Tuesday, 2017-10-03

*** namnh has quit IRC00:02
*** takashin has joined #openstack-nova00:02
*** ijw has joined #openstack-nova00:05
*** sapcc-bot has quit IRC00:05
*** sapcc-bot has joined #openstack-nova00:06
*** namnh has joined #openstack-nova00:06
openstackgerritJay Pipes proposed openstack/nova master: rp: de-ORM ResourceProvider.get_by_uuid()  https://review.openstack.org/50902500:06
openstackgerritJay Pipes proposed openstack/nova master: rp: Move RP._get|set_aggregates() to module scope  https://review.openstack.org/50902600:06
openstackgerritJay Pipes proposed openstack/nova master: rp: Remove RP.get_traits() method  https://review.openstack.org/50902700:06
openstackgerritJay Pipes proposed openstack/nova master: rp: move RP._set_traits() to module scope  https://review.openstack.org/50902800:06
openstackgerritJay Pipes proposed openstack/nova master: rp: remove CRUD operations on Inventory class  https://review.openstack.org/50902900:06
openstackgerritJay Pipes proposed openstack/nova master: rp: streamline InventoryList.get_all_by_rp_uuid()  https://review.openstack.org/50903000:06
openstackgerritJay Pipes proposed openstack/nova master: rp: remove dead code in Allocation._create_in_db()  https://review.openstack.org/50903100:06
openstackgerritJay Pipes proposed openstack/nova master: rp: remove ability to delete 1 allocation record  https://review.openstack.org/50903200:06
openstackgerritJay Pipes proposed openstack/nova master: rp: fix up AllocList.get_by_resource_provider_uuid  https://review.openstack.org/50903300:06
openstackgerritJay Pipes proposed openstack/nova master: rp: remove bad comment in AllocList.create_all()  https://review.openstack.org/50903400:06
openstackgerritJay Pipes proposed openstack/nova master: rp: rework AllocList.get_all_by_consumer_id()  https://review.openstack.org/50903500:06
openstackgerritJay Pipes proposed openstack/nova master: rp: remove _HasAResourceProvider mixin  https://review.openstack.org/50903600:06
*** jaypipes has quit IRC00:07
*** hieulq has quit IRC00:09
*** hieulq has joined #openstack-nova00:10
*** namnh has quit IRC00:10
*** mriedem has quit IRC00:21
*** sdague has quit IRC00:22
*** gbarros has quit IRC00:28
*** acormier has joined #openstack-nova00:29
*** itlinux has joined #openstack-nova00:35
*** yamahata has quit IRC00:37
*** chyka has quit IRC00:41
*** chyka has joined #openstack-nova00:42
*** lnxnut has joined #openstack-nova00:42
*** chyka has quit IRC00:46
*** chyka has joined #openstack-nova00:49
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove 400 as expected error  https://review.openstack.org/50903900:51
*** lnxnut has quit IRC00:53
*** chyka has quit IRC00:54
*** edmondsw has joined #openstack-nova01:01
*** tbachman has quit IRC01:03
*** sapcc-bot3 has joined #openstack-nova01:04
*** sapcc-bot has quit IRC01:05
*** phuongnh has joined #openstack-nova01:05
*** acormier has quit IRC01:07
*** tbachman has joined #openstack-nova01:07
*** acormier has joined #openstack-nova01:07
*** adisky_ has joined #openstack-nova01:07
*** gjayavelu has quit IRC01:09
*** edmondsw has quit IRC01:17
*** penick has joined #openstack-nova01:18
*** acormier has quit IRC01:25
*** penick has quit IRC01:25
*** acormier has joined #openstack-nova01:26
*** itlinux has quit IRC01:29
*** itlinux has joined #openstack-nova01:29
*** acormier_ has joined #openstack-nova01:30
*** acormier has quit IRC01:30
*** itlinux has quit IRC01:48
*** lnxnut has joined #openstack-nova01:50
*** lbragstad has joined #openstack-nova01:51
openstackgerritmelanie witt proposed openstack/nova-specs master: Propose counting quota usage from placement  https://review.openstack.org/50904201:51
*** szaher has quit IRC01:57
*** lnxnut has quit IRC02:01
*** baoli has quit IRC02:02
*** baoli has joined #openstack-nova02:03
*** szaher has joined #openstack-nova02:07
*** acormier_ has quit IRC02:18
*** tbachman has quit IRC02:20
*** tbachman has joined #openstack-nova02:25
*** itlinux has joined #openstack-nova02:31
*** bnemec has quit IRC02:39
*** baoli has quit IRC02:43
*** itlinux has quit IRC02:44
*** yamamoto has joined #openstack-nova02:45
*** hongbin has joined #openstack-nova02:50
*** baoli has joined #openstack-nova02:51
*** baoli has quit IRC02:57
*** FL1SK has quit IRC02:57
*** baoli has joined #openstack-nova02:58
*** lnxnut has joined #openstack-nova02:59
*** hongbin has quit IRC03:00
*** hongbin has joined #openstack-nova03:00
*** erlon has quit IRC03:02
*** lnxnut has quit IRC03:07
*** vladikr has quit IRC03:08
openstackgerritTony Breeds proposed openstack/nova master: [DNM] Yesting the legacy-requirements job  https://review.openstack.org/50905403:12
*** baoli has quit IRC03:14
*** esberglu has joined #openstack-nova03:17
*** esberglu has quit IRC03:17
*** lbragstad has quit IRC03:17
*** sam_nowitzki has joined #openstack-nova03:29
*** gbarros has joined #openstack-nova03:32
*** thorst has joined #openstack-nova03:33
*** gyee has joined #openstack-nova03:37
*** gyee has quit IRC03:38
*** vladikr has joined #openstack-nova03:38
*** crushil has joined #openstack-nova03:43
openstackgerritPhilip Choi proposed openstack/nova master: commit ce26619af3f7ac8c55567d6aa8125b77e457af19 Author: Philip Choi <phchoi@cisco.com> Date:   Fri Sep 8 10:52:42 2017 +0800  https://review.openstack.org/50906003:45
*** udesale has joined #openstack-nova03:45
*** Dinesh_Bhor has joined #openstack-nova03:47
*** thorst has quit IRC03:50
*** vladikr has quit IRC03:56
*** gouthamr has joined #openstack-nova04:01
*** lnxnut has joined #openstack-nova04:04
*** ijw has quit IRC04:06
*** FL1SK has joined #openstack-nova04:09
*** esberglu has joined #openstack-nova04:10
*** esberglu has quit IRC04:14
*** mingyu has quit IRC04:17
*** acormier has joined #openstack-nova04:18
*** mingyu has joined #openstack-nova04:19
*** mingyu has quit IRC04:19
*** lnxnut has quit IRC04:21
*** oanson has quit IRC04:21
*** acormier has quit IRC04:23
*** gbarros has quit IRC04:28
*** oanson has joined #openstack-nova04:30
*** sree has joined #openstack-nova04:30
*** sridharg has joined #openstack-nova04:32
*** hongbin has quit IRC04:32
*** yamahata has joined #openstack-nova04:33
*** armax has quit IRC04:35
*** diga has joined #openstack-nova04:38
*** mdnadeem has joined #openstack-nova04:39
*** claudiub|2 has joined #openstack-nova04:45
*** ijw has joined #openstack-nova04:47
*** ijw_ has joined #openstack-nova04:50
*** gouthamr has quit IRC04:50
*** crushil has quit IRC04:51
*** ijw has quit IRC04:53
*** ijw_ has quit IRC04:58
*** pcaruana has joined #openstack-nova05:01
*** markvoelker_ has joined #openstack-nova05:03
*** namnh_ has joined #openstack-nova05:03
*** psachin has joined #openstack-nova05:04
*** markvoelker has quit IRC05:05
*** ratailor has joined #openstack-nova05:07
*** namnh_ has quit IRC05:07
*** manasm has joined #openstack-nova05:10
*** masber has joined #openstack-nova05:12
*** lpetrut has joined #openstack-nova05:16
*** edand has joined #openstack-nova05:16
*** lnxnut has joined #openstack-nova05:18
*** mingyu has joined #openstack-nova05:20
*** pcaruana has quit IRC05:23
*** mingyu has quit IRC05:25
*** lnxnut has quit IRC05:28
*** oanson has quit IRC05:36
*** oanson has joined #openstack-nova05:36
*** markvoelker_ has quit IRC05:37
*** markvoelker has joined #openstack-nova05:37
*** Eran_Kuris has joined #openstack-nova05:41
*** markvoelker has quit IRC05:42
*** tetsuro has joined #openstack-nova05:45
*** hieulq has quit IRC05:49
*** phuongnh has quit IRC05:49
*** manasm has quit IRC05:49
*** phuongnh has joined #openstack-nova05:49
*** hieulq has joined #openstack-nova05:49
*** manasm has joined #openstack-nova05:51
*** josecastroleon has joined #openstack-nova05:53
*** rcernin has joined #openstack-nova05:53
*** mingyu has joined #openstack-nova05:55
*** jamesdenton has quit IRC05:56
*** lpetrut has quit IRC06:09
*** Oku_OS-away is now known as Oku_OS06:09
*** mvk has quit IRC06:13
*** sahid has joined #openstack-nova06:13
*** rletrocquer has quit IRC06:13
*** Apoorva has joined #openstack-nova06:23
*** namnh_ has joined #openstack-nova06:24
*** nikhil has quit IRC06:24
*** lnxnut has joined #openstack-nova06:24
*** hoonetorg has quit IRC06:27
*** namnh_ has quit IRC06:28
*** aloga has quit IRC06:29
*** lajoskatona has joined #openstack-nova06:30
*** aloga has joined #openstack-nova06:30
*** Apoorva has quit IRC06:30
*** lnxnut has quit IRC06:34
*** slaweq has quit IRC06:35
*** slaweq has joined #openstack-nova06:36
*** ociuhandu has joined #openstack-nova06:38
*** slaweq has quit IRC06:40
*** hoonetorg has joined #openstack-nova06:41
*** slaweq has joined #openstack-nova06:42
*** mlakat has joined #openstack-nova06:43
*** pcaruana has joined #openstack-nova06:44
*** hferenc has quit IRC06:48
*** cfriesen has quit IRC06:54
*** esberglu has joined #openstack-nova06:55
*** esberglu has quit IRC06:55
*** lpetrut has joined #openstack-nova06:56
*** Eran_Kuris has quit IRC07:00
*** lpetrut has quit IRC07:05
openstackgerritMerged openstack/nova master: check query param for service's index function  https://review.openstack.org/48949207:07
openstackgerritMerged openstack/nova master: Do not setup conductor in BaseAPITestCase  https://review.openstack.org/50812007:08
*** ragiman has joined #openstack-nova07:11
*** migi_ is now known as migi07:13
*** tesseract has joined #openstack-nova07:17
*** hferenc has joined #openstack-nova07:19
*** ralonsoh has joined #openstack-nova07:24
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove 400 as expected error  https://review.openstack.org/50903907:27
*** brault has joined #openstack-nova07:28
*** sree has quit IRC07:30
*** lnxnut has joined #openstack-nova07:31
*** sree has joined #openstack-nova07:31
*** sree has quit IRC07:35
*** markvoelker has joined #openstack-nova07:38
*** jpena|off is now known as jpena07:39
*** lnxnut has quit IRC07:41
openstackgerritTakashi NATSUME proposed openstack/nova-specs master: Abort Cold Migration  https://review.openstack.org/33473207:44
*** brault has quit IRC07:46
*** brault has joined #openstack-nova07:46
bauzasgood specs day to everyone07:47
openstackgerritRadoslav Gerganov proposed openstack/nova master: VMware: serial console log (completed)  https://review.openstack.org/45063607:48
*** esberglu has joined #openstack-nova07:48
*** esberglu has quit IRC07:53
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Move 'ips' field from Subnet object to VIF object  https://review.openstack.org/50849807:53
*** josecastroleon has quit IRC07:55
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix test_get_volume_config method  https://review.openstack.org/48946707:58
*** takashin has left #openstack-nova08:00
*** alexchadin has joined #openstack-nova08:00
*** Eran_Kuris has joined #openstack-nova08:02
*** sree has joined #openstack-nova08:05
*** manasm has quit IRC08:06
*** mvk has joined #openstack-nova08:10
*** markvoelker has quit IRC08:12
*** ociuhandu has quit IRC08:14
*** boris_42_ has quit IRC08:15
openstackgerritsahid proposed openstack/nova master: pci: update PciDevice object field 'address' to accept NULL  https://review.openstack.org/50817508:19
openstackgerritsahid proposed openstack/nova master: pci: add for PciDevice object new field mdev  https://review.openstack.org/50817608:19
openstackgerritsahid proposed openstack/nova master: pci: generalize object unit-tests for different framework  https://review.openstack.org/50817708:19
openstackgerritsahid proposed openstack/nova master: pci: add support for mdev device type request  https://review.openstack.org/50817808:19
openstackgerritsahid proposed openstack/nova master: pci: generalize stats unit-tests for different framework  https://review.openstack.org/50817908:19
openstackgerritsahid proposed openstack/nova master: pci: add support for mdev devices type in devspec  https://review.openstack.org/50818008:19
openstackgerritsahid proposed openstack/nova master: pci: add support for resource pool stats of mdev devices  https://review.openstack.org/50818108:19
openstackgerritsahid proposed openstack/nova master: pci: make manager to accept handling mdev devices  https://review.openstack.org/50818208:19
openstackgerritsahid proposed openstack/nova master: libvirt: update PCI node device to report mdev devices  https://review.openstack.org/50818308:19
openstackgerritsahid proposed openstack/nova master: libvirt: report mdev resources  https://review.openstack.org/50818408:19
openstackgerritsahid proposed openstack/nova master: libvirt: add support to start vm with using mdev (vGPU)  https://review.openstack.org/50818508:19
openstackgerritsahid proposed openstack/nova master: functional: rework fakelibvirt host pci devices  https://review.openstack.org/50818608:19
openstackgerritsahid proposed openstack/nova master: functional: resuse SRIOV funtional tests for MDEV devices  https://review.openstack.org/50818708:19
openstackgerritsahid proposed openstack/nova master: WIP - functional: rewrite class to generate pci devices  https://review.openstack.org/50883508:19
*** oanson has quit IRC08:21
*** oanson has joined #openstack-nova08:22
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Change Subnet.dhcp_server to contain multiple IPs  https://review.openstack.org/50910708:25
*** alexchadin has quit IRC08:27
*** lucas-afk is now known as lucasagomes08:30
*** alexchadin has joined #openstack-nova08:35
*** trinaths has joined #openstack-nova08:37
*** mvk has quit IRC08:38
*** lnxnut has joined #openstack-nova08:39
*** spectr has joined #openstack-nova08:41
*** manasm has joined #openstack-nova08:42
*** tpatil has joined #openstack-nova08:44
*** derekh has joined #openstack-nova08:44
*** cdent has joined #openstack-nova08:48
*** lnxnut has quit IRC08:48
*** tpatil has quit IRC08:50
*** mvk has joined #openstack-nova08:54
*** gszasz has joined #openstack-nova08:56
*** diga has quit IRC09:00
*** markvoelker has joined #openstack-nova09:09
*** alexchadin has quit IRC09:10
*** tssurya has joined #openstack-nova09:11
openstackgerritJan Zerebecki proposed openstack/nova master: Fix wording of debug message for future releases  https://review.openstack.org/50826109:13
*** ociuhandu has joined #openstack-nova09:16
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Add support for Windows network commands  https://review.openstack.org/48740509:17
*** mvk has quit IRC09:20
openstackgerritJan Zerebecki proposed openstack/nova master: Fix wording of debug message for future releases  https://review.openstack.org/50826109:22
*** sdague has joined #openstack-nova09:23
*** yamamoto has quit IRC09:24
*** priteau has joined #openstack-nova09:25
*** diga has joined #openstack-nova09:25
*** igordc has quit IRC09:26
*** igordc has joined #openstack-nova09:27
cdentbauzas: that’s a great looking dog you have09:27
bauzascdent: hah thanks09:28
bauzasEurasier FTW09:28
*** TuanLA has joined #openstack-nova09:28
bauzasFWIW, since it's specs review day, I'm just fixing my own dashs because of Gerrit 2.13 :)09:30
*** alexchadin has joined #openstack-nova09:31
bauzasin case people face the same problem, sdague explained the issue in http://lists.openstack.org/pipermail/openstack-dev/2017-September/122277.html09:31
*** mvk has joined #openstack-nova09:34
*** Eran_Kuris has quit IRC09:35
*** belmoreira has joined #openstack-nova09:39
*** yamamoto has joined #openstack-nova09:41
*** markvoelker has quit IRC09:42
*** rmart04 has joined #openstack-nova09:44
*** edand has quit IRC09:44
*** sree has quit IRC09:45
*** sree has joined #openstack-nova09:45
*** yamamoto has quit IRC09:46
*** lpetrut has joined #openstack-nova09:46
*** lnxnut has joined #openstack-nova09:46
*** yamamoto has joined #openstack-nova09:46
*** Eran_Kuris has joined #openstack-nova09:47
*** sree has quit IRC09:50
*** udesale has quit IRC09:51
*** yamamoto has quit IRC09:51
*** alexchadin has quit IRC09:53
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova-specs master: Intel Fortville Dynamic Device Personalization (DDP)  https://review.openstack.org/50300109:53
*** yassine has quit IRC09:56
*** lnxnut has quit IRC09:56
*** jangutter has joined #openstack-nova09:57
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova-specs master: Intel Fortville Dynamic Device Personalization (DDP)  https://review.openstack.org/50300109:59
*** yamamoto has joined #openstack-nova10:00
*** priteau has quit IRC10:00
*** chyka has joined #openstack-nova10:00
*** lpetrut_ has joined #openstack-nova10:04
*** chyka has quit IRC10:05
*** lpetrut has quit IRC10:06
*** yamahata has quit IRC10:08
*** phuongnh has quit IRC10:13
*** edand has joined #openstack-nova10:14
*** belmoreira has quit IRC10:16
*** sam_nowitzki has quit IRC10:17
*** mingyu has quit IRC10:18
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova-specs master: Enable SR-IOV NIC offload feature discovery  https://review.openstack.org/50489510:20
*** claudiub has joined #openstack-nova10:22
*** claudiub|2 has quit IRC10:25
*** brault has quit IRC10:29
openstackgerritChris Dent proposed openstack/nova-specs master: Fix issues for post-allocations spec  https://review.openstack.org/50913610:31
cdentefried: that ^ should fix some of your comments, thanks for them10:32
*** pooja_jadhav has quit IRC10:33
*** esberglu has joined #openstack-nova10:33
*** esberglu has quit IRC10:33
openstackgerritStephen Finucane proposed openstack/nova master: Remove dead code of api.fault notification sending  https://review.openstack.org/50516410:37
*** oanson has quit IRC10:38
*** markvoelker has joined #openstack-nova10:39
*** oanson has joined #openstack-nova10:40
*** aloga has quit IRC10:40
*** aloga has joined #openstack-nova10:43
openstackgerritStephen Finucane proposed openstack/nova master: doc: Add documentation for cpu_realtime, cpu_realtime_mask  https://review.openstack.org/50205610:45
stephenfingibi, bauzas: Any chance of a (re-)review of ^10:46
openstackgerritChris Dent proposed openstack/nova-specs master: Spec for limiting GET /allocation_candidates  https://review.openstack.org/50454010:52
*** lnxnut has joined #openstack-nova10:53
*** tbachman has quit IRC10:54
*** alexchadin has joined #openstack-nova10:57
tetsuroThe patch I brought in the latest PTG is ready for review https://review.openstack.org/#/c/465160/.10:59
tetsuroThis is a bug that falls into silent error with NUMATopology when virt_type is not set to kvm.10:59
*** edand has quit IRC11:02
*** lnxnut has quit IRC11:02
*** udesale has joined #openstack-nova11:03
*** edand has joined #openstack-nova11:06
*** spectr has quit IRC11:07
*** nicolasbock_ has joined #openstack-nova11:09
*** mingyu has joined #openstack-nova11:09
*** READ10 has quit IRC11:12
*** markvoelker has quit IRC11:12
*** yassine has joined #openstack-nova11:13
*** priteau has joined #openstack-nova11:14
*** lucasagomes is now known as lucas-hungry11:16
gibistephenfin: done. +211:18
*** nicolasbock_ has quit IRC11:18
*** smatzek has joined #openstack-nova11:18
gibistephenfin: thanks for the respin of the api.fault check. Is there something on master that will help pleasing jenkins? Should I rebase my other patches as well?11:18
*** yangyapeng has joined #openstack-nova11:25
*** esberglu has joined #openstack-nova11:27
*** nicolasbock_ has joined #openstack-nova11:31
*** esberglu has quit IRC11:31
*** yangyapeng has quit IRC11:34
*** yangyapeng has joined #openstack-nova11:35
*** chyka has joined #openstack-nova11:35
*** yangyapeng has quit IRC11:39
*** chyka has quit IRC11:40
*** tetsuro has quit IRC11:47
*** vladikr has joined #openstack-nova11:47
*** edmondsw has joined #openstack-nova11:50
*** brault has joined #openstack-nova11:50
*** tbachman has joined #openstack-nova11:54
openstackgerritChris Dent proposed openstack/nova master: Update RT aggregate map less frequently  https://review.openstack.org/48963311:54
*** udesale has quit IRC11:57
*** artom has joined #openstack-nova11:57
*** thorst has joined #openstack-nova11:58
*** lnxnut has joined #openstack-nova11:59
*** thorst_ has joined #openstack-nova12:00
*** liverpooler has joined #openstack-nova12:00
*** liverpooler has quit IRC12:00
*** liverpooler has joined #openstack-nova12:01
*** thorst has quit IRC12:02
*** artom has quit IRC12:03
*** chyka has joined #openstack-nova12:05
*** TuanLA has quit IRC12:07
*** markvoelker has joined #openstack-nova12:09
*** lnxnut has quit IRC12:09
*** thorst_ has quit IRC12:10
*** jmlowe has quit IRC12:10
*** chyka has quit IRC12:10
*** jaypipes has joined #openstack-nova12:10
*** john5223_ has joined #openstack-nova12:11
*** markvoelker has quit IRC12:12
*** markvoelker has joined #openstack-nova12:12
*** jmlowe has joined #openstack-nova12:14
*** manasm has quit IRC12:16
*** lucas-hungry is now known as lucasagomes12:17
*** jpena is now known as jpena|lunch12:19
*** thorst has joined #openstack-nova12:20
*** udesale has joined #openstack-nova12:22
*** thorst has quit IRC12:24
*** spectr has joined #openstack-nova12:24
*** aloga has quit IRC12:25
*** aloga has joined #openstack-nova12:26
*** jamesdenton has joined #openstack-nova12:26
*** baoli has joined #openstack-nova12:27
*** dave-mccowan has joined #openstack-nova12:28
*** erlon has joined #openstack-nova12:32
*** manasm has joined #openstack-nova12:33
*** catintheroof has joined #openstack-nova12:36
*** catintheroof has quit IRC12:36
*** catintheroof has joined #openstack-nova12:36
stephenfinsahid: Can you reword the comment here? I'm not sure what you mean https://review.openstack.org/#/c/461456/5/nova/virt/hardware.py12:40
*** catintheroof has quit IRC12:41
*** baoli has quit IRC12:41
*** baoli has joined #openstack-nova12:42
*** hemna__ has joined #openstack-nova12:42
*** thorst has joined #openstack-nova12:44
*** rmart04 has quit IRC12:44
sahidstephenfin: yes i will, there are several points on what i'm not agree with that change12:45
*** alexchadin has quit IRC12:46
stephenfinsahid: Cool, thanks. I don't personally see the downside so there's a chance I'm missing something12:46
*** mdnadeem has quit IRC12:48
*** acormier has joined #openstack-nova12:48
sahidwell firstable we add new syntax to provide the same fonctionnality, the previous one already brought bugs, adding a new one is not going to help12:49
*** baoli has quit IRC12:49
*** smatzek has quit IRC12:50
*** pchavva has joined #openstack-nova12:50
*** chyka has joined #openstack-nova12:51
*** esberglu has joined #openstack-nova12:51
sahidthen with this new syntax, we allow users to specify the set of vCPUs used to apply the mask on where in my sense that shouldn't be allowed, the mask is applied on the set of vCPUs of the guest which Nova do provide12:51
*** smatzek has joined #openstack-nova12:52
*** smatzek has quit IRC12:52
*** smatzek has joined #openstack-nova12:52
*** udesale has quit IRC12:52
*** acormier has quit IRC12:53
stephenfinYeah, they do achieve basically the same thing. However, from a usability perspective, I do think this second pattern is more intuitive12:54
*** tbachman has quit IRC12:54
stephenfinI mean, if the option was called 'cpu_realtime_excludes' or something, then the current pattern would make sense (and the carat wouldn't be necessary)12:54
*** catintheroof has joined #openstack-nova12:54
*** lyan has joined #openstack-nova12:55
stephenfinHowever, while nova definitely should keep track of total number of CPUs, I don't see why we should allow "explicit exclude-implicit include", but not "explicit include-implicit exclude"12:55
*** alexchadin has joined #openstack-nova12:55
*** chyka has quit IRC12:55
stephenfin"explicit include-explicit exclude" is an odd one, but it's no harm and already works, so if someone's silly enough to do it I don't see why we shouldn't just let them12:56
stephenfinThis is a usability improvement and nothing more, IMO12:57
*** ratailor has quit IRC12:57
sahidok that is the subjective point, i can hear it. what about the fact we add more and more code to achieve the same result?12:57
*** acormier has joined #openstack-nova12:59
*** acormier has quit IRC13:04
*** mriedem has joined #openstack-nova13:07
*** manasm has quit IRC13:07
*** yangyapeng has joined #openstack-nova13:07
*** lnxnut has joined #openstack-nova13:07
*** mdnadeem has joined #openstack-nova13:08
*** alexchadin has quit IRC13:09
*** yangyapeng has quit IRC13:09
mriedemblarg my nova-specs dashboard doesn't work with new gerrit13:13
*** alexchadin has joined #openstack-nova13:13
cdentmriedem: bauzas had some links earlier today about ways to fix that13:14
bauzasmriedem: yeah, I had to modify my own dashes13:14
cdentmriedem: http://lists.openstack.org/pipermail/openstack-dev/2017-September/122277.html13:14
sdaguebauzas / mriedem are you using the ones from upstream?13:14
bauzasmriedem: tl;dr: labeling your votes doesn't work, you need to use another one13:15
sdagueI tried to fix everything in that repo13:15
sdaguebut, I'm sure they could use tweaks13:15
stephenfinYeah, https://github.com/openstack/gerrit-dash-creator/blob/master/dashboards/nova-specs.dash looks good to me13:15
stephenfinand I'm using https://github.com/openstack/gerrit-dash-creator/blob/master/dashboards/nova.dash daily13:15
bauzassdague: I'm using the same queries than yours, but I use https://review.openstack.org/#/settings/preferences13:15
*** MarginHu has joined #openstack-nova13:16
sdaguebauzas: ok, cool13:16
*** yangyapeng has joined #openstack-nova13:16
sdaguebauzas: yeh, I end up just using browser bookmarks for them as they sync between chrome13:16
bauzassdague: so, tbh I modified my queries thanks to you :)13:16
*** kylek3h has joined #openstack-nova13:16
*** yamamoto has quit IRC13:16
bauzasthat, plus Zuul v3 \o/13:17
*** lnxnut has quit IRC13:17
*** smatzek has quit IRC13:17
*** awaugama has joined #openstack-nova13:17
bauzasis anyone currently working on moving our jobs to the nova repo btw. ?13:17
bauzassdague: do you know that ^ ?13:18
*** eharney has joined #openstack-nova13:18
*** liverpooler has quit IRC13:19
jaypipessdague, mriedem: is it worth rechecking anything at the moment?13:19
sdaguejaypipes: I don't know, I rechecked a few things, I'll let you know if anything works13:19
mriedemjaypipes: don't think so13:19
jaypipessdague: seeing a lot of POST_FAILURE stuff right now...13:19
sdaguebauzas: no, but given that zuul v3 is still not passing many things, it didn't seem really useful to change the queries yet13:19
mriedemwhat i have rechecked is not queueing up13:19
*** yangyap__ has joined #openstack-nova13:19
sdaguejaypipes: yeh... that's what I saw from stuff that hit lastnight13:20
jaypipesmordred: any particular activity that would be useful from us in identifying issues with Zuulv3?13:20
bauzassdague: well, you're right13:20
*** yangyapeng has quit IRC13:20
bauzasjaypipes: there is an etherpad13:20
bauzasjaypipes: https://etherpad.openstack.org/p/zuulv3-migration-faq13:21
jaypipesbauzas: cheers13:21
*** mdnadeem has quit IRC13:21
* bauzas took a couple of time to look at all the zuul v3 docs :)13:21
*** bnemec has joined #openstack-nova13:21
sdaguemriedem: you want me to restrict down the specs dashboard to only stuff proposed for queens?13:21
*** smatzek has joined #openstack-nova13:22
bauzassdague: it's another dash I guess13:22
bauzassdague: honestly, I'm directly querying for that13:22
*** baoli has joined #openstack-nova13:22
mriedemsdague: no that's fine13:23
*** yangyap__ has quit IRC13:24
*** jpena|lunch is now known as jpena13:24
sdaguehttps://goo.gl/JxQn1r is an attempt to slice things off13:25
*** yangyapeng has joined #openstack-nova13:25
sdagueso queens for everything, then a bucket at the bottom for non queens stuff13:25
openstackgerritJohn Garbutt proposed openstack/nova master: Re-use existing ComputeNode on ironic rebalance  https://review.openstack.org/50855513:26
*** yangyapeng has quit IRC13:27
*** jmlowe has quit IRC13:27
openstackgerritLee Yarwood proposed openstack/nova-specs master: Libvirt: Native LUKS decryption by QEMU  https://review.openstack.org/49082413:28
*** lpetrut_ has quit IRC13:30
*** manasm has joined #openstack-nova13:30
*** smatzek has quit IRC13:32
*** smatzek has joined #openstack-nova13:33
*** smatzek has quit IRC13:34
*** yamamoto has joined #openstack-nova13:35
*** edand has quit IRC13:35
jaypipescdent: there isn't a call to get the aggregates for >1 resource provider is there?13:37
* cdent tries to remember13:38
jaypipescdent: nm, no there isn't...13:38
* cdent decompresses archived memories13:38
cdentjaypipes: I believe you are correct13:38
jaypipescdent: was thinking if there was, we could further simplify your agg perf patch to do a single batch call.13:38
jaypipescdent: but meh, no worries :)13:38
*** brault has quit IRC13:39
*** brault has joined #openstack-nova13:39
cdentI had a version which was more brute force (but as a result required multiple requests) and decided that was not the way to go.13:40
*** felipemonteiro has joined #openstack-nova13:40
jaypipescdent: no, this looks quite good. ++13:40
*** mdnadeem has joined #openstack-nova13:40
*** ijw has joined #openstack-nova13:40
*** lbragstad has joined #openstack-nova13:41
jaypipesmriedem, dansmith: https://review.openstack.org/#/c/489633/ looks like a small, well-scoped patch with a good performance benefit.13:41
bauzascdent: just a slight concern about possibly overthinking about times with https://review.openstack.org/#/c/496853/313:42
cdentbauzas: your etag fear has been noted on your api merit badge worksheet as a demerit13:42
*** alexchadin has quit IRC13:43
*** alexchadin has joined #openstack-nova13:43
*** brault has quit IRC13:44
bauzashonestly, I just want to make sure we have a very small implementation for that13:44
cdentbauzas: on the last-modified time: since the time info is already there, seems like we may as well use it, because the spec says we should (if possible) include the _real_ time of update13:44
bauzassure, I understand that13:44
bauzasbut IMHO, keeping it simple for Queens isn't bad13:44
cdentsince we dismissed etags as part of placement long ago, I think we should stick with that plan, I only include the option to be complete13:44
bauzasunless you really wanna cache13:44
bauzasand then, we should possibly use etags13:44
bauzasso, my thoughts are : do the very small change and just use the current time, or use Etags :p13:45
cdentI think the last-modified time is useful metadata for generic users of the placement service, maybe not in nova-scheduler, but for random unknowns. And since I’ve alrady got a working implementation, it’s not too much effort to finish it13:45
*** mmehan has joined #openstack-nova13:46
bauzascdent: well, it needs then to leak out the DB details to the object13:46
bauzaswe do that for a lot of stuff of course13:46
cdentone sec13:46
bauzasbut I thought our placement objects shouldn't be using that13:46
bauzasanyway, I don't want to nitpick over it13:47
cdentbauzas: this is the wip, is not hard: https://review.openstack.org/#/c/495380/7/nova/objects/resource_provider.py13:47
cdentwe alraedy have those fields13:47
bauzasyeah I know13:47
bauzasusing a mixin isn't hard13:47
bauzasit's just we're exposing those DB details out to the API13:47
bauzascdent: is it the first OpenStack project doing that ? (please use your API SIG hat :p )13:48
cdentI don’t understand what you mean by “exposing those db details out the to the api”. You mean last modified time? Why is that a problem?13:48
sdaguemriedem: ok, I'm really struggling about why on - https://review.openstack.org/#/c/501017/2/specs/queens/approved/flavor-description.rst13:49
sdaguebecause that's really already there with flavor name13:49
jaypipesedleafe, bauzas, cdent, stephenfin, dansmith, mriedem: I think the only thing remaining on https://review.openstack.org/#/c/497713/10/specs/queens/approved/add-trait-support-in-allocation-candidates.rst is to agree on the name of the parameter. The options are traits=, required=, required_traits=. Let's just pick one. Please #vote for your first and second pick. I'll go first.13:49
jaypipes#vote 1. required=, 2. traits=13:50
bauzasI need to review that spec13:50
dansmith#vote 1. required 2. required= 3. requires=13:50
dansmithjaypipes: I thought everyone was okay with required=?13:50
*** josecastroleon has joined #openstack-nova13:50
jaypipesdansmith: I don't believe mriedem was and I know edleafe wasn't.13:50
edleaferequired= or traits= are fine with me13:50
dansmithedleafe: said he was13:50
cdentbauzas In my limited review of openstack apis, there’s a mix of who does and does not use last-modifed. If we’re looking to limit the amount of work in Queens, we could choose not to do this spec, it isn’t really required in any way.13:50
edleaferequires= is definitely not13:50
jaypipesor was it requires=...13:50
bauzasrequired= for me13:50
jaypipesyeah, sorry13:50
stephenfin#vote 1. traits=, 2. required=13:51
cdentmriedem expresed dislike for required, yes?13:51
dansmithjaypipes: I think mriedem said required= was okay too13:51
edleafeverbs don't work13:51
bauzasbecause we could implement preferred= later13:51
dansmithbauzas: exactly13:51
edleafepreferred won't be in placement, right?13:51
edleafethat's a weigher issue13:51
openstackgerritJohn Garbutt proposed openstack/nova master: Re-use existing ComputeNode on ironic rebalance  https://review.openstack.org/50855513:51
cdent#vote 1. required 2. required_traits13:52
mriedemi said i cared less about required= even though i didn't like it13:52
bauzasedleafe: I'm not advocating for it now13:52
mriedemi was -1 on the ?required='' thing13:52
bauzasedleafe: I'm just saying it *could* be possible13:52
dansmithedleafe: I don't think so, we still have to pull out things that might have that over things that don't at all, right?13:52
jaypipesedleafe: no plans *currently*, but we still would likely need a corresponding parameter for preferred to pass to the scheduler of course.13:52
cdentyeah, I fixed the required=‘’ thing, thank goodness13:52
mriedemsdague: i can't say how often people are changing flavor names randomly13:52
edleafejaypipes: ok, I thought we were just thinking of the placement api13:52
mriedemthere were a few operators that said they would use this on the spec13:52
jaypipesmriedem: what's your vote then? traits=?13:53
sdaguemriedem: sure, so let's just make name mutable13:53
bauzascdent: honestly, like I said, I don't want to take too much time on that spec, +Wd :)13:53
sdaguemriedem: I guess, it feels really weird to add a **3rd** 255 character string to flavor13:53
mriedemsdague: from a ux perspective i don't think name is the thing you want to have 255 characters of detail in13:53
*** liverpooler has joined #openstack-nova13:53
*** liverpooler has quit IRC13:53
sdaguemriedem: because it's called name?13:53
mriedemwe have server name and description13:53
*** liverpooler has joined #openstack-nova13:54
mriedemi don't expect the name of a resource to be super detailed13:54
mriedemjaypipes: huh? i can't handle 3 conversations at once plus reviews right now.13:54
sdaguemriedem: we do, but servers' don't also have a server_id field that's a 255 character string13:54
mriedemjaypipes: i thought agreement was on required=?13:54
*** liverpooler has quit IRC13:54
mriedemsince we could have preferred later13:54
*** baoli_ has joined #openstack-nova13:54
jaypipesmriedem: got it. ok, it's settled then.13:54
jaypipesI just wanted to hold a final vote.13:54
mriedemsdague: i don't know why flavor is the weirdo resource with a 255 char id field either13:55
sdaguemriedem: because flavor_id == server.name13:55
mriedembut id and name are generally short things13:55
sdagueand flavor.name == server.description13:55
sdagueor, at least should be treated as such13:55
*** liverpooler has joined #openstack-nova13:55
mriedemhow many clouds are filling out the full flavor.name to express everything in that field today?13:56
*** baoli has quit IRC13:56
mriedemincluding details about baremetal and extra specs13:56
sdaguemriedem: don't know, but if that's the concern I'd make the microversion start returning name as a description field instead, and make it mutable after that point13:57
*** brault has joined #openstack-nova13:57
mriedemi'm not going to do that13:57
mriedemwe also embed the flavor name in the instance details now,13:57
mriedemso if name starts becoming big ass description, then your output in things like nova show are going to bloat up13:58
sdaguemriedem: but that can already be an issue today13:58
*** acormier has joined #openstack-nova13:58
*** READ10 has joined #openstack-nova13:58
*** spectr has quit IRC13:58
*** crushil has joined #openstack-nova13:58
sdagueif you put a 64 char constraint on id and name at the same time, I'd be fine with it. But I think choose your own adventure on 3 255 character fields is going to lead to more confusion13:59
efriedcdent Nice, thanks.13:59
*** spectr has joined #openstack-nova13:59
*** cfriesen has joined #openstack-nova13:59
bauzasjaypipes: question in https://review.openstack.org/#/c/497713/1013:59
mriedemsdague: i don't see how it's confusing,13:59
mriedemname is the name,13:59
mriedemid is a uuid by default13:59
mriedemdescription is a description, name and description are different things13:59
*** kylek3h has quit IRC13:59
mriedemin english14:00
*** kylek3h has joined #openstack-nova14:00
bauzasjaypipes: tl;dr say I have a host with 2 PFs, and only one tagged with a trait14:00
sdaguebut your concern was also bloat because 255 characters should not be used14:00
bauzasjaypipes: if I'm asking for 1 VF with that specific trait, I wouldn't get that host ?14:00
jaypipesbauzas: responding on the review...14:00
bauzascool thanks :)14:00
sdagueso, if that's what you believe, then put a 40 character limit on flavor_id, a 64 on name, and add description as the mutable big one14:01
*** jmlowe has joined #openstack-nova14:01
mriedemsdague: that would be an upgrade issue for anyone that has larger values for those fields already, as a workaround14:01
bauzasjaypipes: oh and FWIW, just discovered https://www.instagram.com/itsdougthepug/14:01
sdaguemriedem: start with it as json schema enforcement14:01
mriedemthat might e ok14:02
efriedmriedem sdague jaypipes bauzas edleafe Put me to work, guys.14:02
mriedemreview specs14:03
edleafeefried: paint my house14:03
efriedRight right; any particular ones?  (Is there a dashboard to look at?14:03
efriededleafe Be there in 90 minutes14:03
jaypipesbauzas: :) on dougthepug14:04
bauzassorry, I'm living in the countryside14:04
sdaguemriedem: ok, well that's my current counter proposal in the spec. If we really believe people should only be doing flavor_id as uuid and name should be short, then enforce that on new types past that microversion.14:04
bauzasso, in case that's something people know like since 2 years, :p14:05
*** smatzek has joined #openstack-nova14:05
mriedemefried: https://goo.gl/QidAVs14:07
*** chyka has joined #openstack-nova14:08
bauzasjaypipes: sorry, my question wasn't clear14:09
bauzasjaypipes: if I'm taking your example14:09
*** smatzek has quit IRC14:09
mriedemlyarwood: a few small things in https://review.openstack.org/#/c/490824/ to update14:09
bauzasjaypipes: what if I'm having a node that is having 2 PFs, each of them having 8 VFs ?14:10
bauzasjaypipes: in a nested world, I'd have a root RP (the node) and 2 chidren (the PFs)14:10
*** smatzek has joined #openstack-nova14:10
*** smatzek has quit IRC14:10
bauzasjaypipes: then, if I'm asking for required=HW_NIC_OFFLOAD_TSO, should I get that node as a candidate?14:11
*** smatzek has joined #openstack-nova14:11
*** burt has joined #openstack-nova14:11
bauzasjaypipes: because alex_xu is saying NO to this https://review.openstack.org/#/c/497713/10/specs/queens/approved/add-trait-support-in-allocation-candidates.rst@4414:11
bauzasjaypipes: to clarify, only *one* PF would have HW_NIC_OFFLOAD_TSO14:12
bauzasthe other PF would have other traits14:12
edleafejaypipes: on https://review.openstack.org/#/c/498830/9/specs/queens/approved/return-selection-objects.rst - are you saying that I should drop the limits field entirely?14:12
*** nikhil_ has joined #openstack-nova14:13
*** chyka has quit IRC14:13
jaypipesedleafe: yes, and add a numa_limits field that is a NUMATopologyLimits object.14:13
*** nikhil_ is now known as Guest6184914:13
edleafejaypipes: ok, I'll revise14:13
*** jmlowe has quit IRC14:13
*** Guest61849 is now known as nikhil_k14:14
jaypipesbauzas: when you say "what if I'm having a node that is having 2 PFs" are you really saying "what if I am *requesting* an instance that has two PFs"?14:14
*** lnxnut has joined #openstack-nova14:14
bauzasjaypipes: no, that's probably where I'm unclear14:15
sdaguemriedem: does https://review.openstack.org/#/c/466595 solve cburgess's desire as well?14:15
*** jmlowe has joined #openstack-nova14:15
bauzasjaypipes: nevermind the SR-IOV language14:15
*** thorst has quit IRC14:15
bauzasand keep things simple : one root RP and 2 children RPs14:15
efriedbauzas jaypipes If I'm understanding the algorithm correctly, we'll hit the PF that has the right trait; and then look up that guy's root_provider_id to include in the result set.14:15
mriedemsdague: that is the exact same spec/blueprint we've shot down numerous times,14:16
mriedemwhich i commented on back in may14:16
efriedSo we don't even consider the child PF RP that lacks the trait in question.14:16
bauzasjaypipes: each of those children RPs would have an inventory of 'foo': 814:16
efriedmriedem Paperwork question: I don't see esberglu's https://review.openstack.org/#/c/503061/ in that dashboard.  (Expected to see it under "needs final +2")14:16
bauzasjaypipes: but only one would have a trait 'MISC_FOO'14:16
jaypipesbauzas: k14:17
bauzasjaypipes: my question is, if I'm asking in my query for resources=FOO:1&required=MISC_FOO, could I get that node ?14:17
jaypipesbauzas: yes14:17
jaypipesbauzas: Alex is talking about *aggregates* on line 43-44.14:18
mriedemsdague: -1ed again14:18
sdaguemriedem: sure, but if there really are a bunch of different operators that are all coming forward with "please we need this" it would be good to figure out who they are all14:18
jaypipesbauzas: he's being explicit that aggregates don't have traits associated with them. only RPs do.14:18
*** artom has joined #openstack-nova14:18
bauzasah, shiy14:18
mriedemsdague: don't know who they all are now, at one point i found the various proposals and blueprints for this, and ML threads,14:18
mriedemthat could be collected, but i'm not going to do it today,14:19
jaypipesbauzas: ah, hold up, I see now where your confusion lies14:19
mriedemit could probably be better served as a user survey question, but i do'nt really want to tie us to the response14:19
sdaguemriedem: right, that's fair, not actually asking you to do it14:19
jaypipesbauzas: one sec.14:19
mriedemwe've said, since boston,14:19
efriedjaypipes bauzas Yeah, from that point of view, the traits *do* in fact propagate upwards.14:19
efriedsort of14:19
sdaguebut cburgess might be the right volunteer for that activity14:19
sdaguegiven his interest14:19
mriedemimplement the cinder ephemeral backend and you can pass the volume type through extra specs - but that doesn't really pertain to non-ephemeral volumes we create on behalf of the user14:20
jaypipesbauzas: so, your confusion stems from the sentence "However, traits defined on a child14:20
jaypipesRP do not apply to the parent (ancestor) RPs"14:20
jaypipesbauzas: what he's saying is correct, but weird :)14:20
bauzasit doesn't bubble up14:20
*** namnh has joined #openstack-nova14:20
bauzasso I just want to make sure we walk in the tree14:20
efriedjaypipes bauzas I take that to mean: if the parent RP has *inventory*, you wouldn't get that *inventory*14:20
efriedIt would be weird (but not forbidden) for a parent to have inventory in the same resource class as its child.14:21
*** namnh has quit IRC14:21
jaypipesbauzas: the trait itself doesn't apply to the parent specifically, but when looking for providers to return in allocation candidates, any traits for any provider in a *tree* will be considered as being "traits of the tree"14:21
efriedThat's the case where the traits don't propagate upwards.14:21
*** namnh has joined #openstack-nova14:21
bauzaswell, I'd say I would easily imagine the same resources:FOO:9&required=MISC_FOO not getting the node if only the child is having 8 FOOs per child14:22
mriedemstephenfin: this has been around for a month without any patches https://bugs.launchpad.net/nova/+bug/171401714:22
openstackLaunchpad bug 1714017 in OpenStack Compute (nova) "User guide was not migrated to the nova repo" [Medium,In progress] - Assigned to Pavlukhin Max (mpavlukhin)14:22
mriedemstephenfin: so if you're keen maybe you want to take that over,14:23
jaypipesbauzas: that is correct.14:23
*** itlinux has joined #openstack-nova14:23
jaypipesbauzas: quantitatively it's not possible.14:23
mriedemstephenfin: keeping in mind we need to backport whatever the changes are14:23
mriedemso maybe step 1 is just import the user guide docs and backport, then step 2 is re-arrange and clean them up14:23
mriedemsince some of the config drive user guide docs are really for admins14:23
bauzasjaypipes: but I had the same concern without traits, say I have 2 children with each of them having 8 FOOs14:23
jaypipesbauzas: but remember that the max_unit/min_unit part of hte inventory records will solve that query problem.14:23
stephenfinmriedem: I was going to, but Takashi NATSUME (I don't know his IRC nick) might be on it14:23
bauzasjaypipes: if I'm asking for 9 FOOs, I wouldn't get them even if I could have 8 in a child and 1 in the other child14:24
*** lnxnut has quit IRC14:24
stephenfinsee https://bugs.launchpad.net/nova/+bug/172087314:24
openstackLaunchpad bug 1714017 in OpenStack Compute (nova) "duplicate for #1720873 User guide was not migrated to the nova repo" [Medium,In progress] - Assigned to Pavlukhin Max (mpavlukhin)14:24
*** Eran_Kuris has quit IRC14:24
*** jmlowe has quit IRC14:24
bauzasjaypipes: mmm, I don't see how that helps but okay :)14:24
jaypipesbauzas: if you request 9 foo and there are 2 child providers having 8 foo inventories, each of those inventory records would have max_unit = 8 and that would mean a failing WHERE condition on max_unit <= $requested_amount14:24
bauzasjaypipes: I certainly understand *why* it's failing14:25
bauzasjaypipes: but I can easily imagine operators asking for spreading their resources14:25
efriedyeah, that's a bug.14:25
mriedemstephenfin: ok14:25
stephenfinIf they haven't done anything by next week, I'll pick it up14:25
*** yamamoto has quit IRC14:26
bauzassnap, I need to disappear for my daily daddy work14:26
*** itlinux has quit IRC14:26
jaypipesbauzas: sorry, I'm not following you still. :(14:26
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Move 'ips' field from Subnet object to VIF object  https://review.openstack.org/50849814:26
* bauzas rushing out for 20 mins :(14:26
jaypipesefried: do you think bauzas is talking about the resources1=, required1= stuff we discussed at the PTG?14:27
*** cleong has joined #openstack-nova14:28
stephenfinjaypipes: Fancy taking a look at this some time today? https://review.openstack.org/#/c/502056/14:28
stephenfin(it's docs)14:28
*** krtaylor has quit IRC14:29
efriedjaypipes A side effect thereof, yes.14:29
*** sambetts|afk is now known as sambetts14:29
efriedjaypipes bauzas The two scenarios starting on L83 here: https://etherpad.openstack.org/p/nova-multi-alloc-request-syntax-brainstorm14:30
efried...as far as spreading resources across RPs.14:30
*** hongbin has joined #openstack-nova14:31
*** thorst has joined #openstack-nova14:31
jaypipesefried: ok. I don't disagree that any of those scenarios are important. just that it's kind of a distraction at this point and an acknowledged thing we'll need to look at at a later time14:32
efriedFor the other discussion - how do traits "propagate" - perhaps it should be phrased: "With nested resource providers, traits defined on a parent RP are assumed to belong to all its child (descendant) RPs >>for purposes of inventory allocation<<. Traits defined on a child RP do not apply to the parent (ancestor) RPs.  >>Although inventory is only claimed from a RP having the requested traits, the entire RP tree is returned in the provider summary for14:33
efriedthat allocation candidate.<<"14:33
efriedor something like that.14:34
* cdent stays out of this for the sake of progress ;)14:35
efriedcdent Glad you're staying out of it.14:36
*** felipemonteiro has quit IRC14:37
efriedbauzas jaypipes We may want to defer this discussion; replace that chunk with "how traits are handled in nested RPs is out of scope and will be discussed in a subsequent spec".  But that spec needs to land in Queens.14:37
sdaguemelwitt: https://review.openstack.org/#/c/486947/ is a quotas spec that would be great to have your commentary on14:37
efriedjaypipes The "numbered resources querystring" spec would be a reasonable place for that discussion.14:38
mriedemdansmith: you should probably go through this https://review.openstack.org/#/c/433603/14:39
dansmithmriedem: ugh, do I have to?14:39
mriedemand finish your peas14:39
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova-specs master: Network bandwitdh resource provider  https://review.openstack.org/50230614:40
efriedjaypipes How does the scheduler determine that a particular RP is a compute host for the purposes of landing an instance?14:42
efried(Yes, this is ultimately relevant to the discussion)14:43
*** MarginHu has quit IRC14:43
*** priteau has quit IRC14:44
jaypipesefried: right now, the scheduler has a mapping of hostname to compute node UUIDs. the hostname is the same as the nova-compute service RPC topic queue14:45
dansmithmriedem: gawd moooom14:45
jaypipesefried: https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L64014:45
efriedjaypipes Okay, cool.  Cause when we have more than just compute host RPs, that's going to be pretty crucial.14:46
jaypipesefried: we already do...14:46
*** corey_ has joined #openstack-nova14:46
jaypipesefried: Ironic nodes and shared storage providers are not compute nodes.14:46
efriedjaypipes Nested, then.  Where that ties in is, you'll get your inventory from the child RP, but the whole tree is going to be (explicitly or implicitly via root_provider_id) part of your provider summaries.  That's going to have to be how the scheduler figures out which compute node it should go to.14:47
*** cleong has quit IRC14:47
mriedemcdent: L105 here https://review.openstack.org/#/c/506552/4/specs/queens/approved/allow-update-instance-keypair.rst14:48
mriedemPUT /servers/{server_id/14:48
efriedjaypipes Cause at some point, there's gonna be a model where the compute node RP actually doesn't have *any* resources (e.g. VCPU and MEMORY_MB belong to NUMA node RPs under the compute host; DISK_GB belongs to a shared RP somewhere; etc.)14:48
mriedemif the server is not found by id, it's a 404, but if something in the request body isn't found, then it's a 400, right?14:48
cdentmriedem: correct, that’s the general rule14:48
jaypipesefried: the allocation_requests part of the GET /allocation_candidates HTTP response has the exact allocations against the exact resource providers that the instance will consume from (compute node providers and otherwise)14:48
cdentif the uri fails to hit, 404, otherwise 40014:48
*** esberglu has quit IRC14:49
*** esberglu has joined #openstack-nova14:49
efriedjaypipes Right, but see above (..:48:16).  The compute host RP may not actually be in the list of exact RPs the instance will consume from).14:50
*** jmlowe has joined #openstack-nova14:50
*** alexchadin has quit IRC14:50
mriedemsdague: i see you found the competing instance keypair update specs14:50
mriedemand i know you saw the ML thread14:50
* artom never did figure out if they were competing or complementary14:50
mriedemwhat i don't know is what most deployments do about cloud-init behavior14:50
mriedemif they refresh on reboot or only on new build14:51
jaypipesefried: the compute node provider will always be the root_provider_id, though.14:51
efriedjaypipes Right.  That's my point.  Generically, the scheduler will have to find the compute host by backtracking to the root_provider_id.14:51
jaypipesefried: and the provider_summaries part of the GET /allocation_candidates HTTP response informs the caller that a particular provider is a child of another.14:51
jaypipesefried: yes, that is true.14:52
jaypipesefried: and expected.14:52
efriedjaypipes Right.  That's another question I've had, though: will provider summaries include the whole tree, or only the "exact resource providers that the instance will consume from"?14:52
jaypipesefried: the placement doesn't know or care that a particular provider represents a compute node.14:52
efriedright - the scheduler has to figure that out, I get that.14:52
efriedhence my original leading question.14:53
jaypipesefried: placement doesn't care about anything other than inventory and allocation records...14:53
jaypipesefried: so it's the scheduler's responsibility to understand those things.14:53
*** ijw has quit IRC14:53
jaypipesefried: currently, the scheduler does that by keeping that map of service RPC to compute node UUID in memory. It will need to be adapted to understand nested/root providers14:53
*** esberglu has quit IRC14:53
efriedjaypipes Will provider summaries include the whole tree, or only the exact RPs being consumed from?14:54
jaypipesefried: the exact RPs that could be consumed from and their parents.14:54
jaypipesefried: which is needed for callers to piece the tree together.14:54
efriedup to the root14:54
efriedI dig that.  Saves an extra call from the scheduler.14:55
jaypipesefried: I wasn't planning on making provider_summaries into a tree structure, though. was relying on the caller to do that as needed.14:55
*** eharney has quit IRC14:56
efriedI don't see the need for the whole tree - except that's easier to build, cause you'll already have that code for that API (forget which one) that returns a whole tree.14:56
efriedThe chain of parents is simple enough, but a separate algorithm.14:56
bauzasefried: jaypipes: sorry I had to disappear due to family business14:56
bauzasbut yeah let's punt that discussion14:56
* jaypipes gets out his red typo pen for ralonsoh's Network "bandwitdh" spec.14:56
*** MikeW has left #openstack-nova14:56
*** manasm has quit IRC14:56
jaypipesbauzas: commented on the spec.14:57
ralonsohjaypipes: hello. About https://review.openstack.org/#/c/502306/2/specs/queens/approved/bandwidth-resource-provider.rst@111. I don't understand what you mean. Should ML2 plugin be able to make RP claims? Or create/delete allocation records?14:57
bauzasthe thoughts that I had was about saying "what if I'm asking for more than the inventory amount supported by one child"14:57
*** brault has quit IRC14:57
cdentjaypipes: I just had a quick confused run through there. Looks like some fundamental misconceptions.14:57
jaypipesralonsoh: heh, I was just making a joke about the repeated misspellings of the word "bandwidth" :)14:58
ralonsohjaypipes: sorry!14:58
jaypipesIncluding in the spec title ;)14:58
jaypipesralonsoh: it's cool, I'm just kidding with ya14:58
* cdent wonders what the name of dandysmith’s horse is14:58
jaypipescdent: "Horse".14:59
cdentsuch disappoint14:59
jaypipesralonsoh: I'll comment on the spec, k?14:59
ralonsohjaypipes: perfect! and thanks15:00
jaypipesralonsoh: no problemo15:00
mriedemstephenfin: you've made the diff on this impossible https://review.openstack.org/#/c/496160/15:00
mriedemto compare to the pike spec15:00
*** brault has joined #openstack-nova15:00
mriedemstephenfin: if you're re-proposing a spec from a previous release, please leave the stylistic changes for a separate patch15:00
stephenfinmriedem: How so?15:00
mriedemyou change all the line wrapping15:01
mriedemi think you are going to need to setup a dr visit about your serious ocd issue15:01
jaypipeswait... mriedem is calling someone ocd?15:02
cdentmy thoughts exactly15:02
stephenfinmriedem: Ah, I can undo those, but the only functional changes are the things that were already commented on15:02
mriedemstephenfin: already approved15:04
mriedembut for next time15:04
stephenfinmriedem: (y)15:04
mriedemjaypipes: i'm not the one that -1s your changes for putting closing ] and } and ) on separate lines :P15:05
jaypipesmriedem: lol, tru dat15:05
jaypipesmriedem: that would be edleafe and dansmith, or as I call them "The Parens Posse"15:05
*** Oku_OS is now known as Oku_OS-away15:06
*** eharney has joined #openstack-nova15:08
* edleafe is happy to be part of a gang15:09
cdentdan’s down with tpp15:09
cfriesentpp ya you know me...15:09
*** thorst has quit IRC15:10
sdaguemriedem: yeh, I don't know either15:10
*** armax has joined #openstack-nova15:10
*** chyka has joined #openstack-nova15:10
sdaguealso, I think there is lots of confusion about "rebuild"15:10
sdaguebecause even in the ML thread where someone said "yes, this would be great!"15:10
sdagueand I asked "on just rebuild or you want reboot", A: "we never use rebuild"15:10
*** thorst has joined #openstack-nova15:11
mriedemi'm trying to get some more answers in the ops channel15:11
*** thorst has quit IRC15:11
mriedemand the answer i'm getting there is, "we don't use cloud-init to manage keys in the guest"15:11
sdagueum... so huh what?15:12
*** rcernin has quit IRC15:12
sdaguemriedem: joined there late, who said that?15:12
mriedembloomberg runs a script which is in the image that refreshes the users for the vm, or something15:12
sdagueyeh, cool, so they don't use the keypairs api then15:13
sdaguewhich, honestly, makes sense15:13
jaypipesmriedem: probably they mean there's a single key they use for chef and then rely on chef to inject/write other user keys.15:13
sdaguedo the whole thing out of band15:13
*** mdnadeem has quit IRC15:14
mriedemcburgess: do you have input on this keypair update thing? do you configure cloud-init in the images to refresh keys on reboot or only new build of the guest?15:14
sdaguemriedem: I think the question probably should be A) do you regularly use the keypairs api15:15
mriedembasically what i don't like about Kevin's spec is the 2-3 step dance, which varies from cloud to cloud based on how cloud-init is configured in the image15:15
mriedembecause it's (1) update server with new key_name, (2) reboot and check if that worked, else (3) rebuild15:15
*** kfarr has joined #openstack-nova15:15
*** priteau has joined #openstack-nova15:15
mriedemrestricting it to just rebuild is definitely simpler15:15
mriedemmaybe infra users could help here, since they also have a use case for updating keypair in an instance15:16
*** dave-mccowan has quit IRC15:17
mriedemi've also considered this might be something that needs to be discussed at the summit with ops and users in the room15:18
*** trinaths has left #openstack-nova15:18
cfriesencdent: for https://review.openstack.org/#/c/508164/1/specs/queens/approved/symmetric-allocations.rst I assume that with the new microversion the data would be *required* to be in the dict form?  The spec makes it sound optional.15:19
*** dave-mccowan has joined #openstack-nova15:19
*** jmlowe has quit IRC15:19
sdaguemriedem: it's not really clear to me that there is any more feedback there then virtually15:20
*** esberglu has joined #openstack-nova15:20
mriedemsdague: isn't the point of the forum to get the devs and ops and users in the same room?15:20
*** pcaruana has quit IRC15:20
*** priteau has quit IRC15:20
jaypipesstephenfin: done15:21
*** tbachman has joined #openstack-nova15:21
*** lnxnut has joined #openstack-nova15:21
*** felipemonteiro has joined #openstack-nova15:22
melwittsdague: ack, will review that spec15:22
*** felipemonteiro_ has joined #openstack-nova15:23
cdentcfriesen: sorry was on the phone. the idea is that if you want to do it the old way you can always use the old microversion15:23
*** jmlowe has joined #openstack-nova15:23
stephenfinjaypipes: sahid had a similar complaint about "correctly configured host". I'm working on a larger doc a la '/admin/cpu-topologies': do I need to have that done before this merges?15:24
cdentcfriesen: is there a specific place where I could make that more clear?15:24
stephenfinI didn't include it there and there's a million and one steps necessary and I didn't want to bloat that doc, which is a reference-style doc15:24
stephenfin*as there's15:24
openstackgerritAndrey Volkov proposed openstack/nova master: AZ operations: check host has no instances  https://review.openstack.org/50920615:24
*** tbachman has quit IRC15:25
jaypipesstephenfin: I'd be cool with a comment on the commit message pointing to or referencing it15:26
*** yamamoto has joined #openstack-nova15:26
jaypipesstephenfin: be sure to mention the moon cycle and solar flares.15:26
*** lajoskatona has quit IRC15:27
*** felipemonteiro has quit IRC15:27
sdaguemriedem: you get a really small micro slice of the devs and operators that also had a big travel budget and time15:27
*** brault has quit IRC15:27
*** crushil has quit IRC15:28
jaypipessdague: micro slice, huh? is that a new container technology? :)15:28
*** itlinux has joined #openstack-nova15:29
sdaguereally more of a uni-kernel ....15:29
mriedemok, so what's the point in going to the summit at all?15:29
mriedembut we've been here before, i don't mean to digress15:30
gibimriedem: I'm thinking about skipping the notification subteam meeting not to disturb your spec review focus. Is that OK for you?15:31
*** lnxnut has quit IRC15:31
mriedemgibi: yeah15:31
*** itlinux has quit IRC15:31
melwittmriedem: fwiw I know that at yahoo they inject keypairs via cloud-init and have wanted to be able to pass new userdata on a rebuild to update those15:31
sdaguemriedem: because you get higher bandwidth ability to work through hard things15:31
gibimriedem: OK15:31
*** tbachman has joined #openstack-nova15:31
cfriesencdent: review updated15:33
*** yamamoto has quit IRC15:34
*** penick has joined #openstack-nova15:35
*** coreywright has quit IRC15:37
*** crushil has joined #openstack-nova15:39
*** mvk has quit IRC15:40
*** coreywright has joined #openstack-nova15:40
*** user_am has joined #openstack-nova15:41
*** user_am is now known as amodi15:42
*** amodi is now known as archit15:42
*** namnh has quit IRC15:42
*** priteau has joined #openstack-nova15:42
*** namnh has joined #openstack-nova15:43
*** tbachman has quit IRC15:43
*** priteau has quit IRC15:43
*** itlinux has joined #openstack-nova15:44
*** dave-mcc_ has joined #openstack-nova15:44
*** brault has joined #openstack-nova15:45
*** psachin has quit IRC15:45
*** dave-mccowan has quit IRC15:46
*** namnh has quit IRC15:47
jaypipesefried: you had a nice little ascii diagram in one of your review comments of a provider tree... which review/spec was that?15:48
efriedjaypipes This is the one I just did a few minutes ago: https://review.openstack.org/#/c/502306/15:48
*** kuzko has quit IRC15:48
mriedemmelwitt: i'm confused, we don't allow passing in new user_data during rebuild15:49
mriedemoh, "have wanted"15:49
jaypipesefried: donkey shame.15:49
*** tbachman has joined #openstack-nova15:49
efriedjaypipes bitter15:49
mriedemmelwitt: maybe you want to comment on this then, https://review.openstack.org/#/c/509013/ because i said we wouldn't allow psasing user_data during rebuild even if we remove the ability to pass personality files during rebuild15:50
melwittmriedem: yeah, just another data point on how ppl do ssh keys15:50
melwittmriedem: o rly? I thought at the PTG what came out of that discussion was that we'd be allowing user_data during rebuild15:51
*** tbachman_ has joined #openstack-nova15:52
*** archit has quit IRC15:52
*** gbarros has joined #openstack-nova15:53
melwittL441 on https://etherpad.openstack.org/p/nova-ptg-queens15:53
openstackgerritChris Dent proposed openstack/nova-specs master: Add spec for symmetric GET and PUT of allocations  https://review.openstack.org/50816415:53
*** tbachman has quit IRC15:54
*** tbachman_ is now known as tbachman15:54
mriedemmelwitt: i know, see the ML thread i just started15:55
*** baoli_ has quit IRC15:58
*** baoli has joined #openstack-nova16:01
*** lucasagomes is now known as lucas-afk16:01
*** gyee has joined #openstack-nova16:02
openstackgerritEd Leafe proposed openstack/nova-specs master: Return Selection Objects  https://review.openstack.org/49883016:02
*** archit_ has joined #openstack-nova16:05
*** smatzek has quit IRC16:07
sahidmriedem: can you have this in your list for the spec review? https://review.openstack.org/#/c/485522/16:07
*** namnh has joined #openstack-nova16:07
sahidjaypipes: ^ perhaps you can have a look, the code is ready but you might want this to wait for one of your work in-progress16:07
*** smatzek has joined #openstack-nova16:07
jaypipesmriedem, dansmith, bauzas, cdent: OK, I'm good with https://review.openstack.org/#/c/498830/. I say ship it.16:07
mriedemi'm still in keypair update land16:08
jaypipesheh, ok :)16:08
*** yassine has quit IRC16:08
*** hemna__ has quit IRC16:09
*** smatzek_ has joined #openstack-nova16:10
*** smatzek_ has quit IRC16:10
bauzasjaypipes: edleafe: I'm still not sold on the cell_uuid usefulness but meh16:11
jaypipesbauzas: I can see what edleafe was saying about being beneficial to the superconductor.16:11
*** namnh has quit IRC16:12
*** smatzek has quit IRC16:12
openstackgerritJohn Garbutt proposed openstack/nova-specs master: Support traits in the Ironic driver  https://review.openstack.org/50705216:12
bauzasjaypipes: sure, but adding a new field because of that doesn't seem very nice16:12
jaypipesbauzas: I don't think it hurts.16:13
*** dave-mcc_ is now known as dave-mccowan16:13
bauzasif we don't persist it, for sure16:13
bauzasbut between a versioned field and just an object var, I'd tend to prefer an object variable16:14
*** tonygunk has joined #openstack-nova16:14
bauzasif that's just for helping to not lookup16:14
bauzasanyway, an implementation detail16:14
*** yamahata has joined #openstack-nova16:14
mriedemjaypipes: edleafe: confused about something in https://review.openstack.org/#/c/505209/16:15
dansmithjaypipes: edleafe bauzas: yeah, edleafe's explanation makes sense to me.. if we make it an actual CellMapping object them we're sending credentials over the RPC wire, and we don't really need to do that, so just the uuid seems fine to me16:15
mriedemdo we or do we not change GET /allocation_candidates, and if we do, is it just the response body that changes to show the root provider in the response?16:15
bauzasdansmith: if we can avoid a lookup, then okay16:16
jaypipesmriedem: you are correct.16:16
bauzasdansmith: the real problem I have with that is that (host, node, cell) is a single tuple16:17
bauzasI mean, those are interdependent16:17
jaypipesmriedem: well, not even the root provider ID. rather, the parent_provider_uuid will be included in the provider_summaries section and providers that aren't contained in allocation_requests will appear in the provider_summaries (if they are parents of allocated providers)16:17
mriedemjaypipes: ok then i'm going to update the nested rp spec quick to point out that distinction and then i'm +W16:17
bauzashere, we're creating 3 distinct fields but where 2 are related to the third16:17
jaypipesmriedem: ++16:17
bauzasmriedem: well, good point16:17
edleafemriedem: we don't change the API call as we do for GET /resource_providers. Both will have changed response bodies to include root/parent16:18
edleafemriedem: that's noted in the first paragraph of that section16:18
bauzasedleafe: mriedem's point is that it's unclear16:19
*** krtaylor has joined #openstack-nova16:19
*** brault has quit IRC16:19
*** xyang1 has joined #openstack-nova16:19
mriedemedleafe: ok i guess "of appropriate placement REST APIs." is your way of saying GET /resource_providers and GET /allocation_candidates16:19
mriedembauzas: right, it says, "There is no change proposed to `GET /allocation_candidates`" but clearly there is16:20
mriedemso i'll just update to say that the filter parameter won't be added to GET /allocation_candidates16:20
bauzasmriedem: ping me when you're done and I +216:20
edleafemriedem: ok, I can clarify the wording16:20
edleafeoh wait, are you going to update?16:21
bauzasin the mean time, I'm disappearing for dinner16:21
openstackgerritMatt Riedemann proposed openstack/nova-specs master: Re-propose nested resource providers spec  https://review.openstack.org/50520916:21
mriedemedleafe: ^16:21
edleafeheh, question answered16:21
mriedemif that looks ok i'll +W16:21
edleafemriedem: looks like you forgot to clean up L23516:22
openstackgerritStephen Finucane proposed openstack/nova-specs master: Add 'move-nova-cmds-to-cliff' spec  https://review.openstack.org/43360316:22
*** sahid has quit IRC16:22
*** smatzek has joined #openstack-nova16:22
stephenfinmriedem: I just dropped the nova-status bit. We can look at that separately down the line16:23
openstackgerritEd Leafe proposed openstack/nova-specs master: Re-propose nested resource providers spec  https://review.openstack.org/50520916:23
edleafemriedem: fixed it. Guess you need to re-+2 it16:23
mriedemedleafe: done16:23
openstackgerritMatt Riedemann proposed openstack/nova-specs master: Re-propose nested resource providers spec  https://review.openstack.org/50520916:23
*** smatzek_ has joined #openstack-nova16:23
edleafemriedem: well, looks good now16:24
*** jmlowe has quit IRC16:24
ralonsohcdent, jaypipes: I was mixing concepts and terms in https://review.openstack.org/#/c/502306. Thanks for your reviews16:25
jaypipesralonsoh: np16:25
cdentralonsoh: it’s (too) easy to do16:25
ralonsohjaypipes: I always have this in mind: https://youtu.be/LVkknWuGq_I?t=84116:25
*** yamahata has quit IRC16:25
*** smatzek has quit IRC16:27
mriedemedleafe: we don't really need both https://blueprints.launchpad.net/nova/+spec/return-selection-objects and https://blueprints.launchpad.net/nova/+spec/return-alternate-hosts do we?16:27
*** gbarros has quit IRC16:27
mriedemalternate hosts depends on the selection object stuff, and that's just an object model thing16:27
*** smatzek_ has quit IRC16:28
*** lnxnut has joined #openstack-nova16:29
edleafemriedem: the selection object spec only came about because of disagreement over a) whether it is needed at all, and b) what it should look like. It was simply a way to reach consensus.16:29
mriedemok, but all code is going to be written under the return-alternate-hosts bp right?16:29
edleafealternate hosts could work with the unstructured glob16:29
*** smatzek has joined #openstack-nova16:29
mriedemmaybe i just don't know how you're going to split this up16:30
stephenfinjaypipes: Per comments on that doc you reviewed, you should probably look at https://review.openstack.org/#/c/461456/516:30
edleafeit's just the result of switching direction in the middle. Moving forward, let's put all the code under the return-alternate-hosts bp16:31
jaypipesstephenfin: you mean my "clear as mud..." comment?16:31
stephenfinclaudiub: Think this is something you could tackle, given that you are the Hyper-V guy https://bugs.launchpad.net/nova/+bug/166000116:31
openstackLaunchpad bug 1660001 in OpenStack Compute (nova) " Hyper-V PCI Passthrough" [Low,Confirmed]16:31
stephenfinjaypipes: Correct16:31
jaypipesstephenfin: heh, ok, will do. :)16:32
*** sree has joined #openstack-nova16:32
stephenfinfwiw, I get where sahid is coming from but I think it's a helpful enough usability improvement to warrant inclusion16:32
stephenfinjaypipes: Spot on16:32
* stephenfin calls it for the evening16:32
*** smatzek has quit IRC16:33
mriedemoh btw who is going to be brave and just push this through? https://review.openstack.org/#/c/457532/16:35
*** smatzek has joined #openstack-nova16:35
*** lnxnut has quit IRC16:36
*** tesseract has quit IRC16:37
*** kenperkins has joined #openstack-nova16:38
*** smatzek has quit IRC16:40
claudiubstephenfin: docimpact, huh. hm, adding details about how to configure assignable PCI devices on Hyper-V in doc/source/admin/pci-passthrough.rst should be enough, IMO. any other thing is the same as other drivers.16:41
*** peter-hamilton has joined #openstack-nova16:41
claudiubstephenfin: will send a patch today / tomorrow16:41
jaypipesmriedem: done.16:44
*** smatzek has joined #openstack-nova16:44
*** smatzek has quit IRC16:46
cfriesenmriedem: I've reviewed https://review.openstack.org/#/c/381912 (Strict isolation of group of hosts for image and flavor) as promised.16:46
*** abalutoiu has joined #openstack-nova16:48
openstackgerritMerged openstack/osc-placement master: CLI for resource providers  https://review.openstack.org/45753216:50
*** smatzek has joined #openstack-nova16:51
*** slaweq_ has quit IRC16:51
*** priteau has joined #openstack-nova16:52
*** smatzek has quit IRC16:52
jaypipesholy shit, the osc thing merged.16:52
cdentlimited tess16:52
openstackgerritChris Dent proposed openstack/nova-specs master: Spec for limiting GET /allocation_candidates  https://review.openstack.org/50454016:52
cfriesenmriedem: I had an alternate proposal...rather than "strict" flags on the flavor/image (which would affect all the other image properties/flavor extra-specs) I think it'd make more sense to have a "strict" prefix on the metadata key in the host aggregate.16:53
cfriesenthat way you could be strict on some keys but not on others16:53
*** priteau has quit IRC16:54
efriedcdent Nits/questions https://review.openstack.org/#/c/508164/16:54
cdentroger con aye16:54
*** smatzek has joined #openstack-nova16:55
*** smatzek has quit IRC16:55
*** derekh has quit IRC16:55
*** smatzek has joined #openstack-nova16:56
openstackgerritMatt Riedemann proposed openstack/nova-specs master: List/show all server migration types  https://review.openstack.org/48902916:56
*** esberglu has quit IRC16:58
*** sree has quit IRC16:58
*** yamahata has joined #openstack-nova16:58
*** esberglu has joined #openstack-nova16:58
cfriesenI just noticed something really weird...can someone confirm?  It looks like AggregateInstanceExtraSpecsFilter will *not* match if a flavor specifies something and a host is not in an aggregate or the aggregate metadata doesn't have what the flavor is looking for.17:00
*** sree has joined #openstack-nova17:00
cfriesenBut it looks like AggregateImagePropertiesIsolation *will* match if the image specifies something that isn't in the aggregate metadata17:00
cfriesenAggregateInstanceExtraSpecsFilter loops over the flavor extra-specs looking for a match, but AggregateImagePropertiesIsolation loops over the aggregate metadata looking for a match.17:01
*** crushil_ has joined #openstack-nova17:02
*** exarr has joined #openstack-nova17:02
*** baoli has quit IRC17:02
exarrAnyone around I can ask about rabbit connection problems?17:02
openstackgerritMerged openstack/nova-specs master: Return Alternate Hosts  https://review.openstack.org/50427517:02
exarrInvalid credentials it says. So I readd the user, change password, check the transtport_url details, all seems correct.17:02
exarrrabbit logs say "AMQPLAIN login refused: user 'openstack' - invalid credentials"17:02
exarrSeems clear cut, huh?17:03
*** baoli has joined #openstack-nova17:03
*** ralonsoh has quit IRC17:03
*** edand has joined #openstack-nova17:04
*** nikhil_k has quit IRC17:04
*** kuzko has joined #openstack-nova17:04
*** ijw has joined #openstack-nova17:05
*** liangy has joined #openstack-nova17:06
*** gouthamr has joined #openstack-nova17:06
*** sree has quit IRC17:07
*** lyan has quit IRC17:07
*** sree has joined #openstack-nova17:07
*** namnh has joined #openstack-nova17:08
*** jpena is now known as jpena|off17:10
*** sree has quit IRC17:11
*** hemna__ has joined #openstack-nova17:12
*** namnh has quit IRC17:12
dansmithmriedem: stephenfin: so what is the deal on this nova-manage spec? we're going to just dump syntax compatibility across one release?17:13
*** Swami has joined #openstack-nova17:13
mriedemi wondered about that too, since it said it was unavoidable when moving to cliff17:14
mriedemwhich is kind of a non-starter for me17:14
dansmithseems kinda bad to me17:14
mriedemyou can't break everyone's tooling17:14
mriedemjust because we don't like our under the hood impl17:14
dansmithright, that's my concern17:14
mriedemwell, unless it's cells v1 :)17:14
mriedemthen break away!17:14
*** ociuhandu has quit IRC17:15
*** kukacz has quit IRC17:20
sdaguethe flagging structure should be the same more or less17:20
sdagueI guess the devil in the details there, but I wasn't imaging a big shift17:20
openstackgerritLee Yarwood proposed openstack/nova-specs master: Libvirt: Native LUKS decryption by QEMU  https://review.openstack.org/49082417:21
*** baoli has quit IRC17:21
melwittI'd think we'd need a deprecation cycle for changing nova-manage syntax. is that clock already in motion?17:21
*** baoli has joined #openstack-nova17:22
mriedemsdague: cdent: this needs some more thought https://review.openstack.org/#/c/507762/17:22
mriedemplus, it came up in newton17:22
mriedemand there are some issues to overcome17:22
clarkbas a drive by comment, the use of stevedore (via cliff aiui) in osc is what makes it so painfully slow to use for commands17:22
clarkbso please don't use entrypoints for cli commands17:23
mriedemdansmith: you too on https://review.openstack.org/#/c/507762/ because it would require paging instance actions across cells...17:23
melwittgood to know17:23
mriedemwhich we know is fun17:23
mriedemmelwitt: there were some deprecations made in pike17:23
mriedemto prep for this in queens17:23
cdentah, fart, forgot about cross cell biz17:23
mriedemit's not only that - it's that we literally don't update the updated_at column for intsance actions17:24
mriedemso filtering on changes-since for that is kind of dumb17:24
*** efried has quit IRC17:24
dansmithmriedem: ack will look17:24
sdagueclarkb: it's not going to be entry points17:24
mriedemlike i said, it's come up before17:24
*** slaweq_ has joined #openstack-nova17:24
cdentmriedem, why is that not a bug? (as in, to fix)17:24
sdagueclarkb: it's just going to be non custom cli parsing17:24
sdagueclarkb: there is nothing about stevedore here17:24
mriedemKevin_Zheng is on holiday this week but i'll also bug him on the wechat-o-sphere17:24
dansmithmriedem: instance action list is per instance, right? so no paging across cells?17:25
clarkbsdague: I thought cliff had some baked in ideas of registering commands via stevedore as part of its command parsing17:26
clarkbsdague: but maybe its optional17:26
*** kukacz has joined #openstack-nova17:27
cdentmriedem: “ it's that we literally don't update the updated_at column for intsance actions” <- Is there a reason why for that? Isn’t that what updated_at means?17:27
melwittI didn't think we updated individual instance action records. aren't they just written once?17:28
*** slaweq_ has quit IRC17:28
*** slaweq_ has joined #openstack-nova17:28
melwittlike, "instance created" "instance rebooted" "instance rebuilt" it's not like you go back and edit those17:28
*** gjayavelu has joined #openstack-nova17:29
*** crushil_ has quit IRC17:29
dansmithupdated_at is built into oslo.db17:30
dansmithso if it's never set on those, it's because we never did a save, AFAIK17:30
dansmiththere are other things missing in that spec,17:30
dansmithlike it doesn't show them actually using the marker they say they're going to,17:30
*** hferenc has quit IRC17:30
*** slaweq_ has quit IRC17:30
*** gabor_antal has quit IRC17:30
*** gabor_antal_ has joined #openstack-nova17:31
dansmithbut I assume they really mean start_time, which is its own field17:31
dansmithseparate from the usual three default ones17:31
*** hferenc has joined #openstack-nova17:31
dansmiththere is a finish_time on that object too17:31
melwittyeah, I was trying to say I don't think instance action rows are ever saved, they're created/written once and that's it17:31
sdaguemelwitt: how is end time set then?17:31
sdagueor, only if it's set all at once17:32
dansmithwe can update them actually17:32
dansmithsets the finish time and calls action.update()17:32
melwittoh, weird17:32
dansmithI'd bet we don't finish all the actions we start though.. like everything else:P17:33
*** lnxnut has joined #openstack-nova17:34
*** crushil has quit IRC17:35
*** ijw has quit IRC17:36
mriedemi'd have to read back through the specs, but we create the action record and then we just deal with the events17:38
mriedemwhich are different records17:38
mriedemso there is event_start and event_finish17:39
mriedemand i don't see us ever updating the action record when the event is finished17:40
*** itlinux has quit IRC17:41
*** ociuhandu has joined #openstack-nova17:41
dansmithmriedem: https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L622417:41
dansmithohh, the action and event I'm getting confused I guess17:41
dansmithwe actually just return the action from the action_finish() thing17:42
dansmithnot sure why or how that makes sense17:42
dansmithbut I guess that means we're never updating the _action_17:42
*** itlinux has joined #openstack-nova17:42
dansmithit doesn't matter though, because changes-since should key on the start_time I would think17:42
*** crushil has joined #openstack-nova17:42
dansmithor it could be on max(start_time, finish_time) in case we ever set finish_time17:43
dansmithbut not updated_at I wouldn't think17:43
*** lnxnut has quit IRC17:43
melwittI think the question was, why isn't updated_at updated and I was saying because it's never updated17:43
cdenttautology alert17:44
cdentbut: yes17:44
melwitthaha yeah17:44
dansmithmelwitt: right, I get that, and it's because we're not updating it from action_finish17:44
dansmithmelwitt: I had found action_event_finish() when looking for action_finish() which _does_17:44
dansmithI dunno _why_ we're not updating it during the finish, but..17:44
*** jmlowe has joined #openstack-nova17:45
melwittwell, because each action is just "instance create started" "instance create finished" and they're separate right?17:45
dansmithregardless, I would expect changes-since on this kind of thing to use the inbuilt fields17:45
dansmithnot the default one17:45
melwittthere's nothing to update about it, it's just there17:45
dansmithmelwitt: except we have a finish method that doesn't finish it, and a finish_time field we never set, apparently17:45
mriedemdansmith: i don't think anything actually calls action_event_finish()17:45
mriedemi remember talking to laski about that at one point17:46
mriedemand it was like, something something tasks...we never used it17:46
dansmithfine, extend my argument to why we have those in addition to why they do nothing :)17:46
mriedemthe EventReporter context manager calls objects.InstanceActionEvent.event_finish_with_failure17:46
dansmithit's beside the point though, IMHO17:46
melwittI think I don't know what instance action events are. I'm just thinking of the "instance actions" which are a basic log of things that have happened to the instance17:47
*** sambetts is now known as sambetts|afk17:47
*** sridharg has quit IRC17:47
mriedemthe events are tied to the action17:47
dansmithright, an action is a high level thing,17:48
mriedemsee the EventReporter class17:48
dansmithevents are small steps17:48
*** itlinux has quit IRC17:48
mriedemever notice the @wrap_instance_event decorator on the compute manager methods?17:48
mriedemthat creates the event start on entry, and event finish on exit17:48
mriedemthe action record is created in the API before casting to compute where the real business happens, and that real bidness is recorded with events17:48
mriedemon the action17:48
melwittokay, so I guess someone forgot to ever finish_time instance actions after creating the data model17:51
mriedemjesus i thought this was merged in pike https://review.openstack.org/#/c/480746/17:51
mriedemso ^ gives an example of how a user can use these things for polling when an operation is complete17:51
melwittthat's leet17:52
*** itlinux has joined #openstack-nova17:54
mriedemalso, note that if we ever rename methods in the compute manager which have the @wrap_instance_event decorator, we break that api17:55
dansmithwhich we have, I think17:55
melwittlike elite use of that API17:55
dansmithhasn't gibi complained?17:55
mriedemdon't remember17:55
dansmithI think when we moved things from compute to conductor there was a thing about that17:55
dansmithI don't recall if there was an override to fix it or just "meh, this is an internal api"17:56
mriedemfor events or notifications?17:56
*** bnemec has quit IRC17:56
dansmithoh yeah I guess I'm thinking of notifications17:56
dansmithbut similar deal17:56
*** itlinux has quit IRC17:58
*** bnemec has joined #openstack-nova18:01
*** dave-mccowan has quit IRC18:01
mriedemheh coincidentally https://bugs.launchpad.net/nova/+bug/171956118:01
openstackLaunchpad bug 1719561 in OpenStack Compute (nova) "Instance action's updated_at doesn't updated when action created or action event updated." [Undecided,In progress] - Assigned to Yikun Jiang (yikunkero)18:01
*** slaweq_ has joined #openstack-nova18:01
mriedemi guess they found this out on their own...18:01
*** dave-mccowan has joined #openstack-nova18:03
*** baoli has quit IRC18:06
mriedemhuh we have no indexes on the events table18:06
*** hferenc has quit IRC18:06
*** gabor_antal_ has quit IRC18:06
*** baoli has joined #openstack-nova18:06
*** hferenc has joined #openstack-nova18:07
*** jmlowe has quit IRC18:07
*** gabor_antal has joined #openstack-nova18:07
*** jmlowe has joined #openstack-nova18:09
*** dave-mccowan has quit IRC18:10
mriedemmight be time to fire up the ol' create 1000 instances and list some things out of the api machine18:10
cdentmriedem: one of the last times you did that there were 409s coming out of placement, did that ever get narrowed down, or did the rest of the world come rushing in and flush that for a while?18:11
*** ijw has joined #openstack-nova18:12
*** dave-mccowan has joined #openstack-nova18:12
mriedemi've got a patch18:12
mriedembut zuul has been zuuling it up18:12
cdentthat’s the recreate patch, right, not the “here’s my theory” patch?18:13
mriedemit's a recreate yeah18:14
mriedemto see if we can find out from the logs where the 409 comes from18:14
mriedemwhat triggers it i mean18:14
cdentcool, I just wanted to be sure I hadn’t missed a revelation18:15
*** gabor_antal has quit IRC18:15
*** gabor_antal has joined #openstack-nova18:16
*** ijw has quit IRC18:16
*** jmlowe has quit IRC18:17
*** jmlowe has joined #openstack-nova18:17
mriedemcdent: if you scroll down to the bottom http://logs.openstack.org/18/507918/4/check/legacy-tempest-dsvm-neutron-full/8f969e2/logs/devstacklog.txt.gz you'll see what's going on18:18
mriedemmy check is failing, because i've apparently failed at bash18:19
mriedemsdague: maybe you know how to bashtastify this ^ https://review.openstack.org/#/c/507918/5/stack.sh18:19
cdentbash failing is an easy thing to do18:20
*** esberglu has quit IRC18:20
*** esberglu has joined #openstack-nova18:21
*** markmcclain has quit IRC18:21
*** itlinux has joined #openstack-nova18:21
*** esberglu has quit IRC18:21
sdaguemriedem: I can look after, I have a call think shortly18:21
*** esberglu has joined #openstack-nova18:22
mriedemit could just be we didn't actually fail, or i timed out too early18:22
*** priteau has joined #openstack-nova18:22
*** markmcclain has joined #openstack-nova18:23
mriedemthis is where it starts in the scheduler http://logs.openstack.org/18/507918/4/check/legacy-tempest-dsvm-neutron-full/8f969e2/logs/screen-n-sch.txt.gz#_Sep_28_20_12_08_48624918:24
mriedemlook at all of those beautiful uuids18:24
mriedemi seem to have also totally f'ed up the logging of the request id18:25
mriedembecause the req-47868c4f-61d3-4c32-9ece-94fb6ad35404 used for the instance create it also being logged for running periodic tasks...18:26
mriedemcdent: heh, yeah, i think it actually scheduled the 500 instances18:27
cdentyou want it to break, no break. want it to work, break.18:28
mriedem"Starting instance..." shows up 500 times in n-cpu18:28
mriedemwell, i could bump it to 1000 instances18:28
* cdent raises pinkie to mouth18:28
mriedemok bombs away with 100018:30
mriedemso there is no unique constraint on instance_actions.request_id,18:32
mriedembut we key off it in the API https://developer.openstack.org/api-ref/compute/#show-server-action-details18:33
mriedemso there should probably be a unique constraint on that column yes?18:33
*** mingyu has quit IRC18:33
mriedemcomment in the code even says,18:34
mriedem"The intention is that there will only be one of these per user request.  A18:34
mriedem    lookup by (instance_uuid, request_id) should always return a single result."18:34
mriedembut don't bother enforcing that in the table schema...18:34
exarrAnyone around I can ask about rabbit connection problems? :-(18:35
exarrInvalid credentials it says. So I readd the user, change password, check the transtport_url details, all seems correct.18:35
exarrrabbit logs say "AMQPLAIN login refused: user 'openstack' - invalid credentials"18:35
exarrSeems clear cut, huh?18:35
*** baoli has quit IRC18:35
*** slaweq_ has quit IRC18:35
*** slaweq_ has joined #openstack-nova18:35
*** baoli has joined #openstack-nova18:37
mriedemexarr: are you using ocata+?18:40
*** lnxnut has joined #openstack-nova18:41
mriedemif so, you have to make sure the transport_url in your cell mappings records match18:41
*** tbachman has quit IRC18:41
*** slaweq_ has quit IRC18:42
*** sean-k-mooney has quit IRC18:44
openstackgerritEric Berglund proposed openstack/nova master: WIP(5): PowerVM driver: ovs vif  https://review.openstack.org/42251218:45
*** priteau has quit IRC18:47
*** smatzek has quit IRC18:48
*** smatzek has joined #openstack-nova18:50
*** smatzek has quit IRC18:50
*** lnxnut has quit IRC18:50
*** smatzek has joined #openstack-nova18:52
exarrmriedem: Ocata, yes.18:56
exarrJust updated to latest today18:56
*** mvk has joined #openstack-nova18:56
exarrtransport_url in the cell mapping? oooh. Let me have a look, thanks :-)18:56
mriedemnova-manage cell_v2 list_cells18:59
mriedemexarr: if you had to change the transport_url in config, the cell mappings are probably stale18:59
mriedemand that's what's being used when switching rpc context at runtime18:59
mriedemyou can use the nova-manage cell_v2 update_cell command to update the transport_url for a given cell if needed19:00
mriedemesberglu: i've left some comments in https://review.openstack.org/#/c/503061/ but you can address those in a follow up if you want, or don't, whatever19:00
*** peter-hamilton has quit IRC19:04
openstackgerritMerged openstack/nova-specs master: PowerVM Driver Integration (Queens)  https://review.openstack.org/50306119:05
exarrmriedem: hmmm. I think I have updated as you suggested. However I am still encountering the same problem.19:05
exarrIt's very much as though the credentials are simply incorrect, or rabbit is not allowing the authentication ptotocol.19:07
exarrCan I check the cell mapping details somehow? :-/19:07
mriedemnova-manage cell_v2 list_cells --verbose19:08
mriedemwill dump the cells and their transport_url as they are stored in the db19:08
mriedemkeep in mind that if you update the cell's transport_url, you'll have to restart some services because those values are cached in memory19:08
*** smatzek has quit IRC19:08
exarrmriedem: ahhh. I'll restart the machine, that should help.19:08
exarrthanks for helping a noob.19:09
esberglumriedem: I can fix it quick in that review. Unless you guys don't want to have to re-review, you seem busy atm19:10
*** smatzek has joined #openstack-nova19:10
esbergluIn that case I'll take care of it later19:10
mriedemexarr: also fyi https://docs.openstack.org/nova/pike/user/cells.html#faqs19:10
*** smatzek has quit IRC19:10
mriedemsee #219:10
exarrmriedem: I could bloody kiss you.19:11
dansmithew, a bloody kiss?19:11
dansmithsounds terrible.19:11
mriedemonly on goth thursdays please19:11
*** slaweq_ has joined #openstack-nova19:12
*** MasterOfBugs has joined #openstack-nova19:13
*** smatzek has joined #openstack-nova19:15
*** hemna__ has quit IRC19:22
*** slaweq_ has quit IRC19:22
*** slaweq_ has joined #openstack-nova19:22
openstackgerritChris Dent proposed openstack/nova-specs master: Add spec for symmetric GET and PUT of allocations  https://review.openstack.org/50816419:23
penickAnd now I have an image of mriedem dressed up as The Crow.19:28
cdentpenick: my eyes!19:29
penickYou're welcome.19:29
*** READ10 has quit IRC19:30
*** ijw_ has joined #openstack-nova19:30
openstackgerritChris Dent proposed openstack/nova-specs master: Add spec for symmetric GET and PUT of allocations  https://review.openstack.org/50816419:32
*** catintheroof has quit IRC19:32
*** vvargaszte has joined #openstack-nova19:32
*** catintheroof has joined #openstack-nova19:32
melwittI wish I knew why we didn't already have disk quota in nova19:33
penickI thought it used to be there, but left when cinder split out19:34
openstackgerritChris Dent proposed openstack/nova master: [placement] gabbi tests for shared custom resource class  https://review.openstack.org/48520919:34
melwittlet me check the archived scrolls19:35
penickI think it was in folsom, but got 'et by the Grizzly19:35
cdentI’m now picturing melwitt (in the boots) sitting in dark loft with mriedem as crow “I wish I knew why…”19:35
cdentmriedem gets up slowly from his crushed velvet chair19:36
cdentfist clenched he walks to the broken window19:36
cdentand shouts into the city19:36
penickI just assumed that was how his home office was configured19:37
sdagueI really don't think it was ever implemented19:37
penickmriedem is an enigma wrapped in a tortilla19:37
sdagueI thought I ran this code stream a couple months ago19:37
sdaguebecause quota only matters for the things you have the least of19:38
sdagueusing local disk that's just there doing nothing on computers was never the limiting factor for rax19:38
melwittyeah, so far not finding anything in the folsom-eol tag19:38
mriedemdisk quota also gets goofed up when you're talking about boot from volume19:38
* melwitt slowly backs away19:39
sdaguemriedem: right, there is quota on volumes, because that's off the expensive storage units19:39
mriedemis it that time of the week to link in that change again?19:39
cdentthe words dripped from mriedem’s black lips19:39
*** vvargaszte has quit IRC19:39
melwittso not only will ppl be complaining about consuming > 0 disk with BFV they'll also complain about consuming > 0 quota with BFV19:40
sdagueyeh, well landing that would clearly be a prereq19:40
sdagueI guess disk quota only really makes sense when your instances are all on network storage like ceph or nfs, right?19:41
melwittI have at least two patches that are like "the patches that shall not be named"19:41
sdaguewhich basically didn't work back then anyway19:41
melwittthat one and then there's the thrice reverted bug fix one19:41
*** sapcc-bot3 has quit IRC19:42
*** sapcc-bot has joined #openstack-nova19:42
*** efried has joined #openstack-nova19:43
mriedemheh https://review.openstack.org/#/c/218639/ still around19:43
melwittsdague: hm, yeah ... the spec says they want to be able to bill for storage and restrict storage and currently the only way to do that is with cinder volumes19:43
mriedemwhen is the last time someone ran the auto-abandon script19:43
sdaguemriedem: it's been a while19:44
mriedemseveral other bugs like this: create instance, attach volume, snapshot instance (the snapshot image has bdms in it now), create another instance, rebuild that 2nd instance with the snapshot image with image-defined-bdms, kablammo19:45
sdaguemelwitt: right, I guess if you have flavors with a wide range of disk storage sizes, vs. coupling them to the mem/cpu sizing19:45
*** gszasz has quit IRC19:46
sdaguemriedem: so... I feel like there were enough interesting side conversations about the key_pair thing that they probably warrent a summary email, they will not capture very cleanly in 2 specs and 4 irc threads19:47
sdagueI can write that up if you like19:47
melwittyeah, must be. I haven't really heard this come up before, so it's probably like you said, for most ppl the limiting factor is cpu/mem19:47
*** lnxnut has joined #openstack-nova19:48
mriedemsdague: that's probably a good idea19:48
*** bnemec has quit IRC19:49
*** edand has quit IRC19:52
mriedemis any of the accessIPv4 still valid?19:53
mriedem*accessIPv4 stuff19:53
mriedemthat seems to be another one of those legacy things which we don't test but it's all over the api19:53
mriedemlike diskConfig19:53
sdagueit's user facing19:54
sdagueand user updatable19:54
sdaguesometimes it gets filled out, depending on the config19:54
sdagueif you update it yourself, we don't validate it, it's just reflected19:55
sdaguemriedem: ok, because it's been a while, a rebuild doesn't go through the scheduler, right?19:55
*** archit_ is now known as archit19:55
*** edand has joined #openstack-nova19:55
sdagueyou get to rebuild in place, which means it's probably faster because your base image is already tehre19:55
mriedemsdague: rebuild (not evacuate) bypasses the scheduler yes19:56
sdagueI'm trying to make sure I cover all the reasons this could be interesting19:56
*** jangutter has quit IRC19:58
*** jangutter has joined #openstack-nova19:58
*** lnxnut has quit IRC19:59
sdaguealso, out of curiosity, because of the user scoping of keys, if you rebuild someone else's instance in the same project today, does it actually scope check the key and not add it?20:00
penickthat's an interesting question20:02
mriedemsdague: you mean in the proposed spec?20:03
mriedemor today?20:03
mriedembecause you can't change the key on rebuild today, hence the spec20:04
sdaguemriedem: no20:04
sdagueI mean, you build a server with your key20:05
sdagueI'm in your project20:05
sdagueI rebuild your server20:05
sdaguebut I'm not you20:05
sdagueand I don't have access to your key20:05
sdaguewhat happens20:05
sdaguedo we "never check that scope"20:06
sdaguedo we "check the scope and not add the key"20:06
mriedemjust because you rebuild my server doesn't mean you get access to my key to ssh into it, right?20:06
mriedemso i'm not seeing the issue20:07
*** liverpooler has quit IRC20:08
openstackgerritMatt Riedemann proposed openstack/nova master: api-ref: remove redundant preserve_ephemeral mention from rebuild docs  https://review.openstack.org/50927320:09
penickmriedem but isn't there a spec to add userdata to rebuild?20:10
penickOr maybe I took crazy pills20:10
mriedemthere is a spec to remove personality files from rebuild,20:10
melwittpenick: they're talking about the nova keypair API.20:10
*** pchavva has quit IRC20:10
*** catintheroof has quit IRC20:10
mriedemwe're really talking about like 5 things20:11
*** catintheroof has joined #openstack-nova20:11
mriedemthere is a spec proposed to pass a new key_name during rebuild,20:11
mriedemthere is a spec proposed to remove peronality files from server create and rebuild20:11
cfriesenjaypipes: thanks for the suggestions on https://review.openstack.org/#/c/46820 (proposing  "hw:realtime_cpu_set").  I imagine we'd need to support both "set" and "mask" for a release or two?20:11
mriedemthe question came up if we should allow passing new user_data during rebuild if we're not going to allow passing personality files anymore, and the spec currently states that we will not allow that20:11
penickYeah, my concern was if I can rebuild your instance, and that rebuild pulls in your user ssh keys, and If I can do something to inject my user into your instance on rebuild then i'd gain access to your ssh keys.20:12
melwittpenick: if two ppl are in the same project and one person A creates an instance with their --key_name in it and the second person B does a rebuild on person A's instance, what happens to the key20:12
cfriesenpenick: resources are owned by projects, not individual users20:12
mriedemcfriesen: except keys20:12
mriedempenick: i think your question belongs in https://review.openstack.org/#/c/375221/20:13
mriedemin the security impact section20:13
*** smatzek has quit IRC20:13
cfriesenpenick: mriedem: logically though any keys should belong to the project, not the individual user20:13
mriedemi think what would happen here is if you specified a new key_name during rebuild, we are going to be looking up that keypair in the api to see if it exists,20:14
mriedemthat check is going to use the context.user_id,20:14
*** smatzek has joined #openstack-nova20:14
mriedemand if you're rebuilding my server, and don't have access to my keys, that keypair lookup will fail and you'll get an error20:14
mriedemnow if user A and user B are in the same project, and both have a key named mykey, that could work, but the user A mykey and user B mykey should still be different20:14
penickmelwitt: yeah that's what i'm wondering20:15
cfriesenmriedem: why did we ever tie keys to users?20:15
*** smatzek_ has joined #openstack-nova20:16
mriedemyou will have to ask someone else20:16
mriedembut, why wouldn't we?20:16
mriedemkeys are private20:16
penickcfriesen: it makes sense for user ssh keys. Just not for headless keys or server keys20:16
melwittprobably bc when you create one via the API it gives you the private key20:16
cfriesenmriedem: instances are owned by the project.  keys are used to access instances.  therefore keys should be owned by the project20:17
mriedemare we venturing into the application credentials hullabaloo?20:17
sdagueit is only the public keys that we store20:17
sdagueso it's not really a security risk20:17
sdaguebut it's funny that you can't list those keys for other users20:17
sdaguebut they can be put on a server for you20:18
cfriesenmelwitt: sure, then you dump the private key into some local "project team storage area" with permission controls20:18
*** smatzek has quit IRC20:18
dansmithdoes this matter at all? we'd have to do some pretty major surgery to change the ownership of keys, and it would be hard to support old microversions for them20:18
sdaguecfriesen: keys attached to users are from "the before time"20:18
dansmithunless we're really going to change key ownership, we might as well talk about problems we're going to solve20:19
sdaguedansmith: yeh, it's just an interesting edge question about how the system works because of the scope mismatch20:20
*** smatzek_ has quit IRC20:20
dansmithbut then we can move on once identified right? :)20:20
*** david-lyle has quit IRC20:20
sdagueanyway, I tried to build a summary email from all the conversations I saw today20:21
cfriesendansmith: fair enough...so in mriedem's scenario would we allow user B to replace user A's "mykey" with user B's "mykey" on a rebuild?20:21
*** artom has quit IRC20:21
*** david-lyle has joined #openstack-nova20:21
dansmithIMHO, if you specify a key, it's looked up based on your context, and if that's your key instead of the original one, then so be it20:21
sdaguedansmith: sure, I just think it might be an unexpected thing20:21
dansmithif you don't, then we should keep the same I think20:21
*** itlinux has quit IRC20:22
dansmithsdague: if you specify a key? you can't list other people's keys so what other intent could the user have if they specify one?20:22
penickWell for preserving the host key, the private key will need to be stored as well20:22
*** edand has quit IRC20:22
sdaguepenick: we don't do that20:22
penickthere's no point in keeping a public ssh host key, if you don't preserve the private key20:22
dansmithpenick: we do nothing with the host key20:22
dansmiththese are all user access keys we're talking about20:22
penickOh hell, I completely misunderstood this spec.20:22
*** corey_ has quit IRC20:22
sdaguethis is all just the root ssh key20:22
sdagueroot user20:23
dansmithnot even root,20:23
sdaguebeing configed on cloud init20:23
dansmithwhatever user the image has for access20:23
sdaguedansmith: yeh, right20:23
mriedemcfriesen: ok, so i think what would probably need to happen for rebuild, is if another user does the rebuild, then we have to lookup the keypair by the instance.user_id,20:24
mriedemnot the context.user_id20:24
dansmithand doesn't specify a key.. agreed20:24
dansmithwe just have to be careful not to expose non-project-user keys when we're doing that override,20:25
dansmithsince it happens deep in the db layer I think20:25
sdagueyeh, I just wonder what happens in the current code20:25
dansmithsdague: we fail I think20:25
*** david-lyle has quit IRC20:25
penickahhh ok. So not a security issue then. Usually people want to preserve the ssh host key between reimages. I do think this is an operability issue if keys stay scoped to only a user, instead of a tenant.20:25
mriedemyou can't specify a new key on rebuild today, so i'm lost as to what the concerns are about the current code20:25
*** kenperkins has quit IRC20:25
dansmithI thought we break if another user tries to rebuild today, no?20:26
sdaguedansmith: I don't know if we do, that was the question20:26
sdaguerebuild is an instance operation that's project scoped20:26
sdagueand keys are user scoped20:26
sdagueand I actually haven't checked if B rebuilds a server that A built with A's key if it: a) still has that key, b) has no key, c) errors out20:27
sdagueI'll do some testing around that in the morning, my end of day is now20:27
dansmithsdague: right, I thought it was an oopsie on our part that other users couldn't rebuild successfully20:28
mriedemon a rebuild today, we'll use the original key associated with the instance when it was created20:28
mriedemregardless of who rebuilds the isntance20:28
dansmithpeople want a key during rebuild for taking ownership I think, but that's a destructive operation and not a good argument, IMHO20:28
sdaguemriedem: ok, cool20:28
dansmithmriedem: okay, i thought there was a whole big stink about us not being able to look up the key again on rebuild if triggered by another user20:28
sdaguedansmith: no, it was a question because of the scoping difference.20:29
mriedemas far as i know, TODAY, we don't ever look up the key during rebuild,20:29
*** cdent has quit IRC20:29
mriedembecause you can't change it20:29
mriedemthe only thing i think that happens today,20:30
sdaguemriedem: well, it has to be looked up to go in the config drive, right?20:30
mriedemis when driver.spawn happens, we rebuild config drive, and that's going to call the InstanceMetadata code to get the keypair20:30
mriedemto shove into the config drive20:30
mriedemsdague: yes20:30
mriedemand that is the keypair that's stashed in the instance20:31
mriedemjust like a flavor20:31
dansmithbut we store that on the instance20:31
*** eharney has quit IRC20:31
dansmithmaybe this was broken before and now isn't20:31
sdaguewe store keypair name20:31
dansmithsdague: no we store the whole thing20:31
*** dave-mccowan has quit IRC20:31
mriedemwe store the whole gd thing20:31
dansmithsdague: keypairs live in the api db and compute can't get them20:31
sdagueok, so yeh, maybe cells v2 did move this around and fix a thing20:31
dansmithum, ya'll'er welcome?20:32
*** itlinux has joined #openstack-nova20:32
sdaguedansmith: so, the use case that people seem to want is because keeping IP and device model (mac address) is useful to folks20:32
dansmithbut changing ownership via key, yeah20:33
mriedemyou also keep your volumes attached20:33
sdaguemriedem: right20:33
dansmithI get why people want that20:33
dansmithI wish they didn't want it, but..20:33
sdagueI'd still like a more detailed set of use cases in the spec.20:34
*** jmlowe has quit IRC20:34
openstackgerritMatt Riedemann proposed openstack/nova master: api-ref: add note about rebuild not replacing volume-backed root disk  https://review.openstack.org/50928220:36
mriedem^ is my answer to the buttload of duplicate bugs20:37
*** kenperkins has joined #openstack-nova20:40
*** catintheroof has quit IRC20:42
*** catintheroof has joined #openstack-nova20:42
openstackgerritMatt Riedemann proposed openstack/python-novaclient stable/newton: Fix aggregate_update name and availability_zone clash  https://review.openstack.org/50781620:43
sdaguemriedem: so rebuild on bfv is just a reboot?20:44
sdagueit might be better to actually make that a 40020:44
sdaguebecause it's not actually doing what the user expects20:44
dansmithsdague: but evacuate (rebuild) on BFV is hiiiiiiighly utilized20:45
*** archit has quit IRC20:45
dansmithwhich is a little different, granted, but..20:45
mriedemcan't specify a new image on evacuate20:45
dansmithrebuild on bfv with a key becomes more useful20:45
dansmithright I know20:45
mriedemthe issue here is you specify a different image on rebuild20:45
cfriesensdague: you can specify a different personality on rebuild20:45
dansmithah, if you specify an image sure20:45
mriedemi really need to just get a devstack setup and play with some of this a bit, because a lot of the bug reports are really old and confusing, and mixing issues20:45
sdaguemriedem: yeh20:46
sdagueI was thinking about doing that tomorrow morning20:46
mriedemthe linked bug in there shows a recreate where the rebuild doesn't fail but doesn't change the root disk either20:46
sdagueand just mapping out the space a bit20:46
mriedemit does change the instance image ref20:46
cfriesenmriedem: I think we hit that one....was sort of confusing due to the mismatch20:46
mriedembut, that's probably because we change that in the api20:46
mriedemi bet the rebuild actually fails on the compute20:47
mriedembecause we can't detach the root disk20:47
mriedemalso, unrelated, i just updated an approved change and removed something in the commit message, and it re-applied the +W on the patch20:49
*** flanders_ has joined #openstack-nova20:49
mriedemwhich seems like odd new behavior20:49
sdagueif it's litterally exactly the same patch, I though all the votes come back20:50
sdaguebecause you +W PS120:51
sdagueand 6 is the same as 1 bit for bit, the votes pop back20:51
melwittI had thought commit messages counted as part of the patch in the past20:51
sdaguemessage contents20:51
sdaguenot metadata20:51
sdaguethe commit message is also the same20:52
melwittmriedem changed the commit message in PS620:52
melwittoh, you're saying it's the same as PS120:53
melwittI see now20:53
*** smatzek has joined #openstack-nova20:55
*** smatzek has quit IRC20:56
*** smatzek has joined #openstack-nova20:57
*** itlinux has quit IRC20:57
openstackgerritMerged openstack/nova-specs master: Libvirt: Native LUKS decryption by QEMU  https://review.openstack.org/49082420:59
melwittI put up a spec for counting instances, CPU, RAM from placement https://review.openstack.org/#/c/509042 and it seems like we'd need a new query in placement to be able to figure out "instances"20:59
melwittunless I'm missing something20:59
mriedemmelwitt: we have the consumers table20:59
mriedemalthough, that's going to contain migration records now too...20:59
melwittoh, I was wondering about that. it's not merged yet right? I didn't find anything in the code or the placement api-ref21:00
mriedemconsumers is in the api-ref for placement21:00
* melwitt looks again21:00
*** crushil has quit IRC21:00
mriedemoh nvm it's not21:00
mriedemthere must be a change up for that21:00
mriedemavolkov probably knows21:01
*** ragiman has quit IRC21:01
*** avolkov has quit IRC21:01
dansmithmriedem: and anything else that isn't an instance21:01
melwittI hadn't seen anything about consumers in https://github.com/openstack/nova/tree/master/nova/api/openstack/placement/handlers either21:01
dansmithlike, anything could claim some space on a cinder provider that isn't an instance21:02
exarrAnyone able to offer a decent link to multi-regions for ocata? (Ubuntu)21:02
exarr(as in to learn)21:02
melwittokay, that's what I was wondering, whether we could assume consumers are instances. so apparently not21:03
*** jmlowe has joined #openstack-nova21:03
dansmiththat's why they're called consumers and not instances :P21:03
melwittwell, yeah. I thought someone had said we could derive it from allocations and I guess I didn't know how else other than consumers21:03
*** itlinux has joined #openstack-nova21:03
dansmithwell same deal, allocations could be for other things21:04
melwittinstance mappings would work but they will also contain deleted instances21:04
*** boris_42_ has joined #openstack-nova21:05
melwittmy initial thought was, put a deleted column on instance mapping. but that comes with the challenge of "if delete fails, don't set it." maybe if we hooked it up to instance.destroy() and set the flag from there it would work21:07
*** namnh has joined #openstack-nova21:10
*** crushil has joined #openstack-nova21:11
*** david-lyle has joined #openstack-nova21:11
*** namnh has quit IRC21:15
jaypipescfriesen: yeah, no way around that (supporting both for a release or two).21:17
*** baoli has quit IRC21:19
efriedjaypipes Where do we stand on specs like https://review.openstack.org/#/c/485522/ which include a) enhancements to the existing PCI manager code; which also talk about b) new trait- or inventory-ish additions to the [pci]passthrough_whitelist in same ?21:22
efriedjaypipes There's a kind of push for (a) to do stuff we want sooner than it would be possible under NRP/GDM.  And (b) is compounding a problem we're explicitly getting rid of... but getting rid of *later*; i.e. it'll all go away at once, so does it matter that we add to it now?21:23
openstackgerritMatt Riedemann proposed openstack/nova-specs master: Add ScaleIO ephemeral storage backend  https://review.openstack.org/49592221:23
*** slaweq_ has quit IRC21:27
cfriesenjaypipes: pretty much what I figured.  I'll try and respin along the lines of your suggestion.  Also, could you take a look at https://review.openstack.org/#/c/339715/ ?21:28
dansmithmelwitt: deletes are finished on compute nodes, which can't access that bit21:29
*** smatzek has quit IRC21:29
dansmithmelwitt: and we also said we weren't going to do soft deleted stuff in the api db, because it will only get _more_ out of sync than it does today21:29
melwittdansmith: if it's in Instance.destroy wouldn't that happen in conductor? or you're saying cell conductor not supposed to access API DB21:30
dansmithmelwitt: cell conductor can't talk to the api db right21:31
melwittyeah, just brainstorming. AFAIK there's no way to determine instances from placement. I guess could count allocations of "CPU" or something like that?21:31
dansmithexcept if something outside of nova has allocations for your tenant,21:31
dansmithlike bifrost or mogan :)21:31
melwittwould they not use a different resource class than our canned ones?21:32
dansmithit's not NOVA_CPU, it's CPU21:32
dansmithand MEMORY_MB and DISK_GB21:32
dansmith(actually VCPUS, but you get the idea)21:33
openstackgerritMerged openstack/nova-specs master: Add a spec for minimal cache headers in placement  https://review.openstack.org/49685321:33
melwittso we should abandon the idea of trying to count resources in placement? if something outside nova can consume CPU and RAM, then we wouldn't want to compare those against nova quota21:34
*** gyee has quit IRC21:34
dansmithI guess that's a good point21:34
*** smatzek has joined #openstack-nova21:35
dansmithat ptg we were talking about using placement to make sure that nova and straight ironic uses don't step on each other's compute nodes21:35
dansmithand this'd be a similar thing if mogan or zun or something like that was in the picture21:36
mriedemdansmith: another fun paging spec https://review.openstack.org/#/c/506030/321:36
dansmithwe could just say that you have to have different tenants, but that's not always going to work and counting usage by other services as quota in nova is just not right at all21:36
openstackgerritMerged openstack/nova-specs master: Add ScaleIO ephemeral storage backend  https://review.openstack.org/49592221:36
dansmithmriedem: ah yeah that one has to legit be cells aware21:37
*** armax has quit IRC21:37
*** claudiub has quit IRC21:38
*** liangy has quit IRC21:38
*** lyan has joined #openstack-nova21:39
mriedemso begins my great recheckaning21:40
dansmithmriedem: so do I just get all future pagination specs to review?21:41
dansmithI'm so thrilled.21:41
*** priteau has joined #openstack-nova21:41
*** priteau has quit IRC21:41
mriedemdansmith: you are mr multi-cells21:41
mriedemso yes21:41
mriedemto be fair, these were all approved back in newton apparently, but never merged21:41
mriedemoh yeah, on that migrations paging one,21:42
mriedemi noted that the marker they propose won't work21:42
*** ijw_ has quit IRC21:43
melwittI know, we could have consumer classes in a class column! "nova_instance"21:45
mriedemit would be like osc21:45
dansmithso, I thought at one point about having each instance consume one instance type21:45
*** david-lyle has quit IRC21:45
dansmithit's breaking the model though21:45
dansmithwhat we need, IMHO, is the thing I suggested back in bristol when we were talking about this,21:46
dansmithwhich is a service type on a consume (and maybe resource provider too),21:46
dansmithso we know "this consumer is an instance" and "this provider is a compute node"21:46
dansmithbut jaypipes shot me down21:46
melwittyeah, on the surface I think it makes sense to be able to have some more info about consumers21:46
dansmithIMHO, doing quotas in placement isn't critical and probably not worth that much trouble, at least at the moment21:48
dansmithif people actually start deploying multiple cells and hit dead cells and perf issues, then we might do this to fix that, or we might have to do something totally differently21:48
dansmithdepending on what that data tells us21:48
melwittyeah, I didn't expect it to be this complex when I proposed it, so I'm cool with punting it for later21:48
dansmithyeah, I should have thought of the other-things-consume-stuff thing earlier21:49
dansmithso blame me21:49
*** smatzek has quit IRC21:51
*** mmehan has quit IRC21:51
*** smatzek has joined #openstack-nova21:52
*** itlinux has quit IRC21:53
*** esberglu has quit IRC21:53
*** aries has quit IRC21:54
*** lnxnut has joined #openstack-nova21:55
*** smatzek has quit IRC21:56
*** edmondsw has quit IRC21:57
*** slaweq_ has joined #openstack-nova21:58
jaypipesefried: I believe I already commented on that particular spec?21:59
jaypipesdansmith: actually it's VCPU, not VCPUS :)21:59
dansmithjaypipes: yeah yeah22:00
jaypipesdansmith: I originally had the can_host attribute of the resource_providers table but edleafe shot me down and said we could just use a trait to indicate a sharing provider.22:01
jaypipesedleafe is now under the bus that jaypipes was under.22:01
dansmithjaypipes: yeah can_host was wrong22:01
jaypipesalso, jaypipes now heads to dinner with wifey.'22:01
dansmithwe could do it on providers with traits for sure22:01
*** xyang1 has quit IRC22:02
melwittwe could have a who_dat attribute on consumers22:02
efriedjaypipes You did, on a couple that had that issue, but didn't address that issue specifically.22:06
efriedI mean, I'd like to be able to say, "-1: we're not mucking with the PCI manager." and/or, "-1: we're not going to do any (more) overloading of the whitelist for non-whitelisty stuff."22:07
*** lnxnut has quit IRC22:07
efriedBut I don't feel that "we" have fully landed on either policy.22:07
dansmithI do22:08
*** yassine has joined #openstack-nova22:08
efrieddansmith In the negative?22:13
dansmithefried: I feel like both your -1 reasons are supported by the feelings of people who have expressed feelings on the matter22:13
*** burt has quit IRC22:15
*** tonygunk has quit IRC22:15
*** tonygunk has joined #openstack-nova22:16
*** krtaylor has quit IRC22:17
*** gouthamr has quit IRC22:21
*** acormier has quit IRC22:22
*** claudiub has joined #openstack-nova22:23
*** lyan has quit IRC22:28
*** slaweq_ has quit IRC22:31
*** penick has quit IRC22:31
*** lbragstad has quit IRC22:34
*** slaweq_ has joined #openstack-nova22:40
mriedeminstance action events in action http://paste.openstack.org/show/622590/22:42
mriedemso apparently rebuilding a bfv instance does not fail22:42
mriedemsdague: ^22:42
mriedemcreated a bootable volume with an image, created a server from that volume, then rebuilt it with the same image that is in the volume, no failures22:43
mriedemit won't replace the root disk, but it doesn't explode22:46
*** mingyu has joined #openstack-nova22:49
*** gouthamr has joined #openstack-nova22:50
*** jaypipes has quit IRC22:50
*** jmlowe has quit IRC22:51
*** jmlowe has joined #openstack-nova22:52
*** armax has joined #openstack-nova22:53
*** jmlowe has quit IRC22:53
*** kenperkins has quit IRC22:53
*** mingyu has quit IRC22:53
*** sdague has quit IRC22:56
cfriesenmriedem: I think I found something a bit "off", though most of the time it shouldn't cause problems.  In online_data_migrations() we call aggregate_obj.migrate_aggregates(), which copies the data over to the api_db and then calls db.aggregate_delete() but never deletes the entries in 'aggregate_hosts' or 'aggregate_metadata'22:56
*** chyka_ has joined #openstack-nova22:57
cfriesenmriedem: normally we'd only delete aggregates that don't have any hosts in them, but the migration code doesn't do that check22:57
*** jmlowe has joined #openstack-nova22:58
*** chyka has quit IRC23:00
*** awaugama has quit IRC23:00
*** catintheroof has quit IRC23:01
*** chyka_ has quit IRC23:01
cfriesenwhoops, I'm wrong about not deleting the aggregate_metadata, that's handled in aggregate_delete().  But unless I'm missing something else I don't think we delete all the entries in table 'aggregate_hosts'23:03
*** lnxnut has joined #openstack-nova23:04
*** hongbin has quit IRC23:05
*** namnh has joined #openstack-nova23:07
*** ijw has joined #openstack-nova23:09
*** slaweq_ has quit IRC23:13
*** lnxnut has quit IRC23:14
*** claudiub has quit IRC23:15
*** felipemonteiro_ has quit IRC23:17
*** itlinux has joined #openstack-nova23:21
*** slaweq_ has joined #openstack-nova23:24
*** bnemec has joined #openstack-nova23:25
*** chyka has joined #openstack-nova23:26
*** abalutoiu has quit IRC23:27
*** acormier has joined #openstack-nova23:29
*** chyka has quit IRC23:30
*** acormier has quit IRC23:31
*** acormier has joined #openstack-nova23:31
*** kenperkins has joined #openstack-nova23:31
openstackgerritMerged openstack/python-novaclient stable/newton: Fix aggregate_update name and availability_zone clash  https://review.openstack.org/50781623:32
*** itlinux has quit IRC23:32
*** takashin has joined #openstack-nova23:33
*** acormier has quit IRC23:39
*** mriedem has quit IRC23:39
*** acormier has joined #openstack-nova23:39
*** namnh has quit IRC23:42
*** acormier has quit IRC23:44
*** david-lyle has joined #openstack-nova23:44
*** Swami has quit IRC23:44
*** namnh has joined #openstack-nova23:46
*** slaweq_ has quit IRC23:56
*** markvoelker has quit IRC23:59

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