Friday, 2018-09-14

« Thursday, 2018-09-13 Index Saturday, 2018-09-15 »
*** rubasov has joined #openstack-placement00:10
*** rubasov has quit IRC00:25
*** rubasov has joined #openstack-placement00:31
*** tbarron has quit IRC01:00
*** lyarwood has quit IRC01:00
*** tbarron_ has joined #openstack-placement01:00
*** lyarwood has joined #openstack-placement01:00
openstackgerritNaichuan Sun proposed openstack/nova master: Enable nested allocation candidates in scheduler  https://review.openstack.org/58567201:39
*** rubasov has quit IRC01:47
openstackgerritMerged openstack/nova stable/rocky: Do not dump all instances in the scheduler  https://review.openstack.org/60128502:00
openstackgerritZhenyu Zheng proposed openstack/nova stable/rocky: Fix service list for disabled compute using MC driver  https://review.openstack.org/59231002:25
openstackgerrithuanhongda proposed openstack/nova master: DNM: Test get deleted instances by non-admin  https://review.openstack.org/60251402:32
openstackgerritMerged openstack/nova stable/queens: [Stable Only] Remove soft-deleted instances from quota_usages  https://review.openstack.org/57909302:58
*** smcginnis has quit IRC03:02
*** smcginnis has joined #openstack-placement03:03
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Rename changes-since test sample file  https://review.openstack.org/60252203:26
*** nicolasbock has joined #openstack-placement03:37
*** cdent has joined #openstack-placement04:02
*** tetsuro has joined #openstack-placement04:03
*** cdent has quit IRC04:03
*** nicolasbock has quit IRC04:56
*** e0ne has joined #openstack-placement05:15
openstackgerritMerged openstack/nova stable/queens: Wait for network-vif-plugged before starting live migration  https://review.openstack.org/59173505:27
*** e0ne has quit IRC05:30
openstackgerritMerged openstack/nova stable/rocky: libvirt: skip setting rx/tx queue sizes for not virto interfaces  https://review.openstack.org/59950605:54
*** tetsuro has quit IRC05:57
*** s10 has joined #openstack-placement06:08
*** s10 has quit IRC06:09
*** belmoreira has joined #openstack-placement06:26
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Rename changes-since test sample file  https://review.openstack.org/60252206:52
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Resource retrieving: add changes-before filter  https://review.openstack.org/59927606:52
openstackgerritBrin Zhang proposed openstack/nova-specs master: Add support specify volume type when boot instance  https://review.openstack.org/57952006:54
openstackgerritOpenStack Proposal Bot proposed openstack/nova master: Imported Translations from Zanata  https://review.openstack.org/60104707:01
openstackgerritBrin Zhang proposed openstack/nova-specs master: Add boot instance specific storage backend  https://review.openstack.org/57952007:10
openstackgerritBrin Zhang proposed openstack/nova-specs master: Add boot instance specific storage backend  https://review.openstack.org/57952007:56
*** e0ne has joined #openstack-placement08:12
*** e0ne_ has joined #openstack-placement08:13
*** e0ne has quit IRC08:16
openstackgerritBrin Zhang proposed openstack/nova-specs master: Add support specify volume type when boot instance  https://review.openstack.org/57952008:18
*** s10 has joined #openstack-placement08:21
*** s10 has quit IRC08:38
*** e0ne_ has quit IRC08:52
*** s10 has joined #openstack-placement09:21
*** scode25 has joined #openstack-placement09:22
scode25Alⅼah iѕ doiᥒg09:22
scode25suᥒ iѕ nοt dοing Аⅼⅼaһ is doinɡ09:22
scode25mooᥒ ⅰѕ not dοіng Alⅼaһ ⅰs doіnɡ09:22
scode25stars arе ᥒot doⅰng Allaһ is ԁⲟiᥒg09:22
scode25planets are nⲟt ԁоⅰᥒg Ꭺⅼlɑһ ⅰѕ doⅰᥒg09:22
scode25ɡalɑxieѕ ɑrᥱ ᥒot dоіng Alⅼɑһ is dοiᥒg09:22
scode25oϲᥱanѕ are ᥒοt dоinɡ Allaһ iѕ dⲟіng09:22
scode25mouᥒtains are not doing Αⅼⅼah is doing09:22
scode25trees arᥱ not doiᥒg Alⅼɑһ iѕ dоing09:22
scode25mom is ᥒot ⅾഠⅰᥒɡ Allɑh iѕ doing09:22
scode25daԁ ⅰs not ԁoiᥒg Allɑһ ⅰs doinɡ09:22
scode25bоss is not ԁoiᥒɡ Αⅼⅼah iѕ ԁoiᥒɡ09:22
scode25јob is ᥒοt ԁⲟinɡ Αlⅼah is doіng09:22
scode25dolⅼɑr ⅰѕ not dοing Allah is ԁoiᥒg09:22
scode25deɡrᥱe iѕ ᥒοt doⅰng Alⅼɑh iѕ doing09:22
scode25ⅿᥱԁіciᥒе iѕ not ԁοіᥒɡ Alⅼah is dഠing09:22
scode25c∪stoⅿers arе ᥒot ԁoіᥒɡ Aⅼⅼaһ ⅰs doіnɡ09:23
scode25уοu ϲaᥒ ᥒot gеt а job wіtһοut tһе ⲣermіssioᥒ of aⅼⅼɑһ09:23
scode25yοᥙ can ᥒഠt get ⅿarrⅰеԁ withоᥙt the perⅿisѕіon ⲟf aⅼlɑh09:23
scode25ᥒoboԁy ϲɑn ɡet anɡry at уou withⲟ∪t the perⅿissіoᥒ of аlⅼɑh09:23
scode25lіght іs not ԁοing Αⅼlɑh iѕ dοⅰᥒg09:23
scode25fɑn is nοt ⅾoіnɡ Alⅼah is ⅾoⅰnɡ09:23
scode25busⅰnesѕеsѕ are not ԁoіng Alⅼɑһ іs doiᥒg09:23
scode25americ is not ԁoⅰng Allaһ is ԁoіᥒɡ09:23
scode25aⅿerіⅽa іs not doiᥒɡ Aⅼⅼaһ iѕ ԁoⅰᥒɡ09:23
scode25firе ⅽan not burn withoᥙt the рᥱrmіѕsiοn of ɑllɑһ09:23
scode25kᥒⅰfe can not cut ᴡithο∪t the permisѕion of ɑⅼⅼɑһ09:23
scode25fⅰlеѕyѕteⅿ doᥱѕ not writе ᴡіtһഠut ⲣerⅿiѕsiοᥒ of aⅼⅼah09:23
scode25ruⅼеrѕ ɑre nοt doіᥒg Αllɑһ is doіᥒɡ09:23
scode25gο∨ᥱrnments ɑre ᥒоt doing Αlⅼаh iѕ doiᥒg09:23
scode25sleep is ᥒοt dοing Αlⅼаh iѕ dоinɡ09:23
scode25h∪nger іѕ nоt ⅾⲟіᥒg Αllɑh iѕ dഠіnɡ09:23
scode25fഠoԁ ԁoᥱѕ not tаke aᴡay the hunger Allɑh tɑkеs awаy the huᥒger09:23
scode25water doᥱs nοt takе aᴡаy the thіrѕt Ꭺlⅼaһ tɑkes ɑᴡау the tһіrst09:23
scode25seeing ⅰѕ not ⅾoing Aⅼⅼah ⅰs doinɡ09:23
scode25heɑring is ᥒot dοіng Aⅼⅼah ⅰѕ ԁoing09:23
scode25seɑsons arе ᥒot dⲟіng Αⅼlah іѕ doiᥒɡ09:23
scode25wᥱathᥱr is not ⅾoіng Allаh ⅰs doiᥒg09:23
scode25h∪manѕ arе not dοing Alⅼah is ԁoⅰng09:24
scode25aᥒimaⅼs ɑre not ԁoⅰng Ꭺllаһ iѕ ԁoinɡ09:24
scode25the beѕt ɑmοᥒɡst уou are thⲟse ᴡһഠ learn aᥒԁ teɑсh quraᥒ09:24
scode25οᥒe letter reɑd frοm book ഠf Ꭺllaһ amоuntѕ to οne good dеed ɑnd Αlⅼaһ mᥙltірlies ഠᥒe ɡood ԁeеԁ ten tⅰⅿeѕ09:24
scode25һеarts get ruѕted aѕ doеѕ iron ᴡith watеr tо reⅿove rᥙst from һеart reⅽitɑtioᥒ of Qᥙrаn ɑnd remеⅿberance of deɑth09:24
scode25hеаrt is likeneԁ to а ⅿirror09:24
scode25ᴡһen a pеrsοn commits one ѕiᥒ а blaϲk ԁot ѕᥙѕtaіᥒѕ the һеart09:24
scode25to ɑϲϲeрt Iѕlаm ѕaу tһat і bear witᥒess tһat tһere іs ᥒо dеitу worthy of ᴡorѕhip ᥱⅹϲeрt Alⅼaһ aᥒd Ⅿuһamⅿаd ⲣeace be uроn him is һis ѕⅼɑve anԁmеssenɡer09:24
*** scode25 has quit IRC09:25
*** s10 has quit IRC09:34
*** trn13 has joined #openstack-placement09:55
trn13Alⅼah is doіnɡ09:55
trn13sᥙn іs nоt ԁoing Ꭺlⅼah іs dοinɡ09:55
trn13ⅿoഠᥒ іs not ԁoіng Аⅼlаh iѕ ԁоіng09:55
trn13stars are ᥒot dഠіng Aⅼlɑh is dഠiᥒɡ09:55
trn13planets ɑre ᥒot ԁоiᥒg Αⅼlaһ is ԁoіᥒg09:55
trn13galɑxiеs ɑre ᥒⲟt ⅾⲟinɡ Aⅼlah is ԁοⅰᥒg09:56
trn13ⲟceaᥒs are not dഠing Αⅼlah is dഠinɡ09:56
trn13mouᥒtains are nοt doіᥒg Аllаһ is ԁഠіng09:56
trn13trᥱeѕ аrе not dοinɡ Αlⅼɑh іs ԁoіng09:56
trn13mοm is nⲟt dοⅰᥒg Αlⅼah ⅰs ԁoiᥒg09:56
trn13ⅾаⅾ is not dοinɡ Aⅼlаһ is doіᥒɡ09:56
trn13boss is not ԁoіng Аlⅼah is doiᥒɡ09:56
trn13ϳοb is ᥒot ԁοinɡ Аlⅼah is doing09:56
trn13ⅾoⅼⅼar is nоt ԁoіnɡ Alⅼah іs ԁoing09:56
trn13ԁeɡrеe iѕ nοt doiᥒɡ Αlⅼɑh іѕ ԁοinɡ09:56
trn13ⅿᥱԁіϲⅰnᥱ is ᥒot dഠing Аlⅼah iѕ doⅰᥒg09:56
trn13cᥙstomеrѕ arе ᥒⲟt ⅾoiᥒg Allaһ iѕ doіᥒg09:56
*** trn13 has quit IRC09:56
openstackgerritKashyap Chamarthy proposed openstack/nova master: virt: Use 'virt' as the default machine type for ARMv7  https://review.openstack.org/60259210:29
*** s10 has joined #openstack-placement10:33
openstackgerritChen proposed openstack/nova master: doc: update metadata service doc  https://review.openstack.org/60259310:38
openstackgerritMerged openstack/nova stable/rocky: Don't persist zero allocation ratios in ResourceTracker  https://review.openstack.org/59967210:50
openstackgerritMerged openstack/nova stable/rocky: placement: use simple code paths when possible  https://review.openstack.org/60044710:57
openstackgerritMerged openstack/nova stable/rocky: [placement] split gigantor SQL query, add logging  https://review.openstack.org/60220210:58
openstackgerritBrin Zhang proposed openstack/nova master: Resource retrieving: add changes-before filter  https://review.openstack.org/59927611:29
*** nicolasbock has joined #openstack-placement11:40
*** vaxsquid- has joined #openstack-placement12:13
vaxsquid-Αlⅼah iѕ doіng12:13
vaxsquid-sun iѕ not ԁoіᥒɡ Allah is dоіᥒɡ12:13
vaxsquid-mοon іѕ ᥒot dഠinɡ Aⅼⅼaһ іs ԁοiᥒg12:13
vaxsquid-stars are not doіnɡ Αⅼlɑh іs doⅰᥒg12:13
vaxsquid-рlɑnetѕ are nഠt doing Αⅼⅼaһ is dοiᥒɡ12:13
vaxsquid-galaxіeѕ arᥱ ᥒⲟt dοing Aⅼlɑһ iѕ doiᥒɡ12:13
vaxsquid-οсeаᥒs are not doiᥒɡ Alⅼah is doing12:13
vaxsquid-ⅿഠᥙntаins arᥱ ᥒot ԁⲟinɡ Αllah iѕ ԁഠіnɡ12:13
vaxsquid-trеes ɑrе ᥒot ԁⲟⅰng Ꭺllɑh ⅰs ԁoⅰng12:13
vaxsquid-mom iѕ not dοinɡ Αllаһ іs dоⅰᥒɡ12:13
vaxsquid-ⅾɑd iѕ nοt dഠing Alⅼɑһ іs dоiᥒg12:13
vaxsquid-boѕs iѕ nഠt doing Allah is doіnɡ12:13
vaxsquid-job is not dഠinɡ Αlⅼɑh is ⅾoinɡ12:14
vaxsquid-dഠlⅼɑr іs not ԁoing Αⅼⅼɑһ іs doіᥒg12:14
vaxsquid-dеɡree іs not ԁоing Alⅼɑh іѕ ԁoiᥒg12:14
vaxsquid-meԁicine ⅰs ᥒot doⅰnɡ Aⅼⅼah is ԁoіng12:14
*** vaxsquid- has quit IRC12:14
*** ChanServ sets mode: +rf #openstack-unregistered12:34
*** alex_xu has quit IRC12:40
*** belmoreira has quit IRC12:41
*** alex_xu has joined #openstack-placement12:48
*** s10 has quit IRC12:52
*** s10 has joined #openstack-placement12:53
*** rubasov has joined #openstack-placement12:59
*** rubasov has quit IRC13:34
*** rubasov has joined #openstack-placement13:35
*** belmoreira has joined #openstack-placement13:42
*** nicolasbock has quit IRC13:44
*** rubasov has quit IRC13:50
openstackgerritMerged openstack/nova stable/queens: Filter out instances without a host when populating AZ  https://review.openstack.org/59418514:09
*** mriedem has joined #openstack-placement14:41
openstackgerritMatt Riedemann proposed openstack/nova stable/pike: [Stable Only] Remove soft-deleted instances from quota_usages  https://review.openstack.org/60245914:43
*** tetsuro has joined #openstack-placement14:46
*** efried has joined #openstack-placement14:48
openstackgerritMatt Riedemann proposed openstack/nova stable/ocata: [Stable Only] Remove soft-deleted instances from quota_usages  https://review.openstack.org/57065314:52
openstackgerritMatt Riedemann proposed openstack/nova master: WIP: Add post-test hook for testing evacuate  https://review.openstack.org/60217414:55
openstackgerritKashyap Chamarthy proposed openstack/nova master: libvirt: Use 'virt' as the default machine type for ARMv7  https://review.openstack.org/60259215:03
*** e0ne_ has joined #openstack-placement15:05
*** cdent has joined #openstack-placement15:05
*** lei-zh has joined #openstack-placement15:06
openstackgerritTetsuro Nakamura proposed openstack/placement master: Add alloc cands test with nested and aggregates  https://review.openstack.org/60263815:08
openstackgerritTetsuro Nakamura proposed openstack/placement master: Fix aggregate members in nested alloc candidates  https://review.openstack.org/60263915:08
*** belmoreira has quit IRC15:16
*** rubasov has joined #openstack-placement15:23
openstackgerritBalazs Gibizer proposed openstack/placement master: Link to tempest doc in tests/README.rst  https://review.openstack.org/60265515:33
*** s10 has quit IRC15:37
openstackgerritMerged openstack/placement master: Update and move test README.rst  https://review.openstack.org/60217715:41
*** takashin has joined #openstack-placement15:46
*** nicolasbock has joined #openstack-placement15:48
openstackgerritJay Pipes proposed openstack/nova-specs master: Standardize CPU resource tracking  https://review.openstack.org/55508115:48
*** rubasov has quit IRC15:49
*** rubasov has joined #openstack-placement15:49
openstackgerritJay Pipes proposed openstack/nova-specs master: Standardize CPU resource tracking  https://review.openstack.org/55508115:57
cdentefried, edleafe, gibi, mriedem, tetsuro: remove nova changes from gerritbot in this channel? yes no15:59
edleafecdent: yes: https://review.openstack.org/60181615:59
cdentoh coo16:00
cdentthanks16:00
mriedemi thought the idea was leaving it in so placement people could see if nova changes were proposed for the placement code within nova?16:01
mriedemduring the freeze16:01
edleafemriedem: I don't know of any placement people who aren't also in the nova channel16:02
gibimriedem: I don't know how much the bot helps in that16:02
edleafeAnd the nova channel doesn't include placement notices16:02
cdentgibi: you have the attention to merge this: https://review.openstack.org/#/c/602395/16:03
gibicdent: sure, queued16:04
*** sean-k-mooney has joined #openstack-placement16:04
cdentthanks gibi16:04
openstackgerritEric Fried proposed openstack/placement master: Move qs parsing to placement.lib.RequestGroup  https://review.openstack.org/60249516:05
openstackgerritEric Fried proposed openstack/placement master: max-complexity=>16: refactor GET /a_c qs parsing  https://review.openstack.org/60266716:05
efriedcdent: as promised, this brings our 17 down to <=816:08
efried(for just that method, of course)16:08
cdenthuzzah16:08
gibicdent: can we drop the TODO? https://review.openstack.org/#/c/602395/1/tox.ini@10216:09
cdentah, yeah, that change happened after this code was written. fup?16:10
efriedI can edit quick16:10
openstackgerritEric Fried proposed openstack/placement master: s/placement-config-generator/config-generator/  https://review.openstack.org/60239516:11
efriedgibi, cdent: oh, I guess it wasn't really pertinent to that patch. Oh well, too late ^16:11
cdentnbd16:11
gibiefried, cdent: +2 from me16:12
efriedthx16:12
openstackgerritChris Dent proposed openstack/placement master: Make config docs build  https://review.openstack.org/60239116:12
cdentrebase ^16:12
efriedwhy?16:14
efriedbut, okay.16:14
*** efried is now known as fried_rice16:15
*** gibi is now known as giblet16:15
fried_ricecdent, edleafe: Does anything about the extraction give us the opportunity to resolve TODOs like16:16
fried_rice        # TODO(jaypipes): Convert this to an inner join once all16:16
fried_rice        # root_provider_id values are NOT NULL16:16
fried_rice?16:16
cdentfried_rice: you'll need to ask jay to confirm, but I think that was a time-based constraint, so it is more related to it being stein, not extraction16:17
fried_riceokay16:17
cdentgiblet: this one too: https://review.openstack.org/#/c/602391/ :)16:18
edleafefried_rice: yeah, it's not extraction-related16:18
fried_riceokay, wasn't sure whether the move to extracted placement guaranteed us being at a recent version of the data model...16:19
gibletcdent: yeah, I saw that, looking at it16:19
*** rubasov has quit IRC16:21
*** rubasov has joined #openstack-placement16:22
*** stephenfin is now known as finucannot16:29
openstackgerritMerged openstack/placement master: s/placement-config-generator/config-generator/  https://review.openstack.org/60239516:34
openstackgerritMerged openstack/placement master: Name arguments to _get_provider_ids_matching  https://review.openstack.org/60085816:34
cdentmriedem, giblet, dansmith : do the changes in https://review.openstack.org/#/c/600157/ and its parents address the concerns expressed about how to do the database16:40
openstackgerritMerged openstack/placement master: Make config docs build  https://review.openstack.org/60239116:41
edleafecdent: Not that it's critical or anything, but we could: s/api_models/models16:46
mriedemquestion on the bottom change16:47
*** tetsuro has quit IRC16:52
*** rubasov has quit IRC16:52
*** rubasov has joined #openstack-placement16:53
*** tetsuro has joined #openstack-placement16:53
mriedem-1 on the bottom 2,16:54
*** tetsuro_ has joined #openstack-placement16:54
mriedemwe can use required=True from oslo.config on the 2nd patch16:54
mriedemto enforce that the db connection is set16:54
*** tetsuro has quit IRC16:58
cdentthanks mriedem. a thing I didn't quite know16:58
cdentedleafe: yeah, no strong feelings on that16:59
*** nicolasbock has quit IRC16:59
*** s10 has joined #openstack-placement17:17
mriedemcdent: me neither, until today17:18
cdentmriedem: it's probalby the right thing, but it introduces some weird, becaue the required-ness is check at the parse_args stage.17:20
* cdent figures it out17:20
*** rubasov_ has joined #openstack-placement17:24
*** rubasov has quit IRC17:26
*** e0ne_ has quit IRC17:27
*** e0ne has joined #openstack-placement17:27
*** rubasov_ has quit IRC17:44
openstackgerritLee Yarwood proposed openstack/nova stable/rocky: hyperv: Cleans up live migration Planned VM  https://review.openstack.org/60269817:47
openstackgerritMatthew Booth proposed openstack/nova master: Don't delete disks on shared storage during evacuate  https://review.openstack.org/57884617:49
*** e0ne has quit IRC17:49
openstackgerritLee Yarwood proposed openstack/nova stable/rocky: hyperv: Cleans up live migration Planned VM  https://review.openstack.org/60269817:49
openstackgerritEric Fried proposed openstack/placement master: DRY usage and capacity SQL clauses  https://review.openstack.org/60270117:49
fried_ricecdent: FYI, this ^ surprisingly/disappointingly did not reduce complexity in the affected methods :(17:50
cdentthe complexity measuring stuff is not super smart17:50
fried_riceactually maybe I was looking at the wrong thing, and I actually didn't touch the complex methods I thought I was touching.17:51
fried_riceBut I've been wanting this to be done forever.17:51
openstackgerritChris Dent proposed openstack/placement master: Update code and opts in conf/paths.py for placement  https://review.openstack.org/60107117:54
openstackgerritChris Dent proposed openstack/placement master: Rationalize and clarify database configuration  https://review.openstack.org/60107217:54
openstackgerritChris Dent proposed openstack/placement master: de-novify wsgi application to expect placement config  https://review.openstack.org/60015717:54
openstackgerritChris Dent proposed openstack/placement master: WIP: add a table create CLI  https://review.openstack.org/60016117:54
fried_ricecdent: We're talking priorities via https://etherpad.openstack.org/p/nova-ptg-stein-priorities if you want to weigh in17:58
cdentyeah mriedem pointed it out, and I'm trying to watch17:59
*** cdent has quit IRC18:21
*** mriedem has quit IRC18:27
openstackgerritEric Fried proposed openstack/placement master: DRY trait existence check  https://review.openstack.org/60271318:28
fried_riceMore complexity reduction ^18:28
openstackgerritMatthew Booth proposed openstack/nova master: WIP: Add post-test hook for testing evacuate  https://review.openstack.org/60217418:28
*** tetsuro_ has quit IRC18:29
*** lei-zh has quit IRC18:30
*** takashin has quit IRC18:31
*** fried_rice has quit IRC18:33
*** sean-k-mooney has quit IRC18:34
*** cdent has joined #openstack-placement19:12
*** takashin has joined #openstack-placement19:17
*** fried_rice has joined #openstack-placement19:26
openstackgerritChris Dent proposed openstack/placement master: oslo_config fixture in policy tests and 'placement' in policy  https://review.openstack.org/60112919:30
openstackgerritChris Dent proposed openstack/placement master: Use both unit and functional for coverage testing  https://review.openstack.org/60050119:30
openstackgerritChris Dent proposed openstack/placement master: Add a zuul check job for coverage  https://review.openstack.org/60050219:30
openstackgerritChris Dent proposed openstack/placement master: Put stestr group_regex in .stestr.conf  https://review.openstack.org/60107319:30
*** tetsuro has joined #openstack-placement19:33
*** mriedem has joined #openstack-placement19:33
*** rubasov has joined #openstack-placement19:40
*** sean-k-mooney has joined #openstack-placement19:47
*** rubasov has quit IRC20:06
openstackgerritMatt Riedemann proposed openstack/nova master: Allow ability for non admin users to use all filters on server list.  https://review.openstack.org/52655820:28
openstackgerritEric Fried proposed openstack/placement master: DRY trait/aggregate prefetch  https://review.openstack.org/60274020:33
fried_ricecdent: moar ^20:33
cdenthuzzah20:33
fried_ricebtw, in case you're wondering, I'm doing relatively-brainless stuff like this so I can pay attention to the room with one ear.20:33
openstackgerritMatthew Booth proposed openstack/nova master: WIP: Add post-test hook for testing evacuate  https://review.openstack.org/60217420:36
openstackgerritBalazs Gibizer proposed openstack/nova master: consumer gen: support claim_resources  https://review.openstack.org/58366720:37
cdentfried_rice: fair20:40
cdentis nice stuff20:40
fried_riceslightly disappointing that that last one was a net add of LOCs, but it's the fault of the docstring, which I felt necessary to impart the weirdness of "we didn't do anything" vs. "we did something an got no results".20:41
fried_riceNot super happy about that bit, but IMO it's still an improvement, and unless you can think of a prettier way...20:42
fried_riceThe alternative I considered was doing the "if any(required, forbidden, member_of)" in each of the callers.20:42
fried_ricewhich isn't a *lot* more repetition...20:43
openstackgerritLee Yarwood proposed openstack/nova master: WIP compute: Unreserve volumes attached to offloaded instances before detach  https://review.openstack.org/60274320:44
openstackgerritMerged openstack/placement master: Update code and opts in conf/paths.py for placement  https://review.openstack.org/60107120:44
fried_ricetetsuro: O master of latent bugs...20:50
fried_riceCan you look at this with me please: https://github.com/openstack/placement/blob/master/placement/objects/resource_provider.py#L412220:50
fried_riceIt looks to me like the condition on L4135 will never hit20:51
fried_ricebecause by then we've trimmed alloc_request_objs down to limit20:51
mriedemcdent: just asking for a test wrinkle follow up in https://review.openstack.org/#/c/601072/20:51
tetsurofried_rice: I'm good to be assigned to look into the bug, but I'm not catching up what you are talking about.20:53
tetsurofried_rice: you mean the limit qparam doesn't work at all?20:53
fried_ricetetsuro: I haven't proven the bug, just think we might have one.20:53
fried_ricetetsuro: I mean that the limit param will slice down the *allocation requests*, but will not trim down the *provider summaries* accordingly.20:54
tetsuroAh, I got that.20:54
fried_ricebecause that condition will always be False20:54
fried_riceBecause we've already reduced the alloc_request_objs based on the limit20:54
fried_riceDo you agree?20:54
tetsuroI can look into your concern.20:55
fried_ricetetsuro: I wanted to brainstorm a way to craft a test case to prove this.20:55
fried_riceSince we can't rely on the ordering of allocation requests, even for the non-random thing...20:56
cdentmriedem: so, just to be clear, you're happy with the wsgi app doing a hard fail without catching the exception, you just want the test to be more explicit about what it is testing?20:56
fried_ricePerhaps what's needed is a unit test.20:56
tetsuroI see. Yeah, sometimes the test is not 100% with hashable things.21:00
openstackgerritMerged openstack/placement master: Rationalize and clarify database configuration  https://review.openstack.org/60107221:01
cdentredid os-resource-classes using cookiecutter, if we choose to go that way: https://github.com/cdent/os-resource-classes21:04
fried_ricetetsuro: I'm factoring out the limit stuff so I can pass completely mocked-up (but controllable) lists to it and writing a unit test.21:07
tetsuroI was wondering if we can have limit filter earlier in the code path.21:08
tetsurofried_rice: ...so that we don't have to go through into all huge numbers of iterations. But...so far it seems to be impossible.21:10
fried_ricetetsuro: Oh, I have the fix21:11
fried_ricebut first I need to prove there's a bug :)21:11
tetsuroCool.21:11
mriedemcdent: yeah21:12
mriedemex = self.assertRaises(...)21:12
cdentyeah, already done21:12
mriedemself.assertIn('something', six.text_type(ex))21:12
mriedemok21:12
cdentjust wanted to be sure that was what you meant21:13
openstackgerritChris Dent proposed openstack/placement master: Test for missing database configuration message  https://review.openstack.org/60275821:14
cdentmriedem: there it is21:15
edleafecdent: That's a lot of stuff to hold a simple list :) (re:os-resource-classes)21:17
cdentall that other crap was added to support the existing uses, see: https://review.openstack.org/#/c/584084/21:18
cdentbecause, as dan put it, the y2k problem jay created21:18
fried_ricetetsuro: oboy. <=. So even when we've trimmed it, the = part still hits. That was a relatively lot of work for nothing.21:22
edleafecdent: yeah, I know, but it still seems funny. All the more reason to combine that and os-traits into a single lib21:23
cdentmelwitt has recommended (on the ptg etherpad) an os-constants but I'm not clear on the "costs" of that21:24
fried_ricecdent: I'm pretty clear on that.21:24
melwittI wouldn't call it a recommendation, it was just an idea21:24
fried_riceShort-term, the cost is nearly identical to that of os-resource-classes being separate.21:25
fried_ricewhich is as follows:21:25
melwittor a question even21:25
cdentI already had the code for os-resource-classes in a thing: so making it cookiecuttered as no big deal. Is easy to throw away21:25
edleafefried_rice: I tend to think in long-term cognitive loads21:25
fried_riceos-constants gets a module called resource_classes, which is exactly everything in proposed os-resource-classes21:25
fried_riceand21:25
cdentmelwitt: yeah, sorry, recommend was the wrong word there. I think someone here in the tc room said it at the same time I was typing. It was also the wrong word here in the room21:25
fried_riceos-constants imports os_traits at the base level.21:25
melwittcdent: ok, just wanted to make sure no one thought I was saying "do it this way"21:26
edleafefried_rice: so this would be for all constants OpenStack-wide, not just placement?21:26
fried_riceI wouldn't be opposed to that long-term, shrug. If you want to call it os-placement-const I wouldn't object to that either.21:26
cdentone plan was to talk to lyarwood and ask him about the relative costs of the various options from his perspective21:27
fried_riceAnyway, once that's released, we can continue to add traits to os-traits until we decide we want to get rid of os-traits; then we freeze os-traits and dup it into os-[placement-]const, where subsequent trait development continues thereafter.21:27
edleafefried_rice: I don't care if it's os-placement-const, os-placement, or placementlib. Just as long as the name indicates what it is21:27
* cdent just wants the shortest path to getting this issue off the list21:32
melwittalso, I made up the name os-constants, I don't think it should be actually named that. brain was just fried and trying to call it something21:33
melwittI updated the etherpad to say "os-whatever" so it's clear it was a placeholder21:35
cdentmriedem or melwitt: if https://review.openstack.org/#/c/600157/ gets in then the devstack and grenade experiments don't need as much in the way of depends21:35
melwittack, thanks21:35
openstackgerritEric Fried proposed openstack/placement master: Refactor: separate limiting GET /a_c results  https://review.openstack.org/60277021:38
cdentfried_rice: I suspect it is simply friday at the end of PTG but I'm struggling to parse the test in ^21:46
cdenthowever, I do think that it an ideal case where unit tests is right and that extraction is a huge win21:47
cdents/extraction/method extraction/21:47
cdentanyway, I'll look at it again next week...21:47
fried_ricecdent: I could have used real objects rather than mocks, but it didn't seem worth it. But is that what you're having trouble with?21:48
fried_riceI could sling some comments in there.21:48
fried_riceThe point of the test is to make sure that, absent randomization, the 'limit' thing strips the allocation requests *and* the provider summaries correctly.21:48
cdentthe mocks are fine, I just couldn't parse what it was actually doing. note: friday friday21:48
fried_riceThe latter is the part I thought was buggy, but proved myself wrong.21:48
* cdent nods21:49
cdentI suspect part of my problem is not readfing "sum" as provider summary. I keep trying to add things together21:49
fried_riceugh, sorry about that. I had to make var names short because that method call string is so bloody long.21:50
cdenti'd probably read it fine next week (after gettting over jetlag). It might just be today21:50
fried_ricecool cool21:50
fried_ricecdent: With the patches I put up today, our max-complexity goes to 1521:51
fried_riceI'm writing the patch to knock that down right now.21:51
fried_rice(that is, the limit goes to 15; our actual most complex method is 14)21:51
fried_rice(i.e. -3)21:51
cdentnice.21:52
cdentI think it is very helpful, especially if it enables tests like these21:53
cdentit can help move the db stuff into more a "trusted black box"21:54
mriedemcdent: re https://review.openstack.org/#/c/600157/ yes i've got the tab open, but need to put my brain into it21:54
cdentmriedem: coolio, thanks21:54
cdentthe main guts are the stuff you already merged. that patch is basically "make it work at all21:55
cdentmriedem: the stuff at https://review.openstack.org/#/c/602484/ exercises it21:55
*** mriedem has quit IRC21:56
*** takashin has left #openstack-placement21:58
*** tetsuro has quit IRC21:59
*** N3l1x has quit IRC22:05
*** sean-k-mooney has quit IRC22:07
openstackgerritEric Fried proposed openstack/placement master: Reduce max-complexity to 15  https://review.openstack.org/60277722:10
cdentfried_rice: there are quite a few other tools that can drive this sort of work if it is a thing you like doing22:13
fried_ricecdent: Drive what sort of work? Looking for repeated code?22:13
cdenteither repeated or overtly complex22:14
openstackgerritEric Fried proposed openstack/placement master: Move qs parsing to placement.lib.RequestGroup  https://review.openstack.org/60249522:16
openstackgerritEric Fried proposed openstack/placement master: max-complexity=>16: refactor GET /a_c qs parsing  https://review.openstack.org/60266722:16
openstackgerritEric Fried proposed openstack/placement master: DRY usage and capacity SQL clauses  https://review.openstack.org/60270122:16
openstackgerritEric Fried proposed openstack/placement master: DRY trait existence check  https://review.openstack.org/60271322:16
openstackgerritEric Fried proposed openstack/placement master: DRY trait/aggregate prefetch  https://review.openstack.org/60274022:16
openstackgerritEric Fried proposed openstack/placement master: Refactor: separate limiting GET /a_c results  https://review.openstack.org/60277022:16
openstackgerritEric Fried proposed openstack/placement master: Reduce max-complexity to 15  https://review.openstack.org/60277722:18
fried_ricecdent: Okay, I bit the bullet and put all of those in a single series so I didn't have to have three depends-ons on the top one22:18
cdentmakes good sense22:18
*** cdent has quit IRC22:24
openstackgerritMerged openstack/placement master: Link to tempest doc in tests/README.rst  https://review.openstack.org/60265522:29
*** mriedem has joined #openstack-placement22:40
mriedem+2 on https://review.openstack.org/#/c/600157/ but holding back the +W for jay and dansmith22:47
openstackgerritMatt Riedemann proposed openstack/nova master: WIP: Add post-test hook for testing evacuate  https://review.openstack.org/60217422:58
openstackgerritMatt Riedemann proposed openstack/nova master: Allow ability for non admin users to use all filters on server list.  https://review.openstack.org/52655823:05
*** fried_rice has quit IRC23:09
*** s10 has quit IRC23:39
*** s10 has joined #openstack-placement23:40
*** s10 has quit IRC23:40
*** s10 has joined #openstack-placement23:40
*** s10 has quit IRC23:41
*** s10 has joined #openstack-placement23:41
*** s10 has quit IRC23:42
*** s10 has joined #openstack-placement23:42
« Thursday, 2018-09-13 Index Saturday, 2018-09-15 »

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