Friday, 2015-05-08

dansmithI can certainly provide review of "I have no idea if that works"00:00
*** achanda_ has joined #openstack-nova00:00
dims_need proof (as in logs generated when run under Apache)?00:01
dansmithcan you give me the link so I can look? :)00:01
dansmithI don't need proof no00:01
dims_https://review.openstack.org/#/c/179316/00:02
*** vilobhmm1 has quit IRC00:02
*** vilobhmm1 has joined #openstack-nova00:02
*** Marga_ has quit IRC00:03
melwittpark_heijlong: I get 501 for 'nova interface-list vm1' you said you get 501 for 'nova interface-attach' too, right?00:03
openstackgerritSean Dague proposed openstack/nova: don't report service group connection events as errors in dbdriver  https://review.openstack.org/18095600:04
openstackgerritSean Dague proposed openstack/nova: downgrade log messages for memcache server (dis)connect events  https://review.openstack.org/18125900:04
*** achanda has quit IRC00:04
dims_thanks dansmith00:04
*** apmelton is now known as apmelton_away00:04
dansmithmelwitt: that's what I tested00:04
melwittokay, hm00:05
*** achanda_ has quit IRC00:05
dansmithdims_: happy?00:08
*** IanGovett has quit IRC00:09
dims_dansmith: perfect :)00:09
alex_xucfriesen: hi, is there any plan discuss https://review.openstack.org/168982 on summit?00:09
*** otter768 has joined #openstack-nova00:10
dansmithdims_: reviewed your related nova patch too00:12
tjones1dims_: mind giving this one some love again?  your +1 got lost in the thousands of rebases…00:13
tjones1https://review.openstack.org/#/c/127427/00:13
dims_dansmith: thanks, will share with you the result of the 2 patches when this finishes - https://review.openstack.org/#/c/181260/00:13
dims_tjones1: ack00:13
dansmithdims_: that doesn't work does it?00:14
dansmithdims_: that's controlled by d-g right?00:14
dims_dansmith: we should see 2 log files under apache (same location as keystone)00:14
dansmithdims_: what I mean is, that's going to get overridden by the localrc that d-g generates, isn't it?00:15
openstackgerritSean Dague proposed openstack/nova: don't report service group connection events as errors in dbdriver  https://review.openstack.org/18095600:15
openstackgerritSean Dague proposed openstack/nova: downgrade log messages for memcache server (dis)connect events  https://review.openstack.org/18125900:15
sdaguealex_xu: thanks for the quick catch00:15
*** otter768 has quit IRC00:15
alex_xusdague: you are welcome00:15
dims_dansmith: i touched up stackrc in this review, so should see it activated in some jobs at least00:16
*** tjones1 has quit IRC00:16
dansmithhmm, okay00:16
*** ociuhandu has quit IRC00:18
sdaguedansmith: while I don't have the energy to dive into it tonight, I realized that I can probably just turn that grenade guest creation / verify loop up to 11 and hopefully make this shake out a lot more often. I've seen it locally at least once when I was doing the refactor.00:21
dansmithokay00:21
dansmithwell, yeah, I'm done for the day00:21
dansmithso tomorrow00:21
sdagueyeh, I'll have hopefully some data by the time you get in00:21
dansmithokay00:22
*** takedakn has quit IRC00:22
*** ddaskal has joined #openstack-nova00:23
park_heijlongmelwitt: yep, interface-list and interface-attach, both got 50100:23
dims_jogo: remember i mentioned the new oslo configuration generation utility? if you get a chance, please take a look at the output when you get a chance (https://review.openstack.org/#/c/180013/)00:25
*** andreykurilin__ has quit IRC00:26
*** salv-orlando has joined #openstack-nova00:26
melwittpark_heijlong: okay, thanks. I think we can remove the nova project from that bug then. if something weird is happening in cells, then it'll be a different bug00:27
melwittpark_heijlong: that is, I don't think anything got fixed by the other patch you mentioned. that patch is about raising NotFound instead of BadRequest00:28
alex_xubauzas: melwitt: just saw the irc log, we have bug for v2.1 api ref...there is no reference for os-services. and I actually found more... like os-pci disabled in v2.1 but still in the doc...00:28
alex_xulet me try to fix them today00:28
melwittcool alex_xu00:28
*** jamielennox|away is now known as jamielennox00:29
eliqiao1yes, project_id doen't show up in v2.1 api ref.. this is an issue too.00:29
melwittpark_heijlong: I'll update the bug00:30
*** salv-orlando has quit IRC00:30
*** park_heijlong has quit IRC00:31
*** tsekiyam_ has joined #openstack-nova00:31
*** promulo__ has joined #openstack-nova00:31
sdaguealex_xu: I fixed your nits on that last patch round as well, given that it was still up in my editor00:34
*** tsekiyama has quit IRC00:34
*** promulo_ has quit IRC00:35
*** mtanino has quit IRC00:36
*** tsekiyam_ has quit IRC00:36
*** zhenguo has joined #openstack-nova00:38
*** penick has quit IRC00:39
*** Raildo_ has quit IRC00:41
*** alanf-mc has quit IRC00:46
alex_xusdague: ok, thanks00:47
*** erlon has quit IRC00:51
*** tjones1 has joined #openstack-nova00:57
tjones1dims_: thanks00:57
*** yamahata has joined #openstack-nova01:00
*** gzhai has quit IRC01:00
*** gzhai has joined #openstack-nova01:01
*** VW_ has joined #openstack-nova01:03
*** mkoderer has quit IRC01:05
*** mkoderer has joined #openstack-nova01:13
*** sambol has joined #openstack-nova01:14
*** park_heijlong has joined #openstack-nova01:16
openstackgerritEli Qiao proposed openstack/nova: Libvirt: Show real vCpuCompareResult instead of a url link  https://review.openstack.org/16567101:16
*** jwang_ has joined #openstack-nova01:18
*** otter768 has joined #openstack-nova01:18
*** jwang has quit IRC01:21
*** zhangtralon has joined #openstack-nova01:22
*** dsanders has quit IRC01:27
*** tjones1 has quit IRC01:28
*** aix has quit IRC01:31
*** VW_ has quit IRC01:31
*** kaisers has quit IRC01:44
*** zzzeek has quit IRC01:44
*** patrickeast has quit IRC01:45
*** unicell has quit IRC01:49
*** winston-d_zzZ is now known as winston-d01:53
*** neelashah has joined #openstack-nova01:53
*** kaisers has joined #openstack-nova01:57
*** dims_ has quit IRC02:00
*** popw has joined #openstack-nova02:00
*** dims has joined #openstack-nova02:00
*** dims has quit IRC02:01
*** vilobhmm1 has quit IRC02:01
*** pixelbeat has quit IRC02:01
*** jwcroppe has joined #openstack-nova02:03
*** baoli has joined #openstack-nova02:05
*** baoli has quit IRC02:06
*** baoli has joined #openstack-nova02:07
*** david-lyle has joined #openstack-nova02:10
*** VW_ has joined #openstack-nova02:10
*** VW_ has quit IRC02:13
*** annegentle has joined #openstack-nova02:15
*** patrickeast has joined #openstack-nova02:19
*** annegentle has quit IRC02:20
*** browne has quit IRC02:23
*** achanda has joined #openstack-nova02:29
*** angdraug has quit IRC02:32
*** sambol has quit IRC02:39
*** r-daneel has quit IRC02:40
*** annegentle has joined #openstack-nova02:40
openstackgerritEli Qiao proposed openstack/nova-specs: Validate the service state before deleting it  https://review.openstack.org/16327402:44
openstackgerritEli Qiao proposed openstack/nova: Remove db layer hard-code permission checks for v2.1 cells  https://review.openstack.org/15035002:48
*** salv-orlando has joined #openstack-nova02:54
*** sambol has joined #openstack-nova02:54
*** zsais has joined #openstack-nova02:54
*** browne has joined #openstack-nova02:55
*** dsanders has joined #openstack-nova03:01
*** dims has joined #openstack-nova03:01
*** ashishjain has joined #openstack-nova03:02
ashishjainhi03:02
ashishjainSomehow  I am unable to access the metadata information from a deployed instance03:03
ashishjainthis does not work03:03
*** salv-orl_ has joined #openstack-nova03:03
ashishjaincurl http://169.254.169.254/openstack/latest/user_data03:03
ashishjainanyone faced a similar issue03:03
*** salv-orlando has quit IRC03:04
*** ozialien has quit IRC03:06
*** dims has quit IRC03:06
*** salv-orl_ has quit IRC03:08
*** baoli has quit IRC03:16
*** salv-orlando has joined #openstack-nova03:18
*** tzumainn_ has quit IRC03:19
*** achanda has quit IRC03:22
*** salv-orlando has quit IRC03:22
*** eliqiao1 is now known as eliqiao03:24
*** tojuvone has joined #openstack-nova03:26
*** VW_ has joined #openstack-nova03:31
*** tralon_ has joined #openstack-nova03:32
*** yasu_ has joined #openstack-nova03:36
*** marun has quit IRC03:37
*** vilobhmm1 has joined #openstack-nova03:41
*** unicell has joined #openstack-nova03:42
*** annegentle has quit IRC03:43
*** annegentle has joined #openstack-nova03:44
*** zsais has quit IRC03:47
*** otter768 has quit IRC03:48
*** unicell has quit IRC03:48
*** unicell has joined #openstack-nova03:49
*** patrickeast has quit IRC03:50
*** deepthi has joined #openstack-nova03:50
*** sambol has quit IRC03:53
*** ddaskal has quit IRC03:55
*** dsanders has quit IRC03:56
*** VW_ has quit IRC03:56
*** Marga_ has joined #openstack-nova03:58
*** Marga_ has quit IRC04:00
*** Marga_ has joined #openstack-nova04:00
*** ashishjain has quit IRC04:01
*** annegentle has quit IRC04:02
*** xyang1 has quit IRC04:06
*** sdake has quit IRC04:13
*** VW_ has joined #openstack-nova04:14
*** mmedvede has quit IRC04:14
*** baoli has joined #openstack-nova04:17
*** Alexandra_ has joined #openstack-nova04:17
*** changbl has joined #openstack-nova04:18
*** neelashah has quit IRC04:21
*** baoli has quit IRC04:22
*** achanda has joined #openstack-nova04:22
*** rushiagr_away is now known as rushiagr04:27
*** achanda has quit IRC04:28
*** VW_ has quit IRC04:28
*** mmedvede has joined #openstack-nova04:28
*** dsanders has joined #openstack-nova04:42
*** achanda has joined #openstack-nova04:42
*** tralon_ has quit IRC04:51
*** zhangtralon has quit IRC04:51
*** Sukhdev has joined #openstack-nova04:56
*** nkrinner has joined #openstack-nova04:57
*** cfriesen has quit IRC04:57
*** gokrokve has joined #openstack-nova04:58
*** ferest has joined #openstack-nova04:59
*** Longgeek has joined #openstack-nova05:00
*** annegentle has joined #openstack-nova05:03
*** ildikov has quit IRC05:04
*** VW_ has joined #openstack-nova05:04
*** dave-mccowan has quit IRC05:05
*** dulek has joined #openstack-nova05:05
*** emagana has joined #openstack-nova05:07
*** annegentle has quit IRC05:09
*** tralon_ has joined #openstack-nova05:09
*** zhangtralon has joined #openstack-nova05:09
*** gokrokve has quit IRC05:09
*** gokrokve has joined #openstack-nova05:10
*** VW_ has quit IRC05:12
*** gokrokve has quit IRC05:13
*** gokrokve has joined #openstack-nova05:13
*** tralon_ has quit IRC05:16
*** zhangtralon has quit IRC05:17
*** ildikov has joined #openstack-nova05:17
*** baoli has joined #openstack-nova05:17
*** gokrokve has quit IRC05:18
*** salv-orlando has joined #openstack-nova05:19
*** baoli has quit IRC05:22
*** salv-orlando has quit IRC05:22
*** ferest has quit IRC05:25
*** bkopilov has quit IRC05:27
*** VW_ has joined #openstack-nova05:28
*** Longgeek has quit IRC05:29
*** Sukhdev has quit IRC05:30
*** ildikov has quit IRC05:30
*** jyuso1 has quit IRC05:32
*** VW_ has quit IRC05:36
*** gokrokve has joined #openstack-nova05:40
*** gokrokve has quit IRC05:42
*** gokrokve has joined #openstack-nova05:42
*** emagana has quit IRC05:44
*** vilobhmm1 has quit IRC05:44
*** bkopilov has joined #openstack-nova05:44
*** emagana has joined #openstack-nova05:45
*** otter768 has joined #openstack-nova05:49
*** emagana has quit IRC05:49
*** Nikolay_St has quit IRC05:49
*** park_heijlong has quit IRC05:51
*** kiran-r has joined #openstack-nova05:51
*** otter768 has quit IRC05:53
*** dsanders has quit IRC05:56
*** wverdugo500 has joined #openstack-nova05:56
openstackgerritEli Qiao proposed openstack/nova: Set correct power state when evaucate do failed due to recreate not supported  https://review.openstack.org/17315005:56
*** Alexandra_ has quit IRC05:58
*** Kennan has quit IRC06:03
*** Kennan has joined #openstack-nova06:03
*** annegentle has joined #openstack-nova06:04
*** belmoreira has joined #openstack-nova06:04
*** vilobhmm1 has joined #openstack-nova06:05
*** tralon_ has joined #openstack-nova06:06
*** zhangtralon has joined #openstack-nova06:07
*** pkoniszewski has joined #openstack-nova06:10
*** annegentle has quit IRC06:10
*** park_heijlong has joined #openstack-nova06:10
*** salv-orlando has joined #openstack-nova06:13
*** Longgeek has joined #openstack-nova06:13
*** oro has joined #openstack-nova06:13
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex  https://review.openstack.org/17963806:21
*** park_heijlong has quit IRC06:25
*** sudipto has joined #openstack-nova06:26
*** jyuso1 has joined #openstack-nova06:30
*** vilobhmm1 has quit IRC06:30
*** markvoelker has quit IRC06:33
*** ajayaa has joined #openstack-nova06:34
*** oro has quit IRC06:37
*** kiran-r has quit IRC06:37
*** achanda has quit IRC06:38
*** coreywright has quit IRC06:39
*** zhangtralon has quit IRC06:41
*** ildikov has joined #openstack-nova06:43
*** Marga_ has quit IRC06:44
*** rms_13 has quit IRC06:45
*** mpsairam has joined #openstack-nova06:46
*** bkopilov has quit IRC06:46
openstackgerritAnkit Agrawal proposed openstack/nova: Detach volume after deleting instance with no host  https://review.openstack.org/14573806:50
*** bfic has joined #openstack-nova06:56
*** ankit_ag has joined #openstack-nova07:00
*** armax has quit IRC07:02
*** rushiagr is now known as rushiagr_away07:04
*** annegentle has joined #openstack-nova07:05
*** Mike_D_laptop has joined #openstack-nova07:06
*** pask81 has joined #openstack-nova07:10
*** annegentle has quit IRC07:10
*** pkoniszewski has quit IRC07:12
*** markus_z has joined #openstack-nova07:14
*** jyuso1 has quit IRC07:17
*** jichen has joined #openstack-nova07:18
*** eglynn has quit IRC07:22
*** mmedvede has quit IRC07:25
*** apuimedo has joined #openstack-nova07:27
*** jyuso1 has joined #openstack-nova07:27
*** coreywright has joined #openstack-nova07:32
*** mmedvede has joined #openstack-nova07:35
*** jamielennox is now known as jamielennox|away07:36
openstackgerritTakenori Yoshimatsu proposed openstack/nova: Avoid AttributeError at instance.info_cache.delete  https://review.openstack.org/16790507:39
*** ajayaa has quit IRC07:40
*** rushiagr_away is now known as rushiagr07:40
*** kiran-r has joined #openstack-nova07:41
*** mpavone has joined #openstack-nova07:41
*** jlanoux has joined #openstack-nova07:45
*** otter768 has joined #openstack-nova07:50
*** angdraug has joined #openstack-nova07:51
*** angdraug has quit IRC07:51
*** otter768 has quit IRC07:54
*** mpsairam has left #openstack-nova07:57
*** lucasagomes has joined #openstack-nova07:58
*** bfic has quit IRC07:59
*** mpsairam has joined #openstack-nova07:59
*** ildikov has quit IRC08:01
*** browne has quit IRC08:02
*** ajayaa has joined #openstack-nova08:03
*** tobe has joined #openstack-nova08:05
*** annegent_ has joined #openstack-nova08:06
*** markvoelker has joined #openstack-nova08:07
openstackgerritMarkus Zoeller (markus_z) proposed openstack/nova: hypervisor support matrix: add feature "serial console"  https://review.openstack.org/18091208:09
*** kiranr has joined #openstack-nova08:10
*** kiran-r has quit IRC08:10
*** annegent_ has quit IRC08:11
*** park_heijlong has joined #openstack-nova08:16
*** ildikov has joined #openstack-nova08:17
*** eglynn has joined #openstack-nova08:20
*** danpb has joined #openstack-nova08:21
*** yingjun has joined #openstack-nova08:24
*** bfic has joined #openstack-nova08:26
*** bfic has quit IRC08:28
*** bfic has joined #openstack-nova08:29
*** bfic1 has joined #openstack-nova08:30
*** bfic has quit IRC08:30
*** redbeard has joined #openstack-nova08:30
*** tobe has quit IRC08:31
*** derekh has joined #openstack-nova08:37
*** Nikolay_St has joined #openstack-nova08:37
*** markvoelker has quit IRC08:38
*** bfic1 has quit IRC08:46
*** bkopilov has joined #openstack-nova08:51
openstackgerritStephen Ahn proposed openstack/nova: libvirt: Rollback status when cold migrate failed  https://review.openstack.org/14426108:52
*** dguitarbite has quit IRC08:56
*** dbite has quit IRC08:56
*** dguitarbite has joined #openstack-nova08:57
*** jwcroppe has quit IRC08:58
*** jwcroppe has joined #openstack-nova08:58
*** bfic has joined #openstack-nova09:03
*** yingjun has quit IRC09:03
*** ildikov has quit IRC09:04
*** bfic has quit IRC09:06
*** yingjun has joined #openstack-nova09:07
*** annegentle has joined #openstack-nova09:07
*** bfic has joined #openstack-nova09:11
*** annegentle has quit IRC09:12
*** bfic has quit IRC09:15
*** bfic has joined #openstack-nova09:17
*** ildikov has joined #openstack-nova09:20
openstackgerritSailaja proposed openstack/nova: Project quota update should not be less than user quota.  https://review.openstack.org/18133409:20
ekarlsoheya guys, what service is it that has responsability for dns stuff ?09:21
johnthetubaguyekarlso: it it designate?09:21
johnthetubaguys/it it/is it/09:22
*** oro has joined #openstack-nova09:23
ekarlsojohnthetubaguy: ya :)09:23
ekarlsowanting to make a nova shim for it09:24
johnthetubaguyekarlso: what does that mean?09:24
ekarlsowe had a commit to make a nova integration for your dns piece into designate09:25
*** kiranr is now known as kiran-r09:26
ekarlsojohnthetubaguy: https://github.com/openstack/nova/blob/master/nova/network/minidns.py < looking at that atm09:26
ekarlsowhat stuff is it that calls that kind of dns stuff ?09:26
*** IanGovett has joined #openstack-nova09:27
johnthetubaguyekarlso: oh so thats a different thing to what I thought you were thinking09:27
johnthetubaguyekarlso: thats some code we are trying to deprecate, and failing09:27
johnthetubaguyekarlso: that is the DNS server that nova network gives out from its DHCP responses, if I remember correctly09:28
johnthetubaguyekarlso: neutron has something too, but I am unsure about the details09:28
*** dulek_ has joined #openstack-nova09:28
johnthetubaguyekarlso: I am not an expert in nova-network though, I would have to go digging myself09:28
*** dulek has quit IRC09:29
*** bfic has quit IRC09:29
ekarlsojohnthetubaguy: if you say you are trying to deprecate it09:30
ekarlsoi'm wondering if it's any use for us to try to spend time on it :/09:30
openstackgerritEli Qiao proposed openstack/nova: Libvirt: Show real vCpuCompareResult instead of a url link  https://review.openstack.org/16567109:30
*** yingjun has quit IRC09:30
*** bfic has joined #openstack-nova09:32
*** tralon__ has joined #openstack-nova09:33
*** bfic has quit IRC09:34
*** ndipanov has joined #openstack-nova09:35
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Sync latest code from oslo-incubator  https://review.openstack.org/15005809:37
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Reuse uuidutils frim oslo_utils  https://review.openstack.org/18134009:37
*** tralon_ has quit IRC09:37
*** kiwik-chenrui has joined #openstack-nova09:38
*** RuiChen has quit IRC09:41
*** mpsairam has left #openstack-nova09:44
*** alexpilotti has joined #openstack-nova09:48
*** otter768 has joined #openstack-nova09:51
sudiptojohnthetubaguy, bauzas I have an observation w.r.t the scheduler filters. Let's say I have a image with a certain specified arch_type - I see that in  a mixed environment, the filter scheduler runs all the filters on each of the hosts - despite the fact that the arch_type should have invalidated the host that don't support the format completely. Would you agree with the observation?09:51
johnthetubaguysudipto: not sure thats actually true, at least it didn't used to be true09:51
*** pask81 has quit IRC09:52
johnthetubaguysudipto: I thought it was all about the ordering of the filters09:52
johnthetubaguysudipto: try put the filter that removes the most hosts first09:52
sudiptojohnthetubaguy, Ok. How'z the order determined? Is it always via the nova.conf?09:52
johnthetubaguysudipto: now its possible that got changed, possibly accidentally09:52
johnthetubaguysudipto: yes, I believe so09:53
johnthetubaguysudipto: unless you use the default order09:53
*** ociuhandu has joined #openstack-nova09:54
sudiptojohnthetubaguy, yeah - the order seems to be the same as nova.conf. But I wonder - why it ran against both the hosts, atleast if the ImageFilter disqualifies the image - should it still go via the remaining filters?09:54
johnthetubaguysudipto: I don't think it should09:55
sudiptoAlso, i think i spoke to soon, let me verify if the order is actually followed or not.09:55
*** otter768 has quit IRC09:55
sudiptojohnthetubaguy, I am on Juno, maybe i am looking at older code :(09:55
*** park_heijlong has quit IRC09:55
johnthetubaguysudipto: quite possibly09:55
johnthetubaguysudipto: although I thought it was like this for ages, https://github.com/openstack/nova/blob/master/nova/filters.py#L7409:56
sudiptojohnthetubaguy, yeah - looks like it. I have an environment with a power server and a x86 server. The image used for x86 - is being run against the power servers too - but let me confirm.09:57
johnthetubaguysudipto: hmm, maybe its just not filtering out how you expect09:58
sdaguesudipto: make sure the image metadata you are applying is exactly what's in the code that's being filtered on. Also, realize, very few people have mixed arch clouds, so it wouldn't surprise me if that conditional managed to be wrong10:02
*** coolsvap|afk is now known as coolsvap10:03
*** pask81 has joined #openstack-nova10:06
*** annegent_ has joined #openstack-nova10:08
*** annegent_ has quit IRC10:13
*** kiwik-chenrui has quit IRC10:15
*** moshele has joined #openstack-nova10:15
*** zhenguo has quit IRC10:17
*** zhangtralon has joined #openstack-nova10:18
openstackgerritSailaja proposed openstack/nova: Project quota update should not be less than user quota.  https://review.openstack.org/18134910:19
*** tralon__ has quit IRC10:19
ekarlsojohnthetubaguy: u got a clue on who it is that's responsible for that part of nova ?10:23
johnthetubaguyekarlso: we all are :)10:23
johnthetubaguyekarlso: there is a nova-scheduler subteam meeting you could join and ask them10:23
johnthetubaguyekarlso: what are you seeing? do you have a unit test that proves what you are seeing?10:24
ekarlsojohnthetubaguy: according to our last talks about this designate commited as far as I know to make a nova <> designate integration for the dns code you have10:24
ekarlsonow i'm trying to decide if the effort is needed or not :|10:24
johnthetubaguyekarlso: oh sorry, I crossed wires in my head :)10:24
johnthetubaguyekarlso: we are trying to find an owner for that bit, beagles might be keen, although not 100% sure10:25
johnthetubaguyekarlso: I would worry about neutron and designate, and see how the nova-network talk goes at the summit10:25
johnthetubaguyekarlso: now as for designate finding out about what the hostname should be, nova probably has that information, but it sounds a bit like we should evolve the interface with neutron to support that kind of thing10:26
ekarlsojohnthetubaguy: I guess the integration I mean is the "nova help | grep dns" part10:28
ekarlsoare you saying that is going away ? :D10:28
johnthetubaguyekarlso: I had assumed we left that to neutron now, but thats a good question… I would have to go digging10:29
ekarlsojohnthetubaguy: would be awesome ;P10:29
sudiptojohnthetubaguy, yeah sure.10:30
johnthetubaguyekarlso: I guess its this bit? https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/contrib/floating_ip_dns.py10:30
johnthetubaguyekarlso: https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L173710:31
johnthetubaguyekarlso: hmm, the neutron implementation is not… totally awesome10:31
ekarlsohah :p10:31
ekarlsojohnthetubaguy: that whole area rly sounds like a big mess :p10:31
johnthetubaguyekarlso: it really is, just digging into that now10:32
*** pixelbeat has joined #openstack-nova10:32
ekarlsojohnthetubaguy: furthermore I dont think neutron has any ./care for dns10:32
johnthetubaguyekarlso: thats a feature gap that hasn't come up: https://github.com/openstack/nova/blob/master/nova/network/api.py#L42610:32
johnthetubaguyekarlso: I thought they did dns based on the ip address?10:32
ekarlsojohnthetubaguy: I know they write some stuff to dnsmasq but if that's dns I dunno10:33
ekarlsojohnthetubaguy: it seems all they do with dnsmasq is to write out the dhcp entries and the dns nameservers for the subnet :)10:34
ekarlsojohnthetubaguy: I guess the "dns" part that nova has is some nova-network legacy stuff :p10:34
johnthetubaguyekarlso: it is yeah, thats the bit I linked to just above10:35
johnthetubaguyekarlso: it only works with nova-network basically10:35
johnthetubaguyekarlso: which is fun, as thats not come up in the parity discussion yet (as far as I remember)10:35
ekarlsojohnthetubaguy: sounds like something that should be removed ;P10:35
*** markvoelker has joined #openstack-nova10:35
johnthetubaguyekarlso: its a supported public API, thats tricky10:36
ekarlsojohnthetubaguy: hmmm true enough10:36
ekarlsoso it's dns then for floating ip's..10:36
*** Kevin_Zheng has quit IRC10:36
johnthetubaguyekarlso: thats the main bit I can see, which makes sense with the old model, where floating ips are your only public addresses10:37
johnthetubaguyekarlso: have you seen this: https://review.openstack.org/#/c/9015010:37
ekarlsojohnthetubaguy: nope -,,-10:38
ekarlsojohnthetubaguy: I guess I'll have to go and talk with baldwin :D10:38
*** markvoelker has quit IRC10:40
*** fawadkhaliq has joined #openstack-nova10:41
ekarlsojohnthetubaguy: anyways it sounds to me at least that doing a nova to designate integration at that layer at least sounds like making more technical debt10:41
johnthetubaguyekarlso: +110:42
johnthetubaguyekarlso: at least it looks that way10:42
johnthetubaguyekarlso: I would push for doing neutron + nova + designate first10:42
*** dulek___ has joined #openstack-nova10:43
*** dulek_ has quit IRC10:43
ekarlsojohnthetubaguy: yeah, it sounds like that :P10:44
*** baoli has joined #openstack-nova10:49
*** pask81 has quit IRC10:49
*** pask81 has joined #openstack-nova10:51
*** deepthi has quit IRC10:51
*** baoli has quit IRC10:53
*** moshele has quit IRC10:55
*** zhangtralon has quit IRC10:58
openstackgerritAnthony PERARD proposed openstack/nova: libvirt-xen: Replace vda by xvda in BDM v1  https://review.openstack.org/18016111:00
*** moshele has joined #openstack-nova11:00
*** pask81 has quit IRC11:02
*** pask81 has joined #openstack-nova11:02
*** ajayaa has quit IRC11:05
*** mwagner_lap has joined #openstack-nova11:06
*** annegent_ has joined #openstack-nova11:08
*** rfolco has joined #openstack-nova11:09
*** annegent_ has quit IRC11:13
*** dims has joined #openstack-nova11:13
openstackgerritHans Lindgren proposed openstack/nova: Add a VolumeUsage object  https://review.openstack.org/18044211:14
openstackgerritHans Lindgren proposed openstack/nova: Replace conductor proxying calls with the new VolumeUsage object  https://review.openstack.org/18044311:14
openstackgerritHans Lindgren proposed openstack/nova: Remove and deprecate unused conductor method vol_usage_update()  https://review.openstack.org/18044411:14
*** moshele has quit IRC11:15
*** VW_ has joined #openstack-nova11:16
*** bfic has joined #openstack-nova11:16
*** tojuvone has quit IRC11:17
*** ajayaa has joined #openstack-nova11:22
*** tdurakov_ has joined #openstack-nova11:23
*** dims has quit IRC11:28
johnthetubaguygilliard_afk alex_xu jaypipes updated the API summit session etherpads, https://etherpad.openstack.org/p/YVR-nova-api-2.0-3rd-party https://etherpad.openstack.org/p/YVR-nova-api-2.1-in-liberty11:29
* johnthetubaguy goes for lunch11:29
*** dims has joined #openstack-nova11:29
alex_xujohnthetubaguy: thanks!11:29
*** fawadkhaliq has quit IRC11:31
openstackgerritDaniel Berrange proposed openstack/nova: objects: add a FlexibleBoolean field type  https://review.openstack.org/18108511:31
openstackgerritDaniel Berrange proposed openstack/nova: objects: treat EnumField class as abstract  https://review.openstack.org/18108611:31
openstackgerritDaniel Berrange proposed openstack/nova: objects: define the ImageMeta & ImageMetaProps objects  https://review.openstack.org/7623411:31
openstackgerritDaniel Berrange proposed openstack/nova: objects: convert HVSpec object to use enums  https://review.openstack.org/18136811:31
*** fawadkhaliq has joined #openstack-nova11:34
alex_xudanpb: hi, just want to ask is there any plan discussion this https://review.openstack.org/168982? if there is one let me I'm interesting join the discussion, thanks in advance!11:35
*** markvoelker has joined #openstack-nova11:36
alex_xudanpb: sorry, I mean discussion at summit...sorry for missing some words...11:39
danpbi'm not going to be at the summit so i don't know11:40
*** b3nt_pin has joined #openstack-nova11:41
alex_xudanpb: ok. go it, thanks, maybe I should have talk with Chris at summit first. a lot of debate on https://review.openstack.org/168982 ...11:42
*** aswadr has joined #openstack-nova11:42
alex_xus/go it/got it/...11:44
*** eliqiao1 has joined #openstack-nova11:46
*** yasu_ has quit IRC11:49
*** lucasagomes is now known as lucas-hungry11:50
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: [WIP] Add rootwrap daemon mode support (Switched off)  https://review.openstack.org/18069511:50
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: [WIP] Run using rootwrap daemon as True  https://review.openstack.org/18080911:51
*** otter768 has joined #openstack-nova11:52
tdurakov_hi bauzas!11:52
tdurakov_got a question for you about https://bugs.launchpad.net/nova/+bug/121494311:53
openstackLaunchpad bug 1214943 in OpenStack Compute (nova) "Live migration should use the same memory over subscription logic as instance boot" [High,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza)11:53
*** pask81 has quit IRC11:53
*** pask811 has joined #openstack-nova11:53
*** artom has joined #openstack-nova11:54
*** ndipanov has quit IRC11:56
*** otter768 has quit IRC11:56
alex_xujohnthetubaguy: jaypipes gilliard_afk eliqiao nova api meeting in 4 minutes. free to join if you have something want to talk about.11:56
eliqiao1sure. I am here alex_xu11:57
eliqiao1eliqiao: is a robot :)11:58
alex_xueliqiao1: :)11:58
*** pask811 has quit IRC11:58
alex_xueliqiao1: not sure Ken'ichi whether online after few mins11:58
*** tdurakov_ has quit IRC11:59
eliqiao1alex_xu: do we still have 1 meeting before summit?12:00
*** annegentle has joined #openstack-nova12:00
*** fawadk has joined #openstack-nova12:01
*** kiranr has joined #openstack-nova12:02
*** fawadkhaliq has quit IRC12:02
*** kiran-r has quit IRC12:02
sdagueI'm looking through some of these cpu features -> scheduler specs. Is there a reason that there isn't more of a focus on tools to make it simpler to manage the correct aggregates for a deployment instead of baking every possible match detail into the image -> hypervisor flow?12:04
*** fawadk has quit IRC12:06
*** dave-mccowan has joined #openstack-nova12:06
*** tojuvone has joined #openstack-nova12:07
*** markvoelker has quit IRC12:07
*** markvoelker has joined #openstack-nova12:07
*** sdake has joined #openstack-nova12:07
*** tojuvone has quit IRC12:07
*** tojuvone has joined #openstack-nova12:08
*** sdake_ has joined #openstack-nova12:10
*** derekh is now known as derekh_afk12:11
*** tdurakov_ has joined #openstack-nova12:13
*** sdake has quit IRC12:14
*** MVenesio has joined #openstack-nova12:14
*** claudiub has joined #openstack-nova12:15
*** nithyag_ has joined #openstack-nova12:16
*** dulek___ is now known as dulek_12:16
*** dulek_ is now known as dulek12:16
*** kiranr is now known as kiran-r12:16
*** NithyaG has quit IRC12:18
*** fawadkhaliq has joined #openstack-nova12:19
dimsjohnthetubaguy: i've lined up blueprint, spec and reviews for oslo_rootwrap as daemon (https://blueprints.launchpad.net/nova/+spec/nova-rootwrap-daemon-mode). please take a look when you get a chance12:19
*** baoli has joined #openstack-nova12:19
*** mtanino has joined #openstack-nova12:20
*** dprince has joined #openstack-nova12:22
*** baoli has quit IRC12:24
*** tdurakov_ has quit IRC12:27
johnthetubaguysdague: I have a summit session of the future of flavor extra specs and image properties, its getting a little out of hand12:27
sdaguejohnthetubaguy: ok great12:28
*** sudipto has quit IRC12:28
johnthetubaguysdague: if a discoverability rating was a thing, it would almost be zero, which sucks12:28
*** IanGovett1 has joined #openstack-nova12:28
sdagueright12:29
sdagueyou actually have to read the code to figure out half of them12:29
*** IanGovett has quit IRC12:30
*** VW_ has quit IRC12:31
*** ferest has joined #openstack-nova12:32
*** VW_ has joined #openstack-nova12:36
*** ZZelle has joined #openstack-nova12:37
danpbjohnthetubaguy: this changes is solving it for image props  https://review.openstack.org/7623412:37
danpbthe same will need to be done for flavour extra specs too at some point12:37
*** rushiagr is now known as rushiagr_away12:37
*** tdurakov_ has joined #openstack-nova12:38
danpbdansmith has been pushing for this imagemetaprops object to be created a while12:39
*** aysyd has joined #openstack-nova12:40
johnthetubaguydanpb: yeah, its a good first step, remember the earlier versions of that12:41
johnthetubaguydanpb: kinda want to agree the wider goals there12:42
*** ankit_ag has quit IRC12:42
johnthetubaguydanpb: do you end up saving it in the instance object like flavor yet? instead of system_metadata stuff?12:42
*** aswadr has quit IRC12:43
danpbits not that far along - its just a standalone object for now that is populated from the existing dicts12:43
*** raildo has quit IRC12:43
danpblots of tedious work to incrementally convert the drivers to use the new object too12:44
*** sdake has joined #openstack-nova12:44
danpband then finally look at better integration with the instance object12:44
danpbdidn't want todo the latter as the first step, since that would lead to a quite enourmous patch diff12:44
johnthetubaguydanpb: yeah, agreed12:45
danpbthe sooner we get this done the better though - every time i look at this i find more horrors to deal with12:46
johnthetubaguydanpb: agreed, it should be a blueprint of its own though, its not really mentioned in the request spec blueprint its been related too12:46
*** kylek3h has joined #openstack-nova12:47
johnthetubaguydanpb: if we get it a blueprint, folks know its important code to review, thats all I am thinking here, not process for the sake of it12:47
*** sdake_ has quit IRC12:47
johnthetubaguydanpb: regardless of the summit session output, we need to do that work you have up, so thats all good12:47
danpbyeah, sylvain needs this object for some of the request-spec blueprint work too12:48
*** edleafe is now known as figleaf12:51
*** raildo has joined #openstack-nova12:51
*** lpetrut has joined #openstack-nova12:52
*** baoli has joined #openstack-nova12:53
*** lucas-hungry is now known as lucasagomes12:55
*** krtaylor has quit IRC12:56
*** thangp has joined #openstack-nova12:57
*** baoli has quit IRC12:57
*** baoli has joined #openstack-nova12:58
thomasemjaypipes: Ah, I was just messing with extensions in novaclient and was having trouble using novaclient.client as a library for a testing suite I'm writing. It seemed like the extensions really only affected CLI operations, which seems to be supported here: https://github.com/cloudbuilders/simple_nova_extension12:58
thomasemSo, was just wondering if that were the case, if there were plans to improve support for using novaclient as a library, alternatives I'm not aware of?12:58
thomasemimprove support for extensions when using novaclient as a library*12:59
thomasem:)12:59
*** busterswt has joined #openstack-nova13:00
*** derekh_afk is now known as derekh13:00
*** dulek has quit IRC13:00
*** annegentle has quit IRC13:00
*** annegentle has joined #openstack-nova13:02
*** ociuhandu has quit IRC13:05
*** nkrinner has quit IRC13:06
*** krtaylor has joined #openstack-nova13:08
*** rushiagr_away is now known as rushiagr13:09
openstackgerritJorge Niedbalski R. proposed openstack/nova: Detach iSCSI latest path for latest disk  https://review.openstack.org/13538213:11
*** burt has joined #openstack-nova13:12
*** fawadk has joined #openstack-nova13:12
*** fawadkhaliq has quit IRC13:12
*** vladikr has joined #openstack-nova13:15
*** fawadk is now known as fawadkhaliq13:15
*** EmilienM|afk is now known as EmilienM13:15
sdagueeliqiao1: for https://bugs.launchpad.net/nova/+bug/140146913:18
openstackLaunchpad bug 1386535 in OpenStack Compute (nova) "duplicate for #1401469 DELETE/os-serivce/<id> should judge the service state" [Low,Confirmed] - Assigned to Eli Qiao (taget-9)13:18
sdagueis that same error going to come back to the compute?13:18
*** dims has quit IRC13:19
*** dims has joined #openstack-nova13:20
*** dguitarbite has quit IRC13:21
*** dguitarbite has joined #openstack-nova13:23
*** VW_ has quit IRC13:23
*** VW_ has joined #openstack-nova13:24
dansmithjohnthetubaguy: agree that exra_specs is at least as big of a disaster as image_props13:27
*** popw1 has joined #openstack-nova13:27
*** popw has quit IRC13:28
*** bfic has quit IRC13:28
*** popw1 has quit IRC13:29
danpbdansmith: its arguably worse because we have a schedular filter that allows matching on arbitrary user defined extra specs13:29
eliqiao1sdague: hi, yes.13:29
danpbso there's no way we can 100% enforce what's allowed in extra specs :-(13:30
dansmithdanpb: ues13:30
eliqiao1hi dansmith:13:32
dansmitheliqiao1: hello13:32
eliqiao1dansmith: I have some question on https://review.openstack.org/#/c/160075/, which you help reviewd before. can you plese help to take a look at ?13:33
*** dguitarbite has quit IRC13:33
eliqiao1dansmith:  and another object releated patch may request you help to review https://review.openstack.org/#/c/178056/13:34
dansmitheliqiao1: yes, when I have coffee I will13:35
*** dansmith is now known as superdan13:35
*** coolsvap is now known as coolsvap|afk13:37
*** mriedem_away is now known as mriedem13:37
eliqiao1dansmith: cool, thanks :)13:37
sdaguedanpb: what if we had an API end point that exposed what was supported in the system?13:39
sdaguefrom a match perspective13:39
superdanthat's kinda what I want,13:40
superdanbut you have to account for the fact that you might have mixed hypervisors13:40
superdanso it gets a little tricky I think13:40
superdanwhen we have cellsv2,13:40
superdanit might give us a thing to say "if you're going to have heterogeneity , do it on cell boundaries so that we have a thing we can use as a construct in the api to express differences13:41
sdaguesuperdan: right, so that needs a way to be expressed13:41
sdagueis there a reason not to do that with aggregates?13:41
superdanI think that some people don't want to use them for that reason, right?13:41
sdagueit seems weird to require a cell boundary for it13:41
superdanand/or maybe related to aggregates not being supported in current cells13:41
sdagueI don't know13:42
sdagueyeh, well there is that issue13:42
superdanso,13:42
superdancell boundary is not nearly as big of a deal in v2 as v113:42
superdanso it's not as heavy of a requirement13:42
superdanbut yeah, maybe it needs not to be so physical13:42
superdananyway13:42
superdanwe agree on the need13:42
sdagueagreed13:42
*** mtanino has quit IRC13:42
sdagueis there a session to discuss this at summit?13:42
superdanto discuss extra_specs?13:43
superdanI think johnthetubaguy has one yeah13:43
*** erlon has joined #openstack-nova13:44
*** annegentle has quit IRC13:44
*** ociuhandu has joined #openstack-nova13:45
*** annegentle has joined #openstack-nova13:45
*** unicell1 has joined #openstack-nova13:46
*** markus_z has quit IRC13:47
*** coolsvap has joined #openstack-nova13:47
*** unicell has quit IRC13:48
*** eliqiao1 has quit IRC13:48
*** eharney has joined #openstack-nova13:49
*** MVenesio has quit IRC13:50
superdaneliqiao: okay done, sorry that that notification of your reply got missed13:50
*** openstackgerrit has quit IRC13:51
*** dims is now known as dimsum__13:51
*** openstackgerrit has joined #openstack-nova13:51
superdansdague: so, I haven't looked at the code, but I'm confused about how dims got devstack to run his apache config without modifying d-g13:51
superdanor, project-config I guess13:51
*** flwang has quit IRC13:52
dimsum__superdan: :)13:52
*** signed8bit has joined #openstack-nova13:52
*** otter768 has joined #openstack-nova13:52
*** _gryf is now known as _gryf|afk13:53
superdandimsum__: I still think it needs an experimental comment until we're doing it more than once13:53
superdandimsum__: add that and I'll +2 straight away13:53
superdanalso, really happy that it does work13:53
superdansmells like future stuff13:53
*** ociuhandu has quit IRC13:53
dimsum__superdan: in the nova review right?13:54
superdanyeah13:54
dimsum__ack13:54
superdanthe single run of devstack should be enough to justify the devstack change I'd think13:54
dimsum__superdan: yep13:54
johnthetubaguysuperdan: working on that, yeah, trying to get a plan13:55
superdanI think that we could do a lot of good,13:56
superdanif we just expose the things that are supported by the code13:56
superdanand not necessarily by the hypervisor13:56
superdanlike API schema13:57
superdan"these are the things you can say"13:57
superdanand then at least they're forced to ask for the right thing, even if we can't honor them on the current system13:57
*** otter768 has quit IRC13:57
superdanlater we could come back with "supported: True|False" or something13:57
danpbsuperdan: yeah, it would be really nice to expand on the support matrix into something much finer grained that apps can usefully query13:58
johnthetubaguydanpb: and should be linked to tempest tests that validate the tick really, I think14:00
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Helper scripts for running under Apache2  https://review.openstack.org/17967214:01
*** fawadkhaliq has quit IRC14:02
*** htruta has joined #openstack-nova14:04
*** tojuvone has quit IRC14:05
*** ildikov has quit IRC14:05
*** subscope has quit IRC14:06
*** smccully has joined #openstack-nova14:07
*** baoli has quit IRC14:08
*** baoli has joined #openstack-nova14:09
*** fawadkhaliq has joined #openstack-nova14:09
*** b3nt_pin has quit IRC14:10
openstackgerritDan Smith proposed openstack/nova: Refactor tests that use compute's deprecated run_instance() method  https://review.openstack.org/17995014:12
openstackgerritDan Smith proposed openstack/nova: Drop compute RPC 3.x support  https://review.openstack.org/17995114:12
openstackgerritDan Smith proposed openstack/nova: Update kilo version alias  https://review.openstack.org/18112714:12
*** bnemec has joined #openstack-nova14:13
*** b3nt_pin has joined #openstack-nova14:13
*** exploreshaifali has joined #openstack-nova14:16
*** annegentle has quit IRC14:16
*** kiran-r has quit IRC14:16
*** kiran-r has joined #openstack-nova14:17
superdandimsum__: what is it mriedem says?14:19
superdandimsum__: I'm +1 so hard14:19
superdaner, damn14:19
superdan+214:20
*** lpetrut has quit IRC14:20
dimsum__thanks superdan!14:20
superdanyep14:20
mriedemso hard14:20
*** baoli has quit IRC14:20
*** ildikov has joined #openstack-nova14:21
*** gokrokve_ has joined #openstack-nova14:21
*** subscope has joined #openstack-nova14:21
openstackgerritJohn Garbutt proposed openstack/nova-specs: api: relax validation for unversioned requests  https://review.openstack.org/17324314:21
sdaguedimsum__: should those be in cmd instead?14:21
*** baoli has joined #openstack-nova14:21
superdansdague: they're not runnable14:22
dimsum__right, i equated cmd to command line14:22
superdanright, me too14:22
*** ddaskal has joined #openstack-nova14:22
*** kiran-r has quit IRC14:22
*** cfriesen has joined #openstack-nova14:22
superdanI can see them going under doc/ or something, but I think that most python packages arrange them so you can run them straight from wherever they're installed if the package is properly installed14:23
superdan"run" -> include them from apache or whatever14:23
mriedemsuperdan: you want to approve that blueprint first?14:23
sdagueso... maybe I'd feel better about them if there was some other README in that directory so that people didn't try to use it as a generic namespace for other code14:23
mriedemhttps://blueprints.launchpad.net/nova/+spec/run-nova-services-under-apache214:23
superdanthey're kinda config samples, but they also should just work14:23
dimsum__./keystone/server/wsgi.py << that's what the keystone guys use14:24
superdansdague: you're just concerned about the added wsgi directory?14:24
superdanmriedem: so that spec seemed to be a bigger thing to me14:24
superdanmriedem: this is just a couple of example maybe-working wsgi scripts you can use if they work, hence the experimental tag request14:24
*** gokrokve has quit IRC14:25
sdaguesuperdan: and ensuring it's clear that we keep separation between the wsgi helpers and other more generic code related to wsgi14:25
*** baoli_ has joined #openstack-nova14:25
*** apmelton_away is now known as apmelton14:25
superdansdague: okay14:25
superdansdague: I *think* I've seen $repo/wsgi as a pattern before14:25
*** shakamunyi has joined #openstack-nova14:25
superdanbut it doesn't really matter where they are to me14:25
superdanalso fine with a README, although I doubt that will prevent people from polluting it :)14:25
johnthetubaguyso I do wonder if we want a tiny spec for this… just to record the conversation really14:26
superdanjohnthetubaguy: I thought we already agreed it was just a thing,14:26
*** armax has joined #openstack-nova14:26
superdanjohnthetubaguy: and that if we needed changes, we'd spec them14:26
superdanam I imagining that?14:26
sdaguethe README at least helps people trying to do the right thing to remember "oh, right don't stick other stuff in there"14:27
superdansdague: it provides us a thing to link to if we -2 an add there14:27
dimsum__sdague: ack, will add a README14:27
johnthetubaguysuperdan: probably not, I have drowsy today, I just wanted to ask the question14:27
sdagueI also think we probably want to make the paste and conf file readable out of env, because then you could location of them in the apache.conf easily14:28
*** baoli has quit IRC14:28
sdaguewhich is the way I'd expect sites to actually configure it14:28
sdagues/could location/could set location/14:28
*** lpetrut has joined #openstack-nova14:28
*** bilco105 has joined #openstack-nova14:29
openstackgerritJohn Garbutt proposed openstack/nova: devref: add information to clarify nova scope  https://review.openstack.org/17862314:29
*** bilco105 has left #openstack-nova14:29
*** fawadkhaliq has quit IRC14:29
*** exploreshaifali has quit IRC14:29
superdandimsum__: so is the blueprint just for this?14:29
*** BadCub_Hiding is now known as BadCub14:29
openstackgerritJohn Garbutt proposed openstack/nova: devref: update the nova architecture doc  https://review.openstack.org/17737414:29
superdanI swear I thought the blueprint was to cover changes needed, but it doesn't look like it has any detail14:30
openstackgerritJohn Garbutt proposed openstack/nova: devref: discuss current state of error handling  https://review.openstack.org/17863514:30
superdanmriedem: you can drop your -2 now :)14:31
*** gokrokve_ has quit IRC14:31
* mriedem has gone on vacation14:31
*** gokrokve has joined #openstack-nova14:31
*** ddaskal has quit IRC14:32
*** bnemec is now known as beekneemech14:32
mtreinishmriedem: mexico again?14:32
*** emagana has joined #openstack-nova14:32
mriedemwhat am i dropping the -2 for? i mean, i'm being a procedural dick here, i know, but14:33
mriedemit's either part of a blueprint or it's not, right?14:33
openstackgerritBrent Eagles proposed openstack/nova-specs: VIF plug script support for Nova  https://review.openstack.org/16246814:33
dimsum__mriedem: the review points to the blueprint and the blueprint is approved just now by superdan :)14:34
superdanmriedem: I just approved the blueprint, regardless of if it's covered or not14:34
mriedemoh was looking at status14:35
mriedemderp14:35
superdanmriedem: I certainly wasn't asking you to prematurely stop being a procedural dick14:35
mriedemok, good14:35
mriedemit's all i have most days14:35
*** fawadkhaliq has joined #openstack-nova14:35
dimsum__mriedem: i got the rootwrap daemon working as well14:37
mriedem-514:37
*** tsekiyama has joined #openstack-nova14:37
openstackgerritJohn Garbutt proposed openstack/nova: devref: add a description of flavors  https://review.openstack.org/17862814:37
dimsum__not a good idea?14:37
mriedemjoking14:37
dimsum__ha :)14:38
*** sambol has joined #openstack-nova14:38
*** Marga_ has joined #openstack-nova14:38
kashyapmriedem: Heya, when you get a moment, re: this bug - https://bugs.launchpad.net/nova/+bug/126291414:39
openstackLaunchpad bug 1262914 in OpenStack Compute (nova) "Unnecessary data copy during cold snapshot" [Medium,Confirmed]14:39
*** Marga_ has quit IRC14:39
*** zsais has joined #openstack-nova14:39
*** Marga_ has joined #openstack-nova14:39
mriedemok?14:39
kashyapI was just testing it w/ current git,14:40
kashyapI think the complaint there is14:40
kashyapthis part:14:40
kashyap"qemu-img convert -f qcow2 -O qcow2 /home/kashyapc/src/cloud/data/nova/instances/aa20be6e-de39-4a15-9f95-9844ec9af5a9/disk /home/kashyapc/src/cloud/data/nova/instances/snapshots/tmp2h6al2/1e00639002e2420ba3747145f06511d8"14:40
kashyapExected by Nova, when you do a `nova image-create cirrvm1 snap2 --poll` on an offline Nova instance.14:40
kashyapNow, that `convert` command is just a `cp`, as you might observe if you look closely.14:41
*** Guest58709 has joined #openstack-nova14:41
mriedemyeah it's converting qcow2 to qcow2?14:41
mriedemwhy even run it?14:42
superdanwhat if you have three levels deep?14:42
kashyapmriedem: I'm getting there. . . This commit 3a3ad54014:42
superdanpresumably a copy would maintain the linkage to the base base image, but the convert will actually flatten it, right?14:42
kashyap-                                  '-O', 'raw', from_path, tmp_path)14:42
kashyap+                                  '-O', 'qcow2', from_path, tmp_path)14:42
mriedemi'll admit to not being a qemu-img expert14:42
kashyapmriedem: See there? It removed the raw -> qcow2 conversion14:42
*** oro has quit IRC14:43
kashyapNow, both the source _and_ destination are qcow2, the `convert` ends up being a `cp`.14:43
*** browne has joined #openstack-nova14:44
*** matrohon has joined #openstack-nova14:44
kashyapsuperdan: I'm trying to read the surrounding code to see what's going on14:44
mriedemkashyap: shouldn't you be talking to pixelbeat about this then?14:44
kashyapmriedem: Sure, if he's around.14:45
mriedemthe virt disk api and imagebackend/imagecache stuff in nova is not my forte14:45
mriedemi'm actually trying to figure out in our internal ci testing why guestfs is taking 63 seconds on our first test run14:45
kashyapHeh, np. I'm just going through some of the upstream libvirt Nova bugs and stumbled on this14:45
mriedemblowing up rpc14:45
*** packet has joined #openstack-nova14:45
kashyapmriedem: For guestfs performance, here's a very nice guide from Rich Jones - http://libguestfs.org/guestfs-performance.1.html14:46
kashyapIIRC, first run of guestfs takes time14:46
*** bkopilov has quit IRC14:46
mriedemyeah the first run is taking like 76 seconds sometimes, after that it's 12 seconds14:46
mriedembut 76 > rpc_response_timeout (60) so our first test always fails with messagingtimeout14:47
superdandoesn't it start a guest?14:47
mriedemthe test?14:48
superdanI thought libguestfs started a guest to access the fs sometimes14:48
mriedem2015-05-07 11:47:44.032 6723 DEBUG oslo_concurrency.lockutils [req-779e9374-5b28-455b-ad84-6ac97c4d5c52 - - - - -] Lock "fcf1f91d0b7f9c6c30b2658707c638193193d067" released by "copy_qcow2_image" :: held 76.051s inner /usr/lib/python2.7/site-packages/oslo_concurrency/lockutils.py:45614:48
*** artom has quit IRC14:48
*** Mike_D_laptop has quit IRC14:49
kashyapmriedem: superdan Yes, it starts a tiny appliance14:49
kashyapYou can try:14:49
mriedemi'm trying to get a recreate with guestfs.debug=True14:50
kashyapexport LIBGUESTFS_DEBUG=114:50
kashyap export LIBGUESTFS_TRACE=114:50
*** hemnafk is now known as hemna14:50
kashyapAnd, invoke the `guestfish` tool (`libguestfs-test-tool` should do) multiple times and time it.14:50
*** Mike_D_laptop has joined #openstack-nova14:50
mriedemthis is where the majority of the time is taken http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/vfs/guestfs.py#n7914:51
kashyapmriedem: was pinging Rich Jones on #libguestfs,14:53
*** subscope has quit IRC14:54
*** rwmjones has joined #openstack-nova14:54
danpbmriedem: superdan: libguestfs always starts a guest for everything it does14:54
danpbthe actual guest startup overhead is tiny - if kvm is available it'll be 3-4 seconds max14:54
kashyapmriedem: "in openstack terms, the answer is, don't worry about it" :-)14:54
superdandanpb: okay, I thought it could use some userspace filesystem implementations it could handle14:54
danpbif only qemu it will be upto 10 secs14:54
kashyapmriedem: rwmjones is here.14:54
rwmjonestry doing:14:55
rwmjonestime guestfish -a /dev/null run14:55
mriedemdanpb: this would be qemu14:55
danpb/but/ sometimes guestfs has to rebuild the appliance image and that can take a little while14:55
danpbit caches the appliance so you only have that hit once, unless the kernel / rpms / dpkgs installed have changed14:55
rwmjonesright, so very first time it is run after yum or apt-get has installed a package, it has to rebuild the appliance from scratch14:55
danpbin the case of our CI testing, we'll always have that hit once14:55
rwmjoneswhich is like 60 seconds or something14:55
danpbsince the VMs are always clean vms14:55
mriedemrwmjones: 60+ seconds is exactly what i'm seeing14:56
danpbin real world deployments this'll never be an issue14:56
rwmjonesthe docs for this are: http://libguestfs.org/guestfs-performance.1.html & http://libguestfs.org/libguestfs-make-fixed-appliance.1.html14:56
danpbmriedem: what's the problem this is causing ?14:56
mriedemdanpb: i looked at the logs in devstack/tempest runs and on first hit in the compute logs it was still less than a second to startup/use guestfs14:56
mriedemdanpb: on the first instance boot (first test), it locks something up causing an rpc messaging timeout14:56
mriedembecause: 2015-05-07 11:47:44.032 6723 DEBUG oslo_concurrency.lockutils [req-779e9374-5b28-455b-ad84-6ac97c4d5c52 - - - - -] Lock "fcf1f91d0b7f9c6c30b2658707c638193193d067" released by "copy_qcow2_image" :: held 76.051s inner /usr/lib/python2.7/site-packages/oslo_concurrency/lockutils.py:45614:56
rwmjoneshmm, that's annoying :-(14:57
mriedemand the rpc_response_timeout is 6014:57
mriedemafter this first time, tests are fine14:57
rwmjonesso what you could do to work around that:14:57
danpbmriedem: hmm, perhaps we could prime the cache before running the tests14:57
rwmjonesas the same users that runs nova guests (is it root?), do:14:57
rwmjonesguestfish -a /dev/null run14:57
mriedemfyi14:57
mriedem[root@compute-debug-4a971879 ~]# time guestfish -a /dev/null run14:57
mriedemreal    0m54.595s14:57
mriedemuser    0m36.190s14:57
mriedemsys     0m4.707s14:57
danpbeg something in devstack could trigger it14:57
rwmjonesand that will build the appliance, then when nova comes to run, it won't need to build it again14:57
*** nelsnelson has joined #openstack-nova14:57
danpbyeah, make devstack run   that command i reckon14:58
*** READ10 has joined #openstack-nova14:58
rwmjonesnote also the appliance is cached in /var/tmp/.guestfs-<UID>/ so you can actually have a look at it14:58
mriedem^ was as root, let me run as nova14:58
danpbmriedem: you have a pointer to the logs - the logs are a bit misleading as to when guestfs is used14:58
mriedemdanpb: this is internal ci14:58
danpbbut if you point me to them i can confirm it14:58
*** mtecer has joined #openstack-nova14:58
mriedemi could paste something14:59
mriedemif i know what i'm looking for14:59
danpblet me just check my local env to tell you what to look for14:59
mtecerHey guys, I am stuck with Kilo installation, I followed docs, and installing on CensOS 7. Any way I can get help here ? :)15:00
mriedemmtecer: more for #openstack channel15:01
mtecerThank you, even if it is "TRACE nova.api.openstack" ?15:02
mtecerswitching to #openstack now.15:02
*** ajayaa has quit IRC15:02
*** mtanino has joined #openstack-nova15:03
danpbmriedem: 2015-05-08 16:03:59.280 DEBUG nova.virt.disk.vfs.guestfs [req-a3c87284-213b-44d8-8e18-dbf3acb46c29 None None] Setting up appliance for /home/berrange/src/cloud/data/nova/instances/986f7eb0-b818-4ffd-87ff-c5391cbea8e1/disk qcow2 from (pid=22757) setup /home/berrange/src/cloud/nova/nova/virt/disk/vfs/guestfs.py:16915:05
danpbmriedem: 2015-05-08 16:04:23.180 DEBUG nova.virt.disk.vfs.guestfs [req-a3c87284-213b-44d8-8e18-dbf3acb46c29 None None] Tearing down appliance from (pid=22757) teardown /home/berrange/src/cloud/nova/nova/virt/disk/vfs/guestfs.py:22115:05
danpbthat's an example from my local env - 25 second delay here (i did   rm -rf /var/tmp/.guestfs*   to remove the cache)15:05
mriedemdanpb: kvm or qemu?15:05
openstackgerritBrent Eagles proposed openstack/nova-specs: VIF plug script support for Nova  https://review.openstack.org/16246815:06
danpbkvm15:06
rwmjonesdanpb: the supermin in Fedora is much more efficient than supermin in RHEL 7.1 -- it uses librpm to look up the dependencies15:06
mriedemdanpb: ok, we're nested virt here so qemu15:06
danpbthis is on real hardware15:06
danpband using fedora 20 too15:06
mriedemthis is rhel 7.1 btw15:06
*** subscope has joined #openstack-nova15:06
danpblibguestfs-1.26.10-1.fc20.x86_6415:06
*** Mike_D_laptop has quit IRC15:06
rwmjoneswhereas RHEL 7.1's supermin runs 'rpm' commands to follow the dependencies, making it real slow15:06
rwmjonesI think there's a BZ for it, just a sec15:07
danpbbut anyway, if you want to stop this affecting test times, we'd need to prime the cache from devstack.sh15:07
mriedemdanpb: libguestfs-1.28.1-1.18.el7.x86_6415:07
*** tonytan4ever has joined #openstack-nova15:07
rwmjonesno we didn't manage to add the new librpm code in supermin for RHEL 7.215:08
mriedemsupermin5-5.1.10-1.2.el7.x86_6415:08
*** josecastroleon has quit IRC15:08
rwmjoneshang on, I have to check the rhel dist-git15:08
rwmjonesbasically, no15:09
rwmjonesso the only workaround for RHEL 7 is going to be to insert a run of 'guestfish -a /dev/null run' as the nova user, somewhere before you run the tests15:10
openstackgerritjichenjc proposed openstack/nova: Don't update RT status when set instance to ERROR  https://review.openstack.org/18142615:11
*** Nikolay_St has quit IRC15:11
mriedemi should check the fedora 21 job we have to see if this is affected there15:12
kashyapmriedem: I think you also added an F21 job to experimental queue, maybe one of the runs there? https://review.openstack.org/#/c/171795/15:13
*** Marga_ has quit IRC15:13
mriedemkashyap: that's what i'm looking at15:13
mriedemhttp://logs.openstack.org/32/147332/3/experimental/check-tempest-dsvm-f21/67b350d/logs/screen-n-cpu.txt.gz#_2015-04-14_14_22_20_19815:14
mriedemLock "1355138edb53373ed3b690c6853cf94bd2aa4614" released by "copy_qcow2_image" :: held 0.084s15:14
*** ajayaa has joined #openstack-nova15:16
*** belmoreira has quit IRC15:18
*** fawadk has joined #openstack-nova15:20
mriedemrwmjones: i'm seeing this on two different CI systems, both rhel 7.1, always the first test, so yeah - definitely looks like the problem. i can easily throw in su - nova -c 'guestfish -a /dev/null run' and see if that fixes it15:21
rwmjonesmriedem: I think that's your best bet15:21
rwmjonesif it doesn't though, let me know15:22
mriedemi'm not sure why the fedora 21 job in the community ci doesn't hit this also15:22
rwmjonesmriedem: because ..15:22
rwmjones16:06 < rwmjones> danpb: the supermin in Fedora is much more efficient than supermin in RHEL 7.1 -- it uses librpm to look  up the dependencies15:22
rwmjones16:06 < rwmjones> whereas RHEL 7.1's supermin runs 'rpm' commands to follow the dependencies, making it real slow15:22
mriedemrwmjones: so...could i use just install a newer supermin from epel?15:22
mriedemcould rebuild my own i guess from fedora source rpm15:23
rwmjonesfairly sure it's not in epel, but yes the fedora srpm would work on RHEL, although I guess it's best to stick with RHEL only packages and fix them if there's a problem15:23
danpbwell it is still going to have a delay - it doesn't solve it completely15:23
danpbso i still vote for just priming the cache in devstack15:23
rwmjonescorrect, the guestfish command should solve the problem completely15:24
rwmjoneson RHEL & Fedora15:24
*** fawadkhaliq has quit IRC15:24
*** fawadkhaliq has joined #openstack-nova15:25
*** shakamunyi has quit IRC15:25
*** tjones1 has joined #openstack-nova15:25
mriedemyeah that's what i'm going to go with now15:25
openstackgerritMarian Horban proposed openstack/nova: Disassociating of floating IPs corrected  https://review.openstack.org/14935815:26
*** oro has joined #openstack-nova15:26
*** fawadk has quit IRC15:27
superdanjaypipes: wanna compare tab bars?15:28
superdanjaypipes: http://imgur.com/6vlHysd15:28
*** busterswt has quit IRC15:28
danpbsuperdan: hehe, i try to cull tabs when i get to about 100 open tabs :-)15:28
*** Marga_ has joined #openstack-nova15:28
superdanthat's after clearing my morning load15:28
openstackgerritEric Brown proposed openstack/nova: Minor edits to support-matrix doc  https://review.openstack.org/18143215:29
*** busterswt has joined #openstack-nova15:29
superdandanpb: I'm glad there's no tab count displayed in my browser :)15:29
mriedemrwmjones: danpb: kashyap: btw, thanks again - this has been bugging me b/c i was thinking there were lock issues in nova or something here, we just found out last night it was guestfs15:29
superdanmaybe a DHS-style green-orange-yellow-red-black system would be good for tab count15:29
openstackgerritjichenjc proposed openstack/nova: Use instance.save to update instance  https://review.openstack.org/18143415:30
*** VW_ has quit IRC15:30
*** VW_ has joined #openstack-nova15:31
openstackgerritMarian Horban proposed openstack/nova: WIP: Graceful shutdown WSGI server  https://review.openstack.org/12916715:33
jrollsuperdan: you don't have a tab problem until they're so small that favicons don't show :)15:34
superdanjroll: well, that's where I was this morning before I made it through the backlog15:34
*** VW_ has quit IRC15:35
*** mwagner_lap has quit IRC15:35
*** mtecer has quit IRC15:35
superdanjroll: didn't think about capturing it until later, but jaypipes is +3h so was hoping his was already plowed too :)15:35
jrollsuperdan: yep, I know the feeling :P15:35
*** VW_ has joined #openstack-nova15:35
jaypipessuperdan: that's nothin.15:36
superdana browser extension that would pop up a thing saying "I think you need to handle some of the existing tabs before you open more" would be nice15:36
superdanjaypipes: let's see it, else it doesn't exist15:37
*** Marga_ has quit IRC15:37
jaypipessuperdan: replete with timestamp so you know I'm not cheating :) http://i.imgur.com/9W4FUqi.png15:39
superdanjaypipes: yeah, you have a problem15:39
jaypipessuperdan: actually, I'm quite happy with that. used to not be able to see any of the favicons at all...15:39
superdanwhat's the little angry monster about 1/3 in from the right?15:39
superdanheh15:39
kashyapmriedem: Np. And, I'll let you know if I have an update on the other bug I mentioned.15:39
*** Guest58709 is now known as annegentle15:40
*** VW_ has quit IRC15:40
*** VW_ has joined #openstack-nova15:40
*** VW_ has quit IRC15:40
*** VW_ has joined #openstack-nova15:41
danpbsuperdan: btw, i'm wondering if there's anything i missed with this  change https://review.openstack.org/#/c/181086/3/nova/objects/vcpu_model.py15:41
jaypipessuperdan: this one? http://blog.davidvassallo.me/2012/05/05/kvm-brctl-in-linux-bringing-vlans-to-the-guests/15:41
danpbsuperdan: i'm converting some fields from strings to enums and that requires a version bump15:41
superdanjaypipes: heh, yeah15:41
danpbbut AFAICT, there's nothing i need to worry about with back compatibility15:41
danpbsince it doesn't look like Field class names are encoded anywhere15:41
*** jaypipes is now known as leakypipes15:42
superdandanpb: which one is moving?15:42
superdanlooks like they're Enums moving to special Enums, no?15:42
danpbsuperdan: yeah15:42
superdandanpb: you said strings to enums15:43
danpbsorry, it was the fourth patch in the series which moved from String -> Enum subclass15:43
superdanokay, so in general: don't do that. *however*:15:43
superdanif you do and we are *sure* that the enum covers everything it could have ever been, then it should be okay, although it does need a bump15:44
superdanif the enum doesn't, but we can convert it, then we need a subclassed enum that can handle the conversion if we receive an older thing15:44
superdanand if it can't be reliably converted, then it needs to be a new field, i.e. change the name15:45
danpbok, i think we should be safe then - the enum code has implemented the coerce method to canonicalize the few cases we know are unusal15:45
superdanokay15:45
superdanI was just looking at the latest imagemeta one when you pinged15:45
danpbso we're just moving canonicalization from the callers, into the object model15:45
superdanI need to make sure the version check is considering .valid_values15:45
superdancool15:45
danpbyeah there's a test case which checks the valid_values / version hash15:46
*** dsanders has joined #openstack-nova15:46
danpbsuperdan: the existing test_fingerprint() method in test_fields.py does that15:46
openstackgerritjichenjc proposed openstack/nova: Don't update RT status when set instance to ERROR  https://review.openstack.org/18142615:46
superdandanpb: because valid_values is in the repr() for EnumField.. yeah, cool15:47
* danpb is slowly starting to understand this magic properly15:47
*** ildikov has quit IRC15:47
*** zzzeek has joined #openstack-nova15:48
superdanheh15:48
*** VW_ has quit IRC15:48
*** VW_ has joined #openstack-nova15:49
openstackgerritsean mooney proposed openstack/nova: fix AggregateTypeAffinityFilter multi values support  https://review.openstack.org/14188315:52
*** mpavone has quit IRC15:52
*** tdurakov_ has quit IRC15:52
*** VW_ has quit IRC15:52
*** READ10 has quit IRC15:53
leakypipesalex_xu: still around? pls see my last comment on https://review.openstack.org/#/c/177112/. Perfect opportuntity for you or gilliard_afk to push a guideline proposal to the api-wg repo on the 403 vs 400 overquota thing and have a decision made.15:53
*** VW_ has joined #openstack-nova15:53
*** otter768 has joined #openstack-nova15:53
*** VW_ has quit IRC15:54
*** VW_ has joined #openstack-nova15:54
openstackgerritBrent Eagles proposed openstack/nova-specs: VIF plug script support for Nova  https://review.openstack.org/16246815:55
*** openstackstatus has quit IRC15:56
*** Mike_D_laptop has joined #openstack-nova15:56
*** patrickeast has joined #openstack-nova15:57
*** openstackstatus has joined #openstack-nova15:57
*** ChanServ sets mode: +v openstackstatus15:57
*** kylek3h has quit IRC15:57
*** otter768 has quit IRC15:58
*** kylek3h has joined #openstack-nova15:59
*** annegentle has quit IRC16:02
*** jlanoux has quit IRC16:03
*** Nikolay_St has joined #openstack-nova16:06
*** Marga_ has joined #openstack-nova16:07
*** tdurakov_ has joined #openstack-nova16:08
*** unicell1 has quit IRC16:09
*** Mike_D_laptop has quit IRC16:11
*** Nic has joined #openstack-nova16:12
*** armax has quit IRC16:14
*** kiran-r has joined #openstack-nova16:15
*** dan_ has joined #openstack-nova16:15
*** dan_ is now known as Guest6478716:15
*** shakamunyi has joined #openstack-nova16:16
*** apuimedo has quit IRC16:17
*** vilobhmm1 has joined #openstack-nova16:18
*** browne has quit IRC16:21
alex_xuleakypipes: got it, propose for tag guideline about exceed limit right? not the generic explain for 400 vs 403.16:21
*** Marga_ has quit IRC16:21
*** Marga_ has joined #openstack-nova16:22
alex_xuleakypipes: actually sdague said will submit some review for 400 vs 40316:23
*** dboik has joined #openstack-nova16:27
*** dboik_ has quit IRC16:27
superdandanpb: just to clarify, you're going to leave them here, including the compute imports, but with plans to follow up soon and remove the imports, yes?16:28
*** ildikov has joined #openstack-nova16:28
*** Mike_D_laptop has joined #openstack-nova16:28
superdanalaski: I'm super excited that nova-api seems to maybe work under apache16:29
danpbsuperdan: yep, leave them in place now and once this series is merged i'll submit patches to move the constants into the enum16:29
superdanalaski: any chance you guys will start smoke testing that soon?16:29
*** exploreshaifali has joined #openstack-nova16:29
danpbsuperdan: as the latter will result in a fairly large diff for the virt driver callers16:29
superdandanpb: okay16:29
*** dboik_ has joined #openstack-nova16:30
alaskisuperdan: it's awesome to see.  with the summit upcoming and a few other things in motion atm I don't know about soon, but soon-ish I hope16:30
*** Marga_ has quit IRC16:30
*** tdurakov_ has quit IRC16:30
superdanalaski: well, I didn't mean in the next week, but yeah16:30
superdancool16:30
*** openstack-nova14 has joined #openstack-nova16:31
*** tdurakov_ has joined #openstack-nova16:33
*** dboik has quit IRC16:33
*** shakamunyi has quit IRC16:33
openstack-nova14Can somebody help me I followed these instructions from mirantis https://www.mirantis.com/openstack-portal/express-openstack-portal/quickly-create-openstack-image/ to built my own Centos 5.11.qcow2 but when i import in glance and start VM i get :unable to access resume device (/dev/volgroup00/logvol01) qcow2  I am unable to find an answer googling16:34
openstack-nova14Please, advise if this is not the right forum to ask this question16:34
mriedemopenstack-nova14: see topic please16:34
*** Sukhdev has joined #openstack-nova16:34
*** openstack-nova14 has quit IRC16:35
openstackgerritjichenjc proposed openstack/nova: Ignore Cinder error when shutdown instance  https://review.openstack.org/18145416:36
openstackgerritjichenjc proposed openstack/nova: Add DiskNotFound and VolumeNotFound test  https://review.openstack.org/18145516:36
*** unicell has joined #openstack-nova16:37
*** dboik_ has quit IRC16:39
*** Marga_ has joined #openstack-nova16:39
*** gokrokve has quit IRC16:40
*** dboik has joined #openstack-nova16:40
*** yamahata has quit IRC16:41
*** lucasagomes is now known as lucas-beer16:41
*** Mike_D_laptop has quit IRC16:42
*** Marga_ has quit IRC16:43
*** harlowja has quit IRC16:44
*** Mike_D_laptop has joined #openstack-nova16:44
*** harlowja has joined #openstack-nova16:44
kashyapdansmith: or anyone, when you have a moment, can you validate if I'm understanding of this code is correct here: https://bugs.launchpad.net/nova/+bug/1262914/comments/416:45
openstackLaunchpad bug 1262914 in OpenStack Compute (nova) "Unnecessary data copy during cold snapshot" [Medium,Confirmed]16:45
*** Nic has quit IRC16:46
*** tjones1 has quit IRC16:46
superdankashyap: dansmith doesn't work on fridays16:46
kashyapDarn16:46
kashyap:-)16:46
*** Nic has joined #openstack-nova16:46
superdanit's fine with me, he's a jerk16:46
*** packet has quit IRC16:46
* kashyap assumed (wrongly, as you can see) he _did_ make a mental habit of referring to your other nick on Fridays :-)16:47
*** shakamunyi has joined #openstack-nova16:47
kashyapsuperdan: Sheesh, Launchpad screwed up my formatting that I did locally in plain text :(16:48
superdankashyap: like I said before, I think the convert is not necessarily just a copy16:48
*** armax has joined #openstack-nova16:49
kashyapsuperdan: Sure. /me recalls reading in some source file where it flattens a chain to a single iage.16:49
superdanright16:50
*** tjones1 has joined #openstack-nova16:50
*** fawadkhaliq has quit IRC16:50
superdanyeah, just verified16:50
*** achanda has joined #openstack-nova16:50
kashyapIf the source file were a raw file, of course, the convert will move it to qcow2, thought I noted that there too16:51
*** fawadkhaliq has joined #openstack-nova16:51
*** tjones1 has quit IRC16:51
kashyapsuperdan: Do you see what is the reporter complaining there, then?16:52
*** tjones1 has joined #openstack-nova16:52
superdanI'm reading16:52
kashyapThe "unnecessary copy" he's referring to16:52
*** ferest has quit IRC16:52
*** derekh has quit IRC16:52
*** rushiagr is now known as rushiagr_away16:53
superdankashyap: I'm trying to ascertain what your point is.. you think it is unnecessary or it's not?16:53
superdanhe asserts it's unnecessary16:53
superdanI think it's necessary16:53
kashyapsuperdan: I don't think it's unneccessary16:53
kashyapYeah, so I agree with you.16:53
superdanhah, okay16:54
* superdan works out the boolean logic there16:54
kashyap:-) /me just felt annoyed reading just prose without a pointer to a test16:54
superdanI also assume the point of the copy is so that the instance can immediately be restarted while the upload proceeds16:54
*** danpb has quit IRC16:54
*** jogo is now known as fashgordon16:55
kashyapYeah (althogh, I didn't think of it until you mentioned it).16:55
superdanfashgordon: going a little different today?16:55
*** mwagner_lap has joined #openstack-nova16:55
*** fawadkhaliq has quit IRC16:55
kashyapAnyhow, have a fine weekend folks!16:56
*** fashgordon is now known as flashgordon16:56
*** alaski is now known as lascii16:56
superdankashyap: commented16:57
kashyapCool, thank you.16:57
*** gokrokve has joined #openstack-nova16:57
*** gokrokve has quit IRC16:58
*** gokrokve has joined #openstack-nova16:58
*** annegentle has joined #openstack-nova16:59
leakypipesalex_xu: k, that works too :)17:02
*** Marga_ has joined #openstack-nova17:03
*** Marga_ has quit IRC17:03
*** Marga_ has joined #openstack-nova17:03
*** yamahata has joined #openstack-nova17:06
*** bkopilov has joined #openstack-nova17:06
openstackgerritjichenjc proposed openstack/nova: Remove policy check security_group_default_rule  https://review.openstack.org/18146717:06
*** shakamunyi has quit IRC17:07
*** browne has joined #openstack-nova17:07
openstackgerritjichenjc proposed openstack/nova: Remove policy check security_group_default_rule  https://review.openstack.org/18146717:08
*** wasmum has quit IRC17:09
*** subscope has quit IRC17:09
*** albertom has joined #openstack-nova17:14
*** isd has joined #openstack-nova17:16
*** shakamunyi has joined #openstack-nova17:18
leakypipesyay, melwitt gains +W super-powers.17:20
*** IanGovett1 has quit IRC17:20
*** annegentle has quit IRC17:21
*** IanGovett has joined #openstack-nova17:22
*** shakamunyi has quit IRC17:23
*** matrohon has quit IRC17:27
*** artom has joined #openstack-nova17:30
tjones1woo hoo - congrats melwitt17:31
*** isd has quit IRC17:32
*** Mike_D_laptop has quit IRC17:33
*** shakamunyi has joined #openstack-nova17:34
superdanwoot17:34
*** albertom has quit IRC17:34
superdanmelwitt: easy test-only thing to test your +W on: https://review.openstack.org/#/c/179949/5 :D17:36
*** penick has joined #openstack-nova17:36
*** fawadkhaliq has joined #openstack-nova17:37
*** isd has joined #openstack-nova17:38
*** bkopilov has quit IRC17:39
melwittthanks all :D17:39
*** Sukhdev has quit IRC17:40
flashgordonmelwitt: welcome!17:40
*** Sukhdev has joined #openstack-nova17:41
*** IanGovett has quit IRC17:42
*** sandywalsh has left #openstack-nova17:43
*** annegentle has joined #openstack-nova17:43
russellbmelwitt: nice work :)17:44
*** ildikov has quit IRC17:45
superdandibs on the tweet, as soon as I learn her nick17:45
*** IanGovett has joined #openstack-nova17:45
*** browne has quit IRC17:45
melwittthanks flashgordon russellb :)17:46
*** flwang has joined #openstack-nova17:46
penickCongrats melwitt!17:47
*** ildikov has joined #openstack-nova17:53
melwittthanks penick :)17:53
*** otter768 has joined #openstack-nova17:54
*** gokrokve has quit IRC17:58
*** browne has joined #openstack-nova17:59
*** otter768 has quit IRC17:59
*** ildikov has quit IRC18:00
*** alexpilotti has quit IRC18:01
*** busterswt has quit IRC18:02
*** alexpilotti has joined #openstack-nova18:02
*** claudiub has quit IRC18:04
*** busterswt has joined #openstack-nova18:04
*** alexpilotti has quit IRC18:07
figleafyay melwitt!18:09
*** ajayaa has quit IRC18:09
*** wasmum has joined #openstack-nova18:11
*** ijw has joined #openstack-nova18:11
*** coolsvap has quit IRC18:15
*** ildikov has joined #openstack-nova18:15
superdanall this yay-ing and yet no RTs: https://twitter.com/get_offmylawn/status/59673315975404339218:16
*** kiran-r has quit IRC18:17
*** isd has quit IRC18:17
*** vilobhmm1 has quit IRC18:19
*** vilobhmm1 has joined #openstack-nova18:19
*** rwsu_ has quit IRC18:24
*** Marga_ has quit IRC18:30
figleafsuperdan: ok, happy now?18:30
*** Marga_ has joined #openstack-nova18:30
*** redbeard has quit IRC18:31
*** annegentle has quit IRC18:31
superdanfigleaf: well, I'm never "happy", but I'm happier, yes18:31
figleafsuperdan: every little bit helps18:31
* superdan nods18:32
*** vilobhmm1 has quit IRC18:33
*** wasmum has quit IRC18:33
*** vilobhmm1 has joined #openstack-nova18:33
russellbmelwitt: did you sign up for twitter for superdan's tweet? heh18:33
*** isd has joined #openstack-nova18:34
melwittrussellb: it looks like I did but no :P I made the account a long time ago and realized I couldn't think of things to say18:34
*** Marga_ has quit IRC18:35
*** Marga_ has joined #openstack-nova18:35
superdanit's okay, I'm pretty sure I had to sign up for twitter to retweet something of russellb's18:36
superdanor something like that18:36
russellbha, or at least put up with some harassment until you signed up18:36
superdanI don't recall18:37
russellbI also do not recall.18:37
*** Longgeek has quit IRC18:37
superdaneither way, not my proudest moment, so let's not talk about it18:37
*** Marga_ has quit IRC18:37
melwittlol18:37
*** Longgeek has joined #openstack-nova18:37
*** exploreshaifali has quit IRC18:38
superdan:)18:38
*** Longgeek has quit IRC18:39
*** marun has joined #openstack-nova18:39
*** Longgeek has joined #openstack-nova18:39
*** fawadkhaliq has quit IRC18:39
*** vilobhmm1 has quit IRC18:40
*** vilobhmm1 has joined #openstack-nova18:42
superdanyeah, melwitt has been on longer than I have18:43
*** vilobhmm1 has quit IRC18:43
superdanand I think the fact that she has zero tweets goes pretty well with her personality :)18:43
*** armax has quit IRC18:43
melwitthaha :D18:43
*** vilobhmm1 has joined #openstack-nova18:44
flashgordonsuperdan dimsum__: https://review.openstack.org/#/c/179672/6 I think it would be really neet to see that become the default devstack setup this cycle18:44
superdanmelwitt: we're both ahead of mriedem though, maybe russellb should work on him a bit18:45
superdanflashgordon: that'd be cool18:45
russellbmriedem: you're not on twitter yet?!18:45
superdanhe *just* got a smartphone like six months ago18:45
* russellb proposes an amendment to the TC charter stating that a twitter account is a pre-req for ATC status18:46
superdanheh18:46
flashgordondimsum__ superdan: I guess the next step is to get a non-voting job for a few weeks running to burn in the code and then switch over the main jobs?18:46
melwittsuperdan: twitter's missing out on a gold mine then18:46
*** isd has quit IRC18:46
russellbsuperdan: so basically just a year behind you then18:46
superdanmelwitt: ikr? imagine all the nothing-but-link-to-youtube-movie-clip tweets18:46
superdanrussellb: that's a falsehood, I had a smartphone when you were in diapers18:46
melwittlol!18:46
superdanor what qualified for a smartphone at the time18:47
superdanhttp://www.palminfocenter.com/images/Treo650_apps_S.jpg18:47
russellbso you're saying you're old?18:47
russellbheh, never had one of those18:47
superdansee!18:47
superdanI still have my Pilot 500018:48
*** lpetrut has left #openstack-nova18:48
melwittleakypipes: I find I can't +2 a thing I +1 in the past https://review.openstack.org/#/c/159626/18:49
superdanmelwitt: log out of gerrit18:49
leakypipesmelwitt: yeah, what superdan said :)18:49
*** armax has joined #openstack-nova18:49
* superdan hushes the crowd in anticipation18:49
melwittit didn't work :\18:50
* superdan narrates in a soft voice: "We see now as melwitt refreshes her gerrit cookie..."18:50
superdanwat18:50
tjones1lol18:50
russellbflop18:50
mriedemwhat?18:50
superdanmelwitt: can you +2 on other things?18:50
mriedemoh twitter18:50
*** alanf-mc has joined #openstack-nova18:51
melwittsuperdan: yeah, I did the spawnfixture one already18:51
superdanoh, I missed it18:51
superdanmelwitt: I don't think "have previously +1d" is a thing18:51
superdanmelwitt: maybe drop your vote to zero first?18:51
melwittsuperdan: that did it. thanks :)18:52
superdansure enough: https://review.openstack.org/#/c/179949/18:52
superdanmelwitt: weird, but cool18:52
superdanwell, thanks gerrit, totally screwed the drama18:52
jrollsuperdan: I'm super shocked by your twitter username18:52
superdanjroll: really?18:53
superdanget_thefuckoffmylawn seemed a tad much18:53
jrollSUPER SHOCKED.18:53
jrollheh18:53
b3nt_pindo you even have a lawn?18:53
* b3nt_pin ducks18:53
superdanb3nt_pin: I do, and I have literally shouted at children to get off it18:54
b3nt_pinsuperdan, awesome!18:54
*** fawadkhaliq has joined #openstack-nova18:54
* b3nt_pin been there - is probably known as the crazy grumpy old guy on the corner18:54
openstackgerritVilobh Meshram proposed openstack/nova-specs: Replace Zookeeper, Memcache servicegroup driver with respective Tooz drivers  https://review.openstack.org/13860719:00
*** isd has joined #openstack-nova19:01
*** browne has quit IRC19:03
*** browne has joined #openstack-nova19:04
*** wasmum has joined #openstack-nova19:06
*** Marga_ has joined #openstack-nova19:09
*** penick has quit IRC19:09
*** salv-orlando has quit IRC19:12
dimsum__flashgordon: sounds like a plan!19:17
*** penick has joined #openstack-nova19:19
*** eglynn has quit IRC19:25
*** albertom has joined #openstack-nova19:25
*** mwagner_lap has quit IRC19:26
*** lpetrut has joined #openstack-nova19:26
*** BobH has quit IRC19:27
*** BobH has joined #openstack-nova19:27
*** BobH has quit IRC19:28
vilobhmm1leakypipes, harlowja : https://blueprints.launchpad.net/nova/+spec/servicegroup-api-control-plane19:31
leakypipesvilobhmm1: rock on. :)19:31
leakypipesthx !19:31
harlowjavilobhmm1 cool, i think the idea would be to have specs connected to that (with details of each 'finding')19:31
vilobhmm1harlowja : agree19:31
harlowjaarchaology man19:31
harlowja:)19:32
harlowjaindiana vilobhmm1 jones19:32
vilobhmm1harlowja : :P19:32
vilobhmm1leakypipes : please let me know your feedback on https://review.openstack.org/#/c/138607/ whenever you are free…updated the title as we discussed19:33
leakypipesvilobhmm1: will do, thank you!19:33
*** vladikr has quit IRC19:33
vilobhmm1thanks19:33
*** Mike_D_laptop has joined #openstack-nova19:36
*** salv-orlando has joined #openstack-nova19:38
*** Sukhdev has quit IRC19:41
*** apuimedo has joined #openstack-nova19:41
*** busterswt has quit IRC19:43
*** penick has quit IRC19:48
*** bkopilov has joined #openstack-nova19:49
*** penick has joined #openstack-nova19:53
*** Mike_D_laptop has quit IRC19:54
*** apuimedo has quit IRC19:54
flashgordonsuperdan: you are aware you reviewed an old version of https://review.openstack.org/#/c/178623/ right?19:55
*** otter768 has joined #openstack-nova19:55
superdanflashgordon: yes19:55
superdanI'm doing the current one now19:55
superdanflashgordon: I wanted to agree and disagree with your points19:56
flashgordonsuperdan: just checking19:56
*** dprince has quit IRC19:57
*** otter768 has quit IRC20:00
*** busterswt has joined #openstack-nova20:02
*** packet has joined #openstack-nova20:06
*** ijw has quit IRC20:07
*** alanf-mc has joined #openstack-nova20:10
*** Sukhdev has joined #openstack-nova20:11
*** prometheanfire has joined #openstack-nova20:13
prometheanfirenova requires barbicanclient to generate configs, but barbicanclient is only in test-requirements for kilo20:13
*** VW_ has quit IRC20:15
*** VW_ has joined #openstack-nova20:15
flashgordonmelwitt: I assume you will be going back over your +1s and turning some of those into +2s?20:16
*** VW_ has quit IRC20:16
*** VW_ has joined #openstack-nova20:17
flashgordonprometheanfire: you mean to generate sample configs20:17
*** shakamunyi has quit IRC20:17
*** signed8b_ has joined #openstack-nova20:19
prometheanfireyes20:19
flashgordonprometheanfire: I think that is expected, and don't think we can get around taht20:19
*** tpeoples has joined #openstack-nova20:19
prometheanfirejust move barbicanclient from test-reqs to reqs?20:20
flashgordonwe can make the script gracefully fall back and say options for X are missing though20:20
flashgordonprometheanfire: building sample config doesn't mean its required by default20:20
*** signed8bit has quit IRC20:20
prometheanfirefailing gracefully for those options works too20:20
prometheanfireright now it fails with an empty file20:21
flashgordondimsum__: ^20:21
flashgordonprometheanfire: can you file a bug against nova and I can triage it and get it to the right place for you20:21
flashgordonprometheanfire: but in general to build a sample config you will need test-requirements installed20:21
prometheanfirewell, in nova's case it's only barbicanclient that's missing :P20:22
prometheanfireI added it to package dependencies though20:22
flashgordonsure, but in general there will be other cases20:22
tpeoplesmriedem: for https://review.openstack.org/#/c/132408/1/nova/virt/libvirt/driver.py, why did we not safe decode the xml variable on line 4372?  i am seeing resumes fail for VMs with unicode names with UnicodeDecodeErrors on that line.  using safe_decode on "xml" fixes the issue.20:22
prometheanfireright20:22
*** sdake_ has joined #openstack-nova20:22
flashgordonso barbican is only required if you are using the encrypted volume stuff20:22
flashgordonmaking the client a dependency to enable users to use that option makes sense to me20:23
flashgordonprometheanfire: I assume this is for gentoo?20:23
prometheanfireofc20:23
flashgordonofc?20:23
prometheanfireofcourse20:23
flashgordonahh20:23
mriedemtpeoples: idk, b/c i guess it wasn't the issue we hit at the time of the fix?20:24
flashgordonprometheanfire: once you file a bug send me the link and I will get it triaged20:24
prometheanfireworking on it now20:24
*** annegentle has joined #openstack-nova20:25
tpeoplesmriedem: i guess i would have thought if the other two places were failing the first would have as well, that's why i asked. thought maybe there was a reason we didn't include it there too. going to open a bug and submit a fix myself, jfyi20:25
openstackgerritArtom Lifshitz proposed openstack/nova-specs: Support --all-tenants in floatingips-list  https://review.openstack.org/17138920:25
*** annegentle has quit IRC20:26
*** sdake has quit IRC20:26
mriedemtpeoples: sounds good20:26
*** annegentle has joined #openstack-nova20:26
mriedemtpeoples: it's in a condition so maybe the original bug didn't have xml passed in20:26
mriedemnot sure, it was awhile ago20:26
tpeoplesah, yeah, that's likely it20:27
mriedemtpeoples: or maybe different versions of libvirt or something20:27
tpeoplesthanks20:27
flashgordondimsum__: every  time I use oslo.config it makes me sad20:27
prometheanfirehttps://bugs.launchpad.net/nova/+bug/145326820:27
openstackLaunchpad bug 1453268 in OpenStack Compute (nova) "dynamic nova.conf generation does not work without python-barbicanclient installed" [Undecided,New]20:27
melwittflashgordon: yes, have changed a few to +2 already so far20:29
flashgordonmelwitt: cool20:30
prometheanfireflashgordon: I'll leave that to you then :D20:30
flashgordonprometheanfire: if you want to fix it by all means go for it20:31
prometheanfireat the moment I have a long to list :P20:31
flashgordonprometheanfire: no worries, I tagged it as low-hanging-fruit since it should be pretty straight forward to fix  and hopefully someone will pick it up20:32
mriedemwtf, should tox -e genconfig have python-barbicanclient in the venv?20:34
mriedemb/c it's in test-requirements.txt20:34
flashgordonmriedem: https://twitter.com/OpenStackMN/status/595738339623489536/photo/120:34
prometheanfireworksforme20:34
mriedemflashgordon: i'm lighting up the twitters20:34
flashgordonmriedem: I don't think gentoo wan'ts to use tox20:34
flashgordonbeing distros and pip don't get along20:34
mriedemflashgordon: mestery mangled my last name though20:34
mesterymriedem: You need a less complicated last name :P20:35
*** mriedem is now known as mriderman20:35
mesterymriderman: Seriously, good talk Tuesday night20:35
flashgordonhaha20:35
mridermanmestery: you're not the first and won't be the last person to throw extra r's in my last name20:35
melwittcool pic20:35
prometheanfirewe install from pypi just fine though :P20:36
superdanmestery: oh my god, don't inflate his head any further20:36
mesterylol20:36
mridermani had a manager for 4 years that couldn't pronounce my last name20:36
mriderman"this is matt riderman, he works on something or other, idk"20:36
prometheanfirewe just run this manually, instead of tox20:36
mtreinishsuperdan: just come to the talk at summit so you can boo him20:36
prometheanfire        ./tools/config/generate_sample.sh -b ./ -p nova -o etc/nova || die20:36
superdanmtreinish: will do20:37
flashgordonmtreinish: do we get to heckle?20:37
mridermanprometheanfire: does it work with tox?20:37
prometheanfireprobably20:37
prometheanfirebut we don't use tox20:37
mridermanprometheanfire: if it works with tox it's not a bug20:37
prometheanfireeven for testing20:37
mridermanneither do we20:37
*** ijw has joined #openstack-nova20:37
mridermansphinx docs builds won't work either for you20:37
flashgordonmtreinish: link to the talk so I can add it to my cal?20:37
mridermanif you don't have everything installed20:37
* mriderman wishes we used tox in our rpm builds for docs20:37
prometheanfirewe have some packages that split docs out from testing, but nova isn't one of them20:38
mtreinishflashgordon: http://openstacksummitmay2015vancouver.sched.org/event/c3cda30c3005eb76818688a4d15ad963#.VU0euFn99jE20:38
mridermanprometheanfire: so what i'm saying is, this isn't nova's problem20:38
mtreinishflashgordon: feel free to heckle20:38
*** kaisers has quit IRC20:38
flashgordonmtreinish: why didn't you include docker and NFV in the title?20:39
flashgordonjust for fun20:39
prometheanfiremriderman: nova probably just needs that split out20:39
prometheanfiremriderman: feel free to comment on the bug20:39
mridermanprometheanfire: i did, i marked it invalid20:39
mtreinishflashgordon: heh, I think one of the titles we were discussing had all the buzz words20:39
mridermanflashgordon: i think i actually suggested that20:39
prometheanfireyou are such a pal20:39
flashgordonmtreinish: I can't make it :( won't be around on Monday have a graduation to go to20:39
mridermanflashgordon: debugging NFV in your docker containers for great enterprise bottom line success20:39
*** thangp has quit IRC20:40
mriderman2.020:40
flashgordonmriderman: exactly20:40
mridermanprometheanfire: well this is like saying, nova unit tests don't work w/o installing all of the python dependencies in my site-packages20:40
*** tpeoples has quit IRC20:40
*** fawadkhaliq has quit IRC20:41
*** kaisers has joined #openstack-nova20:41
prometheanfiremaybe I should make a spec stating that example conf generation should be standard and not testing20:41
mridermanprometheanfire: that's cross project you know, right?20:41
*** tpeoples has joined #openstack-nova20:41
mridermani'm not saying this is awesome or anything20:41
mridermanit blows20:41
prometheanfiredoesn't change my opinion on it :P20:42
mridermanthe alternative probably is nova.conf doesn't have barbicanclient options in it20:43
mridermanin the sample anyway20:43
prometheanfireright20:43
mridermanand then someone will open a bug b/c they are missing20:43
prometheanfirecircles20:43
mridermanyes20:44
prometheanfirewhich is why a spec could be nice20:44
mridermanthe config sample is always a catch 22 pain in everyone's ass20:44
prometheanfireya20:44
prometheanfireI don't disagree20:44
mridermanso maybe there is a spec for people that need the config sample generated but can't use tox, and therefore can only get the options defined within nova into nova.conf20:46
mridermaneverything external to nova is ignored20:46
* prometheanfire shrugs20:46
prometheanfirewe could do what some distros do and ship ancient configs isntead of autogeneration20:47
mridermanwhich is misleading and a problem in it's own right20:47
*** shakamunyi has joined #openstack-nova20:48
mridermanb/c then you try using a deprecated option and things don'twork and you open support tickets20:48
*** armax has quit IRC20:48
prometheanfireyep20:48
*** zsais has quit IRC20:48
prometheanfireI think end users should be able to auto-gen an example or have one provided20:48
prometheanfireI'm fine with leaving it as is, we (gentoo) are going to dynamically generate the config at install time, we'll have a doc use flag for it and seperate out only the needed deps from test-requirements20:51
*** htruta has quit IRC20:51
*** pixelbeat has quit IRC20:52
*** jichen has quit IRC20:52
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Drop explicit suds dependency  https://review.openstack.org/18155420:53
*** penick has quit IRC20:54
*** mriderman has quit IRC20:54
*** mriedem has joined #openstack-nova20:59
*** raildo has quit IRC21:00
*** prometheanfire has left #openstack-nova21:00
mgagneRegarding bug #1328546 can someone point me out where (commit id) it got fixed in Nova?21:01
openstackbug 1328546 in neutron "Race condition when hard rebooting instance" [Medium,Invalid] https://launchpad.net/bugs/1328546 - Assigned to Li Ma (nick-ma-z)21:01
*** signed8bit has joined #openstack-nova21:02
*** shakamunyi has quit IRC21:03
*** penick has joined #openstack-nova21:04
*** VW_ has quit IRC21:06
*** signed8b_ has quit IRC21:06
*** VW_ has joined #openstack-nova21:06
*** VW_ has quit IRC21:06
*** lascii is now known as alaski21:08
*** eharney has quit IRC21:09
*** tdurakov_ has quit IRC21:10
*** Longgeek has quit IRC21:10
*** apuimedo has joined #openstack-nova21:11
*** mriedem has quit IRC21:11
*** mriedem has joined #openstack-nova21:12
*** apuimedo has quit IRC21:13
*** apuimedo has joined #openstack-nova21:14
*** emagana has quit IRC21:14
openstackgerritJoel Coffman proposed openstack/nova: libvirt: Disconnect dm-crypt on instance suspend/stop  https://review.openstack.org/14148521:16
openstackgerritJoel Coffman proposed openstack/nova: libvirt: Clean up unit tests for _hard_reboot  https://review.openstack.org/17739821:16
openstackgerritJoel Coffman proposed openstack/nova: libvirt: Remove unnecessary JSON conversions  https://review.openstack.org/17743721:16
*** artom has quit IRC21:17
*** sambol has quit IRC21:18
*** lucas-beer has quit IRC21:18
*** ildikov has quit IRC21:21
*** shakamunyi has joined #openstack-nova21:21
dimsum__flashgordon: responded on the oslo-config-generator review. we need to take a stand on live with what we have (obsolete generator.py) or pick the newer style explicit config file and multiple entrypoints (list_opts). if it is latter, we should actively define smaller groups of related options to make things better slowly. if it is former, we just abandon the patch and call it a day21:23
*** cbader has quit IRC21:25
*** sdake_ has quit IRC21:26
*** figleaf is now known as edleafe21:27
*** signed8bit has quit IRC21:29
*** rwsu has joined #openstack-nova21:30
*** rwsu has quit IRC21:30
*** rwsu has joined #openstack-nova21:31
*** armax has joined #openstack-nova21:31
*** ildikov has joined #openstack-nova21:34
*** BobBall is now known as BobBall_AWOL21:36
*** VW_ has joined #openstack-nova21:37
*** rwsu has quit IRC21:39
melwittI saw this critical bug related to sample generation but don't understand it well enough. has anyone else taken a look at the patch? https://review.openstack.org/#/c/177949/21:41
*** erlon has quit IRC21:41
*** busterswt has quit IRC21:42
superdanmelwitt: I think the general problem is that our configs don't get oslo options included21:42
superdanmelwitt: I think there is a lot wrapped up in this (see dimsum__'s comment above)21:43
superdanand I don't follow it close enough to really know what the deal is21:43
melwitthm, okay21:44
superdanhow's that for unhelpful? :)21:45
melwittit's good! ;)21:45
superdandimsum__ can probably give a more complete overview21:45
*** albertom has quit IRC21:46
*** aysyd has quit IRC21:49
*** mriedem has quit IRC21:49
*** rfolco has quit IRC21:50
*** baoli_ has quit IRC21:53
*** otter768 has joined #openstack-nova21:56
*** lpetrut has quit IRC21:58
*** rwsu has joined #openstack-nova21:58
*** nelsnelson has quit IRC21:59
*** otter768 has quit IRC22:01
cfriesenthis is offtopic, but is anyone aware of a linux feature that would cause a file to appear different when read?  The overall size is the same, but the contents look different.  Sort of like address space randomization but when reading the file, not when executing it.\22:01
bauzasmelwitt: I was off today (WWII day holiday ) but congrats :)22:04
*** shakamunyi has quit IRC22:04
*** barra204 has joined #openstack-nova22:05
melwittthank you bauzas :)22:05
*** browne1 has joined #openstack-nova22:06
*** browne has quit IRC22:07
*** oro has quit IRC22:07
*** EmilienM is now known as EmilienM|afk22:08
*** Sukhdev has quit IRC22:11
*** alexpilotti has joined #openstack-nova22:14
*** beekneemech has quit IRC22:18
*** penick has quit IRC22:19
*** dimsum__ has quit IRC22:21
openstackgerritmelanie witt proposed openstack/nova: Send Instance object to cells instance_update_at_top  https://review.openstack.org/17651822:22
*** annegentle has quit IRC22:23
*** ijw has quit IRC22:24
openstackgerritmelanie witt proposed openstack/nova: Send Instance object to cells instance_update_at_top  https://review.openstack.org/17651822:27
*** busterswt has joined #openstack-nova22:28
*** tpeoples has quit IRC22:31
*** ijw has joined #openstack-nova22:32
*** burt has quit IRC22:35
*** busterswt has quit IRC22:35
*** annegentle has joined #openstack-nova22:36
*** busterswt has joined #openstack-nova22:38
*** Nic has quit IRC22:39
*** busterswt has quit IRC22:42
*** VW_ has quit IRC22:44
*** VW_ has joined #openstack-nova22:45
*** isd has quit IRC22:47
*** VW_ has quit IRC22:53
*** vipuls has joined #openstack-nova22:53
*** VW_ has joined #openstack-nova22:53
*** VW_ has quit IRC22:54
*** VW_ has joined #openstack-nova22:54
*** rwsu has quit IRC22:56
*** VW_ has quit IRC22:59
*** VW_ has joined #openstack-nova22:59
*** VW_ has quit IRC23:02
*** VW_ has joined #openstack-nova23:02
*** dboik has quit IRC23:03
*** vilobhmm1 has quit IRC23:04
*** dboik has joined #openstack-nova23:04
*** vilobhmm1 has joined #openstack-nova23:04
*** pixelbeat has joined #openstack-nova23:06
*** VW_ has quit IRC23:08
*** VW_ has joined #openstack-nova23:08
*** annegentle has quit IRC23:09
*** VW_ has quit IRC23:12
*** VW_ has joined #openstack-nova23:12
*** VW_ has quit IRC23:14
*** VW_ has joined #openstack-nova23:14
*** patrickeast has quit IRC23:15
*** packet has quit IRC23:15
*** VW_ has quit IRC23:16
*** VW_ has joined #openstack-nova23:17
*** VW_ has quit IRC23:17
*** VW_ has joined #openstack-nova23:18
*** rwsu has joined #openstack-nova23:22
*** barra204 has quit IRC23:22
openstackgerritmelanie witt proposed openstack/nova: Send Instance object to cells instance_update_at_top  https://review.openstack.org/17651823:24
*** IanGovett1 has joined #openstack-nova23:24
*** VW_ has quit IRC23:25
*** VW_ has joined #openstack-nova23:26
*** IanGovett has quit IRC23:27
*** VW_ has quit IRC23:31
*** VW_ has joined #openstack-nova23:31
*** marun has quit IRC23:32
*** VW_ has quit IRC23:34
*** VW_ has joined #openstack-nova23:34
*** mestery has quit IRC23:34
*** heyongli has quit IRC23:36
*** VW_ has quit IRC23:37
*** heyongli has joined #openstack-nova23:37
*** VW_ has joined #openstack-nova23:37
*** VW_ has quit IRC23:40
*** VW_ has joined #openstack-nova23:41
*** VW_ has quit IRC23:44
*** VW_ has joined #openstack-nova23:45
*** alexpilotti_ has joined #openstack-nova23:45
*** hemna is now known as hemnafk23:46
*** jodah- has joined #openstack-nova23:46
*** alexpilotti has quit IRC23:47
*** alexpilotti_ is now known as alexpilotti23:47
*** VW_ has quit IRC23:47
*** VW_ has joined #openstack-nova23:48
*** tsekiyama has quit IRC23:51
superdanmelwitt: two things:23:53
superdanfirst, I don't see how the changes to the info_cache function in rpcapi handle the change you're making23:53
superdanyou used to send the fresh-from-the-db thing,23:54
superdanand now you're sending the stale instance pre-save, right?23:54
*** VW_ has quit IRC23:54
superdanmelwitt: and second, I wanna make sure you know what I'm getting at with the _from_db_object() thing23:54
*** VW_ has joined #openstack-nova23:55
melwittsuperdan: yes, okay. I thought you meant the change from sqlalchemy object to nova object23:55
superdanmelwitt: from_db_object(... {'uuid': 'foo'}   will fail because you don't have "availability_zone" and all the other things in your dict23:55
melwittsuperdan: so, I was aware this is different, but I did it to be consistent with what alaski suggested in the instance object to clone self early and send that instead of copying the old changed fields into the new instance retrieved from db23:56
*** rwsu has quit IRC23:56
*** VW_ has quit IRC23:57
melwittsuperdan: er, setting the changed field list in the new from db cloned object23:57
*** otter768 has joined #openstack-nova23:57
*** VW_ has joined #openstack-nova23:57
superdanhmm, I don't get it23:57
superdanhttps://review.openstack.org/#/c/176518/3/nova/objects/instance.py,cm23:59
superdanyou mean there?23:59
melwittsuperdan: in objects/instance.py stale_instance is a self clone that happens at the beginning, before calling db.instance_update_and_get_original and that's sent to cells23:59
superdanhe was talking about instance not info_cache there23:59
superdanright23:59
superdanokay, I guess maybe that's what we want to do23:59
superdansince we're replicating up23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!