Monday, 2019-03-04

*** luksky has quit IRC00:16
*** vishwanathj has joined #openstack-nova00:19
*** brinzhang has joined #openstack-nova00:23
*** jmlowe has joined #openstack-nova00:37
*** temka has quit IRC00:45
*** ileixe has joined #openstack-nova00:47
*** wolverineav has joined #openstack-nova00:55
*** wolverineav has quit IRC01:00
*** markvoelker has joined #openstack-nova01:02
*** sdake has joined #openstack-nova01:06
*** markvoelker has quit IRC01:07
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (3)  https://review.openstack.org/57410401:10
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (4)  https://review.openstack.org/57410601:10
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (5)  https://review.openstack.org/57411001:11
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (6)  https://review.openstack.org/57411301:11
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (7)  https://review.openstack.org/57497401:11
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (8)  https://review.openstack.org/57531101:11
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (9)  https://review.openstack.org/57558101:11
*** mmethot has quit IRC01:27
*** lchabert has quit IRC01:42
*** sdake has quit IRC01:44
*** sdake has joined #openstack-nova01:50
*** lchabert has joined #openstack-nova01:51
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (10)  https://review.openstack.org/57601701:55
*** wolverineav has joined #openstack-nova01:56
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (11)  https://review.openstack.org/57601801:56
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (12)  https://review.openstack.org/57601901:56
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (13)  https://review.openstack.org/57602001:57
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (14)  https://review.openstack.org/57602701:57
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (15)  https://review.openstack.org/57603101:57
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (16)  https://review.openstack.org/57629901:57
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (17)  https://review.openstack.org/57634401:58
*** sdake has quit IRC01:59
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (18)  https://review.openstack.org/57667301:59
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (19)  https://review.openstack.org/57667601:59
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (20)  https://review.openstack.org/57668901:59
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (21)  https://review.openstack.org/57670902:00
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (22)  https://review.openstack.org/57671202:00
*** wolverineav has quit IRC02:01
*** markvoelker has joined #openstack-nova02:03
*** sdake has joined #openstack-nova02:05
*** yedongcan has joined #openstack-nova02:05
*** markvoelker has quit IRC02:06
*** sdake has quit IRC02:08
*** Dinesh_Bhor has joined #openstack-nova02:31
*** Dinesh_Bhor has quit IRC02:31
*** Dinesh_Bhor has joined #openstack-nova02:34
*** vishwanathj has quit IRC02:35
*** chason has quit IRC02:36
*** chason has joined #openstack-nova02:38
*** sdake has joined #openstack-nova02:42
*** wolverineav has joined #openstack-nova02:42
*** sdake has quit IRC02:44
openstackgerritBrin Zhang proposed openstack/nova master: Remove the string check of the flavor attribute 'swap'  https://review.openstack.org/63901202:44
*** tbachman has joined #openstack-nova02:46
*** wolverineav has quit IRC02:50
*** sdake has joined #openstack-nova02:51
*** psachin has joined #openstack-nova02:52
*** vishwanathj has joined #openstack-nova02:55
*** wolverineav has joined #openstack-nova03:05
*** sridharg has joined #openstack-nova03:14
*** mlavalle has quit IRC03:14
openstackgerritTakashi NATSUME proposed openstack/nova master: Add descriptions of numbered resource classes and traits  https://review.openstack.org/62149403:23
*** sdake has quit IRC03:24
*** sdake has joined #openstack-nova03:26
*** sdake has quit IRC03:41
*** wolverineav has quit IRC03:47
*** wolverineav has joined #openstack-nova03:48
*** janki has joined #openstack-nova03:52
*** vishwanathj has quit IRC03:52
*** wolverineav has quit IRC04:14
*** marst has joined #openstack-nova04:18
*** udesale has joined #openstack-nova04:20
*** jmlowe has quit IRC04:29
*** tbachman has quit IRC04:32
*** marst has quit IRC04:35
*** udesale has quit IRC04:39
*** udesale has joined #openstack-nova04:41
*** tbachman has joined #openstack-nova04:41
*** lbragstad has quit IRC04:44
*** wolverineav has joined #openstack-nova04:52
*** lbragstad has joined #openstack-nova04:54
*** Sundar has joined #openstack-nova04:54
*** sdake has joined #openstack-nova05:09
*** wolverineav has quit IRC05:23
*** udesale has quit IRC05:39
*** udesale has joined #openstack-nova05:39
*** sdake has quit IRC05:48
*** sdake has joined #openstack-nova05:54
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix an error when generating a host ID  https://review.openstack.org/63993605:55
*** tbachman has quit IRC05:55
*** sdake has quit IRC06:01
*** tbachman has joined #openstack-nova06:10
*** tbachman_ has joined #openstack-nova06:12
*** ratailor has joined #openstack-nova06:14
*** tbachman has quit IRC06:15
*** tbachman_ is now known as tbachman06:15
*** yedongcan has quit IRC06:27
*** tkajinam_ has joined #openstack-nova06:36
*** tkajinam has quit IRC06:38
*** udesale has quit IRC06:46
*** udesale has joined #openstack-nova06:47
*** igordc has quit IRC06:47
*** markvoelker has joined #openstack-nova06:48
*** wolverineav has joined #openstack-nova06:54
*** wolverineav has quit IRC07:02
*** Luzi has joined #openstack-nova07:02
*** udesale has quit IRC07:05
*** belmoreira has joined #openstack-nova07:05
*** udesale has joined #openstack-nova07:08
*** udesale has quit IRC07:09
*** udesale has joined #openstack-nova07:09
*** slaweq has joined #openstack-nova07:15
*** markvoelker has quit IRC07:21
*** ircuser-1 has quit IRC07:22
*** jangutter has joined #openstack-nova07:23
*** wolverineav has joined #openstack-nova07:27
*** wolverineav has quit IRC07:31
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147607:39
*** wolverineav has joined #openstack-nova07:41
*** wolverineav has quit IRC07:47
*** bhagyashris_ has joined #openstack-nova07:50
*** sapd1 has joined #openstack-nova07:57
*** lchabert has quit IRC07:59
*** tbachman has quit IRC08:00
*** lchabert has joined #openstack-nova08:00
openstackgerritMichael Still proposed openstack/nova master: Implement simple prometheus metrics for nova.  https://review.openstack.org/64068808:01
*** tesseract has joined #openstack-nova08:05
*** luksky has joined #openstack-nova08:07
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147608:10
*** panda|ruck|off is now known as panda|ruck08:12
*** rpittau|sardegna is now known as rpittau08:13
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147608:14
*** dpawlik has joined #openstack-nova08:16
*** markvoelker has joined #openstack-nova08:18
*** tetsuro has joined #openstack-nova08:18
*** shilpasd has joined #openstack-nova08:18
*** sridharg has quit IRC08:20
*** helenafm has joined #openstack-nova08:23
*** pcaruana has joined #openstack-nova08:25
*** tosky has joined #openstack-nova08:27
*** giblet is now known as gibi08:33
*** wolverineav has joined #openstack-nova08:33
*** rha has joined #openstack-nova08:36
*** wolverineav has quit IRC08:37
*** phasespace has quit IRC08:39
*** dtantsur|afk is now known as dtantsur08:44
*** ralonsoh has joined #openstack-nova08:46
*** lbragstad has quit IRC08:46
*** tssurya has joined #openstack-nova08:47
*** tkajinam_ has quit IRC08:48
*** markvoelker has quit IRC08:50
*** rchurch_ has joined #openstack-nova08:51
*** rchurch has quit IRC08:53
*** awalende has joined #openstack-nova08:53
openstackgerritmelanie witt proposed openstack/nova master: Add user_id field to InstanceMapping  https://review.openstack.org/63335008:56
openstackgerritmelanie witt proposed openstack/nova master: Populate InstanceMapping.user_id during migrations and schedules  https://review.openstack.org/63857408:56
openstackgerritmelanie witt proposed openstack/nova master: Add online data migration for populating user_id  https://review.openstack.org/63335108:56
openstackgerritmelanie witt proposed openstack/nova master: Add get_counts() to InstanceMappingList  https://review.openstack.org/63807208:56
openstackgerritmelanie witt proposed openstack/nova master: WIP Count instances from mappings and cores/ram from placement  https://review.openstack.org/63807308:56
openstackgerritmelanie witt proposed openstack/nova master: Use instance mappings to count server group members  https://review.openstack.org/63832408:56
tssuryamelwitt: you are online still ? :o08:56
melwitttssurya: yeah :( wanted to get this uploaded before mriedem morning08:57
melwittgoing to go to bed now08:57
tssuryayea good night! thanks08:57
gibisean-k-mooney: hi! I can start digging the pf interface name autodetection now. If you have any extra info top of what we discussed on Friday then hit me with it08:57
gibimelwitt: good night!08:57
melwittgoodnight everyone :)08:58
* gibi still have a full day ahead :D08:58
*** janki has quit IRC09:00
*** janki has joined #openstack-nova09:00
*** udesale has quit IRC09:04
openstackgerritBalazs Gibizer proposed openstack/nova master: Add pf_interface_name tag to passthrough_whitelist  https://review.openstack.org/62531109:10
openstackgerritBalazs Gibizer proposed openstack/nova master: Ensure that bandwidth and VF are from the same PF  https://review.openstack.org/62354309:10
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636009:10
*** ttsiouts has joined #openstack-nova09:11
openstackgerritMichael Still proposed openstack/nova master: Implement simple prometheus metrics for nova.  https://review.openstack.org/64068809:15
*** zhubx has quit IRC09:28
*** zhubx has joined #openstack-nova09:29
*** derekh has joined #openstack-nova09:34
*** bhagyashris_ has quit IRC09:34
*** sapd1 has quit IRC09:41
*** jaosorior has joined #openstack-nova09:41
*** jistr is now known as jistr|sick09:42
openstackgerritLajos Katona proposed openstack/python-novaclient master: Add support for microversion v2.72  https://review.openstack.org/63723409:45
*** markvoelker has joined #openstack-nova09:47
*** IvensZambrano has joined #openstack-nova09:54
*** cdent has joined #openstack-nova09:57
*** wolverineav has joined #openstack-nova10:10
openstackgerritBoxiang Zhu proposed openstack/nova master: Raise error when force_raw_images and images_type are incompatible  https://review.openstack.org/64027110:13
*** tetsuro has quit IRC10:15
*** wolverineav has quit IRC10:15
*** dtantsur has quit IRC10:18
*** dtantsur has joined #openstack-nova10:18
*** takashin has quit IRC10:19
*** markvoelker has quit IRC10:21
*** kashyap has joined #openstack-nova10:35
kashyapstephenfin: Hi, when you get a moment, I am sure a bug here is required (it might only cause confusion): https://review.openstack.org/#/c/637578/1/nova/virt/libvirt/driver.py ("Omit needless check on 'CONF.serial_console'")10:37
kashyap(And the small refactorings we discussed there should be done as follow-ups.)10:37
aspiersmorning kashyap :)10:38
kashyapaspiers: Morning10:38
lyarwoodmdbooth: https://review.openstack.org/#/q/topic:bug/1803961+(status:open+OR+status:merged) - can you take a look at this today if you have time?10:38
aspierskashyap: https://review.openstack.org/#/c/640483/1 is ready for review now10:39
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix wrong consumer type in logging  https://review.openstack.org/64072310:39
kashyapaspiers: Thanks for reworking: https://review.openstack.org/#/c/633855/7/nova/virt/libvirt/host.py10:42
kashyapaspiers: Yep, looking at the 'emulator' one10:42
*** takashin has joined #openstack-nova10:45
*** ratailor has quit IRC10:51
*** ratailor has joined #openstack-nova10:52
*** cdent has quit IRC10:54
*** luksky has quit IRC11:00
openstackgerritStephen Finucane proposed openstack/nova master: Add nits from Id2beaa7c4e5780199298f8e58fb6c7005e420a69  https://review.openstack.org/64072911:01
openstackgerritStephen Finucane proposed openstack/nova master: doc: Rework 'config-drive' user doc  https://review.openstack.org/64073011:01
gibisean-k-mooney: I made some progress with the pf_interface_name auto detection but I realized that this makes the feature virt driver dependent. As far as I see xenapi also supports VF and PF passthrough but I don't know how to autodetect pf_interface_name in xenapi :/11:01
openstackgerritZhenyu Zheng proposed openstack/nova master: Add compute service support for attach/detach root volume  https://review.openstack.org/61475011:04
*** ttsiouts has quit IRC11:04
*** ttsiouts has joined #openstack-nova11:05
kashyapgibi: It is OK to update a spec file retroactively, yes?  Specifically something cosmetic like changing the name of config attribute11:05
gibikashyap: it is even encuraged to update the spec file if the design diverged11:06
kashyapgibi: E.g. I'm suggesting the patch poster to change from 'cpu_models' to 'cpu_model_list' (to better distinguish from the existing 'cpu_model')11:06
kashyaphttps://review.openstack.org/#/c/637834/4/nova/conf/libvirt.py@54811:06
kashyapgibi: Good, thought so.  Because that's the most sensible thing.11:06
HD|Laptophey all11:06
HD|Laptopis there a way to force nova to only use Cinder for all kinds of block storage, both for the images it fetches from Glance as well as for "ordinary" servers created via a simple "openstack server create" without manually allocating/attaching a Cinder volume?11:06
*** jangutter_ has joined #openstack-nova11:07
lyarwoodHD|Laptop: no, it has been discussed at recent Forum / PTGs etc but nothing has been worked on AFAIK.11:07
HD|Laptopoh damn. 'cause all my compute nodes have 128GB SSDs only and that's it, for storage I have a nice 80TB QNAP NAS :'D11:08
HD|Laptopanother thing: can I somehow specify additional parameters for qemu/kvm? I need to match https://github.com/kholia/OSX-KVM/blob/master/boot-macOS.sh11:09
*** jangutte_ has joined #openstack-nova11:09
*** jangutter has quit IRC11:09
HD|Laptop(and yes I do only run this on original Macs, 9x 2010-era Mac Pro cheesegraters, so I don't violate licenses)11:09
*** ttsiouts has quit IRC11:09
lyarwoodHD|Laptop: you can supply additional QEMU params via libvirt but I don't recall a way of doing that part via Nova11:10
* lyarwood looks11:10
HD|Laptopthanks :)11:10
*** jangutter_ has quit IRC11:11
lyarwoodHD|Laptop: yeah I can't see a way of passing extra args from Nova to libvirt and down to QEMU tbh11:14
kashyapHD|Laptop: lyarwood: There is no way that Nova provides it either.11:14
kashyapIf you go behind libvirt (or Nova's) back and use QEMU command-line options, and if something breaks, you get to keep the pieces :-)11:15
lyarwoodkashyap: could you use a custom machine type for this?11:15
*** udesale has joined #openstack-nova11:15
kashyaplyarwood: Machine types won't help in this case.  (They are a particular chipset with additional devices)11:16
* kashyap reads the context from HD|Laptop 11:16
lyarwoodkashyap: yeah just for the additional devices I mean11:17
kashyapThat's a bit too heavy-weight :-)11:17
kashyap(Earlier typo: s/additional/default/)11:17
openstackgerritStephen Finucane proposed openstack/nova master: WIP: Add initial version of metadata configuration  https://review.openstack.org/64073311:17
*** markvoelker has joined #openstack-nova11:18
kashyaplyarwood: Aside: there's already some discusions upstream to propose a separate machine type for 'x86' that is "legacy-free"11:19
kashyap(Just like how AArch64 has a machine type called 'virt', which is expressely designed for virtual machines.)11:19
*** ileixe has quit IRC11:19
HD|Laptopkashyap: ah damn. How is qemu called in nova, anyway? Tried to read myself through the libvirt driver, but I couldn't even find out where it gets "-machine pc-i440fx-3.1" from, there isn't even any single mention of that machine type anywhere in the nova git repository11:23
HD|LaptopMaybe I can do a hot-patch in there...11:23
*** zhubx has quit IRC11:25
kashyapHD|Laptop: Nova models every guest feature via libvirt classes / objects, which lets libvirt build an opinionated  QEMU command-line.11:25
*** zhubx has joined #openstack-nova11:25
*** jangutte_ is now known as jangutter11:25
kashyapHD|Laptop: No specific machine types are hard-coded in Nova, obviously.  They're selected by what your host hypervisor supports11:26
mdboothzhubx: Are you the author of https://review.openstack.org/#/c/640271/ ?11:26
mdboothlyarwood: Looking now.11:26
sean-k-mooneygibi: you are correct that it would make it virt driver dependent in a way but no more then manually adding the pf_interface_name. for xenapi the compute service i belive will be runnin in dom0 i have not deployed xen in quite a long time but i think you could use sysfs the same way. after all if the sriov_nic agent is running in dom0 and can see the interface netdevs then you can also see the vf and11:27
HD|Laptopkashyap: ah, ok. where do I have to look for the final call to libvirt for starting the qemu process?11:27
sean-k-mooneylookup the pf netdev the same way11:27
mdboothzhubx: If so, I just hit that again. I think this is the right approach and it'll be good to go if you can fix up a few things and add unit tests.11:27
kashyapHD|Laptop: For machine type handling in Nova, see this: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L4221,L424811:27
stephenfinkashyap: Is required or is not required?11:28
kashyapHD|Laptop: There is no "final call" in Nova; libvirt takes care of it.11:28
kashyapstephenfin: Not required.11:28
kashyapHD|Laptop: If you want additional command-line options that Nova currently doesn't have; then you have to model that into Nova by providing libvirt config class objects.11:29
kashyapstephenfin: (Sorry, saw the typo in my sentence in the scrollback :-))11:30
sean-k-mooneykashyap: that is allowed to a point. we do not allow the use of the qemu:args element to be genearted to allow raw qemu commands to be added11:30
mdboothlyarwood: Did you speak to mriedem about removing the timestamp check from the retype tempest test?11:30
sean-k-mooneykashyap: i have tried in the past to enable features that are not supported by libvirt yet or on older versions11:30
mdboothI'll take your work for it that they're flaky, but IIUC the purpose was to assert functionally that the volume contains the same data after retype.11:31
mdboothlyarwood: IIRC mriedem was especially keen on this.11:31
lyarwoodmdbooth: I had no idea he was attached to it tbh11:32
lyarwoodmdbooth: I'll bring it up with him this afternoon11:32
lyarwoodmdbooth: and yeah it has been pretty awful for me locally testing this11:33
mdboothYeah, I can imagine. Seems to work ok in the gate, though.11:33
kashyapsean-k-mooney: Yeah, that's what I meant: Nova not allowing libvirt's <qemu:commandline> element which allows arbitrary pass-through.11:34
*** cdent has joined #openstack-nova11:34
sean-k-mooneyyep that is the element i was thinking of11:34
kashyapsean-k-mooney: (And yeah, I've used it for one-off hacks in Nova in the past.)11:34
sean-k-mooneyya i used it to demonstarte that you can run ovs-dpdk without hugepages before i got funding to have the libvirt support for file backed memory developed11:35
sean-k-mooneyi never got time to add the nova support for file backed memory the way i wanted it but redhat did it a release or two later for other uses11:36
sean-k-mooney<qemu:commandline> is still useful for POCs of thing that will be comming soon so you can write the nova side of the feature and just swap out the xml generation code when it lands in libvirt11:38
kashyapsean-k-mooney: Yeah, is useful in many scenarios; I consider it as a Swiss Army knife.11:39
*** shilpasd has quit IRC11:41
*** thgcorrea has joined #openstack-nova11:43
gibisean-k-mooney: thanks. I think I will skip the xen support now due to the shortness of time. But I feel that I will manage to push the libvirt support today11:49
*** luksky has joined #openstack-nova11:50
*** markvoelker has quit IRC11:50
sean-k-mooneygibi: good to hear. i ended up spending my weekend hacking time on figuring out how to get vs-code to debug eventlet code which i did but as a result i didnt get much time to look at the auto generation11:51
gibisean-k-mooney: no worries you helped a lot by point out the code pieces I need11:51
*** udesale has quit IRC11:52
*** udesale has joined #openstack-nova11:53
openstackgerritZhenyu Zheng proposed openstack/nova master: Add compute service support for attach/detach root volume  https://review.openstack.org/61475011:54
*** wolverineav has joined #openstack-nova11:59
*** ttsiouts has joined #openstack-nova11:59
*** wolverineav has quit IRC12:04
cdentthanks for the response mdbooth, I was doing a run through of patches that touch vmware related stuff (there's an automated thingie that tells me about such stuff). I wasn't sure of the state of that one12:10
mdboothcdent: thanks12:10
openstackgerritZhenyu Zheng proposed openstack/nova master: Detach/Attach root volume API changes  https://review.openstack.org/62398112:13
*** janki has quit IRC12:16
*** janki has joined #openstack-nova12:16
*** udesale has quit IRC12:20
*** udesale has joined #openstack-nova12:21
*** brinzhang has quit IRC12:21
*** priteau has joined #openstack-nova12:30
*** ttsiouts_ has joined #openstack-nova12:36
*** phasespace has joined #openstack-nova12:37
*** udesale has quit IRC12:38
*** ttsiouts has quit IRC12:40
*** udesale has joined #openstack-nova12:41
*** sapd1 has joined #openstack-nova12:45
*** markvoelker has joined #openstack-nova12:48
*** edmondsw has joined #openstack-nova12:49
kashyapaspiers: Have you noticed this: https://www.redhat.com/archives/libvir-list/2019-February/msg01730.html (New Feature: Intel MKTME Support)12:53
kashyapaspiers: It ("MKTME": Multi-Key Total Memory Encryption) is Intel's equivalent to AMD SEV12:53
*** ratailor has quit IRC12:53
*** ratailor has joined #openstack-nova12:54
aspierskashyap: ah interesting, so another alternative to the SGX approach?12:54
kashyapaspiers: I don't know much (if anything) about Intel's SGX12:54
kashyapaspiers: But the MKTME isn't yet merged in the kernel (https://lwn.net/Articles/758313/)12:55
kashyap(Near as I see)12:55
kashyaps/MKTME/"Support for MKTME"/12:55
sean-k-mooney SGX is intels secure enclave technology that prevent the host kernel and userpace processes form acessing meory that is in the enclave and assigend to a specifc process12:55
*** awalende has quit IRC12:56
sean-k-mooneythe SGX enclaves are created and managed via system management mode so even the host kernel cannot read or write to if it is not the owner of the enclve but i dont be think SGX actully encrypts the memory12:57
kashyapsean-k-mooney: Ah, thanks for the nice summary12:57
sean-k-mooneyso sgx + mktme would give someting simlar but maybe more secure then SEV as teh memory is not just encrypted but partitioned12:59
sean-k-mooneyi dont realy know anyting about mktme however12:59
*** artom has joined #openstack-nova13:00
artom~o~13:01
*** jistr|sick is now known as jistr|sick|mtg13:01
* mdbooth wonders how the kernel does a context switch to memory it can't access13:01
mdboothPerhaps there's some super-small ring -1 which is allowed to do context switches?13:01
*** cdent has quit IRC13:03
sean-k-mooneymdbooth: by the way https://review.openstack.org/#/c/634276/ may have been related to some of the port binding issue you were having with failed migrations13:04
*** awalende has joined #openstack-nova13:05
mdboothsean-k-mooney: Nice, thanks.13:06
*** udesale has quit IRC13:08
sean-k-mooneymdbooth:for sgx. i dont think so. memory in sgx enclaves is not swapable. so other the createing and destroying them there isnt anything that a ring -1 e.g. hypervior layer process would need to do13:09
*** ratailor has quit IRC13:09
*** udesale has joined #openstack-nova13:09
sean-k-mooneyas far as i know the sgx enclaves are mapped into the process virtual memory via the iommu. as such i dont think there is a context switch requried for a process to acess it. that said its been 2 years since i looked at how sgx worked in detail13:12
mdboothlyarwood: https://review.openstack.org/#/c/639331/13:14
mdboothlyarwood: I'd like to get mriedem's opinion on that.13:14
mdboothGrr, gerrit seems to bogosort the results of a change query. The order changes every time I look at it.13:15
openstackgerritMerged openstack/nova stable/rocky: Fix legacy-grenade-dsvm-neutron-multinode-live-migration  https://review.openstack.org/64018613:20
openstackgerritMerged openstack/nova master: api-ref: explain aggregate set_metadata semantics  https://review.openstack.org/64046013:20
openstackgerritMerged openstack/nova master: Remove mox in unit/network/test_neutronv2.py (3)  https://review.openstack.org/57410413:20
*** markvoelker has quit IRC13:21
*** cdent has joined #openstack-nova13:21
*** sdake has joined #openstack-nova13:24
lyarwoodmdbooth: sorry was on the phone13:26
lyarwoodmdbooth: ack thanks13:26
mdboothlyarwood: You've had the same failure 3 times in a row on the migration tempest test, btw.13:26
mdboothlyarwood: failure to delete type because it's still in use.13:26
lyarwoodmdbooth: yeah clean up is racing13:26
mdboothHave you investigated that already?13:26
lyarwoodmdbooth: next on my list13:27
*** udesale has quit IRC13:36
*** marst has joined #openstack-nova13:37
*** wolverineav has joined #openstack-nova13:47
*** lbragstad has joined #openstack-nova13:49
*** wolverineav has quit IRC13:52
*** tetsuro has joined #openstack-nova13:55
*** marst has quit IRC13:55
openstackgerritMerged openstack/nova master: Fixes race condition with privsep utime  https://review.openstack.org/62574114:00
efriedn-sch meeting now in #openstack-meeting-alt14:00
openstackgerritMerged openstack/nova master: Optimize populate_queued_for_delete online data migration  https://review.openstack.org/63984014:04
openstackgerritMerged openstack/nova master: Remove wrong description for auto resize confirm  https://review.openstack.org/63835714:04
openstackgerritMerged openstack/nova stable/rocky: Fix overcommit for NUMA-based instances  https://review.openstack.org/63319714:04
*** awaugama has joined #openstack-nova14:06
*** mmethot has joined #openstack-nova14:09
*** sdake has quit IRC14:14
*** ttsiouts_ has quit IRC14:18
*** markvoelker has joined #openstack-nova14:18
*** jmlowe has joined #openstack-nova14:18
*** ttsiouts has joined #openstack-nova14:19
*** janki has quit IRC14:19
*** ttsiouts has quit IRC14:21
*** ttsiouts has joined #openstack-nova14:21
*** janki has joined #openstack-nova14:24
openstackgerritJim Rollenhagen proposed openstack/nova stable/rocky: ironic: stop hammering ironic API in power sync loop  https://review.openstack.org/64077114:24
openstackgerritJim Rollenhagen proposed openstack/nova stable/rocky: ironic: check fresh data when sync_power_state doesn't line up  https://review.openstack.org/64077214:24
jrollturns out we never backported that first one >.>14:25
*** sdake has joined #openstack-nova14:25
*** sdake has quit IRC14:25
*** pcaruana has quit IRC14:31
*** marst has joined #openstack-nova14:31
stephenfinlyarwood: Could you look at https://review.openstack.org/#/c/636919/ today?14:32
*** phasespace has quit IRC14:32
*** beekneemech is now known as bnemec14:33
*** mriedem has joined #openstack-nova14:36
mriedemwho's ready to rush some crap in14:37
*** sdake has joined #openstack-nova14:38
sean-k-mooneyi dont know its only monday :P we have 3 whole days left to rush crap in14:38
stephenfinmriedem: Just tell me what I need to blindly +W14:39
* stephenfin grabs blindfold14:39
openstackgerritMatt Riedemann proposed openstack/nova master: Add nits from Id2beaa7c4e5780199298f8e58fb6c7005e420a69  https://review.openstack.org/64072914:44
openstackgerritMatt Riedemann proposed openstack/nova master: doc: Rework 'config-drive' user doc  https://review.openstack.org/64073014:44
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147614:49
*** sdake has quit IRC14:49
*** markvoelker has quit IRC14:50
*** tetsuro has quit IRC14:53
openstackgerritStephen Finucane proposed openstack/nova master: objects: Store InstancePCIRequest.numa_policy in DB  https://review.openstack.org/62444414:53
openstackgerritStephen Finucane proposed openstack/nova master: WIP: fakelibvirt: Add ability to generate fake PCI devices  https://review.openstack.org/64040914:53
*** sdake has joined #openstack-nova14:55
* alex_xu can sleep well with learning something new on request group14:56
mriedemsounds like a nightmare to me14:56
bauzasmriedem: give me some crap, I'm thirsty14:57
openstackgerritKashyap Chamarthy proposed openstack/nova master: qemu: Make disk image conversion dramatically faster  https://review.openstack.org/64078115:00
kashyapmdbooth: ^ If you have time.15:00
kashyap(Change itself is a "one-worder"; but the commit message is long, with a back-of-the-envelope "benchmark" & context)15:01
*** takashin has left #openstack-nova15:01
*** mlavalle has joined #openstack-nova15:02
*** jistr|sick|mtg is now known as jistr|sick15:02
*** eharney has joined #openstack-nova15:04
*** ttsiouts has quit IRC15:06
*** ttsiouts has joined #openstack-nova15:07
*** sapd1 has quit IRC15:09
*** sapd1 has joined #openstack-nova15:10
*** ttsiouts has quit IRC15:11
*** sdake has quit IRC15:11
*** priteau has quit IRC15:11
*** ralonsoh has quit IRC15:11
*** cfriesen has joined #openstack-nova15:12
*** ralonsoh has joined #openstack-nova15:12
*** ralonsoh has quit IRC15:12
*** ralonsoh has joined #openstack-nova15:13
openstackgerritBalazs Gibizer proposed openstack/nova master: Ensure that bandwidth and VF are from the same PF  https://review.openstack.org/62354315:13
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636015:13
*** tbachman has joined #openstack-nova15:13
*** Luzi has quit IRC15:13
gibisean-k-mooney, stephenfin: removed the whitelis tag config extension and added the auto detection of the pf interface name ^^15:14
*** sdake has joined #openstack-nova15:15
gibiany feedback are welcome15:16
artommriedem, well, *my* crap is going to remain unrushed, and will smell delicious by the time it lands ;)15:16
artomefried, btw, I'm +1 on https://review.openstack.org/#/c/635440/9, but I left a question inline15:17
mriedemumm, what actually puts the instance into ERROR state in ^?15:20
artomGremlins?15:22
jroll.v 3715:22
jrollblurhg, sorry15:22
* kashyap wonders if he's going to win the "longest commit message" award in Nova: https://review.openstack.org/#/c/640781/15:22
kashyap:D15:22
* kashyap bbiab15:22
cfriesensean-k-mooney: is it expected that PCI aliases will be configured in the nova-api nova.conf the same as all compute nodes?15:23
sean-k-mooneyyes15:23
cfriesenmriedem: there you go ^15:23
mriedemisn't that....weird?15:25
*** pcaruana has joined #openstack-nova15:25
mriedemthe computes can have different pci devices right? so wouldn't the alias config be per-compute?15:25
cfriesenmriedem: I assume the idea is that a given alias should mean the same thing across the whole cloud, even if not all computes have that device.15:29
mriedemok i guess it's true https://docs.openstack.org/nova/latest/admin/pci-passthrough.html#configure-nova-api-controller15:31
mriedemand documented that way15:31
sean-k-mooneymriedem: it is weird but its needed for reasons15:31
sean-k-mooneyim trying to rememebr why15:31
sean-k-mooneystephenfin:  do you rememebr ?15:31
sean-k-mooneyi think i had something to do with either hardwar offloaded ovs or pci numa policies15:32
stephenfinsean-k-mooney: Why they have to be specified on the API node?15:33
sean-k-mooneycfriesen: it was related to scheduling. we needed the content of teh alias to aloow the schulers to make desissions15:33
sean-k-mooneystephenfin: yes15:33
sean-k-mooneyi think it was so the numa toplogy filter could take the pci_numa policy into effect15:33
cfriesenI'm not complaining, it's helpful for the flavor/image validation. :)15:34
stephenfinNah, it's because it's needed for move operations15:34
stephenfinSee b4ce2d9f12ef6d50837e4133dff09fa43fd152d215:34
*** wolverineav has joined #openstack-nova15:35
sean-k-mooneystephenfin: well the move operations e.g. cold migration need it for schudleing15:35
sean-k-mooneymriedem: the alias does not contain the pci addresses. so it works independenly form the compute nodes for most usecause15:37
sean-k-mooneyit does mean if you put the pci vendor and product id in the alias that it has to match across the compute nodes but you are better off haveing 1 alias per device model anyway15:38
*** wolverineav has quit IRC15:40
*** luksky has quit IRC15:40
mriedemsure15:41
mriedemartom: btw, how is your downstream numa live migration whitebox tempest testing stuff passing if the intel people testing it are finding issues? just different issues from what the CI would hit?15:42
*** janki has quit IRC15:42
sean-k-mooneymriedem: i responded on the mailing list15:42
sean-k-mooneymriedem: they were using virsh edit to view the xml instead of virsh dumpxml15:43
sean-k-mooneyvirsh dumpxml shows the current state of the vm15:43
sean-k-mooneyvirsh edit shows the xml that the vm would have it you were to reboot it15:43
sean-k-mooneyit looks like when we update the xml as part of a migration virsh edit still shows the original xml15:44
sean-k-mooneyso that is the reason for the delta. the whitebox test use virsh dumpxml which i belive is correct15:44
*** markvoelker has joined #openstack-nova15:48
*** cdent has quit IRC15:48
mriedem(8:39:26 AM) stephenfin: mriedem: Just tell me what I need to blindly +W15:52
mriedemstephenfin: don't go blind on this, but https://review.openstack.org/#/c/623543/15:53
mriedemsince gibi abandoned https://review.openstack.org/#/c/625311/15:53
stephenfinmriedem: Yup, already on it (gibi pinged us)15:53
mriedemok that's a big change so i'll hold off until someone more familiar with the pf/vf/pci stuff has gone through it15:53
* stephenfin wonders if adrianc fancies taking a look at it too15:54
*** wolverineav has joined #openstack-nova15:57
gibimriedem: sorry about the size of that patch, it organicly grow to that15:58
sean-k-mooneygibi: is your current plan to support but the manual tag in the whitelist + the auto disovery of the tag?15:59
*** sdake has quit IRC16:00
sean-k-mooneygibi: im ok with that by the way as a fallback mechanisum ignoring the similar nameing of pf_interface_name and devname16:00
sean-k-mooneyoh you have abandoned it never mind im looking at the wrong patch16:01
bauzasgibi: I just started reviewing https://review.openstack.org/#/c/623543/ and I have a concern16:01
gibisean-k-mooney: I dropped the whitelist support16:01
bauzasgibi: do we agreed on having a pci claim providing a reschedule ?16:01
bauzas(for this I mean)16:02
gibibauzas: fixing the PCI filter would be a really big work as the filters doesn't know about allocation candidates16:02
sean-k-mooneygibi: cool that was my perference but i could live with it if it was just a fallback for if we could not auto detect it.16:02
bauzasgibi: sure, I'm just saying we haven't this in the spec, right?16:02
gibisean-k-mooney: libvirt autodetection was pretty small change so I went for it16:02
gibibauzas: yeah, devil is in the details so this edge case surfaced during implementation16:03
gibibauzas: I'm planning to retrofit the spec16:03
bauzasgibi: okay, my point is that I want to make sure people are okay with this16:03
gibibauzas: sure, I also want to make sure, hence the big commit message16:04
bauzas(fwiw, I have the same problem with any NUMA affinity for vGPUs ;) )16:04
bauzaschanging the filter is difficult16:04
gibiif we end up fearing the re-schedule then I can drop this patch and document the limitation16:04
bauzasgibi: I'm not really bad with this16:05
bauzasalso, https://review.openstack.org/#/c/636360/19 is the API microversion that really creates the new feature, right?16:05
gibibauzas: yes, everything is rejected until https://review.openstack.org/#/c/636360/1916:06
gibibauzas: everything involving ports with resource request16:06
bauzasyeah, I'm seeing with https://review.openstack.org/#/c/636360/19/nova/api/openstack/common.py16:06
gibibauzas: except server delete16:06
gibibauzas: yepp, that is the key16:07
*** wolverineav has quit IRC16:07
*** macza has joined #openstack-nova16:07
*** burt has quit IRC16:07
*** luksky has joined #openstack-nova16:11
stephenfingibi: Left a series of nits plus a suggestion to s/pf_interface_name/parent_ifname/, but otherwise +2 on https://review.openstack.org/#/c/623543/ Thanks for tackling that16:15
bauzasgibi: just to make it clear, if you solve https://review.openstack.org/#/c/623543/39//COMMIT_MSG by the compute claim, that means that you check the compute version for all the move ops ?16:16
* bauzas tries to look at the spec to see whether we only accept compute versions16:16
gibistephenfin: I'm OK to rename the key to parent_ifname16:17
stephenfinand I'm OK to re-review if you do :)16:17
bauzasdon't see any of this in https://specs.openstack.org/openstack/nova-specs/specs/stein/approved/bandwidth-resource-provider.html#upgrade-impact16:17
gibibauzas: I think you found a problem16:17
*** tobeass-urdin is now known as tobias-urdin16:17
gibibauzas: we did not planned to check the compute version16:18
gibibauzas: right now if you boot a server and the scheduler select a host that is still Rocky then the pci claim fails as the new pf_interface_name key will be missing from the pci spec16:19
gibibauzas: so no resource inconsistency but more re-schedule will happen16:19
*** mrch_ has quit IRC16:20
bauzasgibi: wait16:20
gibibauzas: ohh no. the InstancePCIRequest will not request the new key either if the request lands on an old compute16:20
bauzasgibi: yeah that16:20
*** markvoelker has quit IRC16:21
bauzasit's old code so it doesn't know about anything16:21
gibibauzas: yeah, the InstancePCIRequest change is also in the compute/manager16:21
gibiBUT if we have an old compute then the old compute will not have bandwidth inventory16:22
gibiso placement will not select it16:22
gibi(assuming that nova-compute is updated along with the neutron agents)16:23
bauzasgibi: that was the original reason why we didn't need a check16:23
bauzasbecause placement checks it for free16:23
bauzasbut16:23
bauzasfor a classic move operation, you can end up with the late-check not being read16:23
mriedemthere is no move support for this in stein16:24
*** wolverineav has joined #openstack-nova16:24
gibibauzas: ^^16:24
mriedemgibi: and we enforce that in the api right?16:24
mriedemyeah i remember16:24
gibimriedem: yes, the moves are rejected16:24
gibimriedem: even after the 2.7216:24
bauzasmriedem: what happens with my instance ? am I able to migrate it with dropping the bandwidth check, or do i get some ERROR ?16:24
mriedemyou get a 400 or 40916:25
bauzasah, that's an API enforcement16:25
mriedemi can't remember which16:25
bauzasgood then16:25
* gibi makes a note about possible compute version check needs for move16:25
mriedemhttps://github.com/openstack/nova/blob/master/nova/api/openstack/compute/evacuate.py#L123 40016:25
bauzasso that works, at least if neutron agent is upgraded with compute service, which is usually what's done16:25
mriedemif the neutron agent is upgraded before nova-compute it could be reporting inventory before the compute is ready to handle those types of allocations...16:26
mriedembut i'm not sure how realistic that is16:26
gibibauzas: even if they ugrade the neutron agent they can skip the bw config in the agent until nova-compute is upgraded16:26
bauzasworth a note at least16:26
gibibauzas: the whole neutron bw inventory is config based16:26
bauzasyeah16:26
mriedemhow much of the compute-level stuff is virt-driver specific?16:27
bauzasso, ops can upgrade whenever they want, and once they're done, they set configs16:27
mriedemjust the multi-VF on same PF thing?16:27
bauzasgibi: planned to write some ops docs btw. ?16:27
bauzasthat would also help me reviewing all of this :)16:27
gibibauzas: yeah, rubasov already started it for neutron and I will chime in16:27
bauzasI'd be more than happy to review it16:28
gibimriedem: good question, I think the SRIOV + bandwidth support is now become virt driver specify16:28
gibibauzas: here is a neutron WIP doc patch https://review.openstack.org/#/c/640390/16:28
dansmithmriedem: hmm, aren't the two things in runway slot 2 and 3 part of the same effort?16:28
mriedemdansmith: yeah, spot instances16:28
dansmithyeah16:29
mriedemseparate blueprints for whatever reason16:29
dansmithhrm16:29
dansmiththey're part of the same stack in gerrit, so.. kinda seems like that's one thing:)16:29
gibimriedem: if there is a InstancePCIRequest then it will have the new tag included even if the compute host only has a single PF16:29
mriedemgibi: but maybe a compute without a supporting virt driver wouldn't configure the neutron agent to report that inventory16:29
artommriedem, I wasn't checking the database instance_numa_topology16:29
artomSo the pins are updated and correct16:29
gibimriedem: SRIOV without bw is kept intact16:30
artomBut Windriver are telling me the DB isn't updated, despite apply_migration_context() being called16:30
gibimriedem: so if no resource request in the sriov port then we don't need the special key from the virt driver16:30
mriedemartom: are you saving the instance changes after calling apply_migration_context()?16:30
mriedemb/c ^ doesn't persist the changes, it just updates the object fields16:31
artommriedem, wait, seriously?16:31
*** wolverineav has quit IRC16:31
artommriedem, huh, but instance.save() is still called after in post_live_migration16:31
mriedemdon't take my word for it16:31
bauzasgibi: if that's virt-specific, you should document it in https://docs.openstack.org/nova/latest/user/feature-classification.html16:31
gibibauzas: thanks for the pointer16:31
artommriedem, ah, you're right, it's not saved16:32
*** gyee has joined #openstack-nova16:32
mriedemi'm more worried that we send a server create request to a compute that doesn't support it16:32
mriedemand we don't have any kind of feature capability for this port requested resources stuff16:32
mriedemgibi: bauzas: ^16:32
mriedemand we just drop the request on the floor16:32
dansmithartom: so does that mean some testing gap exists because we never re-migrate the instance a second time or something?16:33
mriedemthinking about https://review.openstack.org/#/c/538498/ coincidentally16:33
bauzasmriedem: that's my -116:33
dansmithartom: meaning.. how could we be failing to save that and think that this works?16:33
artomBut there's an instance.save 5 lines after https://review.openstack.org/#/c/634606/37/nova/compute/manager.py@692516:33
gibimriedem: yeah if neutron agent is configured to report bandwidth but the nova virt driver does not support it then it will be NoValidHost after a bunch of reschedule16:33
bauzasmriedem: but the question is : should we leave it as docs, or enforce it ?16:33
artomdansmith, I was just testing the XML is updated correctly16:33
dansmithartom: oh :(16:33
artomdansmith, yeah, testing coverage gap :(16:34
mriedemgibi: bauzas: at this point it's probably best to just doc it and clean it up in train with a capability tag a la https://review.openstack.org/#/c/538498/16:34
*** cdent has joined #openstack-nova16:34
bauzasI'm good with that16:35
mriedemaspiers: efried: btw who is on the hook for documenting ^ now that it's going to be codified?16:35
gibibauzas, mriedem, sean-k-mooney, stephenfin: this is the cons of the auto detection of the pf_interface_name. It makes the feature virt driver specific. If we go with the a new tag in the passthrough_whitelist then the solution is virt driver agnostic16:35
*** imacdonn_ has joined #openstack-nova16:35
mriedembut then you have to configure more crap16:35
*** imacdonn_ is now known as imacdonn16:35
gibimriedem: exactly16:36
mriedemthis is pretty magical unicorn and i expect 99% of openstack users for nfv stuff are using libvirt anyway16:36
gibimriedem: I'm OK to add the capability in Train16:36
openstackgerritMerged openstack/nova stable/rocky: Fix race in test_volume_swap_server_with_error  https://review.openstack.org/64059516:36
mriedemdansmith: regarding the spot instance stuff i have skimmed some of it in the past (over a month ago) noting some obvious concerns, but haven't been back on it since, nor has anyone else from the core team it looks like, so that one is too risky at this point for me, especially without any integration (tempest) testing16:38
bauzasmriedem: gibi: I'm not opposed to have this feature be virt-driver specific for the moment and provide a new abstract trait later16:38
mdboothlyarwood: I was looking into your tempest failures but got distracted. Just getting back into it. Did you manage to look by any chance?16:38
dansmithmriedem: yeah, same, I was just skimming over it, thinking that since it was two pieces, maybe half of it was simple stuff we could merge, but it seems all tied together16:39
mdboothLooking at one of them in particular, I note that the type is deleted after a volume of that type is deleted, but before cinder logs that the delete completed. However, it could be, for eg, that cinder has already deleted the volume from the db but doesn't log until some backend stuff is done? I'm still looking. Looks weird.16:40
dansmithartom: so yeah, I dunno why that save() doesn't get you there, but the migration context stuff is complex, so you might just be tripping over some assumption16:40
dansmithartom: honestly, I'd start trying to reproduce (or disprove) that with a functional test before rattling a fix16:41
artomdansmith, yeah, first order is modifying my tests to catch that16:41
artom(Or writing new func tests for the same purpose)16:41
artomThen debugging16:41
dansmithartom: ++16:41
mriedembtw i do a lot of ^ type stuff in the cross-cell resize functional testing to make sure the db is what i expect it to be16:41
*** rpittau is now known as rpittau|afk16:42
mriedemsince i'm farting around with 2 dbs16:42
gibibauzas: cool, then your current -1 on the patch is just procedural while you reviewing the rest of the patch?16:42
*** helenafm has quit IRC16:43
artommriedem, I think the DB asserting part will be relatively easy, it's the running 2 services part that's finicky16:43
artomdansmith, while I have you, can we context switch to your thoughts on https://review.openstack.org/#/c/639396/?16:44
dansmithartom: ugh, I really don't want to... I know that is going to be ugly because of whatever has moved out of that scope, but I'm really (really) not in favor of changing how that works.. like REALLY.16:45
bauzasgibi: I left a new comment as a soft -116:46
gibibauzas: looking16:46
bauzasgibi: but I'm continuing to review so I can +1 once I'm done16:46
gibibauzas: sure16:46
bauzasgibi: just askign docd16:46
gibibauzas: I will let you go through it before I start updating the patch16:46
artomdansmith, yeah :/16:46
artomdansmith, as I said in the review, I don't see a way around *some* change, because we race on setting up the event waiter16:47
gibibauzas: would you accep the documentation in a separate patch?16:47
artomIf we don't get to the driver code by the time Neutron sends us the event, it's lost16:47
dansmithartom: yeah, but neutron is sending the event in response to things we're doing, which just means whatever we do to trigger it needs to happen within the context16:48
bauzasgibi: yup16:48
gibibauzas: coolio16:48
bauzasI can turn into a +1 if you prefer16:48
gibibauzas: don't worry, I will respin anyhow16:48
artomdansmith, yeah - but we're poking Neutron in the compute manager, and setting up the waiter in the virt driver...16:49
gibibauzas: just let me know if you spot any other issues16:49
dansmithartom: I understand16:49
artomFrom the virt driver, I should say. It's a call to the compute manager16:49
artomHrmm, there's no RPC, mind you16:49
artomJust... a whole bunch of intermediary steps16:49
*** wolverineav has joined #openstack-nova16:49
bauzasgibi: I'm on it16:50
gibibauzas: thanks a lot16:50
bauzasI wish I could help better, that's it16:50
artomdansmith, anyways, I'll try to come up with a least bad way of doing it, we'll have a strawman to pick apart16:51
artomAt least you're on board with the need for change ;)16:52
mriedemartom: so we likely need 2 context managers to wait for separate events right?16:54
*** wolverineav has quit IRC16:54
mriedemone for the port binding host change, and one for the vif plug16:54
mriedemthat port binding host change happens in several move operations we'd care about (at least resize and evacuate)16:54
mriedemand unshelve i think...16:54
artommriedem, hrmm, you lost me, how would 2 context managers solve this?16:54
mriedemcouldn't we create a helper method for that neutronv2.api method that updates the host in the port binding to register an event callback16:55
mriedemartom: we have 2 events to wait for yeah?16:55
mriedemso you wait for the first (port binding host change), then call the driver to spawn or whatever which does the vif plug16:55
mriedemwhich triggers the event we've always waited for16:55
artommriedem, uhh, I thought it was Nova that PUT'ed the post binding host change?16:56
mriedemit does16:56
artomAnd that got us a vif-plugged as a "response"?16:56
mriedemand then we get an event from neutron that we're not waiting for16:56
*** dpawlik has quit IRC16:56
mriedemand then we actually plug vifs which produces another event yeah?16:57
dansmithartom: was it not updating instance_extra.migration_context, or was it not persisting the applied-from-context instance attributes?16:57
artomWe are, sort us, we just start waiting after it gets to us, but yeah, you're basically right16:57
dansmithartom: I had assumed the former16:57
artomdansmith, the latter16:57
dansmithah16:57
dansmithartom: which attributes of those specifically?16:57
artomdansmith, so I haven't looked into it in detail, but I'm told the new NUMA topology16:58
dansmithnuma_topology?16:58
dansmithyou probably need to do something to make those dirty,16:58
artom...16:58
* artom nopes right out.16:58
dansmithsince they're nested objects, they probably only save if the object itself is dirty16:58
dansmithartom: you need to give them "the talk"16:58
artomRight, tell them they've been naughty and need to be saved from their sins?16:59
mdboothlyarwood: Ok, I can see in the tempest logs that tempest receives a 404 for GET volume *before* deleting the type, and I can see the same in the cinder logs. However, c-vol doesn't log the completion of the delete until some time later, which is *after* the attempted deletion of the type.17:00
dansmiththey must not be marked as changed in the migration context, so I'm thinking just applying them to the instance and then calling save doesn't have them included in obj_what_changed()17:00
openstackgerritMatt Riedemann proposed openstack/python-novaclient master: Microversion 2.71 - show server group  https://review.openstack.org/64065717:00
artommriedem, wait, plugging the vifs produces *another* event?17:00
artommriedem, I need to flow through the code again in that case, because it means I don't have a complete picture17:01
sean-k-mooneyartom: binding the vif produces a network-chaged event17:01
mriedemartom: that was the original reason for adding the context manager in the virt driver17:01
artomOh, so binding -> network-changed17:01
artomvig plugging -> vif-plugged17:01
mriedemi don't think that's correct17:01
sean-k-mooneywhen the neutron agent plugs the vif we get a network-vif-plugged event17:01
mriedemif they were different events, we wouldn't be missing the 2nd event17:01
* artom is hungry and can't think straight17:02
dansmiththe problem is the vif shows up on the destination, neutron notices and immediately sends the vif-plugged event right?17:02
mdboothmriedem: lyarwood was having trouble with the timestamp tests being flaky and removed them in https://review.openstack.org/#/c/639331/ . I vaguely recall you previously expressing an opinion on these tests.17:02
sean-k-mooneyi dont think we need to move the handeling of the network-chaged event17:02
dansmithbefore nova is ready to start it17:02
mriedemas far as i understand the problem, we change the port binding host and then plug vifs, which normally happens close enough that neutron sends a single network-vif-plugged event which we're waiting for by the time we plugged vifs and registered a callback in the virt driver17:02
sean-k-mooneydansmith: yep basically17:02
dansmithmriedem: and this is happening in the migrate case in a different place than it happens during spawn and that's the issue, right?17:03
sean-k-mooneyneutron sees the port on the sorce because we have not clean it up yet so when the rollback of the migration happens and we set the host back to the source neturon send the vif-plugged event right away17:03
dansmithoh, it's the rollback case only?17:04
artomdansmith, revert resize is the downstream bug17:05
artomHrmm, do we actually have 2 problems here?17:05
bauzasgibi: left yet another comment https://review.openstack.org/#/c/623543/3917:05
sean-k-mooneydansmith: artom is fixing the rollback case. there is another race with sdn controler that send vif-plugged when you bind the port that this will also fix17:05
dansmithokay, I'm even less concerned then :)17:05
bauzasgibi: can't we do all of this on the libvirt driver anyway?17:05
dansmiththe rollback case could just not wait for events since it's already on the source, I would think.. that would be a band-aid worth doing I would think17:06
bauzasgibi: but I guess it's not possible because you wanna claim() this17:06
gibibauzas: looking17:06
artomdansmith, that'd be a weird bit of into to pass down to the virt driver, no?17:07
artom... cuz we just call spawn there17:07
dansmithwe already pass a similar flag in places don't we?17:08
mriedemmdbooth: i don't really remember that, but i'm assuming that timestamp stuff was copied from other scenario tests, and the commit message doesn't really explain the problem with the timestamp stuff17:08
mriedemjust "it's bad" isn't very good17:08
artomdansmith, I'd have to check17:08
dansmithmaybe I'm thinking about within the driver, but still17:08
mriedemdansmith: upstream it's at least known to be happening in evacuate17:08
mriedemhttp://status.openstack.org/elastic-recheck/#181378917:08
gibibauzas: yeah we need the info to drive the pci claim17:09
dansmithmriedem: on evacuate, not evacuate and then restart-on-original-host right?17:09
sean-k-mooneyya i was going to say the same. but dan is right we could skip waiting in this case but i kindo of like not have another flag to pass down17:09
*** psachin has quit IRC17:09
gibibauzas: does compute/manager have access to the ProvideTree?17:09
bauzasgibi: but you understand my concern, right?17:09
dansmithmriedem: the evac case seems ripe for just moving where we wait, but I haven't gone digging17:09
sean-k-mooneythe flag would only fix the rollback case. artom change would fix the cisco aci/old odl case17:09
bauzasgibi: that's the problem, I don't think so17:09
gibibauzas: I understand that we call placement a lot17:09
sean-k-mooneyin addtion to rollbacks17:09
*** wolverineav has joined #openstack-nova17:09
dansmithartom: thinking of this:                                    vifs_already_plugged=False,17:10
bauzasgibi: and say you have two pci requests for the same interface...17:10
mriedemgibi: bauzas: pretty sure the compute manager has a handle to the provider tree,17:10
bauzasit's a bit sad to query it twice17:10
mriedembecause it has to clear it on sighup17:10
*** tbachman has quit IRC17:10
mriedemdansmith: i don't know the answer to that offhand17:10
gibimriedem: ooh17:10
artomdansmith, ah, OK. It's still only a bandaid tho...17:10
* dansmith has to jump on a call17:10
mriedemgibi: bauzas: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L55117:10
artomdansmith, thanks, that'll allow me to eat :P17:11
gibibauzas: sure your concern is valid either I would like to use a ProviderTree or add a local cache17:11
gibimriedem: looking17:12
mdboothmriedem: The purpose of the timestamp stuff IIUC is to assert that the contents of the volume are preserved across the migration. I personally think it's an important test, but if it's problematic in practise we'd need to be pragmatic, e.g. removing or guarding it. NP if you don't have a strong opinion, lyarwood and I can fling fedoras at each other internally :)17:12
gibibauzas, mriedem: so here in the compute/manager we only need the name of the RP and we know the UUID of it. So I guess I should be able to use the ProvideTree cache for that17:13
mriedemwell, give a justifiable reason in the commit message to remove it17:13
mriedemneed another core on the 2.71 novaclient change https://review.openstack.org/#/c/640657/ so we can get that into the final release17:14
bauzasgibi: the cache is a non-public information from the reportclient17:14
bauzasgibi: at the moment, I mean17:14
bauzasmriedem: on it17:15
gibibauzas: but I call reportclient anyhow so I can call something else on the reportclient that uses the internal cache17:15
*** agopi has joined #openstack-nova17:15
gibibauzas: it I fail then I will add a local cache to my code to query each RP once17:16
bauzasgibi: sure, I'm just saying that if you wanna use its cache, you have to make its attribute accessible17:16
bauzaseither thru a getter or whatever17:16
*** wolverineav has quit IRC17:16
gibibauzas: I will go with a new method (if not exists already) in reportclient that reads an RP from the cache17:17
bauzasI dislike at most accessing internal attributes in python17:17
bauzaseven if that's allowed17:17
bauzasgibi: fine with me17:17
bauzasmy point being "please don't just access self.reportclient.provider_tree" or whatever it's named17:18
*** markvoelker has joined #openstack-nova17:18
*** tbachman has joined #openstack-nova17:19
*** Sundar has quit IRC17:20
gibibauzas: understood17:20
bauzassorry but I have to bail out17:20
bauzas++17:20
*** awalende has quit IRC17:21
*** sdake has joined #openstack-nova17:22
*** awalende has joined #openstack-nova17:22
gibibauzas: thanks a lot17:24
gibibauzas: I will bail soon as well17:24
gibibauzas: I will finish updating the patch tomorrow morning17:25
*** awalende has quit IRC17:26
*** dtantsur is now known as dtantsur|afk17:29
openstackgerritStephen Finucane proposed openstack/nova master: fakelibvirt: Add ability to generate fake PCI devices  https://review.openstack.org/64040917:29
openstackgerritStephen Finucane proposed openstack/nova master: tests: Create PCI tests for NUMA'y tests  https://review.openstack.org/64082617:29
*** dpawlik has joined #openstack-nova17:30
openstackgerritChris Friesen proposed openstack/nova master: Flavor extra spec and image properties validation  https://review.openstack.org/62070617:31
cfriesenmriedem: this addresses almost all your comments.  will split out the hardware.py stuff as you asked in the next revision17:32
*** dpawlik has quit IRC17:34
*** wolverineav has joined #openstack-nova17:46
* gibi leaves for today17:47
*** mvkr has quit IRC17:48
*** mrch_ has joined #openstack-nova17:51
*** markvoelker has quit IRC17:51
openstackgerritChris Dent proposed openstack/nova master: Correct lower-constraints.txt and the related tox job  https://review.openstack.org/62297217:56
openstackgerritKashyap Chamarthy proposed openstack/nova master: qemu: Make disk image conversion dramatically faster  https://review.openstack.org/64078117:57
*** tssurya has quit IRC17:58
mriedemcfriesen: wouldn't a deactivate image (even if it doesn't change) on resize fail the resize if the dest host can't download the image (and it's not cached)?17:58
cfriesenmriedem: good question.  I seem to remember some scenarios where it copies it from the cache on the source...but let me check.18:00
mriedemmaybe if you're using shared storage?18:04
mriedembut not everyone does18:04
*** derekh has quit IRC18:05
mriedemi believe the libvirt driver code falls back to pull from the source if it's not in the dest image cache18:05
mriedemor tries to18:05
mriedemmdbooth probably knows better18:05
* mriedem runs to lunch, bbl18:05
*** mriedem has quit IRC18:05
*** wolverineav has quit IRC18:06
*** ociuhandu has joined #openstack-nova18:11
*** wolverineav has joined #openstack-nova18:14
aspiersgah, looks like functional tests for https://review.openstack.org/#/c/538498/ failed in the gate18:18
aspiershttp://logs.openstack.org/98/538498/22/gate/nova-tox-functional/d76478b/testr_results.html.gz has a bunch of errors like "oslo_config.cfg.NoSuchOptError: no such option sync_on_startup in group [placement_database]"18:18
aspiersdoes that make sense to anyone18:18
aspiers?18:18
*** dpawlik has joined #openstack-nova18:19
*** Sundar has joined #openstack-nova18:22
*** ralonsoh has quit IRC18:22
* aspiers finds https://github.com/openstack/placement/commit/da36ad16e15181eb8b72f04cb3b90f8009411c23 and looks at cdent and efried18:23
*** dpawlik has quit IRC18:24
*** whoami-rajat has quit IRC18:24
cdentaspiers: that's interesting, it should default to false18:24
cdentexcept that's not what the error says, it says the conf option doesn't exist18:25
aspiersright18:25
aspierstotally weird that this should appear in the gate but not in the check18:25
cdentwhich means the gate doesn't have the right code...18:25
sean-k-mooney aspiers ist it posble that it was usign the inrepo placement?18:25
sean-k-mooneyit does not exist here https://github.com/openstack/nova/blob/master/nova/conf/placement.py18:26
cdentsean-k-mooney: the fix should be using an indepdent conf18:27
cdentand according to the failures it's using the external placement18:27
cdenthowever18:27
cdentit might be that the fixture needs to be tweakd.18:27
cdentaspiers: if you have the time can you try to run the functional tests locally on a clean tox?18:28
* cdent tries too18:28
aspierscdent: sure18:28
sean-k-mooneywell looking at code search it only exists in placement http://codesearch.openstack.org/?q=sync_on_startup&i=nope&files=&repos=18:28
aspiersright18:28
aspierscdent: but the zuul check already did that and passed ...18:29
cdentaspiers: yes, but that change (to add that config) merged to placement several minutes ago so maybe it wasn't there in check18:30
*** tesseract has quit IRC18:30
sean-k-mooneycdent: if it only merged a few minuts ago its not in pypi18:31
sean-k-mooneythe last relase was feb 13th https://pypi.org/project/osc-placement/#history18:31
sean-k-mooneysorry wrong repo18:32
sean-k-mooneyjanuary 7th https://pypi.org/project/openstack-placement/#history18:32
cdentsean-k-mooney: it should be installed a tox-sibling, not from pypi18:32
cdentotherwise it wouldn't work as a fixture-to-use-master18:32
cdentaspiers: i've repeated here18:32
cdentso I reckon it's a bug in the fixture18:33
sean-k-mooneyperhaps in which case then ya it could be jsut a a timing window18:33
aspiersah18:33
sean-k-mooneyaspiers: do you have a depends on on that placmenet change18:33
sean-k-mooneyaspiers: that would have prevented any race18:33
aspiersno I didn't know about the change until just now18:33
cdentsean-k-mooney: not in the case18:33
cdentthe fixture has a real bug18:33
sean-k-mooneycdent: oh ok18:33
sean-k-mooneyaspiers: generally depens on helps bug not in all cases18:34
aspiersis it possible to do rechecks on the gate?18:34
sean-k-mooneyyep18:34
sean-k-mooneyjust a normal recheck18:34
sean-k-mooneywe used to have reverify years ago but not anymore18:34
sean-k-mooneyreverify just ran gate18:35
sean-k-mooneyrecheck will run check and gate18:35
cdentaspiers: I'll see about fixing this18:38
cdentas in, now18:38
sean-k-mooneyim going to grab dinner. be back in an hour or so18:39
*** panda|ruck is now known as panda|ruck|off18:41
*** markvoelker has joined #openstack-nova18:48
cdentah. aspiers, kind of figured it out. Those tests which are failing are using the placement database fixture directly, not through the Placement fixture, thus the the config opt is not registered, leading to plenty of  confusion18:49
cdentaspiers: https://bugs.launchpad.net/nova/+bug/181856018:53
openstackLaunchpad bug 1818560 in OpenStack Compute (nova) "Nova's use of the placement database fixture from test_report_client doesn't register opts" [Critical,Confirmed]18:53
cdentin case I wasn't clear there, the fix is on the placement side18:53
*** pcaruana has quit IRC18:56
*** whoami-rajat has joined #openstack-nova18:58
*** jlvillal has quit IRC18:58
*** jlvillal has joined #openstack-nova18:59
*** IvensZambrano has quit IRC19:01
*** mdbooth_ has joined #openstack-nova19:04
*** mdbooth has quit IRC19:07
*** sdake has quit IRC19:07
*** mriedem has joined #openstack-nova19:11
cdentaspiers: there are two levels of problem here19:15
*** wolverineav has quit IRC19:15
*** wolverineav has joined #openstack-nova19:16
cdentrequires fixes in both placement and nova, mabye19:16
cdentstill looking to try to figure out the cleanest way19:16
*** agopi has quit IRC19:18
*** agopi_ has joined #openstack-nova19:18
aspierscdent: awesome, thanks!19:19
mriedemdansmith: am i missing what you're asking for here? https://review.openstack.org/#/c/636412/1/nova/scheduler/client/report.py@175819:19
dansmithmriedem: ah, sorry I missed that the original change was already including one direction19:20
*** markvoelker has quit IRC19:21
mriedemi love one direction19:21
aspiersX-D19:21
cfriesenmriedem: in the case of resize, I think the fact that sometimes it *can* go through (if the image is already cached, or we're using "raw") means that we can't fail the operation due to image inactivity.19:22
mriedemcfriesen: sure19:22
mriedemnot a huge sticking point for me19:22
*** awalende has joined #openstack-nova19:23
*** openstackgerrit has quit IRC19:23
*** openstackgerrit has joined #openstack-nova19:25
openstackgerritChris Friesen proposed openstack/nova master: Flavor extra spec and image properties validation  https://review.openstack.org/62070619:25
openstackgerritChris Friesen proposed openstack/nova master: Improve existing flavor and image metadata validation  https://review.openstack.org/64084119:25
openstackgerritChris Friesen proposed openstack/nova master: Flavor extra spec and image properties validation from API  https://review.openstack.org/62070619:26
cfriesenfigured I'd differentiate the two titles a bit better19:27
eanderssonnova-hostname (from metadata) is set by dhcp_domain right?19:27
eanderssonThat is deprecated19:27
eanderssonWhat is supposed to populate that field?19:27
*** awalende has quit IRC19:28
mriedemeandersson: you just stepped into something19:28
mriedemeandersson: https://bugs.launchpad.net/nova/+bug/169801019:29
openstackLaunchpad bug 1698010 in OpenStack Compute (nova) "neutron-based instances should not use the nova-network 'dhcp_domain' option" [High,In progress] - Assigned to Stephen Finucane (stephenfinucane)19:29
sean-k-mooneyeandersson: if you dont have designate deployed i dont think neutron will populate that value.19:30
sean-k-mooneyignoring the nova bug19:30
melwittdansmith, mriedem: I updated the user_id patch to make the field non-nullable. unit tests are all broken and some/all of the next patch need to be squashed in, but the next patch passes grenade/tempest. I hoped for a sanity check from you before I go ahead and do any commit squashing and unit test fixing https://review.openstack.org/63335019:32
cdentaspiers: you should be proud, you've found a hole in the space time continuium, I'm looping back on myself19:32
aspierscdent: just don't destroy the universe in the process19:33
*** eharney has quit IRC19:33
cdentefried: are you back yet? I think I might need to consult with you on this19:35
cdentthe issue I'm trying to contend with this that at https://github.com/openstack/nova/blob/master/nova/tests/functional/test_report_client.py#L124 that CONF is a nova conf, not a placement CONF19:36
cdentso now that we've added a new conf item to placement things are awry19:36
openstackgerritAdam Spiers proposed openstack/nova master: Parse <emulator> elements from virConnectGetCapabilities()  https://review.openstack.org/64048319:40
mriedemmelwitt: it looks like what i'd expect19:42
melwittmriedem: thank ya19:43
mriedemjackding: are you able to rebase this? https://review.openstack.org/#/c/603844/19:44
mriedemjaypipes: ^ i think i answered your question in there19:44
dansmithmelwitt: I gotta run to an appt but will try to look when I get back19:45
melwittdansmith: ok, thank you19:45
jackdingmriedem: will do19:46
eanderssonthanks mriedem sean-k-mooney19:49
cfriesenjaypipes: FYI, this was based on an instance seen in the wild.19:50
*** agopi_ is now known as agopi19:50
cfriesenjaypipes: (re: your question in jackding's review)19:51
openstackgerritJack Ding proposed openstack/nova master: Correct instance port binding for rebuilds  https://review.openstack.org/60384419:53
jackdingmriedem: rebase done. Thanks19:54
mriedemcfriesen: real soft -1 on https://review.openstack.org/#/c/640841/19:56
mriedemjust update an existing api unit test and i'm +219:56
cfriesencool, will respin19:56
jaypipescfriesen: that's why I didn't -1 it...20:00
jaypipescfriesen: I wasn't 100% sure about it.20:00
*** thgcorrea has quit IRC20:00
jaypipesjackding: I've +2'd 603844 since mriedem was +2 on it previously.20:04
*** Sundar has quit IRC20:06
eanderssonSomething got a lot more expensive for us after moving to Rocky20:08
eanderssonIt looks like Neutron is getting a lot more calls20:08
*** jmlowe has quit IRC20:08
eanderssonbut difficult to say what is causing cpu load to increase so drastically20:08
eanderssonWe upgraded Nova, Neutron and Cinder20:09
eanderssonand CPU load has gone up 5x20:09
*** wolverineav has quit IRC20:09
eanderssonAnyone seen this?20:09
sean-k-mooneyeandersson: are you using OSP?20:09
eanderssonWhat is OSP?20:09
sean-k-mooneyredhat openstack plathform20:10
eanderssonnah20:10
eanderssonRunning something closer to Kolla20:10
sean-k-mooneyok we had a perfromce issue  awhile ago that was related to not enableing memcahce to cache keystone auth tokens the slowed everything down20:10
eanderssonWe do have that enabled20:14
openstackgerritChris Dent proposed openstack/nova master: Use a placement conf when testing report client  https://review.openstack.org/64085320:14
eanderssonAssuming you don't need that con computes right?20:14
sean-k-mooneythe compute nodes do hit the neutron and cinder apis20:14
sean-k-mooneyi dont know if you need/should configure caching on those too20:15
cdentaspiers: https://review.openstack.org/#/q/topic:bug/1818560 should fix it. mriedem: that's a gate blocker fix for functional/test_report_client20:15
mriedemis that just preventing https://review.openstack.org/#/c/538498/ from merging or are all nova changes now broken?20:16
cdentmriedem: I think all20:16
mriedemwhat changed/regressed?20:17
cdentthe fact that nova's conf was being used to run a placement wsgi app was revealed20:17
cdentby the fact that we added a config on the placement side20:17
mriedemlink to the placement change?20:18
cdentso it was going to happen as soon as that divergence happened and it just so happens it was today20:18
*** markvoelker has joined #openstack-nova20:18
cdentmriedem: it's linked from both commit messages at the topic above20:18
mriedemhttps://review.openstack.org/#/c/619050/20:18
mriedemok i was looking at the bug20:18
cdentthe bug description wasn't quite right, because I was trying to be hasty20:19
*** dpawlik has joined #openstack-nova20:20
cdentI've udpated the title and description20:23
efriedcdent: I'm back now. How can I help?20:23
cdentefried: I figured it out, but if you could review https://review.openstack.org/#/q/topic:bug/1818560  to confirm I figured it out the right way, that would be good20:24
*** dpawlik has quit IRC20:25
efried...20:25
mriedemcdent: comments inline but if it passes it passes20:28
mriedemwe likely need to talk about test coverage for the externally-used placement fixture20:29
mriedem*at some point20:29
cdentyes20:30
cdentbut I figured at this stage of the game it was better to just fix it20:30
mriedemyeah i agree20:30
openstackgerritChris Friesen proposed openstack/nova master: Improve existing flavor and image metadata validation  https://review.openstack.org/64084120:30
sean-k-mooneymriedem: cdent i dont know if we could somehow run the nova functional tests on change to the placement repo or perhaps just that fixture in the placmeent repo?20:34
efriedcdent: ugh, I had wanted to get rid of either PlacementFixture or PlacementDirect, and now we've made them inextricably interdependent.20:34
cdentefried: we already had both and we're trying to make a quick fix20:35
cdentthat doesn't remove the possibility of fixing it better late20:35
efriedyeah, I get it.20:35
cdentI personally think we should get rid of PlacementDirect, probably by exposing the intercept functionality in PlacementFixture20:35
openstackgerritMatt Riedemann proposed openstack/nova master: Flavor extra spec and image properties validation from API  https://review.openstack.org/62070620:35
cdentPlacementDirect is for a _very_ specific thing, in which a placement conf (a real one) is expected to exist20:36
cdenthowever I have seen zero evidence that PlacementDirect ever should have existed, so I'm not inclined to put a lot of energy into it20:36
cdent(for now)20:36
cdents/get rid of/get rid of (on the nova side)/20:37
sean-k-mooneyby the way i assume the current lag in jobs starting in the gate is jsut the usual 3 days to feature freeze lag?20:37
cfriesenno, zuul has a personal grudge20:38
melwittspeaking of test coverage, mnaser: how goes it with the OSA job patch? https://review.openstack.org/63585220:38
mriedembtw http://status.openstack.org/elastic-recheck/data/integrated_gate.html "Generated at: 2019-02-08" so we don't know the true status of the gate right now20:38
mriedemclark is investigating20:39
*** BlackDex has quit IRC20:39
sean-k-mooneycfriesen: :) first jenkins and now zuul well at least the ci is consitnt20:39
mnasermelwitt: that test should be working and pulling down nova20:39
mnaserwe could probably add it as non voting20:39
mnaseri need to find time to integrate os_placement into the main tests/roles but20:39
mnasera lot of our cores were laid off so we're struggling in osa land </320:39
melwittoh :(20:40
melwittI'm sorry to hear that20:40
cfriesenThis is a standalone disk performance optimization patch...has +1 from mdbooth.  https://review.openstack.org/#/c/616692/20:40
mnaseryeah, so just trying to keep things afloat, the placement role is ready but ill make sure we get a job that continues to deploy using in-repo placement, and another one that deploys using out-of-repo20:40
melwittmnaser: ok, I wasn't sure what the job in that patch is missing/why WIP20:41
mnasermelwitt: i wanted to leave it that way in case someone was to make an OSA change that starts deploying with out-of-repo by default and then the job becomes redundant20:41
mnaseri need to put sometime to add an extra tox target to deploy with in-repo placement, which will ensure we don't accidentally start testing from out-of-tree when we switch the default20:42
melwittoh, I see20:43
*** tbachman has quit IRC20:43
mnaseri'll get around that eventually20:44
mnaser(and i can help someone if they want to help do it as well)20:44
melwittok, and that's different than the TODO here? https://review.openstack.org/#/c/635852/1/.zuul.yaml@21120:45
mnasermelwitt: oh no that's exactly that todo :P20:46
*** jmlowe has joined #openstack-nova20:46
melwitt"without placement" confused me bc whether it's in-repo or out, I thought that's still "with placement" :P20:47
mnaserpretty much involves creating a new tox.ini target in openstack/openstack-ansible-os_nova and creating an extra var like "nova_use_in_repo_placement=True"20:47
mnaseryeah that should say "without os_placement" (which is our extra role)20:47
melwittahhh20:47
melwittand the role is already merged? I can try the tox target thing but I have no clue what I'm doing20:48
*** BlackDex has joined #openstack-nova20:48
*** markvoelker has quit IRC20:51
*** sdake has joined #openstack-nova20:51
openstackgerritChris Dent proposed openstack/nova master: Use a placement conf when testing report client  https://review.openstack.org/64085320:55
*** dpawlik has joined #openstack-nova20:55
cdentmriedem, efried: it will soon be past my bedtime, if you need/want more changes to that ^ and it's partner please feel free20:56
efriedack20:56
*** IvensZambrano has joined #openstack-nova20:57
efriedcdent: what's up with the merge failure?20:57
cdentit's because I submitted the dependent change at the same time20:58
openstackgerritKashyap Chamarthy proposed openstack/nova-specs master: cpu-model-selection: s/cpu_models/cpu_model_list/  https://review.openstack.org/64086620:59
*** dpawlik has quit IRC21:00
*** wolverineav has joined #openstack-nova21:02
*** whoami-rajat has quit IRC21:04
*** awaugama has quit IRC21:14
*** cdent has quit IRC21:25
*** eharney has joined #openstack-nova21:28
*** mvkr has joined #openstack-nova21:29
mriedemcfriesen: some questions in https://review.openstack.org/#/c/620706/21:31
mriedemcfriesen: happy to discuss here21:31
cfriesenchecking21:31
openstackgerritguang-yee proposed openstack/nova master: pass endpoint interface to Ironic client  https://review.openstack.org/64087921:31
openstackgerritguang-yee proposed openstack/nova master: pass endpoint interface to Ironic client  https://review.openstack.org/64087921:33
cfriesenmriedem: I've got a change queued up for get_pci_requests_from_flavor.  can add to existing or do as followup.  Happy to split out PciInvalidAlias.21:34
cfriesenrest of comments look reasonable, will address later today.21:35
mriedemcfriesen: i'd split as much as possible for review sanity21:35
*** hemna has joined #openstack-nova21:35
cfriesenthe get_pci_requests_from_flavor change is pretty straightforward, just hadn't pushed it yet since I was expecting to respin this.21:36
*** sdake has quit IRC21:41
*** sdake has joined #openstack-nova21:43
openstackgerritMatt Riedemann proposed openstack/nova master: doc: cleanup pci.alias references  https://review.openstack.org/64088221:43
*** markvoelker has joined #openstack-nova21:49
*** mriedem has quit IRC21:50
*** tbachman has joined #openstack-nova21:51
*** rnoriega has quit IRC21:52
*** klindgren_ has joined #openstack-nova21:52
*** rnoriega has joined #openstack-nova21:52
*** klindgren has quit IRC21:53
*** smcginnis has quit IRC21:53
*** zzzeek has quit IRC21:54
*** zzzeek has joined #openstack-nova21:54
openstackgerritMerged openstack/nova master: Add nits from Id2beaa7c4e5780199298f8e58fb6c7005e420a69  https://review.openstack.org/64072921:58
*** wolverineav has quit IRC22:03
*** wolverineav has joined #openstack-nova22:06
*** betherly has joined #openstack-nova22:07
*** wolverineav has quit IRC22:11
*** betherly has quit IRC22:12
sean-k-mooneyfrom the description of https://review.openstack.org/#/c/603844/ im not sure if its needed due to https://review.openstack.org/#/c/634276/22:14
sean-k-mooneyit look like we are fixign the same or related bugs in different ways on both the nova and neutron side.22:15
sean-k-mooneythat said i need to review https://review.openstack.org/#/c/603844/ properly22:15
*** wolverineav has joined #openstack-nova22:18
*** markvoelker has quit IRC22:21
*** Sundar has joined #openstack-nova22:29
*** zhubx has quit IRC22:29
*** zhubx has joined #openstack-nova22:30
openstackgerritEric Fried proposed openstack/nova master: Stop using PlacementDirect  https://review.openstack.org/64088722:31
efriedmriedem, cdent: A different solution ^22:31
sean-k-mooneyefried: ignore whitespacce helps alot22:38
efriedsean-k-mooney: Indeed.22:39
efriedAre you telling me that having seen my comment already?22:39
sean-k-mooneyinitally i tought you rewrote most of the file. and yes i was about to ask if they were fucntuionally the same after then i made that change looked and commented on irc22:40
sean-k-mooneyso basically you removed the _interceptor context manager which used placement direct and instead repace teh client with  the versionCheckingReportClient after enabling the placemetnFixture22:43
*** agopi has quit IRC22:44
*** sdake has quit IRC22:50
sean-k-mooneyefried: im more or less trusting that you ran the fucntional test locally with master placement and it worked but ya droping placement direct form the fuctional test makes sense.22:51
sean-k-mooneyis there a similar followup to remove placement direct form placement itself or is there a usecase to keep it in placement22:52
efriedsean-k-mooney: I ran those two suites locally. I didn't run the whole world.22:52
efriedletting zuul do that.22:52
efriedWe don't know whether there's a real use case, cdent and I were talking about that (again) earlier.22:52
sean-k-mooneyhehe ya thats resonable :) zull is rather slow currently22:52
efriedIt was deemed a hard requirement for something-or-other (we can't remember exactly what) for extraction, so we did it.22:52
efriedbut at this point it's not clear that it's needed at all.22:53
*** slaweq has quit IRC22:53
efriedI want to say maybe it was going to be a prerequisite for cinder using it in certain cases22:53
efriedbut then in denver they decided they weren't going to use placement for those cases22:53
efriedso it's possible it will never be needed.22:53
efriedcdent did point out that it would probably be expected to work better from "real deployment" than from test22:54
*** smcginnis has joined #openstack-nova22:54
sean-k-mooneyplacement direct just allows you to bypass the placement api is thtat correct?22:54
efriedanyway, at this point I think the plan is to let it languish, see if anyone decides they want/need it, and if/when they do, find out if it's broken for whatever problem they're trying to solve.22:54
efriedsean-k-mooney: You're not bypassing it. It behaves as a real placement service. It just doesn't go "over the wire" and uses an in-memory database (I think).22:55
efriedsean-k-mooney: It really does effectively the same thing as PlacementFixture, just in a slightly different way.22:55
sean-k-mooneyoh i tougth you could use placemetn direct to talk to a real database too22:55
efriedI don't understand all the "wsgi intercept" magic.22:55
*** tbachman has quit IRC22:55
efriedoh, yeah, I think you can.22:55
efriedin this case we didn't want to do that, because we wanted to use a fixture.22:56
*** dpawlik has joined #openstack-nova22:56
*** tkajinam has joined #openstack-nova22:56
sean-k-mooneyso addmitly flawed understading was it was a way to run a placement api service in process to directly acess and maniupluate placement without needing to go over the wire22:56
sean-k-mooneye.g. just skip the http transprot part22:57
efriedyes22:57
efriedsorry, I was focused in on the change I just made, where I believe we were faking the database as well.22:57
sean-k-mooneyoh ya in the fuctional test i woudl expect us to be using a fake deb or a per test sqlite db22:58
*** tbachman has joined #openstack-nova22:58
efriedsean-k-mooney: Yeah, https://review.openstack.org/#/c/640887/1/nova/tests/functional/test_report_client.py@a9522:58
efriedmocking the db22:58
sean-k-mooneyright makes sense.22:59
*** dpawlik has quit IRC23:00
*** slaweq has joined #openstack-nova23:01
*** luksky has quit IRC23:02
*** slaweq has quit IRC23:05
*** Sundar has quit IRC23:12
*** sdake has joined #openstack-nova23:12
*** mlavalle has quit IRC23:16
*** betherly has joined #openstack-nova23:16
*** markvoelker has joined #openstack-nova23:18
*** betherly has quit IRC23:20
*** openstackgerrit has quit IRC23:28
*** tosky has quit IRC23:28
*** marst has quit IRC23:32
*** sdake has quit IRC23:35
*** dpawlik has joined #openstack-nova23:37
*** dklyle has quit IRC23:39
*** dpawlik has quit IRC23:41
*** mriedem has joined #openstack-nova23:43
mriedemefried: i'm +2 on cdent's gate fixes https://review.openstack.org/#/q/topic:bug/1818560+(status:open+OR+status:merged) - i see you have an alternative but i can't grok that right now23:45
mriedemand we could do your alternative later yeah?23:45
sean-k-mooneymriedem: looking at cdents change yes there woudl be no real conflcit with merging that quick and then efried change23:46
*** macza has quit IRC23:48
sean-k-mooneyefried: change looks resoable to me as it just drops teh use of placement direct an swaps to useing the palcement fixture instead but i also dont review the report clinet stuff often so im +1 but i also am not that familar those tests23:48
sean-k-mooneymriedem: ^23:49
*** eharney has quit IRC23:51
sean-k-mooneymriedem: for what its work like at the live ci log it just passed all the functional test and should report back in about 3-4 mins23:51
*** markvoelker has quit IRC23:51
sean-k-mooneyhttp://logs.openstack.org/87/640887/1/check/nova-tox-functional/1ba43e5/testr_results.html.gz23:51
sean-k-mooneythat ^ is the result of efried alternitve23:52
mriedemthe gate is broken and it's 6pm and i literally do not want to spend the time reviewing a new change23:52
*** wolverineav has quit IRC23:52
mriedemi only logged back on to +2 those changes23:52
mriedemand now i'm dropping of23:52
mriedem*off23:52
mriedemo/23:52
*** mriedem has quit IRC23:52
sean-k-mooneymriedem: sure  night o/23:53
*** IvensZambrano has quit IRC23:53

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