Friday, 2018-11-23

*** erlon has quit IRC00:00
*** sambetts_ has quit IRC00:12
*** sambetts_ has joined #openstack-nova00:13
*** cdent has quit IRC00:15
*** ileixe has joined #openstack-nova00:55
openstackgerritArtom Lifshitz proposed openstack/nova-specs master: Re-propose numa-aware-live-migration spec  https://review.openstack.org/59958700:56
*** brinzhang has joined #openstack-nova01:21
openstackgerritYikun Jiang proposed openstack/nova master: Remove live_migration_progress_timeout config  https://review.openstack.org/61914201:23
alex_xustephenfin: do you know whether we support resize for numa topology stuff? I didn't see where we update the numa topo in req_spec with new flavor when resize02:00
*** cfriesen has quit IRC02:12
*** ileixe has quit IRC02:22
*** ileixe has joined #openstack-nova02:24
*** owalsh has quit IRC02:25
*** mrsoul has joined #openstack-nova02:27
*** owalsh has joined #openstack-nova02:39
*** mhen has quit IRC02:46
*** tonyb has quit IRC02:49
*** mhen has joined #openstack-nova02:50
*** whoami-rajat has joined #openstack-nova02:52
*** tonyb has joined #openstack-nova02:52
*** cfriesen has joined #openstack-nova03:05
*** psachin has joined #openstack-nova03:12
*** udesale has joined #openstack-nova03:13
*** ileixe has quit IRC03:58
*** ileixe has joined #openstack-nova03:59
*** ileixe has quit IRC04:00
*** janki has joined #openstack-nova04:29
*** ivve has joined #openstack-nova04:35
*** sridharg has joined #openstack-nova05:04
*** psachin has quit IRC05:13
*** ileixe has joined #openstack-nova05:14
*** psachin has joined #openstack-nova05:14
*** abhishekk has joined #openstack-nova05:33
*** tonyb[m] has joined #openstack-nova05:36
*** yikun has quit IRC05:41
*** pcaruana has quit IRC05:44
*** ratailor has joined #openstack-nova05:45
*** janki has quit IRC05:55
*** janki has joined #openstack-nova05:56
*** ralonsoh has joined #openstack-nova06:23
openstackgerritYikun Jiang proposed openstack/nova master: Remove live_migration_progress_timeout config  https://review.openstack.org/61914206:32
openstackgerritMerged openstack/nova stable/rocky: Imported Translations from Zanata  https://review.openstack.org/61475706:49
*** cfriesen has quit IRC06:57
*** david-lyle has joined #openstack-nova07:03
*** dklyle has quit IRC07:04
*** pcaruana has joined #openstack-nova07:08
*** rmart04 has joined #openstack-nova07:11
*** rmart04 has quit IRC07:19
*** Luzi has joined #openstack-nova07:23
*** dtantsur|afk is now known as dtantsur07:38
openstackgerritYongli He proposed openstack/nova-specs master: add 'show-server-group' spec  https://review.openstack.org/61225507:47
*** rcernin has quit IRC07:48
*** slaweq has joined #openstack-nova07:49
openstackgerritYongli He proposed openstack/nova-specs master: add spec "show-server-numa-topology"  https://review.openstack.org/61225607:51
*** rmart04 has joined #openstack-nova07:59
*** yonglihe has joined #openstack-nova08:01
*** jangutter has joined #openstack-nova08:05
*** rmart04 has quit IRC08:11
*** sridharg has quit IRC08:13
*** awalende has joined #openstack-nova08:14
*** sridharg has joined #openstack-nova08:17
*** helenaAM has joined #openstack-nova08:27
*** brault has joined #openstack-nova08:58
*** mdbooth has joined #openstack-nova09:01
openstackgerritZhenyu Zheng proposed openstack/nova master: Bump compute service to indicate attach/detach root volume is supported  https://review.openstack.org/61475009:01
*** mdbooth has quit IRC09:03
*** mdbooth has joined #openstack-nova09:11
*** cdent has joined #openstack-nova09:16
*** k_mouza has joined #openstack-nova09:17
*** alexchadin has joined #openstack-nova09:19
*** erlon has joined #openstack-nova09:24
*** alexchadin has quit IRC09:29
lyarwoodmdbooth: morning, can I bug you about fake_imagebackend this morning?09:37
mdboothlyarwood: You may09:37
* mdbooth is at home without his external monitors today, btw, waiting for a tradesman09:38
lyarwoodmdbooth: cool, so, https://github.com/openstack/nova/blob/master/nova/tests/unit/virt/libvirt/fake_imagebackend.py#L124-L183 and the way it's currently returning a func over an actual class09:39
lyarwoodmdbooth: do you recall why you wired it up this up instead of returning some fake Image class?09:39
mdboothHave I got a patch up which changes that?09:39
mdboothThe issue, iirc, is that you'd need to have a closed class09:39
mdboothi.e. the function it returns is a closure09:40
lyarwoodahhhhh09:40
mdboothIf you returned a class, the class would also have to be a closure over the same data09:40
mdboothHowever, I vaguely recall I may have a patch which touches this09:40
lyarwoodright, so I have that WIP change on top of another series from you where I'm attempting to use the static is_shared_block_storage method from the image classes that I obviously can't when it's a func09:41
mdboothHmm, if I do I can't see it right now09:41
mdboothI'm *convinced* I've written a patch for this already09:42
*** alexchadin has joined #openstack-nova09:43
mdboothIf only we could land cleanup changes in non-geological time I wouldn't lose track of this stuff.09:43
*** tssurya has joined #openstack-nova09:43
*** derekh has joined #openstack-nova09:44
mdboothlyarwood: I'm pretty sure that whenever I wrote it, it was for exactly the issue you just hit, so I'm guessing we've worked on this before.09:45
lyarwoodkk looking through https://review.openstack.org/#/q/owner:%22Matthew+Booth+%253Cmbooth%2540redhat.com%253E%22+file:%22nova/tests/unit/virt/libvirt/fake_imagebackend.py%22 now09:47
lyarwoodmdbooth: yeah I really don't see it there, maybe you didn't post it?09:49
mdboothlyarwood: Possible09:49
mdboothlyarwood: Which bug are you working on?09:51
lyarwoodmdbooth: https://launchpad.net/bugs/1414895 & https://review.openstack.org/#/c/618478/109:52
openstackLaunchpad bug 1414895 in OpenStack Compute (nova) "failed compute node didn't delete instance's path directory in init_host" [Low,In progress] - Assigned to Lee Yarwood (lyarwood)09:52
mdboothlyarwood: Nope, I'm coming up blank.09:53
mdboothlyarwood: Lemme have a quick look at the code to refresh my memory.09:53
lyarwoodmdbooth: basically with images_type = rbd we never cleanup the instance directory after an evacuation09:54
lyarwoodmdbooth: we also don't have migrate_data set so I wanted to call down into the local backend to see if it's safe to remove the instance directory09:54
lyarwoodmdbooth: I don't understand the use of migrate_data here if I'm honest, given how the imagebackend is local to the current compute, I know we can't mix but it still seems pointless to stash in migrate_data09:55
*** jangutter has quit IRC09:56
kashyapHey folks, those who dwell on the Nova API part: any specific reason why the custom server metadata value is limited to 255 bytes here: https://github.com/openstack/nova/blob/master/nova/compute/api.py#L343,L36009:58
*** jangutter has joined #openstack-nova09:59
mdboothlyarwood: is_shared_block_storage() is a bug, btw09:59
mdboothlyarwood: Any use of it is pretty much guaranteed to be wrong in some circumstances09:59
lyarwoodmdbooth: how so?10:00
lyarwoodmdbooth: it's only True for the Rbd backend10:01
lyarwoodmdbooth: which is always shared10:01
mdboothlyarwood: Not necessarily shared between everything, though10:02
mdboothIt's broken if you have more than 1 ceph pool10:03
lyarwoodmdbooth: that's still not a reason to leave the instance directory around on the source node of an evacuation10:04
lyarwoodmdbooth: or to not clean the instance directory up10:04
mdboothlyarwood: Ok, from very vague memory I picked an arbitrary imagebackend, probably Flat, and made it return a closed subclass of that10:04
mdboothOr perhaps it wasn't a subclass... perhaps it was an autospec'd Mock10:05
lyarwoodmdbooth: kk thanks, I'll take a swing at that today10:05
openstackgerritMichael Still proposed openstack/nova master: Remove utils.execute() calls from xenapi.  https://review.openstack.org/61970010:06
openstackgerritMichael Still proposed openstack/nova master: Remove utils.execute() from libvirt remotefs calls.  https://review.openstack.org/61970110:06
openstackgerritMichael Still proposed openstack/nova master: Remove utils.execute() from quobyte libvirt storage driver.  https://review.openstack.org/61970210:06
openstackgerritMichael Still proposed openstack/nova master: Move nova.libvirt.utils away from using nova.utils.execute().  https://review.openstack.org/61970310:06
openstackgerritMichael Still proposed openstack/nova master: Imagebackend should call processutils.execute directly.  https://review.openstack.org/61970410:06
openstackgerritMichael Still proposed openstack/nova master: Remove final users of utils.execute() in libvirt.  https://review.openstack.org/61970510:06
*** yikun has joined #openstack-nova10:07
*** tssurya has quit IRC10:11
*** abhishekk has quit IRC10:18
*** evrardjp has joined #openstack-nova10:20
*** psachin has quit IRC10:26
*** psachin has joined #openstack-nova10:27
*** brinzhang has quit IRC10:30
*** dpawlik_ has quit IRC10:32
*** dpawlik has joined #openstack-nova10:37
kashyapmdbooth: Hi, got a moment to borrow your eyes on my understanding of a bit certain code?10:39
mdboothkashyap: Sure10:40
kashyapmdbooth: So the context is I'm trying to fix this bug: https://bugs.launchpad.net/nova/+bug/178013810:41
openstackLaunchpad bug 1780138 in OpenStack Compute (nova) "Don't assume the guest machine type to be of 'pc'" [Medium,Confirmed] - Assigned to Kashyap Chamarthy (kashyapc)10:41
kashyapmdbooth: See the "What will break?" section to get to the "tofu of the matter".10:41
mdboothIs that like a less-substantial meat of the matter?10:42
kashyapHeh10:42
kashyapmdbooth: My first step is to see in _get_guest_config() here: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L519410:43
kashyapIf the host capabilities (including the machine type for a given QEMU binary) will be in memory.10:43
kashyapLet me show an example of what I mean:10:43
kashyapGiven this line of code there:  caps = self._host.get_capabilities()10:46
*** lpetrut has joined #openstack-nova10:46
kashyapWill we have both the 'host' _and_ 'guest' part of the capabilities output as shown here: https://kashyapc.fedorapeople.org/virsh-capabilities.txt10:46
mdboothok10:48
kashyapI need to extract the default machine type (caps.guest.machine) for the QEMU on the given host, from this snippet:10:48
kashyap  <guest>10:48
kashyap    <os_type>hvm</os_type>10:48
kashyap    <arch name='x86_64'>10:48
kashyap      <wordsize>64</wordsize>10:48
kashyap      <emulator>/usr/bin/qemu-system-x86_64</emulator>10:48
kashyap      <machine maxCpus='255'>pc-i440fx-3.0</machine>10:48
kashyap      [...]10:48
*** sambetts_ has quit IRC10:49
*** sambetts has joined #openstack-nova10:49
kashyapSo my question is do you know if we'll have the "guest.arch.machine" as well in memory.  (And not just the <host> part)10:50
kashyapIf you're deeply buried in something, disregard me, I'll keep duking around :-)10:50
mdboothkashyap: Sorry, repair guy just turned up10:52
kashyapmdbooth: Yeah, saw your note about it earlier.  Keep going10:52
kashyapI also need to shortly step out to get some groceries10:53
mdboothkashyap: To 'answer' your question, I have no idea :)10:55
kashyapmdbooth: No worries, I'm trying to figure out how on earth to try it; maybe break-points and do 'print' of "caps", etc10:55
kashyapI'll let you know when / if I see the light10:56
sean-k-mooneykashyap: i belive you can use the machive type to set the arch but not sure. this was a capablit i taught we always had but recently i have been less sure if we actully support it or not10:57
kashyapsean-k-mooney: Heya, I'm not trying to set the arch via machine type.  What I want to know is, what _exactly_ are the contents of "caps"10:58
kashyap[Where caps = self._host.get_capabilities()]10:58
kashyapAnd how do I "see" it10:58
sean-k-mooneykashyap: i would hope it was be what is show in virsh cabalities for that emalator type10:59
*** k_mouza has quit IRC10:59
kashyapsean-k-mooney: So, 'virsh capabilities' has two parts: <host> and <guest> (per emulator)10:59
kashyapE.g. https://kashyapc.fedorapeople.org/virsh-capabilities.txt10:59
kashyapsean-k-mooney: Given that 'caps' holds self._host.get_capabilities(), will it have _both_ parts is what I'm wondering.11:00
sean-k-mooneykashyap: i added the arm package and this got added to virsh capablities http://paste.openstack.org/show/735969/11:01
kashyapsean-k-mooney: That's on your host, and is expected :-).11:02
kashyapSo, I'll assume Nova will have the entire 'virsh capabilities' output from the host.11:03
*** mdbooth has quit IRC11:03
kashyapWhen it runs: self._host.get_capabilities()11:03
sean-k-mooneyyes but you can list the falgs of the different cpus yes and get there capablities11:03
*** mdbooth has joined #openstack-nova11:05
kashyapsean-k-mooney: Nod; my main goal is to extract the default machine from the caps XML and update the code in _get_guest_config() to reflect that11:07
*** udesale has quit IRC11:07
kashyap(Instead of assuming the machine the machine type is "pc", when 'guest.os_mach_type == None' -- as noted in the code snippet in the bug.)11:07
* kashyap --> groceries; bbiab11:08
*** mdbooth has quit IRC11:09
*** alexchadin has quit IRC11:09
*** jchhatbar has joined #openstack-nova11:13
* cdent waves11:13
sean-k-mooneyo/11:14
kashyapcdent: Heya, wonder if you saw my earlier question about Nova API and metadata value limit to 255 bytes11:15
cdentkashyap: I didn't see it. I also know almost nothing about anything :)11:15
*** janki has quit IRC11:15
kashyap10:58 < kashyap> Hey folks, those who dwell on the Nova API part: any specific reason why the custom server metadata value is limited to 255 bytes here: https://github.com/openstack/nova/blob/master/nova/compute/api.py#L343,L36011:15
kashyapHeh, okay; I'll do a `git blame` and find the culprit ;-)11:16
cdentIt's probably an arbitrary choice of the "there needs to be some limit and we all love the number 255" sort11:16
kashyapHehe11:17
*** k_mouza has joined #openstack-nova11:17
*** mdbooth_ has joined #openstack-nova11:19
*** mdbooth_ is now known as mdbooth11:20
*** psachin has quit IRC11:26
sean-k-mooneykashyap: if you like im sure we could change it to 4211:26
*** jchhatba_ has joined #openstack-nova11:29
*** jchhatbar has quit IRC11:32
*** cdent has quit IRC11:49
*** tssurya has joined #openstack-nova11:53
jangutterclearly, 640KB should be enough for everybody (second time today I said this).11:58
*** ileixe has quit IRC12:01
openstackgerritMatthew Booth proposed openstack/nova master: Fix configure() called after DatabaseAtVersion fixture  https://review.openstack.org/61972312:06
mdbooth^^^ was a pig to find, causes non-deterministic failures in unit tests based on the order they run in a worker12:06
*** cdent has joined #openstack-nova12:16
*** pcaruana has quit IRC12:26
*** mdbooth has quit IRC12:34
*** awalende_ has joined #openstack-nova12:57
*** awalende has quit IRC13:01
*** skatsaounis has quit IRC13:06
*** cdent has quit IRC13:12
openstackgerritElod Illes proposed openstack/nova master: Transform scheduler.select_destinations notification  https://review.openstack.org/50850613:17
*** ratailor has quit IRC13:20
*** cdent has joined #openstack-nova13:22
*** davidsha has joined #openstack-nova13:40
*** whoami-rajat has quit IRC13:48
*** skatsaounis has joined #openstack-nova13:51
*** tbachman has joined #openstack-nova13:55
*** tbachman has quit IRC14:02
*** cfriesen has joined #openstack-nova14:37
*** awalende_ has quit IRC14:41
*** cfriesen has quit IRC14:42
*** ratailor has joined #openstack-nova14:43
*** Luzi has quit IRC14:53
*** cfriesen has joined #openstack-nova15:01
*** dpawlik has quit IRC15:06
*** mmethot has quit IRC15:09
openstackgerritBalazs Gibizer proposed openstack/nova master: Extend RequestGroup object for mapping  https://review.openstack.org/61952715:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Calculate RequestGroup resource provider mapping  https://review.openstack.org/61623915:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Fill the RequestGroup mapping during schedule  https://review.openstack.org/61952815:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Pass resource provider mapping to neutronv2 api  https://review.openstack.org/61624015:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Recalculate request group - RP mapping during re-schedule  https://review.openstack.org/61952915:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Send RP uuid in the port binding  https://review.openstack.org/56945915:16
openstackgerritBalazs Gibizer proposed openstack/nova master: Test boot with more ports with bandwidth request  https://review.openstack.org/57331715:16
*** lpetrut has quit IRC15:22
*** cdent has quit IRC15:23
*** mchlumsky has joined #openstack-nova15:39
*** mmethot has joined #openstack-nova15:41
*** pcaruana has joined #openstack-nova16:00
*** maciejjozefczyk has joined #openstack-nova16:01
*** cdent has joined #openstack-nova16:02
*** dpawlik has joined #openstack-nova16:04
*** dave-mccowan has joined #openstack-nova16:05
*** maciejjozefczyk has quit IRC16:06
*** dpawlik has quit IRC16:08
*** ratailor has quit IRC16:29
*** dave-mccowan has quit IRC16:45
*** jchhatba_ has quit IRC16:54
*** helenaAM has quit IRC16:58
*** jangutter has quit IRC17:02
openstackgerritMatthew Booth proposed openstack/nova master: Fix configure() called after DatabaseAtVersion fixture  https://review.openstack.org/61972317:06
*** dtantsur is now known as dtantsur|afk17:07
lyarwoodhmm is tox -e debug also broken for others at the moment?17:10
lyarwoodUnhandled error: TypeError: Can't instantiate abstract class SubclassSignatureTestCase with abstract methods _get_base_class17:10
lyarwoodis all I'm getting, even running oslo_debug_helper directly17:11
*** sridharg has quit IRC17:15
*** slaweq has quit IRC17:16
*** rmart04 has joined #openstack-nova17:34
*** davidsha has quit IRC17:35
*** k_mouza_ has joined #openstack-nova17:38
*** k_mouza has quit IRC17:42
*** k_mouza_ has quit IRC17:42
*** derekh has quit IRC18:00
*** erlon has quit IRC18:01
*** awalende has joined #openstack-nova18:06
*** ralonsoh has quit IRC18:08
*** rmart04 has quit IRC18:09
*** awalende has quit IRC18:10
*** rmart04 has joined #openstack-nova18:10
*** tssurya has quit IRC18:23
*** cdent has quit IRC18:25
*** rmart04 has quit IRC18:27
artomlyarwood, maybe try with -r ?18:56
*** k_mouza has joined #openstack-nova18:56
*** k_mouza has quit IRC19:00
openstackgerritJack Ding proposed openstack/nova-specs master: Flavor Extra Spec and Image Properties Validation  https://review.openstack.org/61854219:11
*** artom has quit IRC19:19
*** artom has joined #openstack-nova19:19
*** david-lyle has quit IRC19:47
*** tbachman has joined #openstack-nova20:30
*** slaweq has joined #openstack-nova20:40
*** dave-mccowan has joined #openstack-nova20:55
*** xek__ has quit IRC21:01
*** xek__ has joined #openstack-nova21:02
*** xek__ has quit IRC21:03
*** xek__ has joined #openstack-nova21:04
*** xek__ has quit IRC21:05
*** imacdonn has quit IRC21:06
*** imacdonn has joined #openstack-nova21:06
*** xek has joined #openstack-nova21:06
*** xek has quit IRC21:09
*** xek has joined #openstack-nova21:09
*** xek has quit IRC21:11
*** xek_ has joined #openstack-nova21:11
*** dave-mccowan has quit IRC21:11
*** xek_ has quit IRC21:12
*** slaweq has quit IRC21:21
*** pcaruana has quit IRC21:46
*** Yingxin has quit IRC22:03
*** awalende has joined #openstack-nova22:06
*** awalende has quit IRC22:11
openstackgerritLee Yarwood proposed openstack/nova master: libvirt: Always cleanup the instance dir when using block image backends  https://review.openstack.org/61847822:12
openstackgerritLee Yarwood proposed openstack/nova master: fixtures: Return a mocked class instead of method within fake_imagebackend  https://review.openstack.org/61980422:12
*** mchlumsky has quit IRC22:36
*** lbragstad has joined #openstack-nova22:53
openstackgerritLee Yarwood proposed openstack/nova master: fixtures: Return a mocked class instead of method within fake_imagebackend  https://review.openstack.org/61980422:56
openstackgerritLee Yarwood proposed openstack/nova master: libvirt: Always cleanup the instance dir when using block image backends  https://review.openstack.org/61847822:56
*** eandersson has joined #openstack-nova23:10
*** mmethot has quit IRC23:13
*** slaweq has joined #openstack-nova23:18
*** xek has joined #openstack-nova23:25
*** slaweq has quit IRC23:27
openstackgerritAndreas Karis proposed openstack/nova-specs master: Remove doc about metadata from 'share-pci-between-numa-nodes'  https://review.openstack.org/61980723:56

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