheJuliathe other advantage of handling it in the client is then we can properly downgrade if we hit an un-upgraded api server after the fact.... instead of possibly dying hard.15:29
rlooTheJulia: the 'exposed retry logic' that I am thinking of, if we don't land the version-list-part, is that the code loops through the list of versions from highest to lowest, to get the supported version. it isn't that difficult.15:29
dtantsur20 lines of code vs 2, something like that15:29
rlooTheJulia: i am not disagreeing that we shouldn't support it in the client code. i am just saying that it isn't urgent to do for tomorrow. if the patch doesn't merge.15:29
*** dprince has joined #openstack-ironic15:30
TheJuliaI would like it to merge, I'd like to fix the code in the nova virt driver, if nova doesn't want us to, then really it is all kind of pointless15:30
*** cjloader has joined #openstack-ironic15:31
rlooTheJulia: the changes in the client to 'get the version' is needed for tomorrow. i think with that, nova code can be changed.15:31
TheJuliaThe bottom line is not doing so, is just pushing a problem back which ultimately doesn't help anyone15:31
TheJuliarloo: no, it can't, because we still have an overriding static pin inside of our code15:31
TheJuliawe _HAVE_ to try and negotiate on something that is not 1.915:32
rlooTheJulia: huh? well, probably not worth spending time in this discussion. i suspect it would be more fruitful to work on each patch to see how far we get.15:32
TheJuliaand have the ability to fallback15:32
*** Guest28399 is now known as mgagne15:33
*** mgagne has joined #openstack-ironic15:33
rlooTheJulia: i don't see how we cannot do the above, w/o providing support for the user to specify a list of versions15:33
*** cjloader has quit IRC15:33
rlooTheJulia: it seems to me that if we land https://review.openstack.org/#/c/531029 we are good.15:33
patchbotpatch 531029 - python-ironicclient - Facilitate latest Rest API use15:33
rlooTheJulia: but i could be wrong. I only started looking at this yesterday and you've been working on it for awhile.15:34
*** cjloader has joined #openstack-ironic15:34
TheJuliarloo: dmitry has comments early on in that patchset that might be impacting my position that we need both15:34
TheJuliarloo: I did also change the logic a little to fail out on maximum locally known, which might make it useful.15:35
TheJuliathe downside of supporting latest is what if behavior changes in 1.39, and python-ironicclient is accidently upgraded15:36
dtantsurthis ^^^15:36
TheJuliaif the client is locked in to support 1.33, 1.34, 1.35, then 1.39 is never possible15:36
*** yolanda has quit IRC15:37
*** etingof has joined #openstack-ironic15:48
*** crushil has quit IRC15:50
*** SerenaFeng has joined #openstack-ironic15:50
*** crushil has joined #openstack-ironic15:51
TheJulia looks like the oneview change that was in the gate is going to have to be rechecked15:51
TheJuliamerge_failure15:51
TheJulialikely gerrit going down... maybe15:52
*** e0ne has joined #openstack-ironic15:52
*** yangyapeng has joined #openstack-ironic15:53
mgoddardhi dtantsur: remember this: https://git.openstack.org/cgit/openstack/nova/commit/?id=c92337bdf80fea4c0a8ebb433bacec4cc07f7a9415:59
mgoddarddtantsur: oh, and this: v16:00
mgoddardhttps://git.openstack.org/cgit/openstack/nova/commit/?id=d25feca90ec4bad6ec9ececedced63b9f00b4c8716:00
dtantsurmgoddard: hi, I do16:00
mgoddarddtantsur: currently seeing warnings because the node is reporting no inventory, so compute can't create allocations against the resource provider in placement16:01
mgoddardseen that before?16:01
*** mjura has quit IRC16:01
dtantsurnot sure16:02
dtantsuron a meeting now, mind pasting your warnings?16:02
NobodyCamGood Morning Ironic'ers16:03
mgoddarddtantsur: http://paste.openstack.org/show/652292/16:03
mgoddarddtantsur: it's not urgent16:03
*** marios has quit IRC16:05
*** AlexeyAbashkin has quit IRC16:05
*** marios has joined #openstack-ironic16:05
rloogood morning NobodyCam16:07
NobodyCamMoring rloo :)16:07
rloodtantsur, TheJulia: I +2'd the first patch, 531029, but to be honest, i am uncomfortable giving it a +A.16:07
*** mjturek has quit IRC16:07
rloodtantsur: so you can do the honours :)16:08
dtantsurheh :) will do, lemme finish with a meeting16:08
*** mjturek has joined #openstack-ironic16:17
*** lucas-brb is now known as lucasagomes16:23
*** jaganathan has quit IRC16:35
*** pcaruana has quit IRC16:35
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Devstack - Add ironic-neutron-agent  https://review.openstack.org/53011716:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Add support to bind type vlan networks  https://review.openstack.org/53234916:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Add dsvm job  https://review.openstack.org/53040916:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Docs - Add multi-tenant devstack - n-g-s vlans  https://review.openstack.org/53445916:38
openstackgerritHarald Jensås proposed openstack/networking-baremetal master: Make the agent distributed using hashring and notifications  https://review.openstack.org/52470916:38
*** e0ne has quit IRC16:40
*** e0ne has joined #openstack-ironic16:40
*** SerenaFeng has quit IRC16:40
*** radez has quit IRC16:44
*** hjensas has joined #openstack-ironic16:47
*** radez has joined #openstack-ironic16:49
*** jcoufal has quit IRC17:02
*** jcoufal has joined #openstack-ironic17:04
*** serlex has quit IRC17:05
*** akhilaki has joined #openstack-ironic17:05
*** akhilaki_ has joined #openstack-ironic17:05
*** jlabarre has quit IRC17:08
*** etingof has quit IRC17:13
*** hjensas has quit IRC17:14
TheJuliarechecked 535296 since zuul never posted results and one of the jobs failed :(17:26
TheJuliaone of the jobs that likes timing out failed....17:26
mgoddard:( thanks TheJulia17:27
openstackgerritMerged openstack/python-ironicclient master: Facilitate latest Rest API use  https://review.openstack.org/53102917:28
*** yolanda has joined #openstack-ironic17:35
*** etingof has joined #openstack-ironic17:36
*** trown|ruck is now known as trown|outtypewww17:40
*** trown|outtypewww is now known as trown|lunch17:40
mjturekdtantsur: fyi - I think I actually have the intermittent failure for building lshw fixed. I was doing 'touch lshw.1' when make builds the file to 'src/lshw.1' https://review.openstack.org/#/c/501799/ could you re-review?17:46
patchbotpatch 501799 - ironic-python-agent - Use lshw in place of dmidecode for the default har...17:46
dtantsurmjturek: will do17:46
mjturekthx!17:46
*** milan_ has quit IRC17:50
*** mjturek has quit IRC17:52
*** mjturek has joined #openstack-ironic17:52
openstackgerritSam Betts proposed openstack/ironic-inspector master: [WIP][DNM] Some beginnings towards removing construct from inspector  https://review.openstack.org/53747217:57
sambettsdtantsur: ^^17:58
dtantsur\o/17:58
dtantsurwill check after dinner17:58
sambettsdtantsur: seems like its not too difficult to recreate what we need from construct, but I've not been through all the cases we currently handle yet17:59
dtantsurcool18:00
*** amarao has quit IRC18:03
*** derekh is now known as derekh_afk18:07
*** sambetts is now known as sambetts|afk18:08
*** hjensas has joined #openstack-ironic18:08
*** tesseract has quit IRC18:08
*** jlabarre has joined #openstack-ironic18:15
bfourniedtantsur: just saw this ^, it looks like construct 2.8.22 is causing problems with CI?18:18
*** vsaienk0 has quit IRC18:19
*** e0ne has quit IRC18:28
pas-habfournie: yes, patch to requirements is https://review.openstack.org/#/c/537378/18:31
patchbotpatch 537378 - requirements - Block construct >2.8.2118:31
*** cjloader has quit IRC18:32
bfourniepas-ha: thanks, is it worth fixing the exceptions for the issue introduced with 2.8.22?18:32
*** cjloader has joined #openstack-ironic18:32
*** fragatina has quit IRC18:32
pas-haeither we fix those - and still make a patch to requirements to require >=2.8.22, or cap it in requirements at 2.8.21 while sambetts is working on getting rid of it completely18:33
pas-hainspector seems to be the only project in openstack/* depending o it anyway18:33
bfourniepas-ha: k, yeah I think it is, getting rid of it entirely by not be that easy though18:34
* pas-ha -> home, have a good day|night all18:34
bfourniepas-ha: good night18:34
*** cjloader has quit IRC18:38
*** cjloader has joined #openstack-ironic18:38
*** mjturek has quit IRC18:52
*** marios has quit IRC18:52
*** mjturek has joined #openstack-ironic18:53
*** trown|lunch is now known as trown18:54
dtantsurfolks, can someone review https://review.openstack.org/#/c/460564/ please? this is loosely related to routed networks18:58
patchbotpatch 460564 - ironic - Allow setting {provisioning,cleaning,rescuing}_net...18:58
dtantsurand it has been around since April, I've just updated it recently18:59
dtantsurwow this is weird http://logs.openstack.org/75/536875/5/check/ironic-tempest-dsvm-pxe_ipmitool-postgres/d72ee02/logs/devstacklog.txt.gz#_2018-01-24_16_27_06_38819:00
dtantsurI wonder if we broken PostgreSQL with the recent traits patch19:01
dtantsurmgoddard: ^^^19:01
*** jtomasek has quit IRC19:05
*** lucasagomes is now known as lucas-afk19:10
*** nicodemos has quit IRC19:13
mgoddarddtantsur: hmm, that does look traity19:19
mgoddarddtantsur: I'd guess it's this: https://review.openstack.org/#/c/532268/13/ironic/db/sqlalchemy/api.py@46619:20
patchbotpatch 532268 - ironic - Add RPC object for traits (MERGED)19:20
rloomgoddard: this is where it seems to start failing: https://review.openstack.org/#/c/532268/819:21
patchbotpatch 532268 - ironic - Add RPC object for traits (MERGED)19:21
mgoddardrloo, dtantsur: we should be able to do the joined query separate from the update. I'll investigate19:21
rloothx mgoddard. sorry if this is going to be another long day for you.19:21
dtantsurthanks!19:21
rloodtantsur: wrt your stuff, i'm guessing you're going to take off soon. and you're not around tomorrow?19:22
dtantsurrloo: yep, not much wifi coverage19:22
rloodtantsur: if so, then we work on the classic driver migration stuff next week after you're back?19:22
*** vsaienk0 has joined #openstack-ironic19:22
*** jkilpatr has quit IRC19:22
dtantsurrloo: there are things to review, like https://review.openstack.org/536875 and the next patches19:23
patchbotpatch 536875 - ironic - Switch the CI to hardware types19:23
rloodtantsur: damn. do they have the same topic?19:23
dtantsurrloo: I will try addressing comments tomorrow and on Friday in the evening19:23
dtantsurrloo: did I ever mention that I'm horrible at gerrit topics? :)19:23
rloodtantsur: https://review.openstack.org/#/q/topic:bug/1690185+(status:open+OR+status:merged)19:23
rloodtantsur: 6 more patches?19:24
dtantsuroh, I'm not that bad :)19:24
*** sasha2 has quit IRC19:24
dtantsuryep, these 6 are all ready to review, I *think* the CI failures are unrelated19:24
dtantsurwell, one needs an update, I'll do it around right now19:24
rloodtantsur: ok, on my list to review them all, this week or Monday the latest.19:24
dtantsurthanks!19:25
bfourniedtantsur: posting fix for the ironic-inspector construct exception issue shortly, I don't think we need any workarounds or need to remove construct19:25
*** sasha2 has joined #openstack-ironic19:25
bfourniedtantsur: is there an upstream bug for that? I assume no19:26
*** pcaruana has joined #openstack-ironic19:27
*** pcaruana has quit IRC19:27
dtantsurbfournie: I'm not sure personally19:28
*** e0ne has joined #openstack-ironic19:28
bfourniedtantsur: adding one...19:28
*** jlabarre has quit IRC19:30
*** muttley has joined #openstack-ironic19:32
*** vsaienk0 has quit IRC19:32
*** AlexeyAbashkin has joined #openstack-ironic19:34
*** e0ne_ has joined #openstack-ironic19:35
*** jlabarre has joined #openstack-ironic19:36
*** jkilpatr has joined #openstack-ironic19:37
*** e0ne has quit IRC19:38
*** harlowja has joined #openstack-ironic19:38
*** AlexeyAbashkin has quit IRC19:38
*** fragatina has joined #openstack-ironic19:40
openstackgerritBob Fournier proposed openstack/ironic-inspector master: Remove use of construct lib FieldError exception  https://review.openstack.org/53762719:41
openstackgerritDmitry Tantsur proposed openstack/ironic master: Deprecate classic drivers  https://review.openstack.org/53692819:45
dtantsurokay, that should be it ^^^19:46
*** dtantsur is now known as dtantsur|afk19:46
dtantsur|afko/19:46
*** muttley has quit IRC19:48
*** athomas has quit IRC19:50
*** etingof has quit IRC19:53
*** etingof has joined #openstack-ironic20:07
rpiosoShould the base class definition of a new, optional method return a value or raise an UnsupportedDriverExtension exception?20:09
rpiosos/class/hardware interface class/20:10
openstackgerritMark Goddard proposed openstack/ironic master: Fix node update with PostgreSQL  https://review.openstack.org/53763720:13
mgoddardrloo: hopefully that should fix our postgres issue ^^^20:14
rloomgoddard: thx, will look in a few minutes.20:15
rloomgoddard: i reviewed the client patch and it needs an update. if you feel like it. i think we can wait for jroll to update it later/tomorrow though.20:15
openstackgerritJohn L. Villalovos proposed openstack/ironic-lib master: Add missing 'self' parameter to class methods  https://review.openstack.org/53764020:19
*** e0ne_ has quit IRC20:19
openstackgerritMerged openstack/python-ironicclient master: Allow API user to define list of versions  https://review.openstack.org/53127120:20
TheJuliais there a legitimate reason we are running cleaning in a multinode job?20:22
TheJuliaI think we need to consider extending some timeouts at least, since the last attempt to land rpc for traits blew out because the nodes were still cleaning20:23
TheJuliarloo: ^^^ ?20:23
TheJuliajroll: if you happen to remember20:23
rlooTheJulia: i think we just need to run cleaning in *a* test20:24
TheJuliaWe have toooooo many long running jobs :( and on multinode... it is insane20:24
TheJuliabecause that spins.... ?5? vms per node20:24
rlooTheJulia: my guess is the multinode test is testing the same stuff the main grenade test. it should be sufficient to test cleaning in the single-node grenade test20:24
rlooTheJulia: i mean, if you're paranoid and a tester, you want to test EVERYTHING!20:25
rlooTheJulia: cuz who knows what might break. it's a black box.20:25
rlooTheJulia: you know, what if a node is being cleaned, the conductor gets shut down/upgraded, and another conductor inherits the node being cleaned and ... well, i know, we're not explicitly testing that :)20:26
TheJuliarloo: it is cleaning that is happening as part of initial stand-up20:27
rlooTheJulia: having said that, what is cleaning doing? I've already forgotten. We should just mock something, add a cleaning thing in ipa that just waits 1 minute or something before finishing.20:27
rlooTheJulia: we could just turn off cleaning via the configs, for multinode grenade. it is non voting now anyway, right?20:28
TheJuliathis is not multinode grenade20:28
TheJuliait is just multinode20:28
rlooTheJulia: oh, sorry. i missed that.20:28
TheJuliano, my fault, I only said multinode20:28
mjturekI hate to ask, but should I be asking for a feature freeze exception on this patch? Is it too late? https://review.openstack.org/#/c/501799/20:29
patchbotpatch 501799 - ironic-python-agent - Use lshw in place of dmidecode for the default har...20:29
rloomjturek: is it a feature or a bug?20:29
rloomjturek: looks like a bug. no need for ffe20:29
mjturekrloo: is the rule basically, if it's an RFE, needs a FFE?20:30
rloomjturek: yup. FFE = feature freeze extension. For features. RFE = request for Feature Enhancement I think. confusing...20:30
mjturekand thank you :)20:30
* TheJulia considers it a bug/defficency for non-x8620:30
jrollrloo: TheJulia: I'm finally back, what can I help out with?20:30
TheJulialiquor? lots of liquor?20:31
mjturekawesome, thank you!20:31
rloojroll: answer TheJulia's question about cleaning ^^20:31
* jroll fetches the bourbon20:31
rloojroll: update your traits client patch20:31
*** hw_wutianwei has quit IRC20:31
rloojroll: review https://review.openstack.org/#/c/537637/. traits broke postgres20:31
jrollTheJulia: I don't know why we currently run cleaning20:31
patchbotpatch 537637 - ironic - Fix node update with PostgreSQL20:31
jrolllol postgres20:32
rloojroll: let me know when you'd like more stuff to do. and welcome back :D20:32
*** hw_wutianwei has joined #openstack-ironic20:32
jroll:P20:32
TheJuliaomfg, we only don't clean on standalone runs20:33
rlooTheJulia: for now, i see two choices, turn of cleaning, or extend timeout. maybe something to discuss (again) at ptg.... sigh.20:33
TheJuliahow much disk time are we wasting!?!?!?!?!20:33
rlooTheJulia: wow, are we that thorough?20:33
rloomy positive spin on that... :-(20:33
openstackgerritJulia Kreger proposed openstack/ironic master: Be more sane about cleaning  https://review.openstack.org/53765020:45
*** jtomasek has joined #openstack-ironic20:45
TheJuliaso the main grenade job also runs 7 VMs and cleaning enabled :\20:46
TheJuliajroll: rloo: jlvillal: and anyone else  https://review.openstack.org/53765020:47
patchbotpatch 537650 - ironic - Be more sane about cleaning20:47
TheJuliaI pretty much think that just needs to be landed at this point, there is no way we're going to handle a busy gate without cutting at least some fat off of our jobs20:49
rlooTheJulia: saner is always better! looking...20:50
* jroll tries to remember the benefit of master multinode20:53
*** jlabarre has quit IRC20:55
openstackgerritJulia Kreger proposed openstack/ironic master: Be more sane about cleaning  https://review.openstack.org/53765020:58
TheJuliajroll: replied20:58
jrollTheJulia: yeah fair20:59
* jroll is +1 either way20:59
*** e0ne has joined #openstack-ironic21:00
TheJuliarloo: jlvillal updated commit message, please review, if so I'll rebase and re-approve one of our troublesome changes on top of it21:01
rlooTheJulia: i just reviewed21:01
rlooTheJulia: do you want to fix that one thing (which is more important that the commit, heh). or i can just +2. let me know.21:02
TheJuliaI can fix the text later :)21:03
TheJuliaI just want to get things stacked and running so hopefully things land more reliable21:03
TheJuliareliably21:03
rlooTheJulia: ok21:03
*** jtomasek has quit IRC21:03
rlooTheJulia: heh. looks like i ended up voting on ps2 anyway :)21:04
rloojlvillal: if you want to use your recheck powers, 535296.21:07
jlvillalrloo, Sure :)21:07
TheJuliawait21:07
openstackgerritYusef Shaban proposed openstack/sushy master: Restores sushy session functionality.  https://review.openstack.org/53591021:08
rloojlvillal: ^^ TheJulia: did you want to rebase that?21:08
TheJuliayeah, we rebase it on top of the test changes21:08
TheJuliasince it has hit cleaning timeout and overall job timeout twice on that21:09
rlooTheJulia: ok, actually i think it is fine for jlvillal to add to his recheck thing.21:09
TheJuliabut then we might for for hours with a bad run due to recheck21:09
jlvillalTheJulia, rloo: Rebasing won't impact the 'recheck' script. It will only recheck if there is a failure.21:10
jlvillalAnd someone already did a recheck on it.21:10
jlvillalGetting details for patch: 53529621:10
jlvillalhttps://review.openstack.org/#/c/535296/ Add RPC API and conductor manager for traits21:10
jlvillalZuul voted: -2 at 2018-01-23 22:22:03-08:00 localtime21:10
jlvillalCheck to see if we should post 'recheck' message ...21:10
jlvillalBuild Failed date: 2018-01-23 22:22:03-08:00 localtime21:10
patchbotpatch 535296 - ironic - Add RPC API and conductor manager for traits21:10
jlvillalrecheck date: 2018-01-24 09:25:15-08:00 localtime21:10
jlvillal'recheck' message already posted at: 2018-01-24 03:57:38-08:00, 2018-01-24 09:25:15-08:0021:10
TheJuliayeah, and it is running, will fail shortly. If we rebase, then the new config will be in place21:10
*** amarao has joined #openstack-ironic21:11
TheJulialooks like you two want to wait for results to come back before approving21:12
TheJuliaso then lets just recheck 535296 again by hand.....21:12
TheJuliabecause we won't have results for at least four hours21:12
rlooTheJulia: OH. right, if we rebase, we'll have to wait for that to merge first. So yeah, let's keep rechecking until then.21:12
jlvillalTheJulia, I think should do the rebase most likely. The patch has already failed in zuul.21:13
rlooTheJulia: i wanted to see the results first, but I guess it doesn't matter. it'll merge if they pass, and we aren't that worried about non-voting jobs.21:13
jlvillalhttp://zuul.openstack.org/21:13
jlvillalPatch failed on multinode job21:13
jlvillalhttps://review.openstack.org/53529621:14
patchbotpatch 535296 - ironic - Add RPC API and conductor manager for traits21:14
rlooheh, TheJulia: you probably don't want that topic for https://review.openstack.org/#/c/537650/21:16
patchbotpatch 537650 - ironic - Be more sane about cleaning21:16
TheJuliaheh21:18
TheJuliayay for git review21:18
*** e0ne has quit IRC21:20
openstackgerritJulia Kreger proposed openstack/ironic master: Add RPC API and conductor manager for traits  https://review.openstack.org/53529621:23
* TheJulia crosses fingers21:23
*** amarao has quit IRC21:30
*** dprince has quit IRC21:30
*** weshay|mtg is now known as weshay|rover21:34
*** trown is now known as trown|ruck21:34
*** threestrands has joined #openstack-ironic21:41
*** threestrands_ has joined #openstack-ironic21:44
*** threestrands_ has quit IRC21:44
*** threestrands_ has joined #openstack-ironic21:44
*** threestrands has quit IRC21:47
*** trown|ruck is now known as trown|outtypewww22:03
*** baha has quit IRC22:04
*** jjohnson2 has quit IRC22:08
*** rmcall has quit IRC22:09
*** Goneri has quit IRC22:11
*** akhilaki_ has quit IRC22:13
*** akhilaki has quit IRC22:13
*** rcernin has joined #openstack-ironic22:14
*** yangyapeng has quit IRC22:16
*** bfournie has quit IRC22:19
*** cdearborn has quit IRC22:25
*** crushil has quit IRC22:26
*** rcernin has quit IRC22:29
*** jrist has quit IRC22:29
*** jrist has joined #openstack-ironic22:30
*** rcernin has joined #openstack-ironic22:31
openstackgerritJulia Kreger proposed openstack/ironic master: Implementation for UEFI iSCSI boot for ILO  https://review.openstack.org/46828822:52
openstackgerritJulia Kreger proposed openstack/ironic master: Implementation for UEFI iSCSI boot for ILO  https://review.openstack.org/46828822:58
*** yangyapeng has joined #openstack-ironic23:01
*** sasha2 has quit IRC23:16
*** derekh_afk has quit IRC23:33
*** rpioso is now known as rpioso|afk23:42
openstackgerritMerged openstack/ironic-lib master: Updated from global requirements  https://review.openstack.org/53705323:54
*** rbudden has quit IRC23:55

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