| Haomeng | devananda:ping | 00:12 |
|---|---|---|
| NobodyCam | Haomeng: he is flying today | 00:18 |
| Haomeng | NobodyCam, ok thanks, I think he is going toHK now:) | 00:23 |
| Haomeng | just one concern, based on devananda comments, we can run some customized sql to do some special query, | 00:24 |
| Haomeng | this is the comments "basically must be implemented at DB layer: UPDATE .. SET col=value WHERE node.id=xxx AND node.instance_uuid=NULL; if rows_affected != 1: raise" | 00:24 |
| Haomeng | do you have idea if our sqlalchemy support such special sql query? | 00:24 |
| Haomeng | NobodyCam, I found the fix byhim - http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/baremetal/db/sqlalchemy/api.py#n190, so thankyou:) | 00:26 |
| Haomeng | no question now:) | 00:27 |
| *** matsuhashi has joined #openstack-ironic | 00:29 | |
| NobodyCam | nope not hk yet | 00:34 |
| NobodyCam | :) | 00:35 |
| Haomeng | :) | 00:40 |
| *** nosnos has joined #openstack-ironic | 01:05 | |
| *** matsuhashi has quit IRC | 01:11 | |
| *** matsuhashi has joined #openstack-ironic | 01:12 | |
| *** sjing has joined #openstack-ironic | 01:19 | |
| sjing | Morning, Ironic | 01:42 |
| Haomeng | Morning Jing:) | 01:43 |
| sjing | Hi Hao Meng | 01:48 |
| Haomeng | :) | 01:50 |
| *** lnxnut_ has joined #openstack-ironic | 02:01 | |
| *** lnxnut has quit IRC | 02:01 | |
| *** lnxnut_ has quit IRC | 02:03 | |
| *** michchap has joined #openstack-ironic | 02:29 | |
| *** rloo has quit IRC | 02:53 | |
| *** kobier has joined #openstack-ironic | 03:22 | |
| *** crandquist has quit IRC | 03:26 | |
| *** prekarat has joined #openstack-ironic | 04:00 | |
| *** matsuhashi has quit IRC | 04:12 | |
| *** matsuhashi has joined #openstack-ironic | 04:13 | |
| *** matsuhashi has quit IRC | 04:17 | |
| *** nosnos has quit IRC | 04:17 | |
| *** nosnos_ has joined #openstack-ironic | 04:17 | |
| *** nosnos_ has quit IRC | 04:22 | |
| *** nosnos has joined #openstack-ironic | 04:23 | |
| *** nosnos has quit IRC | 04:26 | |
| *** nosnos has joined #openstack-ironic | 04:29 | |
| *** nosnos_ has joined #openstack-ironic | 04:33 | |
| *** nosnos has quit IRC | 04:37 | |
| *** matsuhashi has joined #openstack-ironic | 05:04 | |
| *** slogneo has joined #openstack-ironic | 05:23 | |
| *** nosnos_ has quit IRC | 05:28 | |
| *** nosnos has joined #openstack-ironic | 05:28 | |
| *** anniec has joined #openstack-ironic | 05:30 | |
| *** anniec has quit IRC | 05:31 | |
| *** anniec has joined #openstack-ironic | 05:39 | |
| *** anniec has left #openstack-ironic | 05:39 | |
| *** slogneo has quit IRC | 05:42 | |
| *** anniec_ has joined #openstack-ironic | 05:55 | |
| *** nosnos_ has joined #openstack-ironic | 06:06 | |
| *** nosnos has quit IRC | 06:08 | |
| *** michchap has quit IRC | 06:10 | |
| *** michchap has joined #openstack-ironic | 06:13 | |
| *** michchap has quit IRC | 06:13 | |
| *** michchap has joined #openstack-ironic | 06:14 | |
| *** martyntaylor has joined #openstack-ironic | 06:33 | |
| *** jianingy has joined #openstack-ironic | 06:44 | |
| *** nosnos_ has quit IRC | 07:22 | |
| *** nosnos has joined #openstack-ironic | 07:22 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ironic: Change Port create API to EAFP https://review.openstack.org/54537 | 07:40 |
| *** rushiagr has joined #openstack-ironic | 07:45 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ironic: Change Port create API to EAFP https://review.openstack.org/54537 | 07:46 |
| *** bauzas has joined #openstack-ironic | 07:50 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ironic: Change Port create API to EAFP https://review.openstack.org/54537 | 07:55 |
| *** prekarat has left #openstack-ironic | 08:04 | |
| *** prekarat has joined #openstack-ironic | 08:04 | |
| *** jianingy is now known as jianingy_afk | 08:05 | |
| *** anniec_ has quit IRC | 08:06 | |
| *** blamar_ has joined #openstack-ironic | 08:11 | |
| *** blamar has quit IRC | 08:12 | |
| *** blamar_ is now known as blamar | 08:12 | |
| *** matsuhashi has quit IRC | 08:14 | |
| *** matsuhas_ has joined #openstack-ironic | 08:17 | |
| *** jistr has joined #openstack-ironic | 08:28 | |
| *** sjing has quit IRC | 08:40 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ironic: Fix i18N compliance https://review.openstack.org/54687 | 08:47 |
| *** matsuhas_ has quit IRC | 08:54 | |
| *** matsuhashi has joined #openstack-ironic | 08:55 | |
| *** matsuhashi has quit IRC | 08:59 | |
| *** rushiagr has quit IRC | 09:02 | |
| *** matsuhashi has joined #openstack-ironic | 09:02 | |
| *** rushiagr has joined #openstack-ironic | 09:03 | |
| *** prekarat has quit IRC | 09:04 | |
| *** rushiagr has left #openstack-ironic | 09:11 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Fix i18N compliance https://review.openstack.org/54689 | 09:12 |
| *** prekarat has joined #openstack-ironic | 09:19 | |
| *** lucasagomes has joined #openstack-ironic | 09:22 | |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/python-ironicclient: Fix i18N compliance https://review.openstack.org/54689 | 09:27 |
| *** matsuhashi has quit IRC | 09:28 | |
| *** matsuhashi has joined #openstack-ironic | 09:29 | |
| *** derekh has joined #openstack-ironic | 09:30 | |
| *** matsuhashi has quit IRC | 09:33 | |
| *** matsuhashi has joined #openstack-ironic | 09:42 | |
| *** ndipanov_gone is now known as ndipanov | 09:46 | |
| *** matsuhashi has quit IRC | 10:04 | |
| *** matsuhashi has joined #openstack-ironic | 10:08 | |
| *** matsuhashi has quit IRC | 10:12 | |
| *** lucasagomes has quit IRC | 10:19 | |
| *** derekh has quit IRC | 10:20 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add driver interfaces validation https://review.openstack.org/54593 | 10:22 |
| *** lucasagomes has joined #openstack-ironic | 10:22 | |
| *** matsuhashi has joined #openstack-ironic | 10:26 | |
| *** derekh has joined #openstack-ironic | 10:35 | |
| *** matsuhashi has quit IRC | 10:46 | |
| *** matsuhashi has joined #openstack-ironic | 10:47 | |
| *** matsuhashi has quit IRC | 10:51 | |
| *** nosnos has quit IRC | 10:51 | |
| *** romcheg has joined #openstack-ironic | 10:53 | |
| romcheg | Hi folks | 10:53 |
| romcheg | lucasagomes: Are you around? | 10:54 |
| Haomeng | how are you romcheg:) | 10:54 |
| romcheg | Haomeng: I'm quite good, thanks! | 10:54 |
| romcheg | Yourself? | 10:54 |
| Haomeng | yes, i am fine:) | 10:55 |
| Haomeng | I think so:) | 11:01 |
| Haomeng | they will be prepare to HK:) | 11:01 |
| Haomeng | romcheg, will you attend HK summit? | 11:02 |
| romcheg | Haomeng: yes, I will | 11:02 |
| Haomeng | cool:) | 11:02 |
| Haomeng | will fly on next mon? | 11:03 |
| romcheg | No, I have a flight on this Saturday morning so I arrive to HK Sunday morning | 11:04 |
| romcheg | Ukraine is very far so I need a day to deal with the jet lag :) | 11:04 |
| Haomeng | :) | 11:05 |
| Haomeng | take care, and enjoy the journey:) | 11:06 |
| romcheg | Thanks | 11:06 |
| Haomeng | welcome you to China:) | 11:06 |
| Haomeng | :) | 11:06 |
| romcheg | How long does it take you to get to HK? | 11:06 |
| Haomeng | about 3.5 hrs from Beijing To HK | 11:08 |
| Haomeng | but I willnot attend:) | 11:08 |
| romcheg | Ahh | 11:08 |
| Haomeng | else will gudie all you friends on HK for shopping or something:) | 11:08 |
| Haomeng | :) | 11:08 |
| romcheg | Do people speak/understand English in HK in general? | 11:09 |
| Haomeng | yes,almostly | 11:11 |
| Haomeng | HK is Internationali city | 11:12 |
| Haomeng | almost people can say some basic English:) | 11:12 |
| Haomeng | so dont worry about | 11:12 |
| *** anniec has joined #openstack-ironic | 11:16 | |
| romcheg | Found a bug: https://bugs.launchpad.net/ironic/+bug/1246680 | 11:27 |
| Haomeng | this should be i18nissue | 11:29 |
| Haomeng | looks likenow it does not support non-english chars | 11:29 |
| Haomeng | can you paste the details about the post api call arguments | 11:30 |
| *** michchap has quit IRC | 11:31 | |
| *** matsuhashi has joined #openstack-ironic | 11:31 | |
| Haomeng | you just want ton input unicode with '\u', right? | 11:32 |
| *** michchap has joined #openstack-ironic | 11:34 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 11:37 |
| lucasagomes | romcheg, ping | 11:38 |
| Haomeng | lucasagomes:pong:) | 11:39 |
| lucasagomes | Haomeng, romcheg morning u guys :) | 11:40 |
| Haomeng | did you have Lunch? | 11:40 |
| Haomeng | good afternoon lucasagomes:) | 11:40 |
| *** prekarat has quit IRC | 11:41 | |
| lucasagomes | :) not yet... almost noon here | 11:41 |
| Haomeng | I got your comments about my fix, will work out new patch:) thanks) | 11:41 |
| Haomeng | :) | 11:41 |
| lucasagomes | Haomeng, ah np :) but I'm sure there might be many other strings that weren't i18n yet | 11:41 |
| Haomeng | yes, agree with you | 11:41 |
| Haomeng | so I will not use 'closed-bug' tag:) | 11:42 |
| Haomeng | got it | 11:42 |
| Haomeng | will try to find more string without "_(" enablement | 11:42 |
| lucasagomes | Haomeng, take a look here: https://wiki.openstack.org/wiki/GitCommitMessages#Including_external_references | 11:43 |
| lucasagomes | it's more like a Partial-Bug: #<id> | 11:43 |
| Haomeng | ok, I will use this correct one:) | 11:44 |
| *** prekarat has joined #openstack-ironic | 11:48 | |
| Haomeng | lucasagomes, thank for your reference | 11:51 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 11:54 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for root (/) https://review.openstack.org/54717 | 11:55 |
| lucasagomes | Haomeng, you welcome :) | 11:55 |
| Haomeng | :) | 11:55 |
| *** matsuhashi has quit IRC | 11:56 | |
| *** matsuhashi has joined #openstack-ironic | 11:59 | |
| *** michchap has quit IRC | 12:04 | |
| *** michchap has joined #openstack-ironic | 12:04 | |
| romcheg | Haomeng: regarding to the bug | 12:05 |
| Haomeng | yes | 12:05 |
| romcheg | that is not an i18N issue | 12:05 |
| *** yuriyz has joined #openstack-ironic | 12:05 | |
| romcheg | We debugged it here and found that the problem is in the object model | 12:05 |
| Haomeng | object model does not accept the unicode, right? | 12:06 |
| Haomeng | ok, if you find the root cause, you can assign to yourself:) | 12:06 |
| romcheg | Haomeng: thanks | 12:07 |
| Haomeng | welcome:) | 12:07 |
| Haomeng | :) | 12:07 |
| romcheg | I will assign yuriyz to the issue then because he has found the reason first | 12:07 |
| Haomeng | yes, make sense:) I am fine:) | 12:08 |
| Haomeng | dont worry:) | 12:08 |
| romcheg | * while I was having my lunch :) | 12:08 |
| Haomeng | :) enjoy your lunch:) | 12:08 |
| romcheg | thanks | 12:09 |
| Haomeng | your local time should be around 12:09? | 12:09 |
| Haomeng | welcome:) | 12:09 |
| romcheg | Haomeng: no, it's 14:09 | 12:10 |
| Haomeng | sorry:) | 12:10 |
| romcheg | it's EET time zone | 12:10 |
| Haomeng | ithink maybe this line " 'description': utils.str_or_none," line is root cause, right? | 12:10 |
| Haomeng | in Chassis.py | 12:10 |
| romcheg | Haomeng: yes, but the problem is quite deeper: the same type specifier is used for the things that cannot contain unicode as well | 12:11 |
| romcheg | e.g., hostnames | 12:12 |
| Haomeng | I see such issue in Ceilometer | 12:12 |
| Haomeng | let me compare with Ceilometer Object code | 12:12 |
| romcheg | So there are two problems: 1) we cannot just change the type specifier to support unicode. 2) on py3 the same type specifier supports unicode so it will cause problems for the attributes that are ascii-only | 12:14 |
| Haomeng | yes | 12:15 |
| romcheg | yurizis about to fix py3 compatibility and add a new type specifier | 12:15 |
| Haomeng | great | 12:15 |
| romcheg | yuriyz even | 12:15 |
| Haomeng | he give my fix some good comments:) | 12:16 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 12:21 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for root (/) https://review.openstack.org/54717 | 12:23 |
| *** jdob has joined #openstack-ironic | 12:27 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for chassis https://review.openstack.org/54720 | 12:28 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 12:30 |
| *** prekarat has quit IRC | 12:32 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 12:33 |
| * lucasagomes always forget something >.< | 12:33 | |
| lucasagomes | sorry for spamming | 12:34 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for node https://review.openstack.org/54714 | 12:34 |
| Haomeng | no problem:) welcome your fix:) | 12:36 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Improve the API doc spec for port https://review.openstack.org/54725 | 12:41 |
| *** rloo has joined #openstack-ironic | 12:52 | |
| *** jbjohnso has joined #openstack-ironic | 13:15 | |
| *** jdob has quit IRC | 13:24 | |
| *** jdob has joined #openstack-ironic | 13:24 | |
| *** linggao has joined #openstack-ironic | 13:28 | |
| *** anniec_ has joined #openstack-ironic | 13:33 | |
| *** anniec has quit IRC | 13:35 | |
| *** anniec_ is now known as anniec | 13:35 | |
| linggao | Morning max_lobur | 13:42 |
| *** jdob has quit IRC | 13:42 | |
| max_lobur | Morning linggao | 13:43 |
| *** jdob has joined #openstack-ironic | 13:43 | |
| *** matsuhashi has quit IRC | 13:43 | |
| max_lobur | Morning Everyone | 13:43 |
| linggao | Morning all. | 13:43 |
| *** matsuhashi has joined #openstack-ironic | 13:43 | |
| linggao | max_lobur, I saw your comments. They are very good. Can we bring it to davananda and NobodyCam for their approval? | 13:44 |
| *** matsuhashi has quit IRC | 13:44 | |
| *** matsuhas_ has joined #openstack-ironic | 13:44 | |
| max_lobur | linggao, Glad you liked =) Sure, why not | 13:45 |
| linggao | I can be either way. | 13:45 |
| linggao | I am not good at comming up with a good API/CLI | 13:46 |
| linggao | I can make the plumbing work. :) | 13:46 |
| romcheg | Morning linggao and max_lobur | 13:46 |
| romcheg | How are you? | 13:46 |
| linggao | Morning romcheg. | 13:47 |
| linggao | A cloudy day here. | 13:47 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Allows unicode description for chassis https://review.openstack.org/54737 | 13:54 |
| *** matsuhas_ has quit IRC | 13:55 | |
| *** matsuhashi has joined #openstack-ironic | 13:56 | |
| max_lobur | Morning romcheg :) How's your weather today? :) | 13:57 |
| romcheg | max_lobur: I think it's the same as yours :) | 13:57 |
| max_lobur | Indeed :) | 13:58 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Accessing a subresource that parent does not exist https://review.openstack.org/54742 | 14:19 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Accessing a subresource that parent does not exist https://review.openstack.org/54742 | 14:20 |
| *** dwi has joined #openstack-ironic | 14:28 | |
| *** matsuhashi has quit IRC | 14:35 | |
| *** matsuhashi has joined #openstack-ironic | 14:36 | |
| *** matsuhashi has quit IRC | 14:41 | |
| *** matsuhas_ has joined #openstack-ironic | 14:41 | |
| NobodyCam | Good morning Ironic | 14:47 |
| *** matsuhas_ has quit IRC | 14:47 | |
| *** matsuhashi has joined #openstack-ironic | 14:48 | |
| lucasagomes | morning NobodyCam | 14:49 |
| *** ecozzi has joined #openstack-ironic | 14:49 | |
| NobodyCam | morning lucasagomes | 14:50 |
| *** matsuhashi has quit IRC | 14:53 | |
| GheRivero | morning Ironic! | 14:53 |
| NobodyCam | good morning GheRivero | 14:54 |
| lucasagomes | morning GheRivero | 14:54 |
| NobodyCam | woo hoo Ironic has landed (in tripleO) | 15:09 |
| *** martyntaylor has left #openstack-ironic | 15:12 | |
| * NobodyCam gets mo coffee | 15:14 | |
| NobodyCam | and seems its walkies time... brb | 15:19 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Allows unicode description for chassis https://review.openstack.org/54737 | 15:22 |
| romcheg | Morning NobodyCam | 15:33 |
| NobodyCam | morning romcheg | 15:35 |
| yuriyz | Morning/evening all | 15:42 |
| NobodyCam | morning yuriyz :) | 15:43 |
| lucasagomes | NobodyCam, kudos! | 15:43 |
| NobodyCam | :-p | 15:43 |
| lucasagomes | ironic and ironic client | 15:44 |
| lucasagomes | double kudos! | 15:44 |
| NobodyCam | :-p | 15:45 |
| NobodyCam | working on nova-ironic element now | 15:45 |
| *** bauzas has quit IRC | 15:54 | |
| dkehn | NobodyCam: I need to add in the ml2 portion of the tripleo-image-elements/elements/neutron/os-config-applier/etc/neutron/plugins, does one just add the subdir and ini file and it gets built or is there more? | 16:02 |
| NobodyCam | dkehn: yep that file would end up at /etc/neutron/plugins | 16:04 |
| dkehn | NobodyCam: just wonder if just adding is all there is to it? | 16:04 |
| NobodyCam | os-apply-config | 16:05 |
| NobodyCam | make sure that it is in element-deps | 16:05 |
| *** anniec has quit IRC | 16:06 | |
| *** ben_duyujie has joined #openstack-ironic | 16:06 | |
| dkehn | u mean the os-apply-config is there | 16:08 |
| dkehn | I don't see any direct reference to neutron in there | 16:09 |
| NobodyCam | ya, they updated the name from os-config-applier to os-apply-config | 16:10 |
| NobodyCam | if you changing a existing element git mv it to the new name | 16:11 |
| NobodyCam | and note that update in the commit message | 16:11 |
| NobodyCam | SpamapS: is that correct ^^^^ | 16:11 |
| max_lobur | linggao, I'm leaving earlier today, sorry beforehand if I'll miss the discussion of your patches | 16:11 |
| max_lobur | I'll check chat history tomorrow | 16:12 |
| linggao | max_lobour. we need to wait for davananda. | 16:12 |
| max_lobur | yeah, I know | 16:12 |
| linggao | It's better you bring it up to him. | 16:13 |
| linggao | It's not in a hurry. we can do it tomorow. | 16:13 |
| max_lobur | OK, then I will ping him tomorrow | 16:13 |
| max_lobur | Thanks | 16:13 |
| linggao | thank you max_lobur. | 16:13 |
| *** derekh has quit IRC | 16:14 | |
| linggao | we better have max_lobur, devadana, NobodyCam and lifeless together to come to a conclusion for the instance uuid CLI, REST API and internal API. I'll make the plumbing work :-). | 16:16 |
| NobodyCam | anyone ever seen this: /opt/stack/venvs/nova/bin/activate: line 8: _OLD_VIRTUAL_PATH: unbound variable | 16:17 |
| devananda | g'morning, all | 16:22 |
| NobodyCam | good morning devananda | 16:26 |
| devananda | max_lobur: thanks for the suggestion - your point regarding returning a one-element list is quite convincing | 16:27 |
| devananda | max_lobur: i hadn't considered a filter to get_one. I dont know if other projects are doing that, but it seems reasonable to allow, eg "ironic node-show [--instance] <uuid>" | 16:30 |
| devananda | max_lobur: however, using the same parameter and "try node.uuid; then try node.instance_uuid" is illogical to me. over-simplified | 16:31 |
| linggao | Good morning, davanada, I think max_lobur left already. | 16:33 |
| devananda | hi linggao | 16:33 |
| devananda | linggao: i've posted my suggestion here https://review.openstack.org/#/c/53262/6/ironic/api/controllers/v1/node.py | 16:41 |
| devananda | lucasagomes: if you have a minute, your feedback on the REST API discussion ^ would be great | 16:42 |
| devananda | i'm not sure if my suggestion violates REST principles any less than max's. node.instance_uuid is a property of node, but can we make a GET'able path out of it, too? | 16:43 |
| linggao | davananda, I like it. This will make a lot of people happy. | 16:43 |
| devananda | linggao: thanks for being patient while we all discuss this :) | 16:45 |
| linggao | davananda, :). then we also need to modify the associated and unassociated case in REST API. | 16:45 |
| devananda | linggao: i like max' suggestion for those -- GET /v1/nodes/?instance_associated=[True|False] | 16:46 |
| devananda | or something like that | 16:46 |
| devananda | it's clear that it's a filter, it returns a list of nodes, and the parameter means only one thing | 16:47 |
| linggao | or \/v1/nodes/instance?associated=[True|False] ? | 16:47 |
| lucasagomes | ok I quickly read max comments | 16:48 |
| lucasagomes | something similar already happens in our API with ports | 16:48 |
| lucasagomes | both uuid and address could be used as ID | 16:48 |
| lucasagomes | for e.g: http://fpaste.org/50814/ | 16:48 |
| lucasagomes | or http://fpaste.org/50815/ | 16:49 |
| devananda | ah | 16:49 |
| devananda | because of db.sqla.api.add_port_filter | 16:50 |
| lucasagomes | yes | 16:50 |
| devananda | drat | 16:50 |
| lucasagomes | now, there's good and bad args on both sides | 16:50 |
| lucasagomes | if ID means something unique of that object, maybe it also doesnt do any harm in having more than one ID | 16:51 |
| lucasagomes | although it's doesnt really sounds very consistent | 16:51 |
| devananda | those cases are a matter of "different format of ID". Which is still magic and could be confusing | 16:55 |
| devananda | in this case, theyr'e both UUID | 16:55 |
| linggao | right | 16:55 |
| devananda | technicaly it is possible to get the same value in both fields | 16:55 |
| devananda | there isn't (and cant be) a UX covering both columns | 16:56 |
| devananda | brb | 16:57 |
| devananda | back | 17:08 |
| NobodyCam | wb back | 17:09 |
| lucasagomes | devananda, www.w3.org/Protocols/rfc2616/rfc2616-sec10.html | 17:18 |
| lucasagomes | check the 303 | 17:18 |
| linggao | devananda, which one do you want to bless? :-) /v1/nodes/?instance_associated=[True|False] or /v1/nodes/instance/?associated=[True|False] | 17:25 |
| devananda | lucasagomes: are you suggesting we implement 303 for things like "GET /v1/ports/<MAC ADDR/" ? | 17:27 |
| lucasagomes | devananda, no I think that /v1/port/<mac addr> is bad to be honest | 17:27 |
| lucasagomes | because mac address can be changed | 17:27 |
| devananda | right | 17:28 |
| lucasagomes | so i can be changed and cna be assigned to another port for e.g | 17:28 |
| devananda | lucasagomes: this generates teh same question as isntance_uuid though | 17:28 |
| lucasagomes | so that ports/<mac address> now is the id of another port | 17:28 |
| lucasagomes | devananda, yes | 17:28 |
| devananda | do we have a list-like query tjhat returns 0-or-1, eg /ports/?mac=xxx | 17:28 |
| lucasagomes | I don't have the right answer yet | 17:28 |
| lucasagomes | but IMO it should not be used | 17:28 |
| devananda | or a sub-path, eg, /ports/mac/xxxx/ | 17:28 |
| lucasagomes | devananda, yes | 17:29 |
| lucasagomes | I would -1 max idea | 17:29 |
| lucasagomes | I'll comment to that review | 17:29 |
| devananda | thanks | 17:29 |
| lucasagomes | I don't know the right answer yet | 17:29 |
| devananda | k | 17:29 |
| lucasagomes | but I'm sure that approach is not the right one | 17:29 |
| lucasagomes | cause those are changeable ID's so should not be used like we are using it now | 17:30 |
| devananda | indeed | 17:30 |
| devananda | I consider that an API bug | 17:30 |
| lucasagomes | devananda, +1 | 17:30 |
| devananda | (due to some magic in the db implementation) | 17:30 |
| * devananda files it | 17:30 | |
| lucasagomes | the say I see it, it's def a bug | 17:30 |
| lucasagomes | the way* | 17:31 |
| *** jistr has quit IRC | 17:31 | |
| devananda | https://bugs.launchpad.net/ironic/+bug/1246816 | 17:35 |
| devananda | I think it boils down to -- is there anything unRESTful about "GET /path/object/property/<changeable-value>/" returning a single object or 404? The alternative seems to be "GET /path/object/?property=<changeable-value>" returning a (possibly empty) list of object IDs, and requiring a second GET to pull the details | 17:37 |
| lucasagomes | devananda, yea, that's the question I don't know the answer if that's unrestful or not, but doesn't really sounds correct to me | 17:42 |
| lucasagomes | because that URL should be unique and would redirect you _always_ to the same object | 17:43 |
| lucasagomes | but in the case of a changable-value it does not | 17:43 |
| devananda | yea | 17:43 |
| lucasagomes | the ?property=<changeble-value> sounds reasonable to me | 17:43 |
| *** ben_duyujie has quit IRC | 17:43 | |
| devananda | yea | 17:44 |
| lucasagomes | but one thing about REST/UNREST is that (at least what I've seen until now that I have being playing with the Ironic API) is that nobody has the answer to that | 17:44 |
| devananda | the mroe i think about the implications of /path/object/property/<changeable-value>/ the more it does not make sense | 17:44 |
| lucasagomes | yes | 17:45 |
| devananda | because subsequent GET to the same URL could yield very different results | 17:45 |
| lucasagomes | I knew this thing about the port for sometime by now, apologies I never brought it up... but just now reading max comments that I actually started thinking about it | 17:45 |
| devananda | 1) here's node X. 2) 404-not-found. 3) here's node Y. .... | 17:45 |
| devananda | as a client, getting a string of responses like ^ for the same URL would be maddening | 17:46 |
| lucasagomes | devananda, yes it's _totally_ unpredictable | 17:46 |
| linggao | then davananda, we need to get back to /va/nodes/?instance_uuid=<uuid> instead of /v1/nodes/instance/<uuid> ? | 17:46 |
| lucasagomes | linggao, both seems reasonable but I would +1 the /nodes/?instance_uuid=<uuid> in the moment | 17:47 |
| devananda | linggao: yes. where /v1/nodes/?instance_uuid=<value> returns a LIST of node UUID | 17:47 |
| devananda | linggao: list may be empty | 17:47 |
| lucasagomes | I don't want to create a subresource for each attribute in the resource | 17:47 |
| devananda | however this still doesn't solve the "get [un]associated" API question | 17:47 |
| devananda | since that isn't a property | 17:47 |
| devananda | it's more of a metaproperty | 17:47 |
| lucasagomes | devananda, +1 /nodes (without id) is a collection, and collection return lists | 17:48 |
| lifeless | morning | 17:48 |
| devananda | morning | 17:48 |
| lucasagomes | so yea even a list of 1 item is fine for me | 17:48 |
| devananda | lucasagomes: list of 0 items is still a valid list, IMO | 17:48 |
| lucasagomes | devananda, oh yea | 17:48 |
| lucasagomes | in case the instace_uuid is non-existent | 17:49 |
| linggao | then are we still doing node-show --instance <uuid> ? | 17:49 |
| lucasagomes | lifeless, morning | 17:50 |
| linggao | morning lifeless. | 17:50 |
| lucasagomes | lifeless, if you get a minute to read the discussing above and give us ur pov would be great :) | 17:50 |
| devananda | lucasagomes: also, 404, while not classified as a permanent error, might be a bit of a harsh response when searching on a mutable field. | 17:52 |
| lucasagomes | devananda, indeed, imo it's like a query... he searched for something and it returned 0 items, so the return would be a empty list | 17:52 |
| lucasagomes | it's not an error | 17:53 |
| lucasagomes | its the result of that query | 17:53 |
| yuriyz | lucasgomes: +1 for non-id fields | 17:55 |
| lifeless | I think for any url it should be clear if it's a search (returns a list always, 404 only if the /contex/ of the search is unknown) | 17:56 |
| lifeless | or if it's a get - returns the thing or 404 if the thing-or-the-context is unknown | 17:56 |
| lifeless | note that 404's can be negatively cached by proxies | 17:57 |
| yuriyz | lucas, we dont expose db id's for users in api, but 'ironic node-show 1' works, its OK? | 18:02 |
| lifeless | lucasagomes: so there's a lot of context above | 18:02 |
| lifeless | lucasagomes: whats the key things that are at stake? | 18:03 |
| SpamapS | NobodyCam: correct. We will deprecate the os-config-applier name entirely.. someday. | 18:04 |
| NobodyCam | :) awesome TY SpamapS :) | 18:04 |
| NobodyCam | dkehn: you see that ^^^ | 18:04 |
| dkehn | yepper | 18:04 |
| lucasagomes | yuriyz, yea it's because of the same magic filters that makes port returns using the macadress | 18:06 |
| lucasagomes | yuriyz, but I would avoid it just to be more consistent u know | 18:06 |
| yuriyz | yes, +1, this magic only for internal use | 18:06 |
| lucasagomes | like uuids for users and only internally we would the id cause of the db operations | 18:06 |
| lucasagomes | lifeless, it started cause of this review here https://review.openstack.org/#/c/53262/ where someone suggested us to have a similar behavior to that v1/ports/<mac address> but for v1/nodes/<instance uuid> | 18:08 |
| dkehn | lifeless: still showing what I've got in github' | 18:08 |
| lucasagomes | yuriyz, devananda lifeless I came to the office today so I have to go home | 18:08 |
| *** romcheg has quit IRC | 18:08 | |
| lifeless | lucasagomes: ok, ciao. | 18:09 |
| lucasagomes | can we continue that tomorrow? or u guys cna continue and then I will read the scrollback | 18:09 |
| NobodyCam | have a good ride home lucasagomes | 18:09 |
| lifeless | I don't think v1/nodes/instanceuuid makes sense | 18:09 |
| lifeless | there should be one canonical location to do GET from | 18:09 |
| lifeless | vs search | 18:09 |
| lucasagomes | lifeless, yea me too, there's some comments that worth reading on that review | 18:09 |
| lucasagomes | NobodyCam, cheers mate! I see ya later | 18:09 |
| yuriyz | good night lucasagomes | 18:11 |
| lucasagomes | night yall! | 18:11 |
| *** lucasagomes has quit IRC | 18:11 | |
| *** yuriyz has quit IRC | 18:15 | |
| *** openstackgerrit has quit IRC | 18:25 | |
| *** romcheg has joined #openstack-ironic | 18:40 | |
| * NobodyCam kicks cable modem and reboot its | 18:41 | |
| *** ndipanov is now known as ndipanov_gone | 18:58 | |
| *** vkozhukalov has joined #openstack-ironic | 19:01 | |
| * devananda steps out for coffee | 19:13 | |
| NobodyCam | :) | 19:15 |
| *** openstackgerrit has joined #openstack-ironic | 19:18 | |
| *** rushiagr has joined #openstack-ironic | 19:24 | |
| *** vkozhukalov has quit IRC | 19:36 | |
| NobodyCam | devananda: you around | 19:41 |
| NobodyCam | devananda: lol :-p https://github.com/NoBodyCam/nova-ironic/blob/master/nova-ironic/install.d/89-nova-ironic-services | 19:42 |
| devananda | back | 19:47 |
| NobodyCam | wb | 19:47 |
| NobodyCam | devananda: that kinda the process you recall? | 19:49 |
| romcheg | devananda: Hi | 20:00 |
| romcheg | I just pushed a big update to the tempest patch | 20:00 |
| devananda | NobodyCam: yep! looks about right | 20:00 |
| devananda | NobodyCam: after that, you'd need to service xxx restart | 20:00 |
| NobodyCam | that at dib stage so still have to boot | 20:00 |
| NobodyCam | :) | 20:00 |
| devananda | NobodyCam: also, i was having problems with DNS resolution, so I hacked 8.8.8.8 into /etc/resolv.conf | 20:00 |
| NobodyCam | yes | 20:01 |
| devananda | NobodyCam: not at dib stage. after heat stack-create undercloud, i would run that in the instance | 20:01 |
| romcheg | There are 4 patches now that need to be merged to enable gate testing for Ironic | 20:01 |
| devananda | i maen, you could do it in dib stage. but i prefer to keep th image general, then specialize it for a given test run | 20:02 |
| romcheg | No one has taken a look at them until now | 20:02 |
| romcheg | devananda: can you please poke someone? | 20:02 |
| devananda | romcheg: great. can you post links here? | 20:02 |
| devananda | i will poke away | 20:02 |
| NobodyCam | just for me testing right now. | 20:02 |
| romcheg | Tempest tests for Ironic API: https://review.openstack.org/#/c/48109 | 20:03 |
| romcheg | Pre-cache Ironic to slaves: https://review.openstack.org/#/c/54569 | 20:03 |
| romcheg | Enable Ironic in devstack-gate: https://review.openstack.org/#/c/53899 | 20:03 |
| romcheg | Enable tempest tests in the experimental pipeline: https://review.openstack.org/#/c/53917 | 20:03 |
| NobodyCam | devananda: I will prob start adding to the nova bindings patch | 20:03 |
| devananda | romcheg: just to confirm, no dependency on any ironic or python-ironiccilent patches right now? | 20:05 |
| romcheg | devananda: No | 20:05 |
| devananda | k k | 20:06 |
| romcheg | devananda: There is one bug in Ironic that should be fixed to make all tempest tests to pass but I don't think that will be a problem | 20:06 |
| romcheg | I'm talking about this one: https://review.openstack.org/#/c/54737/ | 20:06 |
| *** hughsaunders has joined #openstack-ironic | 20:21 | |
| *** jdob has quit IRC | 20:31 | |
| *** romcheg has left #openstack-ironic | 20:36 | |
| *** rushiagr has quit IRC | 20:41 | |
| *** michchap has quit IRC | 21:01 | |
| *** linggao has quit IRC | 21:03 | |
| *** michchap has joined #openstack-ironic | 21:03 | |
| *** jbjohnso has quit IRC | 21:23 | |
| * NobodyCam is back | 21:48 | |
| NobodyCam | with hairs cut | 21:48 |
| NobodyCam | great email devananda | 22:04 |
| *** jbjohnso has joined #openstack-ironic | 22:30 | |
| NobodyCam | starts building a new undercloud image and goes for quick walkies | 22:41 |
| openstackgerrit | Haomeng,Wang proposed a change to openstack/ironic: Fix i18N compliance https://review.openstack.org/54687 | 23:39 |
| NobodyCam | ahhh | 23:50 |
| NobodyCam | No distributions at all found for python-ironicclient | 23:50 |
| NobodyCam | guess we would actualy need to do a release for that to have worked | 23:51 |
| devananda | yep! heh | 23:51 |
| NobodyCam | :-p | 23:51 |
| NobodyCam | hehe | 23:51 |
| *** harlowja has quit IRC | 23:55 | |
| *** harlowja has joined #openstack-ironic | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!