| *** naohirot has joined #openstack-ironic | 00:04 | |
| *** dkehn__ has joined #openstack-ironic | 00:14 | |
| *** dkehn has quit IRC | 00:14 | |
| *** chlong has joined #openstack-ironic | 00:15 | |
| *** igordcard_ has joined #openstack-ironic | 00:34 | |
| *** igordcard_ has quit IRC | 00:35 | |
| *** igordcard_ has joined #openstack-ironic | 00:36 | |
| *** igordcard_ has quit IRC | 00:37 | |
| *** igordcard_ has joined #openstack-ironic | 00:39 | |
| *** igordcard has joined #openstack-ironic | 00:39 | |
| *** saripurigopi has joined #openstack-ironic | 00:45 | |
| *** saripurigopi has quit IRC | 00:46 | |
| *** igordcard has quit IRC | 00:48 | |
| *** saripurigopi has joined #openstack-ironic | 00:51 | |
| *** gridinv has joined #openstack-ironic | 00:53 | |
| *** saripurigopi has quit IRC | 01:00 | |
| *** marzif has quit IRC | 01:00 | |
| *** igordcard has joined #openstack-ironic | 01:01 | |
| *** krtaylor has quit IRC | 01:02 | |
| *** gridinv has quit IRC | 01:06 | |
| *** krtaylor has joined #openstack-ironic | 01:15 | |
| *** diegolp has quit IRC | 01:23 | |
| *** logan2 has joined #openstack-ironic | 01:25 | |
| *** diegolp has joined #openstack-ironic | 01:33 | |
| *** kan_ has joined #openstack-ironic | 01:34 | |
| *** dguerri`away has quit IRC | 01:41 | |
| *** igordcard has quit IRC | 01:43 | |
| *** radek__ has joined #openstack-ironic | 01:44 | |
| *** dguerri`away has joined #openstack-ironic | 01:45 | |
| *** dguerri`away is now known as dguerri | 01:45 | |
| *** rameshg87 has joined #openstack-ironic | 01:53 | |
| rameshg87 | good morning ironic | 01:53 |
|---|---|---|
| pshige | rameshg87: morining :) | 01:53 |
| rameshg87 | pshige: o/ | 01:53 |
| Haomeng | morning rameshg87, pshige :) | 02:00 |
| pshige | Haomeng: morning :) | 02:01 |
| Haomeng | pshige: :) | 02:01 |
| rameshg87 | Haomeng: o/ | 02:01 |
| *** mtanino has quit IRC | 02:02 | |
| rameshg87 | Haomeng: do you wish to +A https://review.openstack.org/#/c/184944/ ? it has 2x+2s now :) | 02:09 |
| Haomeng | rameshg87: sure:) | 02:09 |
| rameshg87 | thanks | 02:09 |
| Haomeng | rameshg87: welcome:) | 02:10 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add generic RAID configuration spec for liberty https://review.openstack.org/173214 | 02:10 |
| pshige | http://libertydesignsummit.sched.org/thierry#.VWPQb_ntmko | 02:12 |
| pshige | mispaste ... | 02:16 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add inband RAID configuration spec for liberty https://review.openstack.org/173218 | 02:20 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add new boot interface in Ironic https://review.openstack.org/168698 | 02:25 |
| pshige | Haomeng: gate-bifrost came to work at last. Would you like to get things done on bifrost? https://review.openstack.org/#/q/project:openstack/bifrost,n,z | 02:32 |
| Haomeng | pshige: ok, you mean for my patch https://review.openstack.org/#/c/185294/ ? | 02:33 |
| Haomeng | pshige: ok, I will review others | 02:34 |
| pshige | I mean all the thing if you can | 02:34 |
| Haomeng | pshige: these days I am working on bifrost also | 02:34 |
| pshige | thanks! | 02:34 |
| Haomeng | pshige: ok, sure, thanks for your reminder:) | 02:34 |
| pshige | bifrost seems to be interesting! :) | 02:35 |
| *** ramineni has joined #openstack-ironic | 02:38 | |
| Haomeng | pshige: yes, I think it should be best candidate to deploy and undercloud os, maybe we can integrate it with discovery, then it will more useful | 02:41 |
| Haomeng | pshige: I have more idea, should we have same standalone ironic support for dib elements, then dib can build standalone ironic image or iso, or usb drive, which can be used directly without installation, not sure if this good idea:) | 02:46 |
| Haomeng | pshige: and I want to enable proxy support for bifrost to install pip/yum/apt etc via proxy, and try to add to 'all' env proxy var not working, any idea? | 02:48 |
| openstackgerrit | Merged openstack/ironic: Update pyremotevbox documentation https://review.openstack.org/184944 | 02:52 |
| openstackgerrit | Merged openstack/bifrost: Fix documentation in bifrost https://review.openstack.org/182514 | 03:00 |
| *** radek__ has quit IRC | 03:07 | |
| *** saripurigopi has joined #openstack-ironic | 03:22 | |
| pshige | Haomeng: good! | 03:27 |
| *** davideagnello has joined #openstack-ironic | 03:32 | |
| *** gridinv has joined #openstack-ironic | 03:34 | |
| *** davideagnello has quit IRC | 03:37 | |
| *** gridinv has quit IRC | 03:39 | |
| *** gridinv has joined #openstack-ironic | 03:43 | |
| *** yuikotakada has joined #openstack-ironic | 03:56 | |
| *** gridinv has quit IRC | 03:57 | |
| *** dtantsur|afk is now known as dtantsur | 04:02 | |
| dtantsur | Morning Ironic | 04:02 |
| pshige | dtantsur: morning :) | 04:03 |
| dtantsur | o/ | 04:04 |
| *** alexpilotti has joined #openstack-ironic | 04:06 | |
| *** achanda has joined #openstack-ironic | 04:12 | |
| *** blakec has joined #openstack-ironic | 04:13 | |
| *** Nisha has joined #openstack-ironic | 04:15 | |
| *** blakec has quit IRC | 04:17 | |
| *** eghobo has joined #openstack-ironic | 04:23 | |
| *** absubram has joined #openstack-ironic | 04:25 | |
| *** links has joined #openstack-ironic | 04:27 | |
| *** achanda has quit IRC | 04:30 | |
| *** absubram has quit IRC | 04:31 | |
| *** absubram has joined #openstack-ironic | 04:32 | |
| *** alexpilotti has quit IRC | 04:41 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Virtual Media Deploy module for iRMC Driver https://review.openstack.org/151958 | 04:42 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Refactor ilo.common._prepare_floppy_image() https://review.openstack.org/185494 | 04:44 |
| *** absubram has quit IRC | 04:45 | |
| *** absubram has joined #openstack-ironic | 04:46 | |
| *** zz_natorious is now known as natorious | 04:46 | |
| *** pradipta has joined #openstack-ironic | 04:48 | |
| *** absubram has quit IRC | 04:52 | |
| *** natorious is now known as zz_natorious | 04:57 | |
| *** davideagnello has joined #openstack-ironic | 05:00 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add localboot support for iscsi_irmc driver https://review.openstack.org/163738 | 05:02 |
| *** penick has joined #openstack-ironic | 05:03 | |
| *** penick has quit IRC | 05:03 | |
| *** logan2 has quit IRC | 05:08 | |
| *** chenglch has joined #openstack-ironic | 05:09 | |
| *** davideagnello has quit IRC | 05:16 | |
| *** smoriya has quit IRC | 05:20 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add whole disk image support for iscsi_irmc driver https://review.openstack.org/181771 | 05:21 |
| openstackgerrit | Merged openstack/ironic: Delete neutron ports when the node cleaning fails https://review.openstack.org/178570 | 05:28 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add IPA support for iscsi_irmc driver https://review.openstack.org/181807 | 05:29 |
| *** itamarl has joined #openstack-ironic | 05:29 | |
| *** ukalifon1 has joined #openstack-ironic | 05:36 | |
| *** rameshg87 is now known as rameshg87-lunch | 05:40 | |
| Nisha | rameshg87-lunch, hi | 05:46 |
| *** e0ne has joined #openstack-ironic | 05:46 | |
| *** itamarl has quit IRC | 05:46 | |
| *** itamarl has joined #openstack-ironic | 05:47 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Refactor ilo.common._prepare_floppy_image() https://review.openstack.org/185494 | 05:47 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Rename discoverd -> inspector https://review.openstack.org/185499 | 05:49 |
| dtantsur | \o/ | 05:49 |
| *** e0ne has quit IRC | 05:53 | |
| Nisha | dtantsur, o/ | 05:56 |
| dtantsur | o/ | 05:56 |
| pshige | Nisha: o/ | 05:59 |
| Nisha | pshige, o/ | 06:00 |
| openstackgerrit | Haomeng,Wang proposed openstack/bifrost: supports redhat yum installation https://review.openstack.org/185294 | 06:04 |
| *** chlong has quit IRC | 06:30 | |
| *** radek__ has joined #openstack-ironic | 06:34 | |
| *** mdbooth has quit IRC | 06:35 | |
| *** mdbooth has joined #openstack-ironic | 06:41 | |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185508 | 06:42 |
| *** itamarl_ has joined #openstack-ironic | 06:43 | |
| *** itamarl has quit IRC | 06:44 | |
| *** itamarl_ is now known as itamarl | 06:44 | |
| openstackgerrit | Haomeng,Wang proposed openstack/bifrost: supports redhat yum installation https://review.openstack.org/185294 | 06:45 |
| *** Haomeng|2 has joined #openstack-ironic | 06:48 | |
| *** ifarkas has joined #openstack-ironic | 06:50 | |
| *** Haomeng has quit IRC | 06:51 | |
| *** itamarl has quit IRC | 06:52 | |
| *** itamarl has joined #openstack-ironic | 06:52 | |
| *** yuanying_ has joined #openstack-ironic | 06:54 | |
| *** yuanying has quit IRC | 06:54 | |
| *** rameshg87-lunch has quit IRC | 06:55 | |
| *** achanda has joined #openstack-ironic | 07:04 | |
| *** logan2 has joined #openstack-ironic | 07:06 | |
| *** yuanying_ has quit IRC | 07:12 | |
| *** yuanying has joined #openstack-ironic | 07:13 | |
| openstackgerrit | Haomeng,Wang proposed openstack/bifrost: supports redhat yum installation https://review.openstack.org/185294 | 07:13 |
| *** eghobo has quit IRC | 07:15 | |
| *** logan2 has quit IRC | 07:17 | |
| *** gridinv has joined #openstack-ironic | 07:25 | |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185520 | 07:28 |
| *** pas-ha has joined #openstack-ironic | 07:28 | |
| *** achanda has quit IRC | 07:30 | |
| *** itamarl has quit IRC | 07:31 | |
| *** itamarl has joined #openstack-ironic | 07:31 | |
| *** ndipanov has quit IRC | 07:35 | |
| *** jistr has joined #openstack-ironic | 07:39 | |
| *** edcragg has joined #openstack-ironic | 07:39 | |
| *** e0ne has joined #openstack-ironic | 07:39 | |
| *** e0ne is now known as e0ne_ | 07:40 | |
| *** ndipanov has joined #openstack-ironic | 07:42 | |
| openstackgerrit | Merged openstack/ironic: Do not add auth token in context for noauth API mode https://review.openstack.org/185376 | 07:42 |
| *** achanda has joined #openstack-ironic | 07:48 | |
| *** e0ne_ is now known as e0ne | 07:49 | |
| *** itamarl has joined #openstack-ironic | 07:49 | |
| *** edcragg has quit IRC | 07:53 | |
| *** itamarl_ has joined #openstack-ironic | 07:55 | |
| *** dlpartain has joined #openstack-ironic | 07:55 | |
| *** viktors|afk is now known as viktors | 07:56 | |
| *** itamarl has quit IRC | 07:57 | |
| *** itamarl_ is now known as itamarl | 07:57 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Do not preserve partitions after ephemeral https://review.openstack.org/182688 | 08:02 |
| openstackgerrit | Merged openstack/ironic: Disable meaningless sort keys in list command https://review.openstack.org/182563 | 08:02 |
| *** dlpartain has left #openstack-ironic | 08:02 | |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185523 | 08:03 |
| *** yuanying has quit IRC | 08:09 | |
| *** yog__ has joined #openstack-ironic | 08:14 | |
| openstackgerrit | Zhai, Edwin proposed openstack/ironic-specs: New meter for Node Manager https://review.openstack.org/176109 | 08:14 |
| *** lucasagomes has joined #openstack-ironic | 08:15 | |
| *** romcheg has joined #openstack-ironic | 08:16 | |
| *** derekh has joined #openstack-ironic | 08:20 | |
| *** bkero has quit IRC | 08:21 | |
| *** bkero has joined #openstack-ironic | 08:21 | |
| *** gpocentek has joined #openstack-ironic | 08:25 | |
| openstackgerrit | baiyuan proposed openstack/ironic: Add documentation for getting a node's console https://review.openstack.org/176642 | 08:25 |
| *** edcragg has joined #openstack-ironic | 08:33 | |
| *** e0ne is now known as e0ne_ | 08:42 | |
| *** e0ne_ is now known as e0ne | 08:43 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Fix sync local state periodic task for standalone mode https://review.openstack.org/185533 | 08:45 |
| lintan_ | hi lucasagomes? | 08:53 |
| lucasagomes | lintan_, hi there | 08:54 |
| lucasagomes | lintan_, thanks for the review, I know the "vendor" in the name sucks :-( | 08:54 |
| Haomeng|2 | lucasagomes: welcome back:) | 08:54 |
| lucasagomes | Haomeng|2, hey there! thanks, how it's going? | 08:55 |
| lintan_ | lucasagomes, That's fine for me :) I am looking at the API thing, so why do we need links stuff? | 08:55 |
| Haomeng|2 | lucasagomes: I did not attend the summit, but want to know our discusstion in summit, so can you help to share the url? | 08:55 |
| *** lazy_prince has joined #openstack-ironic | 08:55 | |
| Haomeng|2 | lucasagomes: I checked http://bit.ly/ironic-whiteboard, looks like there is no this summit content | 08:56 |
| Haomeng|2 | lucasagomes: thank you Lucas:) | 08:56 |
| lintan_ | Haomeng|2, you can follow this https://etherpad.openstack.org/p/liberty-ironic-design-summit-ideas | 08:57 |
| Haomeng|2 | lintan_: ok, thanks:) | 08:57 |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185523 | 08:57 |
| Haomeng|2 | lintan_: do we have more etherpad, not sure if this one cover all sessions:) | 08:58 |
| lucasagomes | lintan_, oh links is something that we first copied from the nova api http://developer.openstack.org/api-ref-compute-v2.html | 08:59 |
| *** bcornec has joined #openstack-ironic | 08:59 | |
| lintan_ | Haomeng|2, yes, we have the detail log on theses sections list on Working sessions and "Fishbowl" slots | 08:59 |
| lucasagomes | lintan_, it's to help clients to obtain instead of having to create URLs | 08:59 |
| lucasagomes | Haomeng|2, yeah lintan_ etherpad is correct. There's also the etherpads from specific design sessions (will grab some links) | 08:59 |
| lintan_ | lucasagomes, OK, I will have a look at it, thanks :) | 09:00 |
| lucasagomes | Haomeng|2, https://etherpad.openstack.org/p/liberty-ironic-network-isolation https://etherpad.openstack.org/p/liberty-ops-ironic-feedback https://etherpad.openstack.org/p/liberty-ironic-cycle-goals https://etherpad.openstack.org/p/liberty-ironic-nova-virt-driver https://etherpad.openstack.org/p/liberty-ironic-rack-to-ready-state | 09:00 |
| *** pradipta has quit IRC | 09:01 | |
| *** logan2 has joined #openstack-ironic | 09:02 | |
| *** subscope has joined #openstack-ironic | 09:03 | |
| *** leopoldj has joined #openstack-ironic | 09:04 | |
| *** sambetts has joined #openstack-ironic | 09:06 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Override PXE kernel cmdline via Glance property https://review.openstack.org/178229 | 09:17 |
| *** itamarl_ has joined #openstack-ironic | 09:37 | |
| *** itamarl has quit IRC | 09:38 | |
| *** itamarl_ is now known as itamarl | 09:38 | |
| *** naohirot has quit IRC | 09:56 | |
| *** itamarl has quit IRC | 10:01 | |
| *** e0ne is now known as e0ne_ | 10:08 | |
| *** Nisha has quit IRC | 10:10 | |
| *** e0ne_ has quit IRC | 10:19 | |
| *** MattMan has joined #openstack-ironic | 10:21 | |
| *** e0ne has joined #openstack-ironic | 10:28 | |
| Haomeng|2 | lucasagomes: ok, thank you Lucas:) | 10:38 |
| *** e0ne is now known as e0ne_ | 10:39 | |
| *** marzif has joined #openstack-ironic | 10:39 | |
| *** e0ne_ has quit IRC | 10:49 | |
| *** itamarl has joined #openstack-ironic | 10:52 | |
| sambetts | o/ hello all | 10:57 |
| saripurigopi | Hi all | 11:01 |
| *** chenglch has quit IRC | 11:04 | |
| *** ramineni has quit IRC | 11:05 | |
| TheJulia | good morning | 11:14 |
| dtantsur | sambetts, saripurigopi, TheJulia, morning! | 11:20 |
| *** saripurigopi has quit IRC | 11:26 | |
| sambetts | dtantsur: o/ | 11:29 |
| *** thrash|g0ne is now known as thrash | 11:30 | |
| lucasagomes | sambetts, TheJulia saripuri<tab> morning | 11:33 |
| sambetts | lucasagomes: o/ did you have a good trip back? | 11:33 |
| lucasagomes | sambetts, hey man, yeah pretty good. No delays this time | 11:33 |
| *** jistr is now known as jistr|class | 11:33 | |
| lucasagomes | how was ur trip back? | 11:33 |
| sambetts | lucasagomes: Pretty good, flew through the night and didn't get much sleep, but everything was pretty good :) | 11:34 |
| lucasagomes | nice one :-) | 11:35 |
| *** lucasagomes is now known as lucas-hungry | 11:40 | |
| * TheJulia ponders breakfast | 11:41 | |
| *** t0msen has joined #openstack-ironic | 11:56 | |
| t0msen | hi. someone experiences with ironic and keystone v3? | 11:57 |
| *** e0ne has joined #openstack-ironic | 12:00 | |
| *** trown|outttypeww is now known as trown | 12:02 | |
| *** dguerri is now known as dguerri`away | 12:10 | |
| *** logan2 has quit IRC | 12:12 | |
| *** ekarlso has quit IRC | 12:16 | |
| *** ekarlso has joined #openstack-ironic | 12:16 | |
| openstackgerrit | baiyuan proposed openstack/ironic: Add documentation for getting a node's console https://review.openstack.org/176642 | 12:18 |
| *** dprince has joined #openstack-ironic | 12:18 | |
| *** yog__ has quit IRC | 12:29 | |
| *** achanda has quit IRC | 12:33 | |
| *** lazy_prince has quit IRC | 12:38 | |
| *** lazy_prince has joined #openstack-ironic | 12:39 | |
| *** lucas-hungry is now known as lucasagomes | 12:41 | |
| *** mjturek1 has joined #openstack-ironic | 12:41 | |
| *** achanda has joined #openstack-ironic | 12:41 | |
| *** cdearborn has joined #openstack-ironic | 12:44 | |
| *** killer_prince has joined #openstack-ironic | 12:49 | |
| *** dguerri`away is now known as dguerri | 12:50 | |
| *** lazy_prince has quit IRC | 12:51 | |
| lucasagomes | t0msen, hi there, not v3. But ask maybe someone in the channel knows something | 12:51 |
| *** lazy_prince has joined #openstack-ironic | 12:53 | |
| *** killer_prince has quit IRC | 12:56 | |
| *** jistr|class is now known as jistr | 13:00 | |
| *** vdrok has quit IRC | 13:00 | |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185523 | 13:02 |
| *** achanda has quit IRC | 13:05 | |
| openstackgerrit | Syed Ismail Faizan Barmawer proposed openstack/ironic-lib: Merge lastet disk partitioning code from ironic https://review.openstack.org/162162 | 13:10 |
| *** e0ne is now known as e0ne_ | 13:10 | |
| *** kan_ has quit IRC | 13:13 | |
| *** e0ne_ has quit IRC | 13:21 | |
| *** logan2 has joined #openstack-ironic | 13:21 | |
| *** MattMan has left #openstack-ironic | 13:22 | |
| *** rloo has joined #openstack-ironic | 13:24 | |
| *** e0ne has joined #openstack-ironic | 13:26 | |
| *** links has quit IRC | 13:29 | |
| *** lazy_prince has quit IRC | 13:33 | |
| *** [1]cdearborn has joined #openstack-ironic | 13:37 | |
| openstackgerrit | baiyuan proposed openstack/ironic: fix node-get-console returns url always start with http https://review.openstack.org/185523 | 13:39 |
| *** GheRivero has left #openstack-ironic | 13:40 | |
| *** zz_natorious is now known as natorious | 13:41 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add pluggable credentials storage https://review.openstack.org/185074 | 13:44 |
| *** natorious is now known as zz_natorious | 13:45 | |
| *** vdrok has joined #openstack-ironic | 13:48 | |
| *** cdearborn has quit IRC | 13:48 | |
| *** pas-ha has quit IRC | 13:49 | |
| *** logan2 has quit IRC | 13:54 | |
| *** yuikotakada has quit IRC | 13:55 | |
| *** bnemec has joined #openstack-ironic | 13:57 | |
| *** rloo has quit IRC | 14:01 | |
| *** rloo has joined #openstack-ironic | 14:02 | |
| *** jjohnson2 has joined #openstack-ironic | 14:09 | |
| BadCub | Morning folks | 14:12 |
| *** saripurigopi has joined #openstack-ironic | 14:13 | |
| TheJulia | good morning BadCub | 14:13 |
| BadCub | Morning TheJulia :) | 14:14 |
| NobodyCam | good morning Ironicers | 14:15 |
| jroll | morning BadCub TheJulia NobodyCam lucasagomes etc etc :) | 14:16 |
| lucasagomes | BadCub, NobodyCam jroll morning! | 14:16 |
| NobodyCam | morning TheJulia jroll lucasagomes and BadCub :) | 14:16 |
| BadCub | Morning jroll lucasagomes | 14:16 |
| TheJulia | good morning NobodyCam | 14:16 |
| BadCub | mornin NobodyCam :) | 14:16 |
| *** saripurigopi has quit IRC | 14:17 | |
| jroll | so much info to collect from last week | 14:17 |
| * jroll chugs coffee | 14:17 | |
| sambetts | Morning BadCub, NobodyCam, jroll | 14:17 |
| NobodyCam | morning sambetts | 14:17 |
| NobodyCam | :) | 14:17 |
| BadCub | morning sambetts :) | 14:17 |
| jroll | \o sambetts | 14:18 |
| NobodyCam | how was everyones flight back home | 14:18 |
| rloo | morning and hello ironickers TheJulia, BadCub, NobodyCam, jroll, sambetts, lucasagomes :) | 14:19 |
| jroll | hiya rloo :) | 14:20 |
| TheJulia | NobodyCam: uneventful :) | 14:20 |
| *** saripurigopi has joined #openstack-ironic | 14:20 | |
| lucasagomes | rloo, ahoy | 14:20 |
| *** [1]cdearborn has quit IRC | 14:21 | |
| BadCub | morning rloo :) | 14:21 |
| NobodyCam | morning rloo :) | 14:22 |
| NobodyCam | TheJulia: thats good news! | 14:22 |
| sambetts | Morning rloo :) | 14:24 |
| sambetts | NobodyCam: Good thanks, yours? | 14:25 |
| rloo | hey lucasagomes, did you see the email from michael (nova) about using depends-on for patches that need specs approved? | 14:25 |
| rloo | lucasagomes: http://lists.openstack.org/pipermail/openstack-dev/2015-May/064604.html | 14:26 |
| lucasagomes | rloo, oh I did not | 14:26 |
| * lucasagomes looks | 14:26 | |
| jroll | oh, I like that | 14:26 |
| dtantsur | morning jroll, rloo, NobodyCam, BadCub and everyone :) | 14:27 |
| rloo | jroll, lucasagomes: the only problem I see with that is if a patch is approved but doesn't match the approved spec. | 14:27 |
| NobodyCam | morning dtantsur | 14:28 |
| jroll | rloo: as in the depends-on is wrong? | 14:28 |
| lucasagomes | rloo, interesting... Yeah that's def a good idea | 14:29 |
| jroll | oh, never mind, I see | 14:29 |
| BadCub | morning dtantsur :) | 14:29 |
| lucasagomes | jroll, sorry I meant to review the release cycle patch | 14:29 |
| rloo | jroll, lucasagomes: it might be worth trying anyway, cuz no one likes -2 | 14:29 |
| lucasagomes | do you have it handy there? | 14:29 |
| jroll | the links are what I like about it, I still don't think we should approve code before specs | 14:29 |
| jroll | yeah | 14:29 |
| jroll | lucasagomes: https://review.openstack.org/#/c/185171/ | 14:29 |
| lucasagomes | rloo, +1 yeah the -2 is def too harsh | 14:30 |
| jroll | I wish people didn't see -2 as hurtful :( | 14:30 |
| jroll | it's a process thing, has nothing to do with the code or the person | 14:30 |
| *** cdearborn has joined #openstack-ironic | 14:31 | |
| rloo | jroll: agree, but we're human and they do | 14:31 |
| jroll | right. makes me sad. | 14:31 |
| cdearborn | Hey folks, having problems running tox. The py34 tests fail with: db type could not be determined | 14:33 |
| cdearborn | error: testr failed (3) | 14:33 |
| cdearborn | any thoughts? | 14:33 |
| cdearborn | on fedora 21 | 14:33 |
| cdearborn | This is after following the dev quickstart instructions here: http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html | 14:35 |
| NobodyCam | cdearborn: what were the failures? can you paste(.openstack.org) them so we can see? | 14:37 |
| cdearborn | NobodyCam, sure: http://pastebin.com/XZqpDAvC | 14:38 |
| cdearborn | From what I have been able to glean through google, the "db type could not be determined" error could be because the test results db was created by py27, and py34 is trying to read it | 14:39 |
| jroll | ohhh | 14:40 |
| jroll | you could blow away .testrepository or whatever it's called | 14:40 |
| jroll | that's where testr stores its db | 14:40 |
| NobodyCam | yea! | 14:40 |
| NobodyCam | tox -r too | 14:40 |
| jroll | idk if -r rebuilds the testr db, though | 14:41 |
| *** e0ne is now known as e0ne_ | 14:41 | |
| *** e0ne_ has quit IRC | 14:41 | |
| NobodyCam | brb ... | 14:42 |
| lucasagomes | sorry was in a short call | 14:43 |
| lucasagomes | jroll, yeah apparently people see it as hurtful | 14:43 |
| jroll | right. | 14:44 |
| lucasagomes | so I think it would be good if we can start using depends-on | 14:44 |
| lucasagomes | perhaps we need a guideline about it | 14:44 |
| *** dguerri is now known as dguerri`away | 14:44 | |
| lucasagomes | https://wiki.openstack.org/wiki/Ironic/Specs_Process | 14:44 |
| lucasagomes | and mention it in the template as well | 14:45 |
| sambetts | lucasagomes: and APIImpact and DocImpact flags in commit messages too probably | 14:45 |
| jroll | I don't think that's a good solution, personally, but it can't hurt | 14:45 |
| *** achanda has joined #openstack-ironic | 14:45 | |
| jroll | I prefer to be the change I want to see in the world | 14:45 |
| lucasagomes | yeah | 14:45 |
| lucasagomes | I don't know any better way to do it either | 14:45 |
| lucasagomes | not with the current tools we have | 14:45 |
| rloo | lucasagomes, jroll: use depends-on instead of -2, right? I'm good with that, but do we need to vote on it or mention in email too? | 14:48 |
| rloo | would also make it easier to find the link to the specs; linking to the BP is a pain... | 14:49 |
| jroll | rloo: I don't think hurt feelings should change the process, but I'm +1 on *also* using depends-on | 14:49 |
| *** zz_natorious is now known as natorious | 14:50 | |
| rloo | jroll: so depends-on and -2. that seems like overkill. | 14:50 |
| *** romcheg has left #openstack-ironic | 14:50 | |
| jroll | rloo: then just -2, maybe, idk | 14:51 |
| rloo | jroll: the other reason for not -2, is cuz it is hard to remove the -2 if you aren't the person that added it. | 14:51 |
| jroll | if a good comment about *why* the -2 is being put there, it shouldn't hurt feelings | 14:51 |
| rloo | jroll: one person said that it prevents folks from reviewing their patches. Which it shouldn't but maybe it does. It prevents me but even w/o it, I wouldn't look until a spec is approved. | 14:52 |
| lucasagomes | I don't know how customizable is gerrit. But maybe worth throwing an idea to the infra folks about adding a new flag to indicate that the code is waiting for the spec to land? | 14:53 |
| jroll | rloo: yeah, I don't think the -2 prevents people from reviewing except if reviewers have their own rules around it | 14:53 |
| lucasagomes | something that persist between new patch sets but that all members of the core group could lift as well | 14:53 |
| lucasagomes | or that is automatically lift after the spec is merged | 14:54 |
| rloo | lucasagomes. yeah, a better solution is to do something with gerrit. Not a -2, but 'cannot land' that any core can remove. I guess. | 14:55 |
| lucasagomes | yeah, we can talk to infra see if that's possible I believe | 14:56 |
| *** jjohnson2_ has joined #openstack-ironic | 14:56 | |
| rloo | might be worth seeing what people say in that email thread. | 14:56 |
| jroll | maybe one of you should reply to that thread with the problem statement from our perspective | 14:57 |
| cdearborn | tried deleting .testrepository & running tox -epy34, and got past that error. I'm now getting a ton of "TypeError: 'xxx' object is not iterable" errors that are not appearing in the gate. Any thoughts? | 14:57 |
| jroll | but I don't think it's going to realistically solve any problems except the "other cores can't remove -2", which I don't believe has slowed things down in practice | 14:57 |
| *** jjohnson2 has quit IRC | 14:58 | |
| *** e0ne has joined #openstack-ironic | 14:58 | |
| cdearborn | http://paste.openstack.org/show/237938/ | 14:58 |
| * NobodyCam shaks fist at his vpn :-p | 15:01 | |
| NobodyCam | shakes even | 15:01 |
| rloo | jroll: ok, i'll bite... | 15:01 |
| TheJulia | NobodyCam: is it not behaving? | 15:01 |
| *** hemnafk is now known as hemna | 15:01 | |
| *** mtanino has joined #openstack-ironic | 15:01 | |
| jroll | does corp vpn ever behave? :) | 15:01 |
| NobodyCam | lol ... operator error .... E_LACK_OF_COFFEE | 15:02 |
| jroll | cdearborn: try "tox -r" to reinstall dependencies etc | 15:02 |
| NobodyCam | jroll: +++ | 15:02 |
| TheJulia | jroll: every other thursday | 15:03 |
| NobodyCam | ahh it only tuesday thats why its not working | 15:03 |
| *** pas-ha has joined #openstack-ironic | 15:05 | |
| *** jjohnson2_ is now known as jjohnson2 | 15:05 | |
| rloo | jroll: remind me, you are against not using -2 because ...? | 15:06 |
| jroll | rloo: because "it might hurt people's feelings" is a bad reason to change that. I'd rather change the perception. | 15:07 |
| jroll | (imho) | 15:07 |
| rloo | jroll: but the depends-on does the same/similar thing. prevent patch from merging until spec is approved. | 15:07 |
| rloo | jroll: except in the corner case of a patch being approved, spec changing after that and being approved, and no one noticing. | 15:08 |
| jroll | rloo: sure, I'm not strictly opposed to switching, for a good reason | 15:09 |
| *** t0msen has quit IRC | 15:11 | |
| rloo | jroll: so I think that 'might hurt people's feelings' is actually a valid (not saying how 'good') reason. But it means we don't need to find/remind the -2 person to undo it. I agree I don't think it has caused problems, but it would be one less thing to deal with. | 15:13 |
| jroll | rloo: well, instead we would have to deal with teaching people how to use, and enforcing the use of, depends-on | 15:13 |
| rloo | jroll: true, but I think that would be a good thing to do anyway, AND it makes it easier for me to find the link to the spec. | 15:13 |
| jroll | fair | 15:14 |
| jroll | I'm eager to hear others' opinions on it | 15:14 |
| rloo | jroll: yeeeees. although i'm slightly afraid to add to that existing email thread. | 15:16 |
| jroll | rloo: heh, it will be fine :) | 15:16 |
| *** dkehn__ is now known as dkehnx | 15:16 | |
| jroll | rloo: something like "on that note, we've been talking about how to make -2 better" | 15:16 |
| rloo | jroll: ok, deep breath... :) | 15:17 |
| jroll | :) | 15:18 |
| *** itamarl has quit IRC | 15:20 | |
| *** achanda_ has joined #openstack-ironic | 15:22 | |
| *** achanda has quit IRC | 15:22 | |
| * krotscheck didn't read backscroll, but now would ironic's new release strategy work with the oslo library freeze? | 15:22 | |
| krotscheck | *how | 15:23 |
| krotscheck | (y'all might have covered this already) | 15:23 |
| krotscheck | Just... abandon global requirements altogether? | 15:23 |
| *** krtaylor has quit IRC | 15:23 | |
| jroll | krotscheck: we haven't talked a ton about that, I encourage you to ask that question here :) https://review.openstack.org/#/c/185171/ | 15:23 |
| krotscheck | Aaaalllrightey | 15:23 |
| lucasagomes | jroll, re versioning. We are going to use our own internal version right? | 15:27 |
| *** viktors is now known as viktors|afk | 15:27 | |
| lucasagomes | I mean, right now in the setup.cfg we use the openstack one | 15:28 |
| lucasagomes | version = 2015.2 | 15:28 |
| lucasagomes | I wonder how we can version Ironic outside OpenStack | 15:29 |
| * lucasagomes looks at swift setup.cfg | 15:29 | |
| lucasagomes | yeah they don't have a version there hmmm | 15:29 |
| devananda | g'morning, all | 15:30 |
| * devananda reads backscroll | 15:31 | |
| NobodyCam | good morning devananda | 15:31 |
| rloo | morning devananda | 15:32 |
| lucasagomes | devananda, morning | 15:32 |
| rloo | jroll: sent out email. decided not to mention your stance about 'hurting people's feelings' isn't a good reason :-) | 15:32 |
| * devananda likes depends-on for patches waiting on specs, too | 15:32 | |
| *** saripurigopi has quit IRC | 15:33 | |
| devananda | jroll: re: -2 "preventing reviews" -- some reviewers on some projects simply filter out any patch that has -1 or -2, because their queue is so long | 15:35 |
| devananda | jroll: so a constructive comment, while it should help avoid hurt feelings, would still result in less (possibly no) people looking at that patch until the -2 is removed | 15:36 |
| pshige | oh | 15:36 |
| pshige | devananda: morning :) | 15:37 |
| jroll | devananda: sure, but that's a concious choice reviewers are making | 15:37 |
| jroll | devananda: e.g. rloo still won't look at patches without an approved spec | 15:37 |
| devananda | lucasagomes, krotscheck: if we switch to independent release model, we'd start with release number 11, I think. | 15:38 |
| devananda | jroll: it's not a conscious choice any longer - its encoded in the tooling those reviewers use | 15:39 |
| devananda | so they dont even *see* the patch | 15:39 |
| lucasagomes | right, but we have to somehow map a release number to the openstack versioning ? say 2016.1 for the next release | 15:39 |
| * lucasagomes needs to read more how swift do that | 15:39 | |
| dtantsur | devananda, morning! | 15:39 |
| jroll | I wonder how pip/pbr/etc is going to handle this when we go backwards in versions (2015.x -> 1.x or whatever) | 15:39 |
| dtantsur | devananda, may I have your ack on https://review.openstack.org/#/c/185442/ ? | 15:39 |
| jroll | lucasagomes: AIUI swift ignores the "global" version | 15:39 |
| lucasagomes | I know it's possible, I just need to read more about how other openstack projects (swift) does it | 15:39 |
| lucasagomes | jroll, right | 15:40 |
| jroll | so you say "5.x is compatible with kilo" or something | 15:40 |
| lucasagomes | seems fair | 15:40 |
| *** jistr has quit IRC | 15:40 | |
| devananda | "or something" :) | 15:41 |
| devananda | dtantsur: looking | 15:41 |
| pshige | :) | 15:41 |
| devananda | dtantsur: rename looks fine. | 15:43 |
| devananda | dtantsur: re: stable branches, how are you tracking backports? | 15:43 |
| *** e0ne is now known as e0ne_ | 15:43 | |
| dtantsur | devananda, manually cherry-pick stuff I consider critical :) | 15:44 |
| devananda | dtantsur: and do you intend to rename the project on launchpad and pip? | 15:44 |
| dtantsur | devananda, I do (for pip I'd rather create a new one) | 15:45 |
| *** leopoldj has quit IRC | 15:45 | |
| devananda | cool | 15:45 |
| jroll | devananda: are there notes somewhere on the outcome of friday's neutron/ironic talks? heard something about a weekly meeting? | 15:45 |
| NobodyCam | brb | 15:46 |
| *** krtaylor has joined #openstack-ironic | 15:46 | |
| *** e0ne_ is now known as e0ne | 15:46 | |
| lucasagomes | if you guys have some time, mind taking a look at https://review.openstack.org/#/c/183491/ ? | 15:46 |
| lucasagomes | I want to use that to avoid having nova to fetch the whole node object when pooling Ironic | 15:47 |
| devananda | jroll: sukhdev took an action item to set up a weekly meeting between the folks interested, on irc with all the usual trappings of such a meeting | 15:47 |
| * lucasagomes wonders also if that requires a microversion | 15:47 | |
| rloo | lucasagomes: just wondering -- why isn't there a bug for that? doesn't it change the API? | 15:47 |
| lucasagomes | rloo, it doesn't | 15:48 |
| rloo | lucasagomes: isn't the fields=xxx new? | 15:48 |
| lucasagomes | everything works as before, we just added a new optional parameter for the GETs | 15:48 |
| jroll | devananda: cool, anything else I should be aware of? | 15:48 |
| devananda | jroll: we will need to track all the dependencies somewhere (not yet determined), including cloud-init (v2) changes, and set up a CI job for it using OVS | 15:49 |
| rloo | lucasagomes: i notice sometimes that people push up patches w/o bug or spec links. When is it appropriate to do that? Cuz this one seems to at least need a bug. | 15:49 |
| lucasagomes | rloo, yes it is. What I mean by it doesn't change the API is that it doesn't breaks anyone api | 15:49 |
| *** derekh_ has joined #openstack-ironic | 15:50 | |
| lucasagomes | rloo, why? | 15:50 |
| lucasagomes | rloo, it's not, it's an enhancement | 15:50 |
| rloo | lucasagomes: yeah, it doesn't break any api (assuming that if before, 'fields=xx' was ignored | 15:50 |
| devananda | jroll: we also talked about whether LAG / port bonding should be considered part of the MVP for this, or whether we can punt on that to reduce complexity of the initial implementation | 15:50 |
| rloo | lucasagomes: so it is ok to submit patches for enhancements, w/o a bug or spec? | 15:50 |
| lucasagomes | I think so | 15:50 |
| rloo | lucasagomes: seems like it needs a docimpact anyway? | 15:50 |
| lucasagomes | if it doesn't affect multiple areas of the project (for a spec) and is not a bug | 15:51 |
| rloo | lucasagomes: so we don't have a way of tracking "enhancements". I'm fine with that as long as people know what enhancement is vs bug fix vs feature. | 15:51 |
| devananda | lucasagomes: so this is fixing a bug in nova, isn't it? | 15:51 |
| jroll | devananda: thanks. I tend to think this isn't horribly useful without bonding, but I see why we might want to do it. will think on it. | 15:52 |
| lucasagomes | devananda, it will be used to fix a bug in Nova | 15:52 |
| devananda | lucasagomes: so it should have a related-to tag, no? | 15:52 |
| lucasagomes | so we can say it's related | 15:52 |
| devananda | jroll: I think that was the concensus :) | 15:52 |
| lucasagomes | devananda, right | 15:52 |
| jroll | heh | 15:52 |
| lucasagomes | yeah I think it's fair to have a related bug flag | 15:52 |
| lucasagomes | rloo, so the docimpact flag I still don't understand it much... I think yeah we can add | 15:53 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Rename discoverd -> inspector https://review.openstack.org/185499 | 15:53 |
| * devananda thinks we need some high level tracking of all the features / changes / refactoring that we discussed at the summit | 15:53 | |
| lucasagomes | but basically we don't have docs in OpenStack | 15:53 |
| lucasagomes | the docimpact flag creates a bug against the openstack docs | 15:53 |
| *** derekh has quit IRC | 15:53 | |
| lucasagomes | so it's basically creating a bug for soething that doesn't exist | 15:53 |
| rloo | lucasagomes: I *think*, if we had some document process :-), that the docimpact flag would be a way to make sure documents are updated to reflect any changes that ought to be documented. | 15:53 |
| devananda | So. re docs, I spoke with the doc team a couple times, including in one of their cross project sessions on this | 15:54 |
| rloo | lucasagomes: eg, so when liberty goes out, we could mention eg 'new fields option to specify ....' | 15:54 |
| devananda | they liked my suggestion and are moving forward with it | 15:54 |
| devananda | tldr; we'll keep our docs in our tree, but syndicate // publish them to the openstack-docs project | 15:54 |
| devananda | so that they get included in all the right places | 15:54 |
| lucasagomes | devananda, right, so re DocImpact flag. Do we need it? | 15:55 |
| * lucasagomes is not against it | 15:55 | |
| rloo | devananda: will someone (pshige?) document that process of syndicating/publishing... | 15:55 |
| devananda | doc-impact flag creates a bug in the doc project | 15:55 |
| devananda | rloo: when it's developed, yes :) | 15:55 |
| lucasagomes | right, yeah that's my concern | 15:56 |
| *** achanda has joined #openstack-ironic | 15:56 | |
| lucasagomes | cause we have been using that flag, but we don't actually have docs there | 15:56 |
| devananda | for a code change that requires us to update our docs, we should continue to, well, require the developer to make that change in the same patch series | 15:56 |
| lucasagomes | so saying it's not updated is like odd | 15:56 |
| *** achanda_ has quit IRC | 15:56 | |
| devananda | and for that, we dont need to use the docimpactflag | 15:56 |
| devananda | so probably we wont need it at all, but there could, in principle, be things that require it -- like if the docs team were to be maintaining some fancy graphics for us, and we change something that makes them need to rebuild it | 15:57 |
| lucasagomes | right, and for API the docs is auto generated | 15:57 |
| devananda | but that's a big IF, since they aren't doing that now | 15:57 |
| devananda | lucasagomes: well, yes, but the API docs are one of the things that the doc team needs to do s ome work to pull into their repos | 15:57 |
| lucasagomes | gotcha | 15:57 |
| devananda | brb | 15:58 |
| pshige | did you call me? | 15:59 |
| dtantsur | calling it a (loooong) day, g'night | 15:59 |
| *** eghobo has joined #openstack-ironic | 15:59 | |
| *** dtantsur is now known as dtantsur|afk | 16:00 | |
| pshige | gotcha | 16:00 |
| pshige | night! | 16:00 |
| lucasagomes | dtantsur|afk, night | 16:01 |
| *** rwsu has joined #openstack-ironic | 16:02 | |
| rloo | lucasagomes: not sure I understood when we should or shouldn't use docimpact, from that exchange. I like it cuz I can grep for it to see if there is anything "useful" to mention in eg the release notes. | 16:03 |
| NobodyCam | night dtantsur|afk | 16:03 |
| *** e0ne is now known as e0ne_ | 16:04 | |
| lucasagomes | rloo, IMHO we should never use it (as things currently are) | 16:04 |
| lucasagomes | cause we don't have docs in openstack, so it doesn't make sense to open a bug for something that doesn't exist saying we need to update it | 16:05 |
| lucasagomes | I mean, a bug against the openstack docs project | 16:05 |
| lucasagomes | we can have a bug in ironic saying we don't have documentation for X feature | 16:06 |
| lucasagomes | cause we are the owners of the docs | 16:06 |
| *** e0ne_ is now known as e0ne | 16:06 | |
| rloo | lucasagomes: fine with me. it seemed like a good habit to have, since it seemed to be used across openstack and i had hopes that we would have docs folks dealing with our docs. | 16:06 |
| *** krtaylor has quit IRC | 16:08 | |
| lucasagomes | rloo, yeah, that's very unique in OpenStack land I think... I think all other integrated projects does have docs | 16:08 |
| lucasagomes | even non integrated, I know zaqar have docs | 16:08 |
| lucasagomes | rloo, but I'm interested in knowing how this idea of keeping the docs in our tree will play out. Seems a good idea but I sometimes I doubt that engineers can actually right good documentation for end users | 16:10 |
| *** alexpilotti has joined #openstack-ironic | 16:10 | |
| lucasagomes | it requires some skills that I think we don't have (or at last I don't have it) | 16:10 |
| rloo | lucasagomes: I thought deva mentioned that some doc people would support us by reviewing the docs, but not sure. I agree with you, engineers are not doc writers. | 16:11 |
| lucasagomes | rloo, oh right yeah that would be great if they help us out | 16:12 |
| rloo | lucasagomes: we'll see! | 16:13 |
| *** zz_jgrimmn is now known as jgrimm | 16:13 | |
| *** romcheg has joined #openstack-ironic | 16:14 | |
| sambetts | rloo, lucasagomes: I have extended a page in this in the last cycle, it's a pretty nice format though not complete yet, I would love to see something similar for ironic http://docs.openstack.org/networking-guide/ | 16:15 |
| lucasagomes | sambetts, looks neat :-) | 16:16 |
| lucasagomes | is it rst? | 16:16 |
| sambetts | yup all rst | 16:17 |
| rloo | sambetts: nice. i think that's what I would like to see with our liaison with the docs folks; what 'holes' to fill in. | 16:17 |
| lucasagomes | sambetts, ah nice. I remember a while ago people having to deal with XML to update the docs | 16:19 |
| lucasagomes | RST sounds way better | 16:19 |
| cinerama | hi folks | 16:19 |
| *** romcheg has left #openstack-ironic | 16:19 | |
| sambetts | lucasagomes: it was the same in Neutron I think, the XML scared people off | 16:19 |
| pshige | basically docs people writes manuals and dev people give them information and triage bugs | 16:20 |
| lucasagomes | sambetts, nice, yeah I wish we had something like that for Ironic too | 16:21 |
| lucasagomes | looks really good | 16:21 |
| pshige | I will contact Karin Levenstein tomorrow | 16:21 |
| pshige | I'm now sleepy | 16:21 |
| rloo | pshige: thx. (whoever karin is :-)). Go to sleep! | 16:25 |
| devananda | lucasagomes: docs are still in XML / docbook format, but they're close to converting to RST. and yea, I agree that would be easier for us devs to contribute to | 16:25 |
| lucasagomes | devananda, +1 | 16:25 |
| lucasagomes | devananda, btw, re microversioning. When to bump it? | 16:25 |
| pshige | Karin leads to write Install guides | 16:25 |
| lucasagomes | is it for every change? Or to non-backward compat ones? | 16:25 |
| lucasagomes | I think it's for everychange due the "logical name" having a microversion but I want to confirm | 16:26 |
| * lucasagomes will have to update it's patch too | 16:26 | |
| rloo | thx pshige. | 16:26 |
| sambetts | Has the ironic mailer had anything sent on it recently, I don't seem to be getting any emails :/ | 16:27 |
| rloo | lucasagomes: I think it was meant for every change (based on nova spec) but I don't really see the value of it for every change. | 16:27 |
| devananda | sambetts: yes | 16:28 |
| devananda | rloo, lucasagomes: how about "required for any non-backwards-compatible change, and strongly encouraged for any significant features" ? (and yes, that's subjective) | 16:29 |
| rloo | devananda: I like that. | 16:29 |
| lucasagomes | devananda, right, I think it's good | 16:29 |
| lucasagomes | cause it makes sense to bump for non-backward compat | 16:29 |
| lucasagomes | but less sense to bump for everything | 16:29 |
| lucasagomes | so +1 | 16:30 |
| sambetts | hmm, the last email I received was the 20th of April :/ | 16:30 |
| NobodyCam | sambetts: humm thatz not good | 16:30 |
| devananda | sambetts: I sent a message yesterday re: no meeting last night | 16:30 |
| NobodyCam | oh didn't do the meeting recap | 16:31 |
| NobodyCam | are we keeping the split meetings? | 16:31 |
| rloo | if you have a feature that is backwards compatible, why would you want to version it vs not versioning it means you can get that feature w/o up'ing the version. | 16:31 |
| rloo | sambetts: yeah, there's email today too | 16:32 |
| rloo | sambetts: look what you missed: http://lists.openstack.org/pipermail/openstack-dev/2015-May/date.html | 16:33 |
| rloo | NobodyCam: we didn't get around to discussing the meetings. I'm assuming status quo? | 16:33 |
| jlvillal | lucasagomes: Thanks for updating the tox.ini patch for httpretty | 16:34 |
| lucasagomes | jlvillal, oh no problem | 16:34 |
| lucasagomes | jlvillal, thank you for fixing the gate with it :-) | 16:34 |
| NobodyCam | rloo: :) | 16:34 |
| NobodyCam | morning jlvillal | 16:34 |
| jlvillal | lucasagomes: My pleasure :) | 16:34 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: API: Get a subset of fields from Ports and Chassis https://review.openstack.org/183549 | 16:34 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: API: Get a subset of fields from Nodes https://review.openstack.org/183491 | 16:34 |
| sambetts | thank rloo, I don't know why I'm not getting the mails, I'm getting others, I recived 5 in the last 30 mins | 16:34 |
| jlvillal | NobodyCam: Morning. Just getting my system turned on. After a few reboots.... | 16:34 |
| rloo | sambetts: oh yeah. we blacklisted you cuz you weren't participating in ironic. just joking. | 16:35 |
| lucasagomes | lol | 16:35 |
| sambetts | rloo: ;) | 16:35 |
| lucasagomes | sambetts, check ur filters | 16:35 |
| * jlvillal Has to update to Fedora 22 now... | 16:35 | |
| devananda | NobodyCam: summarizing that poll is on my backlog | 16:35 |
| sambetts | lucasagomes: trying to find the page for that | 16:35 |
| lucasagomes | I remember I had some problems with case sensitiveness | 16:35 |
| lucasagomes | I was filtering [Ironic] (capitalized) but sometimes people uses [ironic] | 16:36 |
| lucasagomes | and so on... so | 16:36 |
| *** davideagnello has joined #openstack-ironic | 16:40 | |
| sambetts | lucasagomes: Yeah, I have all the email dump into a main folder if they dont get filters and there not in my ironic folder or that main one | 16:41 |
| *** trown is now known as trown|lunch | 16:41 | |
| *** e0ne has quit IRC | 16:45 | |
| *** achanda has quit IRC | 16:48 | |
| rloo | back in an hour or so | 16:49 |
| *** rloo has quit IRC | 16:49 | |
| *** ndipanov has quit IRC | 16:49 | |
| *** romcheg1 has joined #openstack-ironic | 16:49 | |
| *** romcheg1 has left #openstack-ironic | 16:49 | |
| * BadCub heads off for a 2 hour call | 16:56 | |
| *** dkehnx has quit IRC | 16:59 | |
| *** dkehn has joined #openstack-ironic | 16:59 | |
| jlvillal | sambetts: I just added a comment to the PEP8 bug you assigned to yourself. | 17:00 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: IPA: Do a soft reboot at the end of the deployment https://review.openstack.org/185667 | 17:00 |
| *** pas-ha has quit IRC | 17:00 | |
| * NobodyCam also joins the call of doom | 17:01 | |
| *** ifarkas has quit IRC | 17:02 | |
| *** mtanino has quit IRC | 17:02 | |
| jroll | lucasagomes: left a thing there | 17:02 |
| lucasagomes | jroll, oh, right check if we can access the result of the last command or something | 17:03 |
| lucasagomes | ? | 17:03 |
| sambetts | jlvillal: Thanks for the comments thats good info to have :) | 17:03 |
| lucasagomes | and consider a failure to access the node actually a success (it rebooted) | 17:03 |
| jlvillal | sambetts: You're welcome. | 17:03 |
| jroll | lucasagomes: yeah, maybe wait a few seconds and then check the result, if you can connect then it's a problem | 17:03 |
| jroll | lucasagomes: though hard to tell how long sync+shutdown will take | 17:04 |
| jroll | lucasagomes: and there could be a situation where systemd takes down IPA but doesn't actually reboot, I think | 17:04 |
| lucasagomes | jroll, ew oh, really? | 17:04 |
| lucasagomes | hmmm | 17:04 |
| jroll | I mean | 17:04 |
| jroll | sometimes the reboot command fails unexpectedly, right? | 17:05 |
| jroll | I've watched people reboot a whole cloud before, it can happen :) | 17:05 |
| lucasagomes | yeah it's possible | 17:05 |
| lucasagomes | jroll, I wil try to add some checks to that | 17:06 |
| *** harlowja has joined #openstack-ironic | 17:07 | |
| lucasagomes | but I don't know what would be the best way to check it yet tho | 17:07 |
| jroll | lucasagomes: cool. I also wonder if we need to mount in /proc here https://github.com/openstack/ironic-python-agent/blob/master/imagebuild/coreos/oem/cloud-config.yml#L59 | 17:07 |
| jroll | since we're in a container | 17:07 |
| jroll | I think it should be mounted in already though, or else /proc/cmdline wouldn't work, maybe nspawn muonts it in by default? | 17:07 |
| lucasagomes | I've tested it in a VM and the reboot worked... I can try some baremetal | 17:08 |
| lucasagomes | see if I can spot something | 17:08 |
| jroll | ok, cool, should be fine then | 17:08 |
| jroll | it *should* work in all cases, but everything fails sometimes | 17:08 |
| lucasagomes | yeah :-( | 17:09 |
| lucasagomes | wonders if I should make it optional as well | 17:09 |
| *** penick has joined #openstack-ironic | 17:09 | |
| lucasagomes | have a config to always do hardreboot | 17:09 |
| jroll | +1 | 17:09 |
| lucasagomes | but seems overkill because, DIB ramdisk have been rebooting itself for a long time and it seems to work | 17:09 |
| lucasagomes | jroll, right ok | 17:09 |
| lucasagomes | so, perhaps we leave hard reboot by default as-is today? | 17:10 |
| jroll | lucasagomes: maybe the option is overkill, idk | 17:10 |
| lucasagomes | and leave to operators to test if soft reboot works for them? | 17:10 |
| jroll | but there's probably hardware that wants a hard reboot out there :/ | 17:10 |
| jroll | well | 17:10 |
| jroll | I guess let's back up, what's the benefit of a soft reboot? | 17:10 |
| lucasagomes | it plays nice with UEFI | 17:10 |
| lucasagomes | and works with WOL :-) | 17:10 |
| lucasagomes | jroll, re uefi https://drive.google.com/file/d/0B5a_VAhX3Hz5WU1qN0FObkdSS00/view | 17:11 |
| jroll | heh | 17:11 |
| lucasagomes | that's ramesh screen | 17:11 |
| jroll | hm | 17:11 |
| jroll | so hard reboot can break uefi?? | 17:11 |
| lucasagomes | in some machines, apparently yes | 17:11 |
| jroll | wtf | 17:11 |
| jroll | that doesn't make sense at all :( | 17:12 |
| lucasagomes | it does, cause UEFI the boot stuff is not in the disk anymore | 17:12 |
| *** derekh_ has quit IRC | 17:12 | |
| lucasagomes | so sync filesystems may not be enough | 17:12 |
| lucasagomes | I wonder if there's a "sync"-like command for it | 17:12 |
| jroll | oh, right | 17:13 |
| lucasagomes | jroll, if you google, uefi hard reboot you will see there's problems with it | 17:13 |
| lucasagomes | but it seems a bit obscure | 17:14 |
| jroll | yeah | 17:15 |
| lucasagomes | jroll, making it optional seems the best way | 17:18 |
| lucasagomes | anyway I will call it a day the patch is marked as WIP until we get more info about it | 17:18 |
| lucasagomes | I'd like to get ramesh to test it out on the machines he saw the problem with UEFI | 17:19 |
| lucasagomes | have a good night everyone! | 17:20 |
| *** lucasagomes is now known as lucas-dinner | 17:23 | |
| *** davideagnello has quit IRC | 17:26 | |
| *** natorious is now known as zz_natorious | 17:28 | |
| *** davideagnello has joined #openstack-ironic | 17:29 | |
| jroll | lucas-dinner: sounds good, have a good night :) | 17:29 |
| *** spandhe has joined #openstack-ironic | 17:30 | |
| *** achanda has joined #openstack-ironic | 17:33 | |
| *** blakec has joined #openstack-ironic | 17:36 | |
| *** eghobo has quit IRC | 17:39 | |
| *** blakec has quit IRC | 17:41 | |
| *** HenryG has quit IRC | 17:42 | |
| *** alexpilotti has quit IRC | 17:42 | |
| *** eghobo has joined #openstack-ironic | 17:43 | |
| *** mtanino has joined #openstack-ironic | 17:44 | |
| *** HenryG has joined #openstack-ironic | 17:46 | |
| *** zz_natorious is now known as natorious | 17:49 | |
| *** trown|lunch is now known as trown | 17:52 | |
| *** blakec has joined #openstack-ironic | 17:57 | |
| *** HenryG has quit IRC | 18:01 | |
| *** HenryG has joined #openstack-ironic | 18:01 | |
| *** rloo has joined #openstack-ironic | 18:02 | |
| *** harlowja_ has joined #openstack-ironic | 18:06 | |
| *** harlowja has quit IRC | 18:06 | |
| *** mtanino is now known as mtanino_away | 18:08 | |
| *** dguerri`away is now known as dguerri | 18:08 | |
| *** davideagnello has quit IRC | 18:08 | |
| *** davideag_ has joined #openstack-ironic | 18:08 | |
| *** martini has joined #openstack-ironic | 18:09 | |
| martini | @TheJulia - where do I submit patches for bifrost? | 18:09 |
| NobodyCam | openstack/bifrost | 18:10 |
| *** cdearborn has quit IRC | 18:10 | |
| jroll | martini: typical openstack procedure, git review, gerrit, etc | 18:11 |
| jroll | it's an openstack project now | 18:11 |
| martini | 'typical openstack procedure' doesn't really help me :-) For pyghmi, I used review.openstack.org:29418/stackforge/pyghmi.git, and I can't find a comparable URL for bifrost | 18:13 |
| jroll | same thing, just /openstack.bifrost.git | 18:13 |
| jroll | err | 18:13 |
| jroll | /openstack/bifrost.git | 18:13 |
| NobodyCam | https://github.com/openstack/bifrost | 18:13 |
| jroll | or clone from github, and when you use 'git review' it will set up the review.o.o remote for you | 18:14 |
| martini | ahh, that helps. Is there a prototypical CONTRIBUTING.md for openstack projects? | 18:15 |
| jroll | maybe http://docs.openstack.org/infra/manual/developers.html#development-workflow ? | 18:16 |
| martini | I mean, do any other projects have an actual 'CONTRIBUTING.md' file in their repo, whether it links to / subsumes that or not? | 18:17 |
| NobodyCam | we have http://docs.openstack.org/developer/ironic/dev/contributing.html | 18:18 |
| martini | Ah, nice, I see the '.gitreview' now. Apparently I was *just* behind enough to not have it | 18:18 |
| jroll | yeah, ironic also has https://github.com/openstack/ironic/blob/master/CONTRIBUTING.rst | 18:19 |
| jroll | which is where I got that link :) | 18:19 |
| martini | jroll: Thanks! That's what I was hoping to see | 18:19 |
| jroll | :) | 18:19 |
| rloo | martini: maybe you could add something similar to bifrost :) | 18:20 |
| TheJulia | martini: ohh, awesome | 18:24 |
| martini | (Planning on it! That's why I asked ;-) | 18:25 |
| rloo | thx martini! | 18:26 |
| *** davideag_ has quit IRC | 18:31 | |
| *** achanda has quit IRC | 18:32 | |
| *** dprince has quit IRC | 18:32 | |
| *** davideagnello has joined #openstack-ironic | 18:33 | |
| blakec | jroll: regarding network-provider spec, has case of mapping of network_provider to node been considered? Similar to deploy driver, some nodes may be on different hardware with a switch (or SDN) requiring a different plugin. Obvious tradeoff between complexity and generality… thoughts? | 18:35 |
| *** jnielsen55 has joined #openstack-ironic | 18:37 | |
| *** krtaylor has joined #openstack-ironic | 18:38 | |
| *** davideagnello has quit IRC | 18:45 | |
| *** edcragg has quit IRC | 18:46 | |
| *** ukalifon2 has joined #openstack-ironic | 18:46 | |
| *** ukalifon1 has quit IRC | 18:46 | |
| *** logan2 has joined #openstack-ironic | 18:47 | |
| *** Haomeng has joined #openstack-ironic | 18:49 | |
| *** davideagnello has joined #openstack-ironic | 18:50 | |
| *** Haomeng|2 has quit IRC | 18:52 | |
| *** e0ne has joined #openstack-ironic | 18:53 | |
| *** ukalifon2 has quit IRC | 19:10 | |
| *** mtanino_away is now known as mtanino | 19:12 | |
| *** barra204 has joined #openstack-ironic | 19:16 | |
| *** cdearborn has joined #openstack-ironic | 19:34 | |
| *** blakec has quit IRC | 19:39 | |
| *** [1]cdearborn has joined #openstack-ironic | 19:48 | |
| *** Sukhdev has joined #openstack-ironic | 19:50 | |
| *** e0ne has quit IRC | 19:55 | |
| *** gridinv_ has joined #openstack-ironic | 19:57 | |
| *** e0ne has joined #openstack-ironic | 20:00 | |
| *** cdearborn has quit IRC | 20:02 | |
| jroll | TIL https://en.wikipedia.org/wiki//dev/full | 20:02 |
| jroll | lucas-dinner: ^ you'll like that | 20:02 |
| *** natorious is now known as zz_natorious | 20:05 | |
| trown | nice, TIL too | 20:05 |
| openstackgerrit | Merged openstack/python-ironicclient: Add node-show-states command https://review.openstack.org/183039 | 20:06 |
| jnielsen55 | quick question. With the current Juno or Kilo Ironic code is there a single compute node that acts as a proxy to the baremetal nodes? | 20:06 |
| jroll | jnielsen55: yes, the recommended set up is to use a single compute node | 20:08 |
| jroll | there's a clustered compute manager in our tree that allows for more, but race conditions abound | 20:08 |
| jnielsen55 | aweseom, thanks jroll | 20:09 |
| jnielsen55 | awesome | 20:09 |
| jroll | np :) | 20:09 |
| *** mtanino has quit IRC | 20:11 | |
| NobodyCam | call of doom is done | 20:11 |
| *** blakec has joined #openstack-ironic | 20:12 | |
| lucas-dinner | jroll, a-ha haha awesome! | 20:14 |
| lucas-dinner | I didn't know about /dev/full, neat | 20:14 |
| jroll | :D | 20:14 |
| *** mtanino has joined #openstack-ironic | 20:14 | |
| openstackgerrit | Radoslaw Smigielski proposed openstack/ironic: enable linux console redirection by default https://review.openstack.org/185729 | 20:18 |
| *** zz_natorious is now known as natorious | 20:22 | |
| *** gridinv_ has quit IRC | 20:36 | |
| *** blakec has quit IRC | 20:37 | |
| rloo | NobodyCam: you're still around? Then it wasn't all doom :-) | 20:43 |
| *** blakec has joined #openstack-ironic | 20:49 | |
| NobodyCam | rloo: heheheh | 20:52 |
| NobodyCam | just super long | 20:52 |
| rloo | NobodyCam: the advantage of attending remotely ;) | 20:52 |
| NobodyCam | YES!!!! | 20:52 |
| *** jgrimm is now known as zz_jgrimm | 20:53 | |
| jnielsen55 | one more quick question. I'm deploying Ironic on Juno. Does Juno use the iSCSI deploy method and not the ironic-python-agent? | 20:55 |
| jroll | IPA is supported on juno, if I remember correctly. However it does require swift. | 20:56 |
| *** natorious is now known as zz_natorious | 20:57 | |
| *** eghobo has quit IRC | 20:58 | |
| *** eghobo has joined #openstack-ironic | 21:00 | |
| *** eghobo_ has joined #openstack-ironic | 21:03 | |
| *** HenryG has quit IRC | 21:04 | |
| *** zz_natorious is now known as natorious | 21:04 | |
| *** natorious is now known as zz_natorious | 21:05 | |
| *** zz_natorious is now known as natorious | 21:06 | |
| *** eghobo has quit IRC | 21:07 | |
| *** HenryG has joined #openstack-ironic | 21:07 | |
| *** lucas-dinner has quit IRC | 21:10 | |
| *** jjohnson2 has quit IRC | 21:11 | |
| *** trown is now known as trown|outttypeww | 21:15 | |
| *** blakec has quit IRC | 21:18 | |
| NobodyCam | brb | 21:21 |
| Sukhdev | devananda: Ping | 21:27 |
| devananda | Sukhdev: pong | 21:27 |
| Sukhdev | devananda: I am looking to setup weekly meetings for Ironi/Neutron | 21:27 |
| *** radek__ has quit IRC | 21:27 | |
| *** blakec has joined #openstack-ironic | 21:27 | |
| Sukhdev | devananda: wanted to check with you if you have any preference to time slot/day | 21:28 |
| devananda | Sukhdev: heh, there's probably no time that I can commit to always make | 21:28 |
| devananda | that said, mondays tend to be meeting day, so piling more on is fine | 21:28 |
| Sukhdev | devananda: :-) | 21:29 |
| devananda | Sukhdev: it looks like neutorn l2 gateway meeting is 10am PT (same time as our weekly meetings, half of the time) | 21:29 |
| devananda | perhaps 11am PT ? | 21:29 |
| devananda | or 10am the alternate week, since it looks like your meeting is only every other week? | 21:30 |
| Sukhdev | devananda: let me check - If I find an available channel, will set it up | 21:30 |
| * devananda throws ideas at the wall | 21:30 | |
| jroll | devananda: y u kill my mondays | 21:30 |
| jroll | :) | 21:30 |
| devananda | jroll: get all the monday out of the way so the rest of the week is friday? ^_^ | 21:31 |
| NobodyCam | :-p | 21:31 |
| Sukhdev | jroll: I can move to Tuesday or Wednesday - if that works better for you guys - | 21:31 |
| Sukhdev | that is the reason to ping before I set something up - | 21:31 |
| jroll | Sukhdev: monday is typically fine, don't mind my sarcasm :) | 21:31 |
| Sukhdev | jroll: ha ha…. cool then.. let me see if I can find an available channel | 21:32 |
| clif_h | maybe someone can fill me in: are the py34 tests broken? | 21:33 |
| clif_h | or am I doing something wrong locally? | 21:33 |
| jroll | someone else had problems this morning | 21:34 |
| jroll | clif_h: tox -r ? | 21:34 |
| NobodyCam | clif_h: I know jlvillal has done a lot of work on our py3 tests | 21:35 |
| Sukhdev | devananda, jroll: mondays 11am PT is available - I am getting ready to grab that slot and will send out the email on ML - just a heads up | 21:35 |
| jroll | Sukhdev: great, ty, will be there | 21:35 |
| jlvillal | clif_h: What kind of error are you seeing? I've done a little work in there. | 21:36 |
| jlvillal | clif_h: And 'tox -r' is a good idea. | 21:36 |
| clif_h | I'm pretty sure this was my first run for 'py34' | 21:38 |
| devananda | jroll, Sukhdev: so I'll be landing in vegas at 9:15am next monday. will try to make the meeting, but there's a chance I'll either miss part of it, or be taking it from a cafe in the airport :p | 21:38 |
| jlvillal | clif_h: Error message? | 21:39 |
| clif_h | jlvillal: https://gist.github.com/ClifHouck/b6e5f2fa705e4adec5a2 | 21:39 |
| jroll | devananda: so, yay for you not being a spof for this? :P | 21:39 |
| devananda | jroll: exactly | 21:40 |
| Sukhdev | devananda: oh I see... | 21:40 |
| jroll | clif_h: oh, blow away .testrepository files | 21:40 |
| devananda | Sukhdev: one downside to that time is it makes it hard for the HP Cloud folks working on integration of our switches | 21:40 |
| devananda | Sukhdev: they're in bangalore, 12hr offset to PT | 21:40 |
| jroll | devananda: I'd argue this meeting should track the non-vendor-specific bits of this (though vendors in these talks could be useful) | 21:41 |
| Sukhdev | devananda: Oh - in that case, we should move to earlier slot - L2GW meeting used to be at 11PT, I moved to earlier slot to accomodate them - this will cause an issue | 21:41 |
| Sukhdev | devananda: I hate to pull west coast guys into meetings on Monday 9am :-) | 21:42 |
| Sukhdev | devananda: How about Tuesday 9am? will that work better? | 21:42 |
| devananda | Sukhdev: there are already 5 meetings in that slot | 21:43 |
| Sukhdev | devananda: yikes!! how about Thursday? | 21:44 |
| *** mtanino has quit IRC | 21:44 | |
| clif_h | jroll: thanks, that seems to have worked | 21:44 |
| jroll | np | 21:44 |
| Sukhdev | unless you believe monday 9am is OK? | 21:44 |
| clif_h | will I have to remove that every time I switch between py27, and py34? | 21:45 |
| devananda | 9am works for me. i'm up well before that | 21:45 |
| devananda | (usually) | 21:45 |
| jlvillal | clif_h: I ran in to that before, I think. I fixed it once and so far it has worked. | 21:45 |
| devananda | jroll: ? | 21:45 |
| jlvillal | clif_h: I remember deleting .testrepository files | 21:46 |
| *** [1]cdearborn has quit IRC | 21:46 | |
| jroll | clif_h: I'm not really sure... I have a feeling when you installed py34 that became your default python for running system tox | 21:46 |
| jroll | devananda: ? | 21:46 |
| *** natorious is now known as zz_natorious | 21:46 | |
| jroll | devananda: the 'np' was for clif if that's your question | 21:46 |
| clif_h | env tox claims its 2.7 | 21:46 |
| jroll | weird, idk then | 21:47 |
| clif_h | I should say 'tox --version' | 21:47 |
| Sukhdev | jroll: I think devananda was asking if 9AM monday works for you? | 21:47 |
| devananda | 9am monday lets us at least get us + eu folks | 21:47 |
| jroll | Sukhdev: oh, I'm usually online by 7 or 8, anything after that is fine with me :) | 21:47 |
| devananda | though i'll miss half of the first meeting for sure, because travel | 21:48 |
| devananda | but that's fine | 21:48 |
| Sukhdev | devananda, jroll: cool - so, I will set it up for Mondays at 9am | 21:48 |
| Sukhdev | devananda: It would have been better, if you were present in the opening meeting :-):-) | 21:48 |
| jroll | devananda: I have a feeling the first meeting will be around goals/expectations/etc, it would be cool if we can finalize that early so you can email in opinions or something | 21:49 |
| Sukhdev | jroll: correct - | 21:49 |
| devananda | maybe the three of us +bertie meet on friday to set the goals, then you two present & discuss on monday | 21:50 |
| Sukhdev | jroll devananda: I will type up the preliminary agenda and let you modify | 21:50 |
| *** Sukhdev has quit IRC | 21:50 | |
| *** eghobo_ has quit IRC | 21:50 | |
| *** Sukhdev has joined #openstack-ironic | 21:51 | |
| jroll | subscope: sounds good, thanks | 21:51 |
| jroll | Sukhdev rather | 21:51 |
| devananda | I just dont want to postpone the starting meeting a week because I'm travelling | 21:51 |
| Sukhdev | devananda: sorry - got disconnected - am back now | 21:51 |
| devananda | Sukhdev: thanks | 21:51 |
| Sukhdev | devananda: Friday will be fine | 21:52 |
| Sukhdev | devananda: let me set up the meeting, set up the agenda and then we can meet on Friday - the rest of the team can meet Monday mornig | 21:53 |
| Sukhdev | devananda, jroll : I am moving into a meeting - will get disconnected again - | 21:54 |
| Sukhdev | devananda jroll: I will find (hopefully) an available channel for Mondays 9am, will send out the ML invite and you let me know what time would you want three of us meet on Friday? | 21:55 |
| jroll | Sukhdev: as of now, my friday is 100% open :) | 21:55 |
| Sukhdev | jroll, devananda: Friday 9am? | 21:56 |
| jroll | that's fine | 21:56 |
| devananda | Sukhdev: WFM | 21:58 |
| devananda | pls send gcal invite | 21:58 |
| jroll | ^^ | 21:58 |
| *** zz_natorious is now known as natorious | 22:00 | |
| *** natorious is now known as zz_natorious | 22:01 | |
| *** zz_natorious is now known as natorious | 22:01 | |
| *** eghobo has joined #openstack-ironic | 22:01 | |
| *** Sukhdev has quit IRC | 22:02 | |
| *** eghobo_ has joined #openstack-ironic | 22:03 | |
| *** Sukhdev has joined #openstack-ironic | 22:04 | |
| *** bandicot has joined #openstack-ironic | 22:05 | |
| *** eghobo has quit IRC | 22:06 | |
| mrda | Morning Ironic | 22:08 |
| *** bandicot has quit IRC | 22:08 | |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: WIP: Cache negotiated api microversion for server https://review.openstack.org/173674 | 22:14 |
| *** enikanorov_ has joined #openstack-ironic | 22:15 | |
| NobodyCam | morning mrda | 22:16 |
| mrda | o/ | 22:17 |
| BadCub | morning mrda :) | 22:18 |
| mrda | \o | 22:18 |
| *** bandicot has joined #openstack-ironic | 22:19 | |
| TheJulia | Good morning mrda | 22:19 |
| mrda | \o/ | 22:19 |
| *** gridinv_ has joined #openstack-ironic | 22:20 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Add 'simplegeneric' to Python 3.4 test-requirements https://review.openstack.org/185759 | 22:22 |
| cinerama | hi mrda | 22:23 |
| mrda | hey cinerama (I've run out of arms :) | 22:23 |
| jroll | morning mrda /o\ | 22:24 |
| jroll | :P | 22:24 |
| mrda | oh yeah, that could've worked. | 22:25 |
| jroll | and hiya cinerama! | 22:25 |
| mrda | \o\ | 22:25 |
| cinerama | everyone feeling reasonably well? | 22:26 |
| * NobodyCam is quite tired today... | 22:33 | |
| *** lifeless has quit IRC | 22:35 | |
| *** krtaylor has quit IRC | 22:37 | |
| *** gridinv_ has quit IRC | 22:37 | |
| mordred | mrda, devananda: btw ... | 22:37 |
| mordred | (moving here, since it's a different topic) | 22:37 |
| mrda | cinerama: pretending that the sumptoms I'm feeling won't become a cold | 22:37 |
| devananda | mrda: tldr; ironicclient generates a new keystone token for each request, or requires the caller to pass in a (valid) token object | 22:38 |
| mordred | we're using dogpile.cache in shade and are pretty happy about it ... but I mention that because we're also moving os-client-config support into ksa | 22:38 |
| devananda | mrda: if that token is bad, iornicclient just errors and its all teh callers problem | 22:38 |
| mrda | symptoms | 22:38 |
| mordred | and os-client-config has support for reading cache settings from the clouds.yaml file | 22:38 |
| mordred | SO | 22:38 |
| mordred | if you did a similar thing, then you could consume the same cache config settings through teh os-client-config/ksa library | 22:38 |
| mrda | mordred: ok, shall take a look | 22:38 |
| devananda | mrda: instead, IIUC, we should use keystone SESSION objects, which handle re-validating themselves when needed | 22:38 |
| mordred | mrda: you wind up needing to do a little bit of work to get it going in the first place - but SpamapS has done all of that in shade, so feel free to either copy that, or poke SpamapS - or something | 22:39 |
| mrda | mordred: ok, shall do | 22:40 |
| devananda | and then, in the case where someone passes in creds instead of a session, we could cache the session in memory (not disk) for the duration of the calling process | 22:40 |
| devananda | would help the nova driver considerably, and any other process (eg, shade, bifrost...) that uses the python lib | 22:40 |
| mordred | ++ | 22:40 |
| jroll | the nova driver already caches the client object, fwiw | 22:40 |
| jroll | this would just remove the 401/403 from the "retry these" list | 22:41 |
| devananda | jroll: yes, but the client object is generating a new keystone TOKEN for each request | 22:41 |
| * mrda is still waking up, hence the short replys. But will re-read after coffee | 22:41 | |
| jroll | (which is good, to be clear) | 22:41 |
| devananda | I mean, it shouldn't be, but the logs indicate it is | 22:41 |
| jroll | devananda: no, we cache the client object, which has a token | 22:41 |
| devananda | hrmm | 22:41 |
| jroll | at least it works in my cloud (tm) | 22:41 |
| jroll | afaict | 22:41 |
| jroll | I could be wrong but I believe clif_h confirmed this with data | 22:42 |
| jroll | which should be trivial to reproduce | 22:42 |
| *** romcheg has joined #openstack-ironic | 22:43 | |
| * devananda looks at recent gate logs to confirm/deny his assertion | 22:43 | |
| clif_h | it should cache it | 22:43 |
| *** ijw has joined #openstack-ironic | 22:44 | |
| clif_h | possibly it caches the client but the client is still getting a token every time? | 22:44 |
| jroll | gate logs do look like a lot of token requests :/ | 22:44 |
| jroll | looking at http://logs.openstack.org/49/183549/4/check/check-tempest-dsvm-ironic-pxe_ssh/f92ecc0/logs/apache/keystone_access.txt.gz | 22:44 |
| mrda | clif_h: IIRC the object is being cached | 22:44 |
| jroll | clif_h: based on the results in our cloud, though, it seems to be caching the token, otherwise we'd still be having issues with it | 22:44 |
| clif_h | maybe there's a config setting | 22:45 |
| jroll | 249 token requests in that log | 22:45 |
| jroll | for one tempest run | 22:45 |
| devananda | might be tempest's client, not nova's use of our client? | 22:45 |
| *** romcheg has quit IRC | 22:45 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Add 'simplegeneric' to Python 3.4 test-requirements https://review.openstack.org/185759 | 22:46 |
| jroll | right, hard to tell | 22:46 |
| * devananda is reading from the tail of http://logs.openstack.org/80/180680/3/check/check-tempest-dsvm-ironic-pxe_ssh/e374f82/logs/screen-n-cpu.txt.gz backwards | 22:46 | |
| jroll | actually, I think we're good | 22:47 |
| jroll | $ curl http://logs.openstack.org/49/183549/4/check/check-tempest-dsvm-ironic-pxe_ssh/f92ecc0/logs/screen-n-cpu.txt.gz | grep 'v2.0/tokens' | wc -l | 22:47 |
| jroll | 2 | 22:47 |
| devananda | one for the ironic token one for the neutron token | 22:48 |
| devananda | yea, looks good | 22:48 |
| * devananda stands correcte | 22:48 | |
| devananda | d | 22:48 |
| jroll | \o/ | 22:48 |
| jroll | great work clif_h :P | 22:49 |
| devananda | so a more accurate tldr; is that we implemented caching and retry in a wrapper class within our nova driver | 22:49 |
| devananda | and that doesn't help any other users of the client | 22:50 |
| devananda | which makes mordred sad | 22:50 |
| mordred | yes | 22:50 |
| mordred | it does make me sad | 22:50 |
| mordred | although | 22:50 |
| mordred | to be fair | 22:50 |
| mordred | I would likely not use your caching | 22:50 |
| *** davideagnello has quit IRC | 22:50 | |
| mordred | since I have a pervassive caching system myself | 22:50 |
| devananda | mordred: if only you didn't use our client ;) | 22:50 |
| mordred | :) | 22:50 |
| *** natorious is now known as zz_natorious | 22:50 | |
| mordred | devananda: actually - we should figure out what it looks like for us to use your caching but to tie it in to our caching | 22:51 |
| *** davideagnello has joined #openstack-ironic | 22:51 | |
| mordred | devananda: because for shade we have multi-cloud stuff sorted in terms of cache keys/collisions | 22:51 |
| mordred | this is almost certainly not important for ironic | 22:51 |
| mordred | BUT | 22:51 |
| mordred | if we figure out the interface between the two of us, it'll likely cover all the use cases known to man | 22:52 |
| jroll | devananda: right, I agree, however we still need most of that logic to retry for a failed token for other variants of retries :) | 22:52 |
| mordred | well, no | 22:52 |
| mordred | you do not need to do that at all | 22:52 |
| jroll | mmm? | 22:52 |
| mordred | once you're on keystone session, session takes care of all of that | 22:52 |
| jroll | we retry on 409 because our locking is insane | 22:52 |
| jroll | etc | 22:52 |
| devananda | mordred: https://github.com/openstack/nova/blob/master/nova/virt/ironic/client_wrapper.py#L59 | 22:52 |
| mordred | it's one of the main reasons to use keystone session | 22:52 |
| jroll | does session do that? | 22:52 |
| mordred | jamielennox: ^^ | 22:52 |
| mordred | jroll: basically, you should never need to know about it | 22:53 |
| mordred | it should be transparent | 22:53 |
| mordred | that's what session gives you | 22:53 |
| jamielennox | hm? | 22:53 |
| devananda | i'd be happy if we replaced that code with sessions | 22:53 |
| mordred | jamielennox: more session discussion | 22:53 |
| jroll | mordred: specifically, can I configure this for KS session? https://github.com/openstack/nova/blob/master/nova/virt/ironic/client_wrapper.py#L113-115 | 22:53 |
| jroll | and if so that's awesome | 22:53 |
| *** e0ne has quit IRC | 22:53 | |
| mordred | jroll: yah - actually, your code doesn't take in to account keystone auth plugins either | 22:53 |
| jamielennox | i think it will only automatically retry 401 | 22:54 |
| jamielennox | because it fetches a new token and re-issues | 22:54 |
| jroll | ^ right, that was my assumption | 22:54 |
| devananda | yea, so we'd need to ALSO retry on 409 | 22:54 |
| mordred | yah. but you do not need to retry on expired token | 22:54 |
| jroll | right. and so my point is that we'll still need most of this logic | 22:55 |
| jamielennox | yea, we might be able to expose a way to add that, but i don't want to add 409 handling by default | 22:55 |
| mordred | gotcha | 22:55 |
| jroll | that isn't to say we shouldn't do it | 22:55 |
| mordred | ++ | 22:55 |
| jroll | but "remove this code" isn't a win here | 22:55 |
| jroll | jamielennox: that would be useful, and I agree with not by default :) | 22:55 |
| jamielennox | shouldn't this logic be ironic client? | 22:55 |
| jamielennox | why is ironic having to wrap its own client? | 22:56 |
| devananda | yea, it basically removes https://github.com/openstack/nova/blob/master/nova/virt/ironic/client_wrapper.py#L123-L129 | 22:56 |
| devananda | jamielennox: yes | 22:56 |
| jamielennox | oh sorry - nova | 22:56 |
| jroll | jamielennox: idk, as a cli user I'd be okay with it bailing in those cases | 22:56 |
| jroll | but I want to be able to restart ironic-api without interrupting nova activity, etc | 22:56 |
| jroll | or not fail a build because stray lock produces a 409 | 22:56 |
| *** dguerri is now known as dguerri`away | 22:57 | |
| jlvillal | jroll: If you have time to look at the changes I made to your patch, it would be appreciated. I believe I only changed the unittests. https://review.openstack.org/130877 | 22:58 |
| jlvillal | jroll: If you think it is good, I was going to add it to the etherpad to be reviewed. | 22:58 |
| jroll | jlvillal: oh ya, will do | 22:59 |
| jlvillal | jroll: Thanks | 22:59 |
| jroll | jlvillal: though I tend not to worry too much about the tests, so long as they work :P | 22:59 |
| jlvillal | :) | 22:59 |
| *** yuanying has joined #openstack-ironic | 22:59 | |
| jroll | jlvillal: also, if you need someone to bug on that, alaski is your dude | 23:00 |
| jlvillal | jroll: Excellent. Thanks. mrda and I have a discussion on Nova bugs right about now. | 23:00 |
| jroll | jlvillal: cool. he's the goto for cells stuff and helped me dig on that downstream iirc | 23:01 |
| mordred | TheJulia, jamielennox, devananda: so - was just talking about the special casing in shade for ironic and non-keystone auth | 23:01 |
| jamielennox | mordred: https://github.com/openstack/python-keystoneclient/blob/master/keystoneclient/auth/token_endpoint.py | 23:01 |
| *** eghobo_ has quit IRC | 23:01 | |
| mordred | jamielennox: so just use that with a token of None or something? | 23:02 |
| *** eghobo has joined #openstack-ironic | 23:02 | |
| jamielennox | mordred: you'd need to test but it should work | 23:02 |
| mordred | (problem we're trying to solve - how to _just_ use the Session code path for ironicclients and still support both ks and non-ks auth) | 23:02 |
| mordred | jamielennox: cool | 23:02 |
| mordred | morganfainberg: ^^ | 23:03 |
| jamielennox | there is a failure case there for auth to indicate that it didn't work it returns None, so it might have to be the empty string or something | 23:03 |
| morganfainberg | oh hai! | 23:03 |
| devananda | hi there! | 23:03 |
| jamielennox | but assumedly if you're not using keystone auth you could enter some arbitrary string for the token value and it wouldn't care | 23:03 |
| devananda | right now, yea, it works with any string | 23:04 |
| devananda | and our API ignores it, basically | 23:04 |
| morganfainberg | jamielennox: or no string. | 23:05 |
| morganfainberg | devananda: thats good. | 23:05 |
| NobodyCam | the readme calls out OS_AUTH_TOKEN - Any value, such as an empty space, is required to cause the client library to send requests directly to the API. | 23:05 |
| morganfainberg | NobodyCam: hm. ok | 23:05 |
| devananda | let's say I want to use ironic under apache auth? (works fine today) what if I want to use ironic with my LDAP server? ... (also, personally, I *dont* wnt to do that, just to be clear) | 23:05 |
| morganfainberg | devananda: lets not build in ldap support in ironic | 23:06 |
| jroll | devananda: I hate to tell you that someone may wish to do that :( | 23:06 |
| morganfainberg | devananda: mod_auth_ldap should suffice for that. | 23:06 |
| devananda | jroll: i'm fine with enabling someone else to do that | 23:06 |
| morganfainberg | devananda: do the same thing we're doing, rely on apache's good support for that kind of thing | 23:06 |
| * jroll forgets if nova did that within nova or some other way | 23:06 | |
| devananda | I should have said "*I* dont want to ..." | 23:06 |
| devananda | morganfainberg: works for me | 23:06 |
| morganfainberg | nothing stops them from layering apache mods in front of ironic | 23:07 |
| devananda | nope | 23:07 |
| jroll | devananda: I'm more sad at the possibility of needing to do that myself, apache or not | 23:07 |
| *** ijw has quit IRC | 23:07 | |
| morganfainberg | devananda: I also would advocate that RBAC [even limited] should be when deployed with keystone [to keep your life simple] | 23:07 |
| morganfainberg | but that is a personal preferance | 23:07 |
| morganfainberg | jroll: ^ | 23:07 |
| jroll | morganfainberg: yep, agree. my downstream is leaking here, feel free to ignore me :) | 23:08 |
| *** ijw has joined #openstack-ironic | 23:08 | |
| morganfainberg | jroll: if there is a good reason to do something else, I'm happy to help advise/entertain how we can get there. | 23:08 |
| * BadCub hands jroll a mop to clean up his leaking downstream | 23:08 | |
| morganfainberg | jroll: but if we can avoid extra insanity i'm happy | 23:09 |
| morganfainberg | BadCub: uh.. you probably want to make sure jroll keeps that mop afterwards. | 23:09 |
| morganfainberg | :P | 23:09 |
| BadCub | lol | 23:09 |
| *** chlong has joined #openstack-ironic | 23:09 | |
| jroll | lol | 23:09 |
| BadCub | morganfainberg: yeah.. No return policy on mops! | 23:09 |
| devananda | on that note, I gotta run | 23:10 |
| BadCub | see ya devananda lol | 23:10 |
| * devananda is careful not to split on the spilled downstream | 23:10 | |
| jroll | morganfainberg: I actually never considered mod_ldap for this bit, I'll probably poke at that | 23:10 |
| NobodyCam | lol | 23:10 |
| devananda | s/split/slip/ | 23:10 |
| NobodyCam | night devananda | 23:10 |
| morganfainberg | jroll: ^_^ | 23:11 |
| jroll | thanks :) | 23:11 |
| morganfainberg | jroll: i'm trying really really hard to keep us from re-inventing wheels around here | 23:11 |
| jroll | morganfainberg: yep, hear you there. | 23:11 |
| morganfainberg | (it's why i lurk everywehre) trying to make sure I can head things off before they become locked in. | 23:12 |
| *** Sukhdev has quit IRC | 23:13 | |
| jroll | +1, I try to do the same :P | 23:13 |
| *** chlong has quit IRC | 23:15 | |
| openstackgerrit | Clif Houck proposed openstack/ironic: Add support for image caching for the agent driver https://review.openstack.org/161832 | 23:19 |
| *** Sukhdev has joined #openstack-ironic | 23:20 | |
| *** hemna is now known as hemnafk | 23:31 | |
| *** zz_natorious is now known as natorious | 23:32 | |
| *** natorious is now known as zz_natorious | 23:33 | |
| *** Haomeng|2 has joined #openstack-ironic | 23:37 | |
| *** Haomeng has quit IRC | 23:40 | |
| * NobodyCam calls it a day.... G'Night ya all... | 23:41 | |
| *** jamielennox is now known as jamielennox|away | 23:42 | |
| *** eghobo has quit IRC | 23:43 | |
| *** chlong has joined #openstack-ironic | 23:44 | |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Add ability to write CSV file https://review.openstack.org/185786 | 23:47 |
| *** lifeless has joined #openstack-ironic | 23:54 | |
| *** Sukhdev has quit IRC | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!