| *** sdake_ has joined #openstack-ironic | 00:02 | |
| *** Sukhdev has joined #openstack-ironic | 00:03 | |
| *** Sukhdev has quit IRC | 00:03 | |
| JayF | I have two, somewhat related, patches for the Ironic virt driver that fix major bugs awaiting review for backport to stable/mitaka: https://review.openstack.org/#/c/323477 + https://review.openstack.org/#/c/323196 your reviews would be appreciated :) -- I know nobody in here can land Nova stable patches, but it certainly won't hurt to have Ironic votes on the patches | 00:04 |
|---|---|---|
| JayF | thanks!! | 00:04 |
| *** vishwanathj has quit IRC | 00:09 | |
| *** vishwanathj has joined #openstack-ironic | 00:10 | |
| *** aNupoisc has quit IRC | 00:10 | |
| *** vishwana_ has joined #openstack-ironic | 00:11 | |
| *** aNupoisc has joined #openstack-ironic | 00:11 | |
| *** garthb_ has joined #openstack-ironic | 00:12 | |
| *** praneshp has quit IRC | 00:13 | |
| *** garthb has quit IRC | 00:15 | |
| *** vishwanathj has quit IRC | 00:15 | |
| *** jrist has quit IRC | 00:16 | |
| *** Sukhdev has joined #openstack-ironic | 00:17 | |
| *** garthb_ has quit IRC | 00:19 | |
| *** ayoung has joined #openstack-ironic | 00:20 | |
| *** Sukhdev has quit IRC | 00:21 | |
| *** jrist has joined #openstack-ironic | 00:21 | |
| *** piet_ has quit IRC | 00:21 | |
| *** david-lyle has quit IRC | 00:25 | |
| *** david-lyle has joined #openstack-ironic | 00:26 | |
| *** hoangcx has joined #openstack-ironic | 00:29 | |
| *** david-lyle has quit IRC | 00:30 | |
| *** hshiina has quit IRC | 00:30 | |
| *** baoli has joined #openstack-ironic | 00:36 | |
| *** Sukhdev has joined #openstack-ironic | 00:37 | |
| *** lindycoder has joined #openstack-ironic | 00:38 | |
| *** amotoki has joined #openstack-ironic | 00:40 | |
| *** amotoki has quit IRC | 00:45 | |
| *** phuongnh has joined #openstack-ironic | 00:48 | |
| *** lindycoder has quit IRC | 00:52 | |
| *** lindycoder has joined #openstack-ironic | 00:58 | |
| *** YiboCai has joined #openstack-ironic | 00:58 | |
| *** hshiina has joined #openstack-ironic | 00:59 | |
| *** mtanino has quit IRC | 01:01 | |
| *** Goneri has joined #openstack-ironic | 01:05 | |
| *** Sukhdev has quit IRC | 01:09 | |
| *** aNupoisc has quit IRC | 01:11 | |
| *** Sukhdev has joined #openstack-ironic | 01:13 | |
| *** Sukhdev has quit IRC | 01:15 | |
| *** spandhe has left #openstack-ironic | 01:17 | |
| *** vishwana_ has quit IRC | 01:18 | |
| *** baoli has quit IRC | 01:23 | |
| *** adu has quit IRC | 01:23 | |
| *** praneshp has joined #openstack-ironic | 01:26 | |
| *** baoli has joined #openstack-ironic | 01:27 | |
| *** adu has joined #openstack-ironic | 01:29 | |
| *** baoli has quit IRC | 01:30 | |
| *** baoli has joined #openstack-ironic | 01:31 | |
| *** praneshp has quit IRC | 01:35 | |
| *** baoli has quit IRC | 01:36 | |
| *** vishwanathj has joined #openstack-ironic | 01:39 | |
| *** vishwanathj has quit IRC | 01:40 | |
| *** vishwanathj has joined #openstack-ironic | 01:42 | |
| *** Sukhdev has joined #openstack-ironic | 01:46 | |
| *** vishwanathj has quit IRC | 01:47 | |
| *** frontrunner has quit IRC | 01:55 | |
| *** joprovost has quit IRC | 01:58 | |
| *** PollyZ has joined #openstack-ironic | 01:59 | |
| *** joprovost has joined #openstack-ironic | 02:00 | |
| *** ijw has quit IRC | 02:00 | |
| *** mtanino has joined #openstack-ironic | 02:01 | |
| *** rbudden has quit IRC | 02:01 | |
| *** PollyZ__ has joined #openstack-ironic | 02:03 | |
| *** baoli has joined #openstack-ironic | 02:06 | |
| *** lindycoder has quit IRC | 02:09 | |
| *** lindycoder has joined #openstack-ironic | 02:09 | |
| *** baoli has quit IRC | 02:09 | |
| *** vishwanathj has joined #openstack-ironic | 02:11 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Network https://review.openstack.org/294559 | 02:12 |
| *** penick has quit IRC | 02:13 | |
| *** baoli has joined #openstack-ironic | 02:15 | |
| *** frontrunner has joined #openstack-ironic | 02:19 | |
| *** vishwana_ has joined #openstack-ironic | 02:20 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add some missed test cases in node object tests https://review.openstack.org/268045 | 02:20 |
| *** vishwanathj has quit IRC | 02:20 | |
| *** PollyZ__ has quit IRC | 02:21 | |
| *** PollyZ has quit IRC | 02:21 | |
| *** vishwana_ has quit IRC | 02:22 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add description filed to ports - DB https://review.openstack.org/324644 | 02:23 |
| *** lindycoder has quit IRC | 02:25 | |
| *** lindycoder has joined #openstack-ironic | 02:26 | |
| *** joprovost1 has joined #openstack-ironic | 02:34 | |
| *** joprovost has quit IRC | 02:36 | |
| *** joprovost1 is now known as joprovost | 02:36 | |
| *** sacharya has joined #openstack-ironic | 02:37 | |
| *** amotoki has joined #openstack-ironic | 02:41 | |
| *** mtanino has quit IRC | 02:46 | |
| *** lindycoder has quit IRC | 02:46 | |
| *** amotoki has quit IRC | 02:46 | |
| *** joprovost1 has joined #openstack-ironic | 02:47 | |
| *** joprovost has quit IRC | 02:49 | |
| *** joprovost1 is now known as joprovost | 02:49 | |
| *** yuanying has quit IRC | 02:50 | |
| *** r-daneel has quit IRC | 02:53 | |
| *** piet has joined #openstack-ironic | 02:57 | |
| *** joprovost1 has joined #openstack-ironic | 02:58 | |
| *** joprovost has quit IRC | 03:00 | |
| *** joprovost1 is now known as joprovost | 03:00 | |
| *** rloo has quit IRC | 03:00 | |
| *** joprovost1 has joined #openstack-ironic | 03:09 | |
| *** piet has quit IRC | 03:11 | |
| *** joprovost has quit IRC | 03:11 | |
| *** joprovost1 is now known as joprovost | 03:11 | |
| *** frontrunner has quit IRC | 03:28 | |
| *** Sukhdev has quit IRC | 03:28 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add a new db_api get_remains_conductors https://review.openstack.org/325026 | 03:29 |
| *** Sukhdev has joined #openstack-ironic | 03:29 | |
| *** aparnav has quit IRC | 03:33 | |
| *** Sukhdev has quit IRC | 03:33 | |
| *** piet has joined #openstack-ironic | 03:35 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Uplink Set https://review.openstack.org/295981 | 03:36 |
| *** appu103_ has joined #openstack-ironic | 03:36 | |
| *** amotoki has joined #openstack-ironic | 03:42 | |
| *** appu103_ is now known as aparnav | 03:46 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Remove insecure connection warnings https://review.openstack.org/323873 | 03:47 |
| *** amotoki has quit IRC | 03:47 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Remove insecure connection warnings https://review.openstack.org/323873 | 03:48 |
| *** yuanying has joined #openstack-ironic | 03:49 | |
| *** lindycoder has joined #openstack-ironic | 03:49 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add description filed to ports - DB https://review.openstack.org/324644 | 03:51 |
| *** adu has quit IRC | 03:52 | |
| *** hparekh has quit IRC | 03:55 | |
| *** appu103_ has joined #openstack-ironic | 03:57 | |
| *** aparnav has quit IRC | 03:57 | |
| *** lindycoder has quit IRC | 03:58 | |
| *** baoli has quit IRC | 03:59 | |
| *** Sukhdev has joined #openstack-ironic | 04:04 | |
| *** vmud213 has joined #openstack-ironic | 04:06 | |
| *** piet has quit IRC | 04:06 | |
| *** hparekh has joined #openstack-ironic | 04:07 | |
| *** harlowja_at_home has joined #openstack-ironic | 04:08 | |
| *** ayoung has quit IRC | 04:20 | |
| *** ishant|away is now known as ishant | 04:32 | |
| *** hoangcx has quit IRC | 04:33 | |
| *** vishwanathj has joined #openstack-ironic | 04:34 | |
| *** links has joined #openstack-ironic | 04:38 | |
| *** amotoki has joined #openstack-ironic | 04:43 | |
| *** e0ne has quit IRC | 04:46 | |
| *** e0ne has joined #openstack-ironic | 04:46 | |
| *** e0ne has quit IRC | 04:46 | |
| *** Sukhdev has quit IRC | 04:47 | |
| *** amotoki has quit IRC | 04:47 | |
| *** hoangcx has joined #openstack-ironic | 04:49 | |
| *** harlowja_at_home has quit IRC | 04:56 | |
| *** rcernin has joined #openstack-ironic | 04:58 | |
| *** rcernin has quit IRC | 05:04 | |
| *** adu has joined #openstack-ironic | 05:05 | |
| *** yuanying has quit IRC | 05:17 | |
| *** praneshp has joined #openstack-ironic | 05:22 | |
| *** joprovost has quit IRC | 05:24 | |
| *** vishwanathj has quit IRC | 05:30 | |
| *** vmud213 has quit IRC | 05:40 | |
| *** vmud213 has joined #openstack-ironic | 05:41 | |
| *** amotoki has joined #openstack-ironic | 05:44 | |
| *** rcernin has joined #openstack-ironic | 05:46 | |
| *** adu has quit IRC | 05:47 | |
| *** amotoki has quit IRC | 05:49 | |
| *** adu has joined #openstack-ironic | 05:54 | |
| divya_ | hi vmud213 | 05:56 |
| divya_ | nova failed to boot instance | 05:56 |
| divya_ | Failed to deploy. Error: 400 Bad Request: Unknown scheme 'swift+config' found in URI (HTTP 400) | 05:56 |
| divya_ | any inputs ? | 05:57 |
| vmud213 | divya: I usualy run it with ironic client. | 06:00 |
| vmud213 | What error s u are getting on the conductor? or glance,swift | 06:00 |
| vmud213 | can u look and paste them if possible | 06:00 |
| *** yuanying has joined #openstack-ironic | 06:05 | |
| *** vishwanathj has joined #openstack-ironic | 06:08 | |
| *** sabeen has quit IRC | 06:08 | |
| divya_ | am using glance as backened store - file instead of swift | 06:09 |
| divya_ | http://paste.openstack.org/show/507518/ | 06:09 |
| divya_ | pasted the ironic-cond error | 06:09 |
| divya_ | pls look into it | 06:09 |
| *** vmud213 has quit IRC | 06:10 | |
| *** vmud213 has joined #openstack-ironic | 06:11 | |
| *** mjura has joined #openstack-ironic | 06:11 | |
| *** sabeen has joined #openstack-ironic | 06:12 | |
| divya_ | hi vmud213 | 06:13 |
| divya_ | n-cpu.log | 06:13 |
| divya_ | am using glance as backened store - file instead of swift | 06:13 |
| divya_ | http://paste.openstack.org/show/507518/ | 06:13 |
| *** vishwanathj has quit IRC | 06:16 | |
| *** openstackgerrit has quit IRC | 06:17 | |
| *** openstackgerrit has joined #openstack-ironic | 06:18 | |
| *** mjura has quit IRC | 06:20 | |
| *** mjura has joined #openstack-ironic | 06:20 | |
| *** phuongnh has quit IRC | 06:21 | |
| *** phuongnh has joined #openstack-ironic | 06:22 | |
| *** ohamada has joined #openstack-ironic | 06:26 | |
| *** yuanying has quit IRC | 06:28 | |
| *** yuanying has joined #openstack-ironic | 06:29 | |
| *** mkoderer has quit IRC | 06:30 | |
| *** yuanying has quit IRC | 06:31 | |
| *** mkoderer has joined #openstack-ironic | 06:31 | |
| *** yuanying has joined #openstack-ironic | 06:31 | |
| *** yuanying has quit IRC | 06:33 | |
| *** yuanying has joined #openstack-ironic | 06:34 | |
| *** harlowja has quit IRC | 06:35 | |
| *** adu has quit IRC | 06:37 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Imported Translations from Zanata https://review.openstack.org/325077 | 06:39 |
| vmud213 | divya: Sry abt the delay.. | 06:43 |
| vmud213 | Cpl of things need to check | 06:43 |
| *** amotoki has joined #openstack-ironic | 06:45 | |
| vmud213 | what is the /etc/glance/glance-ap.conf configured for swift? | 06:45 |
| vmud213 | what swift file it is refering to | 06:45 |
| vmud213 | Looks like glance image-download is failing | 06:46 |
| *** praneshp has quit IRC | 06:49 | |
| *** amotoki has quit IRC | 06:49 | |
| *** kromanenko__ has joined #openstack-ironic | 06:50 | |
| *** Romanenko_K has joined #openstack-ironic | 06:50 | |
| *** kromanenko_ has joined #openstack-ironic | 06:50 | |
| divya_ | thanks vmud213 | 06:50 |
| divya_ | i configured back swift now | 06:51 |
| divya_ | but the new error os Error: Failed to connect to Glance to get the properties of the image | 06:51 |
| divya_ | any idea | 06:51 |
| vmud213 | Are u sure u are using the correct swift account key | 06:55 |
| vmud213 | oh..ok.not much idea Divya. But hard to follow what errors u'r getting unless see the log | 06:56 |
| divya_ | http://paste.openstack.org/show/507523/ | 07:01 |
| divya_ | ironic-cond logs | 07:01 |
| *** harlowja_at_home has joined #openstack-ironic | 07:02 | |
| *** harlowja_at_home has quit IRC | 07:02 | |
| *** openstackgerrit has quit IRC | 07:03 | |
| *** openstackgerrit has joined #openstack-ironic | 07:03 | |
| *** SiRiuS__ has joined #openstack-ironic | 07:06 | |
| *** sacharya has quit IRC | 07:07 | |
| *** phuongnh has quit IRC | 07:11 | |
| *** phuongnh has joined #openstack-ironic | 07:12 | |
| *** jistr is now known as jistr|mtg | 07:22 | |
| *** watanabe_isao has joined #openstack-ironic | 07:25 | |
| *** watanabe_isao has quit IRC | 07:26 | |
| *** watanabe_isao has joined #openstack-ironic | 07:27 | |
| *** watanabe_isao has quit IRC | 07:29 | |
| *** watanabe_isao has joined #openstack-ironic | 07:29 | |
| *** vmud213 has quit IRC | 07:31 | |
| *** vmud213 has joined #openstack-ironic | 07:31 | |
| divya_ | hi vmud213, i am still getting error during deploy -" Failed to connect to Glance to get the properties of the image" | 07:42 |
| divya_ | [glance] swift_temp_url_duration = 3600 swift_container = glance swift_account = AUTH_46bd1bb3a96e42e39d9c53f6c30387fd swift_api_version = v1 swift_endpoint_url = http://172.26.24.110:5000 swift_temp_url_key = password | 07:43 |
| divya_ | these are the glance configuraton in ironic conf | 07:43 |
| divya_ | do i need to update any few? | 07:43 |
| divya_ | no error are observed in glance or swift logs | 07:44 |
| divya_ | only ironic logs have the errors | 07:44 |
| *** vishwanathj has joined #openstack-ironic | 07:44 | |
| *** amotoki has joined #openstack-ironic | 07:45 | |
| *** amotoki has quit IRC | 07:50 | |
| *** yuanying has quit IRC | 07:50 | |
| vmud213 | divya_: two things i'd like to know. | 07:56 |
| vmud213 | First the image that corresponds to the ID in the log exists. | 07:56 |
| vmud213 | and "netstat -tupln|grep 9292" is listing glance-api process | 07:56 |
| *** ChrisMead has quit IRC | 07:58 | |
| *** yonglihe has quit IRC | 07:58 | |
| *** evgenyl has quit IRC | 07:58 | |
| *** kozhukalov has quit IRC | 07:58 | |
| *** rajinir has quit IRC | 07:58 | |
| *** agordeev has quit IRC | 07:58 | |
| *** romcheg has quit IRC | 07:58 | |
| *** sergek has quit IRC | 07:58 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:01 | |
| *** sdake_ has quit IRC | 08:03 | |
| *** mgoddard has joined #openstack-ironic | 08:04 | |
| *** daemontool has joined #openstack-ironic | 08:06 | |
| *** ifarkas has joined #openstack-ironic | 08:06 | |
| *** sacharya has joined #openstack-ironic | 08:08 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 08:10 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 08:10 |
| *** lucas-afk is now known as lucasagomes | 08:11 | |
| lucasagomes | morning all | 08:13 |
| *** sacharya has quit IRC | 08:13 | |
| *** derekh has joined #openstack-ironic | 08:13 | |
| *** jimmy_ has joined #openstack-ironic | 08:16 | |
| jimmy_ | Good morning all. | 08:16 |
| betherly | morning! | 08:16 |
| jimmy_ | When I run "tox -e py27" the stdout is redirected to virtualenv. How can I disable this? | 08:17 |
| jimmy_ | I mean I want to see it in console. | 08:17 |
| openstackgerrit | Anton Arefiev proposed openstack/ironic-inspector: Test commit https://review.openstack.org/325118 | 08:22 |
| *** e0ne has joined #openstack-ironic | 08:24 | |
| *** dtantsur|afk is now known as dtantsur | 08:28 | |
| dtantsur | Morning Ironic | 08:28 |
| lucasagomes | jimmy_, not sure if I get the question, the ouput goes to stdout no? | 08:30 |
| *** mjura has quit IRC | 08:31 | |
| *** MattMan has quit IRC | 08:31 | |
| *** MattMan has joined #openstack-ironic | 08:31 | |
| vsaienko | Morning Ironic, dtantsur, lucasagomes | 08:36 |
| *** betherly_ has joined #openstack-ironic | 08:36 | |
| *** phschwartz_ has joined #openstack-ironic | 08:36 | |
| *** lekha-away_ has joined #openstack-ironic | 08:36 | |
| *** MattMan_ has joined #openstack-ironic | 08:37 | |
| *** ildikov_ has joined #openstack-ironic | 08:38 | |
| *** daemontool has quit IRC | 08:38 | |
| *** clif has joined #openstack-ironic | 08:38 | |
| *** e0ne has quit IRC | 08:38 | |
| *** e0ne has joined #openstack-ironic | 08:39 | |
| *** ildikov has quit IRC | 08:39 | |
| *** johnthetubaguy has quit IRC | 08:39 | |
| *** lekha-away has quit IRC | 08:39 | |
| *** bradjones has quit IRC | 08:39 | |
| *** vishwanathj has quit IRC | 08:39 | |
| *** phschwartz has quit IRC | 08:39 | |
| *** morgabra has quit IRC | 08:39 | |
| *** clif_h has quit IRC | 08:39 | |
| *** joemonk- has quit IRC | 08:39 | |
| *** MattMan has quit IRC | 08:39 | |
| *** watanabe_isao has quit IRC | 08:39 | |
| *** tonyb has quit IRC | 08:39 | |
| *** betherly has quit IRC | 08:39 | |
| *** tonyb has joined #openstack-ironic | 08:39 | |
| *** johnthetubaguy has joined #openstack-ironic | 08:39 | |
| *** ildikov_ is now known as ildikov | 08:39 | |
| *** joemonk has joined #openstack-ironic | 08:39 | |
| *** yuanying has joined #openstack-ironic | 08:39 | |
| *** lekha-away_ is now known as lekha-away | 08:39 | |
| *** bradjones has joined #openstack-ironic | 08:39 | |
| *** bradjones has quit IRC | 08:39 | |
| *** bradjones has joined #openstack-ironic | 08:39 | |
| *** morgabra_ has joined #openstack-ironic | 08:40 | |
| *** betherly_ is now known as betherly | 08:41 | |
| *** mjura has joined #openstack-ironic | 08:43 | |
| *** sabeen has quit IRC | 08:43 | |
| jimmy_ | lucasagomes， By default, I can't see the output. How can I see it? | 08:43 |
| *** sabeen has joined #openstack-ironic | 08:43 | |
| *** jistr|mtg is now known as jistr | 08:43 | |
| alineb | morning all | 08:44 |
| dtantsur | morning alineb | 08:44 |
| dtantsur | and vsaienko | 08:44 |
| *** amotoki has joined #openstack-ironic | 08:46 | |
| alineb | o/ dtantsur | 08:46 |
| *** phuongnh has quit IRC | 08:48 | |
| *** sabeen has quit IRC | 08:48 | |
| *** phuongnh has joined #openstack-ironic | 08:48 | |
| vsaienko | o/ alineb | 08:48 |
| lucasagomes | jimmy_, when you run tox -epy27 you don't see any output at all!/ | 08:49 |
| lucasagomes | ?* | 08:49 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add a new DB API get_remains_conductors() https://review.openstack.org/325026 | 08:50 |
| jimmy_ | lucasagomes, Yes, I see something. | 08:50 |
| lucasagomes | jimmy_, can you paste it? | 08:50 |
| lucasagomes | like do you see the results of the tests? The traceback of the tests that failed ? | 08:50 |
| *** amotoki has quit IRC | 08:51 | |
| alineb | morning vsaienko | 08:51 |
| jimmy_ | lucasagomes, These tests passes... | 08:51 |
| lucasagomes | jimmy_, sure, but when a test fails do you see the traceback I mean? | 08:52 |
| lucasagomes | jimmy_, can you paste the output u see please? | 08:52 |
| jimmy_ | lucasagomes, when a test fails, I can see the traceback. | 08:53 |
| *** rajinir has joined #openstack-ironic | 08:54 | |
| jimmy_ | lucasagomes, How can I see the stdout of the tested code? | 08:55 |
| lucasagomes | jimmy_, I think we are out of sync about the output expectations here | 08:56 |
| lucasagomes | jimmy_, you want to debug the tests ? | 08:56 |
| lucasagomes | because you won't see a "print" in the tests that passes | 08:57 |
| jimmy_ | lucasagomes, yes, I want to debug the tests. | 08:57 |
| lucasagomes | neither logs | 08:57 |
| lucasagomes | ok then | 08:57 |
| jimmy_ | lucasagomes, Yes I mean it. | 08:57 |
| *** agordeev has joined #openstack-ironic | 08:57 | |
| *** kozhukalov has joined #openstack-ironic | 08:57 | |
| *** evgenyl has joined #openstack-ironic | 08:57 | |
| jimmy_ | lucasagomes, How can I see these "print"? | 08:59 |
| * lucasagomes thinks, never did that either | 09:03 | |
| lucasagomes | I know that you can use pdb with testr and all | 09:03 |
| *** romcheg has joined #openstack-ironic | 09:05 | |
| sambetts|afk | thiagop: Seems to have gone ok so far, I'm not using puppet-openstackci right now so that might be why :) | 09:08 |
| *** sambetts|afk is now known as sambetts | 09:08 | |
| jimmy_ | lucasagomes, Thanks for remind me of this way!!! | 09:09 |
| lucasagomes | jimmy_, yeah, but I don't know about the output :-( | 09:09 |
| jimmy_ | lucasagomes, I searched it. | 09:09 |
| lucasagomes | cool :-) | 09:10 |
| sambetts | lucasagomes, jimmy_: if you are trying debug a specfic test and want to use prints statements, just add an assertTrue(False) to the end of the test, and it'll fail the test and you'll get to see the output | 09:10 |
| lucasagomes | sambetts, hah | 09:10 |
| * dtantsur adores our testing tool | 09:10 | |
| lucasagomes | or print to a file | 09:10 |
| jimmy_ | lucasagomes, https://wiki.openstack.org/wiki/Testr#Debugging_.28pdb.29_Tests | 09:10 |
| *** yonglihe has joined #openstack-ironic | 09:11 | |
| divya | still same error "am not getting the error now Failed to connect to Glance to get the properties of the image 2aee8624-f020-49db-af2d-2ea7b117ad08 (HTTP 500)" | 09:11 |
| lucasagomes | jimmy_, yup, yeah with pdb after you set the the points you can print an all | 09:11 |
| jimmy_ | lucasagomes, I will use this way!!! | 09:11 |
| lucasagomes | jimmy_, cool, yeah that's useful | 09:11 |
| sambetts | lucasagomes, dtantsur: nice to see the gate is lovely and stable today -> https://review.openstack.org/#/c/172067/ | 09:12 |
| *** sergek has joined #openstack-ironic | 09:12 | |
| lucasagomes | :-) | 09:12 |
| lucasagomes | on a friday yeah | 09:13 |
| dtantsur | WOW | 09:13 |
| dtantsur | yeah, Friday | 09:13 |
| dtantsur | btw happy Friday everyone | 09:13 |
| sambetts | TGIF! | 09:13 |
| lucasagomes | #pixiesay happy friday -m happy | 09:15 |
| PixieBoots | ʕ^ᴥ^ʔ: happy friday | 09:15 |
| lucasagomes | it's bank holiday on monday here too o/ | 09:15 |
| lucasagomes | sambetts, same there, isn't it? | 09:15 |
| *** ChrisMead has joined #openstack-ironic | 09:16 | |
| *** vdrok has joined #openstack-ironic | 09:19 | |
| divya | Failed to connect to Glance to get the properties of the image 2aee8624-f020-49db-af2d-2ea7b117ad08 | 09:22 |
| divya | not able to come out of this..deploy failing :( :( | 09:22 |
| dtantsur | lucasagomes, do you plan to work on the new 'baremetal create' command or should someone else take it? | 09:23 |
| vdrok | morning ironic | 09:24 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add debug environment to tox https://review.openstack.org/325151 | 09:24 |
| lucasagomes | dtantsur, I think vsaienko would do it ? :-) | 09:24 |
| lucasagomes | but if not I can give it a first stab | 09:25 |
| vdrok | lucasagomes, dtantsur, yuriyz so how do you think, how long should we be backwards compatible with old IPA ramdisks? for one release, like mitaka IPA, newton ironic, or longer, or forever? :) | 09:25 |
| lucasagomes | jimmy_, ^ see my patch there | 09:25 |
| dtantsur | vdrok, one release is pretty standard in openstack | 09:25 |
| lucasagomes | jimmy_, https://review.openstack.org/325151 this will allow you to use pdb and also see prints | 09:25 |
| vdrok | brb | 09:25 |
| lucasagomes | vdrok, yeah I would say 1 release too | 09:26 |
| sambetts | lucasagomes: :/ I don't think it is for England, it looks like an Ireland only holiday | 09:26 |
| vdrok | morning sambetts | 09:26 |
| lucasagomes | sambetts, oh, too bad | 09:26 |
| sambetts | o/ vdrok | 09:26 |
| vdrok | ok, 1 release makes sense | 09:26 |
| lucasagomes | jroll, JayF vdrok I'm thinking on approving https://review.openstack.org/#/c/186700/ , it looks good and just extend the power capabilities with something already supported bu IPMI[tool] | 09:27 |
| lucasagomes | any concerns? | 09:28 |
| vdrok | lucasagomes, I'll take one more look at it in a half an hour or so, but I'm not a core in specs :) | 09:30 |
| lucasagomes | vdrok, cool, np | 09:30 |
| divya | Error finding address for https://172.26.24.110:9292/v1/images/2aee8624-f020-49db-af2d-2ea7b117ad08: [Errno 1] _ssl.c:510: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol | 09:30 |
| *** rcernin is now known as rcernin|lunch | 09:31 | |
| divya | how to resolve this | 09:32 |
| divya_ | Error finding address for https://172.26.24.110:9292/v1/images/2aee8624-f020-49db-af2d-2ea7b117ad08: | 09:32 |
| dtantsur | vsaienko, please let us know if you plan on working on https://bugs.launchpad.net/ironic/+bug/1588339 | 09:33 |
| openstack | Launchpad bug 1588339 in Ironic "[RFE] A command to import Ironic resources from a file" [Wishlist,Confirmed] | 09:33 |
| *** SiRiuS__ has quit IRC | 09:33 | |
| lucasagomes | divya, it's glance related. Plus, as pointed out yesterday this is a *dev* channel not a support channel | 09:35 |
| lucasagomes | dig into it more, find the problem, open a bug if needed | 09:35 |
| lucasagomes | but be more precisely on the questions here and don't flood the channel please | 09:36 |
| -openstackstatus- NOTICE: CI is experiencing issues with test logs, all jobs are currently UNSTABLE as a result. No need to recheck until this is fixed! Thanks for your patience. | 09:37 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add some missed test cases in node object tests https://review.openstack.org/268045 | 09:40 |
| jimmy_ | lucasagomes, oslo_debug_helper works well im my environment!! | 09:43 |
| lucasagomes | jimmy_, cool, if you can please comment on the patch it would be great :-) | 09:43 |
| *** pcaruana has joined #openstack-ironic | 09:44 | |
| jimmy_ | lucasagomes, It seems jenkins is unstable... | 09:44 |
| lucasagomes | :-( | 09:44 |
| lucasagomes | when it's back | 09:44 |
| jimmy_ | lucasagomes, Of course, I will. | 09:45 |
| lucasagomes | thanks a lot | 09:45 |
| *** amotoki has joined #openstack-ironic | 09:47 | |
| sambetts | lucasagomes: I'm looking at that spec, and I'm not sure i see NMI as a power state | 09:51 |
| *** amotoki has quit IRC | 09:51 | |
| sambetts | lucasagomes: I'm stoked about soft reboot, but the NMI part seems like it doesn't fit to me | 09:51 |
| lucasagomes | sambetts, I think this is following the way ipmitool actually sees it? (power on/off/dig) | 09:52 |
| lucasagomes | sambetts, | 09:53 |
| lucasagomes | http://linux.die.net/man/1/ipmitool | 09:53 |
| * lucasagomes is not saying that, just because ipmitool do that we also have to. But we've been basing our api on it for various things, power, boot devices (the name of the boot devices are all based on ipmitool) etc | 09:54 | |
| sambetts | lucasagomes: yeah, it just bugs me because it isn't actually a power state | 09:55 |
| sambetts | and it doesn't change the power state either as far as I can tell | 09:55 |
| lucasagomes | yeah, more like a power "action" or something | 09:55 |
| sambetts | lucasagomes: it just seems like (to me at least) it fits better in the managment interface, unless it actually reboots the machine | 09:57 |
| *** logan_ has joined #openstack-ironic | 09:57 | |
| lucasagomes | sambetts, I don't think it does reboot the machine | 09:58 |
| * lucasagomes have to test | 09:58 | |
| *** logan_ is now known as Guest62971 | 09:58 | |
| sambetts | so its a power action that goes from Power ON to Power ON without a reboot, it doesn't sound like a power action to me | 09:58 |
| *** _milan_ has joined #openstack-ironic | 09:58 | |
| *** _milan_ has quit IRC | 09:59 | |
| *** logan- has quit IRC | 09:59 | |
| *** mkovacik__ has joined #openstack-ironic | 10:01 | |
| *** mkovacik__ is now known as milan_ | 10:01 | |
| *** milan_ has quit IRC | 10:01 | |
| *** milan_ has joined #openstack-ironic | 10:01 | |
| *** Guest62971 is now known as logan- | 10:01 | |
| *** agordeev has quit IRC | 10:02 | |
| *** kozhukalov has quit IRC | 10:02 | |
| *** romcheg has quit IRC | 10:02 | |
| *** evgenyl has quit IRC | 10:02 | |
| *** yonglihe has quit IRC | 10:02 | |
| *** sergek has quit IRC | 10:02 | |
| *** rajinir has quit IRC | 10:02 | |
| *** ChrisMead has quit IRC | 10:02 | |
| lucasagomes | sambetts, yeah, def it's an arguable thing... I'm kinda indifferent, I think I would be happy with either having it on the mgmt interface or as a power command (following what ipmitool does) | 10:04 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Replace vif_portgroup_id with vif_port_id https://review.openstack.org/325197 | 10:04 |
| *** YiboCai has quit IRC | 10:04 | |
| lucasagomes | sambetts, please comment on the spec | 10:04 |
| sambetts | lucasagomes: leaving comments as we speak :) | 10:04 |
| lucasagomes | cool | 10:04 |
| lucasagomes | ty! | 10:04 |
| *** vishwanathj has joined #openstack-ironic | 10:06 | |
| -openstackstatus- NOTICE: CI is experiencing issues with test logs, all jobs are currently UNSTABLE as a result. No need to recheck until this is fixed! Thanks for your patience. | 10:08 | |
| *** ChanServ changes topic to "CI is experiencing issues with test logs, all jobs are currently UNSTABLE as a result. No need to recheck until this is fixed! Thanks for your patience." | 10:08 | |
| sambetts | lucasagomes: do we have a way to set/look at image properties today? so we can enable/disable something based on if an image has a certain feature? | 10:11 |
| *** rajinir has joined #openstack-ironic | 10:15 | |
| *** milan_ has quit IRC | 10:16 | |
| *** Romanenko_K has quit IRC | 10:16 | |
| *** mkoderer has quit IRC | 10:16 | |
| *** slagle has quit IRC | 10:16 | |
| *** lintan has quit IRC | 10:16 | |
| *** xavierr has quit IRC | 10:16 | |
| *** hshiina has quit IRC | 10:17 | |
| *** parlos has quit IRC | 10:17 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Gracefully degrade start_iscsi_target for Mitaka ramdisk https://review.openstack.org/319183 | 10:18 |
| *** agordeev has joined #openstack-ironic | 10:19 | |
| *** kozhukalov has joined #openstack-ironic | 10:19 | |
| *** evgenyl has joined #openstack-ironic | 10:19 | |
| *** rcernin|lunch is now known as rcernin | 10:20 | |
| *** vishwanathj has quit IRC | 10:20 | |
| *** sylwesterB has left #openstack-ironic | 10:20 | |
| *** vishwanathj has joined #openstack-ironic | 10:26 | |
| *** tangchen has quit IRC | 10:27 | |
| *** romcheg has joined #openstack-ironic | 10:29 | |
| *** dtantsur is now known as dtantsur|lunch | 10:30 | |
| vmud213 | Morning all | 10:32 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add debug environment to tox https://review.openstack.org/325151 | 10:34 |
| *** yonglihe has joined #openstack-ironic | 10:35 | |
| *** sergek has joined #openstack-ironic | 10:36 | |
| *** ChrisMead has joined #openstack-ironic | 10:41 | |
| *** ishant is now known as ishant|away | 10:42 | |
| *** phuongnh has quit IRC | 10:44 | |
| *** amotoki has joined #openstack-ironic | 10:48 | |
| *** lucasagomes is now known as lucas-lunch | 10:49 | |
| *** milan_ has joined #openstack-ironic | 10:49 | |
| *** Romanenko_K has joined #openstack-ironic | 10:49 | |
| *** mkoderer has joined #openstack-ironic | 10:49 | |
| *** slagle has joined #openstack-ironic | 10:49 | |
| *** lintan has joined #openstack-ironic | 10:49 | |
| *** xavierr has joined #openstack-ironic | 10:49 | |
| *** orwell.freenode.net changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/developer/ironic/ | Bugs: https://bugs.launchpad.net/ironic" | 10:49 | |
| *** vishwanathj has quit IRC | 10:50 | |
| *** vishwanathj has joined #openstack-ironic | 10:52 | |
| *** amotoki has quit IRC | 10:52 | |
| *** hoangcx has quit IRC | 11:00 | |
| *** thrash|g0ne is now known as thrash | 11:03 | |
| *** vishwanathj has quit IRC | 11:04 | |
| *** vishwanathj has joined #openstack-ironic | 11:05 | |
| *** vdrok has quit IRC | 11:10 | |
| *** vdrok has joined #openstack-ironic | 11:12 | |
| *** vishwanathj has quit IRC | 11:14 | |
| openstackgerrit | Bharath kumar proposed openstack/ironic: adding back the inspect_ports as an optional parameter https://review.openstack.org/325230 | 11:15 |
| *** parlos has joined #openstack-ironic | 11:16 | |
| *** vishwanathj has joined #openstack-ironic | 11:26 | |
| *** tesseract has joined #openstack-ironic | 11:33 | |
| *** d0ugal has quit IRC | 11:34 | |
| *** ishant|away has quit IRC | 11:39 | |
| *** kirankv has quit IRC | 11:39 | |
| *** lazy_prince has quit IRC | 11:39 | |
| jroll | morning y'all | 11:39 |
| sambetts | Morning jim | 11:39 |
| *** kirankv has joined #openstack-ironic | 11:40 | |
| -openstackstatus- NOTICE: CI is experiencing issues with test logs, all jobs are currently UNSTABLE as a result. No need to recheck until this is fixed! Thanks for your patience. | 11:41 | |
| *** ChanServ changes topic to "CI is experiencing issues with test logs, all jobs are currently UNSTABLE as a result. No need to recheck until this is fixed! Thanks for your patience." | 11:41 | |
| jroll | that means it's time to go home, right? | 11:43 |
| sambetts | pretty much heh | 11:43 |
| *** pcaruana has quit IRC | 11:45 | |
| *** vishwanathj has quit IRC | 11:45 | |
| *** ishant|away has joined #openstack-ironic | 11:45 | |
| *** vishwanathj has joined #openstack-ironic | 11:45 | |
| *** lazy_prince has joined #openstack-ironic | 11:46 | |
| *** jimmy_ has quit IRC | 11:46 | |
| *** amotoki has joined #openstack-ironic | 11:48 | |
| *** baoli has joined #openstack-ironic | 11:51 | |
| vmud213 | JayF: I'd like to discuss about https://review.openstack.org/#/c/302143. Just ping me when you are here | 11:52 |
| *** lucas-lunch is now known as lucasagomes | 11:52 | |
| lucasagomes | jroll, morning | 11:52 |
| *** amotoki has quit IRC | 11:53 | |
| *** baoli_ has joined #openstack-ironic | 11:53 | |
| *** baoli has quit IRC | 11:56 | |
| *** appu103_ has quit IRC | 12:00 | |
| *** vishwana_ has joined #openstack-ironic | 12:03 | |
| *** vishwanathj has quit IRC | 12:04 | |
| *** SiRiuS has joined #openstack-ironic | 12:07 | |
| nicodemos | good morning, ironic | 12:08 |
| TheJulia | good morning | 12:09 |
| TheJulia | jroll: I think it means to go back to bed | 12:09 |
| *** sacharya has joined #openstack-ironic | 12:10 | |
| *** dtantsur|lunch is now known as dtantsur | 12:11 | |
| dtantsur | morning jroll, TheJulia, nicodemos, vmud213 and everyone | 12:11 |
| *** trown|outtypewww is now known as trown | 12:11 | |
| TheJulia | good morning dtantsur | 12:11 |
| *** aarefiev has quit IRC | 12:12 | |
| vmud213 | morning dtantsur | 12:12 |
| openstackgerrit | Sam Betts proposed openstack/ironic: [DO NOT MERGE] Testing Cisco CI https://review.openstack.org/289312 | 12:12 |
| *** aarefiev has joined #openstack-ironic | 12:13 | |
| *** dprince has joined #openstack-ironic | 12:14 | |
| *** sacharya has quit IRC | 12:15 | |
| *** daemontool has joined #openstack-ironic | 12:16 | |
| *** joprovost has joined #openstack-ironic | 12:16 | |
| *** vishwana_ has quit IRC | 12:18 | |
| lucasagomes | nicodemos, TheJulia morning | 12:19 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add CLI tests to check JSON response body https://review.openstack.org/291239 | 12:21 |
| TheJulia | good morning lucasagomes | 12:23 |
| xavierr | morning Ironic | 12:25 |
| *** frontrunner has joined #openstack-ironic | 12:25 | |
| xavierr | morning lucasagomes nicodemos TheJulia | 12:25 |
| dtantsur | morning xavierr | 12:25 |
| xavierr | Hey dtantsur :) | 12:25 |
| lucasagomes | xavierr, hey hey :-) | 12:26 |
| *** lindycoder has joined #openstack-ironic | 12:30 | |
| *** alaski is now known as lascii | 12:40 | |
| vdrok | morning vmud213 jroll nicodemos TheJulia xavierr | 12:44 |
| vmud213 | vdrok o/ :) | 12:44 |
| nicodemos | \o vdrok | 12:45 |
| xavierr | hi vdrok o/ | 12:45 |
| nicodemos | lucasagomes, dtantsur, xavierr o/ | 12:45 |
| *** mjturek1 has joined #openstack-ironic | 12:47 | |
| *** piet has joined #openstack-ironic | 12:48 | |
| *** amotoki has joined #openstack-ironic | 12:49 | |
| *** sdake has joined #openstack-ironic | 12:49 | |
| *** daemontool has quit IRC | 12:50 | |
| *** amotoki has quit IRC | 12:54 | |
| *** d0ugal has joined #openstack-ironic | 12:55 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add CLI tests to check JSON response body https://review.openstack.org/291239 | 12:55 |
| *** jcoufal has joined #openstack-ironic | 12:56 | |
| *** vmud213 has quit IRC | 12:56 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Network https://review.openstack.org/294559 | 13:05 |
| *** parlos has quit IRC | 13:07 | |
| vdrok | lucasagomes, sambetts I wonder should we allow things like http://127.0.0.1:6385/v1/nodes/4df80221-8da6-4cbc-8a1d-ef7214401cb7/ports?portgroup=79f9ae41-60e2-4b9f-9319-e465b197a345 in our API? | 13:07 |
| vdrok | or if it is a subcontroller, we just say no additional filtering? | 13:07 |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Uplink Set https://review.openstack.org/295981 | 13:08 |
| lucasagomes | I think that we accept filtering on subcontrollers already, no? | 13:08 |
| *** phschwartz_ is now known as pschwartz | 13:08 | |
| *** pschwartz is now known as phschwartz | 13:09 | |
| *** amotoki has joined #openstack-ironic | 13:09 | |
| vdrok | lucasagomes, ah, yes, I guess we do, api-ref just does not tell that yet | 13:09 |
| lucasagomes | vdrok, yeah :-/ it's undocumented but, urgh, it's some "magic" we have in our code | 13:13 |
| lucasagomes | but deeply, I hope people are not using it heh | 13:14 |
| *** [1]cdearborn has joined #openstack-ironic | 13:15 | |
| vdrok | lucasagomes, then, should we documented at all? :) because now it get weird with ports being subcontroller of both node and portgroup, and portgroup is a subcontroller of a node - https://review.openstack.org/#/c/206244/108/ironic/api/controllers/v1/port.py@277 | 13:15 |
| sambetts | that sounds fine | 13:16 |
| vdrok | s/documented/document it/ | 13:16 |
| *** daemontool has joined #openstack-ironic | 13:16 | |
| lucasagomes | vdrok, I feel a bit divided. But, yeah to be fair I think we should document it + add tests to make sure we don't break it | 13:17 |
| lucasagomes | I don't think we have tests like that on tempest | 13:17 |
| vdrok | nope | 13:17 |
| lucasagomes | my fear is that, if we document it people will use so, we better guarantee that we won't break them in the first place | 13:18 |
| lucasagomes | maybe tests + document | 13:18 |
| TheJulia | good morning vdrok | 13:19 |
| vdrok | o/ :) | 13:20 |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Network https://review.openstack.org/294559 | 13:20 |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Extend python-oneviewclient to Uplink Set https://review.openstack.org/295981 | 13:22 |
| *** wajdi has quit IRC | 13:26 | |
| *** rama_y has joined #openstack-ironic | 13:28 | |
| *** ohamada has quit IRC | 13:31 | |
| *** PollyZ has joined #openstack-ironic | 13:32 | |
| *** joprovost has quit IRC | 13:34 | |
| *** Goneri has quit IRC | 13:34 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add api-ref for new port fields https://review.openstack.org/325299 | 13:35 |
| *** PollyZ has left #openstack-ironic | 13:36 | |
| *** PollyZ__ has joined #openstack-ironic | 13:37 | |
| *** PollyZ__ has quit IRC | 13:37 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add api-ref for new port fields https://review.openstack.org/325299 | 13:39 |
| *** PollyZ has joined #openstack-ironic | 13:39 | |
| *** trown is now known as trown|brb | 13:42 | |
| *** mbound has joined #openstack-ironic | 13:43 | |
| *** thiagop has joined #openstack-ironic | 13:45 | |
| *** mgoddard_ has joined #openstack-ironic | 13:45 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add api-ref for new port fields https://review.openstack.org/325299 | 13:46 |
| *** vdrok has quit IRC | 13:46 | |
| thiagop | Good morning Ironic | 13:48 |
| thiagop | TGIF | 13:48 |
| *** mgoddard has quit IRC | 13:49 | |
| sambetts | Hey thiagop TGIF | 13:49 |
| NobodyCam | Good Morning Ironicers | 13:49 |
| NobodyCam | thiagop morning and Yes TGIF | 13:49 |
| NobodyCam | morning sambetts | 13:49 |
| *** piet has quit IRC | 13:50 | |
| lucasagomes | thiagop, NobodyCam morning | 13:50 |
| NobodyCam | morning TheJulia | 13:50 |
| NobodyCam | morning lucasagomes | 13:50 |
| xavierr | morning NobodyCam | 13:50 |
| NobodyCam | morning xavierr | 13:50 |
| *** dansmith is now known as superdan | 13:51 | |
| *** ametts has joined #openstack-ironic | 13:52 | |
| xavierr | that moment you are looking for ironic on youtube and you see NobodyCam talking about that :) | 13:52 |
| dtantsur | morning thiagop, NobodyCam | 13:53 |
| *** dprince has quit IRC | 13:53 | |
| *** dprince has joined #openstack-ironic | 13:54 | |
| *** piet has joined #openstack-ironic | 13:54 | |
| NobodyCam | doh | 13:56 |
| NobodyCam | morning dtantsur happy Friday | 13:57 |
| dtantsur | Friday is happy as usual, especially with gates broken :D | 13:57 |
| *** trown|brb is now known as trown | 13:57 | |
| NobodyCam | yea | 13:57 |
| NobodyCam | white board has a Check plus for status | 13:58 |
| NobodyCam | just waking up is there a bug I can add to the WhiteBoard? | 13:59 |
| *** ChanServ changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/developer/ironic/ | Bugs: https://bugs.launchpad.net/ironic" | 13:59 | |
| -openstackstatus- NOTICE: Cleanup from earlier block storage disruption on static.openstack.org has been repaired, and any jobs which reported an "UNSTABLE" result or linked to missing logs between 08:00-14:00 UTC can be retriggered by leaving a "recheck" comment. | 13:59 | |
| *** e0ne has quit IRC | 13:59 | |
| *** mgoddard_ has quit IRC | 13:59 | |
| *** e0ne has joined #openstack-ironic | 14:00 | |
| *** mgoddard has joined #openstack-ironic | 14:00 | |
| sambetts | NobodyCam: ^ | 14:00 |
| sambetts | heh | 14:00 |
| NobodyCam | lol | 14:00 |
| *** PollyZ has quit IRC | 14:00 | |
| sambetts | NobodyCam: Morning TGIF | 14:00 |
| NobodyCam | morning sambetts :) | 14:00 |
| *** ayoung has joined #openstack-ironic | 14:00 | |
| NobodyCam | updated status to check - | 14:01 |
| *** mjura has quit IRC | 14:02 | |
| *** rama_y has quit IRC | 14:04 | |
| *** PollyZ has joined #openstack-ironic | 14:08 | |
| *** [1]cdearborn has quit IRC | 14:08 | |
| *** [1]cdearborn has joined #openstack-ironic | 14:09 | |
| milan_ | morning all! | 14:10 |
| milan_ | I'm wondering about this line: https://github.com/openstack/ironic/blob/master/devstack/files/debs/ironic#L12 | 14:10 |
| dtantsur | "morning" milan_ :) | 14:10 |
| milan_ | dtantsur, :P | 14:11 |
| NobodyCam | good morning milan_ | 14:11 |
| milan_ | NobodyCam, morning | 14:11 |
| *** rloo has joined #openstack-ironic | 14:11 | |
| milan_ | w/r that line, ubuntu kernels newer than 3.13 don't seem to require it | 14:11 |
| NobodyCam | good morning and happy Friday rloo :) | 14:11 |
| *** parlos has joined #openstack-ironic | 14:12 | |
| milan_ | moreover, 16.4 LTS doesn't even have that package any more | 14:12 |
| rloo | Hi NobodyCam! I'm not here today (not sure how this started up...) | 14:12 |
| * milan_ not sure what to do about that | 14:12 | |
| *** rloo has quit IRC | 14:12 | |
| *** PollyZ has quit IRC | 14:12 | |
| *** e0ne has quit IRC | 14:14 | |
| *** e0ne has joined #openstack-ironic | 14:14 | |
| milan_ | actually, to make it even more fun: https://github.com/openstack-dev/devstack/blob/master/lib/neutron_plugins/ovs_base#L57 | 14:14 |
| milan_ | so it seems we're getting smarter here than the devstack neutron plugin | 14:15 |
| * milan_ confused | 14:15 | |
| milan_ | any suggestions? should I remove it from ironic's devstack conf? | 14:16 |
| milan_ | sambetts, maybe ^ ;) | 14:17 |
| *** mbound has quit IRC | 14:18 | |
| sambetts | milan_: interesting, we shouldn't need to replicate anything installed by the neutron plugin right? | 14:19 |
| milan_ | sambetts, +1 | 14:19 |
| milan_ | but it's unconditional | 14:19 |
| milan_ | in that file | 14:19 |
| milan_ | not sure what else should be removed | 14:20 |
| milan_ | openvswitch-switch ? | 14:20 |
| *** mjturek1 has quit IRC | 14:20 | |
| * milan_ would like to drop at least the dkms package as it breaks things on 16.4 LTS (my poor grenade VM) | 14:21 | |
| *** Goneri has joined #openstack-ironic | 14:23 | |
| vsaienko | lucasagomes: http://logs.openstack.org/29/299229/8/check/gate-ironic-staging-drivers-dsvm-nv/e6829ef/logs/screen-ir-cond.txt.gz#_2016-06-03_13_32_19_542 | 14:24 |
| sambetts | milan_: most of those depedencies were added over 2 years ago https://review.openstack.org/#/c/70348/ I think we can probably drop at least some of them because they are covered by the other plugins now | 14:25 |
| sambetts | lucasagomes: ^ | 14:25 |
| *** r-daneel has joined #openstack-ironic | 14:25 | |
| *** wajdi has joined #openstack-ironic | 14:25 | |
| lucasagomes | vsaienko, w00t! | 14:25 |
| lucasagomes | that's awesome | 14:25 |
| vsaienko | lucasagomes, I think we can enable all staging drivers by default | 14:26 |
| *** parlos has quit IRC | 14:26 | |
| lucasagomes | sambetts, reading... | 14:26 |
| milan_ | sambetts, lucasagomes preferably we'd remove that from the mitaka as well -- because of grenade testing -- but don't kill me for removing stuff from stable branch right away ;) | 14:26 |
| lucasagomes | vsaienko, on the tests you mean? | 14:27 |
| lucasagomes | milan_, I'm a bit out of context (/me reading scrollback) | 14:27 |
| vsaienko | lucasagomes, yes | 14:28 |
| milan_ | lucasagomes, TL;DR: our devstack/files/debs gives me a headache: we're enforcing a neutron-related package to be installed all the time no matter the circumstances | 14:29 |
| lucasagomes | milan_, if not needed anymore and the base os that we use in the gate is happy with it +1 to remove | 14:29 |
| *** ChrisAusten has joined #openstack-ironic | 14:29 | |
| milan_ | lucasagomes, ack | 14:29 |
| lucasagomes | vsaienko, ++ | 14:29 |
| * milan_ removes | 14:29 | |
| lucasagomes | thanks :D | 14:30 |
| sambetts | :d | 14:31 |
| sambetts | :D | 14:31 |
| *** joprovost has joined #openstack-ironic | 14:32 | |
| *** PollyZ has joined #openstack-ironic | 14:33 | |
| *** e0ne has quit IRC | 14:33 | |
| *** vdrok has joined #openstack-ironic | 14:35 | |
| *** links has quit IRC | 14:36 | |
| *** rama_y has joined #openstack-ironic | 14:37 | |
| *** e0ne has joined #openstack-ironic | 14:37 | |
| jlvillal | vsaienko, ping: https://review.openstack.org/#/c/317082/ Would you be able to respond to the comments? | 14:39 |
| *** mjturek1 has joined #openstack-ironic | 14:39 | |
| vsaienko | jlvillal I will | 14:40 |
| vsaienko | jlvillal I will upload new version soon | 14:41 |
| jlvillal | vsaienko, Spasibo :) | 14:41 |
| *** jcoufal has quit IRC | 14:42 | |
| *** Jabadia has joined #openstack-ironic | 14:43 | |
| *** jcoufal has joined #openstack-ironic | 14:43 | |
| PollyZ | Hi, I am installing dev-stack, even after reading the instruction, I still don't know which driver should I choose | 14:44 |
| NobodyCam | morning jlvillal sambetts | 14:45 |
| jlvillal | NobodyCam, Good morning :) | 14:45 |
| NobodyCam | :) | 14:45 |
| jroll | PollyZ: we typically run devstack with pxe_ssh or agent_ssh, we also support pxe_ipmitool and agent_ipmitool now | 14:46 |
| thiagop | PollyZ: that depends on what you are doing (and wich type or baremetal hardware you are using) | 14:46 |
| *** rcernin has quit IRC | 14:46 | |
| jroll | PollyZ: the sample config will give you a good general start that should work for developing anything that isn't driver-specific | 14:47 |
| NobodyCam | jroll: Morning | 14:47 |
| jroll | hey NobodyCam :D | 14:47 |
| thiagop | PollyZ: If you're only going for a test and not using real hardware, I'd use agent_ssh which is the default on the instruction | 14:47 |
| vdrok | morning jlvillal and NobodyCam | 14:48 |
| NobodyCam | hey vdrok :) morning | 14:48 |
| jlvillal | vdrok, good morning :) | 14:48 |
| *** Goneri has quit IRC | 14:52 | |
| openstackgerrit | Milan Kováčik proposed openstack/ironic: remove neutron stuff from devstack deb packages https://review.openstack.org/325346 | 14:53 |
| lucasagomes | JayF, mind if I pick ur mind for a second? The syslog thing, running out of ideas on coreos :-/ | 14:54 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for openstackclient plugin https://review.openstack.org/301687 | 14:54 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for openstackclient plugin https://review.openstack.org/301687 | 14:55 |
| lucasagomes | JayF, I wonder if a mix between ur suggestion and what is proposed would be acceptable. What about documenting that "distros without systemd should make sure that all logs are redirected under /var/log" | 14:55 |
| lucasagomes | and we continue to use journald for the systemd ones | 14:55 |
| lucasagomes | instead of trying to forward journald -> syslog -> syslogd | 14:56 |
| lucasagomes | fwiw, journald -> syslog is not problem, now coreos doesn't have any syslogd, rsyslog or the old sysklog so :-/ | 14:56 |
| lucasagomes | so the second part, the proxy, is missing | 14:56 |
| jroll | solution: kill the coreos image | 14:56 |
| lucasagomes | jroll, that's another way out heh | 14:57 |
| * jroll sick of the shenanigans | 14:57 | |
| lucasagomes | jroll, the ultimate goal would be to have every distro to log under /var/log, but is it too bad if some distros the logs come from journald and others from /var/log ? | 14:57 |
| TheJulia | I kind of feel a very strong +1 to that, is there a strong case that anyone can make to keep it? | 14:58 |
| lucasagomes | it's more cosmetic IMO, because the problem is having no access to any logs | 14:58 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Remove deprecated alias "root_device_hint" for "raid_device" hook https://review.openstack.org/325351 | 14:58 |
| lucasagomes | TheJulia, the coreos image? Nop, I think the strongest argument is the tooling around it | 14:58 |
| jroll | TheJulia: it's incredibly easy to customize, has ssh access built in, etc | 14:58 |
| *** adu has joined #openstack-ironic | 14:58 | |
| lucasagomes | TheJulia, since we've moved to use chroot there's no security/isolation gain | 14:59 |
| lucasagomes | jroll, would you be ok with the journald + /var/log for non-systemd distros case? | 14:59 |
| jroll | I mean, the agent is meant to destroy a machine, security/isolation was never a reason :) | 14:59 |
| TheJulia | jroll: true, biggest issue I hear from people is using the devuser element to customize their deployment ramdisk when they use diskimage-builder | 14:59 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Remove deprecated alias "root_device_hint" for "raid_device" hook https://review.openstack.org/325351 | 14:59 |
| lucasagomes | jroll, heh fair | 15:00 |
| TheJulia | oh, coreos lacks ipmitool internally so part of the information collection for discovery just doesn't work | 15:00 |
| lucasagomes | jroll, was thinking on the long running agent thingy | 15:00 |
| jroll | TheJulia: right, I want to add ssh keys for my team when I build a ramdisk | 15:00 |
| jroll | lucasagomes: I'm not stongly opinionated, I do agree with jay that distro-specific code in IPA isn't good, but I think eventually we may have no choice | 15:00 |
| TheJulia | jroll: and truthfully, I think that is just everyone needs a bit better doc | 15:00 |
| *** mgoddard_ has joined #openstack-ironic | 15:01 | |
| *** ifarkas has quit IRC | 15:01 | |
| PollyZ | thiagop: it should be only testing | 15:01 |
| jroll | TheJulia: sure | 15:01 |
| jroll | lucasagomes: what about it? | 15:01 |
| lucasagomes | jroll, so it's not really distro specific. I mean, the idea is: If systemd is present use it, if not collect the logs from /var/log | 15:01 |
| lucasagomes | the fallback works for all dsitros | 15:01 |
| TheJulia | jroll: ++ Well, even in a ramdisk it could still be in like a debian chroot.... :) | 15:01 |
| PollyZ | jroll: thank you for your information. working on it :) | 15:01 |
| jroll | PollyZ: np | 15:02 |
| *** joprovost has quit IRC | 15:02 | |
| lucasagomes | most if not all biggest distros now uses systemd, so, the fallback is more to cover the tiny ipa case | 15:02 |
| *** ijw has joined #openstack-ironic | 15:02 | |
| jroll | lucasagomes: let's go back because I don't have context | 15:03 |
| lucasagomes | jroll, ack, maybe if you can skim the comments here https://review.openstack.org/#/c/323511/ ? | 15:03 |
| jroll | yeah | 15:03 |
| *** mtanino has joined #openstack-ironic | 15:04 | |
| *** mgoddard has quit IRC | 15:04 | |
| jroll | lucasagomes: so your original question, What about documenting that "distros without systemd should make sure that all logs are redirected under /var/log" | 15:04 |
| jroll | isn't that what JayF suggested? | 15:05 |
| jroll | " we should instead set a logging standard for ramdisks; i.e. "IPA logs must be in dir X" and then modify the ramdisks to match the standard?" | 15:05 |
| lucasagomes | jroll, he suggested (AFAIUI) that we should proxy the logs from journald to /var/log using syslog | 15:05 |
| lucasagomes | so that the ramdisk would always get the logs from /var/log | 15:05 |
| *** Goneri has joined #openstack-ironic | 15:05 | |
| *** e0ne has quit IRC | 15:05 | |
| jroll | sure, so the problem is you can't do that in coreos? | 15:05 |
| lucasagomes | exactly | 15:06 |
| jroll | hrm | 15:06 |
| lucasagomes | because I can forward the logs to syslog | 15:06 |
| lucasagomes | but coreos does not have any daemon to get from syslog and put it on /var/log | 15:06 |
| lucasagomes | which would be rsyslog, syslogd etc... | 15:06 |
| jroll | :| | 15:06 |
| jroll | well, this is a chroot right | 15:07 |
| jroll | what if we bind-mount /var/log | 15:07 |
| jroll | and redirect IPA logs to /var/log | 15:07 |
| lucasagomes | yeah but I will still miss the system logs | 15:07 |
| lucasagomes | like the logs from boot | 15:07 |
| *** joprovost has joined #openstack-ironic | 15:07 | |
| jroll | 'dmesg > /var/log/xxx' | 15:07 |
| *** e0ne has joined #openstack-ironic | 15:07 | |
| *** e0ne has quit IRC | 15:08 | |
| jroll | OH | 15:08 |
| jroll | urgh, I get it | 15:08 |
| jroll | grr | 15:08 |
| lucasagomes | yeah there's more logs hehe | 15:08 |
| lucasagomes | I would love to have access to as much info as I can get | 15:08 |
| lucasagomes | because it's troubleshooting | 15:08 |
| jroll | the objection is actually calling 'journalctl' from IPA,r ight? | 15:09 |
| lucasagomes | yeah, because tinyipa does have it | 15:09 |
| lucasagomes | that's why I thought about falling back to zip the /var/log + dmesg for distros where it's not present | 15:09 |
| jroll | echo '* * * * * journalctl > /var/log/journal-foo' > crontab | 15:09 |
| jroll | :P | 15:09 |
| lucasagomes | lol | 15:09 |
| dtantsur | that's pretty terrible, unless you do it every 1-2 msec | 15:10 |
| jroll | oh I know | 15:10 |
| dtantsur | as I said on the patch, we already struggle from stripped logs :) | 15:10 |
| lucasagomes | honestly I think that people using ironic will create their ramdisks based on a single distro | 15:10 |
| jroll | does bash have a sleep in ms | 15:10 |
| jroll | ? | 15:10 |
| lucasagomes | I wouldn't mix many distros in my env | 15:10 |
| lucasagomes | so, the logs will be uniform on the same "cloud" | 15:10 |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Correct RAID documentation JSON https://review.openstack.org/321862 | 15:11 |
| lucasagomes | the main problem is not having access to the logs | 15:11 |
| lucasagomes | mix distros == for the deploy ramdisk only, not talking about user image | 15:11 |
| *** mgoddard has joined #openstack-ironic | 15:11 | |
| openstackgerrit | Merged openstack/python-ironic-inspector-client: Updated from global requirements https://review.openstack.org/323916 | 15:11 |
| jroll | echo 'while 1; do journalctl > /var/log/myjournal; sleep .001; done' > some-systemd-unit | 15:11 |
| jroll | (I'm only half serious here) | 15:11 |
| jroll | right, I agree a cloud won't have mixed ramdisk distros | 15:12 |
| jroll | I'm not 100% against calling journalctl, honestly, but I do want jay on board | 15:12 |
| *** mgoddard_ has quit IRC | 15:12 | |
| lucasagomes | jroll, yeah sure. I also want his opnion on it | 15:13 |
| lucasagomes | I just don't want it to be a hard block on it. And if in the future we decide to the kill coreos based ramdisk | 15:13 |
| lucasagomes | we can rethink the idea | 15:13 |
| lucasagomes | it's all internal implementation anyway | 15:14 |
| jroll | I wish we were more opinionated with ipa | 15:14 |
| jroll | single distro, hard requirements on what's in the ramdisk | 15:15 |
| sambetts | if its the SSH part thats keeping people on Coreos over TinyIPA, thats an easy fix | 15:15 |
| lucasagomes | jroll, yeah, we may get there someday | 15:15 |
| jroll | it isn't just ssh, it's the extensibility | 15:16 |
| jroll | I wish I could show our agent builder repo | 15:16 |
| lucasagomes | jroll, be happy that we have only 1 ramdisk implementation at the moment :-) | 15:16 |
| jroll | lucasagomes: >.> | 15:16 |
| * jroll rewrites in go | 15:16 | |
| * lucasagomes is serious | 15:16 | |
| jroll | lucasagomes: are there people that want to implement their own agent? | 15:16 |
| lucasagomes | jroll, I hope not, but we had the bash one before | 15:17 |
| PollyZ | :thiagop: I think the default driver on the instruction is pxe_ssh | 15:17 |
| lucasagomes | and took us > 1 year to get rid of it | 15:17 |
| *** dprince has quit IRC | 15:17 | |
| jroll | ah right | 15:17 |
| *** jcoufal has quit IRC | 15:17 | |
| *** r-daneel has quit IRC | 15:18 | |
| lucasagomes | jroll, TheJulia anyway, thanks for the inputs | 15:18 |
| lucasagomes | I will follow up with JayF when he's available | 15:18 |
| lucasagomes | (review on the spec also appreciated :-) ) | 15:18 |
| jroll | yeah np | 15:19 |
| sambetts | I think the difference between the bash ramdisk and alternate ramdisks is that the bash ramdisk was always intended to be replaced by IPA and therefore we had to use the same driver right, if someone makes an alternate ramdisk I expect them to create their own driver specific to that ramdisk | 15:19 |
| jroll | sambetts: well, it's a protocol, anyone could go build a better ramdisk | 15:20 |
| lucasagomes | sambetts, yea, I was thinking on the bash ramdisk when I said "only 1 implementation" | 15:20 |
| lucasagomes | I hope nobody is working on a separated implementation | 15:20 |
| jroll | also, it wasn't always intended to be replaced by IPA afaik :) | 15:20 |
| lucasagomes | yeah, history | 15:20 |
| * lucasagomes is glad we did tho, bash was a pain | 15:20 | |
| sambetts | ++ | 15:21 |
| jroll | agree | 15:21 |
| *** piet has quit IRC | 15:22 | |
| thiagop | PollyZ: you're right :) | 15:22 |
| *** ChubYann has quit IRC | 15:23 | |
| *** thiagop is now known as thiagop-afk | 15:23 | |
| *** bsilverman has quit IRC | 15:23 | |
| PollyZ | :thiagop on the previous message, you refered agent_ssh as default, so should I use agent or pxe which is the deault driver | 15:25 |
| *** rcernin has joined #openstack-ironic | 15:31 | |
| *** ChubYann has joined #openstack-ironic | 15:35 | |
| *** d0ugal has quit IRC | 15:39 | |
| *** rcernin has quit IRC | 15:39 | |
| JayF | jroll: that's actually the inverse of what I want with IPA | 15:43 |
| JayF | jroll: I want IPA to follow a standard, and as such not care about what is in the ramdisk it's running in | 15:43 |
| jroll | JayF: 'that' being opinionated? | 15:43 |
| jroll | sure | 15:43 |
| *** ccarmack has joined #openstack-ironic | 15:43 | |
| JayF | jroll: I see a day where we need IPA to run on non-x86, maybe even in a non-linux ramdisk | 15:43 |
| JayF | So the "opinion" of IPA should be to explicitly not be distro specific, and follow standards | 15:43 |
| jroll | I think something this low-level, it's very difficult to do that | 15:43 |
| jroll | especially not-linux-specific :| | 15:43 |
| JayF | I think it would've been three lines of config (in the logging case) if pxe image hadn't been slimmed so considerably | 15:44 |
| JayF | (for coreos) | 15:44 |
| JayF | Like in the case of logging, file logging or journald are basically the two standards, so now that we know it's not possible to standardize I think that's reasonable | 15:44 |
| JayF | but we should be vigilant against having distro-specific stuff in IPA | 15:44 |
| JayF | and/or requiring code changes to IPA to run in a different linux distribution | 15:44 |
| jroll | sure | 15:45 |
| jroll | so maybe I should re-phrase | 15:45 |
| jroll | I wish it was possible for us to be more opinionated with IPA | 15:45 |
| JayF | since journald is systemd and a standard, I'm cool with having to do it both ways since doing it one way is technically difficult | 15:45 |
| JayF | heh | 15:45 |
| JayF | I have lots of opinions about IPA ramdisks ... | 15:45 |
| JayF | ... downstream | 15:45 |
| jroll | heh | 15:45 |
| jroll | fair :) | 15:45 |
| *** absubram has joined #openstack-ironic | 15:46 | |
| *** sacharya has joined #openstack-ironic | 15:49 | |
| devananda | morning, all | 15:49 |
| jroll | hai devananda | 15:50 |
| NobodyCam | morning devananda :) happy friday | 15:50 |
| JayF | jroll: did you give me write access to the ironic trello? | 15:52 |
| lucasagomes | devananda, morning | 15:52 |
| jroll | JayF: I did not | 15:52 |
| mat128 | JayF, rama_y: whenever you're ready | 15:52 |
| jroll | JayF: thinking of restricting it to cores for now (see the email I just sent) | 15:53 |
| *** mjturek1 has quit IRC | 15:53 | |
| JayF | jroll: I'm a specs core and an IPA core, are we really going to split that hair? | 15:53 |
| lucasagomes | JayF, so, are you ok with the current proposal + documenting in the image build docs the case for non-systemd distros? | 15:53 |
| JayF | frankly, if I was not a core right now, given my review velocity and quality on IPA, I'd have core on the whole project by now | 15:53 |
| jroll | JayF: oh yeah, specs core ++ | 15:53 |
| rama_y | Good morning mat128, JayF; I am ready | 15:53 |
| * JayF tired of being the last person with partial-core | 15:53 | |
| JayF | lucasagomes: /var/log || journald is fine, since we know coreos image isn't going to cooperate | 15:54 |
| JayF | rama_y: mat128: let me get a hangouts url for us, and i'll pass it on | 15:54 |
| JayF | /win 33 | 15:54 |
| devananda | jroll: keytone v3 and policy support isn't on our priority list currently, but is something I've wanted to hack on - and something we need to do before keystone v2 is completely dropped | 15:54 |
| lucasagomes | JayF, yeah, cool. If we one day kill the coreos then we can revist that and use the proxy | 15:54 |
| jroll | devananda: policy is on the 'small things' list | 15:54 |
| devananda | jroll: cool. also, I've been hacking on it this week. should have patches up for devstack soon, and policy in ironic soon after that | 15:55 |
| jroll | devananda: v3 is also important to me, though you're right that it isn't on any lists | 15:55 |
| JayF | devananda: policy is something I'm working on now | 15:56 |
| JayF | devananda: rama_y, myslef, and mat128 are about to work on the spec for it | 15:56 |
| devananda | JayF: oh? | 15:56 |
| devananda | JayF: I have code working for it already | 15:56 |
| devananda | JayF: let's chat :) | 15:56 |
| openstackgerrit | Chuck Carmack proposed openstack/bifrost: Add redeploy-dynamic playbook https://review.openstack.org/325379 | 15:56 |
| mat128 | devananda: wanna join the workgroup? | 15:57 |
| JayF | devananda: we have a scheduled hangout in 3 minutes | 15:57 |
| mat128 | :) | 15:57 |
| devananda | good timing | 15:57 |
| jroll | lol | 15:57 |
| devananda | yes | 15:57 |
| JayF | devananda: mat128: rama_y: https://hangouts.google.com/hangouts/_/4vwp7752f5fmhozhax746orqpme | 15:57 |
| ccarmack | devananda: wondering if you could review https://review.openstack.org/#/c/325379/ ? | 15:58 |
| *** milan_ has quit IRC | 15:59 | |
| JayF | devananda: mat128: rama_y: https://etherpad.openstack.org/p/ironic-policy-worksession-201600603 | 15:59 |
| dtantsur | morning devananda, JayF | 15:59 |
| JayF | devananda: joining? we'll wait for you if so | 16:00 |
| *** penick has joined #openstack-ironic | 16:01 | |
| dtantsur | jroll, I'm dtantsur on trello I think | 16:01 |
| jroll | dtantsur: added, grabbed lucas too | 16:02 |
| jroll | and devananda | 16:02 |
| jroll | and NobodyCam | 16:02 |
| jroll | people with irc nicks that match trello usernames ++ | 16:03 |
| dtantsur | thnx :) | 16:03 |
| NobodyCam | +++ awesome | 16:03 |
| lucasagomes | thanks | 16:03 |
| * lucasagomes is thinking about shortening his irc handle | 16:04 | |
| lucasagomes | I've "lcs" on freenode, I may use it | 16:04 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for openstackclient plugin https://review.openstack.org/301687 | 16:07 |
| *** sabeen has joined #openstack-ironic | 16:08 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add debug environment to tox https://review.openstack.org/325151 | 16:09 |
| lucasagomes | vdrok, oslotest was already part of our test-requirements ^ (/me missed it) | 16:09 |
| *** garthb has joined #openstack-ironic | 16:11 | |
| *** lindycod_ has joined #openstack-ironic | 16:11 | |
| *** piet has joined #openstack-ironic | 16:12 | |
| *** PollyZ has quit IRC | 16:12 | |
| *** sabeen1 has joined #openstack-ironic | 16:13 | |
| *** trown is now known as trown|lunch | 16:13 | |
| *** kromanenko__ has quit IRC | 16:14 | |
| *** lindycoder has quit IRC | 16:14 | |
| *** Romanenko_K has quit IRC | 16:14 | |
| *** sabeen has quit IRC | 16:15 | |
| *** kromanenko_ has quit IRC | 16:15 | |
| * sambetts -> afk | 16:15 | |
| sambetts | Night all | 16:15 |
| jroll | night \o | 16:15 |
| *** sambetts is now known as sambetts|afk | 16:15 | |
| sambetts|afk | cya jroll | 16:15 |
| *** frontrunner has quit IRC | 16:17 | |
| *** aarefiev has quit IRC | 16:17 | |
| *** bnemec has quit IRC | 16:17 | |
| *** tesseract has quit IRC | 16:19 | |
| openstackgerrit | Yosef Hoffman proposed openstack/ironic-python-agent: lldp-timeout kernel parameter missing ipa- prefix https://review.openstack.org/324608 | 16:19 |
| *** PollyZ has joined #openstack-ironic | 16:19 | |
| *** aarefiev has joined #openstack-ironic | 16:20 | |
| openstackgerrit | Merged openstack/ironic-specs: Update metrics spec to reflect review comments https://review.openstack.org/321153 | 16:20 |
| *** bnemec has joined #openstack-ironic | 16:23 | |
| *** PollyZ has quit IRC | 16:24 | |
| *** frontrunner has joined #openstack-ironic | 16:30 | |
| *** rcernin has joined #openstack-ironic | 16:31 | |
| *** dprince has joined #openstack-ironic | 16:32 | |
| *** [1]cdearborn has quit IRC | 16:33 | |
| *** PollyZ has joined #openstack-ironic | 16:33 | |
| *** rbrud has joined #openstack-ironic | 16:35 | |
| lucasagomes | sambetts|afk, see ya | 16:36 |
| *** PollyZ has quit IRC | 16:38 | |
| *** piet has quit IRC | 16:39 | |
| *** piet has joined #openstack-ironic | 16:39 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add API reference for portgroups https://review.openstack.org/322796 | 16:40 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add api-ref for new port fields https://review.openstack.org/325299 | 16:40 |
| *** SiRiuS has quit IRC | 16:40 | |
| lucasagomes | folks, calling it a day | 16:41 |
| lucasagomes | have a great evening, enjoy the weekend | 16:41 |
| *** r-daneel has joined #openstack-ironic | 16:42 | |
| dtantsur | lucasagomes, g'night | 16:42 |
| dtantsur | also leaving, have a great weekend | 16:42 |
| lucasagomes | it's national holiday on monday here, so I will try to make it to the meeting | 16:42 |
| lucasagomes | dtantsur, g'night! Enjoy the weekend | 16:42 |
| jroll | have a good weekend guys :) | 16:42 |
| *** dtantsur is now known as dtantsur|afk | 16:42 | |
| *** lucasagomes is now known as lucas-afk | 16:42 | |
| *** derekh has quit IRC | 16:44 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add API reference for portgroups https://review.openstack.org/322796 | 16:45 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add api-ref for new port fields https://review.openstack.org/325299 | 16:45 |
| *** bnemec is now known as beekneemech | 16:54 | |
| *** harlowja has joined #openstack-ironic | 16:55 | |
| *** garthb has quit IRC | 16:58 | |
| *** PollyZ has joined #openstack-ironic | 17:03 | |
| *** ElCoyote_ has quit IRC | 17:05 | |
| *** ElCoyote_ has joined #openstack-ironic | 17:06 | |
| SpamapS | TheJulia: whoa, http://logs.openstack.org/34/305634/5/check/gate-bifrost-releasenotes/bd3849a/console.html ... that recheck failed reno in a weird way. any ideas? | 17:07 |
| NobodyCam | morning SpamapS | 17:07 |
| SpamapS | NobodyCam: howdy! | 17:07 |
| SpamapS | TheJulia: https://review.openstack.org/305634 as a reminder. Also hi. ;) | 17:07 |
| TheJulia | SpamapS: wow.... | 17:08 |
| *** _vsaienko has joined #openstack-ironic | 17:09 | |
| *** _vs has joined #openstack-ironic | 17:09 | |
| NobodyCam | doc error? | 17:09 |
| TheJulia | yeah | 17:09 |
| TheJulia | I think so | 17:09 |
| TheJulia | sections defined with no values in the reno file submitted | 17:09 |
| TheJulia | I suspect that would explode it qutie nicely | 17:10 |
| NobodyCam | not line #2 with the > | 17:10 |
| SpamapS | sounds like reno needs to be defensive | 17:10 |
| TheJulia | #2 is valid yaml as far as I can see | 17:10 |
| *** sacharya_ has joined #openstack-ironic | 17:11 | |
| TheJulia | but not the intent of prelude as I understand it | 17:11 |
| SpamapS | NobodyCam: >'s just mean "ignore newlines" | 17:11 |
| TheJulia | yup | 17:11 |
| SpamapS | http://paste.openstack.org/show/507755/ | 17:11 |
| SpamapS | (except the last one) | 17:12 |
| SpamapS | I believe it's called "folded" | 17:12 |
| SpamapS | It's what you use when you want something else to format the data | 17:12 |
| openstackgerrit | Ganesh Maharaj Mahalingam proposed openstack/bifrost: Make ansible installation directory configurable https://review.openstack.org/324789 | 17:12 |
| SpamapS | TheJulia: the reno input is valid yaml, but reno may not allow nulls | 17:12 |
| NobodyCam | yep had to refresh yaml knowledge | 17:12 |
| * SpamapS will try local | 17:13 | |
| TheJulia | SpamapS: that is kind of what I'm thinking actually | 17:13 |
| *** sacharya has quit IRC | 17:13 | |
| *** garthb has joined #openstack-ironic | 17:14 | |
| * SpamapS may have a patch for reno today :) | 17:15 | |
| SpamapS | still breaks on reno master | 17:18 |
| *** yohoffman has joined #openstack-ironic | 17:20 | |
| SpamapS | oh this is annoying | 17:21 |
| *** _vsaienko has quit IRC | 17:21 | |
| *** _vs has quit IRC | 17:21 | |
| SpamapS | I updated to Ubuntu 16.04 so now I can't easily do py34 tests | 17:21 |
| jroll | devananda: what's the difference between doing policy with keystone v2 vs v3, does the client/oslo.policy not just handle it? | 17:28 |
| devananda | jroll: I *think* it will just work in both | 17:30 |
| devananda | jroll: however, the environment setup for me is quite different | 17:30 |
| *** notmorgan has joined #openstack-ironic | 17:30 | |
| notmorgan | o/ | 17:30 |
| devananda | notmorgan: ohhai! | 17:30 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Fix markup in documentation https://review.openstack.org/325432 | 17:30 |
| notmorgan | a wild notmorgan has appeared | 17:30 |
| jroll | devananda: okay, the way JayF described it to me was that it was difficult/undesirable to implement policy support for Keystone v2 | 17:31 |
| TheJulia | ohhh... it is a wild notmorgan | 17:31 |
| TheJulia | Greetings wild notmorgan | 17:31 |
| devananda | notmorgan: great minds think (and time things) alike. JayF already had started looking into adding policy to ironic as well | 17:31 |
| jroll | notmorgan: hai | 17:31 |
| notmorgan | OOOH it's a TheJulia ! woohoo, this is where a bunch of cool people are | 17:31 |
| jroll | duh! | 17:31 |
| notmorgan | jroll: hehe | 17:31 |
| * JayF barely restrains himself from making the obvious "surely that's not morgan" joke | 17:31 | |
| notmorgan | JayF: never restrain from such jokes | 17:32 |
| notmorgan | especially not on fridays | 17:32 |
| notmorgan | but since today is friday.... | 17:32 |
| *** notmorgan is now known as morgan | 17:32 | |
| devananda | hah! | 17:32 |
| morgan | casual nick friday? | 17:32 |
| morgan | :P | 17:32 |
| TheJulia | It feels like a thursday, but it is Friday | 17:32 |
| jroll | heh | 17:32 |
| *** JayF is now known as notJayF | 17:32 | |
| *** devananda is now known as deva | 17:32 | |
| *** notJayF is now known as JasonF | 17:32 | |
| JasonF | There, I'll use my old nick for casual nick friday, lol | 17:33 |
| *** deva is now known as devananda | 17:33 | |
| * TheJulia suspects notjulie would be her choice | 17:33 | |
| * JasonF 's IRC nickname is older than his r/l nickname | 17:33 | |
| morgan | JasonF: so you know if you do not<xxxx> you just make notmyname's life less fun in shared channels :P | 17:33 |
| morgan | anyway | 17:34 |
| morgan | so policy | 17:34 |
| jroll | heh | 17:34 |
| *** devananda is now known as deva | 17:34 | |
| jroll | we have enough j*'s in here that all of our life is not so fun | 17:34 |
| JasonF | https://etherpad.openstack.org/p/ironic-policy-worksession-201600603 is our (very rough) notes from the hour worksession earlier, to give context if some folks don't have it | 17:34 |
| *** deva is now known as devananda | 17:34 | |
| morgan | jroll: i like morgan cause it means people in -infra fall into the mor<tab> trap when talking to mordred | 17:35 |
| morgan | :P | 17:35 |
| devananda | morgan: the question was, I believe, whether, if by implementing support for policy as you and I started, which I know works with keystone v3, whether that will _also_ just work for v2 | 17:35 |
| morgan | or even here. :P | 17:35 |
| morgan | devananda: would work the same for both | 17:35 |
| jroll | hooray | 17:35 |
| morgan | with the exception that all projects are in the "default domain" in keystone v2. | 17:35 |
| morgan | and users must also be* | 17:35 |
| *** [1]cdearborn has joined #openstack-ironic | 17:35 | |
| morgan | * = there is a weird exception to this with auth by ID cases, but assume this is not relevant for your use | 17:36 |
| devananda | JasonF: my gripe on hangouts was that I do not want to, nor presently know how to, set up the right things in keystone v2 to test this | 17:36 |
| devananda | JasonF: and doing it in v3 "just worked" | 17:36 |
| morgan | basically, the token data is extracted, converted into the headers | 17:36 |
| morgan | and oslo_policy consumes that and compares against the policy.json | 17:36 |
| JasonF | devananda: cool, yeah, that makes sense | 17:36 |
| morgan | asserting authz success/failure | 17:36 |
| morgan | now. keep in mind that if you use a dedicated "ironic" domain, that is incompat with v2 (but v2 is deprecated, dying, going away, shuffled off the mortal coil...etc <insert more dead parrot jokes>) | 17:37 |
| devananda | morgan: s/domain/tenant/ ? | 17:38 |
| morgan | devananda: domain is the v3 construct | 17:38 |
| morgan | project is under a domain | 17:38 |
| jroll | morgan: for context, my reaction to "we need v3 to make this work right" was "wat" and "how do I bribe dolph and friends to finally get me v3 downstream" | 17:38 |
| morgan | in v2 all projects (for compat) end up in the domain "default" | 17:38 |
| morgan | jroll: most deployments support v3 now. devstack mostly is all v3, we expect v3-only to be gated on in newton | 17:39 |
| devananda | morgan: so the way we want to isolate access to ironic resources for non-cloud-admin users is to assign them to a specific "baremetal" domain, and create specific roles within that | 17:39 |
| morgan | devananda: right. | 17:39 |
| jroll | morgan: right, I understand, my deployment (rackspace) does not as I'm sure you've heard :) | 17:39 |
| morgan | jroll: which i *think* means you're not fully defcore compliant :P but that... is a different topic for a different time. | 17:40 |
| morgan | (v3 auth is required for defcore last i checked - with exceptions etc) | 17:40 |
| jroll | morgan: well, it's complicated, our user facing thing supports it or something | 17:40 |
| *** htruta has joined #openstack-ironic | 17:40 | |
| * devananda gets a bucket of blue paint for the bikeshed | 17:41 | |
| jroll | idk all the details except that I can't use v2 | 17:41 |
| morgan | jroll: yeah i know you guys have an insane thing ;) | 17:41 |
| morgan | jroll: but that aside | 17:41 |
| *** ijw has quit IRC | 17:41 | |
| morgan | devananda: it better be #0033cc | 17:41 |
| morgan | devananda: /bikeshed | 17:41 |
| morgan | anyway, you can also use a "project" instead of a domain (project==tenant) that users need a role on to access ironic api | 17:42 |
| devananda | jroll: so - the policy checking that I'm hacking on ought to work in keystone v2, however, the way I think we should restrict access (eg, by adding users to a special "baremetal" domain) will not | 17:42 |
| devananda | or maybe it will, sort of? | 17:42 |
| * jroll wonders where we're getting bikesheddy here | 17:42 | |
| morgan | domain_scope is not commonly used (even in keystone) | 17:42 |
| morgan | where most everything else uses project_scope. *however* i think ironic's use is much more in line with a domain -- you don't want resources on this thing, it's a policy-authz only thing | 17:43 |
| *** piet has quit IRC | 17:43 | |
| morgan | and projects tend to have the ability to own resources. | 17:43 |
| jroll | devananda: surely role: would work on both | 17:43 |
| jroll | assign the role to a domain/tenant/project/whatever | 17:44 |
| morgan | i need to check to see if we can assign a role across domains to a user (we should, if we don't... ugh) | 17:44 |
| morgan | jroll: correct. | 17:44 |
| *** daemontool has quit IRC | 17:44 | |
| morgan | jroll: it works on either it's a question of how the policy file references it: domain_id or project_id | 17:44 |
| jroll | morgan: well, the policy file should use roles and then a domain or project with that role has access, no? | 17:45 |
| morgan | (and you ... unfortunately... need to use the _id_ since it's the only thing that is guaranteed "unique" | 17:45 |
| morgan | jroll: yes. you would say something like-- | 17:45 |
| morgan | jroll: bearmetal:node:list => role:admin and project_id:<project_id_expected> | 17:46 |
| jroll | "ironic_admin": "role:ironic_admin or role:admin" | 17:46 |
| devananda | morgan: right -- by combinding role name and project_id | 17:46 |
| *** thrash is now known as thrash|biab | 17:46 | |
| morgan | jroll: be careful, that allows for any one on any project (or domain) with the role "ironic_admin" or "admin" to access the apis | 17:46 |
| jroll | well, I would hope not to use specific project_ids | 17:46 |
| jroll | ah | 17:47 |
| jroll | :/ | 17:47 |
| morgan | yes i know :( | 17:47 |
| devananda | jroll: now you see the conundrum | 17:47 |
| jroll | so if I wanted 10 accounts to have access... what's the best thing? have them all in a project with the right access? | 17:47 |
| morgan | it's a behavior i dislike. i would like a set of "global" roles that cannot ever be attached to a project | 17:47 |
| morgan | jroll: assign the users <role X> on <known project> | 17:47 |
| jroll | also, does "project_id:foo or domain_id:foo" work between both versions? | 17:48 |
| morgan | yes. | 17:48 |
| jroll | ok, cool | 17:48 |
| morgan | because empty values should not *ever* break things | 17:48 |
| morgan | and it's always project_id iirc now | 17:48 |
| jroll | (though I guess a deployer could use whichever matches the right keystone version) | 17:48 |
| morgan | even if it was a v2 token | 17:48 |
| morgan | it's fine to use logical ors in the policy DSL | 17:49 |
| jroll | oh, so project_id works everywhere then? | 17:49 |
| devananda | eg, "ironic_admin": "role:ironic_admin and (project_id:baremetal or domain_id:baremetal)", | 17:49 |
| devananda | ? | 17:49 |
| morgan | devananda: yep | 17:49 |
| morgan | though the parens are superfluous | 17:49 |
| morgan | as and preceeds or in the DSL | 17:49 |
| morgan | but it makes it easier to read | 17:50 |
| devananda | right - but improve readability | 17:50 |
| *** morgan is now known as notmorgan | 17:50 | |
| notmorgan | that should 100% work | 17:50 |
| notmorgan | except ....... | 17:51 |
| notmorgan | you can't use project_names iirc | 17:51 |
| notmorgan | must be the unfriendly uuids | 17:51 |
| notmorgan | unless you explicitly build the context to include the name(s). | 17:51 |
| devananda | wat? | 17:51 |
| notmorgan | and if you use names .. remember you must always specify the domain as well | 17:51 |
| notmorgan | since project names are not globally unique | 17:52 |
| notmorgan | only domains are... | 17:52 |
| devananda | notmorgan: so that should be "project_id:baremetal AND domain_id:baremetal" ? | 17:52 |
| notmorgan | eh. hard to explain on irc... sec. trying to figure that out | 17:52 |
| notmorgan | so. project_id expects the project_id (uuid) always | 17:53 |
| *** vdrok has quit IRC | 17:53 | |
| notmorgan | i do not think we can check against project name from the context provided from keystone middleware at the moment | 17:53 |
| notmorgan | let me see what KSM passes down | 17:53 |
| *** piet has joined #openstack-ironic | 17:54 | |
| *** praneshp has joined #openstack-ironic | 17:54 | |
| notmorgan | so ok. we *do* pass down project_domain_name, domain_name, project_name | 17:55 |
| notmorgan | so we should be able to encode that in the policy dsl. but i have never tried it | 17:55 |
| notmorgan | devananda: if you are using project or domain via name it would be: | 17:56 |
| notmorgan | ironic_admin: "role:ironic_admin and ((project_name:baremetal and domain_name:baremetal) or domain_name:baremetal) | 17:56 |
| notmorgan | i *think*. | 17:57 |
| thiagop-afk | I did some time ago, it worked | 17:57 |
| * notmorgan will need to create a keystone instance with that domain structure and snag a token to check against the policy file. | 17:57 | |
| thiagop-afk | (like 2 years ago...) | 17:57 |
| *** mjturek1 has joined #openstack-ironic | 17:58 | |
| devananda | notmorgan: that looks redundant | 17:58 |
| notmorgan | devananda: the first or is if you're using a role on the project | 17:58 |
| notmorgan | devananda: first part of the or. | 17:58 |
| devananda | notmorgan: yes, but it's superceded by the second part of the or | 17:59 |
| notmorgan | devananda: the second paryt would be if you're using a domain role. | 17:59 |
| *** yolanda has quit IRC | 17:59 | |
| notmorgan | oh wait sorry | 17:59 |
| notmorgan | ironic_admin: "role:ironic_admin and ((project_name:baremetal and project_domain_name:baremetal) or domain_name:baremetal)" | 17:59 |
| notmorgan | there | 17:59 |
| devananda | ahhh | 17:59 |
| devananda | that looks logicaly correct | 17:59 |
| notmorgan | and i don't 100% know if that works as you'd expect from a keystone token -> policy dict | 18:00 |
| devananda | i'll test it! | 18:00 |
| notmorgan | because i don't know what gets shoved in for the matcher off the top of my head (i can figure it out though) | 18:00 |
| * notmorgan nods. | 18:00 | |
| notmorgan | now if you're in v2, you'd need to do project_domain_name:default | 18:01 |
| notmorgan | or... uh | 18:01 |
| notmorgan | sorry project_domain_id:default | 18:01 |
| notmorgan | (because the name is mutable and seems to be either Default or default or something else weird) | 18:01 |
| notmorgan | but the id is always "default" (and cannot be changed) | 18:01 |
| notmorgan | [and i *think* ksm populates "default" in the v2 case... and if it doesn't it's a massive security hole i need to fix] | 18:02 |
| jroll | so I got totally lost here | 18:02 |
| jroll | but | 18:02 |
| *** PollyZ has quit IRC | 18:03 | |
| jroll | I'm thinking the best thing to do here is just use domain_id in the default | 18:03 |
| jroll | and document if you're on v2, use project_id | 18:03 |
| devananda | https://gist.github.com/devananda/beec03ac995a6e780ec66749949592a9 | 18:04 |
| devananda | JasonF: notmorgan: how's that look ^ ? | 18:04 |
| JasonF | that's pretty clean | 18:05 |
| JasonF | can you push up the prototype code you used to get there, too? | 18:05 |
| devananda | oops - it's also backwards | 18:05 |
| devananda | JasonF: not yet, but I will | 18:06 |
| devananda | fixed the gist -- observer should be able to get/list :) | 18:06 |
| notmorgan | devananda: ok so, lets consider. do we want to recommend using a domain or a project? | 18:09 |
| devananda | notmorgan: you tell me | 18:09 |
| notmorgan | project is more v2/v3 compat. | 18:09 |
| notmorgan | (key is "more") | 18:09 |
| notmorgan | hmm. | 18:09 |
| thiagop-afk | devananda: your is_cloud_admin is saying that basically every user that has admin role is cloud admin (in any project) | 18:09 |
| JasonF | thiagop-afk: yes, that's required for backwards compatibility | 18:10 |
| JasonF | thiagop-afk: oooh, you're saying it has to be admin in agmin? | 18:10 |
| notmorgan | devananda: ok, so lets recommend a project. most people don't know how to get a domain scoped role elegantly | 18:10 |
| *** clif is now known as clif_h | 18:10 | |
| notmorgan | devananda: just thinking that domain-scoped token is just not commonly used. | 18:10 |
| devananda | thiagop-afk: the current default policy rule is: "admin_api": "role:admin or role:administrator", | 18:11 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/bifrost: Updated from global requirements https://review.openstack.org/325442 | 18:11 |
| *** trown|lunch is now known as trown | 18:12 | |
| *** PollyZ has joined #openstack-ironic | 18:12 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/324827 | 18:13 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-inspector: Updated from global requirements https://review.openstack.org/324828 | 18:13 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-lib: Updated from global requirements https://review.openstack.org/325450 | 18:13 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/325451 | 18:13 |
| thiagop-afk | devananda: which default? I thought you were using https://github.com/openstack/keystone/blob/master/etc/policy.v3cloudsample.json as basis | 18:13 |
| JasonF | thiagop-afk: in Ironic, today | 18:13 |
| devananda | thiagop-afk: ironic/etc/ironic/policy.json | 18:13 |
| JasonF | https://github.com/openstack/ironic/blob/master/etc/ironic/policy.json | 18:13 |
| thiagop-afk | we have policy on Ironic? Seriously? | 18:14 |
| devananda | yes | 18:14 |
| JasonF | but today it only blocks password viewing | 18:14 |
| thiagop-afk | Never saw that, sorry | 18:14 |
| devananda | it doesn't _do_ much -- but it's been there for a while | 18:14 |
| devananda | all it really does is enforce "are you an admin or not" and "can you see passwords" | 18:14 |
| JasonF | It's going to do a lot more soon :) | 18:14 |
| notmorgan | devananda: and your gist looks correct as long as it tests sanely | 18:15 |
| *** joprovost has quit IRC | 18:20 | |
| thiagop-afk | going out for the day | 18:23 |
| thiagop-afk | see ya | 18:23 |
| *** thiagop-afk has quit IRC | 18:24 | |
| *** cdearborn has joined #openstack-ironic | 18:27 | |
| *** mgoddard has quit IRC | 18:27 | |
| *** ijw has joined #openstack-ironic | 18:28 | |
| *** joprovost has joined #openstack-ironic | 18:32 | |
| JasonF | Hey btw; I usually plug my own reviews, but devananda's api-ref stuff is gorgeous rendered; if you haven't reviewed it yet you should (and it's on my list for today) | 18:33 |
| *** joprovost1 has joined #openstack-ironic | 18:36 | |
| devananda | eg, https://review.openstack.org/#/c/313187/ | 18:37 |
| *** joprovost has quit IRC | 18:37 | |
| *** joprovost1 is now known as joprovost | 18:37 | |
| JasonF | https://review.openstack.org/#/q/topic:api-ref+and+project:openstack/ironic is what I was using, it catches the other patch too | 18:37 |
| devananda | ah - better link | 18:38 |
| *** rbrud has quit IRC | 18:42 | |
| *** [1]cdearborn has quit IRC | 18:42 | |
| *** thrash|biab is now known as thrash | 18:43 | |
| *** dprince has quit IRC | 18:45 | |
| openstackgerrit | Ganesh Maharaj Mahalingam proposed openstack/bifrost: Make ansible installation directory configurable https://review.openstack.org/324789 | 18:46 |
| *** dprince has joined #openstack-ironic | 18:46 | |
| *** PollyZ_ has joined #openstack-ironic | 18:48 | |
| *** rama_y has quit IRC | 18:49 | |
| *** PollyZ has quit IRC | 18:50 | |
| *** adu has quit IRC | 18:55 | |
| TheJulia | gmmaha: is 324789 ready to go? | 18:55 |
| openstackgerrit | Merged openstack/bifrost: Remove invalid directory_mode from ironic install https://review.openstack.org/322267 | 18:55 |
| gmmaha | TheJulia: yes, i think i have got all the things fixed | 18:55 |
| gmmaha | hopefully i havent missed anything | 18:56 |
| gmmaha | i thougth about updating docs, but not sure quite sure hwo to put this down | 18:56 |
| TheJulia | ok, thanks! | 18:56 |
| gmmaha | without setting this flag, the doc is true as is..:) | 18:56 |
| TheJulia | Yeah :) | 19:00 |
| *** rbrud has joined #openstack-ironic | 19:04 | |
| *** rama_y has joined #openstack-ironic | 19:13 | |
| TheJulia | gmmaha: I think 324789 needs a release note or some sort of documentation | 19:13 |
| TheJulia | also, I think your about to merge conflict on testing changes | 19:14 |
| TheJulia | :( | 19:14 |
| gmmaha | TheJulia: ohh.. | 19:14 |
| gmmaha | i can try and rebase the patch.. shouldnt be hard.. | 19:15 |
| gmmaha | and for the doc change, do suggestions on where i should land that? | 19:15 |
| *** piet has quit IRC | 19:16 | |
| openstackgerrit | Merged openstack/bifrost: Unify testing scripts https://review.openstack.org/297424 | 19:16 |
| TheJulia | gmmaha: README.vagrant.rst and likely a note in README.rst I think | 19:21 |
| TheJulia | gmmaha: the revision that merge conflicts just landed | 19:21 |
| gmmaha | TheJulia: thanks.. let me do the rebase and start fixing up the docs | 19:22 |
| *** ametts has quit IRC | 19:22 | |
| openstackgerrit | Chuck Carmack proposed openstack/bifrost: Add redeploy-dynamic playbook https://review.openstack.org/325379 | 19:22 |
| gmmaha | TheJulia: yeah saw that.. much easier change to jump onto and was thinking about unificaiton when i saw the places i had to make the INSTALL_ROOT change | 19:22 |
| *** PollyZ_ has quit IRC | 19:23 | |
| *** garthb_ has joined #openstack-ironic | 19:24 | |
| *** garthb has quit IRC | 19:25 | |
| *** nicodemos is now known as nico-afk | 19:25 | |
| TheJulia | gmmaha: oh, sorry! I wasn't thinking for you to try and take that on | 19:25 |
| TheJulia | I was just kind of tossing out "this would be kind of awesome and useful... we should do this... at some point" if that makes sense | 19:25 |
| gmmaha | TheJulia: :) totally.. i was just thinking that might be a nice thing to have hwen i was making my change | 19:26 |
| cinerama | we actually talked about this yesterday | 19:26 |
| TheJulia | but I can totally see the ansible_install_root too | 19:26 |
| TheJulia | cinerama: sorry for missing that discussion, i've been crazy busy. :( | 19:27 |
| *** yolanda has joined #openstack-ironic | 19:27 | |
| TheJulia | gmmaha: awesome :) | 19:27 |
| TheJulia | greetings yolanda | 19:27 |
| * TheJulia wonders if yolanda has some awesome coffee | 19:28 | |
| yolanda | hi TheJulia , how are you? | 19:28 |
| * yolanda is trying to setup a new laptop | 19:28 | |
| cinerama | gmmaha, i haven't looked at your updated change yet, working on a couple of other things | 19:28 |
| cinerama | gmmaha, but thanks for contributing it! | 19:29 |
| gmmaha | cinerama: aah, my pleasure.. :) | 19:29 |
| gmmaha | and no rush on my change.. will need to update the docs as well.. so should have a final review-able version soon | 19:29 |
| openstackgerrit | Chuck Carmack proposed openstack/bifrost: Add redeploy-dynamic playbook https://review.openstack.org/325379 | 19:29 |
| TheJulia | yolanda: crazy busy :( | 19:30 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Patch to test Grenade. https://review.openstack.org/317751 | 19:30 |
| *** piet has joined #openstack-ironic | 19:30 | |
| yolanda | TheJulia, not sure if that's good or bad :) | 19:30 |
| ccarmack | Hi, I'm new to bifrost… is there a way to have bifrost_inventory call ironic only for a subset of nodes? | 19:34 |
| TheJulia | yolanda: it is a mixed curse I think | 19:35 |
| TheJulia | ccarmack: yes... I think so, one moment, I think there is a standing example | 19:35 |
| ccarmack | TheJulia: thanks :-) | 19:35 |
| *** adu has joined #openstack-ironic | 19:36 | |
| TheJulia | ccarmack: Something like https://github.com/openstack/bifrost/blob/master/playbooks/example-deploy-all-available-nodes.yaml ? :) | 19:36 |
| jroll | it's too nice out, see y'all tomorrow | 19:37 |
| TheJulia | jroll: enjoy! | 19:37 |
| TheJulia | jroll: tomorrow is saturday! | 19:37 |
| jroll | heh | 19:37 |
| jroll | see y'all monday :D | 19:37 |
| jroll | have a good weekend TheJulia | 19:37 |
| ccarmack | TheJulia: I'd like to set BIFROST_INVENTORY_SOURCE to gather data from ironic, but only for a few nodes in my env (otherwise using ironic as the source goes to all my nodes) | 19:37 |
| *** jjohnson2 has joined #openstack-ironic | 19:38 | |
| TheJulia | so out of the box, we just don't have that capability :( | 19:38 |
| ccarmack | TheJulia: ok, thats what I found reading the code…I was going to try extracting the shade function out of bifrost_inventory.py and calling it node by node | 19:40 |
| ccarmack | I tried making a json file for the inventory source, but I don't know things like the nic mac addresses… | 19:40 |
| TheJulia | ccarmack: Maybe node discovery, but we don't have that enabled | 19:42 |
| TheJulia | and coreos can't discover the bmc address :( | 19:43 |
| TheJulia | ccarmack: you could also create a search of sorts via an env var... since you can't pass an argument really to the inventory program, a env var that you could use as a search | 19:44 |
| ccarmack | TheJulia: ok… that's an option… do you think adding this support to bifrost_inventory is a good addition? | 19:45 |
| TheJulia | without fully understanding the scenario, it _might_ be to do some sort of search, or to have an env var that defines a list of nodes to get instead of get_all_machines from shade | 19:47 |
| TheJulia | yeah, would be kind of easy to do either actually if blah exists, collect machines and build the data structure one at a time | 19:47 |
| ccarmack | ok, I might take a crack at that.. | 19:50 |
| TheJulia | ccarmack: feel free to ping me if you need a [in]sanity check ;) | 19:51 |
| * TheJulia wonders how crazy she can make herself look right now | 19:51 | |
| ccarmack | thanks TheJulia | 19:53 |
| TheJulia | no problem :) | 19:57 |
| * TheJulia goes back to reviews | 19:57 | |
| *** joprovost has quit IRC | 19:58 | |
| -openstackstatus- NOTICE: The infrastructure team is taking Gerrit offline for maintenance this afternoon, beginning shortly after 20:00 UTC. We aim to have it back online around 00:00 UTC. | 19:59 | |
| *** rama_y has quit IRC | 20:00 | |
| *** e0ne has joined #openstack-ironic | 20:00 | |
| TheJulia | ugh, well I know what I'm doing in the morning then :) | 20:00 |
| *** joprovost has joined #openstack-ironic | 20:00 | |
| *** absubram has quit IRC | 20:01 | |
| *** adu has quit IRC | 20:01 | |
| NobodyCam | so much for catching up on reviews | 20:01 |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Make boolean usage consistent across playbooks https://review.openstack.org/321310 | 20:03 |
| TheJulia | It has been one of those weeks :( | 20:05 |
| *** baoli_ has quit IRC | 20:06 | |
| *** rama_y has joined #openstack-ironic | 20:07 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/bifrost: Updated from global requirements https://review.openstack.org/325442 | 20:07 |
| * TheJulia blinks | 20:07 | |
| NobodyCam | lol... didn't I just recheck that one | 20:08 |
| -openstackstatus- NOTICE: Gerrit is offline for maintenance until 00:00 UTC | 20:09 | |
| *** ChanServ changes topic to "Gerrit is offline for maintenance until 00:00 UTC" | 20:09 | |
| NobodyCam | there we go | 20:10 |
| cinerama | doh | 20:11 |
| * TheJulia thinks projector + beer == ?? | 20:13 | |
| NobodyCam | ++ | 20:14 |
| *** rcernin has quit IRC | 20:15 | |
| *** wajdi_ has joined #openstack-ironic | 20:18 | |
| *** praneshp has quit IRC | 20:19 | |
| *** wajdi__ has joined #openstack-ironic | 20:20 | |
| *** wajdi has quit IRC | 20:22 | |
| *** wajdi_ has quit IRC | 20:23 | |
| *** wajdi__ has quit IRC | 20:24 | |
| devananda | notmorgan: I am now getting this error. "baremetal:node:get on ..... disallowed by policy (HTTP 500)" | 20:26 |
| devananda | notmorgan: anything immediately come to mind? | 20:27 |
| *** PollyZ has joined #openstack-ironic | 20:28 | |
| notmorgan | huh not off the top of my head | 20:28 |
| *** amotoki has quit IRC | 20:30 | |
| devananda | k | 20:30 |
| PollyZ | when installing devstack, how to get the password for the credentials on the local_conf, or do not need them | 20:30 |
| *** yohoffman has quit IRC | 20:30 | |
| JasonF | PollyZ: What exactly do you mean? The passwords are set in the local conf | 20:32 |
| *** frontrunner has quit IRC | 20:32 | |
| JasonF | PollyZ: also the openrc file should help you get creds sourced into your environment | 20:32 |
| *** persia has quit IRC | 20:33 | |
| *** [1]cdearborn has joined #openstack-ironic | 20:34 | |
| *** persia has joined #openstack-ironic | 20:35 | |
| *** baoli has joined #openstack-ironic | 20:37 | |
| *** flwang1 has quit IRC | 20:38 | |
| *** amotoki has joined #openstack-ironic | 20:39 | |
| *** amotoki has quit IRC | 20:39 | |
| *** PollyZ has quit IRC | 20:41 | |
| *** joprovost has quit IRC | 20:43 | |
| *** lindycod_ has quit IRC | 20:43 | |
| *** dprince has quit IRC | 20:43 | |
| *** ayoung has quit IRC | 20:44 | |
| *** flwang has joined #openstack-ironic | 20:46 | |
| *** e0ne has quit IRC | 20:46 | |
| *** cdearborn has quit IRC | 20:49 | |
| *** wajdi has joined #openstack-ironic | 20:58 | |
| *** piet has quit IRC | 20:59 | |
| *** wajdi has quit IRC | 21:02 | |
| *** Goneri has quit IRC | 21:03 | |
| *** mbound has joined #openstack-ironic | 21:03 | |
| xavierr | guys, while trying to cleaning a node (automatic option) I'm getting this error in last_error field: "Timeout reached while cleaning the node. Please check if the ramdisk responsible for the cleaning is running on the node. " | 21:04 |
| xavierr | any idea how to have it checked? | 21:04 |
| *** ccarmack1 has joined #openstack-ironic | 21:04 | |
| *** ccarmack has quit IRC | 21:08 | |
| *** jjohnson2 has quit IRC | 21:08 | |
| *** mbound has quit IRC | 21:11 | |
| *** fragatin_ has joined #openstack-ironic | 21:19 | |
| *** harlowja has quit IRC | 21:20 | |
| *** fragatina has quit IRC | 21:22 | |
| *** praneshp has joined #openstack-ironic | 21:22 | |
| *** rbrud has quit IRC | 21:22 | |
| *** ElCoyote_ has quit IRC | 21:22 | |
| *** baoli has quit IRC | 21:22 | |
| *** baoli has joined #openstack-ironic | 21:24 | |
| *** ccarmack1 has quit IRC | 21:24 | |
| *** lascii is now known as alaski | 21:27 | |
| *** trown is now known as trown|outtypewww | 21:30 | |
| *** daemontool has joined #openstack-ironic | 21:31 | |
| *** baoli has quit IRC | 21:37 | |
| *** ccarmack has joined #openstack-ironic | 21:38 | |
| *** ccarmack has left #openstack-ironic | 21:38 | |
| *** PollyZ has joined #openstack-ironic | 21:39 | |
| PollyZ | JasonF: I tried with the source, but it gives me "Warning: setting legancy OS TENANT NAME to support cli tools" | 21:40 |
| JasonF | PollyZ: that warning is fine to ignore | 21:40 |
| JasonF | PollyZ: at least for getting started :) | 21:40 |
| PollyZ | but when I run "./stack.sh", it told me "line 494: generate-subunit: command not found", and the rest source command gives me "command not found" | 21:43 |
| PollyZ | JasonF: I should install QEMU? | 21:45 |
| devananda | JasonF: familiar with our api stack? I'm stuck on debugging why I'm not getting the right header in the right place | 21:45 |
| JasonF | PollyZ: are you following the ironic devstack howto to the letter? | 21:46 |
| JasonF | PollyZ: I've never had trouble following that | 21:46 |
| Madasi | PollyZ: did you find https://bugs.launchpad.net/devstack/+bug/1547379 ? | 21:46 |
| openstack | Launchpad bug 1547379 in devstack "./stack.sh: line 488: generate-subunit: command not found" [Undecided,Invalid] | 21:46 |
| JasonF | PollyZ: if yuo're having trouble, I'd blow it away and start over. Make sure to use http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack that guide | 21:47 |
| JasonF | Madasi: nice | 21:47 |
| Madasi | it's got a couple suggestions, but no promises :) | 21:47 |
| JasonF | devananda: I'm good at troubleshooting things, even if I don't know lots about it | 21:47 |
| JasonF | devananda: so try me :) | 21:47 |
| devananda | JasonF: I put a print just after this line https://github.com/openstack/ironic/blob/master/ironic/api/hooks.py#L71 so I can see what the value is for auth_token | 21:47 |
| devananda | JasonF: in my env, it *does not* match what I see in the output of "openstack -v -v baremetal list" | 21:48 |
| devananda | I believe this is why all my auth checks are failing -- the valid token I'm passing to Ironic is, somehow, not making it into pecan.request | 21:48 |
| JasonF | Why not just dump all the headers there? | 21:49 |
| JasonF | And see if the other values are what you expect? | 21:49 |
| JasonF | That might lend a clue to what is happening | 21:49 |
| devananda | JasonF: tried that. having trouble dereferencing the webob magic | 21:50 |
| JasonF | devananda: is this a devstack? | 21:50 |
| devananda | I get a lot of this sort of thing | 21:50 |
| devananda | HEADERS: <webob.headers.EnvironHeaders object at 0x7f6be2af7588> | 21:50 |
| devananda | JasonF: no | 21:50 |
| devananda | JasonF: my plan was: get it working in a python venv first, then get it working in devstack, then push it up | 21:51 |
| JasonF | I was asking more for my ability to locally reproduce | 21:51 |
| JasonF | than anything else | 21:51 |
| devananda | I have python venv for ironic and keystone, both using a local mysql & rabbit (in a VM, technically) | 21:52 |
| JasonF | https://github.com/Pylons/webob/blob/39ac465c84c4c33b1c3c7b9700d6da5dfbf84692/webob/headers.py#L124 | 21:53 |
| JasonF | does .keys show anything interesting/ | 21:53 |
| devananda | nope | 21:53 |
| devananda | another webob object reference | 21:53 |
| JasonF | that seems ... wrongish? | 21:53 |
| devananda | HEADERS: <webob.headers.EnvironHeaders object at 0x7f8eb2ae05f8> | 21:54 |
| devananda | KEYS: <filter object at 0x7f8eb2ae04e0> | 21:54 |
| devananda | AUTH TOKEN: 17e0a13056fb42c0aed39788aabaf6a9 | 21:54 |
| devananda | I agree | 21:55 |
| JasonF | wrap that KEYS call in list() | 21:55 |
| JasonF | and I think it'll make that readable for you | 21:55 |
| devananda | hah! yes | 21:56 |
| JasonF | now is that interesting? | 21:56 |
| JasonF | or is it just another wtf-rabbithole | 21:56 |
| *** _vsaienko has joined #openstack-ironic | 21:57 | |
| *** _vs has joined #openstack-ironic | 21:57 | |
| *** yolanda has quit IRC | 21:57 | |
| PollyZ | JasonF: I used Debian, maybe that is why | 21:58 |
| devananda | JasonF: confirms that that object is real, but the same question remains | 21:58 |
| devananda | here is the content now: | 21:58 |
| devananda | KEYS: ['Accept: application/json', 'X-Auth-Token: e6ce11ef9d214e5db0f35cbb734e2818', 'Connection: keep-alive', 'User-Agent: python-ironicclient', 'X-Openstack-Ironic-Api-Version: 1.6', 'Host: 127.0.0.1:6385', 'Content-Type: applica | 21:58 |
| devananda | tion/json', 'Accept-Encoding: gzip, deflate'] | 21:58 |
| devananda | however -- that's still not the same auth token that my client claims to be sending | 21:59 |
| JasonF | can you gist the full client output ov -vv along with the HEADERS/KEYS/AUTH_TOKEN for that request? | 21:59 |
| * JasonF has a small hunch bit it's a weak one | 22:00 | |
| devananda | notmorgan: does openstackclient represent auth tokens in a different way than they are stored in the keystone db ? | 22:02 |
| devananda | notmorgan: like -- if I'm comparing the output of "openstack -v -v ..." when it prints the REQ curl command, it includes this: -H "X-Auth-Token: {SHA1}150b... | 22:03 |
| devananda | but I do not see that token in the keystone DB | 22:03 |
| devananda | what am I missing? | 22:03 |
| JasonF | $ echo e6ce11ef9d214e5db0f35cbb734e2818 | sha1sum | 22:04 |
| JasonF | 7b3d1bf62894f5cc773b45410dea14cab2dd0835 -a | 22:04 |
| devananda | gah. serioulsy | 22:04 |
| JasonF | I think it sha1's the actual token in the log | 22:04 |
| JasonF | is that it? | 22:04 |
| devananda | nope | 22:04 |
| JasonF | if you wanna gist that output from openstack client I'd be real interested though | 22:05 |
| devananda | i just trunkated the keystone 'token' table and reran the command ONCE | 22:05 |
| devananda | yea, gisting now | 22:05 |
| JasonF | at this point I understand more about our api stack than I ever have before | 22:05 |
| JasonF | wanna keep troubleshooting so I keep learning :) | 22:05 |
| *** JasonF is now known as JayF | 22:05 | |
| *** fragatin_ has quit IRC | 22:09 | |
| devananda | JayF: https://gist.github.com/devananda/c83b23c5a666679997404204f8415ee5 | 22:10 |
| *** fragatina has joined #openstack-ironic | 22:10 | |
| JayF | the tokens match in this example | 22:11 |
| *** [1]cdearborn has quit IRC | 22:11 | |
| JayF | the token in the RESP BODY is db6[snip]446 and that's what was logged | 22:11 |
| devananda | JayF: take a look at line 63 | 22:16 |
| devananda | JayF: you're right - it's the correct auth token id. the sha1sum thing was throwing me off earlier | 22:16 |
| devananda | JayF: however, the local context is still missing all the information about that user | 22:16 |
| notmorgan | devananda: uhm. | 22:16 |
| notmorgan | devananda: if you're using fernet there are not toekns in the db | 22:17 |
| notmorgan | devananda: if tou use uuid, they will be indexed by the short token. | 22:17 |
| notmorgan | id (32bytes hex) | 22:17 |
| notmorgan | devananda: and yes the {sha1} is different than what is in the db | 22:17 |
| notmorgan | that is a security thing | 22:18 |
| devananda | notmorgan: got it | 22:18 |
| JayF | devananda: you don't pass in any of that context, it appears | 22:18 |
| devananda | JayF: I found the problem | 22:18 |
| JayF | devananda: is Ironic supposed to fetch it from the auth token and apply it? | 22:18 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/api/hooks.py#L121 is simply non-functional | 22:18 |
| notmorgan | devananda: it's the sha of the id so the output is sane to log / paste / etc | 22:18 |
| notmorgan | since token ids are considered secure info (they are your AuthZ context) | 22:18 |
| JayF | devananda: TrustedCallHook returns nothing when I search against the pecan codebase | 22:19 |
| devananda | JayF: that's defined in Ironic | 22:19 |
| devananda | the problem seems to be that we're not pulling the user, roles, etc, out of the auth token | 22:19 |
| devananda | but rather, looking for them in the headers | 22:20 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/api/hooks.py#L73 | 22:20 |
| JayF | yeah, traced it back now | 22:20 |
| devananda | that "creds" dict gets stored into the pecan.request.context | 22:20 |
| devananda | which is then checked everywhere else | 22:20 |
| JayF | that makes a lot of sense | 22:20 |
| JayF | Ironic will have to add an api call to keystone to fetch info about the user with token | 22:20 |
| devananda | but policy.enforce() doesn't do anything with the auth token | 22:20 |
| devananda | yes | 22:21 |
| JayF | I suspect other apps have to do the same thing? | 22:21 |
| devananda | I dunno? | 22:21 |
| devananda | notmorgan: thoughts ^ ? | 22:21 |
| notmorgan | devananda: uhm.. sec, catching up | 22:21 |
| JayF | I'll look at the implementation of it in another project | 22:21 |
| notmorgan | middleware should provide all the info you need | 22:22 |
| notmorgan | keystonemiddleware for a token | 22:22 |
| notmorgan | you shouldn't need to ask keystone outside of that unless ytou're doing things other than consuming the token data | 22:22 |
| notmorgan | JayF: don't look at keystone for an example, we're bad, look at nova for how they receive it | 22:22 |
| notmorgan | JayF: or cinder | 22:23 |
| JayF | nova is what I have open | 22:23 |
| * notmorgan nods. | 22:23 | |
| *** _vsaienko has quit IRC | 22:24 | |
| *** _vs has quit IRC | 22:24 | |
| *** daemontool has quit IRC | 22:26 | |
| devananda | notmorgan: something's not doing that, at least in my env | 22:26 |
| notmorgan | devananda: i'll need to circle back on that, time for me to head north | 22:27 |
| devananda | notmorgan: ack. ciao! | 22:27 |
| notmorgan | devananda: and enjoy the weekend :) | 22:27 |
| JayF | devananda: https://github.com/openstack/nova/blob/master/nova/api/auth.py#L74 I can't find in here either where nova is fetching that info | 22:27 |
| JayF | devananda: it looks like it expects them in the headers | 22:27 |
| notmorgan | JayF: that would be what i would expect | 22:27 |
| devananda | JayF: expects them in the headers, when not in the token, i thought? | 22:27 |
| devananda | http://docs.openstack.org/developer/keystonemiddleware/api/keystonemiddleware.auth_token.html#what-auth-token-adds-to-the-request-for-use-by-the-openstack-service | 22:27 |
| devananda | hm, no, never mind | 22:28 |
| devananda | this isn't about the "service user" | 22:28 |
| *** mbound has joined #openstack-ironic | 22:30 | |
| JayF | https://github.com/openstack/nova/blob/2a0fce7a44d4050e112a6b987f24236627632d2c/nova/tests/unit/api/test_auth.py#L38 | 22:31 |
| JayF | even in the tests, they always pass in tenantid, auth token, and a service catalog | 22:31 |
| * JayF is missing something | 22:32 | |
| devananda | hah! | 22:33 |
| devananda | found my problem | 22:34 |
| JayF | https://github.com/openstack/nova/blob/ced89e7b26b3cff323852e1d8a9c6db80334f4dd/etc/nova/api-paste.ini#L77 hmm... looks like api-paste has them using keystonemiddleware.auth_token | 22:34 |
| devananda | if pecan_config.app.enable_acl: | 22:34 |
| devananda | app = acl.install(app, cfg.CONF, pecan_config.app.acl_public_routes) | 22:34 |
| *** PollyZ has quit IRC | 22:34 | |
| devananda | I had removed those two lines because I'm implementing the ACL checks at a different layer | 22:34 |
| devananda | however | 22:34 |
| JayF | ah I had read that code and assumed it was still being run | 22:34 |
| devananda | that's actually where the keystonemiddleware is getting loaded | 22:35 |
| JayF | aha | 22:35 |
| JayF | yep, and I traced it all the way back through for Nova too | 22:36 |
| JayF | IDK if I've been helpful in tshooting this, but this is super helpful to me at least | 22:36 |
| JayF | yep, and the keystonemiddleware AuthProtocol classes handle populating all that data | 22:37 |
| jroll | the middleware should return that stuff | 22:39 |
| * jroll now sees deva found it | 22:39 | |
| *** Jabadia has quit IRC | 22:40 | |
| devananda | aaaaand its working | 22:44 |
| devananda | also, gerrit is down | 22:44 |
| jroll | hence my early friday (why am I here) | 22:44 |
| JayF | niiice | 22:44 |
| devananda | JayF: https://github.com/devananda/ironic/commit/a0250681a233257f2ca4f89767b7f96a8d4bb16d | 22:46 |
| JayF | yep that's essentially identical to how nova is doing it | 22:47 |
| JayF | glorious | 22:47 |
| JayF | devananda: am I going to block you on anything if I don't have that draft spec up until Monday morning/ | 22:50 |
| *** harlowja has joined #openstack-ironic | 22:51 | |
| *** lindycoder has joined #openstack-ironic | 22:52 | |
| *** sabeen has joined #openstack-ironic | 22:54 | |
| *** sabeen1 has quit IRC | 22:55 | |
| devananda | JayF: nope | 22:57 |
| devananda | updated my commit, because that one didn't quite work, and this one is much better: https://github.com/devananda/ironic/commit/86dd3c8e9b06b87972494a6a251e9a029cbae113 | 22:57 |
| JayF | cool; thanks. got tied up in some unrelated things earlier and wanted to push that to the weekend | 22:57 |
| devananda | I'll push the code to gerrit later today / tomorrow | 22:57 |
| devananda | next steps for me: add policy checks to all the other places in the code, then start testing in devstack | 22:57 |
| devananda | I expect there will be a few changes, eg. creating the 'baremetal' tenant in ironic/devstack/lib/ironic and such as well | 22:58 |
| devananda | before we can test and merge this code | 22:58 |
| JayF | absolutely, but from my perspective, I thought I was going to be doing most of this work when I woke up this morning | 22:58 |
| devananda | but that one ^ should be enough to get anyone else started | 22:58 |
| devananda | hehehe :) | 22:58 |
| JayF | and by this afternoon someone else has a proof of concept working | 22:58 |
| JayF | I won the open source community lottery \o/ | 22:58 |
| JayF | lol | 22:58 |
| devananda | JayF: I'd been meaning to hack on this for a few months, and notmorgan and I finally found the rihgt time to sit down together, for the last two days, and brainstorm | 22:59 |
| devananda | anyway, time for me to go offline and enjoy the friday air | 22:59 |
| *** praneshp has quit IRC | 22:59 | |
| JayF | Absolutely, I'm going to do the same and do that spec this weekend with fresher eyes | 22:59 |
| devananda | JayF: thanks (in advance) for writing the spec :) | 22:59 |
| JayF | no problem | 23:00 |
| jroll | devananda: just a nit, json doesn't allow comments, does policy.json? :P | 23:00 |
| JayF | it certainly makes github angry, lol | 23:00 |
| jroll | heh yeah, that's what my thought was | 23:00 |
| devananda | yea, it works just fine ... | 23:00 |
| jroll | interesting | 23:01 |
| * jroll wonders what parser that uses | 23:01 | |
| devananda | inorite? | 23:01 |
| devananda | magic oslo parsers | 23:01 |
| *** jrist has quit IRC | 23:01 | |
| jroll | lovely | 23:01 |
| devananda | FYI - how I'm bootstrapping my local keystone env to test this -- https://gist.github.com/devananda/c7f6fdf105ed9ce948a18223632682a2 | 23:01 |
| *** lindycoder has quit IRC | 23:01 | |
| *** jrist has joined #openstack-ironic | 23:02 | |
| jroll | TIL yaml.safe_load can load json O_o | 23:02 |
| jroll | >>> yaml.safe_load('{"foo": "bar"}') | 23:03 |
| jroll | {'foo': 'bar'} | 23:03 |
| jroll | https://github.com/openstack/oslo.policy/blob/master/oslo_policy/policy.py#L311 | 23:03 |
| *** lindycoder has joined #openstack-ironic | 23:03 | |
| JayF | https://github.com/openstack/nova/blob/master/etc/nova/policy.json https://github.com/openstack/cinder/blob/master/etc/cinder/policy.json https://github.com/openstack/glance/blob/master/etc/policy.json | 23:04 |
| *** r-daneel has quit IRC | 23:04 | |
| JayF | none of those have # comments in the example | 23:04 |
| jroll | nice patch though devananda, I like | 23:05 |
| jroll | JayF: they probably don't realize the parser can handle it, comments are invalid in json | 23:05 |
| JayF | Sure; I'd say if the file ends in .json it should be json though | 23:05 |
| jroll | (one of the arguments for yaml) | 23:05 |
| jroll | also, I'd love to go yaml here (it's supported), but idk if it's worth being different | 23:05 |
| jroll | ++ | 23:05 |
| JayF | I mean, if we do policy in code | 23:05 |
| *** lindycoder has quit IRC | 23:05 | |
| JayF | that file is going to be a hell of a lot smaller | 23:06 |
| jroll | ah true | 23:06 |
| jroll | another ++ for you | 23:06 |
| *** sabeen has quit IRC | 23:08 | |
| *** sabeen has joined #openstack-ironic | 23:08 | |
| *** harlowja has quit IRC | 23:10 | |
| *** harlowja has joined #openstack-ironic | 23:10 | |
| *** sabeen has quit IRC | 23:11 | |
| *** lindycoder has joined #openstack-ironic | 23:11 | |
| *** sabeen has joined #openstack-ironic | 23:11 | |
| *** adu has joined #openstack-ironic | 23:16 | |
| *** causten_ has joined #openstack-ironic | 23:17 | |
| *** ppiela has quit IRC | 23:17 | |
| *** lindycoder has quit IRC | 23:18 | |
| *** ChrisAusten has quit IRC | 23:19 | |
| *** adu has quit IRC | 23:32 | |
| *** mtanino has quit IRC | 23:37 | |
| *** praneshp has joined #openstack-ironic | 23:38 | |
| *** causten_ has quit IRC | 23:42 | |
| *** mbound has quit IRC | 23:46 | |
| *** johnthetubaguy has quit IRC | 23:53 | |
| *** johnthetubaguy has joined #openstack-ironic | 23:53 | |
| *** agordeev has quit IRC | 23:53 | |
| *** evgenyl has quit IRC | 23:53 | |
| *** agordeev has joined #openstack-ironic | 23:54 | |
| *** evgenyl has joined #openstack-ironic | 23:55 | |
| *** garthb_ has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!