| *** pixelbeat has quit IRC | 00:02 | |
| *** Longgeek has joined #openstack-nova | 00:03 | |
| *** otter768 has joined #openstack-nova | 00:03 | |
| *** annegentle has joined #openstack-nova | 00:06 | |
| *** Longgeek has quit IRC | 00:07 | |
| *** IanGovett has joined #openstack-nova | 00:08 | |
| *** rmoe has quit IRC | 00:08 | |
| *** otter768 has quit IRC | 00:08 | |
| *** Nikolay_St has quit IRC | 00:08 | |
| *** penick has joined #openstack-nova | 00:12 | |
| *** IanGovett1 has joined #openstack-nova | 00:14 | |
| *** boris-42 has joined #openstack-nova | 00:15 | |
| *** IanGovett has quit IRC | 00:16 | |
| *** IanGovett1 has quit IRC | 00:19 | |
| *** Longgeek has joined #openstack-nova | 00:19 | |
| *** takedakn has quit IRC | 00:20 | |
| *** penick has quit IRC | 00:20 | |
| *** melwitt has quit IRC | 00:21 | |
| *** mwagner_lap has joined #openstack-nova | 00:21 | |
| *** bkopilov has quit IRC | 00:24 | |
| *** Longgeek has quit IRC | 00:26 | |
| *** rmoe has joined #openstack-nova | 00:27 | |
| *** patrickeast has joined #openstack-nova | 00:27 | |
| *** patrickeast has quit IRC | 00:29 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Reopen socket if its determined to be bad (WIP) https://review.openstack.org/163582 | 00:30 |
|---|---|---|
| *** annegentle has quit IRC | 00:31 | |
| *** patrickeast has joined #openstack-nova | 00:32 | |
| *** FL1SK has quit IRC | 00:36 | |
| *** FL1SK has joined #openstack-nova | 00:37 | |
| *** FL1SK has quit IRC | 00:37 | |
| *** annegentle has joined #openstack-nova | 00:38 | |
| *** FL1SK has joined #openstack-nova | 00:38 | |
| *** park has joined #openstack-nova | 00:40 | |
| *** vishy has quit IRC | 00:41 | |
| *** vmtrooper has joined #openstack-nova | 00:41 | |
| *** vishy has joined #openstack-nova | 00:43 | |
| *** zzzeek has quit IRC | 00:44 | |
| *** vmtrooper has quit IRC | 00:46 | |
| *** mtanino has quit IRC | 00:47 | |
| *** annegentle has quit IRC | 00:48 | |
| openstackgerrit | Ian Wienand proposed openstack/python-novaclient: Add Client object to documentation https://review.openstack.org/163280 | 00:49 |
| *** annegentle has joined #openstack-nova | 00:49 | |
| *** rfolco_ has quit IRC | 00:59 | |
| *** marun has joined #openstack-nova | 01:00 | |
| openstackgerrit | Darren Sanders proposed openstack/nova: Don't fetch kernel/ramdisk files if you already have them. https://review.openstack.org/163661 | 01:03 |
| *** igordcard_ has quit IRC | 01:03 | |
| *** rwsu is now known as rwsu-afk | 01:04 | |
| *** marun has quit IRC | 01:05 | |
| *** hemna_ has joined #openstack-nova | 01:05 | |
| *** apuimedo has quit IRC | 01:08 | |
| *** annegentle has quit IRC | 01:08 | |
| *** promulo__ has joined #openstack-nova | 01:10 | |
| *** royce has joined #openstack-nova | 01:10 | |
| *** promulo has quit IRC | 01:14 | |
| *** gokrokve has quit IRC | 01:15 | |
| *** park has quit IRC | 01:17 | |
| *** dave-mccowan has joined #openstack-nova | 01:20 | |
| openstackgerrit | Anand Shanmugam proposed openstack/nova: Handles exception when unsupported virt-type given https://review.openstack.org/163152 | 01:20 |
| *** mtanino has joined #openstack-nova | 01:21 | |
| *** unicell has quit IRC | 01:25 | |
| *** markvoelker has quit IRC | 01:27 | |
| *** markvoelker has joined #openstack-nova | 01:27 | |
| *** takedakn has joined #openstack-nova | 01:29 | |
| *** markvoelker has quit IRC | 01:32 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Verify all quotas before updating the db https://review.openstack.org/163668 | 01:34 |
| *** asettle-mtg is now known as asettle | 01:37 | |
| *** sbfox has joined #openstack-nova | 01:38 | |
| *** Alexandra_ has joined #openstack-nova | 01:38 | |
| *** Alexandra_ has quit IRC | 01:38 | |
| *** sbfox has quit IRC | 01:39 | |
| *** sdake_ has quit IRC | 01:42 | |
| *** clds_ has quit IRC | 01:42 | |
| openstackgerrit | Thang Pham proposed openstack/nova: VMware: Add support for swap disk https://review.openstack.org/142174 | 01:44 |
| *** popw has joined #openstack-nova | 01:44 | |
| *** clds has joined #openstack-nova | 01:44 | |
| dims_ | jogo: rewrote all the tests, so filed a fresh review (also there was a -W from Sean)... https://review.openstack.org/163668 | 01:48 |
| * jogo hands dims_ a cookie | 01:49 | |
| dims_ | jogo: if you give a mouse a cookie :) http://en.wikipedia.org/wiki/If_You_Give_a_Mouse_a_Cookie | 01:50 |
| dims_ | not sure if you have seen that book :) | 01:50 |
| *** Longgeek has joined #openstack-nova | 01:52 | |
| jogo | haha | 01:52 |
| *** dims_ has quit IRC | 01:53 | |
| *** Longgeek has quit IRC | 01:56 | |
| *** angdraug has quit IRC | 01:56 | |
| *** salv-orlando has quit IRC | 01:58 | |
| *** yamahata has quit IRC | 01:59 | |
| *** armax has quit IRC | 02:01 | |
| *** royce has quit IRC | 02:04 | |
| *** otter768 has joined #openstack-nova | 02:04 | |
| *** armax has joined #openstack-nova | 02:08 | |
| *** otter768 has quit IRC | 02:08 | |
| *** ybathia has quit IRC | 02:09 | |
| *** nelsnelson has joined #openstack-nova | 02:10 | |
| *** Nikolay_St has joined #openstack-nova | 02:14 | |
| *** ijw has quit IRC | 02:14 | |
| *** patrickeast has quit IRC | 02:18 | |
| openstackgerrit | Eric Brown proposed openstack/nova: Use oslo_config choices support https://review.openstack.org/158573 | 02:19 |
| *** haomaiwang has joined #openstack-nova | 02:22 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove unused method _make_stub_method https://review.openstack.org/162839 | 02:27 |
| *** erkules_ has joined #openstack-nova | 02:28 | |
| *** park has joined #openstack-nova | 02:29 | |
| *** vmtrooper has joined #openstack-nova | 02:30 | |
| *** erkules has quit IRC | 02:31 | |
| *** vmtrooper has quit IRC | 02:35 | |
| openstackgerrit | xhzhf proposed openstack/nova: VMware driver: Increasing speed of downloading image https://review.openstack.org/135625 | 02:37 |
| *** kaisers has joined #openstack-nova | 02:40 | |
| openstackgerrit | Tony Breeds proposed openstack/nova: Websocket Proxy should verify Origin header https://review.openstack.org/163033 | 02:40 |
| *** kaisers2 has quit IRC | 02:42 | |
| *** achanda has quit IRC | 02:42 | |
| *** nelsnelson has quit IRC | 02:42 | |
| *** david-lyle is now known as david-lyle_afk | 02:45 | |
| *** sdake_ has joined #openstack-nova | 02:48 | |
| *** royce has joined #openstack-nova | 02:48 | |
| *** Marga_ has quit IRC | 02:51 | |
| *** kaufer has joined #openstack-nova | 02:52 | |
| *** oomichi has joined #openstack-nova | 02:52 | |
| *** rushiagr_away is now known as rushiagr | 02:52 | |
| *** unicell has joined #openstack-nova | 02:53 | |
| *** marun has joined #openstack-nova | 02:53 | |
| oomichi | alex_xu: hi | 02:54 |
| *** dims_ has joined #openstack-nova | 02:54 | |
| alex_xu | oomichi: hi, good morning | 02:54 |
| *** gokrokve has joined #openstack-nova | 02:54 | |
| *** rushiagr is now known as rushiagr_away | 02:54 | |
| oomichi | alex_xu: re: https://review.openstack.org/#/c/159760/4 | 02:54 |
| *** rushiagr_away is now known as rushiagr | 02:55 | |
| oomichi | alex_xu: to be honest, I have the same question as Joe, and I don't have enough idea about that. | 02:55 |
| oomichi | alex_xu: why we need a change of nova/tests/functional/test_api_samples.py ? | 02:56 |
| alex_xu | oomichi: The reason is we have admin checks in db layer, but the sample tests fake the db call | 02:56 |
| alex_xu | oomichi: for now we move the admin checks into the API layer, if without the admin context, that test will failed | 02:57 |
| *** bkopilov has joined #openstack-nova | 02:58 | |
| *** salv-orlando has joined #openstack-nova | 02:59 | |
| oomichi | alex_xu: thanks for your explanation, I will try the patch without a change of nova/tests/functional/test_api_samples.py. | 02:59 |
| alex_xu | oomichi: why we shouldn't change that? | 02:59 |
| *** dims_ has quit IRC | 02:59 | |
| oomichi | alex_xu: just want to clarify the reason. | 03:00 |
| openstackgerrit | Ian Wienand proposed openstack/nova: Add [pci] group and whitelist_sections to configure PCI passthrough https://review.openstack.org/133138 | 03:00 |
| alex_xu | oomichi: ok, got it, I miss understand your works. | 03:00 |
| alex_xu | s/works/words/.... | 03:01 |
| oomichi | alex_xu: now there is a lot of tests which contain ADMIN_API, so that is not so a big matter. | 03:01 |
| alex_xu | oomichi: ok, got it | 03:01 |
| *** zz_jgrimm is now known as jgrimm | 03:03 | |
| *** takedakn has quit IRC | 03:06 | |
| *** takedakn has joined #openstack-nova | 03:06 | |
| *** popw1 has joined #openstack-nova | 03:07 | |
| *** popw has quit IRC | 03:08 | |
| *** dboik_ has quit IRC | 03:10 | |
| *** salv-orlando has quit IRC | 03:10 | |
| alex_xu | oomichi: there is one patch just missing approve https://review.openstack.org/#/c/149527/ | 03:11 |
| *** rushiagr is now known as rushiagr_away | 03:12 | |
| oomichi | alex_xu: +A | 03:13 |
| alex_xu | oomichi: and I listed the policy patches which ready to merge or review at the etherpad https://etherpad.openstack.org/p/apipolicycheck appreciate you can take a look at them, at least hope we can merge all the v2.1 related patch, not sure we can merge all the db patch. | 03:13 |
| alex_xu | oomichi: thanks a lot | 03:13 |
| *** salv-orlando has joined #openstack-nova | 03:19 | |
| *** nelsnelson has joined #openstack-nova | 03:19 | |
| *** bkopilov has quit IRC | 03:19 | |
| *** salv-orlando has quit IRC | 03:23 | |
| *** mtanino has quit IRC | 03:24 | |
| *** salv-orlando has joined #openstack-nova | 03:24 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Move policy enforcement into REST API layer for v2.1 extended_volumes https://review.openstack.org/150258 | 03:26 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove useless policy rules for v2.1 api which removed/disabled https://review.openstack.org/163689 | 03:26 |
| alex_xu | oomichi: ^ I removed the useless rule for swap/attach/detach and os-pci in separated patch | 03:27 |
| *** mdbooth has quit IRC | 03:28 | |
| *** deepthi has joined #openstack-nova | 03:29 | |
| *** salv-orlando has quit IRC | 03:29 | |
| oomichi | alex_xu: thanks, I reviewed https://review.openstack.org/#/c/163689/ , can you check my comment ? | 03:32 |
| *** mdbooth has joined #openstack-nova | 03:34 | |
| *** rajesht has joined #openstack-nova | 03:35 | |
| *** ijw has joined #openstack-nova | 03:36 | |
| *** Longgeek has joined #openstack-nova | 03:37 | |
| *** sdake_ has quit IRC | 03:38 | |
| *** ijw has quit IRC | 03:41 | |
| *** rajesht has quit IRC | 03:44 | |
| *** rajesht has joined #openstack-nova | 03:44 | |
| *** claudiub has quit IRC | 03:46 | |
| openstackgerrit | Li Yingjun proposed openstack/nova: Improve 'attach interface' exception handling https://review.openstack.org/146788 | 03:51 |
| *** nelsnelson has quit IRC | 03:51 | |
| *** rushiagr_away is now known as rushiagr | 03:53 | |
| *** Marga_ has joined #openstack-nova | 03:53 | |
| *** dsanders has quit IRC | 03:56 | |
| *** Longgeek has quit IRC | 04:03 | |
| *** otter768 has joined #openstack-nova | 04:05 | |
| *** Nikolay_St has quit IRC | 04:05 | |
| *** penick has joined #openstack-nova | 04:06 | |
| openstackgerrit | Dave McCowan proposed openstack/nova: Websocket Proxy should verify Origin header https://review.openstack.org/163033 | 04:08 |
| *** sdake_ has joined #openstack-nova | 04:08 | |
| *** otter768 has quit IRC | 04:10 | |
| *** kaufer has quit IRC | 04:10 | |
| *** penick has quit IRC | 04:12 | |
| *** gangil1 has joined #openstack-nova | 04:13 | |
| *** oomichi has quit IRC | 04:13 | |
| *** Marga_ has quit IRC | 04:15 | |
| openstackgerrit | Rick Harris proposed openstack/nova: Fix infinite recursion caused by unnecessary stub https://review.openstack.org/163614 | 04:16 |
| *** Marga_ has joined #openstack-nova | 04:16 | |
| *** gokrokve_ has joined #openstack-nova | 04:18 | |
| *** vmtrooper has joined #openstack-nova | 04:19 | |
| *** gokrokve has quit IRC | 04:21 | |
| *** gokrokve_ has quit IRC | 04:22 | |
| *** vmtrooper has quit IRC | 04:24 | |
| *** devlaps has quit IRC | 04:25 | |
| *** dsanders has joined #openstack-nova | 04:30 | |
| *** ijw has joined #openstack-nova | 04:31 | |
| *** achanda has joined #openstack-nova | 04:31 | |
| *** ijw has quit IRC | 04:37 | |
| *** erlon has quit IRC | 04:41 | |
| *** unicell has quit IRC | 04:44 | |
| *** unicell has joined #openstack-nova | 04:45 | |
| *** marun has quit IRC | 04:47 | |
| *** salv-orlando has joined #openstack-nova | 04:49 | |
| *** marun has joined #openstack-nova | 04:50 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: use instance UUID as instance name https://review.openstack.org/160796 | 04:51 |
| openstackgerrit | garyk proposed openstack/nova: VMware: update test_vm_util to use instance object https://review.openstack.org/160273 | 04:51 |
| openstackgerrit | garyk proposed openstack/nova: VMware: fix VM rescue problem with VNC console https://review.openstack.org/113908 | 04:51 |
| *** nelsnelson has joined #openstack-nova | 04:53 | |
| *** sdake__ has joined #openstack-nova | 04:55 | |
| *** Nikolay_St has joined #openstack-nova | 04:55 | |
| *** bkopilov has joined #openstack-nova | 04:56 | |
| *** Alexandra_ has joined #openstack-nova | 04:57 | |
| *** gokrokve has joined #openstack-nova | 04:58 | |
| *** sdake_ has quit IRC | 04:58 | |
| *** salv-orlando has quit IRC | 05:00 | |
| *** yamahata has joined #openstack-nova | 05:04 | |
| *** gokrokve_ has joined #openstack-nova | 05:04 | |
| *** achanda has quit IRC | 05:04 | |
| *** david-ly_ has joined #openstack-nova | 05:04 | |
| *** cbader02 has joined #openstack-nova | 05:04 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add and use raise_http_not_implemented_error() https://review.openstack.org/163697 | 05:04 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add a hacking rule for consistent HTTP501 message https://review.openstack.org/163698 | 05:04 |
| *** leseb- has joined #openstack-nova | 05:04 | |
| *** omnibus7 has joined #openstack-nova | 05:05 | |
| *** annashen has joined #openstack-nova | 05:06 | |
| *** ijw has joined #openstack-nova | 05:06 | |
| *** devanand1 has joined #openstack-nova | 05:07 | |
| *** telemons1er has joined #openstack-nova | 05:07 | |
| *** mestery_ has joined #openstack-nova | 05:07 | |
| *** promulo has joined #openstack-nova | 05:07 | |
| *** shadyabh1 has joined #openstack-nova | 05:08 | |
| *** sdake__ has quit IRC | 05:08 | |
| *** mestery has quit IRC | 05:08 | |
| *** rmk` has joined #openstack-nova | 05:09 | |
| *** shaohe_robot1 has joined #openstack-nova | 05:09 | |
| *** jamespd has joined #openstack-nova | 05:09 | |
| *** jyuso has quit IRC | 05:09 | |
| *** leseb_ has quit IRC | 05:09 | |
| *** FL1SK has quit IRC | 05:09 | |
| *** david-lyle_afk has quit IRC | 05:09 | |
| *** jyuso has joined #openstack-nova | 05:09 | |
| *** beagles has quit IRC | 05:09 | |
| *** telemonster has quit IRC | 05:09 | |
| *** mikal_ has joined #openstack-nova | 05:09 | |
| *** shaohe_robot has quit IRC | 05:09 | |
| *** devananda has quit IRC | 05:09 | |
| *** jamespd_ has quit IRC | 05:09 | |
| *** gokrokve has quit IRC | 05:09 | |
| *** rmk has quit IRC | 05:09 | |
| *** davidlenwell has quit IRC | 05:09 | |
| *** mariusv has quit IRC | 05:09 | |
| *** gmann has quit IRC | 05:09 | |
| *** shadyabhi has quit IRC | 05:09 | |
| *** promulo__ has quit IRC | 05:09 | |
| *** cbader has quit IRC | 05:09 | |
| *** gangil1 has quit IRC | 05:09 | |
| *** frickler has quit IRC | 05:09 | |
| *** mikal has quit IRC | 05:09 | |
| *** frickler has joined #openstack-nova | 05:09 | |
| *** raginbajin has quit IRC | 05:09 | |
| *** rmk` is now known as rmk | 05:09 | |
| *** directxman12 has quit IRC | 05:09 | |
| *** rmk has quit IRC | 05:09 | |
| *** rmk has joined #openstack-nova | 05:09 | |
| *** mariusv has joined #openstack-nova | 05:09 | |
| *** mariusv has quit IRC | 05:09 | |
| *** mariusv has joined #openstack-nova | 05:09 | |
| *** gmann has joined #openstack-nova | 05:10 | |
| *** b3nt_pin has joined #openstack-nova | 05:10 | |
| *** b3nt_pin is now known as Guest27074 | 05:10 | |
| *** gangil has joined #openstack-nova | 05:11 | |
| *** ijw has quit IRC | 05:12 | |
| *** directxman12 has joined #openstack-nova | 05:12 | |
| *** raginbajin has joined #openstack-nova | 05:13 | |
| *** rushiagr is now known as rushiagr_away | 05:14 | |
| *** marun has quit IRC | 05:19 | |
| *** marun has joined #openstack-nova | 05:22 | |
| *** kiwik-chenrui has quit IRC | 05:22 | |
| *** annashen has quit IRC | 05:24 | |
| *** annashen has joined #openstack-nova | 05:24 | |
| *** markvoelker has joined #openstack-nova | 05:25 | |
| *** Nikolay_St has quit IRC | 05:26 | |
| *** pradipta has joined #openstack-nova | 05:26 | |
| *** achanda has joined #openstack-nova | 05:28 | |
| *** markvoelker has quit IRC | 05:30 | |
| *** lpetrut has joined #openstack-nova | 05:35 | |
| openstackgerrit | Zhenzan Zhou proposed openstack/nova: Replace BareMetalDriver with IronicDriver in option help string https://review.openstack.org/163700 | 05:37 |
| *** ankit_ag has joined #openstack-nova | 05:38 | |
| *** harlowja_ is now known as harlowja_away | 05:41 | |
| *** obondarev_ has joined #openstack-nova | 05:42 | |
| *** obondarev has quit IRC | 05:42 | |
| *** marun has quit IRC | 05:42 | |
| *** eliqiao has joined #openstack-nova | 05:42 | |
| *** annashen has quit IRC | 05:42 | |
| *** nelsnelson has quit IRC | 05:44 | |
| *** Longgeek has joined #openstack-nova | 05:45 | |
| *** wuhg has joined #openstack-nova | 05:56 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for service_get_all_by_* https://review.openstack.org/159762 | 05:56 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for service_get_by_host* https://review.openstack.org/150349 | 05:56 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for service_get_all https://review.openstack.org/160088 | 05:56 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for service_get_by_compute_host https://review.openstack.org/160089 | 05:56 |
| *** browne1 has quit IRC | 05:56 | |
| *** annashen has joined #openstack-nova | 05:57 | |
| *** rushiagr_away is now known as rushiagr | 05:58 | |
| *** adelinatuvenie has joined #openstack-nova | 06:00 | |
| *** abhishekkk has joined #openstack-nova | 06:01 | |
| *** sdake_ has joined #openstack-nova | 06:02 | |
| *** sdake__ has joined #openstack-nova | 06:04 | |
| *** annashen has quit IRC | 06:05 | |
| *** ijw has joined #openstack-nova | 06:06 | |
| *** otter768 has joined #openstack-nova | 06:06 | |
| *** sdake_ has quit IRC | 06:08 | |
| *** vmtrooper has joined #openstack-nova | 06:08 | |
| *** takedakn has quit IRC | 06:09 | |
| *** sdake_ has joined #openstack-nova | 06:10 | |
| *** otter768 has quit IRC | 06:11 | |
| *** ijw has quit IRC | 06:11 | |
| *** vmtrooper has quit IRC | 06:13 | |
| *** sdake__ has quit IRC | 06:13 | |
| *** sdake__ has joined #openstack-nova | 06:14 | |
| *** abhishekk has joined #openstack-nova | 06:17 | |
| *** abhishekk has left #openstack-nova | 06:17 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/162030 | 06:18 |
| *** Longgeek has quit IRC | 06:18 | |
| *** sdake_ has quit IRC | 06:18 | |
| *** gokrokve has joined #openstack-nova | 06:20 | |
| *** eliqiao has quit IRC | 06:21 | |
| *** gokrokve_ has quit IRC | 06:23 | |
| *** eliqiao has joined #openstack-nova | 06:23 | |
| *** rajesht has quit IRC | 06:26 | |
| *** markvoelker has joined #openstack-nova | 06:26 | |
| *** yamahata has quit IRC | 06:26 | |
| *** markvoelker has quit IRC | 06:32 | |
| *** dims_ has joined #openstack-nova | 06:32 | |
| *** sdake_ has joined #openstack-nova | 06:37 | |
| *** dims_ has quit IRC | 06:38 | |
| *** Alexandra_ has quit IRC | 06:38 | |
| *** FL1SK has joined #openstack-nova | 06:39 | |
| *** sabari has quit IRC | 06:40 | |
| *** sdake__ has quit IRC | 06:41 | |
| *** eliqiao has quit IRC | 06:43 | |
| openstackgerrit | Qin Zhao proposed openstack/nova: Refactor show_port() in neutron api https://review.openstack.org/135020 | 06:43 |
| *** irenab has joined #openstack-nova | 06:43 | |
| *** oro has joined #openstack-nova | 06:43 | |
| *** setuid has quit IRC | 06:43 | |
| *** oro_ has joined #openstack-nova | 06:43 | |
| *** eliqiao has joined #openstack-nova | 06:44 | |
| *** dsanders has quit IRC | 06:45 | |
| *** eliqiao has quit IRC | 06:45 | |
| *** eliqiao has joined #openstack-nova | 06:45 | |
| *** setuid has joined #openstack-nova | 06:46 | |
| *** aniket has quit IRC | 06:48 | |
| *** sabari has joined #openstack-nova | 06:49 | |
| *** dsanders has joined #openstack-nova | 06:54 | |
| *** shaohe_robot1 is now known as shaohe_feng | 06:55 | |
| *** sdake__ has joined #openstack-nova | 06:58 | |
| *** sdake_ has quit IRC | 07:02 | |
| *** gangil1 has joined #openstack-nova | 07:04 | |
| *** sabari has quit IRC | 07:04 | |
| *** gangil has quit IRC | 07:05 | |
| *** armax has quit IRC | 07:06 | |
| *** ijw has joined #openstack-nova | 07:06 | |
| *** Marga_ has quit IRC | 07:07 | |
| *** exploreshaifali has joined #openstack-nova | 07:08 | |
| *** sdake__ has quit IRC | 07:08 | |
| *** pkoniszewski has joined #openstack-nova | 07:09 | |
| *** achanda has quit IRC | 07:11 | |
| *** ijw has quit IRC | 07:11 | |
| *** sabari has joined #openstack-nova | 07:13 | |
| *** dsanders has quit IRC | 07:17 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove useless policy rules for v2.1 api which removed/disabled https://review.openstack.org/163689 | 07:17 |
| openstackgerrit | Alex Xu proposed openstack/nova: Move policy enforcement into REST API layer for v2.1 extended_volumes https://review.openstack.org/150258 | 07:17 |
| *** kiwik-chenrui has joined #openstack-nova | 07:19 | |
| *** garyk has joined #openstack-nova | 07:19 | |
| *** achanda has joined #openstack-nova | 07:24 | |
| *** gszasz has joined #openstack-nova | 07:26 | |
| *** jcoufal has joined #openstack-nova | 07:27 | |
| *** salv-orlando has joined #openstack-nova | 07:27 | |
| *** markvoelker has joined #openstack-nova | 07:28 | |
| *** Maike has joined #openstack-nova | 07:28 | |
| *** sahid has joined #openstack-nova | 07:28 | |
| *** markvoelker has quit IRC | 07:33 | |
| *** sabari has quit IRC | 07:33 | |
| *** lpetrut has quit IRC | 07:35 | |
| *** salv-orlando has quit IRC | 07:35 | |
| *** TobiasE has joined #openstack-nova | 07:36 | |
| openstackgerrit | garyk proposed openstack/nova: Ensure that API updates notify compute node https://review.openstack.org/163434 | 07:36 |
| openstackgerrit | Ankit Agrawal proposed openstack/nova: Detach volume after deleting the instance https://review.openstack.org/145738 | 07:38 |
| *** cbader02 has quit IRC | 07:38 | |
| *** david8hu has quit IRC | 07:39 | |
| *** salv-orlando has joined #openstack-nova | 07:39 | |
| *** david8hu has joined #openstack-nova | 07:39 | |
| *** cbader has joined #openstack-nova | 07:39 | |
| chaochin | garyk: Hi Gary, are you there? | 07:41 |
| *** garyk has quit IRC | 07:41 | |
| *** garyk has joined #openstack-nova | 07:41 | |
| *** salv-orl_ has joined #openstack-nova | 07:42 | |
| chaochin | garyk: hello? | 07:42 |
| *** sluo_wfh has joined #openstack-nova | 07:42 | |
| *** sdake_ has joined #openstack-nova | 07:42 | |
| *** salv-orlando has quit IRC | 07:43 | |
| *** achanda has quit IRC | 07:44 | |
| garyk | chaochin: hi | 07:44 |
| *** salv-orl_ has quit IRC | 07:46 | |
| chaochin | garyk: one vmware question. Do you know how to modify hotPlugMemoryLimit? I boot one vm using vmdk file, and notice that its hotPlugMemoryLimit seems to be equal to its current memory size. | 07:46 |
| chaochin | garyk: I can not increase memory size, even if I enables memory hot add | 07:47 |
| garyk | chaochin: are you increasingt he memory size by flavor? | 07:48 |
| *** jamespd has quit IRC | 07:48 | |
| chaochin | garyk: I am attempting to increase memory via vcenter gui | 07:48 |
| chaochin | garyk: the vm is boot by nova | 07:48 |
| *** sdake_ has quit IRC | 07:49 | |
| *** romainh has joined #openstack-nova | 07:49 | |
| chaochin | garyk: I modified vm_utils.py to enable cpu/memory hot add capability in get_create_vm_spec() | 07:50 |
| chaochin | garyk: and then, boot a vm. In vcenter gui, cpu/memory hot add is enabled. I can increase cpu, but can not increase memory size. | 07:51 |
| *** sabari has joined #openstack-nova | 07:51 | |
| chaochin | garyk: on gui, I see the memory hot add limit is same as vm's current memory size (2GB) | 07:52 |
| garyk | chaochin: sec, i am looking | 07:53 |
| chaochin | garyk: I hope to modify it, but I did not find any api to modify this attribute in sdk doc | 07:53 |
| chaochin | garyk: thank you! | 07:53 |
| *** sluo_wfh has quit IRC | 07:53 | |
| *** jlanoux has joined #openstack-nova | 07:54 | |
| garyk | chaochin: please see http://pubs.vmware.com/vsphere-55/index.jsp#com.vmware.wssdk.apiref.doc/vim.vm.ConfigSpec.html#field_detail. I imagine that you are setting memoryHotAddEnabled? | 07:54 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in xenapi/test_vm_utils to pass wiith random PYTHOHASHSEED https://review.openstack.org/163363 | 07:56 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_createserverext to pass with random PYTHONHASHSEED https://review.openstack.org/163362 | 07:56 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Refactored tests in test_neutron_driver to pass with random PYTHONHASHSEED https://review.openstack.org/163361 | 07:56 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed archivation of deleted records https://review.openstack.org/163360 | 07:56 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Fixed tests in test_glance to pass with random PYTHONHASHSEED https://review.openstack.org/163364 | 07:56 |
| *** sdake has joined #openstack-nova | 07:56 | |
| garyk | chaochin: other than that i am not familiar with how the feature works :) | 07:56 |
| *** vmtrooper has joined #openstack-nova | 07:57 | |
| chaochin | garyk: yes. I add 3 lines in get_create_vm_spec(). config_spec.cpuHotAddEnabled = True config_spec.cpuHotRemoveEnabled = True config_spec.memoryHotAddEnabled = True | 07:58 |
| chaochin | garyk: ok, thank you. Let me look for other documents | 07:59 |
| *** jamespd has joined #openstack-nova | 07:59 | |
| *** cbader has quit IRC | 07:59 | |
| *** david8hu has quit IRC | 08:00 | |
| *** cbader has joined #openstack-nova | 08:00 | |
| *** david8hu has joined #openstack-nova | 08:00 | |
| garyk | chaochin: ok. i am not sure how you plan to use this. | 08:01 |
| *** salv-orlando has joined #openstack-nova | 08:01 | |
| *** vmtrooper has quit IRC | 08:02 | |
| *** dsanders has joined #openstack-nova | 08:02 | |
| *** Longgeek has joined #openstack-nova | 08:03 | |
| *** jistr has joined #openstack-nova | 08:03 | |
| *** markus_z has joined #openstack-nova | 08:04 | |
| *** salv-orlando has quit IRC | 08:06 | |
| chaochin | garyk: we have customers who hope to resize vm memory without power it off, so I am looking for way to help them. | 08:06 |
| *** ijw has joined #openstack-nova | 08:06 | |
| *** afazekas_ has joined #openstack-nova | 08:06 | |
| *** lpetrut has joined #openstack-nova | 08:06 | |
| *** otter768 has joined #openstack-nova | 08:07 | |
| chaochin | garyk: because the customer can use vcenter gui to modify memory size without powering off the vm, so they hope openstack can also do that. | 08:07 |
| *** pask81 has joined #openstack-nova | 08:08 | |
| garyk | chaochin: when i set the value as enabled adn tried to update the memory i was not able too. not sure if it is soemthing in my setup. | 08:09 |
| garyk | chaochin: i am not sure how this will fit into the openstack model - you will need to do a resize without a power off. that would require detailed design | 08:10 |
| *** apuimedo has joined #openstack-nova | 08:10 | |
| *** rushiagr is now known as rushiagr_away | 08:11 | |
| chaochin | garyk: yes, it require a design if we hope to implement it in nova. Now I am investigating if it is doable. | 08:11 |
| *** otter768 has quit IRC | 08:11 | |
| *** matrohon has joined #openstack-nova | 08:11 | |
| garyk | chaochin: ok, sounds good. | 08:12 |
| *** ijw has quit IRC | 08:12 | |
| *** Longgeek has quit IRC | 08:12 | |
| chaochin | garyk: cpu and disk can be increased by ReconfigureVM_Task without powering off. But I just notice that the memory of vm boot by Nova can not be increased. If I manually create vm in vcenter gui, its memory can be increased. | 08:14 |
| *** gokrokve has quit IRC | 08:14 | |
| chaochin | garyk: not sure what is the difference... | 08:14 |
| *** gokrokve has joined #openstack-nova | 08:14 | |
| *** pask81 has quit IRC | 08:15 | |
| *** pradipta has quit IRC | 08:16 | |
| *** adelinatuvenie has quit IRC | 08:16 | |
| *** sluo_wfh has joined #openstack-nova | 08:18 | |
| *** takedakn has joined #openstack-nova | 08:19 | |
| *** takedakn has quit IRC | 08:19 | |
| *** gokrokve has quit IRC | 08:19 | |
| garyk | chaochin: i do not understand | 08:19 |
| garyk | chaochin: it may be related to one of the properties of the vm. | 08:20 |
| garyk | maybe the guest needs to support this | 08:20 |
| *** eglynn has quit IRC | 08:20 | |
| garyk | do you have vmware tools in your guest? | 08:20 |
| *** pask81 has joined #openstack-nova | 08:21 | |
| *** Nikolay_St has joined #openstack-nova | 08:21 | |
| *** garyk has quit IRC | 08:25 | |
| *** garyk has joined #openstack-nova | 08:25 | |
| *** mrda is now known as mrda-away | 08:27 | |
| *** markvoelker has joined #openstack-nova | 08:29 | |
| *** dsanders has quit IRC | 08:30 | |
| *** lpetrut has quit IRC | 08:30 | |
| *** markvoelker has quit IRC | 08:34 | |
| *** salv-orlando has joined #openstack-nova | 08:34 | |
| *** tdurakov has quit IRC | 08:35 | |
| lan | Can anyone help take a look at this issue : https://bugs.launchpad.net/nova/+bug/1431201 ? kilo controller conduct juno compute failed. | 08:36 |
| openstack | Launchpad bug 1431201 in OpenStack Compute (nova) "kilo controller cann't conduct juno compute nodes" [Undecided,New] | 08:36 |
| openstackgerrit | Keiichi KII proposed openstack/nova: libvirt: optimize multipath call to identify IQN https://review.openstack.org/163733 | 08:37 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add and use raise_http_not_implemented_error() https://review.openstack.org/163697 | 08:37 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add a hacking rule for consistent HTTP501 message https://review.openstack.org/163698 | 08:37 |
| *** mpavone has joined #openstack-nova | 08:39 | |
| *** erkules_ is now known as erkules | 08:40 | |
| *** erkules has quit IRC | 08:40 | |
| *** erkules has joined #openstack-nova | 08:40 | |
| *** pixelbeat has joined #openstack-nova | 08:41 | |
| *** tdurakov has joined #openstack-nova | 08:41 | |
| *** mpaolino has joined #openstack-nova | 08:44 | |
| *** gokrokve has joined #openstack-nova | 08:45 | |
| *** gokrokve has quit IRC | 08:50 | |
| *** sdake__ has joined #openstack-nova | 08:54 | |
| *** moshele has joined #openstack-nova | 08:55 | |
| *** sdake has quit IRC | 08:57 | |
| *** nellysmitt has joined #openstack-nova | 09:02 | |
| *** lpetrut has joined #openstack-nova | 09:03 | |
| *** ijw has joined #openstack-nova | 09:06 | |
| *** annashen has joined #openstack-nova | 09:06 | |
| *** park has quit IRC | 09:08 | |
| *** annashen has quit IRC | 09:11 | |
| *** ijw has quit IRC | 09:11 | |
| *** gokrokve has joined #openstack-nova | 09:14 | |
| *** eglynn has joined #openstack-nova | 09:15 | |
| *** gokrokve has quit IRC | 09:15 | |
| *** gokrokve has joined #openstack-nova | 09:16 | |
| *** gokrokve has quit IRC | 09:20 | |
| *** derekh has joined #openstack-nova | 09:23 | |
| *** lpetrut has quit IRC | 09:23 | |
| *** lucasagomes has joined #openstack-nova | 09:26 | |
| *** markvoelker has joined #openstack-nova | 09:30 | |
| *** subscope has joined #openstack-nova | 09:30 | |
| *** markvoelker has quit IRC | 09:35 | |
| *** cbader has quit IRC | 09:37 | |
| *** asselin has quit IRC | 09:37 | |
| *** david8hu has quit IRC | 09:37 | |
| bauzas | lan: I'm in | 09:37 |
| bauzas | lan: thanks for raising the bug | 09:37 |
| bauzas | lan: I'll take it | 09:37 |
| *** asselin has joined #openstack-nova | 09:38 | |
| *** cbader has joined #openstack-nova | 09:38 | |
| *** david8hu has joined #openstack-nova | 09:38 | |
| *** Guest27074 is now known as b3nt_pin | 09:41 | |
| lan | bauzas, thanks :) | 09:43 |
| bauzas | lan: would you mind cherry-pick the commit I'll do and test it on your env ? | 09:44 |
| *** royce has quit IRC | 09:44 | |
| bauzas | lan: there are 2 gate tests for upgrades, called grenade and grenade-partial-ncpu | 09:44 |
| bauzas | lan: grenade-partial-ncpu is actually doing what you want to do - ie. have old computes while controller is master | 09:45 |
| lan | bauzas: sure | 09:45 |
| bauzas | lan: what's weird is that grenade-partial-ncpu didn't notice the problem | 09:45 |
| bauzas | lan: so that's probably a coverage problem | 09:45 |
| bauzas | lan: I also replied to your ticket by asking you more verbosity on the 2nd problem you mention | 09:45 |
| *** vmtrooper has joined #openstack-nova | 09:45 | |
| bauzas | lan: would you mind putting more info on that ? | 09:46 |
| lan | bauzas: sure , I will do all I can do. | 09:46 |
| bauzas | lan: thanks | 09:47 |
| bauzas | lan: I tagged your bug as critical, because it breaks the rolling upgrades | 09:47 |
| *** vmtrooper has quit IRC | 09:51 | |
| *** dims__ has joined #openstack-nova | 09:54 | |
| *** Longgeek has joined #openstack-nova | 09:54 | |
| lan | bauzas: I replied to your question in launchpad. It's really a critical problem broken my upgrade test. | 09:57 |
| bauzas | lan: awesome comment, ty | 09:58 |
| openstackgerrit | garyk proposed openstack/nova: libvirt: add in missing translation for exception https://review.openstack.org/163744 | 10:00 |
| openstackgerrit | YangLei proposed openstack/nova: Use the right instance object https://review.openstack.org/163745 | 10:04 |
| *** eliqiao has quit IRC | 10:05 | |
| *** ijw has joined #openstack-nova | 10:06 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 10:07 | |
| *** otter768 has joined #openstack-nova | 10:08 | |
| *** ijw has quit IRC | 10:11 | |
| *** lpetrut has joined #openstack-nova | 10:12 | |
| *** otter768 has quit IRC | 10:13 | |
| *** Longgeek has quit IRC | 10:14 | |
| *** gokrokve has joined #openstack-nova | 10:16 | |
| *** gangil1 has quit IRC | 10:17 | |
| *** jlanoux has quit IRC | 10:18 | |
| *** salv-orl_ has joined #openstack-nova | 10:18 | |
| *** salv-orlando has quit IRC | 10:20 | |
| *** apevec has joined #openstack-nova | 10:21 | |
| *** gokrokve has quit IRC | 10:21 | |
| apevec | nova-core: looks like master OSSA fix for websocket proxy https://review.openstack.org/163033 is good to go now, please review/merge | 10:22 |
| apevec | I'd like to merge Icehouse backport before 2014.1.4 release today | 10:22 |
| *** salv-orl_ has quit IRC | 10:22 | |
| *** andreykurilin_ has joined #openstack-nova | 10:26 | |
| kashyap | apevec, See mdbooth 's last comment - "but TL;DR I don't think this is a bug at all" | 10:27 |
| mdbooth | kashyap: I'm just rowing back on that :) | 10:28 |
| mdbooth | apevec: ^^^ | 10:28 |
| kashyap | Oh, okay | 10:28 |
| mdbooth | Reason is that NoVNC puts the token in a cookie and then subsequently doesn't require it | 10:28 |
| apevec | mdbooth, really, which novnc version? | 10:29 |
| *** jprovazn has joined #openstack-nova | 10:30 | |
| *** park has joined #openstack-nova | 10:30 | |
| *** markvoelker has joined #openstack-nova | 10:31 | |
| mdbooth | apevec: I don't know, but the websocketproxy code expects it to do that and handles it explicitly | 10:33 |
| mdbooth | if not token: | 10:33 |
| mdbooth | # NoVNC uses it's own convention that forward token | 10:33 |
| mdbooth | # from the request to a cookie header, we should check | 10:33 |
| mdbooth | # also for this behavior | 10:33 |
| mdbooth | hcookie = self.headers.getheader('cookie') | 10:33 |
| *** BobBall_AWOL is now known as BobBall | 10:33 | |
| apevec | mdbooth, I guess we need sross (he's novnc upstream maintainer) to answer that | 10:35 |
| mdbooth | apevec: It's there in the code. I'd assume it's true. I can also look now if you like. | 10:36 |
| *** markvoelker has quit IRC | 10:36 | |
| *** jistr has quit IRC | 10:36 | |
| mdbooth | apevec: However, I don't think validating Origin against Host makes any sense. | 10:36 |
| mdbooth | And I think we need to check the port number | 10:36 |
| apevec | tristanC, ^ so looks like that websocket proxy fix will need another round of reviews | 10:38 |
| mdbooth | apevec: Yep, noVNC definitely does this: | 10:39 |
| mdbooth | // If a token variable is passed in, set the parameter in a cookie. | 10:39 |
| mdbooth | // This is used by nova-novncproxy. | 10:39 |
| mdbooth | token = WebUtil.getQueryVar('token', null); | 10:39 |
| mdbooth | if (token) { | 10:39 |
| mdbooth | WebUtil.createCookie('token', token, 1) | 10:39 |
| mdbooth | } | 10:39 |
| mdbooth | I wonder why? | 10:40 |
| mdbooth | It's no more efficient | 10:40 |
| jprovazn | Hi, I have an issue with attaching a volume to an instance, "nova volume-attach" returns 404 albeit the volume is attached properly, from debug output it seems that the problem is that novaclient tries to get the volume through nova's cinder API wrapper: http://paste.openstack.org/show/191791/ (nova volume-show which talks directly to cinder works fine), any idea what's wrong/hot fix this? | 10:41 |
| *** pkholkin has quit IRC | 10:41 | |
| jprovazn | s/hot/how/ | 10:43 |
| *** ndipanov has joined #openstack-nova | 10:45 | |
| *** haomaiwang has quit IRC | 10:48 | |
| *** mpaolino has quit IRC | 10:48 | |
| gmann | sdague: ping | 10:49 |
| sdague | gmann: what's up? | 10:49 |
| *** subscope has quit IRC | 10:49 | |
| gmann | sdague: regarding sample tests merge - https://review.openstack.org/#/c/162016/ | 10:50 |
| gmann | sdague: is that direction looks fine? | 10:50 |
| sdague | ok, if I look later today and send an email? | 10:51 |
| gmann | sdague: yea, sure. | 10:51 |
| sdague | cool, will do | 10:52 |
| *** pkholkin has joined #openstack-nova | 10:52 | |
| gmann | sdague: just wondering about when we run tests twice in tox.ini, how to set something which can tell tests to run for v2 or v21 | 10:52 |
| *** jlanoux has joined #openstack-nova | 10:52 | |
| *** jistr has joined #openstack-nova | 10:53 | |
| gmann | sdague: tried many thing on tox.ini but did find good way. may be i can get some good pointer for that. | 10:53 |
| gmann | sdague: anyways ll check mail tomorrow. Thanks | 10:54 |
| sdague | yes, I was thinking about that as well. We might want to look at testscenarios for this instead | 10:54 |
| mdbooth | apevec: Also, read the whole bug. It seems they've been over this ground already. | 10:55 |
| sdague | which lets you define multiple scenarios for a test and will run it in each of them | 10:55 |
| gmann | sdague: ohhk | 10:56 |
| *** xgsa has quit IRC | 10:56 | |
| *** exploreshaifali has quit IRC | 10:56 | |
| *** _gryf has quit IRC | 10:56 | |
| *** erlon has joined #openstack-nova | 10:58 | |
| gmann | sdague: or can we set env variable in tox.ini apart from senenv section? | 10:59 |
| gmann | sdague: to differentiate two run | 11:00 |
| *** subscope has joined #openstack-nova | 11:00 | |
| sdague | gmann: we could, but it might make it easier on understanding the tests if we used testscenarios | 11:01 |
| *** garyk has quit IRC | 11:04 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Rename and move the v2.1 api policy into separated files https://review.openstack.org/150267 | 11:05 |
| *** garyk has joined #openstack-nova | 11:05 | |
| *** ijw has joined #openstack-nova | 11:06 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Rename and move the v2.1 api policy into separated files https://review.openstack.org/150267 | 11:07 |
| gmann | sdague: yaa. ll find more on testscenarios | 11:07 |
| *** ijw has quit IRC | 11:11 | |
| sdague | bauzas: did you post the logging add patch for scheduling failures? | 11:11 |
| sdague | or should I post something? | 11:12 |
| *** xgsa has joined #openstack-nova | 11:12 | |
| bauzas | sdague: not yet, working on a critical bug that I introduced https://bugs.launchpad.net/nova/+bug/1431201 | 11:13 |
| openstack | Launchpad bug 1431201 in OpenStack Compute (nova) "kilo controller cann't conduct juno compute nodes" [Critical,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 11:13 |
| sdague | ok | 11:14 |
| sdague | I'll poke | 11:14 |
| bauzas | sdague: that said, I can do this very quickly, hold on | 11:14 |
| *** gokrokve has joined #openstack-nova | 11:14 | |
| *** andreykurilin_ has quit IRC | 11:18 | |
| *** aix has joined #openstack-nova | 11:19 | |
| *** gokrokve has quit IRC | 11:19 | |
| sdague | bauzas: ok, I also filed a bug for tracking - https://bugs.launchpad.net/nova/+bug/1431291 | 11:21 |
| openstack | Launchpad bug 1431291 in OpenStack Compute (nova) "Scheduler Failures are no longer logged with enough detail for a site admin to do problem determination" [High,Confirmed] | 11:21 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Level up verbosity to WARN with a filter returns 0 hosts https://review.openstack.org/163771 | 11:22 |
| bauzas | sdague: ^ | 11:22 |
| sdague | bauzas: ok, except that doesn't actually provide enough info | 11:23 |
| openstackgerrit | Pranali Deore proposed openstack/nova: Fix typo in _live_migration_cleanup_flags method https://review.openstack.org/163774 | 11:23 |
| bauzas | sdague: the expected being ? | 11:23 |
| sdague | does that give us the context? | 11:24 |
| *** nellysmi_ has joined #openstack-nova | 11:24 | |
| sdague | I feel like we actually want the NoValidHost exception logged before it's raised | 11:24 |
| sdague | because it doesn't say why it failed | 11:25 |
| bauzas | sdague: mmm, I tend to disagree | 11:25 |
| bauzas | sdague: if operators have a timestamp where they know when it failed, they can match this | 11:25 |
| openstackgerrit | Alexey I. Froloff proposed openstack/nova: Pass user-requested ephemeral size to scheduler https://review.openstack.org/163393 | 11:25 |
| *** nellysmitt has quit IRC | 11:26 | |
| sdague | bauzas: I think you miss the point of the bug, which is we've now made them go through all kinds of hoops when there used to be a much more centralized logged error of why things failed | 11:27 |
| *** deepthi has quit IRC | 11:27 | |
| sdague | ops can always do more work to get answers | 11:27 |
| sdague | that seems like the wrong thing to make them do | 11:27 |
| bauzas | sdague: so what could the magic bullet for them ? | 11:28 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: partial fix for live-migration with config drive https://review.openstack.org/133919 | 11:28 |
| bauzas | *be | 11:28 |
| bauzas | do they need something poping up saying "eh this request failed" ? | 11:28 |
| sdague | I tried to put the info in the bug, please look there | 11:28 |
| bauzas | sdague: ack | 11:28 |
| bauzas | sdague: keep in mind I was an OpenStack operator before turning myself into a dev :) | 11:29 |
| dims__ | sdague: one good thing came out of the live-migration tempest testing! - https://review.openstack.org/133919 | 11:29 |
| bauzas | sdague: oh instance_id I see... | 11:30 |
| sdague | bauzas: and why | 11:30 |
| sdague | the why is really important | 11:30 |
| bauzas | sdague: which kind of why ? Is a filter saying "eh, zero hosts matched" is not that enough ? | 11:31 |
| *** IanGovett has joined #openstack-nova | 11:31 | |
| sdague | that is definitely not enough | 11:31 |
| *** dulek has joined #openstack-nova | 11:31 | |
| sdague | why is 'not enough ram' 'not enough disk' or something like that | 11:31 |
| sdague | the inner why | 11:31 |
| bauzas | sdague: I agree on the point that NoValidHost is a no-brainer because it's either due to a reschedule unsuccessful or a bad filter | 11:31 |
| bauzas | sdague: most of that info is at a debug level | 11:32 |
| sdague | bauzas: yes, that's the bug | 11:32 |
| bauzas | sdague: so I assume they want it to be shown | 11:32 |
| *** shyama has joined #openstack-nova | 11:32 | |
| bauzas | sdague: ack | 11:32 |
| sdague | sifting debug logs is not something that should be required here, we need to roll up the information | 11:32 |
| bauzas | sdague: about the instance_id, that's something we agreed (by we, I mean the nova devs) saying that scheduler doesn't have to know the list of ids | 11:32 |
| *** markvoelker has joined #openstack-nova | 11:33 | |
| sdague | well, that's apparently a problem for problem determination | 11:33 |
| bauzas | sdague: so now, scheduler only knows how many instances are asked, period | 11:33 |
| sdague | because user just sees 'vm in error state' | 11:33 |
| sdague | and they can't backtrack it | 11:33 |
| shyama | Hi I want to understand the utils.synchronized decorator used in nova. I understand it is a file based lock. where can i find more documentation on that? | 11:33 |
| bauzas | sdague: I think that's something we need to pop up in a Nova meeting | 11:34 |
| sdague | that's fine, you want to put that info in the bug as well? | 11:34 |
| bauzas | sdague: sure | 11:34 |
| *** vmtrooper has joined #openstack-nova | 11:34 | |
| sdague | dims__: is the libvirt bug registered | 11:34 |
| sdague | ? | 11:34 |
| bauzas | sdague: that said, we have the request spec which include the first instance info | 11:34 |
| *** rajesht has joined #openstack-nova | 11:37 | |
| openstackgerrit | Michal Rostecki proposed openstack/nova: Adding security groups to existing ports https://review.openstack.org/163780 | 11:37 |
| *** markvoelker has quit IRC | 11:37 | |
| sdague | dims__: I see that nova has a bug for this, but does libvirt? If we keep calling it a libvirt bug, I'd like to see the bugzilla for that | 11:39 |
| *** nihilifer has joined #openstack-nova | 11:39 | |
| *** vmtrooper has quit IRC | 11:39 | |
| * kashyap reads the scroll about 'libvirt' to get enough context. . . | 11:40 | |
| *** exploreshaifali has joined #openstack-nova | 11:41 | |
| kashyap | sdague, You're referring to this - https://bugs.launchpad.net/nova/+bug/1246201 | 11:42 |
| openstack | Launchpad bug 1246201 in OpenStack Compute (nova) "Live migration fails when the instance has a config-drive" [High,In progress] - Assigned to Davanum Srinivas (DIMS) (dims-v) | 11:42 |
| kashyap | ? | 11:42 |
| dims__ | sdague: vfat works, cdrom doesn't. cdrom fix would be in libvirt | 11:42 |
| dims__ | or qemu | 11:43 |
| dims__ | kashyap: magic word to invoke you is libvirt? :) | 11:43 |
| kashyap | If so - I see a corresponding live migration bugs https://bugs.launchpad.net/nova/+bug/1414065 and https://bugs.launchpad.net/nova/+bug/1429220 | 11:43 |
| openstack | Launchpad bug 1414065 in OpenStack Compute (nova) "Nova can loose track of running VM if live migration raises an exception" [High,Fix committed] - Assigned to Daniel Berrange (berrange) | 11:43 |
| openstack | Launchpad bug 1429220 in OpenStack Compute (nova) "libvirt does ensure live migration will eventually complete (or abort)" [High,In progress] - Assigned to Daniel Berrange (berrange) | 11:43 |
| dims__ | sdague: tony was chasing the libvirt issue last i heard | 11:44 |
| johnthetubaguy | garyk: answering you ML email, can I ask I quick question about your launchpad bug question? | 11:44 |
| kashyap | dims__, :-) I try to stay on top of upstreams (as humanly as possible in my interest areas) of libvirt/QEMU, so I just try to pay attention when those projects are mentioned. | 11:44 |
| dims__ | kashyap: thanks!! much appreciated | 11:45 |
| garyk | johnthetubaguy: sure. the mail was just pretty cryptic in my opinon | 11:47 |
| johnthetubaguy | garyk: its basically, no change, same as last time, its happening soon | 11:48 |
| garyk | johnthetubaguy: ok. | 11:48 |
| *** dulek has quit IRC | 11:48 | |
| johnthetubaguy | I just sent a reply that hopefully answers your questions | 11:48 |
| sdague | dims__: sure, I just want to make sure that there is actually a libvirt native bug on this, because we keep calling it a libvirt bug, but do we know if it's something the libvirt community believes is a bug and might fix some day? | 11:49 |
| johnthetubaguy | but I wasn't sure about what you meant by "Is launchpad not a sufficient source for highlighting bugs"? | 11:49 |
| *** markus_z has quit IRC | 11:49 | |
| johnthetubaguy | garyk: do you mean the rc bug tag? | 11:49 |
| *** mpavone has quit IRC | 11:49 | |
| sdague | johnthetubaguy: hey, process question which came up last week. As we hit these various freezes, it would be good to email out current review criteria, I think people are confused about that | 11:50 |
| *** markus_z has joined #openstack-nova | 11:50 | |
| johnthetubaguy | sdague: I have been trying to just link to the official docs on the release page, maybe worth calling out those links explicity? | 11:51 |
| *** oro has quit IRC | 11:51 | |
| *** oro_ has quit IRC | 11:51 | |
| dims__ | sdague: kashyap: last thread was this http://thread.gmane.org/gmane.comp.emulators.libvirt/109171/focus=109246 - i don't know if it ended up in a bug filed against libvirt | 11:51 |
| garyk | johnthetubaguy: my understanding is that the etherpad that you posted has the project priorities and the 'quick hit reviews'. in addition to this all bug should be done via launchpad. that is what i understood from sdague when in the past i tried to add reviews that address serious bugs to the etherpad mentioned | 11:52 |
| johnthetubaguy | garyk: I see the etherpad as purely additive, it basically points to things already tracked in launchpad, the bug list is for trivial stuff, as defined in the etherpad, but I could be getting the wrong end of the stick | 11:54 |
| kashyap | dims__, See the response from DanB and Michal Privoznik. In short - the change proposed there wants to control API behavior in guest XML | 11:54 |
| kashyap | Which is not the right way. And, way seems to be to use an existing migrate API (virDomainMigrate3) | 11:55 |
| dims__ | kashyap: right, but for tracking purposes...there should be a bug or feature logged somewhere? | 11:55 |
| kashyap | dims__, I'll ask on the list if there's a bug already created about it, if not, will ask the original poster to create one. | 11:56 |
| dims__ | kashyap: thanks a lot | 11:56 |
| *** rfolco has joined #openstack-nova | 11:56 | |
| *** aix has quit IRC | 11:59 | |
| *** Longgeek has joined #openstack-nova | 11:59 | |
| *** salv-orlando has joined #openstack-nova | 12:00 | |
| *** mpavone has joined #openstack-nova | 12:01 | |
| *** adalbas has joined #openstack-nova | 12:02 | |
| *** _gryf has joined #openstack-nova | 12:04 | |
| *** mwagner_lap has quit IRC | 12:04 | |
| kashyap | Done. (On the same thread.) | 12:05 |
| *** ijw has joined #openstack-nova | 12:06 | |
| ndipanov | sahid, there? | 12:08 |
| *** salv-orlando has quit IRC | 12:08 | |
| *** baoli has joined #openstack-nova | 12:08 | |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova-specs: Consolidate the APIs for getting consoles https://review.openstack.org/163790 | 12:08 |
| *** salv-orlando has joined #openstack-nova | 12:08 | |
| *** otter768 has joined #openstack-nova | 12:09 | |
| *** ijw has quit IRC | 12:12 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: libvirt: partial fix for live-migration with config drive https://review.openstack.org/133919 | 12:13 |
| *** otter768 has quit IRC | 12:14 | |
| *** gokrokve has joined #openstack-nova | 12:14 | |
| *** markvoelker has joined #openstack-nova | 12:15 | |
| *** gokrokve has quit IRC | 12:16 | |
| *** gokrokve has joined #openstack-nova | 12:16 | |
| *** marun has joined #openstack-nova | 12:17 | |
| dims__ | dhellmann: flaper87: bnemec: i logged a bunch of reviews for removing namespaced imports in oslo libraries. please take a quick look when you have some time - https://review.openstack.org/#/q/owner:%22Davanum+Srinivas+(dims)%22+status:open,n,z | 12:18 |
| *** Longgeek has quit IRC | 12:19 | |
| *** salv-orlando has quit IRC | 12:20 | |
| *** gokrokve has quit IRC | 12:21 | |
| *** ferest has joined #openstack-nova | 12:21 | |
| *** sdake has joined #openstack-nova | 12:23 | |
| *** sdake__ has quit IRC | 12:23 | |
| *** ferest has quit IRC | 12:25 | |
| sahid | ndipanov: yes | 12:27 |
| sahid | reading at your last comment in https://review.openstack.org/#/c/137817/ | 12:28 |
| *** rushiagr_away is now known as rushiagr | 12:28 | |
| sahid | i'm kind of desapointed since this is something with have talk with jaypipes on hangout | 12:29 |
| sahid | ndipanov: do you have an other solution to propose than ratio? | 12:29 |
| sahid | if you don't like the term "limit" perhaps we can use an other name, but i think ratio make sense in our case since we need to handle a no defined number of cell and each cell provide a non uniform number of cpu/memory | 12:31 |
| *** jecarey has quit IRC | 12:34 | |
| *** neelashah has joined #openstack-nova | 12:34 | |
| *** eliqiao has joined #openstack-nova | 12:34 | |
| ndipanov | sahid, so yes we did talk about it | 12:35 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 12:35 | |
| *** dims__ has quit IRC | 12:35 | |
| ndipanov | but I missd that subtlety | 12:35 |
| ndipanov | limits are limts | 12:35 |
| ndipanov | so sending ratios is a no-go from my pov | 12:36 |
| *** dims_ has joined #openstack-nova | 12:36 | |
| ndipanov | sahid, if you want to do something good for that bit of code | 12:36 |
| ndipanov | I suggest moving limits to an object first | 12:36 |
| ndipanov | then we can try out what makes sense | 12:36 |
| *** dprince has joined #openstack-nova | 12:37 | |
| *** dulek has joined #openstack-nova | 12:37 | |
| jaypipes | ndipanov, sahid: good morooning guys | 12:38 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix Juno nodes checking service.compute_node https://review.openstack.org/163797 | 12:38 |
| bauzas | lan: still there ? | 12:38 |
| bauzas | lan: https://review.openstack.org/163797 is a proposal for fixing the first problem | 12:38 |
| sahid | good morning jaypipes | 12:38 |
| bauzas | jaypipes: morning | 12:38 |
| ndipanov | jaypipes, morning | 12:38 |
| bauzas | jaypipes: we have a critical bug re: detach-service | 12:38 |
| bauzas | jaypipes: https://bugs.launchpad.net/nova/+bug/1431201 | 12:39 |
| openstack | Launchpad bug 1431201 in OpenStack Compute (nova) "kilo controller cann't conduct juno compute nodes" [Critical,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 12:39 |
| jaypipes | ndipanov, sahid: so, on that limits topology patch... | 12:39 |
| jaypipes | bauzas: will look at that in a sec. | 12:39 |
| bauzas | jaypipes: yey | 12:39 |
| *** Longgeek has joined #openstack-nova | 12:39 | |
| *** oro_ has joined #openstack-nova | 12:40 | |
| *** Marga_ has joined #openstack-nova | 12:40 | |
| *** oro has joined #openstack-nova | 12:40 | |
| *** marun has quit IRC | 12:41 | |
| *** aysyd has joined #openstack-nova | 12:43 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Allow force-delete irrespective of VM task_state https://review.openstack.org/163798 | 12:44 |
| ndipanov | jaypipes, yeeeeeeeessss :) | 12:45 |
| sahid | ? | 12:45 |
| jaypipes | ndipanov, sahid: answering in the code review, sorry... | 12:46 |
| *** park has quit IRC | 12:47 | |
| *** kaufer has joined #openstack-nova | 12:47 | |
| ndipanov | jaypipes, cool - I need to step out for a bit too... | 12:47 |
| jaypipes | ndipanov: sounds good. should have this reviewed in <30 minutes. | 12:48 |
| *** vipul has quit IRC | 12:48 | |
| jaypipes | apologies for the long delay, guys :( | 12:48 |
| *** sdake has quit IRC | 12:48 | |
| *** vipul has joined #openstack-nova | 12:48 | |
| sahid | np jaypipes, thanks to review it | 12:48 |
| *** sdake has joined #openstack-nova | 12:49 | |
| *** devanand1 has quit IRC | 12:49 | |
| *** devananda has joined #openstack-nova | 12:50 | |
| *** mwagner_lap has joined #openstack-nova | 12:52 | |
| *** rajesht has quit IRC | 12:53 | |
| jaypipes | sahid: OK, done with the review. | 12:56 |
| *** oro_ has quit IRC | 12:57 | |
| *** oro has quit IRC | 12:58 | |
| *** aix has joined #openstack-nova | 12:58 | |
| sahid | jaypipes: ack | 12:59 |
| *** TobiasE1 has joined #openstack-nova | 12:59 | |
| *** Maike_ has joined #openstack-nova | 13:00 | |
| *** bkopilov has quit IRC | 13:01 | |
| *** TobiasE has quit IRC | 13:01 | |
| edleafe | jaypipes: answered your questions in https://review.openstack.org/#/c/160511 | 13:02 |
| *** ptm_away is now known as PaulMurray | 13:02 | |
| *** dboik has joined #openstack-nova | 13:03 | |
| *** TobiasE1 has quit IRC | 13:03 | |
| *** Maike has quit IRC | 13:03 | |
| *** setuid has quit IRC | 13:05 | |
| *** setuid has joined #openstack-nova | 13:05 | |
| *** dboik_ has joined #openstack-nova | 13:05 | |
| *** ijw has joined #openstack-nova | 13:06 | |
| *** raildo has joined #openstack-nova | 13:06 | |
| *** tdurakov has joined #openstack-nova | 13:06 | |
| *** gszasz_ has joined #openstack-nova | 13:07 | |
| *** mestery_ has quit IRC | 13:08 | |
| *** gszasz has quit IRC | 13:08 | |
| bauzas | dansmith: when you'll be enough coffeenated, could we please discuss about https://bugs.launchpad.net/nova/+bug/1431201 re: https://review.openstack.org/#/c/157025/ | 13:08 |
| openstack | Launchpad bug 1431201 in OpenStack Compute (nova) "kilo controller cann't conduct juno compute nodes" [Critical,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 13:08 |
| *** dboik has quit IRC | 13:08 | |
| bauzas | dansmith: sounds like that's looping over IncompatibleObjectVersion | 13:08 |
| PaulMurray | jaypipes, dansmith, ndipanov there's series of 4 patches adding ComputeNode object to RT starting here https://review.openstack.org/#/c/148904/ | 13:08 |
| PaulMurray | jaypipes, dansmith ndipanov any takers ? :) | 13:09 |
| bauzas | PaulMurray: you don't imagine how much I would loved your series implemented by Juno... | 13:09 |
| *** TobiasE has joined #openstack-nova | 13:09 | |
| *** TobiasE has quit IRC | 13:09 | |
| *** tdurakov_ has joined #openstack-nova | 13:09 | |
| *** tdurakov__ has joined #openstack-nova | 13:10 | |
| PaulMurray | edleafe, thanks for your review | 13:10 |
| edleafe | PaulMurray: yw | 13:11 |
| ankit_ag | <garyk> hi, could you please check my reply on https://review.openstack.org/#/c/159062/. Thank you | 13:11 |
| *** TobiasE has joined #openstack-nova | 13:11 | |
| *** dboik_ has quit IRC | 13:11 | |
| *** ijw has quit IRC | 13:12 | |
| *** TobiasE has quit IRC | 13:13 | |
| *** gokrokve has joined #openstack-nova | 13:14 | |
| garyk | ankit_ag: thanks - i still thinkt hat we should swap _get_flavor with instance.flavor - i suggest that you add dansmith and melwitt to the review. they did that code and may be able to clarify. i will chang emy score if this is ok | 13:16 |
| *** gokrokve has quit IRC | 13:18 | |
| jaypipes | PaulMurray: will do in a bit, yes... on a call. | 13:18 |
| ankit_ag | <garyk> I test this patch with the change suggested by you, but it does not work and breaks down the performance because here I've removed system metadata from expected_attrs | 13:19 |
| PaulMurray | jaypipes, thanks | 13:19 |
| *** mestery has joined #openstack-nova | 13:20 | |
| ankit_ag | garyk: matthew has already added dansmith to this review. Thank you | 13:21 |
| *** vmtrooper has joined #openstack-nova | 13:23 | |
| garyk | ankit_ag: thanks! | 13:24 |
| *** Longgeek has quit IRC | 13:24 | |
| garyk | ankit_ag: if use the instance.flav cause a degradation of the performance then i think that dansmith should be aware. that is troubling | 13:24 |
| garyk | ankit_ag: i think that he is only online in an hour or so… | 13:24 |
| ankit_ag | garyk: Sure I'll try to catch him on irc if possible. Thaks for your time and help :) | 13:26 |
| eliqiao | hello all, any cores around to help to merge https://review.openstack.org/#/c/109178/ , this is pending for long time , and it is ready to get merged since the dependency patch has been merge recently. | 13:26 |
| *** vmtrooper has quit IRC | 13:29 | |
| *** eliqiao has quit IRC | 13:29 | |
| dansmith | bauzas: +2 on the conductor fix | 13:31 |
| *** dboik has joined #openstack-nova | 13:32 | |
| *** dboik has quit IRC | 13:32 | |
| *** dboik has joined #openstack-nova | 13:32 | |
| *** Nikolay_St has quit IRC | 13:33 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Rename and move the v2.1 api policy into separated files https://review.openstack.org/150267 | 13:33 |
| dansmith | bauzas: on the second thing, I think you're not backleveling the compute_node according to the rules | 13:34 |
| dansmith | bauzas: so if they ask for version 1.5 you still give them compute_node 1.10, right? | 13:35 |
| *** Nikolay_St has joined #openstack-nova | 13:35 | |
| *** mriedem_away is now known as mriedem | 13:36 | |
| *** shyama has quit IRC | 13:37 | |
| *** annegentle has joined #openstack-nova | 13:37 | |
| *** Longgeek has joined #openstack-nova | 13:37 | |
| *** abhishekkk has quit IRC | 13:39 | |
| *** hemna_ has quit IRC | 13:40 | |
| dansmith | garyk: get_flavor() and instance.flavor are the same under the covers, but yes it would be nice to get it converted to instance.flavor for aesthetics and cleanliness | 13:40 |
| *** dave-mccowan has quit IRC | 13:40 | |
| garyk | dansmith: ok. thanks | 13:41 |
| garyk | ankit_ag: ^^ | 13:41 |
| openstackgerrit | garyk proposed openstack/nova: libvirt: add support for instance update https://review.openstack.org/163818 | 13:43 |
| ankit_ag | dansmith: I did not understood your comment ". This change would cause the tenant usage to be wrong for instances when a flavor has changed" | 13:44 |
| *** abhishekkk has joined #openstack-nova | 13:44 | |
| bauzas | dansmith: yey, just wondering how can I get a IncompatibleObjectVersion exception | 13:44 |
| bauzas | dansmith: sounds like you're right | 13:44 |
| dansmith | ankit_ag: this change pulls the flavor information from the database according to the flavor now, not when the instance was run | 13:44 |
| *** burt has joined #openstack-nova | 13:44 | |
| bauzas | dansmith: I should return a 1.5 | 13:44 |
| bauzas | dansmith: when calling obj_make_compatible | 13:44 |
| ankit_ag | because when we resize an instance, instance_type_id also gets updated | 13:44 |
| dansmith | bauzas: so probably what you want to do is call super() right after you fix the compute node | 13:45 |
| *** adelinatuvenie has joined #openstack-nova | 13:45 | |
| dansmith | bauzas: super() will backlevel it according to the relationship map | 13:45 |
| bauzas | dansmith: /me facepalms | 13:45 |
| dansmith | ankit_ag: yes, I understand | 13:45 |
| bauzas | dansmith: you're probably right, let me check | 13:45 |
| bauzas | dansmith: the main problem is that Grenade didn't disagree | 13:45 |
| bauzas | dansmith: so I don't know how to really test it | 13:46 |
| dansmith | bauzas: this doesn't happen as part of a request, right? it's part of a periodic and/or a restart, right? | 13:46 |
| bauzas | dansmith: I should rather create 2 VMs on my env, and try it directly | 13:46 |
| bauzas | dansmith: yey | 13:46 |
| bauzas | dansmith: that just fails periodically | 13:46 |
| dansmith | bauzas: yeah, so it's going to be harder to test, unfortunately | 13:46 |
| dansmith | bauzas: but with unit tests you can make sure that for 1.5 you get back a service primitive with a compute node primitive at the right version | 13:47 |
| dansmith | bauzas: and then a local test would be prudent | 13:47 |
| bauzas | dansmith: I added a unittest for it, but sounds like I missed the point | 13:47 |
| dansmith | bauzas: you added a unit test for the juno version of service? | 13:47 |
| bauzas | dansmith: nope, for testing 1.10 | 13:47 |
| dansmith | if so, then you just didnt' assert the compute_node version | 13:47 |
| dansmith | right | 13:47 |
| dansmith | bauzas: I missed it too, it's complicated stuff | 13:48 |
| bauzas | dansmith: ok I see, lemme try to see how I can provide it | 13:48 |
| bauzas | thanks for your help | 13:48 |
| *** Marga_ has quit IRC | 13:49 | |
| dansmith | bauzas: sure, glad we found this now and not after kilo :) | 13:49 |
| bauzas | agreed | 13:49 |
| *** dkliban_afk is now known as dkliban | 13:50 | |
| *** Marga_ has joined #openstack-nova | 13:50 | |
| bauzas | dansmith: I will also think about how we can cover a backwards compatibilty for service periodic tasks, either in Grenade or elsewhere | 13:50 |
| bauzas | dansmith: I should even open a ML convo for this | 13:51 |
| dansmith | bauzas: well, we try not to restart the grenade services specifically | 13:51 |
| dansmith | bauzas: and I was thinking that if we did, they would restart under the newer code | 13:51 |
| dansmith | bauzas: so, doing that will require some monkeywork, but I agree of course | 13:51 |
| bauzas | dansmith: agreed, but the question is how to cover this | 13:51 |
| dansmith | yup | 13:52 |
| bauzas | dansmith: because more and more people are looking at rolling upgrades and I'm not happy to see we can still pay a technical debt for it | 13:52 |
| dansmith | potentially with the aiopcn tests to keep them separate | 13:52 |
| bauzas | I pass | 13:52 |
| bauzas | I don't know what's this ? | 13:52 |
| dansmith | multinode testing in the gate | 13:53 |
| bauzas | ooooh | 13:53 |
| dansmith | one node that is an all-in-one devstack, and another node that is just a compute | 13:53 |
| *** abhishekkk has quit IRC | 13:53 | |
| dansmith | if we don't upgrade the compute, then we'd get a very real environment and then we could try doing a restart of the old node after we've run tests to make sure it's still okay | 13:53 |
| bauzas | oh aiopcn = all in one * compute node ? | 13:53 |
| dansmith | yar | 13:53 |
| bauzas | ok, let's open a thread for this | 13:54 |
| *** ankit_ag has quit IRC | 13:55 | |
| *** jprovazn has quit IRC | 13:56 | |
| sdague | can we test this in the functional tree, or are we always going back to an oslo.config var ? | 13:58 |
| dansmith | sdague: it would be very synthetic, because what we need is to test juno code against kilo code | 13:59 |
| sdague | ah, gotcha, and not just the juno messages | 13:59 |
| dansmith | right, the juno behavior | 14:00 |
| sdague | do you have a summary of what code path was missed? | 14:00 |
| dansmith | we're sending back something we meant to be legit and juno is kicking it back saying "no bitch, I said VERSION 1.5" | 14:00 |
| *** baoli has quit IRC | 14:00 | |
| dansmith | sdague: juno asks for 1.5 we send it 1.10, it says, no I wanted 1.5, we say "how about 1.10?" | 14:00 |
| bauzas | dansmith: it comes to me that we should maybe patch the objects by setting a var like in RPC API saying "eh : juno == 1.4" | 14:01 |
| dansmith | bauzas: why does that help? | 14:01 |
| mriedem | issue brought up in here about error handling, looking for some more opinions https://review.openstack.org/#/c/109178/38/nova/compute/manager.py | 14:01 |
| sdague | so that seems like we could in tree test the 'not correctly downgrading' part | 14:01 |
| bauzas | dansmith: dunno, just about visibility | 14:01 |
| dansmith | sdague: we could and we do, we just missed it on this one change | 14:02 |
| *** tdurakov___ has joined #openstack-nova | 14:02 | |
| *** baoli_ has joined #openstack-nova | 14:02 | |
| *** apuimedo has quit IRC | 14:02 | |
| bauzas | sdague: agreed, it was just that I was stupid enough | 14:02 |
| *** baoli_ has quit IRC | 14:02 | |
| *** tdurakov____ has joined #openstack-nova | 14:02 | |
| *** thangp has joined #openstack-nova | 14:02 | |
| dansmith | we have a test that tries to backport an object to every version it claims to support, but it doesn't check nested objects | 14:03 |
| sdague | did the partial cpu job dump some representative error message in the log? We could also fail that test if things like that end up in logs | 14:03 |
| dansmith | so we could make that test recurse, but ... | 14:03 |
| sdague | dansmith: ah, gotcha | 14:03 |
| bauzas | sdague: but we can't hardly assess that all people will be brilliant, my question being then "how we can make sure that dumb people are not enough foolish to break compatibility" | 14:03 |
| *** r-daneel has joined #openstack-nova | 14:03 | |
| dansmith | sdague: no because it only happens when you start the compute node, or during the periodic task | 14:03 |
| *** dave-mccowan has joined #openstack-nova | 14:03 | |
| *** apuimedo has joined #openstack-nova | 14:03 | |
| bauzas | dansmith: to be clear, it logged the recursion | 14:03 |
| dansmith | bauzas: sure | 14:04 |
| bauzas | dansmith: sdague: http://logs.openstack.org/25/157025/5/check/check-grenade-dsvm-partial-ncpu/7766f56/logs/old/screen-n-cpu.txt.gz | 14:04 |
| *** baoli_ has joined #openstack-nova | 14:04 | |
| *** krykowski has joined #openstack-nova | 14:04 | |
| *** gszasz_ has quit IRC | 14:04 | |
| sdague | bauzas: line link please | 14:04 |
| bauzas | but as it was a periodic task, it said "eh, no worries !" | 14:04 |
| bauzas | sdague: sure | 14:04 |
| bauzas | sdague: http://logs.openstack.org/25/157025/5/check/check-grenade-dsvm-partial-ncpu/7766f56/logs/old/screen-n-cpu.txt.gz#_2015-03-03_17_26_51_537 | 14:04 |
| *** tdurakov_ has quit IRC | 14:05 | |
| bauzas | sdague: as I said, it was a PEBKAC | 14:05 |
| *** tdurakov_ has joined #openstack-nova | 14:05 | |
| sdague | bauzas: sure | 14:05 |
| *** tdurakov_____ has joined #openstack-nova | 14:06 | |
| dansmith | bauzas: modifying the backport test to check sub-objects would be some improvement | 14:06 |
| *** Maike_ has quit IRC | 14:06 | |
| *** tdurakov has quit IRC | 14:06 | |
| *** ijw has joined #openstack-nova | 14:06 | |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova: VMware: use get_object_properties_dict from oslo.vmware https://review.openstack.org/163831 | 14:06 |
| *** tdurakov__ has quit IRC | 14:06 | |
| dansmith | bauzas: even if we only go one deep, that should be enough and not have to fully recurse | 14:06 |
| *** tdurakov____ has quit IRC | 14:06 | |
| bauzas | dansmith: agreed | 14:06 |
| *** tdurakov___ has quit IRC | 14:06 | |
| *** tdurakov_____ has quit IRC | 14:06 | |
| *** nelsnelson has joined #openstack-nova | 14:06 | |
| dansmith | I can work on that | 14:06 |
| *** tdurakov_ has quit IRC | 14:06 | |
| *** nelsnelson has quit IRC | 14:07 | |
| bauzas | dansmith: feel free, I would spend more time than necessary for this | 14:07 |
| *** tdurakov has joined #openstack-nova | 14:07 | |
| sdague | it still looks like the periodic trip would catch this if we specified that we should never see that in logs - http://logstash.openstack.org/#eyJzZWFyY2giOiJcIm1heGltdW0gcmVjdXJzaW9uIGRlcHRoIGV4Y2VlZGVkXCIiLCJmaWVsZHMiOltdLCJvZmZzZXQiOjAsInRpbWVmcmFtZSI6IjYwNDgwMCIsImdyYXBobW9kZSI6ImNvdW50IiwidGltZSI6eyJ1c2VyX2ludGVydmFsIjowfSwic3RhbXAiOjE0MjYxNjkxODk0NDh9 | 14:07 |
| *** oro_ has joined #openstack-nova | 14:07 | |
| *** tdurakov has quit IRC | 14:07 | |
| *** oro has joined #openstack-nova | 14:07 | |
| dansmith | well, I was fine to chalk this up as an accident, but if making the test go one deep isn't too hard, it's probably worth it | 14:07 |
| sdague | so adding 'bad compute messages' test at the end of the partial cpu job would probably be a reasonable safety check | 14:07 |
| *** apevec has left #openstack-nova | 14:08 | |
| *** irenab has quit IRC | 14:08 | |
| *** jecarey has joined #openstack-nova | 14:08 | |
| mdbooth | dansmith: So, instance_type_id isn't flavor of the month? ;) | 14:08 |
| dansmith | sdague: well, checking that we never see "maximum recursion depth exceeded" for any reason would be good :) | 14:08 |
| mdbooth | Sorry | 14:08 |
| *** Marga_ has quit IRC | 14:08 | |
| dansmith | mdbooth: heh, nop :) | 14:08 |
| dansmith | er, nope | 14:08 |
| bauzas | sdague: how do you propose to gate it N | 14:08 |
| bauzas | ? | 14:08 |
| mdbooth | dansmith: Did you read the comment about the performance, btw? | 14:08 |
| bauzas | sdague: I was thinking that e-r was only for failures | 14:09 |
| sdague | bauzas: at the end of the grenade job, comb the logs before upgrade | 14:09 |
| dansmith | mdbooth: yes | 14:09 |
| sdague | if you find that message, fail the job | 14:09 |
| mdbooth | dansmith: (Take your word on the Flavor thing, btw) | 14:09 |
| bauzas | sdague: oic | 14:09 |
| sdague | sorry, before 'upload' | 14:09 |
| *** otter768 has joined #openstack-nova | 14:09 | |
| dansmith | mdbooth: so after kilo, we can make a call to drop the system_metadata bits for flavor compatibility and just say everyone needs to make sure they have migrated their flavors (using the tool) before lemming | 14:10 |
| mdbooth | dansmith: I was wondering if we could get closer to a bare db query within current design constraints. | 14:10 |
| dansmith | mdbooth: which should get us to the point where that is (or could be made to be) a single call with a join | 14:10 |
| dansmith | mdbooth: we need to decide if we are okay making that a one-cycle mandatory conversion, but if we are, then we can drop it early in lemming | 14:11 |
| mdbooth | dansmith: How about the object instantiation overhead, though? IIUC that was were the time was going. | 14:11 |
| *** gszasz has joined #openstack-nova | 14:11 | |
| *** gszasz has quit IRC | 14:11 | |
| *** ijw has quit IRC | 14:12 | |
| *** amotoki has joined #openstack-nova | 14:12 | |
| *** shyama has joined #openstack-nova | 14:14 | |
| *** Maike has joined #openstack-nova | 14:14 | |
| *** gszasz has joined #openstack-nova | 14:14 | |
| *** gokrokve has joined #openstack-nova | 14:14 | |
| *** rushiagr is now known as rushiagr_away | 14:14 | |
| *** thomasem_ has quit IRC | 14:14 | |
| dansmith | mdbooth: I'd want to see a profile showing that it's substantial enough to care | 14:14 |
| dansmith | mdbooth: regardless, we can't do anything until we can drop compatibility, so lets talk about optimizing after that is done | 14:14 |
| *** otter768 has quit IRC | 14:15 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Add ability to pause instance during live migration https://review.openstack.org/163836 | 14:15 |
| *** iamjarvo has joined #openstack-nova | 14:18 | |
| alex_xu | hi, any core can give another +2 for those easy patch https://review.openstack.org/150258 https://review.openstack.org/163689 https://review.openstack.org/159376 | 14:18 |
| *** thomasem has joined #openstack-nova | 14:18 | |
| *** gokrokve has quit IRC | 14:19 | |
| *** adelinatuvenie has quit IRC | 14:19 | |
| *** vladikr has joined #openstack-nova | 14:23 | |
| sdague | dansmith: any chance of running through the rest of my test stack today https://review.openstack.org/#/c/161710/ (just 2 patches left) | 14:23 |
| dansmith | sdague: there's a chance, yes | 14:23 |
| mdbooth | So, is InstanceExtra.flavor a json blob or something like that, now? | 14:23 |
| mriedem | it'd be nice if the compute driver's attach_interface method didn't potentially raise a dozen different errors | 14:24 |
| sdague | bauzas: so on your patch - http://logs.openstack.org/97/163797/1/check/check-grenade-dsvm-partial-ncpu/04cd3c4/logs/old/screen-n-cpu.txt.gz?level=TRACE#_2015-03-12_13_10_06_467 is still a thing | 14:25 |
| dansmith | sdague: his patch is only for half the problem | 14:25 |
| dansmith | sdague: the bug had two buglets inside | 14:25 |
| bauzas | sdague: hence the related-bug | 14:25 |
| bauzas | sdague: as I said in a comment, there are 2 problems | 14:25 |
| sdague | bauzas: ok | 14:26 |
| bauzas | and I'm only fixing the first one | 14:26 |
| bauzas | the second one still need to have a proper failing unittest in place before fixing it :) | 14:26 |
| bauzas | I need to reproduce an error | 14:26 |
| sdague | bauzas: you could push the functional code and verify it in the partial test upstream while working on the unit test | 14:27 |
| openstackgerrit | Lingxian Kong proposed openstack/nova: Raise exception when using invalid az for vm creation https://review.openstack.org/163842 | 14:27 |
| mriedem | vishy: russellb: do you happen to remember the story around when neutron vif type handling was added to the libvirt driver back in the day? i.e. why we decided to just do all of this vif stuff locally on the compute node and let the neutron L2 agent poll to pick up the changes and report back to the neutron server, | 14:28 |
| mriedem | rather than talking to neutron directly? | 14:28 |
| bauzas | sdague: eh, I learnt TDD a couple of years ago :) | 14:29 |
| mriedem | vishy: russellb: i assume it was because there was already code that worked for nova-network and linuxbridge, and we just piled on to get neutron working? | 14:29 |
| *** TobiasE has joined #openstack-nova | 14:29 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: neutron: deprecate 'allow_duplicate_networks' config option https://review.openstack.org/163581 | 14:29 |
| mriedem | ^ was already approved if a core can +W, was a trivial rebase | 14:30 |
| sdague | bauzas: yes, what I'm saying is we kind of already have a test that validates the functional code, though it currently requires manual inspection. It would be good to get those results regardless of your unit test, which will be somewhat synthetic | 14:30 |
| sdague | mriedem: +A | 14:31 |
| mriedem | sdague: thanks | 14:31 |
| *** popw1 has quit IRC | 14:31 | |
| garyk | mriedem: in a large majority of deployments there is actually no l2 agent. an external controller will configure the open vswicth. | 14:32 |
| *** gokrokve has joined #openstack-nova | 14:33 | |
| mriedem | garyk: so what is the libvirt vif driver talking to in that case? | 14:33 |
| mriedem | it's still running ovs commands on the compute node | 14:33 |
| mriedem | is the external controller talking to neutron then? | 14:33 |
| garyk | mriedem: there are some ovs commands, but the actual 'network connectivity' is done by a controller | 14:34 |
| garyk | in the neutron ovs case the agent will configure the actual vlan/segmenation id | 14:34 |
| garyk | no, neutron will configure the external controller and that will interface directly with the ovs. | 14:34 |
| mriedem | garyk: so regardless of L2 agent or external controller, it seems we still are doing a thing in the compute node that we assume will get back to neutron somehow so neutron can update it's state, | 14:35 |
| garyk | there are 2 parts. the first is configuring libvirt with the necessary data and the second is ensure that the ovs know what to do with the specific port | 14:35 |
| *** marun has joined #openstack-nova | 14:35 | |
| mriedem | so we're doing stuff 'out of band' from neutrons POV, which seems bad | 14:35 |
| garyk | mriedem: yes, in certain cases we are. | 14:35 |
| mriedem | i.e. racey | 14:35 |
| garyk | i am not 100% sure of the speicifc case you are referring to. can you please point me to the code | 14:36 |
| mriedem | why don't the virt drivers talk to a network API (neutron in this case) when doing plug/unplug? | 14:36 |
| *** achanda has joined #openstack-nova | 14:36 | |
| mriedem | pretty much any of the plug/unplug code in the libvirt vif driver | 14:36 |
| garyk | the plug/unplu needs to configure the xml with the relevant neutron id's | 14:36 |
| *** TobiasE has quit IRC | 14:37 | |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/vif.py#n407 | 14:37 |
| garyk | this is so that the security groups can be used | 14:37 |
| mriedem | garyk: i don't see how that matters | 14:37 |
| mriedem | garyk: but when using neutron we're supposed to be using neutron's security groups aren't we? | 14:38 |
| garyk | neutron is not aware of the host | 14:38 |
| garyk | neutrons security groups in the case of the opensource code use ip tables on the host | 14:38 |
| *** tdurakov has joined #openstack-nova | 14:38 | |
| garyk | this means that the traffic needs to get from the ovs to the iptables | 14:38 |
| garyk | hence the bridge here | 14:38 |
| garyk | in neutron there are now patches in review that will do this in the ovs | 14:39 |
| *** eharney has joined #openstack-nova | 14:39 | |
| garyk | maybe when those land we can think on how to deprecate the code you are refrring to | 14:39 |
| mriedem | yeah i'm not a network guy, but it seems to me that if we are going to have a network service, at a high level we should defer network things to said network service, | 14:39 |
| mriedem | like how we defer image and volume things to the image and volume service | 14:39 |
| *** adelinatuvenie has joined #openstack-nova | 14:39 | |
| mriedem | doing all of this out of band round tripping seems like a big race condition possibility | 14:40 |
| mriedem | this is the reason why we needed the external events stuff that dansmith and arosen worked on, | 14:40 |
| *** mpaolino has joined #openstack-nova | 14:40 | |
| mriedem | if plug_vifs was just a neutron api call and we waited until that came back, it'd be much simpler | 14:41 |
| garyk | i agree that it is problematic. my understanding was that we were going to address this with the neutron/nova interface. sadly this was stuck in the review process of the spec this entire cycle | 14:41 |
| dansmith | mriedem: we'd still need the event, because we can't block on a rest call for ten minutes :) | 14:41 |
| dansmith | mriedem: but otherwise I agree with what you're saying of course :) | 14:41 |
| garyk | the vif plug is the connection of the virtual network to the virtual machine | 14:41 |
| garyk | this should be done int he context of nova | 14:41 |
| garyk | i do not see how neutron can do this. | 14:41 |
| mriedem | dansmith: but we're blocking on the event for 5 minutes right? | 14:42 |
| dansmith | mriedem: yes, I just mean we still need some way to make the request and want for an async response | 14:42 |
| *** apuimedo is now known as apuimedo|away | 14:42 | |
| *** r-daneel has quit IRC | 14:42 | |
| garyk | dansmith: it would have been good if that api returned an id that nova could poll on | 14:42 |
| mriedem | dansmith: ok, seems we could have polled neutron for that info or something, seems like we do that kind of stuff in tempest all the time | 14:42 |
| dansmith | garyk: polling sucks | 14:43 |
| garyk | why? | 14:43 |
| dansmith | because it | 14:43 |
| dansmith | is inefficient? | 14:43 |
| *** unicell1 has joined #openstack-nova | 14:43 | |
| dansmith | it wastes resources on both sides | 14:43 |
| garyk | i think that is an implementation detail. | 14:43 |
| mriedem | true, we don't need to hit the neutron API constantly | 14:43 |
| mriedem | when we have an event callback | 14:43 |
| *** unicell has quit IRC | 14:43 | |
| *** achanda has quit IRC | 14:44 | |
| dansmith | well, that's why I said above that we still need the events mechanism, but it would be good if we were waiting for neutron to actually do the thing and not have us do it | 14:44 |
| garyk | the event callback is error prone in the case that there is a lot of load on the service | 14:44 |
| dansmith | which service? | 14:45 |
| mriedem | garyk: "the vif plug is the connection of the virtual network to the virtual machine" - isn't that happening later in the libvirt driver when the domain xml is created? | 14:45 |
| garyk | the neutron service | 14:45 |
| *** r-daneel has joined #openstack-nova | 14:45 | |
| dansmith | garyk: and us just polling it until it's done is less load? :) | 14:45 |
| garyk | mriedem: i think that it depends on the actual vif driver used. | 14:45 |
| garyk | dansmith: at least one would not get random failures when the connection can actually take place. | 14:46 |
| garyk | i just wonder in cases when an external controoler is used how this affect the events | 14:46 |
| garyk | controller | 14:46 |
| bauzas | dansmith: may I ask a question about nested dependencies ? | 14:47 |
| bauzas | dansmith: I can now reproduce the bug by a unittesr | 14:47 |
| dansmith | bauzas: sure, because I'm only in two conversations at the moment :P | 14:47 |
| dansmith | bauzas: cool | 14:47 |
| bauzas | dansmith: no worries, taking my ticket | 14:47 |
| mriedem | for example: http://git.openstack.org/cgit/openstack/nova/tree/nova/network/linux_net.py#n1367 | 14:48 |
| bauzas | danms = objects.DanSmith() | 14:48 |
| bauzas | danms.lock() | 14:48 |
| mriedem | that's the ovs hybrid case, the only thing used from the libvirt driver there is the instance uuid | 14:48 |
| mriedem | neutron has the device_id == instance_uuid when the port is created | 14:48 |
| dansmith | bauzas: go ahead, I'd rather have your conversation than the others | 14:48 |
| mriedem | so it seems that neutron could easily do http://git.openstack.org/cgit/openstack/nova/tree/nova/network/linux_net.py#n1367 instead of nova | 14:48 |
| dansmith | bauzas: I'm also close to a generic test, and I may have found another inconsistency or two | 14:48 |
| bauzas | dansmith: so I found that compute_node.VERSION sticks with 1.10 | 14:49 |
| bauzas | dansmith: now I want to convert it to an acceptable version | 14:49 |
| mriedem | garyk: unrelated, but can you also comment in this bug? https://bugs.launchpad.net/nova/+bug/1430512 | 14:49 |
| openstack | Launchpad bug 1430512 in OpenStack Compute (nova) "Adding a fixed_ip to an instance with neutron only updates the first port using the first subnet" [Undecided,New] | 14:49 |
| mriedem | garyk: tl;dr is add/remove fixed_ips in the neutronv2 API seems pretty wonky | 14:49 |
| bauzas | dansmith: but calling obj_make_compatible on the Service object doesn't backport it to the appropriate version | 14:49 |
| garyk | mriedem: i think that is spefific for a particular driver (not 100% sure though). i think that libvirt now has the ability to do that interanlly (or it did about 15 months ago) | 14:49 |
| *** apuimedo|away has quit IRC | 14:49 | |
| bauzas | dansmith: even if I'm moving super() at the end of the queue | 14:49 |
| dansmith | bauzas: yeah, you can't change objects.ComputeNode.VERSION because that's the class definition | 14:50 |
| bauzas | dansmith: oic | 14:50 |
| garyk | mriedem: sure, i will take alook | 14:50 |
| bauzas | dansmith: so I should rather check obj_to_primitive() ? | 14:50 |
| dansmith | bauzas: yes | 14:51 |
| dansmith | bauzas: check the version in the primitive | 14:51 |
| bauzas | dansmith: ack | 14:51 |
| *** shyama has quit IRC | 14:51 | |
| garyk | mriedem: at the moment the ipam in neutron works like that. there is currently work to have a ipam driver. that will certain change the way in which this is done | 14:51 |
| bauzas | dansmith: I hope to get my green belt on objects soon then | 14:52 |
| dansmith | bauzas: hah :) | 14:52 |
| *** shyama has joined #openstack-nova | 14:52 | |
| garyk | mriedem: i also think that the way that we handle it in the nova api is also a bit broken. | 14:52 |
| mriedem | bauzas: just be sure to have dansmith explain the magic context that may or may not need to be passed to a method | 14:52 |
| mriedem | :) | 14:52 |
| mriedem | for the 50th time | 14:52 |
| dansmith | AARGH | 14:53 |
| dansmith | :P | 14:53 |
| bauzas | mriedem: lol | 14:53 |
| *** thedodd has joined #openstack-nova | 14:53 | |
| *** Nikolay_St has quit IRC | 14:53 | |
| mriedem | you can't enter the nova objects dojo until you've asked about that | 14:53 |
| *** Nikolay_St has joined #openstack-nova | 14:53 | |
| bauzas | mriedem: I was already aware of it, is that ok ? | 14:53 |
| bauzas | mriedem: because I probably asked the question a while ago :D | 14:53 |
| mriedem | bauzas: it's better than ok | 14:53 |
| *** tonytan4ever has joined #openstack-nova | 14:53 | |
| bauzas | mriedem: that doesn't prevent me to explicitely write methods passing context as args | 14:54 |
| bauzas | :) | 14:54 |
| *** obutenko has joined #openstack-nova | 14:54 | |
| bauzas | mriedem: because I feel the pain of removing that arg :) | 14:54 |
| tonytan4ever | Hello nova folks. | 14:54 |
| dansmith | bauzas: it does, I already merged the patch that will prevent you from doing that :) | 14:54 |
| bauzas | dansmith: \o/ | 14:55 |
| garyk | i still do not understand the argument and removing it. sorry but my simple brain cannot comprehend that | 14:55 |
| tonytan4ever | I am trying to find the nova deploy documentation, and can someone points to it ? | 14:55 |
| bauzas | garyk: contexts are localized to the object's lifetimle | 14:55 |
| mriedem | tonytan4ever: docs.openstack.org | 14:55 |
| bauzas | tonytan4ever: sure | 14:55 |
| *** Longgeek has quit IRC | 14:55 | |
| bauzas | tonytan4ever: http://docs.openstack.org/developer/nova/devref/ | 14:56 |
| tonytan4ever | Thanks | 14:56 |
| *** penick has joined #openstack-nova | 14:56 | |
| bauzas | tonytan4ever: oops, deploy doc ? yey check rather docs.openstack.org in general | 14:56 |
| *** penick has quit IRC | 14:56 | |
| *** mpaolino has quit IRC | 14:57 | |
| bauzas | and garyk: s/localized/tied (sorry for that very bad Frenglish) | 14:57 |
| garyk | i think i understand. | 14:58 |
| *** dboik_ has joined #openstack-nova | 14:58 | |
| garyk | mriedem: this is dodgy: https://github.com/openstack/nova/blob/2015.1.0b2/nova/network/neutronv2/api.py#L768 | 14:58 |
| mriedem | garyk: yes, it sure is :) | 14:59 |
| mriedem | nested loops and it returns on the first one set | 14:59 |
| ndipanov | PaulMurray, hey - reviewed the rt series | 14:59 |
| ndipanov | awesome work | 14:59 |
| mriedem | garyk: the tempest test doesn't validate much in this case, only that it can add a fixed ip and remove it | 14:59 |
| mriedem | w/o blowing up | 14:59 |
| ndipanov | of course, I -1ed all of it | 14:59 |
| ndipanov | but for small little details | 14:59 |
| mriedem | garyk: not that it's usable in any way | 14:59 |
| ndipanov | :) | 15:00 |
| mriedem | garyk: note that code is basically untouched since it was added back in grizzly | 15:00 |
| *** baoli_ has quit IRC | 15:00 | |
| mriedem | so i don't know if it's even used really in the real world | 15:00 |
| *** dboik has quit IRC | 15:00 | |
| garyk | mriedem: i think that the common use case is booting with one network. | 15:01 |
| *** claudiub has joined #openstack-nova | 15:01 | |
| mriedem | garyk: yes, but that's not the only option since juno | 15:02 |
| garyk | i am not sure now nova-network deals with this. it may have been done like this as nova-network only supported booting from one vnic (i may be talking crap) | 15:02 |
| mriedem | garyk: sorry, meant multiple ports on the same network, i'm working on adding a tempest test for that now | 15:02 |
| garyk | in general a ip will be assigned by neutron for each port - that is done when the port is created. in this case we are updating a port | 15:02 |
| mriedem | garyk: i will totally accept "i was drunk as hell back in grizzly" | 15:03 |
| mriedem | as an asnwer | 15:03 |
| claudiub | mriedem: hello, thanks for the review on https://review.openstack.org/#/c/138934/ . I've replied to your comment. :) | 15:03 |
| mriedem | claudiub: so the answer is, there was stuff not tested in hyperv vmops that is now testing with this change? but it looks like it's just moving tests from test_hypervapi.py to test_vmops.py | 15:04 |
| mriedem | claudiub: so it doesn't look like an add to me, but a refactor/move | 15:05 |
| mriedem | claudiub: and i wasn't sure what the criteria was for a test being moved to test_vmops.py | 15:05 |
| claudiub | mriedem: yeah, it is a refactor. using mock instead of mox | 15:05 |
| garyk | mriedem: please see - https://github.com/openstack/nova/blob/b9a3e798dcb474bd733cb7a92219ec186831ae0c/nova/network/quantumv2/api.py | 15:05 |
| claudiub | mriedem: plus, completing the coverage on some methods | 15:06 |
| garyk | this is where it worked. | 15:06 |
| *** ijw has joined #openstack-nova | 15:06 | |
| bauzas | dansmith: so as per my findings, obj_make_compatible doesn't backport nested objects to an acceptable version | 15:06 |
| garyk | now between quantum and neutron someone opened a keg | 15:06 |
| mriedem | claudiub: but my question is, why isn't something like test_attach_volume_logout moved also? | 15:06 |
| garyk | take that back, there is the return :) | 15:06 |
| dansmith | bauzas: I don't know wha tyou mean | 15:06 |
| mriedem | claudiub: since that also uses mox | 15:07 |
| claudiub | mriedem: it is. There is a series of commits that refactors all the tests | 15:07 |
| *** gokrokve_ has joined #openstack-nova | 15:07 | |
| garyk | mriedem: now this is the real culprit - https://github.com/openstack/nova/commit/55af31cfec74244de4efbc7187a548972c4d99f3 | 15:07 |
| claudiub | mriedem: but that would have been a few thousands of LOC | 15:07 |
| mriedem | claudiub: ok, so what was the criteria for choosing the tests that were picked for this one change? | 15:07 |
| bauzas | dansmith: will upload my patch | 15:07 |
| *** bkopilov has joined #openstack-nova | 15:07 | |
| claudiub | mriedem: so, I've split it in several commits. | 15:07 |
| claudiub | mriedem: that commit is the 6th/11 commits | 15:07 |
| mriedem | claudiub: e.g. move all spawn* methods, move all migrate* tests, etc | 15:08 |
| *** baoli has joined #openstack-nova | 15:08 | |
| dansmith | bauzas: okay | 15:08 |
| claudiub | mriedem: tests that were belonging to a particular Ops class. For example, for VMOps, we refactored all the VMOps tests | 15:08 |
| openstackgerrit | sahid proposed openstack/nova: objects: introduce numa topology limits objects https://review.openstack.org/137817 | 15:08 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 15:09 | |
| claudiub | mriedem: also, example, the VolumeOps tests, all the tests related to the volumes | 15:10 |
| mriedem | claudiub: yeah, idk, b/c i see def _test_power_on was volumeops but that's moved in the vmops change | 15:10 |
| *** gokrokve has quit IRC | 15:11 | |
| *** mpaolino has joined #openstack-nova | 15:11 | |
| claudiub | mriedem: yeah, since that is a VMOps method. It is a method related to vm operations | 15:11 |
| *** ijw has quit IRC | 15:11 | |
| mriedem | claudiub: sorry for getting hung up on this, i just don't really know how to review these changes i guess. at a minimum i'd probably just see if any tests were dropped and if not, +2 | 15:12 |
| mriedem | since i'd rather not dig into the guts of the hyperv modules to see what methods are where | 15:12 |
| *** vmtrooper has joined #openstack-nova | 15:12 | |
| edleafe | ndipanov and other cores: care to give this a quick review? PaulMurray's patch and this are intertwined. https://review.openstack.org/#/c/145619/ | 15:12 |
| openstackgerrit | garyk proposed openstack/nova: Neutron: configure requested amount of fixed_ips https://review.openstack.org/163864 | 15:13 |
| garyk | mriedem: ^ (it is wip progress - i need to test….) | 15:14 |
| claudiub | mriedem: That's ok. :) That is how we are managing our hyper-v modules, we divided the code in a few portions, each dealing with a certain part: VM management, volumes, migration, networking, etc. Anyways, thanks for taking a look. :) | 15:16 |
| ndipanov | edleafe, some of these bugs are just depressing | 15:16 |
| edleafe | ndipanov: :) | 15:17 |
| *** yamahata has joined #openstack-nova | 15:17 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: WIP: Fix ComputeNode backport for Service.obj_make_compatible https://review.openstack.org/163867 | 15:17 |
| *** vmtrooper has quit IRC | 15:17 | |
| bauzas | dansmith: ^ above is the new unittest + change in the service object | 15:18 |
| bauzas | dansmith: it just shows me that the cn primitive is set in the rock with 1.10 (didn't said marble !) | 15:18 |
| dansmith | bauzas: okay, trying to dial in this test and then I'll look | 15:18 |
| *** adelinatuvenie has quit IRC | 15:19 | |
| garyk | mriedem: please ping me when you have a few secs | 15:19 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: WIP: Fix ComputeNode backport for Service.obj_make_compatible https://review.openstack.org/163867 | 15:20 |
| *** david-ly_ is now known as david-lyle | 15:23 | |
| *** tonytan4ever has quit IRC | 15:23 | |
| edleafe | ndipanov: thanks for the +2. Any other cores wanna push this through? https://review.openstack.org/#/c/145619/ | 15:24 |
| *** nellysmi_ has quit IRC | 15:24 | |
| *** Maike has quit IRC | 15:27 | |
| *** rushiagr_away is now known as rushiagr | 15:30 | |
| *** eharney has quit IRC | 15:32 | |
| *** rwsu-afk is now known as rwsu | 15:33 | |
| *** neelashah has quit IRC | 15:33 | |
| *** eharney has joined #openstack-nova | 15:33 | |
| *** kaisers has quit IRC | 15:35 | |
| *** dsanders has joined #openstack-nova | 15:35 | |
| *** setuid has quit IRC | 15:35 | |
| *** Nikolay_St has quit IRC | 15:37 | |
| *** setuid has joined #openstack-nova | 15:38 | |
| *** lpetrut has quit IRC | 15:38 | |
| *** lpetrut has joined #openstack-nova | 15:40 | |
| *** zul has quit IRC | 15:40 | |
| *** kaisers has joined #openstack-nova | 15:40 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Removed 'PYTHONHASHSEED=0' from tox.ini https://review.openstack.org/163877 | 15:41 |
| *** exploreshaifali has quit IRC | 15:41 | |
| *** zul has joined #openstack-nova | 15:41 | |
| *** whenry_ has quit IRC | 15:41 | |
| *** dsanders has quit IRC | 15:44 | |
| *** gszasz has quit IRC | 15:44 | |
| garyk | mriedem: ok. a few updates. 1. i have fixed the bug. 2. i have tested and it works. 3. it was added in havana when we only supported one fixed ip. 4. nishbar li ha ding dong | 15:44 |
| dims_ | garyk: oslo.vmware a quick review to remove oslo namespaces when you get a chance please - https://review.openstack.org/#/c/163760/ | 15:45 |
| *** neelashah has joined #openstack-nova | 15:45 | |
| *** neelashah1 has joined #openstack-nova | 15:46 | |
| *** armax has joined #openstack-nova | 15:46 | |
| garyk | dims_: done | 15:48 |
| dims_ | thanks much garyk! | 15:48 |
| *** nelsnelson has joined #openstack-nova | 15:48 | |
| PaulMurray | ndipanov thanks for review - re the comment: https://review.openstack.org/#/c/148904/13//COMMIT_MSG | 15:49 |
| *** neelashah has quit IRC | 15:50 | |
| PaulMurray | ndipanov, are you saying I should drop that sentence from the comment | 15:50 |
| garyk | dims_: np. any idea where mriedem is. he is mia? | 15:51 |
| PaulMurray | ndipanov, or that the patch is not related to the bug report? | 15:51 |
| dims_ | garyk: backscroll says he was around earlier | 15:51 |
| garyk | tx | 15:52 |
| mriedem | garyk: i'm around, will have to check that change later | 15:52 |
| mriedem | trying to fix a bug atm | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_iptables_network to work with random PYTHONHASHSEED https://review.openstack.org/163352 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_neutrounv2 to pass with random PYTHONHASHSEED https://review.openstack.org/160900 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_utils to pass with random PYTHONHASHSEED https://review.openstack.org/163353 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_neutron_driver to pass with random PYTHONHASHSEED https://review.openstack.org/160901 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests to pass in test_inject to pass with random PYTHONHASHSEED https://review.openstack.org/163354 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_cloud to pass with random PYTHONHASHSEED https://review.openstack.org/163355 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_availability_zones to pass with random PYTHONHASHSEED https://review.openstack.org/163356 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_disk_config to work with random PYTHONHASHSEED https://review.openstack.org/160671 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fix for vmware test_driver_api to pass with random PYTHONHASHSEED https://review.openstack.org/163357 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_xenapi to work with random PYTHONHASHSEED https://review.openstack.org/160670 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_simple_tenant_usage to pass with random PYTHONHASHSEED https://review.openstack.org/160897 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_matchers to pass with random PYTHONHASHSEED https://review.openstack.org/163358 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: libvirt test_cpu_info method fixed random PYTHONHASHSEED compatibility https://review.openstack.org/160669 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests in test_wsgi to pass with random PYTHONHASHSEED https://review.openstack.org/160898 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in vmware test_read_write_util to pass with random PYTHONHASHSEED https://review.openstack.org/163359 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Refactored test in linux_net to pass with random PYTHONHASHSEED https://review.openstack.org/160899 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_trusted_filters to pass with random PYTHONHASHSEED https://review.openstack.org/161314 | 15:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests in test_objects to pass with random PYTHONHASHSEED https://review.openstack.org/161313 | 15:52 |
| ndipanov | PaulMurray, well both I think... the bug specifically asks for initializing the compute_node just once and never touch the DB again | 15:54 |
| garyk | mriedem: honestly i am not even sure that we need it. who actually uses this stuff? | 15:54 |
| ndipanov | and we don't do any of that in that patch | 15:54 |
| ndipanov | we just move the code to the beginning of the method | 15:54 |
| mriedem | garyk: well that was something i asked too | 15:54 |
| ndipanov | and make it nicer | 15:54 |
| mriedem | garyk: you implemented it in grizzly :) | 15:54 |
| ndipanov | so I'd just drop any mention of that bug really | 15:54 |
| ndipanov | PaulMurray, | 15:54 |
| ndipanov | ^ | 15:54 |
| garyk | mriedem: i will add unit tests for completeness - so if someone actually does then they can benefit from the work | 15:54 |
| mriedem | garyk: if we removed it, it'd be a regression in the multinic API | 15:54 |
| garyk | mriedem: i did check the the boot with same network and multiple fixed and it works. | 15:55 |
| garyk | mriedem: agree. so i'll fix the tests and we can put this dodgy ege case behind us | 15:55 |
| mriedem | garyk: yeah, i'm working on adding a tempest test for that this week | 15:55 |
| garyk | ok, thanks | 15:55 |
| PaulMurray | ndipanov, I admit, I didn't do it because of the bug report, I noticed I had to move the code and bauzas / edleafe had a pre-existing patch that for it that they thought I could adapt | 15:56 |
| *** Longgeek has joined #openstack-nova | 15:56 | |
| PaulMurray | that's where the link to the bug came from | 15:56 |
| PaulMurray | ndipanov, so I'm happy to remove the link - but there is some relevant discussion in the bug report | 15:56 |
| *** annashen has joined #openstack-nova | 15:56 | |
| PaulMurray | ndipanov, may be worth keeping the link for that? | 15:57 |
| edleafe | PaulMurray: yeah, I just followed bauzas' code when I took it over. I didn't work off of the bug | 15:57 |
| *** browne has joined #openstack-nova | 15:57 | |
| *** annashen has quit IRC | 15:58 | |
| PaulMurray | edleafe, I'm planning to rebase on top of your patch with the changes I am going to make | 15:58 |
| edleafe | PaulMurray: ndipanov: I never understood the bug, to be honest. I didn't see where the service call was ever being made more than once in the RT. | 16:00 |
| ndipanov | PaulMurray, as long as it refernces the discussion I'm fine | 16:00 |
| ndipanov | edleafe, it's not normally, if we have the compute_node set - nothing will be called after that | 16:00 |
| dansmith | bauzas: | 16:00 |
| *** angdraug has joined #openstack-nova | 16:00 | |
| *** thedodd has quit IRC | 16:00 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix backporting objects with sub-objects that can look falsey https://review.openstack.org/163886 | 16:00 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add a test that validates object backports and child object versions https://review.openstack.org/163887 | 16:00 |
| edleafe | PaulMurray: ndipanov: once self.compute_node was created, it is never called again | 16:00 |
| dansmith | bauzas: found three issues while writing that test :) | 16:01 |
| dansmith | bauzas: although it didn't find yours, so I need to look at that | 16:01 |
| PaulMurray | ndipanov, thanks for the other suggestions - you actually pointed out some things I do later in the series that make more sense doing here. | 16:01 |
| PaulMurray | ndipanov, so pretty easy to change | 16:01 |
| PaulMurray | ndipanov, and proabably less confusing that way | 16:01 |
| ndipanov | PaulMurray, yeah I noticed that as well | 16:02 |
| ndipanov | would make the patches more self contained ... | 16:02 |
| ndipanov | also the bug for that object dict attr issue would be nice | 16:02 |
| edleafe | ndipanov: all I saw was we were eliminating some 'if self.compute_node:' logic, since we now assume it has been created | 16:03 |
| PaulMurray | ndipanov, you mean this?: https://review.openstack.org/#/c/163328/ | 16:03 |
| ndipanov | edleafe, yeah tbh - I'd close that bug as invalid after this patch | 16:03 |
| *** andreykurilin has quit IRC | 16:03 | |
| ndipanov | PaulMurray, no | 16:03 |
| ndipanov | the second to last patch in the series I reviewed | 16:04 |
| dansmith | bauzas: ah, it's because I'm setting compute_node ahead of time I think | 16:04 |
| *** tonytan4ever has joined #openstack-nova | 16:04 | |
| PaulMurray | ndipanov, oh, I understand - will do | 16:05 |
| *** ijw has joined #openstack-nova | 16:06 | |
| PaulMurray | ndipanov, the one I just referenced is a nice one though - helps lots when converting tests because most of the tests have nonsense values somewhere | 16:07 |
| *** shyama has quit IRC | 16:08 | |
| *** claudiub has quit IRC | 16:09 | |
| *** eharney has quit IRC | 16:09 | |
| *** unicell1 has quit IRC | 16:09 | |
| *** rmoe has quit IRC | 16:10 | |
| *** TobiasE has joined #openstack-nova | 16:10 | |
| *** dboik_ has quit IRC | 16:10 | |
| *** andreykurilin has joined #openstack-nova | 16:10 | |
| *** pkoniszewski has quit IRC | 16:10 | |
| *** otter768 has joined #openstack-nova | 16:10 | |
| *** dboik has joined #openstack-nova | 16:10 | |
| *** mpaolino has quit IRC | 16:11 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: compute: don't trace on InstanceNotFound in reverts_task_state https://review.openstack.org/163890 | 16:11 |
| *** ijw has quit IRC | 16:12 | |
| *** setuid has quit IRC | 16:13 | |
| *** iamjarvo has quit IRC | 16:14 | |
| *** otter768 has quit IRC | 16:15 | |
| *** bkopilov has quit IRC | 16:15 | |
| *** setuid has joined #openstack-nova | 16:15 | |
| *** lpetrut has quit IRC | 16:17 | |
| *** bkopilov has joined #openstack-nova | 16:17 | |
| *** setuid has quit IRC | 16:18 | |
| *** iamjarvo has joined #openstack-nova | 16:19 | |
| *** iamjarvo has quit IRC | 16:19 | |
| bauzas | PaulMurray: do I need to read your convo, do you need my help ? | 16:19 |
| bauzas | PaulMurray: my IRC client just blinked a lot :) | 16:20 |
| *** iamjarvo has joined #openstack-nova | 16:20 | |
| *** bkopilov has quit IRC | 16:20 | |
| *** setuid has joined #openstack-nova | 16:20 | |
| *** claudiub has joined #openstack-nova | 16:20 | |
| *** kiwik-chenrui has quit IRC | 16:21 | |
| *** kiwik-chenrui has joined #openstack-nova | 16:22 | |
| *** eharney has joined #openstack-nova | 16:22 | |
| *** Nic has joined #openstack-nova | 16:24 | |
| *** annashen has joined #openstack-nova | 16:24 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: WIP(tests) Add CellMapping object https://review.openstack.org/161867 | 16:25 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Split NovaPersistentObject mixin https://review.openstack.org/163605 | 16:25 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 16:26 | |
| *** rfolco has quit IRC | 16:26 | |
| *** Longgeek has quit IRC | 16:26 | |
| *** afazekas_ has quit IRC | 16:27 | |
| *** rmoe has joined #openstack-nova | 16:28 | |
| *** baoli has quit IRC | 16:28 | |
| *** jlanoux has quit IRC | 16:28 | |
| *** irenab has joined #openstack-nova | 16:28 | |
| *** iamjarvo has quit IRC | 16:29 | |
| *** matrohon has quit IRC | 16:30 | |
| *** salv-orlando has joined #openstack-nova | 16:31 | |
| *** annashen has quit IRC | 16:32 | |
| *** garyk has quit IRC | 16:33 | |
| openstackgerrit | Romain Hardouin proposed openstack/nova: heal_instance_info_cache_interval help clearer https://review.openstack.org/163897 | 16:36 |
| openstackgerrit | Chris Friesen proposed openstack/nova: Fix up server group policy validation https://review.openstack.org/162746 | 16:37 |
| *** TobiasE has quit IRC | 16:37 | |
| *** baoli has joined #openstack-nova | 16:38 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_iptables_network to work with random PYTHONHASHSEED https://review.openstack.org/163352 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_neutrounv2 to pass with random PYTHONHASHSEED https://review.openstack.org/160900 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_utils to pass with random PYTHONHASHSEED https://review.openstack.org/163353 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_neutron_driver to pass with random PYTHONHASHSEED https://review.openstack.org/160901 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests to pass in test_inject to pass with random PYTHONHASHSEED https://review.openstack.org/163354 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_cloud to pass with random PYTHONHASHSEED https://review.openstack.org/163355 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_availability_zones to pass with random PYTHONHASHSEED https://review.openstack.org/163356 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_disk_config to work with random PYTHONHASHSEED https://review.openstack.org/160671 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fix for vmware test_driver_api to pass with random PYTHONHASHSEED https://review.openstack.org/163357 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed test in test_xenapi to work with random PYTHONHASHSEED https://review.openstack.org/160670 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_simple_tenant_usage to pass with random PYTHONHASHSEED https://review.openstack.org/160897 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: fixed tests in test_matchers to pass with random PYTHONHASHSEED https://review.openstack.org/163358 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: libvirt test_cpu_info method fixed random PYTHONHASHSEED compatibility https://review.openstack.org/160669 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests in test_wsgi to pass with random PYTHONHASHSEED https://review.openstack.org/160898 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in vmware test_read_write_util to pass with random PYTHONHASHSEED https://review.openstack.org/163359 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Refactored test in linux_net to pass with random PYTHONHASHSEED https://review.openstack.org/160899 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored test in test_trusted_filters to pass with random PYTHONHASHSEED https://review.openstack.org/161314 | 16:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: refactored tests in test_objects to pass with random PYTHONHASHSEED https://review.openstack.org/161313 | 16:39 |
| *** zzzeek has joined #openstack-nova | 16:40 | |
| openstackgerrit | Josh Gachnang proposed openstack/nova: Add support for cleaning in Ironic driver https://review.openstack.org/161474 | 16:42 |
| *** rfolco has joined #openstack-nova | 16:43 | |
| bauzas | jaypipes: mornooning, if you have time for reviewing a critical bugfix https://review.openstack.org/#/c/163797/1 | 16:45 |
| *** jistr has quit IRC | 16:45 | |
| bauzas | jaypipes: that would be appreciated (+W needed) | 16:45 |
| *** tonytan4ever has quit IRC | 16:46 | |
| *** yamahata has quit IRC | 16:47 | |
| jaypipes | bauzas: done. | 16:48 |
| *** markus_z has quit IRC | 16:48 | |
| jaypipes | dansmith: note that I added you as a reviewer on https://review.openstack.org/#/c/161313/. It will be obvious why when you see it. :) | 16:51 |
| bauzas | jaypipes: that's one example of the pain we suffer with doing that kind of split | 16:51 |
| *** Marga_ has joined #openstack-nova | 16:52 | |
| bauzas | jaypipes: as objects are not yet fully in use in the whole Nova codebase | 16:52 |
| *** dsanders has joined #openstack-nova | 16:52 | |
| jaypipes | bauzas: yes, understood. | 16:52 |
| bauzas | jaypipes: woah, sounds I've been beaten a lot by the lenght of the series | 16:52 |
| jaypipes | bauzas: those areas are getting smaller and smaller though... | 16:52 |
| bauzas | jaypipes: I was competing on that and thought I was on a good position but the above https://review.openstack.org/#/c/161313/ blew up my dreams of medals | 16:53 |
| *** Marga__ has joined #openstack-nova | 16:55 | |
| mriedem | Mike_D_wk: replied in https://review.openstack.org/#/c/163181/ again, are you ok with handling that other method in a separate change? | 16:56 |
| *** Marga_ has quit IRC | 16:57 | |
| *** sbfox has joined #openstack-nova | 16:57 | |
| *** Marga__ has quit IRC | 16:58 | |
| *** oro_ has quit IRC | 16:58 | |
| *** Marga_ has joined #openstack-nova | 16:58 | |
| *** mwagner_lap has quit IRC | 16:59 | |
| *** oro has quit IRC | 16:59 | |
| *** afazekas has joined #openstack-nova | 16:59 | |
| *** lpetrut has joined #openstack-nova | 17:00 | |
| *** vmtrooper has joined #openstack-nova | 17:01 | |
| *** sahid has quit IRC | 17:01 | |
| *** krykowski has quit IRC | 17:03 | |
| *** jlanoux has joined #openstack-nova | 17:04 | |
| *** ybathia has joined #openstack-nova | 17:05 | |
| *** harlowja_away is now known as harlowja_ | 17:06 | |
| *** ijw has joined #openstack-nova | 17:06 | |
| *** vmtrooper has quit IRC | 17:06 | |
| *** whenry has joined #openstack-nova | 17:07 | |
| *** yamahata has joined #openstack-nova | 17:09 | |
| *** kaufer has quit IRC | 17:10 | |
| *** ijw has quit IRC | 17:11 | |
| mdbooth | Anybody know which version of eventlet had 'util' in it? | 17:12 |
| mdbooth | Never mind, 0.15 seems to do the trick | 17:13 |
| *** abhishekk has joined #openstack-nova | 17:14 | |
| *** patrickeast has joined #openstack-nova | 17:16 | |
| *** tonytan4ever has joined #openstack-nova | 17:17 | |
| *** takedakn has joined #openstack-nova | 17:19 | |
| openstackgerrit | Sean Dague proposed openstack/nova: Deprecate Nova in tree EC2 APIs https://review.openstack.org/150929 | 17:20 |
| *** tonytan4ever has quit IRC | 17:21 | |
| *** melwitt has joined #openstack-nova | 17:22 | |
| *** kaufer has joined #openstack-nova | 17:23 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 17:25 | |
| *** aix has quit IRC | 17:25 | |
| *** Nikolay_St has joined #openstack-nova | 17:25 | |
| *** devlaps has joined #openstack-nova | 17:26 | |
| *** pask81 has quit IRC | 17:29 | |
| *** wuhg has quit IRC | 17:29 | |
| *** achanda has joined #openstack-nova | 17:32 | |
| *** andreykurilin_ has joined #openstack-nova | 17:33 | |
| *** mtanino has joined #openstack-nova | 17:34 | |
| *** tjones1 has joined #openstack-nova | 17:35 | |
| *** andreykurilin_ has quit IRC | 17:41 | |
| *** mpavone has quit IRC | 17:41 | |
| *** tonytan4ever has joined #openstack-nova | 17:43 | |
| *** angdraug has quit IRC | 17:43 | |
| openstackgerrit | Sean Dague proposed openstack/nova: VMware: Deprecation warning - map one nova-compute to one VC cluster https://review.openstack.org/163426 | 17:44 |
| sdague | dims_: I just respun your thing with the way dhellmann said we should be making standard log deprecation calls | 17:45 |
| dims_ | sdague: thanks! | 17:49 |
| sdague | once it passes tests, I'll +A, as dansmith and mriedem were both +2 on ti | 17:50 |
| dims_ | perfect | 17:51 |
| *** Mike_D_laptop has joined #openstack-nova | 17:52 | |
| *** Marga_ has quit IRC | 17:53 | |
| *** claudiub has quit IRC | 17:54 | |
| *** exploreshaifali has joined #openstack-nova | 17:55 | |
| *** Marga_ has joined #openstack-nova | 17:56 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 17:57 | |
| *** iamjarvo has joined #openstack-nova | 18:00 | |
| *** annegentle has quit IRC | 18:00 | |
| *** Mike_D_laptop has quit IRC | 18:00 | |
| mriedem | sdague: one little comment in here, about a comment https://review.openstack.org/#/c/160423/ | 18:01 |
| mriedem | sdague: would be nice to have that comment so others don't fall into th same trap, but not a biggy | 18:01 |
| mriedem | i would self approve if you updated it | 18:02 |
| sdague | mriedem: I'll update | 18:02 |
| mriedem | k | 18:02 |
| sdague | thanks for the review, update coming shortly | 18:02 |
| *** igordcard_ has joined #openstack-nova | 18:02 | |
| *** derekh has quit IRC | 18:03 | |
| *** dims__ has joined #openstack-nova | 18:03 | |
| *** dims__ has quit IRC | 18:03 | |
| *** dims__ has joined #openstack-nova | 18:04 | |
| openstackgerrit | Sean Dague proposed openstack/nova: implement user negative testing for flavor manage https://review.openstack.org/160423 | 18:05 |
| mriedem | got it | 18:05 |
| dansmith | sdague: is that the set you wanted me to look at? | 18:06 |
| sdague | yeh | 18:06 |
| *** mwagner_lap has joined #openstack-nova | 18:06 | |
| *** ijw has joined #openstack-nova | 18:06 | |
| sdague | mriedem did already though, and jogo had +2s on it previously | 18:06 |
| *** jlanoux has quit IRC | 18:06 | |
| dansmith | looks like I'm off the hook tho | 18:06 |
| dansmith | cool | 18:06 |
| *** dims_ has quit IRC | 18:07 | |
| openstackgerrit | garyk proposed openstack/nova: Neutron: configure requested amount of fixed_ips https://review.openstack.org/163864 | 18:09 |
| *** dims__ has quit IRC | 18:09 | |
| *** Longgeek has joined #openstack-nova | 18:11 | |
| *** otter768 has joined #openstack-nova | 18:11 | |
| *** vladikr has quit IRC | 18:11 | |
| mriedem | ummmm | 18:11 |
| *** ijw has quit IRC | 18:12 | |
| *** annegentle has joined #openstack-nova | 18:12 | |
| mriedem | anyone else seeing this completely explode in test_neutronv2? https://github.com/openstack/nova/commit/92074e0fde7da6262897b08d100fedcfe7e4940f | 18:12 |
| *** lucasagomes has quit IRC | 18:12 | |
| *** annegentle has quit IRC | 18:15 | |
| *** annegentle has joined #openstack-nova | 18:15 | |
| *** baoli has quit IRC | 18:16 | |
| *** otter768 has quit IRC | 18:16 | |
| *** vladikr has joined #openstack-nova | 18:17 | |
| *** achanda has quit IRC | 18:17 | |
| *** eglynn has quit IRC | 18:18 | |
| *** dprince has quit IRC | 18:18 | |
| mriedem | http://paste.openstack.org/show/191892/ | 18:18 |
| *** ndipanov has quit IRC | 18:19 | |
| openstackgerrit | melanie witt proposed openstack/python-novaclient: Add missing servers.create parameter documentation https://review.openstack.org/163923 | 18:19 |
| *** Marga_ has quit IRC | 18:20 | |
| *** penick has joined #openstack-nova | 18:21 | |
| *** bkopilov has joined #openstack-nova | 18:21 | |
| *** achanda has joined #openstack-nova | 18:23 | |
| *** achanda has quit IRC | 18:23 | |
| *** achanda has joined #openstack-nova | 18:24 | |
| sdague | mriedem: there might be an assumed order issue | 18:25 |
| *** afazekas has quit IRC | 18:26 | |
| *** annashen has joined #openstack-nova | 18:30 | |
| *** Longgeek has quit IRC | 18:31 | |
| sdague | I wonder how it got pass upstream testing | 18:31 |
| *** htruta has joined #openstack-nova | 18:35 | |
| *** dims has joined #openstack-nova | 18:35 | |
| *** nellysmitt has joined #openstack-nova | 18:35 | |
| *** claudiub has joined #openstack-nova | 18:36 | |
| *** Sukhdev has joined #openstack-nova | 18:38 | |
| *** mtanino has quit IRC | 18:41 | |
| *** zul has quit IRC | 18:41 | |
| openstackgerrit | Sean Dague proposed openstack/nova: Deprecate Nova in tree EC2 APIs https://review.openstack.org/150929 | 18:41 |
| *** pkoniszewski has joined #openstack-nova | 18:43 | |
| *** zul has joined #openstack-nova | 18:43 | |
| *** dprince has joined #openstack-nova | 18:46 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: fix VM rescue problem with VNC console https://review.openstack.org/113908 | 18:46 |
| *** claudiub has quit IRC | 18:47 | |
| *** sbfox has quit IRC | 18:49 | |
| *** vmtrooper has joined #openstack-nova | 18:50 | |
| *** Marga_ has joined #openstack-nova | 18:50 | |
| *** jcoufal has quit IRC | 18:54 | |
| *** Nikolay_St has quit IRC | 18:55 | |
| *** vmtrooper has quit IRC | 18:55 | |
| *** thedodd has joined #openstack-nova | 18:55 | |
| mriedem | sdague: yeah no idea | 18:56 |
| mriedem | but it's been in since yesterday | 18:56 |
| mriedem | constantly fails for me locally though, i've pulled, rebuilt tox, etc | 18:56 |
| sdague | yeh, it's test ordering, otp now, and I'll poke later | 18:56 |
| *** thangp has quit IRC | 18:58 | |
| *** Nikolay_St has joined #openstack-nova | 19:00 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Revert "Make unit tests inherit from test.NoDBTestCase" https://review.openstack.org/163931 | 19:00 |
| mriedem | sdague: ^ | 19:00 |
| *** jecarey has quit IRC | 19:02 | |
| *** vadorsky has joined #openstack-nova | 19:04 | |
| *** vadorsky has quit IRC | 19:04 | |
| *** jecarey has joined #openstack-nova | 19:04 | |
| *** vadorsky has joined #openstack-nova | 19:05 | |
| *** Nikolay_St has quit IRC | 19:06 | |
| *** ijw has joined #openstack-nova | 19:06 | |
| *** takedakn has quit IRC | 19:10 | |
| *** Marga_ has quit IRC | 19:11 | |
| *** ijw has quit IRC | 19:11 | |
| *** garyk has joined #openstack-nova | 19:12 | |
| *** Nikolay_St has joined #openstack-nova | 19:13 | |
| mriedem | jogo: jaypipes: ^ too | 19:14 |
| jaypipes | mriedem: :( | 19:15 |
| mriedem | jaypipes: i might be crazy, but if you run that one test by itself it should fail | 19:15 |
| *** bwensley has joined #openstack-nova | 19:15 | |
| sdague | mriedem: I might have a narrower fix later today if you can wait | 19:16 |
| jaypipes | mriedem: and why doesn't it fail when run together? :( | 19:16 |
| mriedem | jaypipes: locally if i run all of test_neutronv2 it fails | 19:16 |
| mriedem | or just the test | 19:16 |
| mriedem | i don't know what's different about my local than the gate | 19:16 |
| sdague | I do... otp :) | 19:16 |
| jaypipes | otp? | 19:18 |
| mriedem | on the phone | 19:18 |
| jaypipes | oh | 19:18 |
| mriedem | != odb | 19:18 |
| openstack | mriedem: Error: "=" is not a valid command. | 19:18 |
| jaypipes | lol | 19:19 |
| *** Nikolay_St has quit IRC | 19:19 | |
| mriedem | you got that one :) | 19:19 |
| *** dulek has quit IRC | 19:19 | |
| *** rushiagr is now known as rushiagr_away | 19:20 | |
| *** Nikolay_St has joined #openstack-nova | 19:25 | |
| *** tjones1 has quit IRC | 19:25 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: update test_vm_util to use instance object https://review.openstack.org/160273 | 19:26 |
| openstackgerrit | garyk proposed openstack/nova: VMware: use instance UUID as instance name https://review.openstack.org/160796 | 19:26 |
| *** claudiub has joined #openstack-nova | 19:26 | |
| *** bkopilov has quit IRC | 19:27 | |
| openstackgerrit | Tushar Patil proposed openstack/nova: Save connection_info before volume becomes in-use https://review.openstack.org/163937 | 19:28 |
| *** bkopilov has joined #openstack-nova | 19:32 | |
| *** tjones1 has joined #openstack-nova | 19:34 | |
| *** iamjarvo has quit IRC | 19:37 | |
| *** sbfox has joined #openstack-nova | 19:38 | |
| *** openstackgerrit has quit IRC | 19:40 | |
| *** openstackgerrit has joined #openstack-nova | 19:40 | |
| *** iamjarvo has joined #openstack-nova | 19:42 | |
| mriedem | is there an easy way to unset a specific stub with mox without unstubbing everything? | 19:44 |
| mriedem | i.e. i don't want to call self.mox.UnsetStubs() | 19:45 |
| *** Nikolay_St has quit IRC | 19:45 | |
| *** Marga_ has joined #openstack-nova | 19:45 | |
| *** Nikolay_St has joined #openstack-nova | 19:46 | |
| *** angdraug has joined #openstack-nova | 19:48 | |
| *** oro_ has joined #openstack-nova | 19:49 | |
| *** oro has joined #openstack-nova | 19:49 | |
| openstackgerrit | Chris Friesen proposed openstack/nova: Fix "down" nova-compute service spuriously marked as "up" https://review.openstack.org/163060 | 19:49 |
| *** dougwig has joined #openstack-nova | 19:50 | |
| mriedem | nova meeting in 10 minutes | 19:50 |
| *** Marga_ has quit IRC | 19:50 | |
| melwitt | mriedem: I think it's one hour 10 from now because of daylight savings | 19:51 |
| mriedem | gdi | 19:51 |
| melwitt | lol sorry | 19:51 |
| dougwig | hello nova. devstack's check queue recently added a grenade job including neutron, and the question was raised as to whether it still needs to be doing grenade with nova-net. i didn't see any special case upgrade code in grenade for nova-net, but wanted to check here first before removing the grenade/nova-net job from devstack ? | 19:51 |
| mriedem | dougwig: grenade with both is good imo | 19:52 |
| dougwig | mriedem: can you chime in on here, for posterity? https://review.openstack.org/#/c/160914/ | 19:52 |
| dougwig | if i don't close the loop, anteaya will hunt me down, i'm sure. :) | 19:53 |
| mriedem | uh oh | 19:53 |
| *** gokrokve_ has quit IRC | 19:53 | |
| mriedem | well, question is "Is the upgrade code path any different if nova-net is used instead?" | 19:53 |
| mriedem | i would think yes? | 19:54 |
| mriedem | if we screw up the nova-net rpc api | 19:54 |
| mriedem | and break upgrades | 19:54 |
| *** gokrokve has joined #openstack-nova | 19:54 | |
| jogo | mriedem: can you revert a subset | 19:54 |
| *** marun has quit IRC | 19:54 | |
| mriedem | jogo: oh probably | 19:55 |
| mriedem | i'll have to run a full tox locally first | 19:55 |
| mriedem | to see what else was missed | 19:55 |
| jogo | thanks | 19:55 |
| *** gokrokve_ has joined #openstack-nova | 19:55 | |
| *** dprince has quit IRC | 19:56 | |
| *** bwensley has quit IRC | 19:56 | |
| openstackgerrit | garyk proposed openstack/nova: Neutron: configure requested amount of fixed_ips https://review.openstack.org/163864 | 19:57 |
| bauzas | mriedem: isn't Nova meeting at 2200 UTC ? | 19:58 |
| mriedem | bauzas: yeah i was wrong | 19:58 |
| mriedem | just adjusted my calendar | 19:58 |
| *** bwensley_ has joined #openstack-nova | 19:58 | |
| bauzas | so 2 weeks before having the meeting at midnight for me :) | 19:58 |
| *** jecarey has quit IRC | 19:59 | |
| *** gokrokve has quit IRC | 19:59 | |
| dougwig | mriedem: thanks | 20:00 |
| *** melwitt_ has joined #openstack-nova | 20:02 | |
| *** exploreshaifali has quit IRC | 20:02 | |
| *** mikal_ is now known as mikal | 20:02 | |
| *** melwitt has quit IRC | 20:02 | |
| *** melwitt_ is now known as melwitt | 20:02 | |
| *** garyk has quit IRC | 20:03 | |
| *** setuid has quit IRC | 20:03 | |
| *** baoli has joined #openstack-nova | 20:03 | |
| *** jecarey has joined #openstack-nova | 20:04 | |
| *** sbfox has quit IRC | 20:04 | |
| *** jecarey_ has joined #openstack-nova | 20:04 | |
| *** baoli has quit IRC | 20:04 | |
| *** bwensley_ has quit IRC | 20:05 | |
| *** baoli has joined #openstack-nova | 20:05 | |
| *** exploreshaifali has joined #openstack-nova | 20:05 | |
| *** setuid has joined #openstack-nova | 20:05 | |
| anteaya | bauzas: 2100 utc: https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting | 20:07 |
| bauzas | anteaya: awesome | 20:07 |
| *** jgrimm is now known as zz_jgrimm | 20:07 | |
| bauzas | anteaya: my TB agenda was having 2200 UTC that's much better | 20:08 |
| anteaya | TB? | 20:08 |
| anteaya | tuberculosis | 20:08 |
| anteaya | I hope that is not on your agenda | 20:08 |
| bauzas | eh :) | 20:08 |
| *** jecarey has quit IRC | 20:09 | |
| bauzas | dansmith: around ? | 20:10 |
| *** matrohon has joined #openstack-nova | 20:10 | |
| bauzas | dansmith: I was about to rework on https://review.openstack.org/#/c/163867/ | 20:10 |
| bauzas | dansmith: I need basically to change the target_version to the corresponding version for the compute_node | 20:11 |
| *** sbfox has joined #openstack-nova | 20:11 | |
| bauzas | dansmith: so I could look at the obj_relationships attribute and find the right ComputeNode version compatible with the Service target_version, but was wondering if there was any helper method for this in the base module | 20:12 |
| *** otter768 has joined #openstack-nova | 20:12 | |
| bauzas | ie. 'give me a version and a nested object, and I'll give you the corresponding version for the nested attribute" | 20:12 |
| jaypipes | dansmith, bauzas: https://review.openstack.org/#/c/137817/ could use your review. thx in advance. | 20:13 |
| dansmith | bauzas: back in a bit | 20:13 |
| bauzas | dansmith: np | 20:13 |
| mriedem | jaypipes: is that part of a blueprint? | 20:13 |
| jaypipes | mriedem: kilo-objects? | 20:13 |
| bauzas | mriedem: long story, do you want to hear about it ? :) | 20:13 |
| mriedem | no | 20:14 |
| *** zz_jgrimm is now known as jgrimm | 20:14 | |
| *** annashen has quit IRC | 20:15 | |
| *** Longgeek has joined #openstack-nova | 20:16 | |
| *** otter768 has quit IRC | 20:17 | |
| openstackgerrit | Matthew Booth proposed openstack/nova: db: Don't compare datetime.datetime at > 1 second resolution https://review.openstack.org/156298 | 20:18 |
| openstackgerrit | Matthew Booth proposed openstack/nova: Allow TimeOverride to take a specific time https://review.openstack.org/158796 | 20:18 |
| *** dave-mccowan has quit IRC | 20:19 | |
| openstackgerrit | Matthew Booth proposed openstack/nova: Fix ArchiveTestCase on MySQL due to differing exceptions https://review.openstack.org/156674 | 20:19 |
| *** nellysmitt has quit IRC | 20:22 | |
| *** cbader has quit IRC | 20:22 | |
| *** Longgeek has quit IRC | 20:23 | |
| dansmith | bauzas: there isn't a helper, but it seems like the thing you're doing should also be good enough, so I want to think about how to do that better first | 20:25 |
| dansmith | bauzas: also, I'm almost good on my patch to add the test, so want your feedback on that once I have it working | 20:25 |
| bauzas | dansmith: sure | 20:25 |
| bauzas | dansmith: yey, the problem I think is how to make sure that if we backport a primitive to an older version, it will also backport the nested object primitive to a corresponding ver | 20:26 |
| bauzas | dansmith: and my findings don't show me it's currently possible | 20:26 |
| mriedem | dims: do we need this warning to show up as much as it does? http://goo.gl/rF4fyL | 20:26 |
| dansmith | well, not just that, but the base class just overwrites the primitive, if it thinks it should do the backport | 20:27 |
| *** cbader has joined #openstack-nova | 20:27 | |
| mriedem | dims: https://bugs.launchpad.net/nova/+bug/1431549 | 20:28 |
| openstack | Launchpad bug 1431549 in OpenStack Compute (nova) ""Arguments dropped when creating context" warnings are spamming logs" [Undecided,New] | 20:28 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add a test that validates object backports and child object versions https://review.openstack.org/163887 | 20:28 |
| dansmith | bauzas: I think that will pass now ^ | 20:28 |
| *** Mike_D_laptop has joined #openstack-nova | 20:28 | |
| bauzas | dansmith: ok I see the series, reviewing them | 20:29 |
| openstackgerrit | Matthew Booth proposed openstack/nova: Fix ArchiveTestCase on MySQL due to MySQL read view https://review.openstack.org/156675 | 20:29 |
| sdague | mriedem: so... the reason the tests work in bulk but not in specific is the DatabaseFixture doesn't actually kill all the db stuff on cleanup, it just does the reset | 20:29 |
| sdague | so you might be using someone else's previously setup database | 20:30 |
| dims | mriedem: the LOG.warn was already there - https://github.com/openstack/nova/blob/stable/juno/nova/context.py#L64 | 20:30 |
| sdague | I'll work up a fix for that, and sort out which tests need to rollback | 20:30 |
| dims | mriedem: y i can throttle it | 20:30 |
| mriedem | sdague: ok, i'm nearly done with a py27 run with --concurrency=1 | 20:30 |
| mriedem | once that's done, i'll just revert the failing test classes | 20:31 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Move InstanceClaimTestCase.test_claim_and_audit https://review.openstack.org/136054 | 20:31 |
| sdague | so... I'm surprised that would expose it | 20:31 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Update some ResizeClaimTestCase tests https://review.openstack.org/138686 | 20:31 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Use migration object in resource_tracker https://review.openstack.org/79324 | 20:31 |
| sdague | yeh, now that I did this I've got 48 test fails | 20:31 |
| sdague | sorry, 46 | 20:32 |
| mriedem | sdague: --concurrency=1 didn't expose it, but it did expose some quota races :) http://paste.openstack.org/show/191932/ | 20:32 |
| mriedem | jogo: ^ | 20:32 |
| sdague | mriedem: nice, so honestly, it's probably better if I do this fix instead of that. I'll fix up these 46 tests and this should prevent this kind of issue in the future | 20:33 |
| mriedem | i'll get a bug open for that and a fix up | 20:33 |
| mriedem | sdague: 'that' being the targeted revert? | 20:33 |
| sdague | yeh | 20:33 |
| mriedem | sure | 20:33 |
| sdague | it's probably more than the revert actually | 20:33 |
| jaypipes | sdague: thx Sean. | 20:33 |
| sdague | mriedem: that's not actually a quota race, that's a db getting reused unexpectedly I think | 20:34 |
| jogo | what sdague said | 20:35 |
| mriedem | yeah | 20:35 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1431551 | 20:35 |
| openstack | Launchpad bug 1431551 in OpenStack Compute (nova) "nova.tests.unit.api.openstack.compute.contrib.test_block_device_mapping_v1 hits OverQuota with --concurrency=1" [Undecided,New] | 20:35 |
| mriedem | regardless | 20:35 |
| mriedem | busted | 20:35 |
| *** Marga_ has joined #openstack-nova | 20:35 | |
| mdbooth | mriedem: --concurrency 1 is no concurrency, right? | 20:36 |
| mriedem | sdague: once you have a thing up for https://review.openstack.org/#/c/163931/ i'll drop it | 20:36 |
| mriedem | mdbooth: yes | 20:36 |
| mdbooth | mriedem: Does that often trigger different bugs? Just curious. | 20:36 |
| mriedem | mdbooth: it's exposed some weirdness in tests before | 20:37 |
| mriedem | that have eventually found their way into the gate | 20:37 |
| *** Marga_ has quit IRC | 20:37 | |
| *** Marga_ has joined #openstack-nova | 20:38 | |
| mriedem | dims: huh, i've never noticed that in the amount it logs now | 20:38 |
| *** Sukhdev has quit IRC | 20:38 | |
| *** exploreshaifali has quit IRC | 20:39 | |
| dims | mriedem: ack, will figure out a way to reduce that logging | 20:39 |
| *** vmtrooper has joined #openstack-nova | 20:39 | |
| dims | rather than just yank it | 20:39 |
| *** Mike_D_laptop has quit IRC | 20:39 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: objects: introduce numa topology limits objects https://review.openstack.org/137817 | 20:40 |
| *** dave-mccowan has joined #openstack-nova | 20:40 | |
| jaypipes | mriedem: I just added the Blueprint: kilo-objects line.... ^ | 20:40 |
| *** neelashah1 has quit IRC | 20:42 | |
| *** vladikr has quit IRC | 20:42 | |
| *** mrda-away is now known as mrda | 20:42 | |
| mriedem | https://www.youtube.com/watch?v=5QGb821oAZs&t=74 | 20:43 |
| sdague | mriedem: yeh, should be about another 20 minutes and I'll have the patch | 20:43 |
| sdague | mriedem: are there other bugs registered except 1421551? | 20:44 |
| mriedem | sdague: 1431519 | 20:44 |
| *** vmtrooper has quit IRC | 20:44 | |
| mriedem | 1431551 is the bdm thing | 20:44 |
| *** tonytan4ever has quit IRC | 20:44 | |
| sdague | right, but it's actually the same bug | 20:44 |
| mriedem | alright, we can dupe | 20:45 |
| sdague | or I can close them both with this patch | 20:45 |
| mriedem | that works too | 20:45 |
| dansmith | bauzas: ah, this doesn't catch your case because service.host is not set, and we swallow all errors in do_compute_node() :( | 20:47 |
| openstackgerrit | Chris Friesen proposed openstack/nova: Fix "down" nova-compute service spuriously marked as "up" https://review.openstack.org/163060 | 20:49 |
| *** thedodd has quit IRC | 20:50 | |
| mikal | Morning | 20:50 |
| bauzas | dansmith: argh | 20:50 |
| *** takedakn has joined #openstack-nova | 20:51 | |
| bauzas | dansmith: I'm taking time to review your patch because I need to understand how you metaclass the NovaObject classes | 20:51 |
| dansmith | bauzas: that is all going away in o.vo | 20:51 |
| dansmith | bauzas: I don't like that stuff | 20:51 |
| mriedem | dansmith: is there a patch in the works that fixes this? http://logs.openstack.org/90/163890/1/check/check-tempest-dsvm-cells/896e7e0/logs/screen-n-cell-child.txt.gz?level=TRACE#_2015-03-12_17_44_00_202 | 20:51 |
| *** annashen has joined #openstack-nova | 20:51 | |
| dansmith | mriedem: not from me | 20:52 |
| *** thedodd has joined #openstack-nova | 20:53 | |
| mriedem | k | 20:54 |
| mriedem | i'll take a look | 20:54 |
| *** thedodd has quit IRC | 20:54 | |
| *** marun has joined #openstack-nova | 20:54 | |
| mriedem | since bauzas loves objects so much he could also learn to love cells :) | 20:54 |
| bauzas | mriedem: I had a good teacher with alaski | 20:54 |
| *** penick has quit IRC | 20:55 | |
| bauzas | mriedem: at least now I can understand the differences between all modules :D | 20:55 |
| bauzas | mriedem: but I'm only yellow belt | 20:55 |
| bauzas | mriedem: checking | 20:55 |
| mdbooth | dansmith: Do you remember I submitted a patch to remove that metaclass stuff a few months ago? You killed it because it was an integral part of the design :) | 20:57 |
| mriedem | nova meeting in 3 min | 20:57 |
| mikal | True dat | 20:57 |
| mriedem | bauzas: alaski: dansmith: https://bugs.launchpad.net/nova/+bug/1431562 | 20:57 |
| openstack | Launchpad bug 1431562 in OpenStack Compute (nova) "cells: Error processing message locally: Object action save failed because: Calling remotables with context is deprecated" [Undecided,New] | 20:57 |
| dansmith | mdbooth: eh? | 20:58 |
| *** pixelbeat has quit IRC | 20:58 | |
| *** raildo has quit IRC | 20:58 | |
| dansmith | mdbooth: it is an integral part of the design, in that it serves a purpose that would need a replacement | 20:58 |
| *** dboik_ has joined #openstack-nova | 20:59 | |
| *** sbfox has quit IRC | 20:59 | |
| mdbooth | How do you get gerrit to give you a list of abandoned patches? | 21:00 |
| openstackgerrit | Sean Dague proposed openstack/nova: ensure DatabaseFixture removes db on cleanup https://review.openstack.org/163970 | 21:00 |
| *** cbader has quit IRC | 21:01 | |
| *** dboik has quit IRC | 21:02 | |
| *** harlowja_ has quit IRC | 21:03 | |
| mriedem | change status:open to status:abandoned | 21:03 |
| *** harlowja has joined #openstack-nova | 21:04 | |
| sdague | mriedem: so that passes all the unit tests, running functional on it now, though I think those all require the db anyway | 21:04 |
| *** penick has joined #openstack-nova | 21:04 | |
| *** penick has quit IRC | 21:04 | |
| alaski | mriedem: thanks, looks like dansmiths wedge caught something | 21:04 |
| mriedem | sdague: yeah, the change didn't touch functional | 21:04 |
| mriedem | so i didn't test it | 21:04 |
| *** rfolco has quit IRC | 21:04 | |
| mriedem | alaski: running tox on a change locally now | 21:05 |
| mriedem | alaski: will push that up when done | 21:05 |
| *** moshele has quit IRC | 21:05 | |
| *** patrickeast has quit IRC | 21:05 | |
| mriedem | s/wedge/dragnet/ | 21:05 |
| zzzeek | hey mriedem : was wondering if this issue https://bugs.launchpad.net/nova/+bug/1183523 is the one im seeing where test_archive_deleted_rows_2_tables fails with testtools.matchers._impl.MismatchError: 8 != 5 ? | 21:05 |
| openstack | Launchpad bug 1183523 in OpenStack Compute (nova) "db-archiving fails to clear some deleted rows from instances table" [Medium,Confirmed] | 21:05 |
| *** patrickeast_ has joined #openstack-nova | 21:05 | |
| *** ybathia has quit IRC | 21:05 | |
| zzzeek | mriedem: im getting this failure in tests a lot on my CI server | 21:05 |
| bauzas | alaski: dansmith: sounds like it just needs a slight patch to remove passing the context, that's it | 21:06 |
| mriedem | bauzas: that's what i'm testing | 21:06 |
| dansmith | bauzas: yeah | 21:06 |
| bauzas | mriedem: I don't see anything really related to cells and messaging is just checking the child DB | 21:06 |
| bauzas | mriedem: nothing proxied to cells api | 21:07 |
| alaski | bauzas: looks like line 838 in cells/messaging.py | 21:07 |
| bauzas | alaski: yeah that line | 21:07 |
| bauzas | alaski: just removing that arg should fix the error | 21:08 |
| alaski | yep | 21:08 |
| bauzas | anyway, nova meeting dudes | 21:08 |
| mriedem | alaski: bauzas: yeah...like i said, i have the patch :) | 21:08 |
| mriedem | but tox | 21:08 |
| bauzas | oh oh | 21:08 |
| * bauzas should really create a cells devstack | 21:08 | |
| sdague | mriedem: functional tests all pass locally | 21:09 |
| mriedem | sdague: cool, expected them too | 21:09 |
| mriedem | zzzeek: not sure, nova meeting atm | 21:09 |
| *** aysyd has quit IRC | 21:09 | |
| zzzeek | mriedem: okey | 21:09 |
| mriedem | zzzeek: btw, is it a thing that sqlalchemy < 1.0 doesn't return fkey names for sqlite? | 21:10 |
| mriedem | when doing table.constraints? | 21:10 |
| zzzeek | mriedem: yes | 21:10 |
| mriedem | bauzas: ^ | 21:10 |
| mriedem | there you go | 21:10 |
| bauzas | okay | 21:11 |
| bauzas | mriedem: I saw your comments, ty for them | 21:11 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add CellMapping object https://review.openstack.org/161867 | 21:11 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Objects: use setattr rather than dict syntax in remotable https://review.openstack.org/163974 | 21:11 |
| bauzas | mriedem: was about to do a new upload by today or tomorrow | 21:11 |
| bauzas | trying to hit that before FF | 21:11 |
| mriedem | bauzas: yeah should make it | 21:12 |
| *** romainh has quit IRC | 21:13 | |
| jamielennox | jogo: so https://review.openstack.org/#/c/136931/ had to be rebased | 21:14 |
| jamielennox | jogo: who else would know this stuff i can hit for the other +2 | 21:14 |
| jamielennox | because i've been nursing this one since kilo opened | 21:14 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: cells: don't pass context to instance.save in instance_update_from_api https://review.openstack.org/163978 | 21:14 |
| mriedem | bauzas: alaski: dansmith: ^ | 21:14 |
| *** thedodd has joined #openstack-nova | 21:14 | |
| jogo | jamielennox: not sure but just gave it a +2 again | 21:14 |
| alaski | mriedem: thanks | 21:16 |
| *** romainh has joined #openstack-nova | 21:16 | |
| mriedem | alaski: nice thing the cells job was around to show the failures :) | 21:17 |
| jamielennox | dansmith or sdague: can you have a look at https://review.openstack.org/#/c/136931/ when you get a chance, it's been sitting for a while with one +2 | 21:17 |
| alaski | mriedem: heh, woo | 21:17 |
| *** Marga_ has quit IRC | 21:20 | |
| *** neelashah has joined #openstack-nova | 21:22 | |
| jaypipes | edleafe: no, he was saying to always send an InstanceList object, and for instance_update_info() for a single instance notification, the InstanceList would just contain a single Instance object.' | 21:23 |
| dims | sdague: there's a quota one that jogo dug up yday and i revved up the patch - https://review.openstack.org/#/c/163668/ | 21:23 |
| dims | (with one +2_ | 21:23 |
| dims | (with one +2) | 21:23 |
| *** Marga_ has joined #openstack-nova | 21:24 | |
| edleafe | jaypipes: ah. Ok, that is doable | 21:24 |
| jaypipes | edleafe: yeah, I agree. | 21:25 |
| jogo | jaypipes: we have the nova meeting now | 21:26 |
| jogo | err jamielennox * | 21:26 |
| jogo | so if you want to chime in | 21:26 |
| *** annashen has quit IRC | 21:26 | |
| edleafe | jaypipes: just responded to alaski on gerrit | 21:26 |
| *** exploreshaifali has joined #openstack-nova | 21:27 | |
| edleafe | jaypipes: heh, just noticed that bauzas | 21:28 |
| edleafe | ugh | 21:28 |
| bauzas | edleafe: mmm ? | 21:28 |
| edleafe | jaypipes: heh, just noticed that bauzas' 'update_aggregates' sends two types | 21:28 |
| alaski | edleafe: cool. the naming is odd to me but not that important, if you change to always using instancelist I'm +2 | 21:28 |
| bauzas | edleafe: oh I see your point | 21:29 |
| bauzas | edleafe: yeah I actually accepts either a list or a single item | 21:29 |
| sdague | dims: I left a comment, https://review.openstack.org/#/c/163668 seems wrong | 21:29 |
| bauzas | edleafe: I thought it was payload-saving | 21:30 |
| bauzas | edleafe: I mean, in case we would need to update 2 aggregates, we could send the updates once | 21:30 |
| edleafe | bauzas: me too, but I see alaski's point | 21:30 |
| bauzas | edleafe: which I don't know, link please ? :) | 21:30 |
| edleafe | bauzas: of course. But when you send 1 update, send a 1-item AggregateList | 21:30 |
| edleafe | bauzas: https://review.openstack.org/#/c/160507 | 21:31 |
| bauzas | edleafe: which is a valid point | 21:34 |
| dims | sdague: so here's the situation, if quota_set.items() had 5 items and the 4th item had an issue, we would create_limit for the first 3 and then bail out which is wrong, now with 2 for loops for quota_set.items(), only if the first loop finishes fully do we call any create_limit() | 21:34 |
| bauzas | edleafe: well, we can leave for now for aggregates and I'll change it later on | 21:35 |
| bauzas | edleafe: once your code will be merged | 21:35 |
| sdague | dims: ... but _validate_quota_limit raises an exeception | 21:35 |
| sdague | where are you catching that? | 21:35 |
| edleafe | bauzas: the one benefit I could see is if you got a list instead of a single instance, you could interpret that as a 'fresh' list, and discard stale info | 21:36 |
| dims | sdague: the exception goes back on the wire | 21:36 |
| sdague | dims: and stops processing the for loop | 21:36 |
| dims | yes | 21:36 |
| sdague | isn't that actually the bug | 21:36 |
| sdague | you never get to the next section of code | 21:36 |
| dims | exactly what we want to do | 21:36 |
| sdague | ok... so I at least want a code comment there, because the intent is not clear unless you do a bunch more digging | 21:37 |
| dims | ack will do | 21:37 |
| sdague | especially because we have 2 for loops checking validity | 21:38 |
| sdague | so this will end up a 3 for loop method, which is weird | 21:38 |
| *** vadorsky has quit IRC | 21:38 | |
| *** Marga_ has quit IRC | 21:38 | |
| *** Marga_ has joined #openstack-nova | 21:39 | |
| *** kaufer has quit IRC | 21:42 | |
| *** dboik_ has quit IRC | 21:43 | |
| *** dboik has joined #openstack-nova | 21:44 | |
| *** Sukhdev has joined #openstack-nova | 21:44 | |
| *** lpetrut has quit IRC | 21:44 | |
| *** sbfox has joined #openstack-nova | 21:46 | |
| openstackgerrit | Sean Dague proposed openstack/nova: Deprecate Nova in tree EC2 APIs https://review.openstack.org/150929 | 21:48 |
| *** amotoki has quit IRC | 21:49 | |
| sdague | ok... hopefully that passes all the tests | 21:50 |
| openstackgerrit | Michael Bayer proposed openstack/nova: Repair ArchiveTestCase to pass in all cases https://review.openstack.org/163990 | 21:52 |
| mikal | Ok, so important question | 21:53 |
| mikal | Who first thought of canning worms | 21:53 |
| mikal | And why? | 21:53 |
| zzzeek | im putting just random sets of you on my patch for review | 21:53 |
| mikal | Is it somehow the american dream to be able to can all of the animals in your back yard? | 21:53 |
| zzzeek | i dont care how you fix it but I want my own CI to pass | 21:53 |
| mikal | Do you perhaps can small birds as well? | 21:53 |
| mriedem | mikal: https://www.google.com/webhp?sourceid=chrome-instant&rlz=1C1OPRB_enUS530US530&ion=1&espv=2&ie=UTF-8#safe=off&q=can+of+worms+origin | 21:53 |
| anteaya | mikal: like for bait? | 21:54 |
| zzzeek | b.c. that’s how I make sure SQLAlchemy isn’t breaking | 21:54 |
| mriedem | yes | 21:54 |
| anteaya | mikal: people who lived near lakes who had tourists come up for teh summer | 21:54 |
| anteaya | who knew nothing about fishing but wanted to say they did it | 21:54 |
| mikal | I thought Americans fished with sticks of dynamite? | 21:54 |
| * anteaya lives near a lake and tourists come up for the summer | 21:54 | |
| sdague | yeh, we do them in styrofoam though. They do sound more awesome as "night crawlers" | 21:54 |
| mikal | I'm sure the discovery channel assured me that was common | 21:54 |
| mriedem | cans were in the 50s | 21:55 |
| mriedem | according to google | 21:55 |
| mriedem | up until the 50s | 21:55 |
| sdague | oh the 50s, when tin cans were the signal of modernity | 21:55 |
| dansmith | mikal: heh | 21:55 |
| bauzas | can't we use Pandora box ? | 21:55 |
| bauzas | instead of can of worms ? | 21:56 |
| *** annegentle has quit IRC | 21:56 | |
| sdague | dansmith: 3 other nova cores beat you to the +2 | 21:56 |
| *** mriedem is now known as mriedem_away | 21:56 | |
| dansmith | sdague: what +2? | 21:56 |
| mriedem_away | ec2 | 21:56 |
| sdague | on the ec2 patch :) | 21:56 |
| dansmith | fsck that, push another rev! | 21:56 |
| sdague | well, if I screwed up the tests again, I will tomorrow | 21:56 |
| dansmith | sdague: okay, there's a $20 in it for you if I get advance notice | 21:57 |
| dims | haha | 21:57 |
| sdague | ok, end of day for me. https://review.openstack.org/#/c/163970/ should fix mriedem_away's test issues - https://review.openstack.org/#/c/163970/ so if people want to land that, it would be cool | 21:57 |
| *** neelashah has quit IRC | 21:59 | |
| *** asettle has quit IRC | 22:00 | |
| *** melwitt_ has joined #openstack-nova | 22:01 | |
| *** melwitt_ has quit IRC | 22:02 | |
| *** harlowja has quit IRC | 22:02 | |
| *** melwitt has quit IRC | 22:03 | |
| *** ybathia has joined #openstack-nova | 22:04 | |
| *** apuimedo has joined #openstack-nova | 22:04 | |
| *** harlowja has joined #openstack-nova | 22:04 | |
| *** Longgeek has joined #openstack-nova | 22:08 | |
| *** pcrews has joined #openstack-nova | 22:09 | |
| pcrews | jogo: howdy | 22:10 |
| *** pkoniszewski has quit IRC | 22:11 | |
| jogo | pcrews: so trying to reproduce the quota issues | 22:11 |
| jogo | haven't gotten around to running your script yet, but it sounds like this is related to RPC timeouts and things like that mostly | 22:11 |
| jogo | pcrews: does that align with your understanding? | 22:12 |
| pcrews | jogo: yeah, I have been running some tests myself, with no luck. but that jibes with what I know | 22:12 |
| jogo | I am trying to restart n-* services during the test | 22:13 |
| *** otter768 has joined #openstack-nova | 22:13 | |
| pcrews | have managed to crash nova (yay?), but haven't seen quotas go awry yet. Have been trying create/delete/list_server*/reboot combos with multiple concurrent clients. | 22:13 |
| jogo | I am trying create delete list as well | 22:14 |
| jogo | next I am thinking of working on a a way to make oslo.messaging drop messages on purpose | 22:14 |
| pcrews | I have a few more scenarios to try to stress things / try to hit the good timeouts, but nothing interesting yet. | 22:15 |
| pcrews | that would be pretty solid and useful for this kind of testing. | 22:15 |
| jogo | I just did it ! | 22:15 |
| pcrews | you triggered it? | 22:15 |
| jogo | pcrews: http://paste.openstack.org/show/191961/ | 22:15 |
| jogo | now to better isolate the failure case | 22:16 |
| jogo | http://paste.ubuntu.com/10587937/ + restarting services | 22:16 |
| *** eharney has quit IRC | 22:16 | |
| *** nelsnels_ has joined #openstack-nova | 22:17 | |
| *** annashen has joined #openstack-nova | 22:17 | |
| *** otter768 has quit IRC | 22:18 | |
| pcrews | jogo: that is very cool :) | 22:18 |
| jogo | pcrews: you remember how to reset the quotas? | 22:18 |
| *** takedakn has quit IRC | 22:18 | |
| pcrews | jogo: Alas, I do not | 22:19 |
| *** nelsnelson has quit IRC | 22:19 | |
| pcrews | https://ask.openstack.org/en/question/494/how-to-reset-incorrect-quota-count/ \o/ | 22:19 |
| *** dkliban is now known as dkliban_afk | 22:23 | |
| mtreinish | pcrews: there isn't a nova manage command for that? | 22:24 |
| jogo | mtreinish: not yet | 22:25 |
| *** burt has quit IRC | 22:26 | |
| bauzas | dansmith: so, back to https://review.openstack.org/#/c/163887/2/nova/tests/unit/objects/test_objects.py,cm you got my blessing | 22:26 |
| pcrews | quota-delete Delete quota for a tenant/user so their quota will Revert back to default. | 22:26 |
| bauzas | dansmith: that just shows us that it's broken for my own patch | 22:26 |
| * pcrews just found that, but has never tested it. | 22:27 | |
| bauzas | dansmith: because you just doublecheck that the child field would be backported to an expected previous version | 22:27 |
| *** jecarey_ has quit IRC | 22:28 | |
| *** vmtrooper has joined #openstack-nova | 22:28 | |
| *** patrickeast_ has quit IRC | 22:29 | |
| dansmith | bauzas: well, right, the hard thing to do generally is know that for that case we need to provide a service.host and mock out computenodelist.get...() :( | 22:30 |
| dansmith | bauzas: so this will catch some cases, but we just need good specific testing on yours I think :/ | 22:30 |
| *** annashen has quit IRC | 22:31 | |
| bauzas | dansmith: agreed | 22:31 |
| dansmith | bauzas: so I guess we need to break out the logic of "what version of $child should I send | 22:32 |
| *** andreykurilin_ has joined #openstack-nova | 22:32 | |
| *** annashen has joined #openstack-nova | 22:32 | |
| bauzas | dansmith: you mean in service.obj_make_compatible() ? | 22:32 |
| bauzas | disclaimer: 11.30pm here | 22:32 |
| *** Longgeek has quit IRC | 22:32 | |
| dansmith | bauzas: yeah | 22:33 |
| dansmith | bauzas: go to sleep, we can do that tomorrow :) | 22:33 |
| *** vmtrooper has quit IRC | 22:33 | |
| bauzas | dansmith: that's my original thought, I should try to reproduce your _find_version_mapping helper | 22:34 |
| dansmith | bauzas: well, we just break it out of the base class into a helper I think | 22:35 |
| jogo | pcrews: nova or nova-manage | 22:35 |
| bauzas | dansmith: no worries, wifey is just watching a french equivalent of X-Factor, so I have plenty of time... | 22:35 |
| pcrews | jogo: nova, my bad | 22:35 |
| *** melwitt has joined #openstack-nova | 22:36 | |
| jogo | didn't work | 22:36 |
| *** Marga_ has quit IRC | 22:36 | |
| jogo | think I got a working nova-manage command | 22:38 |
| jogo | hmm | 22:38 |
| bauzas | dansmith: agreed | 22:38 |
| jogo | ohh my bad | 22:38 |
| bauzas | dansmith: for the moment, I'll just copy this into the service object, and we'll see if that makes the deal | 22:39 |
| *** oro_ has quit IRC | 22:39 | |
| jogo | pcrews: your call looks like it works | 22:39 |
| *** oro has quit IRC | 22:39 | |
| dansmith | bauzas: okay | 22:39 |
| *** Nic has quit IRC | 22:39 | |
| dansmith | bauzas: tomorrow I can propose a patch to break out the bits from NovaObject | 22:39 |
| *** thedodd has quit IRC | 22:39 | |
| pcrews | jogo: cool, that will come in handy for future-pcrews as well :) | 22:39 |
| bauzas | dansmith: if so, I would be tempted to add it in the base class and leave you rebase on top of my patch if you agree, because that's a critical bug :/ | 22:40 |
| jogo | pcrews: so I spoke to soon | 22:40 |
| jogo | it resets things to 0 | 22:40 |
| jogo | I have something to fix the quotas I think | 22:40 |
| *** annashen has quit IRC | 22:40 | |
| dansmith | bauzas: I don't think it's so critical that we can't take a day to fix it right the first time | 22:40 |
| jogo | pcrews: yeah that command is bad actually | 22:41 |
| jogo | hmm | 22:41 |
| pcrews | jogo: :( It sounded worthwhile on paper, but it seems par for the course that unborking things is never that easy. | 22:42 |
| jogo | pcrews: ok let me try again | 22:42 |
| jogo | pcrews: that command resets things to 0 | 22:42 |
| jogo | so quotas go to 0 | 22:42 |
| jogo | and after you boot an instance the quotas correct themselves | 22:42 |
| jogo | so lets say you have one instance running already | 22:42 |
| jogo | you do the delete | 22:42 |
| jogo | usage goes to 0 | 22:43 |
| jogo | do an action, quota correct themselves | 22:43 |
| bauzas | dansmith: agreed, provided that's merged before FF... :) | 22:43 |
| bauzas | erm, not FF | 22:43 |
| bauzas | RC1 | 22:43 |
| dansmith | yeah, no problem there I think | 22:44 |
| jogo | pcrews: does that make more sense | 22:45 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Deprecate V1.1, add basic stuff for microversions https://review.openstack.org/152569 | 22:45 |
| *** dims_ has joined #openstack-nova | 22:46 | |
| pcrews | jogo: it does, though that sounds like something that should maybe be better documented to prevent any 'wth just happened' moments. | 22:46 |
| *** dims has quit IRC | 22:46 | |
| *** dboik has quit IRC | 22:48 | |
| jogo | here is a better alternitive http://paste.ubuntu.com/10588043/ | 22:48 |
| *** dboik has joined #openstack-nova | 22:48 | |
| *** r-daneel has quit IRC | 22:48 | |
| *** amotoki has joined #openstack-nova | 22:49 | |
| jogo | pcrews: turns out its only a few lines of code | 22:50 |
| jogo | dansmith: ^ http://paste.ubuntu.com/10588043/ | 22:50 |
| jogo | dansmith: never mind your not on that spec https://review.openstack.org/#/c/161782/2 | 22:50 |
| *** nelsnels_ has quit IRC | 22:50 | |
| *** achanda has quit IRC | 22:52 | |
| *** achanda has joined #openstack-nova | 22:53 | |
| zzzeek | heya so, from waht I’m seeing here, the archive_deleted_rows_for_table() routine is going to DELETE rows from tables unconditionally, meaning, if you run it, its going to wipe your database basically. I feel like that can’t be correct? | 22:54 |
| jogo | zzzeek: it is supposed to delete them from the shadow tables AFAIK | 22:54 |
| openstackgerrit | Michael Bayer proposed openstack/nova: Ensure correct value for "deleted" is used https://review.openstack.org/164009 | 22:54 |
| zzzeek | jogo: hmm. um. | 22:54 |
| *** amotoki has quit IRC | 22:55 | |
| zzzeek | jogo: it does queries like this: | 22:55 |
| zzzeek | INSERT INTO shadow_quota_usages SELECT quota_usages.created_at, quota_usages.updated_at, quota_usages.deleted_at, quota_usages.id, quota_usages.project_id, quota_usages.resource, quota_usages.in_use, quota_usages.reserved, quota_usages.until_refresh, quota_usages.deleted, quota_usages.user_id | 22:55 |
| zzzeek | FROM quota_usages | 22:55 |
| zzzeek | WHERE quota_usages.deleted IS NOT NULL ORDER BY quota_usages.id | 22:55 |
| zzzeek | LIMIT ? OFFSET ? | 22:55 |
| zzzeek | 2015-03-12 17:01:01,218 INFO [sqlalchemy.engine.base.Engine] (7, 0) | 22:55 |
| jogo | pcrews: bbiab, but looks like we have some progress on this | 22:55 |
| zzzeek | 2015-03-12 17:01:01,219 INFO [sqlalchemy.engine.base.Engine] DELETE FROM quota_usages WHERE quota_usages.id in (SELECT T1.id FROM (SELECT quota_usages.id | 22:55 |
| zzzeek | FROM quota_usages | 22:55 |
| zzzeek | WHERE quota_usages.deleted IS NOT NULL ORDER BY quota_usages.id | 22:55 |
| zzzeek | LIMIT ? OFFSET ?) as T1) | 22:55 |
| zzzeek | that’s the opposite | 22:55 |
| zzzeek | FYI “IS NOT NULL”, is, all the rows | 22:55 |
| *** baoli has quit IRC | 22:55 | |
| *** thomasem has quit IRC | 22:55 | |
| pcrews | jogo: ack and ++. Nice find on triggering that bug | 22:56 |
| jogo | zzzeek: have to go AFK but, this corner of the code has caused a lot of issues so wouldn't be surprised if there were more | 22:56 |
| zzzeek | jogo: this woujdl be an entirely non-subtle issue so im not sure what im doing wrong | 22:56 |
| *** achanda_ has joined #openstack-nova | 22:56 | |
| *** annegentle has joined #openstack-nova | 22:56 | |
| zzzeek | jogo: i can defintiely see why the tests dont catch it though | 22:56 |
| *** achanda_ has quit IRC | 22:56 | |
| zzzeek | jogo: bc. theuy work with a mostly blank database | 22:56 |
| *** achanda_ has joined #openstack-nova | 22:56 | |
| *** achanda has quit IRC | 22:57 | |
| *** IanGovett has quit IRC | 22:58 | |
| *** annashen has joined #openstack-nova | 22:58 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: WIP: Fix ComputeNode backport for Service.obj_make_compatible https://review.openstack.org/163867 | 22:58 |
| *** melwitt has quit IRC | 22:59 | |
| *** sdake has quit IRC | 23:00 | |
| *** dims_ has quit IRC | 23:00 | |
| *** otter768 has joined #openstack-nova | 23:00 | |
| *** annashen has quit IRC | 23:00 | |
| *** dims has joined #openstack-nova | 23:00 | |
| *** annashen has joined #openstack-nova | 23:01 | |
| *** erlon has quit IRC | 23:01 | |
| *** apuimedo has quit IRC | 23:01 | |
| *** annegentle has quit IRC | 23:02 | |
| *** apuimedo has joined #openstack-nova | 23:03 | |
| *** melwitt has joined #openstack-nova | 23:05 | |
| *** dims has quit IRC | 23:05 | |
| *** dboik has quit IRC | 23:05 | |
| openstackgerrit | Kent Wang proposed openstack/nova: Adds Not Null constraint to KeyPair name https://review.openstack.org/163230 | 23:06 |
| *** andreykurilin_ has quit IRC | 23:07 | |
| *** achanda has joined #openstack-nova | 23:08 | |
| *** Marga_ has joined #openstack-nova | 23:09 | |
| *** matrohon has quit IRC | 23:10 | |
| *** Marga_ has quit IRC | 23:10 | |
| *** Marga_ has joined #openstack-nova | 23:10 | |
| *** achanda_ has quit IRC | 23:11 | |
| *** igordcard_ has quit IRC | 23:12 | |
| *** nelsnelson has joined #openstack-nova | 23:13 | |
| *** patrickeast has joined #openstack-nova | 23:14 | |
| *** pixelbeat has joined #openstack-nova | 23:14 | |
| *** Marga_ has quit IRC | 23:18 | |
| *** Marga_ has joined #openstack-nova | 23:19 | |
| *** exploreshaifali has quit IRC | 23:21 | |
| *** givi has joined #openstack-nova | 23:22 | |
| *** givi has left #openstack-nova | 23:22 | |
| *** asettle has joined #openstack-nova | 23:23 | |
| jogo | zzzeek: sounds like we need better testing around those bits | 23:25 |
| zzzeek | jogo: nobody uses this routine, right, because itll wipe out your DB right now | 23:26 |
| jogo | zzzeek: so there is a bit of backstory behind the shadow table stuff | 23:26 |
| jogo | mainly we wanted a way to delete old things from the DB | 23:26 |
| zzzeek | jogo: yeah I see what this is doing | 23:26 |
| jogo | but some folks were uncomfortable with just purging so we have the option to store things in shadow table | 23:26 |
| *** otter768 has quit IRC | 23:26 | |
| jogo | zzzeek: what are the values for delete now? | 23:26 |
| zzzeek | jogo: im just trying to understand, noboey uses this yet right? or are proudciotn DBs getting wiped? | 23:26 |
| zzzeek | jogo: we put zeros in that column by default, using a python side rule | 23:27 |
| jogo | isn't delete null until we mark something as deleted? | 23:27 |
| jogo | ohh so its not null ... hehe | 23:27 |
| zzzeek | jogo: not at the moment, maybe this is newer behavior | 23:27 |
| zzzeek | im emailng the list | 23:27 |
| zzzeek | jogo: just do a SELECT from any of your tables, like instance_types | 23:27 |
| zzzeek | I get zeroes | 23:27 |
| jogo | cool, my guess is your right no one really uses this | 23:27 |
| zzzeek | OK | 23:27 |
| *** romainh has left #openstack-nova | 23:27 | |
| jogo | and this sounds like a bug worth fixing at least | 23:27 |
| *** eliqiao has joined #openstack-nova | 23:29 | |
| zzzeek | jogo: its the ultimate worst bug you can have! the “I ran X and it wiped my DB clean” bug! | 23:30 |
| *** hemna is now known as hemnafk | 23:30 | |
| zzzeek | it’s literlally the rm -fr * of databases | 23:30 |
| zzzeek | i was just trying to get my CI to pass and thought this was some innocent race condition in the tests or something, spent all afternoon on this and wow | 23:31 |
| openstackgerrit | Joe Gordon proposed openstack/nova: Stop installing everything to run pep8 https://review.openstack.org/164020 | 23:32 |
| *** achanda has quit IRC | 23:32 | |
| *** achanda has joined #openstack-nova | 23:32 | |
| *** sbfox has quit IRC | 23:32 | |
| openstackgerrit | Joe Gordon proposed openstack/nova: Stop installing everything to run pep8 https://review.openstack.org/164020 | 23:32 |
| jogo | zzzeek: file a bug and I am happy to mark it as super duper critical | 23:33 |
| jogo | because yes you are right | 23:33 |
| jogo | marun: FYI with hacking 0.10 you can do easily do https://review.openstack.org/164020 in neutron | 23:33 |
| openstackgerrit | Josh Gachnang proposed openstack/nova: Add support for cleaning in Ironic driver https://review.openstack.org/161474 | 23:34 |
| marun | jogo: nice! | 23:35 |
| *** melwitt has quit IRC | 23:35 | |
| jogo | should shave time off 'tox -r -epep8' for the naive case -- i.e. no caching etc. | 23:37 |
| claudiub | cyeoh: hello | 23:38 |
| jogo | zzzeek: I can't even call the command | 23:38 |
| zzzeek | oh to run the archiver | 23:39 |
| zzzeek | i never actually *use* openstack commands, i just work with specific tests and Python code | 23:39 |
| zzzeek | i have no idae how to use openstack :) | 23:39 |
| jogo | zzzeek: ubuntu@devstack:~$ nova-manage db archive_deleted_rows 4 | 23:40 |
| zzzeek | right | 23:40 |
| jogo | derp http://paste.openstack.org/show/191972/ | 23:40 |
| *** david-lyle is now known as david-lyle_afk | 23:41 | |
| jogo | full output http://paste.ubuntu.com/10588278/ | 23:41 |
| zzzeek | jogo: yeah i think there is an issue for that as well, but, it woudl be *this issue* that is mostly responsible | 23:41 |
| * jogo thanks pastebinit for being there | 23:41 | |
| *** tjones1 has left #openstack-nova | 23:42 | |
| jogo | zzzeek: do you have a bug filed somewhere? | 23:42 |
| zzzeek | jogo: yes I added you to it, plus the gerrit | 23:42 |
| zzzeek | jogo: yeah here is the issue already reported: https://bugs.launchpad.net/nova/+bug/1183523 | 23:43 |
| openstack | Launchpad bug 1183523 in OpenStack Compute (nova) "db-archiving fails to clear some deleted rows from instances table" [Medium,Confirmed] | 23:43 |
| jogo | this patch ? https://review.openstack.org/#/c/164009/ | 23:44 |
| zzzeek | jogo: well maybe that’s the opposite | 23:44 |
| zzzeek | jogo: yes that will fix it here | 23:44 |
| *** Sukhdev has quit IRC | 23:44 | |
| zzzeek | jogo: but the whole routine is crummy i can make this a lot better | 23:44 |
| *** asettle has quit IRC | 23:45 | |
| *** vladan has left #openstack-nova | 23:47 | |
| jogo | that didn't fix it for me :/ http://paste.ubuntu.com/10588296/ | 23:47 |
| jogo | on http://paste.ubuntu.com/10588298/ | 23:48 |
| zzzeek | jogo: it might still be doing the wrong thing with trying to soft-delete rows that have FK constraints | 23:48 |
| zzzeek | jogo: the query there at least is looking for the right view of deleted: DELETE FROM quota_usages WHERE quota_usages.id in (SELECT T1.id FROM (SELECT quota_usages.id \nFROM quota_usages \nWHERE quota_usages.deleted != %s ORDER BY quota_usages.id \n LIMIT %s) as T1)' (0, 100) | 23:49 |
| *** eliqiao has quit IRC | 23:49 | |
| *** unicell has joined #openstack-nova | 23:49 | |
| *** iamjarvo has quit IRC | 23:50 | |
| zzzeek | jogo: but it woudl have to take FKs into account. it should probably use the ORM and relationships | 23:50 |
| jogo | makes sense | 23:50 |
| jogo | zzzeek: unless you need more data, I am going to go back to triggering quota issues in my env instead | 23:51 |
| zzzeek | jogo: im fine, i just need those tests to pass at some point :) | 23:51 |
| zzzeek | jogo: maybe next week I’ll propose a rewrite of this | 23:51 |
| *** dboik has joined #openstack-nova | 23:52 | |
| yjiang5 | hi, has anyone tried to run grenade locally to reproduce some bug? | 23:52 |
| jogo | zzzeek: cool oh and your bug is now critical | 23:52 |
| yjiang5 | always failed when I try locally. | 23:52 |
| *** dboik_ has joined #openstack-nova | 23:52 | |
| zzzeek | jogo: thanks | 23:52 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Implement instance update logic in Scheduler https://review.openstack.org/160511 | 23:53 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add the instance update calls from Compute https://review.openstack.org/160513 | 23:53 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Modify filters to get instance info from HostState https://review.openstack.org/160512 | 23:53 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add the RPC calls for instance updates. https://review.openstack.org/160507 | 23:53 |
| *** asettle has joined #openstack-nova | 23:53 | |
| jogo | zzzeek: damn I think I just broke my env | 23:54 |
| jogo | zzzeek: by running that command | 23:54 |
| jogo | so you were right | 23:54 |
| *** mtanino has joined #openstack-nova | 23:55 | |
| *** gyee has joined #openstack-nova | 23:56 | |
| *** dboik has quit IRC | 23:56 | |
| *** nelsnelson has quit IRC | 23:57 | |
| *** annegentle has joined #openstack-nova | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!