Tuesday, 2019-01-08

« Monday, 2019-01-07 Index Wednesday, 2019-01-09 »
*** macza has quit IRC00:06
*** takashin has joined #openstack-nova00:10
*** tetsuro has joined #openstack-nova00:19
*** munimeha1 has quit IRC00:30
*** Swami has joined #openstack-nova00:38
Swamiquick question on placement-api with stein. Enabling placement-api on compute node fails while trying to recreate_database. Is there any other option in devstack to prevent the database from recreating while running in the compute node. The placement-api is working fine on all in one node.00:39
*** tbachman has quit IRC00:43
*** ircuser-1 has joined #openstack-nova00:43
*** ileixe has joined #openstack-nova00:43
*** tbachman has joined #openstack-nova00:48
*** Swami has quit IRC00:59
*** wolverineav has quit IRC01:07
*** itlinux has joined #openstack-nova01:08
*** wolverineav has joined #openstack-nova01:09
*** wolverineav has quit IRC01:13
*** brinzhang has joined #openstack-nova01:19
*** wolverineav has joined #openstack-nova01:27
*** wolverineav has quit IRC01:32
openstackgerritCorey Bryant proposed openstack/nova stable/ocata: Ensure rbd auth fallback uses matching credentials  https://review.openstack.org/62701201:37
*** sapd1_x has joined #openstack-nova01:44
*** efried has quit IRC01:50
*** macza has joined #openstack-nova01:55
*** igordc has quit IRC01:55
openstackgerritYikun Jiang proposed openstack/nova master: Per aggregate scheduling weight  https://review.openstack.org/62816301:58
*** ileixe has quit IRC02:01
*** ileixe has joined #openstack-nova02:02
*** ileixe has quit IRC02:04
*** ileixe has joined #openstack-nova02:04
*** eandersson has joined #openstack-nova02:11
*** tbachman has quit IRC02:12
*** tetsuro has quit IRC02:16
openstackgerritHesam Chobanlou proposed openstack/nova master: unused images are always deleted (add to in-tree hper-v code)  https://review.openstack.org/62879302:28
*** tetsuro has joined #openstack-nova02:28
*** macza has quit IRC02:30
openstackgerritTakashi NATSUME proposed openstack/nova master: Add descriptions of numbered resource classes and traits  https://review.openstack.org/62149402:37
*** brinzhang has quit IRC02:37
*** hongbin has joined #openstack-nova02:37
*** psachin has joined #openstack-nova02:41
*** bhagyashris has joined #openstack-nova02:48
*** mdbooth has quit IRC02:54
*** tiendc has joined #openstack-nova02:56
*** ileixe has quit IRC02:58
*** mhen has quit IRC02:59
*** ileixe has joined #openstack-nova02:59
*** mhen has joined #openstack-nova03:02
*** whoami-rajat has joined #openstack-nova03:16
*** macza has joined #openstack-nova03:32
*** macza has quit IRC03:36
*** mdbooth has joined #openstack-nova03:41
*** sapd1_x has quit IRC03:43
*** mdbooth has quit IRC03:45
*** jayashri has quit IRC03:54
*** sapd1_x has joined #openstack-nova03:55
*** hongbin has quit IRC03:57
*** udesale has joined #openstack-nova04:12
*** awaugama has quit IRC04:20
*** spsurya has joined #openstack-nova04:22
*** wxy-xiyuan has quit IRC04:24
*** efried has joined #openstack-nova04:25
*** bzhao__ has quit IRC04:26
*** efried has quit IRC04:34
*** wolverineav has joined #openstack-nova04:40
openstackgerritMerged openstack/nova master: Change to debug repetitive info messages  https://review.openstack.org/62805804:44
*** efried has joined #openstack-nova04:46
*** dims has quit IRC04:47
*** dims has joined #openstack-nova04:48
*** sridharg has joined #openstack-nova04:51
*** tbachman has joined #openstack-nova04:53
openstackgerritsean mooney proposed openstack/nova master: Skip checking of target_dev for vhostuser  https://review.openstack.org/62754004:56
*** dims has quit IRC04:56
*** tbachman_ has joined #openstack-nova04:56
*** dims has joined #openstack-nova04:56
*** tbachman has quit IRC04:58
*** tbachman_ is now known as tbachman04:58
*** spsurya has quit IRC05:10
*** spsurya has joined #openstack-nova05:12
*** ratailor has joined #openstack-nova05:15
*** abhishekk has joined #openstack-nova05:30
*** sunnaichuan has joined #openstack-nova06:19
*** alex_xu has quit IRC06:23
*** gyee has quit IRC06:24
*** tbachman has quit IRC06:27
*** alex_xu has joined #openstack-nova06:29
*** sunnaichuan has quit IRC06:32
*** Luzi has joined #openstack-nova06:44
*** mkrai has joined #openstack-nova06:48
*** tetsuro has quit IRC06:51
*** bzhao__ has joined #openstack-nova06:58
*** rcernin has quit IRC06:58
*** fragatina has joined #openstack-nova07:12
*** moshele has joined #openstack-nova07:30
*** wxy-xiyuan has joined #openstack-nova07:32
*** abhishekk has quit IRC07:32
*** fragatina has quit IRC07:33
*** pcaruana has joined #openstack-nova07:42
*** tiendc has quit IRC07:45
*** sapd1_ has quit IRC07:50
*** slaweq has joined #openstack-nova07:50
*** sapd1_ has joined #openstack-nova07:50
*** wolverineav has quit IRC07:51
*** slaweq has quit IRC07:53
*** slaweq has joined #openstack-nova07:55
*** takashin has left #openstack-nova08:00
*** jungleboyj has quit IRC08:06
*** jungleboyj has joined #openstack-nova08:06
*** ccamacho has joined #openstack-nova08:09
*** priteau has joined #openstack-nova08:12
*** moshele has quit IRC08:14
*** moshele has joined #openstack-nova08:14
*** ralonsoh has joined #openstack-nova08:23
*** xek has joined #openstack-nova08:24
*** helenafm has joined #openstack-nova08:25
*** yan0s has joined #openstack-nova08:25
*** mkrai has quit IRC08:26
*** helenafm_ has joined #openstack-nova08:45
*** helenafm has quit IRC08:48
*** tetsuro has joined #openstack-nova08:51
*** wolverineav has joined #openstack-nova08:51
*** sahid has joined #openstack-nova08:55
*** wolverineav has quit IRC08:55
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Import IP implementation modules outside privsep context  https://review.openstack.org/62902509:04
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Cleanup device at the end of 'test_iproute_object_closes_correctly' test  https://review.openstack.org/62911209:06
*** sapd1_x has quit IRC09:15
*** dtantsur|afk is now known as dtantsur09:21
*** maciejjozefczyk has joined #openstack-nova09:28
*** sapd1_x has joined #openstack-nova09:31
*** derekh has joined #openstack-nova09:33
*** ileixe has left #openstack-nova09:42
*** bhagyashris has quit IRC09:55
*** erlon has joined #openstack-nova10:05
*** fragatina has joined #openstack-nova10:14
*** priteau has quit IRC10:17
*** moshele has quit IRC10:19
*** fragatina has quit IRC10:21
*** tetsuro has quit IRC10:23
*** lpetrut has joined #openstack-nova10:26
*** sapd1_x has quit IRC10:44
*** priteau has joined #openstack-nova10:44
*** liuyulong has joined #openstack-nova10:54
*** liuyulong has quit IRC10:55
*** liuyulong has joined #openstack-nova10:56
*** udesale has quit IRC10:57
*** moshele has joined #openstack-nova10:59
*** ccamacho has quit IRC11:03
*** cdent has joined #openstack-nova11:18
*** lpetrut has quit IRC11:26
*** ccamacho has joined #openstack-nova11:33
openstackgerritKashyap Chamarthy proposed openstack/nova master: libvirt: Support native TLS for migration and disks over NBD  https://review.openstack.org/62521611:39
openstackgerritChris Dent proposed openstack/nova master: Remove placement from nova install docs  https://review.openstack.org/62818411:52
openstackgerritChris Dent proposed openstack/nova master: Remove link to placement configuration from nova config docs  https://review.openstack.org/62818511:52
openstackgerritChris Dent proposed openstack/nova master: Remove placement from contributor doc  https://review.openstack.org/62818611:52
*** ratailor has quit IRC11:57
kashyapsean-k-mooney1: gibi: When any of you get a moment, wonder if I can trick you to look at this change — "libvirt: Support native TLS for migration and disks over NBD" — https://review.openstack.org/62521612:01
kashyap(Or anyone here who has time.)12:03
* cdent waves impotently at kashyap :)12:08
kashyapcdent: Hi, there!  Hehe, don't underestimate yourself :D12:09
* cdent is in the midst of a tricky rebase12:09
*** ccamacho has quit IRC12:20
*** wolverineav has joined #openstack-nova12:28
*** wolverineav has quit IRC12:32
sean-k-mooney1 kashyap sure ill take a look. i think i added this to my list yesterday but did not get a chance to review it12:40
openstackgerritYikun Jiang proposed openstack/nova master: Per aggregate scheduling weight  https://review.openstack.org/62816312:44
*** sayalilunkad has quit IRC12:51
*** sayalilunkad has joined #openstack-nova12:51
*** ccamacho has joined #openstack-nova12:53
*** ccamacho has quit IRC12:54
*** ccamacho has joined #openstack-nova12:54
*** sapd1_x has joined #openstack-nova12:59
*** szaher has joined #openstack-nova13:08
*** yan0s has quit IRC13:09
*** ade_lee has quit IRC13:10
*** udesale has joined #openstack-nova13:20
*** liuyulong_ has joined #openstack-nova13:22
*** sapd1_x has quit IRC13:23
*** yan0s has joined #openstack-nova13:24
*** liuyulong has quit IRC13:28
*** zigo has joined #openstack-nova13:29
*** psachin has quit IRC13:33
openstackgerritChris Dent proposed openstack/nova master: Delete the placement code  https://review.openstack.org/61821513:40
openstackgerritChris Dent proposed openstack/nova master: Update nova docs front page for placement removal  https://review.openstack.org/62818313:40
openstackgerritChris Dent proposed openstack/nova master: Remove placement from nova install docs  https://review.openstack.org/62818413:40
openstackgerritChris Dent proposed openstack/nova master: Remove link to placement configuration from nova config docs  https://review.openstack.org/62818513:40
openstackgerritChris Dent proposed openstack/nova master: Remove placement from contributor doc  https://review.openstack.org/62818613:40
*** trident has quit IRC13:41
*** sapd1_x has joined #openstack-nova13:41
*** trident has joined #openstack-nova13:42
kashyapsean-k-mooney1: Thanks!13:43
*** jhesketh has quit IRC13:45
gibikashyap: do I see correctly that the bluprint for https://review.openstack.org/#/c/625216 haven't been approved yet?13:46
*** szaher has quit IRC13:47
kashyapgibi: Before I left for PTO, melwitt and co said they'd approve it.  And I recall even seeing it is approved13:47
* kashyap triple-checks13:47
*** jhesketh has joined #openstack-nova13:47
gibikashyap: this is still in new phase https://blueprints.launchpad.net/nova/+spec/support-qemu-native-tls-for-live-migration13:47
kashyapgibi: Ah, the "Definition"13:48
kashyapHmm13:48
* kashyap gets the relevant chat transcript13:49
*** szaher has joined #openstack-nova13:52
gibikashyap: no worries, just ping melwitt when she is up and get the administration done13:52
gibikashyap: I've started reading the implementation patch...13:53
kashyapYep, will do.  Thanks for noticing.  I saw the BP, but the "Accepted for Stein" decieved me.13:53
kashyapgibi: It is nothing complex; but the underlying setup is a buffalo13:54
kashyapgibi: I wrote a detailed end-to-end test here (some parts of it are already scripted by installer tools): https://kashyapc.fedorapeople.org/Native-TLS/Setup-for-NBD-and-migration-streams-over-TLS.rst.txt13:55
gibikashyap: thanks for the pointers. I will dissapeare for a meeting but still have the intention to go through your patch13:56
kashyapSure.  I don't count on instant responses :-)  Thanks for the time!13:57
gibikashyap: thank you for your effort!13:57
kashyapmelwitt: When you're about, following up from our pre-break chat here: Mind ACKing this one, assuming I answered all questions: https://blueprints.launchpad.net/nova/+spec/support-qemu-native-tls-for-live-migration13:59
*** trident has quit IRC14:03
*** trident has joined #openstack-nova14:03
maciejjozefczykmelwitt: jaypipes PTAL https://review.openstack.org/#/c/614167/  https://review.openstack.org/#/c/591607 :)14:09
*** yan0s has quit IRC14:09
*** yan0s has joined #openstack-nova14:10
*** moshele has quit IRC14:20
jaypipesmaciejjozefczyk: on it.14:20
*** munimeha1 has joined #openstack-nova14:22
*** moshele has joined #openstack-nova14:28
maciejjozefczykthanks Jay14:28
*** priteau has quit IRC14:33
yongliheJay, thank you giving constructive advice for  spec "show-server-numa-topology," , https://review.openstack.org/#/c/612256/2014:36
jaypipesyonglihe: no problem. I'll do a final review on that shortly.14:36
jaypipesyonglihe: just finishing up maciejjozefczyk's second patch review.14:36
yongliheso efficiency -:)14:37
jaypipes:)14:37
maciejjozefczykjaypipes: after coffee already?14:38
yongliheHi Alex, Matt, this 'show-server-group' spec is quite small, please help on it. you guys already working on that hardly, thanks.14:39
openstackgerritRui Zang proposed openstack/nova-specs master: support virtual persistent memory  https://review.openstack.org/60159614:40
yongliheoh, s/hardly/hard14:40
*** ondrejme has quit IRC14:43
jaypipesmaciejjozefczyk: of course! :) already on cup 3.14:44
jaypipesyonglihe: Matt's on PTO this week I think. Alex's IRC nick is alex_xu14:45
jaypipesyonglihe: and I will review the show-server-group spec too. it's small, like you said.14:45
yonglihethanks, i got to grab alex on Cube -:)14:45
*** Luzi has quit IRC14:45
openstackgerritJie Li proposed openstack/nova master: Support volume-backed server rebuild in compute  https://review.openstack.org/62589314:48
stephenfinjaypipes: Random question: is hugepage tracking in placement a thing you've thought about (distant future, obv)14:49
stephenfin*?14:49
jaypipesstephenfin: https://review.openstack.org/#/c/442718/14:49
jaypipesstephenfin: blast from the past ;)14:50
stephenfin:D14:50
jaypipesstephenfin: short answer is yes, I've wanted to track that stuff in placement but all the NUMA and CPU topology stuff has dragged on with nested and other things14:50
stephenfinjaypipes: Fair. IMO, the CPU side of things gets us more and should be higher priority, for sure14:51
stephenfinfwiw, the reason I ask is '[nova] Mempage fun' on openstack-discuss14:51
*** liuyulong_ has quit IRC14:51
cdentjaypipes: you could totally port that to os-resource-classes now, if you felt so inclined. I'm still as confused as I was on the review about how we count14:52
*** moshele has quit IRC14:52
sean-k-mooney1stephenfin: jaypipes talking about the mempage fun?14:52
stephenfinsean-k-mooney1: Not willingly - I dragged him into it slightly ;)14:53
sean-k-mooney1stephenfin: hugepages in placement is part of sylivans numa in placement spec i think14:53
sean-k-mooney1stephenfin: we have talked about it many times in the past and it was one of the main usecase we created nested resouce provirders for14:54
sean-k-mooney1stephenfin: infact it was higher on or prioity list for nested provider which is why i was surpised we were leading with vgpus in denver14:55
stephenfinsean-k-mooney1: Possibly, I haven't checked14:55
*** awaugama has joined #openstack-nova14:55
stephenfinsean-k-mooney1: Though I'd been thinking we'd report pagesizes on both the root provider and the numa cell providers, but that wouldn't work because they're the same thing viewed in different ways14:55
stephenfinDoesn't matter though. Not anything we're going to be tackling in the immediate future14:56
* stephenfin goes back to fixing that oversubscription issue14:56
sean-k-mooney1stephenfin: it was somthing i really want to fix last/this cycle but with the company move i didnt get to work on it14:57
sean-k-mooney1there were enough depencise however that it likely could nto have been done until this cycle anyway14:57
*** sean-k-mooney1 is now known as sean-k-mooney14:57
openstackgerritJie Li proposed openstack/nova master: Support volume-backed server rebuild in compute  https://review.openstack.org/62589314:58
openstackgerritKashyap Chamarthy proposed openstack/nova master: libvirt: Support native TLS for migration and disks over NBD  https://review.openstack.org/62521615:05
openstackgerritRui Zang proposed openstack/nova-specs master: Virtual persistent memory libvirt driver implementation  https://review.openstack.org/62289315:08
jaypipesmaciejjozefczyk: ok, reviewed https://review.openstack.org/#/c/591607/. If you can make that small change I can re-review quickly.15:13
jaypipescdent: oh, totes. just not something that's been on my priority list :)15:13
openstackgerritRui Zang proposed openstack/nova-specs master: support virtual persistent memory  https://review.openstack.org/60159615:14
maciejjozefczykjaypipes:  checking15:14
cdentjaypipes: sure. i was just pointing it out mostly in a sort of "hey, we got a thing for that now!" way15:14
*** itlinux has quit IRC15:21
*** tbachman has joined #openstack-nova15:23
*** markvoelker has joined #openstack-nova15:23
*** markvoelker has quit IRC15:26
*** sapd1_x has quit IRC15:26
*** mlavalle has joined #openstack-nova15:26
jaypipescdent: ack, totes.15:27
jaypipescdent: that patch I linked to stephenfin was >1 year old, after all :)15:27
cdenttoo much water under the bridge. I feel the abyss of time.15:28
* cdent swoons15:28
jaypipescdent: ha! :)15:29
*** ade_lee has joined #openstack-nova15:31
openstackgerritRui Zang proposed openstack/nova-specs master: Virtual persistent memory libvirt driver implementation  https://review.openstack.org/62289315:33
openstackgerritMaciej Jozefczyk proposed openstack/nova master: Force refresh instance info_cache during heal  https://review.openstack.org/59160715:49
maciejjozefczykjaypipes: update done, about sorting: it's kind of chicken or egg problem, which came first, row in VirtualInterface or device_id set in neutron port_data15:50
*** efried has quit IRC15:55
jaypipesmaciejjozefczyk: +2 :)16:01
maciejjozefczykjaypipes: thanks16:02
jaypipesmaciejjozefczyk: no problem.16:02
adriancsean-k-mooney: Hey, hope you had a good holiday. do you reckon we can converge on libvirt-sriov-live-migration spec by end of M2 ?16:02
jaypipesmaciejjozefczyk: thx for your patience!16:02
*** udesale has quit IRC16:02
*** wolverineav has joined #openstack-nova16:04
maciejjozefczykjaypipes: I put the helmet on because im testing it on 40k instances restored prod db16:04
sean-k-mooneyadrianc: i would hope so16:06
sean-k-mooneyadrianc: and hi o/16:06
gibikashyap: I left a question inline in ttps://review.openstack.org/#/c/62521616:07
sean-k-mooneyadrianc: there are currently no outstanding comments that im aware of for https://review.openstack.org/#/c/605116/ so if people have review bandwith between now and the spec freeze i hope it ill get merged16:07
jaypipesmaciejjozefczyk: :)16:08
kashyapgibi: Thanks, will look in a few.16:08
*** szaher has quit IRC16:08
*** wolverineav has quit IRC16:08
adriancsean-k-mooney: :)  Great, i believe we dont have any more open issues according to the discussions. will it help to patition for a review in the nova weekly meeting on thursday ?16:08
sean-k-mooneymaciejjozefczyk: im really looking forword to your chages hopefully merging soon16:09
*** szaher has joined #openstack-nova16:09
sean-k-mooneymaciejjozefczyk: im less loging forward to haveing to figure out how to backport it to newton downstream but we have customer hitting issue that your patches fix16:09
maciejjozefczyksean-k-mooney: I have backport to Newton16:10
sean-k-mooneymaciejjozefczyk: isnt newton eol upstream16:10
maciejjozefczykbut for: https://review.openstack.org/#/c/614167  without functional tests16:11
maciejjozefczyksean-k-mooney: newton is the newest release I support, I could say ;]16:11
maciejjozefczyksean-k-mooney:  Let me test a bit and I can send you patches if you want16:11
maciejjozefczykbut for Newton I needed to cherrry-pick about 4-5 other patches along with one more to neutron16:12
sean-k-mooneymaciejjozefczyk: that woudl be useful. our donwsteam backport policy still say i have to wait for it to merge on at least master upstream before i can submit it for downsteam with getting an exception for out of tree hotfix so no rush16:12
maciejjozefczykok16:13
maciejjozefczyksean-k-mooney: i'm not that lucky, sometimes we have patches that are in review in upstream, but only when we're confident16:13
*** erlon has quit IRC16:15
sean-k-mooneyif we have ci/release blocking bugs we will sometimes make an excpetion but while this is an annoying issue its not that pressing16:15
maciejjozefczykfor this particular one we have about 20-30 occurrences/day16:15
maciejjozefczykon all regions16:15
sean-k-mooneyouch we have only had one customer report an issue and for them it happended after a power failuer broke quoram on there glara cluster16:16
*** pcaruana has quit IRC16:20
*** itlinux has joined #openstack-nova16:20
stephenfinI'm seeing 'sqlalchemy.exc.NoSuchTableError: migration_tmp' when running unit tests. Any idea why?16:22
stephenfin*Anyone any16:22
sean-k-mooneyoh by the way i figured out why the intel nfv ci is still triggering but skipping all jobs. ie why intel has not disabled it totaly16:29
sean-k-mooneyit is still runing jobs properly on neutron16:29
*** ade_lee is now known as ade_lee_lunch16:30
sean-k-mooneythey have just disabel the nova jobs.16:30
*** erlon has joined #openstack-nova16:36
*** erlon has quit IRC16:36
*** ccamacho has quit IRC16:38
*** gyee has joined #openstack-nova16:38
jaypipesstephenfin: from your ML post about hugepages... "It's perfectly fine to have NUMA without CPU pinning, though not the other way around." <-- is that true? Wouldn't a user want to request dedicated CPU resources regardless of whether those CPUs are associated with a particular NUMA node? I mean, if they don't have a device that has NUMA affinity requirements and they don't have a need for a specific guest NUMA layout/topology, why isn't that a16:47
jaypipeslegit request?16:47
stephenfinjaypipes: Ah, I'm talking about the nova view of things. Nova doesn't currently let you have CPU pinning without a NUMA topology16:49
jaypipesah, yes, much to my chagrin.16:49
sean-k-mooney jaypipes  oversubscription is perfectly fine fore cpu pinning too16:49
stephenfinWhether CPU pinning without taking NUMA affinity into account is a different issue entirely16:49
stephenfinsean-k-mooney: but daft, given why people are using CPU pinning16:49
sean-k-mooneystephenfin: oh you were refering to the cpu pinning bit16:50
sean-k-mooneyneeding numa toplogy16:50
jaypipesstephenfin: it's just the way you wrote that, made it seem that a request for dedicated CPU resources without any NUMA affinity/topology wasn't a valid request.16:50
sean-k-mooneystephenfin: no its not16:50
sean-k-mooneycpu pinning means i have a compute intensive workload16:50
sean-k-mooneyit dows nto mean that workload is memory sencitive16:50
*** imacdonn has quit IRC16:50
stephenfinjaypipes: Sorry, not what I was going for. If you are replying, to that email you might clarify16:51
*** imacdonn has joined #openstack-nova16:51
stephenfinsean-k-mooney: Think you're crossing wires here16:51
jaypipessean-k-mooney: pls see my comment on that topic on https://review.openstack.org/#/c/59995716:51
stephenfinI'm saying oversubscription for pinned CPUs is daft, i.e. allowing two instances to share the same pCPU16:51
jaypipesstephenfin: not a huge deal. if I reply, I'll mention it, but was more just making sure I was understanding you properly.16:52
jaypipesstephenfin: see my comment on https://review.openstack.org/#/c/599957/ :)16:52
stephenfinsean-k-mooney: I have no issues with unbinding CPU pinning from NUMA topologies16:52
sean-k-mooneystephenfin: that i agree with but we were talking about memory oversubsription on the email tread16:52
stephenfinjaypipes: Indeed :) I recall that discussion at the PTG16:52
*** artom has quit IRC16:52
sean-k-mooneyjaypipes: ya i stand by my comments on that spec i dont think we should support that16:53
stephenfinsean-k-mooney: Right, you said "oversubscription is perfectly fine fore cpu pinning too" and I thought you mean CPU oversubscription16:53
stephenfin*meant16:53
sean-k-mooneyjaypipes: that being https://review.openstack.org/#/c/599957/16:53
stephenfinsean-k-mooney: Hence my "that is daft" comment :)16:53
sean-k-mooneystephenfin: no dedicated means you get the whole cpus if we ever wanted to allow pinning and over subsrciption i woudl want to different cpu_policy to indicate that16:54
openstackgerritMerged openstack/nova stable/rocky: Make compute rpcapi version calculation check all cells  https://review.openstack.org/62498216:55
stephenfinsean-k-mooney: yup, I agree :) Just me misinterpreting what you said16:55
sean-k-mooneywe named it shared and dedicated to clearly callout that one allows over subscription or cpus and the other does not so https://review.openstack.org/#/c/599957/ is a definet no form me16:55
sean-k-mooneystephenfin: i just resopned to that mail again pointing out the setting hw:numa_nodes=X does not prevent swaping the memory for 4k pages too incase you did not see16:56
openstackgerritMaciej Jozefczyk proposed openstack/nova master: Add fill_virtual_interface_list online_data_migration script  https://review.openstack.org/61416716:57
openstackgerritMaciej Jozefczyk proposed openstack/nova master: Force refresh instance info_cache during heal  https://review.openstack.org/59160716:57
maciejjozefczykjaypipes: I did a little update after your nits, will be silky smooth now16:57
maciejjozefczykneeded to go, bb ;)16:57
*** artom has joined #openstack-nova16:58
*** macza has joined #openstack-nova16:59
*** maciejjozefczyk has quit IRC17:00
stephenfinsean-k-mooney, jaypipes, sahid: https://bugs.launchpad.net/nova/+bug/181097717:00
openstackLaunchpad bug 1810977 in OpenStack Compute (nova) "Oversubscription broken for instances with NUMA topologies" [Undecided,New]17:00
stephenfinI checked and reverting the patch gets us back to where we want to be. Have a better patch ready but currently trying to figure out why UTs won't run17:01
jaypipesstephenfin: I would mark that bug as affecting me but... it doesn't. :)17:01
sean-k-mooneyim assuming oath does not over subscirbe ram or does not use the numa retlated features17:02
jaypipessean-k-mooney: that would be correct.17:03
sean-k-mooneyjaypipes: that certinly simplfies life.17:03
jaypipessean-k-mooney: indeed.17:04
jaypipessean-k-mooney: for performance-sensitive things, we use Ironic and the tenant does whatever the f**k they want to.17:04
sean-k-mooneyya when thats a option baremetal will always beat even the most tunned vm17:05
jaypipes(which is why 90% of our infrastructure is Ironic unfortunately, because tenants assume they must have baremetal to achieve the performance they need, even though 99% of our tenants haven't actually quantified what "performance" they "need".17:05
openstackgerritMerged openstack/nova-specs master: Add PENDING vm state  https://review.openstack.org/55421217:05
sean-k-mooneyjaypipes: ya, im so familar with the tuneing options at this point that i pick an chooes the ones that make sense when i need them17:07
sean-k-mooneyi normally have at least one of the numa feature enabled even if it is just hw:numa_nodes=117:07
sean-k-mooneyoh the pending state spec has merged cool :)17:08
sean-k-mooneythe rest of the premptable instance work is in the reaper project right?17:09
jaypipesyonglihe: k, reviewed the NUMA spec.17:15
*** yan0s has quit IRC17:20
*** helenafm_ has quit IRC17:25
*** sahid has quit IRC17:31
*** elod has quit IRC17:34
*** elod has joined #openstack-nova17:46
openstackgerritMerged openstack/nova-specs master: Flavor Extra Spec and Image Properties Validation  https://review.openstack.org/61854217:51
*** ade_lee_lunch has left #openstack-nova17:57
*** derekh has quit IRC18:00
openstackgerritChris Dent proposed openstack/nova master: Remove placement from nova install docs  https://review.openstack.org/62818418:07
openstackgerritChris Dent proposed openstack/nova master: Remove link to placement configuration from nova config docs  https://review.openstack.org/62818518:07
openstackgerritChris Dent proposed openstack/nova master: Remove placement from contributor doc  https://review.openstack.org/62818618:07
*** ralonsoh has quit IRC18:07
openstackgerritChris Dent proposed openstack/nova master: Switch to using os-resource-classes  https://review.openstack.org/62827818:07
*** maciejjozefczyk has joined #openstack-nova18:08
*** whoami-rajat has quit IRC18:12
*** panda is now known as panda|off18:17
*** maciejjozefczyk has quit IRC18:19
*** maciejjozefczyk has joined #openstack-nova18:21
*** Swami has joined #openstack-nova18:22
*** cdent has quit IRC18:22
*** igordc has joined #openstack-nova18:23
* stephenfin waves at igordc 18:23
openstackgerritStephen Finucane proposed openstack/nova master: Fix overcommit for NUMA-based instances  https://review.openstack.org/62928118:24
stephenfinsean-k-mooney, gibi, jaypipes: Ta-da ^18:24
stephenfinMakes sense to me, anyway. Lemme know18:24
* stephenfin goes home18:24
*** maciejjozefczyk has quit IRC18:24
*** wolverineav has joined #openstack-nova18:29
*** wolverineav has quit IRC18:30
*** wolverineav has joined #openstack-nova18:31
*** moshele has joined #openstack-nova18:33
sean-k-mooneystephenfin: its almost right comment inline. you messed the case fo explcit small pages e.g. hw:mem_page_size=small|4|4k18:34
*** moshele has quit IRC18:37
stephenfinsean-k-mooney: replied18:40
* stephenfin really goes home now :)18:40
*** dtantsur is now known as dtantsur|afk18:45
*** lbragstad has quit IRC18:52
*** lbragsta_ has joined #openstack-nova18:52
*** lbragsta_ is now known as lbragstad18:55
*** moshele has joined #openstack-nova18:57
*** elod has quit IRC18:59
*** sridharg has quit IRC18:59
*** moshele has quit IRC19:18
lbragstadjohnthetubaguy jaypipes o/ curious if either of you have thoughts on this since the last time we discussed it - http://lists.openstack.org/pipermail/openstack-discuss/2019-January/001518.html19:20
*** elod has joined #openstack-nova19:21
*** amodi_ is now known as amodi19:47
openstackgerritPaul-Emile Element proposed openstack/os-traits master: Add new traits for TPM support  https://review.openstack.org/62929919:52
*** ccamacho has joined #openstack-nova19:54
*** ccamacho has quit IRC19:58
*** gary_perkins has quit IRC20:01
*** igordc has quit IRC20:04
*** gary_perkins has joined #openstack-nova20:06
mnaseroh yikes20:19
mnaserhow safe is it to change the admin_or_owner rule20:19
mnaserto say, something like: "is_admin:True or (role:compute and project_id:%(project_id)s)"20:19
openstackgerritJack Ding proposed openstack/nova master: Improve libvirt image and snapshot handling  https://review.openstack.org/61669220:20
*** awaugama has quit IRC20:25
*** jmlowe has quit IRC20:30
*** jmlowe has joined #openstack-nova20:31
*** fragatina has joined #openstack-nova20:41
*** fragatina has quit IRC20:41
*** spsurya has quit IRC20:41
*** fragatina has joined #openstack-nova20:41
*** fragatina has quit IRC20:45
*** fragatina has joined #openstack-nova20:51
openstackgerritJack Ding proposed openstack/nova-specs master: Fix format flavor-extra-spec-image-property-validation  https://review.openstack.org/62931120:51
*** fragatina has quit IRC20:59
*** xek has quit IRC21:02
*** xek has joined #openstack-nova21:02
*** slaweq has quit IRC21:28
*** artom has quit IRC21:30
*** jamesdenton has quit IRC21:35
*** jamesdenton has joined #openstack-nova21:56
*** fragatina has joined #openstack-nova21:56
*** fragatina has quit IRC21:56
*** fragatina has joined #openstack-nova21:57
*** hongbin has joined #openstack-nova22:01
yonglihejay, thanks.22:07
*** fragatina has quit IRC22:08
*** efried has joined #openstack-nova22:18
*** slaweq has joined #openstack-nova22:23
yonglihejaypipes, sean-k-mooney do we need aligned about sean-k-mooney's suggestion? move some stuff out of the 'nodes' and a more granular policy rule.22:27
openstackgerritYongli He proposed openstack/nova-specs master: add spec "show-server-numa-topology"  https://review.openstack.org/61225622:35
yonglihejaypipes, update per your comments. i working on a more specific policy rule,  that's fine to me.22:37
*** itlinux has quit IRC22:48
*** rcernin has joined #openstack-nova22:53
*** igordc has joined #openstack-nova23:12
*** moshele has joined #openstack-nova23:13
*** burt has quit IRC23:13
openstackgerritYongli He proposed openstack/nova-specs master: add spec "show-server-numa-topology"  https://review.openstack.org/61225623:15
jaypipesyonglihe: I actually think no, it's best just left defaulted to admin-only policy.23:15
jaypipesyonglihe: but I won't hold up the spec for that reason or anything23:15
yongliheok, i need revert it. already add a fine control23:16
jaypipesyonglihe: don't revert :)23:16
jaypipesyonglihe: it's not a big deal23:16
yongliheok23:16
yonglihenice to know your idea about which item cause you add -123:18
yonglihereally nice hint for review.23:18
*** _alastor1 has joined #openstack-nova23:28
*** _alastor_ has quit IRC23:31
openstackgerritMerged openstack/python-novaclient stable/rocky: Fix up userdata argument to rebuild.  https://review.openstack.org/61305723:40
openstackgerritMerged openstack/python-novaclient stable/rocky: Follow up "Fix up userdata argument to rebuild"  https://review.openstack.org/61308623:45
openstackgerritHongbin Lu proposed openstack/nova-specs master: Scheduling support for SR-IOV bond  https://review.openstack.org/62605523:45
*** moshele has quit IRC23:50
*** liuyulong has joined #openstack-nova23:51
*** takashin has joined #openstack-nova23:53
« Monday, 2019-01-07 Index Wednesday, 2019-01-09 »

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