| *** piet_ has quit IRC | 00:00 | |
| *** mbound has quit IRC | 00:00 | |
| *** hoangcx_ has joined #openstack-ironic | 00:07 | |
| *** rbudden has quit IRC | 00:07 | |
| *** hoangcx has quit IRC | 00:10 | |
| *** hoangcx_ is now known as hoangcx | 00:10 | |
| *** garthb_ has quit IRC | 00:14 | |
| *** rcernin has joined #openstack-ironic | 00:17 | |
| *** sdake has quit IRC | 00:18 | |
| *** rama_y has quit IRC | 00:18 | |
| rloo | NobodyCam: maybe. or maybe not. Let me see... | 00:19 |
|---|---|---|
| *** rcernin has quit IRC | 00:22 | |
| *** sacharya_ has quit IRC | 00:23 | |
| NobodyCam | :) | 00:23 |
| *** jaybeale has joined #openstack-ironic | 00:25 | |
| *** mtanino has quit IRC | 00:41 | |
| *** Sukhdev has joined #openstack-ironic | 00:44 | |
| *** Sukhdev has quit IRC | 00:45 | |
| *** rcernin has joined #openstack-ironic | 00:48 | |
| *** rcernin has quit IRC | 00:53 | |
| *** mbound has joined #openstack-ironic | 01:01 | |
| *** rcernin has joined #openstack-ironic | 01:06 | |
| *** mbound has quit IRC | 01:06 | |
| *** hoangcx_ has joined #openstack-ironic | 01:07 | |
| *** ijw_ has quit IRC | 01:07 | |
| *** rcernin has quit IRC | 01:10 | |
| *** hoangcx_ has quit IRC | 01:11 | |
| *** hoangcx_ has joined #openstack-ironic | 01:13 | |
| *** jaybeale has quit IRC | 01:14 | |
| *** hoangcx_ has quit IRC | 01:15 | |
| rloo | NobodyCam: not sure if it is your lucky day or not, but I looked at it :) | 01:21 |
| *** rloo has quit IRC | 01:21 | |
| NobodyCam | Thank you rloo :-) | 01:23 |
| *** sdake has joined #openstack-ironic | 01:28 | |
| *** jaybeale has joined #openstack-ironic | 01:30 | |
| *** sdake has quit IRC | 01:32 | |
| *** wajdi has quit IRC | 01:39 | |
| *** wajdi has joined #openstack-ironic | 01:39 | |
| *** baoli has joined #openstack-ironic | 01:41 | |
| *** wajdi has quit IRC | 01:45 | |
| *** links has joined #openstack-ironic | 01:54 | |
| *** sacharya has joined #openstack-ironic | 01:55 | |
| *** jrist has quit IRC | 01:56 | |
| *** sacharya_ has joined #openstack-ironic | 01:57 | |
| *** sacharya has quit IRC | 02:00 | |
| *** jrist has joined #openstack-ironic | 02:03 | |
| *** PollyZ has joined #openstack-ironic | 02:20 | |
| *** PollyZ has quit IRC | 02:25 | |
| *** hoangcx_ has joined #openstack-ironic | 02:30 | |
| *** hoangcx_ has quit IRC | 02:30 | |
| *** abcxyz has joined #openstack-ironic | 02:31 | |
| *** baoli has quit IRC | 02:32 | |
| *** garthb has joined #openstack-ironic | 02:40 | |
| *** PollyZ has joined #openstack-ironic | 02:42 | |
| *** jaybeale has quit IRC | 02:45 | |
| *** yuanying has quit IRC | 02:50 | |
| *** yuanying has joined #openstack-ironic | 02:53 | |
| *** PollyZ has quit IRC | 02:55 | |
| *** PollyZ has joined #openstack-ironic | 02:55 | |
| *** links has quit IRC | 02:56 | |
| *** PollyZ has quit IRC | 02:58 | |
| *** kalpase has joined #openstack-ironic | 03:00 | |
| *** ijw has joined #openstack-ironic | 03:19 | |
| *** fragatina has quit IRC | 03:22 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: [WIP] Add keystone policy support to Ironic https://review.openstack.org/325599 | 03:22 |
| *** PollyZ has joined #openstack-ironic | 03:30 | |
| *** linuxgeek has quit IRC | 03:39 | |
| *** PollyZ has quit IRC | 03:41 | |
| *** jaybeale has joined #openstack-ironic | 03:43 | |
| *** vmud213 has joined #openstack-ironic | 03:43 | |
| *** yuanying_ has joined #openstack-ironic | 03:48 | |
| *** yuanying has quit IRC | 03:49 | |
| *** moshele has joined #openstack-ironic | 03:49 | |
| *** linuxgeek has joined #openstack-ironic | 03:50 | |
| *** Sukhdev has joined #openstack-ironic | 03:50 | |
| *** abcxyz has quit IRC | 03:57 | |
| *** abcxyz has joined #openstack-ironic | 03:57 | |
| *** jaybeale has quit IRC | 04:05 | |
| *** links has joined #openstack-ironic | 04:05 | |
| *** ijw has quit IRC | 04:15 | |
| *** PollyZ has joined #openstack-ironic | 04:19 | |
| *** vishwanathj has joined #openstack-ironic | 04:27 | |
| *** skramaja has quit IRC | 04:29 | |
| *** sacharya_ has quit IRC | 04:33 | |
| *** PollyZ has quit IRC | 04:35 | |
| *** vmud213 has quit IRC | 04:36 | |
| *** vishwanathj has quit IRC | 04:36 | |
| *** vmud213 has joined #openstack-ironic | 04:39 | |
| *** skramaja has joined #openstack-ironic | 04:45 | |
| *** wajdi has joined #openstack-ironic | 04:46 | |
| *** vmud213 has quit IRC | 04:48 | |
| *** fragatina has joined #openstack-ironic | 04:52 | |
| *** fragatina has quit IRC | 04:53 | |
| *** fragatina has joined #openstack-ironic | 04:53 | |
| *** garthb has quit IRC | 04:55 | |
| openstackgerrit | Nisha Agarwal proposed openstack/proliantutils: Fix the logic for fetching the gpu device count https://review.openstack.org/315402 | 05:03 |
| *** abcxyz has quit IRC | 05:05 | |
| *** abcxyz has joined #openstack-ironic | 05:05 | |
| *** rama_y has joined #openstack-ironic | 05:10 | |
| *** bcornec1 has quit IRC | 05:12 | |
| *** abcxyz has quit IRC | 05:17 | |
| *** bcornec1 has joined #openstack-ironic | 05:24 | |
| *** abcxyz has joined #openstack-ironic | 05:26 | |
| *** abcxyz has quit IRC | 05:37 | |
| *** appu103_ has joined #openstack-ironic | 05:41 | |
| *** abcxyz has joined #openstack-ironic | 05:42 | |
| *** rama_y has quit IRC | 05:44 | |
| *** appu103_ is now known as aparnav | 05:44 | |
| *** ChubYann has quit IRC | 05:51 | |
| *** moshele has quit IRC | 05:53 | |
| *** mbound has joined #openstack-ironic | 05:55 | |
| *** mbound has quit IRC | 06:00 | |
| *** yuanying has joined #openstack-ironic | 06:03 | |
| openstackgerrit | Yuiko Takada Mori proposed openstack/ironic-specs: Nova compatible serial console support https://review.openstack.org/319505 | 06:04 |
| *** yuanying_ has quit IRC | 06:05 | |
| *** flwang has quit IRC | 06:05 | |
| *** aparnav has quit IRC | 06:05 | |
| *** appu103_ has joined #openstack-ironic | 06:06 | |
| *** bcornec1 has quit IRC | 06:07 | |
| *** yolanda has joined #openstack-ironic | 06:13 | |
| *** rbartal has joined #openstack-ironic | 06:13 | |
| *** yolanda_ has joined #openstack-ironic | 06:13 | |
| *** yolanda_ has quit IRC | 06:14 | |
| *** itamarl has joined #openstack-ironic | 06:15 | |
| *** openstackgerrit has quit IRC | 06:18 | |
| *** openstackgerrit has joined #openstack-ironic | 06:18 | |
| *** yuanying has quit IRC | 06:21 | |
| *** bcornec1 has joined #openstack-ironic | 06:23 | |
| *** yuanying has joined #openstack-ironic | 06:28 | |
| *** bcornec2 has joined #openstack-ironic | 06:29 | |
| *** rcernin has joined #openstack-ironic | 06:29 | |
| *** flwang has joined #openstack-ironic | 06:30 | |
| *** abcxyz has quit IRC | 06:33 | |
| *** appu103_ has quit IRC | 06:45 | |
| *** appu103_ has joined #openstack-ironic | 06:45 | |
| openstackgerrit | lokesh s proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 06:45 |
| *** ifarkas has joined #openstack-ironic | 06:49 | |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: [WIP]Introducing Inspector Grenade testing https://review.openstack.org/327667 | 06:50 |
| milan | #pixiesay morning Ironic! -m flexing | 06:51 |
| PixieBoots | ᕙʕ⇀ᴥ⇀ʔᕗ: morning Ironic! | 06:51 |
| *** e0ne has joined #openstack-ironic | 06:51 | |
| *** athomas has joined #openstack-ironic | 06:56 | |
| openstackgerrit | lokesh s proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 06:57 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Move tempest configuration to test-config phase https://review.openstack.org/329779 | 07:03 |
| *** tesseract has joined #openstack-ironic | 07:03 | |
| *** rcernin has quit IRC | 07:04 | |
| *** rcernin has joined #openstack-ironic | 07:04 | |
| *** yuanying has quit IRC | 07:05 | |
| *** yuanying has joined #openstack-ironic | 07:08 | |
| *** milan has quit IRC | 07:17 | |
| *** dhellmann has quit IRC | 07:20 | |
| *** dhellmann has joined #openstack-ironic | 07:20 | |
| *** abcxyz has joined #openstack-ironic | 07:27 | |
| *** Sukhdev has quit IRC | 07:29 | |
| *** openstackgerrit has quit IRC | 07:33 | |
| *** openstackgerrit has joined #openstack-ironic | 07:33 | |
| openstackgerrit | Aparna proposed openstack/proliantutils: Sort values of raid_config for 'share_physical_disks' https://review.openstack.org/326319 | 07:36 |
| *** kalpase has quit IRC | 07:38 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add test to verify ironic multitenancy https://review.openstack.org/269157 | 07:38 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 07:38 |
| *** links has quit IRC | 07:39 | |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Warn on legacy auth options on conductor start https://review.openstack.org/326660 | 07:49 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Update devstack plugin with new auth options https://review.openstack.org/297312 | 07:49 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Migrate to using keystoneauth Sessions https://review.openstack.org/236982 | 07:49 |
| *** dtantsur|afk is now known as dtantsur | 07:50 | |
| dtantsur | Morning Ironic | 07:50 |
| dtantsur | JayF, oh, I didn't tell you? yeah, I've researched the hw manager examples, it was very helpful | 07:51 |
| dtantsur | JayF, I wonder how we could make them a part of our documentation | 07:51 |
| *** jtomasek_ has joined #openstack-ironic | 07:53 | |
| *** links has joined #openstack-ironic | 07:56 | |
| *** ccamacho has joined #openstack-ironic | 07:56 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:03 | |
| *** milan has joined #openstack-ironic | 08:06 | |
| *** kalpase has joined #openstack-ironic | 08:10 | |
| *** kalpase has quit IRC | 08:10 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c https://review.openstack.org/318440 | 08:10 |
| *** jtomasek_ has quit IRC | 08:10 | |
| alineb | morning all | 08:12 |
| *** ohamada has joined #openstack-ironic | 08:16 | |
| *** Romanenko_K has joined #openstack-ironic | 08:16 | |
| *** kromanenko__ has joined #openstack-ironic | 08:16 | |
| *** kromanenko_ has joined #openstack-ironic | 08:16 | |
| *** bcornec1 has quit IRC | 08:19 | |
| *** dmk0202 has joined #openstack-ironic | 08:22 | |
| *** stendulker has joined #openstack-ironic | 08:28 | |
| *** hkominos has joined #openstack-ironic | 08:32 | |
| *** bcornec1 has joined #openstack-ironic | 08:33 | |
| *** abcxyz has quit IRC | 08:34 | |
| hkominos | hello openstackers. Quick question . In order to create an image for Ironic, is it possible to use the same .img images which are used for VMs? (for example the ubuntu cloud images) | 08:34 |
| *** abcxyz has joined #openstack-ironic | 08:35 | |
| dtantsur | hkominos, iirc yes | 08:36 |
| dtantsur | morning alineb | 08:36 |
| alineb | o/ dtantsur | 08:38 |
| *** bcornec1 has quit IRC | 08:39 | |
| hkominos | thank you | 08:41 |
| *** derekh has joined #openstack-ironic | 08:45 | |
| *** bcornec1 has joined #openstack-ironic | 08:46 | |
| openstackgerrit | lokesh s proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 08:49 |
| *** mbound has joined #openstack-ironic | 08:52 | |
| *** abcxyz has quit IRC | 08:57 | |
| *** abcxyz has joined #openstack-ironic | 08:57 | |
| *** electrofelix has joined #openstack-ironic | 08:59 | |
| *** bcornec2 has quit IRC | 09:00 | |
| *** wagiel has joined #openstack-ironic | 09:00 | |
| vdrok | morning ironic, milan dtantsur and alineb | 09:02 |
| wagiel | Hi, can you help me get the local.conf for the devstack with real hardware involved instead of vms? | 09:02 |
| dtantsur | morning vdrok | 09:03 |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Add validation of 'ilo_deploy_iso' in deploy.validate() https://review.openstack.org/329837 | 09:03 |
| * dtantsur never used devstack with hardware | 09:03 | |
| milan | morning vdrok dtantsur alineb wagiel and all the Ironicers :) | 09:04 |
| dtantsur | hey milan | 09:04 |
| alineb | morning vdrok, milan :) | 09:04 |
| vdrok | wagiel: I did not try that myself too, but take a look here - https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L839 | 09:06 |
| vdrok | you basically need to set IRONIC_IS_HARDWARE to true in local conf and deploy driver to be eg agent_ipmitool instead of agent_ssh | 09:07 |
| vdrok | the deploy driver you should use depends on your hardware of course | 09:07 |
| *** jtomasek_ has joined #openstack-ironic | 09:08 | |
| wagiel | yes ive read all that already hoping that someone tried that already and have working example | 09:08 |
| vdrok | wagiel: then all the nodes should be described in hwinfo file, its format is described here - https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L66-L90 | 09:08 |
| vdrok | wagiel: I think sambetts|afk has it, you might want to wait while he's online then :) | 09:09 |
| wagiel | perfect i will try myself and wait | 09:10 |
| *** mgould|afk is now known as mgould | 09:10 | |
| mgould | morning dtantsur hkominos alineb vdrok milan wagiel and everyone else! | 09:10 |
| vdrok | good morning mgould | 09:11 |
| milan | morning mgould :) | 09:11 |
| dtantsur | hey mgould | 09:11 |
| alineb | morning mgould | 09:11 |
| *** jtomasek_ has quit IRC | 09:13 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add --wait flag for provision actions and wait_for_provision_state function https://review.openstack.org/328191 | 09:13 |
| *** abcxyz has quit IRC | 09:14 | |
| *** abcxyz has joined #openstack-ironic | 09:15 | |
| sambetts|afk | Morning all | 09:16 |
| *** sambetts|afk is now known as sambetts | 09:16 | |
| dtantsur | morning sambetts | 09:16 |
| sambetts | wagiel: this is the local.conf I use for deploying real hardware in a devstack | 09:19 |
| sambetts | wagiel: http://paste.openstack.org/show/516174/ | 09:19 |
| milan | morning sambetts :) | 09:19 |
| wagiel | yes!!! thank you, appreciate it | 09:19 |
| *** abcxyz has quit IRC | 09:20 | |
| *** abcxyz has joined #openstack-ironic | 09:20 | |
| sambetts | wagiel: there are parts that are specific to my setup like the IMAGE_URLS etc but the big thing is getting the networking setup right so that its talking on a flat neutron network etc | 09:21 |
| wagiel | yes | 09:22 |
| wagiel | sure i will take that into account | 09:22 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Add test to verify ironic multitenancy https://review.openstack.org/269157 | 09:24 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: DO NOT REVIEW https://review.openstack.org/296432 | 09:24 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add --wait flag for provision actions and wait_for_provision_state function https://review.openstack.org/328191 | 09:25 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add --wait flag for provision actions and wait_for_provision_state function https://review.openstack.org/328191 | 09:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Add --wait flag for provision actions and wait_for_provision_state function https://review.openstack.org/328191 | 09:33 |
| *** ohamada_ has joined #openstack-ironic | 09:34 | |
| *** ohamada has quit IRC | 09:34 | |
| *** athomas has quit IRC | 09:45 | |
| *** bcornec1 has quit IRC | 09:47 | |
| *** bcornec1 has joined #openstack-ironic | 09:47 | |
| *** athomas has joined #openstack-ironic | 09:50 | |
| *** cinerama has quit IRC | 09:51 | |
| hkominos | never used devstack : ( | 09:53 |
| *** cinerama has joined #openstack-ironic | 09:55 | |
| *** bcornec1 has quit IRC | 09:57 | |
| *** moshele has joined #openstack-ironic | 10:03 | |
| *** bcornec1 has joined #openstack-ironic | 10:08 | |
| *** hoangcx has quit IRC | 10:24 | |
| hkominos | hello again openstackers. I am reading the manual and something is not clear. In the manual it is stated that the MAC address of the target host must be provided ( I have fuel with 2 NICS + IMPI ). Which mac does it require. The one in which the private Baremetal network is ? | 10:27 |
| sambetts | hkominos: You need the mac address of the NIC that the node will be PXE booted on | 10:28 |
| hkominos | !. ok that makes more sense. thx | 10:29 |
| openstack | hkominos: Error: "." is not a valid command. | 10:29 |
| hkominos | ! ok that makes more sense. thx | 10:29 |
| openstack | hkominos: Error: "ok" is not a valid command. | 10:29 |
| dtantsur | #pixiesay "ok" is a valid command | 10:30 |
| PixieBoots | ʕ•͡ᴥ•ʔ: "ok" is a valid command | 10:30 |
| hkominos | ok that makes more sense. I assume | 10:30 |
| sambetts | haha | 10:30 |
| hkominos | hahah | 10:30 |
| *** abcxyz has quit IRC | 10:30 | |
| hkominos | There is not a lof of documentation and tutorials going around for ironic | 10:30 |
| hkominos | this is becoming a frustration | 10:31 |
| *** hoangcx has joined #openstack-ironic | 10:31 | |
| *** bcornec1 has quit IRC | 10:36 | |
| openstackgerrit | xiexs proposed openstack/ironic: Trivial comments fix https://review.openstack.org/329870 | 10:39 |
| openstackgerrit | Aline Bousquet proposed openstack/python-ironicclient: Fix error returned by ironic --json node-validate https://review.openstack.org/303964 | 10:57 |
| *** bcornec1 has joined #openstack-ironic | 10:58 | |
| *** vmud213 has joined #openstack-ironic | 11:00 | |
| *** abcxyz has joined #openstack-ironic | 11:03 | |
| *** hoangcx has quit IRC | 11:04 | |
| *** bcornec1 has quit IRC | 11:04 | |
| *** baoli has joined #openstack-ironic | 11:05 | |
| *** baoli has quit IRC | 11:06 | |
| mat128 | good morning group! | 11:13 |
| sambetts | Morning mat128 :) | 11:14 |
| *** dprince has joined #openstack-ironic | 11:14 | |
| *** ccamacho is now known as ccamacho|lunch | 11:16 | |
| *** hparekh has quit IRC | 11:16 | |
| *** stendulker has quit IRC | 11:16 | |
| *** thrash|g0ne is now known as thrash | 11:19 | |
| *** hparekh has joined #openstack-ironic | 11:19 | |
| *** hparekh_ has joined #openstack-ironic | 11:19 | |
| *** hparekh_ has quit IRC | 11:20 | |
| *** bcornec1 has joined #openstack-ironic | 11:23 | |
| sambetts | dtantsur: Re: rloo's comments here, which do you think? https://review.openstack.org/#/c/320584/10/ironic_python_agent/hardware.py should I clear the cache if there's an exception or not? | 11:23 |
| sambetts | dtantsur: I can't decided if it should or no | 11:23 |
| dtantsur | hmmm | 11:23 |
| mat128 | sambetts: what kind of exceptions can get_lldp_info raise? | 11:25 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 11:25 |
| mat128 | oic | 11:26 |
| sambetts | mat128: any error related to opening and reading data from sockets | 11:26 |
| mat128 | yeah | 11:26 |
| mat128 | this is analogous to HTTP proxies IMO | 11:27 |
| mat128 | is the cache invalidated because the origin cannot be reached? | 11:27 |
| sambetts | right | 11:27 |
| sambetts | not sure :/ | 11:27 |
| mat128 | sambetts: it's not, cache is valid until TTL expires | 11:27 |
| mat128 | we dont have such ttl for lldp_data | 11:28 |
| sambetts | my orignal attitude, was only replace the existing data if we have some data to replace it with | 11:28 |
| sambetts | but then I thought that I have no indication in that function that the function has has failed to refresh the cache, so you might be operating on old data | 11:29 |
| mat128 | I'd be curious if there is caching/cache where it's already used | 11:30 |
| mat128 | https://github.com/openstack/ironic-python-agent/commit/15aaa03833208bb8ed8a5e164c164b9e073a4769 | 11:30 |
| mat128 | This brought the module and it's tests, with no usage | 11:30 |
| mat128 | :P | 11:30 |
| sambetts | yup :) I'm adding the usage ;) | 11:30 |
| sambetts | I wonder if I should have it return True if successful and False if the function fails, and make it log to say Failed to refresh the cache so expect None or Old lldp data | 11:31 |
| mat128 | sambetts: the thing I see with this approach is that it's fine from a code point of view | 11:34 |
| mat128 | sambetts: but lets say you are using this in the field, having it return false and log that it was expired "so you got nothing" doesnt help you profile a machine | 11:34 |
| mat128 | you need the data | 11:34 |
| mat128 | sambetts: could we distinguish between "timeout waiting for lldp data" vs "generic socket error" ? | 11:34 |
| mat128 | sambetts: given how this is lldp data and it's main purpose is to identify the machine's peers, I would say any data is good data, it's not like an operator will change cabling during inspection | 11:37 |
| mat128 | if you identified your ToR and the ports you are on, this is exactly the required info | 11:37 |
| *** mgoddard has joined #openstack-ironic | 11:38 | |
| sambetts | right, so a timeout waiting for lldp isn't considered a failure, in that case it returns an empty list for the port, with attidude of LLDP must not be running on this interface that why I've not received any packets on this interface | 11:38 |
| sambetts | so I think the failures we'd see in my new function would be something failed to open socket because reason | 11:39 |
| sambetts | or failed to read data from socket for reason | 11:39 |
| *** bcornec1 has quit IRC | 11:41 | |
| mat128 | sambetts: I know timeout is configureable, lets say this is below what you have configured in your switch, you might end up not getting any data | 11:42 |
| mat128 | sambetts: remember this isn't like lldpd where it runs in the background and always listens | 11:42 |
| sambetts | mat128: exactly, so in that case it logs that it failed to get any LLDP data for certain port, and then sets that ports list of TLVs to [] | 11:43 |
| mat128 | sambetts: what if we previously had lldp data | 11:43 |
| mat128 | sambetts: is it invalidated? (I think thats rloo's point) | 11:44 |
| mat128 | Given how this is LLDP data, I would say no | 11:44 |
| mat128 | Physical topology shouldnt change *during* inspection/profiling | 11:44 |
| jroll | if LLDP info is changing while the agent is running, you have bigger problems :P | 11:44 |
| jroll | mat128++ | 11:44 |
| mat128 | yup :) | 11:45 |
| sambetts | yeah, so my orignal thoughts is I don't think it should be, I say we only replace data if we have data to replace it with | 11:45 |
| mat128 | sambetts: I think that's right. I will provide the explanation in your review so everyone can see it | 11:45 |
| sambetts | thanks mat128 :D | 11:45 |
| sambetts | I need to fix one part of the code anyway, to make that work because right now if it gets any data it'll wipe the whole cache, I'll make it so that it replaces per interface instead of the whole cache | 11:46 |
| *** bcornec1 has joined #openstack-ironic | 11:48 | |
| mat128 | jroll: can you share why lldpd was not used instead? I know you can query it in a machine parsable format and it would take care of all of that on it's own. | 11:49 |
| jroll | mat128: history! | 11:49 |
| jroll | so | 11:49 |
| mat128 | one reason I see is adding a new dependency and bloating the filesystem, but thats it :P | 11:49 |
| vdrok | morning mat128 and jroll | 11:50 |
| mat128 | o/ vdrok | 11:50 |
| jroll | we run IPA in a systemd-nspawn container, as a single process, no init system there | 11:50 |
| mat128 | ahhh yes | 11:50 |
| jroll | so adding lldpd would need an init system | 11:50 |
| mat128 | yeah, got it | 11:50 |
| jroll | also it was quite the fun little thing to figure out :P | 11:50 |
| mat128 | supervisord ;) | 11:51 |
| mat128 | yeah | 11:51 |
| jroll | ikr | 11:51 |
| mat128 | I mean, I'm impressed its contained in 2-3 pages tops | 11:51 |
| mat128 | thought it was much more elaborate than this | 11:51 |
| jroll | I'm not opposed to someone making lldpd work in all the images, but we already have the code, so meh | 11:51 |
| jroll | heh, yeah I was too | 11:51 |
| sambetts | tinyipa has an init system, because its not all Yo Dawg I heard you like containers | 11:52 |
| sambetts | ;) | 11:52 |
| jroll | hey we use chroots now :) | 11:52 |
| mat128 | jroll++ | 11:52 |
| mat128 | exactly what I was about to say | 11:53 |
| jroll | but, the coreos+container thing honestly was focused on making the builder easy and customizable | 11:53 |
| mat128 | we could do it now, but as you said, we already have the code and it works | 11:53 |
| sambetts | the existing solution is pretty elegant really | 11:53 |
| sambetts | certainly going to be smaller than including another dep too | 11:53 |
| jroll | JayF is the linux systems guru and he was like, let's make this not terrible | 11:53 |
| openstackgerrit | Sam Betts proposed openstack/ironic-python-agent: Support LLDP data as part of interfaces in inventory https://review.openstack.org/320584 | 11:54 |
| mat128 | haha | 11:54 |
| *** abcxyz has quit IRC | 11:54 | |
| sambetts | haha not terrible is a good thing :P | 11:54 |
| jroll | also, fun fact for your future irc log grepping, lldp via tcpdump: tcpdump -i p9p2 -nn -vvvv ether proto 0x88cc | 11:54 |
| jroll | because the first time you deploy this and it doesn't work, you'll want that ;) | 11:54 |
| jroll | because you know someone will forget to enable lldp on that one switchport | 11:55 |
| *** moshele has quit IRC | 11:55 | |
| jroll | (make your life easy and do that in the ML2 driver) | 11:55 |
| *** moshele has joined #openstack-ironic | 11:55 | |
| * jroll stops resurrecting nightmares now | 11:55 | |
| sambetts | hahahaha :-P that solution I've put up is actually based on sucessfully using it with inspector to populate local link info | 11:56 |
| jroll | ++ | 11:56 |
| sambetts | the existing one was stagnating so I was like eh I'm just going to push my working solution | 11:56 |
| sambetts | it also didn't have the flexiblity we needed | 11:56 |
| mat128 | jroll: good old "switch not being configured as expected" | 11:57 |
| jroll | :D | 11:57 |
| mat128 | I would add "after we told you to change the template 10 times and change that setting" we still have to write code to do it for you | 11:57 |
| mat128 | lessons learned: manual steps are bound to fail | 11:57 |
| sambetts | heh, inspection in the new world is a giant chicken and egg problem | 11:57 |
| sambetts | because we need lldp data to program the switch, but we need the switch programmed to do inspection to get lldp data | 11:58 |
| mat128 | at some point you decided node X would be plugged in power outlet A, switch port B and C | 11:59 |
| jroll | sambetts: yeah, that initial kick step is hard | 11:59 |
| *** jcoufal has joined #openstack-ironic | 11:59 | |
| mat128 | and if you preconfigure all switch ports for lldp until they are used by a tenant | 11:59 |
| jroll | sambetts: if only ironic could deploy switches and their initial config... :) | 11:59 |
| sambetts | right, so we basically added a step to our guide at that point that says you have to manually put the switch port into the inspection network | 12:00 |
| mat128 | jroll: we seriously discussed deploying toasters | 12:00 |
| jroll | mat128: well, switches seem more productive :D | 12:00 |
| jroll | mat128: NobodyCam wrote a coffeepot driver once | 12:00 |
| mat128 | sambetts: why not have your network plugin make that switch? | 12:00 |
| mat128 | jroll: I wish it was still in-tree with provisioning and cleaning =) | 12:00 |
| jroll | heh | 12:01 |
| *** e0ne has quit IRC | 12:01 | |
| sambetts | mat128: the network plugin doesn't know that switch port is plugged into a BM until we pass the local link info when Ironic does a port create for provisioning | 12:02 |
| *** e0ne has joined #openstack-ironic | 12:02 | |
| jroll | sambetts: seriously though, ironic should be able to do the initial wipe-and-config for a switch. roll in a rack, power it up, give ironic switch info, it resets the switch and lays down initial config (inspector mode) and then inspector goes and finds all the things | 12:02 |
| mat128 | this++ | 12:03 |
| mat128 | Ironic as the bare-metal manager for your datacenter | 12:03 |
| mat128 | :D | 12:03 |
| jroll | and/or ironic resets the switch and tells neutron about it, neutron does the config, don't care | 12:03 |
| jroll | mat128++ | 12:03 |
| mat128 | manages all things baremetal | 12:03 |
| mat128 | jroll: its the same | 12:03 |
| mat128 | ironic "makes it happen" | 12:03 |
| sambetts | jroll: yeah that works if that switch is used for just BM, and at that point we don't need to use the LLDP inspection to know what plugged into what either | 12:03 |
| *** appu103_ has quit IRC | 12:04 | |
| jroll | sambetts: in an ideal world ironic owns the dc | 12:04 |
| * jroll side note we need a cmdb, urgh | 12:04 | |
| sambetts | jroll: works for an undercloud where everything is BM I guess :-p | 12:04 |
| *** nico-afk is now known as nicodemos | 12:05 | |
| *** mgoddard_ has joined #openstack-ironic | 12:05 | |
| nicodemos | good morning | 12:05 |
| dtantsur | spec related: https://review.openstack.org/320003 | 12:05 |
| dtantsur | morning nicodemos | 12:05 |
| jroll | sambetts: it's metal turtles all the way down | 12:05 |
| sambetts | mmm metal turtle soup | 12:05 |
| *** hoangcx has joined #openstack-ironic | 12:06 | |
| mat128 | sambetts: almost http://render.fineartamerica.com/images/rendered/medium/throw-pillow/images-medium/turtle-soup-dale-moses.jpg | 12:07 |
| openstackgerrit | Miles Gould proposed openstack/ironic: Document testing an in-review patch with devstack https://review.openstack.org/329898 | 12:07 |
| sambetts | haha | 12:07 |
| *** mgoddard has quit IRC | 12:08 | |
| *** ccamacho|lunch is now known as ccamacho | 12:10 | |
| *** jjohnson2_ has joined #openstack-ironic | 12:12 | |
| mgould | sambetts: I literally didn't know that was possible until we paired on the TinyIPA stuff, I think it's a good thing for new devs to know :-) | 12:12 |
| mat128 | mgould: what you were working on will be forever in the docs ;) | 12:13 |
| TheJulia | Good morning | 12:13 |
| mat128 | mgould: 295946/15 | 12:13 |
| mat128 | o/ TheJulia | 12:13 |
| mgould | mat128: not my patch :-) | 12:14 |
| sambetts | mgould: sorry, which bit? my brain is a jubbled mess at the moment | 12:14 |
| mgould | sambetts: using devstack to check out an in-progress patch | 12:14 |
| *** mbound has quit IRC | 12:15 | |
| sambetts | mgould: ooh right | 12:15 |
| *** mbound has joined #openstack-ironic | 12:15 | |
| *** daemontool has joined #openstack-ironic | 12:16 | |
| sambetts | mgould: yeah I used to do it manually for ages, e.g. run devstack, cd /opt/stack/X, git fetch and checkout, screen -x, ctrl-c, restart | 12:16 |
| mat128 | sambetts: any idea on testing the lldp stuff in devstack? | 12:16 |
| mgould | yeah, I did that a few times too | 12:16 |
| mat128 | sambetts, mgould: you guys know you can do git review -d 329898, right? | 12:16 |
| mat128 | saves copying the full line from gerrit | 12:16 |
| sambetts | mat128: yup, that works if your devstck machine is configured to use git review :-P | 12:17 |
| sambetts | mat128: I've never tried to use LLDP with virtual BM only on a real test bed | 12:17 |
| mgould | mat128: I do now :-) | 12:17 |
| * sambetts is sad that he can't test his patches properly right now because his testbed is being borrowed because downstream deadlines | 12:18 | |
| mat128 | sambetts: we have simulated LLDP in the past (different product =) by running lldpd on the "host" | 12:18 |
| mgould | mat128: actually, I dimly recall seeing that somewhere, but now can't find the page I saw it on :-( | 12:18 |
| *** baoli has joined #openstack-ironic | 12:19 | |
| mat128 | mgould: lldp or git review? the former was an internal project here so I would be surprised :) | 12:19 |
| sambetts | mat128: our VMs in devstck for virtual BM are plugged into OVS so I wonder if that supports it | 12:19 |
| mgould | mat128: git review | 12:20 |
| *** baoli has quit IRC | 12:20 | |
| mat128 | sambetts: yeah right.. I wouldnt know | 12:20 |
| *** baoli has joined #openstack-ironic | 12:20 | |
| mat128 | mgould: ah ok :) | 12:20 |
| sambetts | something we should look into I guess | 12:20 |
| mat128 | mgould: theres always this: http://docs.openstack.org/infra/git-review/usage.html | 12:20 |
| mgould | if I had seen your internal project docs, I wouldn't admit to it in a public logged channel :-) | 12:21 |
| sambetts | hahaha | 12:21 |
| mat128 | mgould: lmao | 12:21 |
| mat128 | tbh, the test suite was more impressive than the product itself | 12:21 |
| mat128 | we would release new versions with barely trying it on metal hardware | 12:22 |
| mgould | mat128: ah, so it's "./stack.sh; cd /opt/stack/$project; git review -d $change; [manual test steps]"? | 12:22 |
| *** baoli_ has joined #openstack-ironic | 12:22 | |
| mat128 | mgould: yes, as long as you have git-review installed on your devstack box | 12:22 |
| mat128 | as Sam pointed out | 12:22 |
| mat128 | sambetts: there's hope: http://openvswitch.org/pipermail/discuss/2015-May/017659.html | 12:22 |
| mgould | awesome, I'll do that after lunch | 12:22 |
| mgould | mat128: effective CI is a thing of beauty and a joy forever | 12:23 |
| mgould | flaky CI is awful :-( | 12:23 |
| *** trown|outtypewww is now known as trown | 12:23 | |
| mat128 | yup | 12:23 |
| mgould | brb | 12:24 |
| sambetts | flaky CI just leads to distrust in the results, and in OpenStack results in people rechecking over and over again without even checking the results, because they assume it was a transiant failure | 12:25 |
| *** baoli has quit IRC | 12:25 | |
| *** Goneri has joined #openstack-ironic | 12:26 | |
| *** rbudden has joined #openstack-ironic | 12:27 | |
| mat128 | sambetts: the tooling is pretty good here, with elastic-recheck and everything | 12:28 |
| jroll | hey, can some folks review the infiniband spec soon? I'd love sambetts review especially https://review.openstack.org/#/c/265635/ | 12:28 |
| sambetts | jroll: sure I'll take a look | 12:29 |
| jroll | thanks | 12:29 |
| sambetts | mat128: it gets better every day :) | 12:29 |
| jroll | TheJulia: NobodyCam: if you could re-review as well please :) ^ | 12:29 |
| *** mgoddard_ has quit IRC | 12:30 | |
| *** mgoddard has joined #openstack-ironic | 12:30 | |
| *** ionutbalutoiu has joined #openstack-ironic | 12:31 | |
| *** frontrunner has joined #openstack-ironic | 12:31 | |
| mgould | mat128: Elastic Recheck feels like a solution to a problem we shouldn't have :-( | 12:34 |
| mat128 | mgould: unless we control all variables we will always have such issues | 12:34 |
| mat128 | mgould: we rely on external pip packages, apt-get mirrors running correctly, and even more | 12:34 |
| mgould | yeah, I suppose so | 12:35 |
| mgould | "a heavy solution to a problem that shouldn't be so severe", perhaps | 12:35 |
| mat128 | agreed | 12:35 |
| mat128 | if only we could gate public package uploads to pypi =) | 12:35 |
| vsaienko | Morning Ironic'ers! | 12:37 |
| vdrok | \o/ | 12:38 |
| TheJulia | jroll: will do | 12:38 |
| *** wagiel has quit IRC | 12:38 | |
| *** mbound has quit IRC | 12:40 | |
| jroll | TheJulia: thanks :) | 12:41 |
| mat128 | o/ vsaienko | 12:43 |
| vsaienko | do we have a bug about apache restart failures at the ironic gates? | 12:45 |
| jroll | vsaienko: yes, let me find it | 12:46 |
| jroll | vsaienko: https://bugs.launchpad.net/ironic/+bug/1590139 | 12:46 |
| openstack | Launchpad bug 1590139 in Ironic "gate job fails when unable to start apache service during horizon setup" [Undecided,New] | 12:46 |
| sambetts | jroll: I was about to leave a comment about adding vendor_id as a new column saying that port.extra should be used instead because its an implementation specific field, but looking back at past comments I see that it was switched to its own field based on comments from you to avoid json fields, I'm not sure if I agree or not | 12:46 |
| jroll | vsaienko: and the elastic recheck patch https://review.openstack.org/#/c/326739/ | 12:47 |
| jroll | sambetts: I didn't ask for it to be its own field, to be clear :) I asked why it wasn't part of the address, and that somehow devolved into it being its own field | 12:48 |
| jroll | sambetts: port.extra makes sense, I didn't get a chance to respond and say so | 12:48 |
| sambetts | jroll: ah right :-P | 12:49 |
| *** bcornec1 has quit IRC | 12:50 | |
| sambetts | jroll: I think that we're going to have to have a list of supported vendor_ids, because the GUID to mac conversion is vendor specific | 12:51 |
| jroll | sambetts: indeed | 12:51 |
| *** bcornec1 has joined #openstack-ironic | 12:51 | |
| sambetts | jroll: vendor_id plugin system? | 12:51 |
| jroll | oh god no pls | 12:52 |
| sambetts | ;) | 12:52 |
| dtantsur | morning jroll, mat128, vsaienko, TheJulia and whoever I didn't say good morning yet :) | 12:54 |
| jroll | morning dtantsur :) | 12:55 |
| TheJulia | Regarding bug 1590139, I think this is our culprit "+ lib/apache:restart_apache_server:190 : sleep 3" <-- not enough time for the kernel to close out sockets in wait state which can prevent the port being rebound. | 12:55 |
| openstack | bug 1590139 in Ironic "gate job fails when unable to start apache service during horizon setup" [Undecided,New] https://launchpad.net/bugs/1590139 | 12:55 |
| dtantsur | jroll, please review https://review.openstack.org/#/c/306418/ I'm going to hack on it really soon, it blocks the driver composition | 12:56 |
| dtantsur | (the spec to promote agent API) | 12:56 |
| *** lindycoder has joined #openstack-ironic | 12:58 | |
| *** itamarl has quit IRC | 12:58 | |
| jroll | dtantsur: sure, driver comp is also on my hotlist | 12:59 |
| vsaienko | jroll please have look https://review.openstack.org/329908 | 12:59 |
| *** lindycod_ has joined #openstack-ironic | 12:59 | |
| jroll | vsaienko: cool | 13:00 |
| *** bcornec1 has quit IRC | 13:00 | |
| vsaienko | I remember we had the same problem in mos when switched to keystone on wsgi | 13:01 |
| vsaienko | https://github.com/openstack/fuel-library/commit/ff2274e44ab732d943976d1b21c8997dc90a7d94 | 13:01 |
| NobodyCam | Good Morning Ironicers | 13:01 |
| NobodyCam | jroll: will have a look after I get some coffee going | 13:01 |
| dtantsur | morning NobodyCam | 13:01 |
| NobodyCam | morning dtantsur :) | 13:01 |
| *** dprince has quit IRC | 13:02 | |
| vdrok | morning TheJulia and NobodyCam | 13:03 |
| *** lindycoder has quit IRC | 13:03 | |
| NobodyCam | good morning vdrok :) | 13:03 |
| *** bcornec1 has joined #openstack-ironic | 13:03 | |
| *** jcoufal has quit IRC | 13:04 | |
| jroll | NobodyCam: morning, thanks :) | 13:05 |
| NobodyCam | :) | 13:07 |
| *** jvgrant_ has joined #openstack-ironic | 13:09 | |
| *** rloo has joined #openstack-ironic | 13:09 | |
| *** bcornec1 has quit IRC | 13:10 | |
| *** rcernin has quit IRC | 13:14 | |
| NobodyCam | Thank you rloo :) | 13:14 |
| rloo | NobodyCam: yw :) | 13:15 |
| NobodyCam | and Good morning :) lol :) | 13:15 |
| sambetts | jroll: left some comments, hopefully I'm not being mad :-P | 13:15 |
| NobodyCam | morning sambetts | 13:15 |
| sambetts | Hey NobodyCam \o/ | 13:15 |
| rloo | NobodyCam: Gooooood morning to you too! And to sambetts, jroll, TheJulia and everyone else! | 13:15 |
| *** bcornec1 has joined #openstack-ironic | 13:16 | |
| sambetts | Hi rloo! | 13:16 |
| NobodyCam | \o/ | 13:16 |
| dtantsur | rloo, morning | 13:16 |
| vsaienko | morning rloo | 13:19 |
| NobodyCam | morning vsaienko | 13:19 |
| rloo | hi dtantsur, vsaienko | 13:19 |
| vsaienko | o/ NobodyCam | 13:19 |
| NobodyCam | :) | 13:19 |
| *** bcornec1 has quit IRC | 13:22 | |
| *** xavierr-afk is now known as xavierr | 13:23 | |
| xavierr | morning Ironic :) | 13:23 |
| xavierr | morning NobodyCam rloo sambetts dtantsur vsaienko | 13:24 |
| NobodyCam | morning xavierr | 13:24 |
| rloo | morning xavierr | 13:24 |
| *** bcornec1 has joined #openstack-ironic | 13:25 | |
| mgould | morning rloo NobodyCam TheJulia jroll | 13:26 |
| NobodyCam | morning mgould | 13:27 |
| rloo | hi mgould | 13:28 |
| *** rcernin has joined #openstack-ironic | 13:28 | |
| *** sdake has joined #openstack-ironic | 13:29 | |
| *** bcornec1 has quit IRC | 13:30 | |
| *** moshele has quit IRC | 13:32 | |
| *** moshele has joined #openstack-ironic | 13:33 | |
| vsaienko | Ironic'ers just a notice, who don't know multitenancy patches passed grenade and multitenancy gate tests https://review.openstack.org/#/c/296432/. I'm kindly asking to review them as it is highest priority for community :). Thanks in advance! | 13:34 |
| jroll | ++ | 13:34 |
| *** jjohnson2_ has quit IRC | 13:34 | |
| *** joprovost has joined #openstack-ironic | 13:34 | |
| NobodyCam | w00t | 13:35 |
| sambetts | \o/ | 13:35 |
| *** wajdi has quit IRC | 13:35 | |
| NobodyCam | vsaienko: the title is "DO NOT REVIEW" | 13:38 |
| NobodyCam | lol | 13:38 |
| jroll | NobodyCam: that's the patch to collect all the things and test... | 13:38 |
| NobodyCam | :p | 13:39 |
| *** jcoufal has joined #openstack-ironic | 13:39 | |
| *** mbound has joined #openstack-ironic | 13:41 | |
| *** jcoufal_ has joined #openstack-ironic | 13:41 | |
| *** sdake_ has joined #openstack-ironic | 13:42 | |
| *** jcoufal has quit IRC | 13:45 | |
| *** sdake has quit IRC | 13:45 | |
| hkominos | Is there any documentation in which i can see the Nic-Network configuration for an ironic slave? | 13:45 |
| *** mbound has quit IRC | 13:46 | |
| *** PollyZ has joined #openstack-ironic | 13:47 | |
| *** sacharya has joined #openstack-ironic | 13:49 | |
| *** bcornec1 has joined #openstack-ironic | 13:51 | |
| vdrok | morning rloo and xavierr | 13:51 |
| rloo | hi vdrok | 13:51 |
| *** sacharya_ has joined #openstack-ironic | 13:53 | |
| *** rcernin has quit IRC | 13:54 | |
| *** bcornec1 has quit IRC | 13:55 | |
| *** sacharya has quit IRC | 13:56 | |
| *** links has quit IRC | 13:56 | |
| *** bcornec1 has joined #openstack-ironic | 13:58 | |
| *** sacharya_ has quit IRC | 13:59 | |
| *** sacharya has joined #openstack-ironic | 13:59 | |
| *** dprince has joined #openstack-ironic | 13:59 | |
| *** joprovost1 has joined #openstack-ironic | 14:00 | |
| *** jvgrant_ has quit IRC | 14:00 | |
| *** joprovost has quit IRC | 14:01 | |
| *** joprovost1 is now known as joprovost | 14:01 | |
| *** bcornec2 has joined #openstack-ironic | 14:01 | |
| *** jaybeale has joined #openstack-ironic | 14:02 | |
| *** cdearborn has joined #openstack-ironic | 14:02 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-specs: Add spec for interface attach and detach API https://review.openstack.org/317636 | 14:06 |
| *** mtanino has joined #openstack-ironic | 14:06 | |
| *** jvgrant_ has joined #openstack-ironic | 14:07 | |
| *** rcernin has joined #openstack-ironic | 14:09 | |
| jlvillal | vsaienko, That is great on the multi-tenancy work :) Thanks for that! | 14:11 |
| *** bcornec2 has quit IRC | 14:13 | |
| *** hoangcx has quit IRC | 14:13 | |
| *** jvgrant_ has quit IRC | 14:14 | |
| *** rcernin has quit IRC | 14:18 | |
| *** trown is now known as trown|mtg | 14:18 | |
| PollyZ | Good morning for this one https://bugs.launchpad.net/wsme/+bug/1256042, does anyone know if it has been fixed or no | 14:21 |
| openstack | Launchpad bug 1256042 in WSME "Pecan's on_error() hook never gets trigged when used with WSME" [High,Confirmed] - Assigned to Jianing Yang (jianingy) | 14:21 |
| jroll | PollyZ: doesn't appear so, from the status | 14:22 |
| *** moshele has quit IRC | 14:22 | |
| *** moshele has joined #openstack-ironic | 14:23 | |
| *** jrist has quit IRC | 14:23 | |
| PollyZ | jroll: And I couldn't reach to the person who is responible for this | 14:24 |
| *** moshele has quit IRC | 14:24 | |
| rloo | vsaienko, jroll: wrt the multitenancy work -- maybe send an email out to devlist? | 14:24 |
| jroll | PollyZ: they assigned themselves two years ago, I would go ahead and work on it if you want to | 14:25 |
| jroll | rloo: ++ | 14:25 |
| *** dprince has quit IRC | 14:25 | |
| *** joprovost1 has joined #openstack-ironic | 14:28 | |
| *** bcornec2 has joined #openstack-ironic | 14:28 | |
| *** vmud213 has quit IRC | 14:28 | |
| openstackgerrit | Miles Gould proposed openstack/ironic: Document testing an in-review patch with devstack https://review.openstack.org/329898 | 14:29 |
| *** joprovost has quit IRC | 14:29 | |
| xavierr | hey vdrok o/ | 14:29 |
| *** rcernin has joined #openstack-ironic | 14:31 | |
| *** joprovost1 has quit IRC | 14:32 | |
| *** catintheroof has joined #openstack-ironic | 14:32 | |
| vsaienko | rloo +1 | 14:32 |
| PollyZ | jroll: Thank you :) | 14:33 |
| *** jrist has joined #openstack-ironic | 14:37 | |
| *** joprovost has joined #openstack-ironic | 14:39 | |
| *** jaybeale has quit IRC | 14:40 | |
| *** bcornec2 has quit IRC | 14:40 | |
| *** wajdi has joined #openstack-ironic | 14:40 | |
| openstackgerrit | Miles Gould proposed openstack/ironic: Document testing an in-review patch with devstack https://review.openstack.org/329898 | 14:42 |
| *** wajdi has quit IRC | 14:43 | |
| *** wajdi has joined #openstack-ironic | 14:44 | |
| *** bcornec2 has joined #openstack-ironic | 14:44 | |
| *** jvgrant_ has joined #openstack-ironic | 14:45 | |
| *** tloeb has joined #openstack-ironic | 14:47 | |
| *** hoangcx has joined #openstack-ironic | 14:48 | |
| dtantsur | jroll, I started hacking on the agent API thing. do you think we should move the find-node-by-macs thing to objects.Node? | 14:55 |
| dtantsur | my concern is that it also accesses objects.Port | 14:55 |
| jroll | dtantsur: I'm good with that, select from nodes and join on the port | 14:56 |
| jroll | that's been a todo forever | 14:56 |
| dtantsur | cool | 14:57 |
| *** rbartal has quit IRC | 15:04 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Replace vif_portgroup_id with vif_port_id https://review.openstack.org/325197 | 15:06 |
| openstackgerrit | Corey Bryant proposed openstack/ironic: Use assert_called_once_with in test_cleanup_cleanwait_timeout https://review.openstack.org/330001 | 15:08 |
| *** catintheroof has quit IRC | 15:08 | |
| *** sdake_ has quit IRC | 15:08 | |
| JayF | Do we have a doc on how to build a dib image for fedora? | 15:08 |
| JayF | hmm. There's a patch up to add devstack support for doing it, maybe I can learn from there... | 15:09 |
| *** tobias_fiberdata has quit IRC | 15:09 | |
| sambetts | JayF: you should just be able to replace ubuntu with fedora | 15:09 |
| JayF | See you're starting from "knows how to build a dib image at all" | 15:09 |
| JayF | lol | 15:09 |
| sambetts | :-P | 15:10 |
| *** sdake has joined #openstack-ironic | 15:10 | |
| sambetts | ah, is the image for tenant image or IPA image? | 15:10 |
| JayF | ipa image | 15:10 |
| JayF | just wanted to play around some with the other ramdisks since I only use coreos much | 15:10 |
| sambetts | disk-image-create ironic-agent fedora -o ironic-agent | 15:11 |
| sambetts | JayF: ^ | 15:11 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic: Deny some driver_info updates when console is on https://review.openstack.org/330003 | 15:11 |
| *** sdake has quit IRC | 15:11 | |
| JayF | that seems super easy | 15:11 |
| sambetts | JayF: we document it in the inspector docs :) | 15:12 |
| * JayF suspicious of anything that seems easy | 15:12 | |
| sambetts | heh | 15:12 |
| TheJulia | JayF: it truly is rather easy :) | 15:12 |
| *** garthb has joined #openstack-ironic | 15:12 | |
| JayF | and, apparently, supported in devstack now | 15:13 |
| sambetts | rloo: thanks for being on top of reviewing my lldp patch :D | 15:13 |
| JayF | I don't get why we don't have these publishing too, and testing against IPA? | 15:13 |
| JayF | Even if we don't want to publish the image, it looks like we're just project-config away from actually testing dib images in ipa gate | 15:13 |
| JayF | I guess we wanted to figure out the dep stuff before doing that though, eh | 15:13 |
| rloo | sambetts: yw. i think i reviewed too many things yesterday; i seem to have a bunch that i need to re-review etc today :-( | 15:13 |
| *** bcornec2 has quit IRC | 15:14 | |
| sambetts | :( | 15:15 |
| rloo | JayF: do you mean http://docs.openstack.org/developer/ironic/deploy/install-guide.html#buildingdibbaseddeployramdisk | 15:15 |
| *** SiRiuS__ has joined #openstack-ironic | 15:15 | |
| *** bcornec2 has joined #openstack-ironic | 15:20 | |
| *** rloo has quit IRC | 15:23 | |
| *** rloo has joined #openstack-ironic | 15:23 | |
| *** ifarkas has quit IRC | 15:23 | |
| *** bcornec2 has quit IRC | 15:27 | |
| *** bcornec2 has joined #openstack-ironic | 15:28 | |
| *** ChrisAusten has joined #openstack-ironic | 15:30 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] [REALLY WIP] Promote agent vendor passthru to core API https://review.openstack.org/330018 | 15:31 |
| *** jaybeale has joined #openstack-ironic | 15:32 | |
| *** rama_y has joined #openstack-ironic | 15:32 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Replace vif_portgroup_id with vif_port_id https://review.openstack.org/325197 | 15:34 |
| vdrok | dtantsur: re https://review.openstack.org/325197 I think there is no need for a release note as there is no API for portgroups | 15:35 |
| *** moshele has joined #openstack-ironic | 15:35 | |
| dtantsur | vdrok, I'd still create a release note for people who already expect us to use the old name once we get to the API | 15:36 |
| dtantsur | or even: if we had API already, that would probably be a breaking change. now I only ask for a release note :) | 15:36 |
| vdrok | hm, I think I did see a comments on another zhenguo's patch that was doing a similar thing from rloo or yuriyz | 15:36 |
| rloo | vdrok: you called? :) | 15:37 |
| vdrok | yep, https://review.openstack.org/325197 :) | 15:37 |
| vdrok | rloo: release note when there is no API yet | 15:37 |
| rloo | vdrok: how would someone be using this code? | 15:37 |
| vdrok | see dtantsur's messages :) ^^ | 15:38 |
| rloo | dtantsur: who would be expecting us to use this in the API? people that are pulling down unmerged patches? | 15:39 |
| dtantsur | rloo, writing their code for future API for example? | 15:39 |
| dtantsur | rloo, yes. there are a lot of them already | 15:39 |
| vdrok | dtantsur: rloo I thought release notes are mostly for operators? | 15:39 |
| rloo | vdrok: that's what i thought too | 15:40 |
| *** sabeen has joined #openstack-ironic | 15:40 | |
| rloo | vdrok: but i think dtantsur thinks this could break someone's deployment. | 15:41 |
| dtantsur | 1. I don't agree with that; 2. the difference is subtle | 15:41 |
| sambetts | if we add a release note it would look really weird in the compiled release notes right? i.e. we'll have a release note saying its adding it followed by a release note saying its changing it | 15:41 |
| dtantsur | if release notes are only for operators, does it mean we should not make release notes for changes in the driver interface? I don't think so. | 15:41 |
| dtantsur | sambetts, that's sad, but we're doing it all the time | 15:42 |
| jroll | we could fix the existing release note... | 15:42 |
| sambetts | also I don't think things in the port.extra field are part of our offical API | 15:42 |
| vdrok | sambetts: there is not reno saying it was added I think | 15:42 |
| rloo | vdrok, dtantsur, sambetts: so I just looked at our existing release notes, i don't see any reference to vif_portgroup_id: http://docs.openstack.org/releasenotes/ironic/mitaka.html | 15:42 |
| vdrok | ++ | 15:42 |
| JayF | sambetts: did you see the comment yesterday; someone said IPA partitition images work in coreos but not tinyipa | 15:42 |
| JayF | sambetts: you know anything about that? | 15:42 |
| vdrok | JayF: it was me, and it seems it's not the case anymore | 15:43 |
| JayF | ah wonderful! | 15:43 |
| rloo | so i think it would be confusing to mention this change in the release notes. also, it is in .extra and it wasn't documented anywhere. | 15:43 |
| vdrok | JayF: sambetts see https://review.openstack.org/329625 partition image job, it failed, but it's not something with binary dependencies | 15:43 |
| sambetts | things in .extra are part of the offical API its driver specific (now) | 15:43 |
| *** tobias_fiberdata has joined #openstack-ironic | 15:44 | |
| sambetts | aren't* | 15:44 |
| jroll | I don't think a release note is needed here, honestly | 15:44 |
| rloo | vdrok, dtantsur, sambetts: I don't think we should put a reno for that | 15:44 |
| sambetts | I agree | 15:44 |
| vdrok | we just need to convince dtantsur :) | 15:45 |
| rloo | it is only going to confuse people i think. | 15:45 |
| * rloo gives dtantsur a beer | 15:45 | |
| sambetts | we'd be adding a release note fixing something for an incomplete feature | 15:45 |
| sambetts | tbh I think its weird that it was created as its own bug | 15:45 |
| rloo | sambetts: i think the problem is in us not knowing how to really handle LONG features | 15:46 |
| sambetts | ++ | 15:46 |
| dtantsur | sambetts, " things in .extra aren't part of the offical API" <-- wrong | 15:46 |
| sambetts | *cough* feature branch *cough* | 15:46 |
| dtantsur | it should not, but it is | 15:46 |
| dtantsur | * they | 15:47 |
| sambetts | dtantsur: its not though, its just a weird contract we have between ironic and the nova driver | 15:47 |
| vdrok | dtantsur: ok, but still there is no way to set it now, just by using code on review | 15:47 |
| mat128 | reno example of a feature that got introduced and modified: http://docs.openstack.org/releasenotes/ironic-python-agent/current-series.html | 15:47 |
| mat128 | has_carrier | 15:47 |
| rloo | sambetts: yup, in hindsight, that should have been on a feature branch. but in liberty cycle, who would have thought it wouldn't get in til newton or beyond? | 15:47 |
| mat128 | The “has_carrier” flag was added [...] and Stop checking the “has_carrier” field when [...] | 15:47 |
| sambetts | rloo: v.true | 15:47 |
| dtantsur | sambetts, weird contract between ironic and nova is definitely something we have to consider API :) | 15:47 |
| jroll | dtantsur: things in .extra are not *officially supported* parts of our API | 15:48 |
| jroll | they are not subject to backward compat, etc | 15:48 |
| sambetts | dtantsur: trying to fix that https://review.openstack.org/#/c/327046/ | 15:48 |
| dtantsur | jroll, we don't officially support nova or what? | 15:48 |
| openstackgerrit | Merged openstack/ironic: Trivial comments fix https://review.openstack.org/329870 | 15:48 |
| dtantsur | it's part of our contract with nova and essentially everyone following the same | 15:48 |
| dtantsur | sambetts++ | 15:48 |
| jroll | dtantsur: that's a different thing IMO, but nova doesn't use vif_portgroup_id today | 15:48 |
| dtantsur | jroll, yeah, I agree about portgroup, but lets face the truth that the port vif is a part of our API right now.. | 15:49 |
| vdrok | and noone using code from master can use it | 15:49 |
| jroll | dtantsur: it's part of our contract between nova and ironic, but is not part of our officially supported API | 15:49 |
| dtantsur | vdrok, I've removed -1 fwiw | 15:49 |
| jroll | dtantsur: that said, we can't just break that one because we'll break the nova contract | 15:49 |
| dtantsur | right. that's what "supported" means to me | 15:50 |
| vdrok | dtantsur: thanks :) | 15:50 |
| dtantsur | and I clearly expect people using ironic+neutron without nova to do the same | 15:50 |
| *** pcaruana has quit IRC | 15:50 | |
| jroll | dtantsur: let me rephrase, port.extra['vif_port_id'] is supported (which is a bug to me), the rest of .extra is not | 15:50 |
| jroll | sambetts is fixing that bug so three cheers for him | 15:50 |
| dtantsur | good :) btw found that I used it as well: https://github.com/dtantsur/metalsmith/blob/master/metalsmith/deploy.py#L82 | 15:50 |
| jroll | ya | 15:51 |
| sambetts | and vif_port_id is soon going to network driver specifc, just like ipmi_password etc | 15:51 |
| sambetts | are for power | 15:51 |
| jroll | say wha? | 15:51 |
| sambetts | so port.extra should be seen the same as driver_info | 15:51 |
| sambetts | but for network drivers | 15:52 |
| dtantsur | >_< | 15:52 |
| *** jrist has quit IRC | 15:52 | |
| *** trown|mtg is now known as trown | 15:52 | |
| *** dmk0202 has quit IRC | 15:52 | |
| sambetts | jroll: neutron network driver requires it, None driver does not, one of my experimental network drivers also does not | 15:53 |
| *** rcernin has quit IRC | 15:53 | |
| jroll | sambetts: :/ | 15:53 |
| *** Sukhdev has joined #openstack-ironic | 15:53 | |
| sambetts | hence why we need an attach and detach interface API that should call into network driver | 15:53 |
| sambetts | because different driver handle that part differently | 15:54 |
| *** jaybeale has quit IRC | 15:54 | |
| jroll | yep | 15:55 |
| *** SiRiuS__ has quit IRC | 15:55 | |
| sambetts | JayF, vdrok: So agent driver + partition images + tinyIPA is working now? | 15:55 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] [REALLY WIP] Promote agent vendor passthru to core API https://review.openstack.org/330018 | 15:55 |
| dtantsur | things become more and more complex pretty quickly | 15:55 |
| sambetts | dtantsur: ooooh yes :-P | 15:55 |
| *** tesseract has quit IRC | 15:55 | |
| * mgould lols at [REALLY WIP] | 15:56 | |
| vdrok | sambetts: I've referenced https://review.openstack.org/329625 it adds all the bits necessary to run a job properly, see the result of gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv | 15:56 |
| vdrok | it is on tinyipa | 15:56 |
| dtantsur | mgould, :) | 15:56 |
| mgould | sambetts, mat128: my devstack VM now has git-review set up - thanks! | 15:56 |
| sambetts | mgould: awesome | 15:56 |
| * mgould hastily takes a snapshot | 15:56 | |
| *** SiRiuS__ has joined #openstack-ironic | 15:57 | |
| *** ohamada_ has quit IRC | 15:59 | |
| *** garthb has quit IRC | 16:00 | |
| sambetts | vdrok: so http://logs.openstack.org/25/329625/2/check/gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv/e7115a3/logs/ironic-bm-logs/ the image is totally deploying and booting, I'm really confused whats causing that errro | 16:00 |
| vdrok | sambetts: http://logs.openstack.org/25/329625/2/check/gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv/e7115a3/logs/ironic-bm-logs/node-0_console_14-06-2016-20:28:43.txt.gz#_2016-06-14_20_28_05_387 | 16:01 |
| vdrok | ephemeral_mb None | 16:01 |
| *** ijw has joined #openstack-ironic | 16:01 | |
| *** tloeb has quit IRC | 16:01 | |
| vdrok | it depends on a devstack-gate patch setting ephemeral size tho | 16:01 |
| *** jcoufal has joined #openstack-ironic | 16:01 | |
| sambetts | vdrok: sounds like an error on the Ironic side then right? | 16:02 |
| vdrok | yeah, either or devstack gate, or devstack plugin :) | 16:02 |
| vdrok | I'll take a closer look now | 16:02 |
| *** jcoufal_ has quit IRC | 16:03 | |
| sambetts | vdrok: I'm not sure you can depends-on a devstack-gate patch, have you had that work before | 16:03 |
| sambetts | ? | 16:03 |
| sambetts | vdrok: oh maybe you can :/ | 16:04 |
| vdrok | sambetts: idk really, I now that project-config does not work this way, but not sure about devstack-gate, it should work the same way as devstack | 16:04 |
| openstackgerrit | Miles Gould proposed openstack/ironic: Document testing an in-review patch with devstack https://review.openstack.org/329898 | 16:05 |
| *** sdake has joined #openstack-ironic | 16:05 | |
| sambetts | http://logs.openstack.org/25/329625/2/check/gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv/e7115a3/console.html#_2016-06-14_19_57_54_266 that doesn't == https://review.openstack.org/#/c/326061/2 | 16:05 |
| vdrok | sambetts: it works, passed from devstack-gate http://logs.openstack.org/25/329625/2/check/gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv/e7115a3/logs/devstacklog.txt.gz#_2016-06-14_20_22_39_942 | 16:05 |
| *** athomas has quit IRC | 16:05 | |
| vdrok | otherwise it would be 0- | 16:05 |
| vdrok | hmm | 16:06 |
| *** appu103_ has joined #openstack-ironic | 16:06 | |
| sambetts | where does DEVSTACK_GATE_IRONIC_EPHEMERAL come from, do we already have that in our Jenkins job? | 16:06 |
| *** bcornec2 has quit IRC | 16:07 | |
| sambetts | oh yeah I see it in the exported vars | 16:07 |
| vdrok | sambetts: correct, it was merged already | 16:07 |
| vdrok | dunno why commit hash is different | 16:07 |
| sambetts | no :/ | 16:07 |
| sambetts | looking at the tempest test logs I see its been set on the node correctly | 16:08 |
| *** krotscheck is now known as krotscheck_dcm | 16:09 | |
| *** thiagop has joined #openstack-ironic | 16:10 | |
| *** [1]cdearborn has joined #openstack-ironic | 16:10 | |
| sambetts | vdrok: Looking at the agent driver, I can see its looking for ephemeral_mb not ephemeral_gb | 16:11 |
| *** tobias_fiberdata has quit IRC | 16:11 | |
| vdrok | sambetts: yeah I noticed it too | 16:11 |
| sambetts | vdrok: which is why its appearing to be None | 16:12 |
| sambetts | vdrok: because ephemeral_gb is set on the node, and then it does a instance_info.get(ephemeral_mb) | 16:12 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: [WIP]Introducing Inspector Grenade testing https://review.openstack.org/327667 | 16:13 |
| sambetts | vdrok: I'm thinking it might have always been broken | 16:14 |
| *** bcornec2 has joined #openstack-ironic | 16:14 | |
| sambetts | but we've never tested it properly | 16:14 |
| thiagop | Good morning, Ironic | 16:15 |
| *** tobias_fiberdata has joined #openstack-ironic | 16:15 | |
| vdrok | sambetts: yeah. nova is setting it to instance_info, so the s/mb/gb should be the only change needed | 16:15 |
| vdrok | morning thiagop | 16:16 |
| *** milan has quit IRC | 16:16 | |
| vdrok | sambetts: heh https://github.com/openstack/ironic-python-agent/blob/3cf5369cb6285fe32c310fba7c712c690feb4ea4/ironic_python_agent/extensions/standby.py#L92 | 16:17 |
| vdrok | ipa expects mb too | 16:17 |
| *** piet has joined #openstack-ironic | 16:17 | |
| sambetts | -,- oh dear... | 16:17 |
| vdrok | and ironic-lib too | 16:18 |
| sambetts | vdrok: take a look at ironic/drivers/modules/iscsi_deploy.py L212 | 16:18 |
| sambetts | vdrok: thats what we should be doing in that driver | 16:18 |
| dtantsur | ooops | 16:18 |
| vdrok | +++ | 16:19 |
| sambetts | we are missing a ephemeral_mb = 1024 * int(i_info['ephemeral_gb'] | 16:19 |
| sambetts | ) | 16:19 |
| vdrok | sambetts: same for root_mb | 16:20 |
| NobodyCam | morning thiagop | 16:20 |
| sambetts | vdrok: yeah, and swap | 16:20 |
| vdrok | sambetts: ah, missed this one - https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/deploy_utils.py#L1139 | 16:21 |
| vdrok | so everything is ok except ephemeral | 16:21 |
| *** derekh has quit IRC | 16:22 | |
| vdrok | swap_mb is mb already from the beginning | 16:22 |
| sambetts | vdrok: we should add a line into that function adding ephemeral_mb to i_info | 16:23 |
| vdrok | sambetts: yep, will file a bug and a patch now | 16:23 |
| sambetts | Sweet \o/ yay for CI catching stuff! | 16:23 |
| sambetts | JayF: ^ if you didn't follow the conversation | 16:23 |
| sambetts | JayF: basically agent driver is passing ephemeral_gb instead of ephemeral_mb as IPA expects | 16:24 |
| *** bcornec2 has quit IRC | 16:25 | |
| *** cdearborn has quit IRC | 16:25 | |
| *** kromanenko__ has quit IRC | 16:26 | |
| *** kromanenko_ has quit IRC | 16:26 | |
| *** Romanenko_K has quit IRC | 16:26 | |
| openstackgerrit | Nisha Agarwal proposed openstack/proliantutils: Fix the logic for fetching the gpu device count https://review.openstack.org/315402 | 16:30 |
| *** dprince has joined #openstack-ironic | 16:31 | |
| *** piet has quit IRC | 16:32 | |
| *** piet has joined #openstack-ironic | 16:33 | |
| *** tobias_fiberdata has quit IRC | 16:34 | |
| *** piet has quit IRC | 16:36 | |
| *** piet has joined #openstack-ironic | 16:37 | |
| *** trown is now known as trown|lunch | 16:39 | |
| *** mgould is now known as mgould|afk | 16:40 | |
| *** abcxyz has joined #openstack-ironic | 16:41 | |
| *** sdake has quit IRC | 16:42 | |
| *** mbound has joined #openstack-ironic | 16:43 | |
| *** hoangcx has quit IRC | 16:43 | |
| *** bcornec2 has joined #openstack-ironic | 16:44 | |
| *** cdearborn has joined #openstack-ironic | 16:44 | |
| *** yolanda has quit IRC | 16:46 | |
| *** [2]cdearborn has joined #openstack-ironic | 16:50 | |
| JayF | sambetts: so that means partition images are broken for all of IPA right now, right? | 16:50 |
| JayF | sambetts: we gotta test that in the gate :( | 16:50 |
| *** jvgrant_ has quit IRC | 16:50 | |
| sambetts | JayF: I would assume so yes, and thats what the job that caught this is for right? I guess we need to move away from the whole instantly blame TinyIPA stigma | 16:51 |
| *** mtanin___ has joined #openstack-ironic | 16:54 | |
| *** mtanino has quit IRC | 16:55 | |
| JayF | sambetts: I don't see those jobs failing? I'm confused | 16:55 |
| JayF | sambetts: was it just one patch breaking something vs all ci being broken? | 16:56 |
| sambetts | JayF: so right now that job still boots whole disk images | 16:56 |
| *** milan has joined #openstack-ironic | 16:56 | |
| vdrok | ugh, we have a broken unittest in master\ | 16:56 |
| sambetts | JayF: vdroks patch fixes that | 16:56 |
| *** rpioso has joined #openstack-ironic | 16:57 | |
| thiagop | jlvillal: QA meeting in 2m? | 16:57 |
| krtaylor | jroll, the stackalytics column in the thrid party test systems table is now currrent -> https://wiki.openstack.org/wiki/Ironic/Drivers#3rd_Party_CI_required_implementation_status | 16:57 |
| jroll | krtaylor: amazing, I was going to ask you about that in 3 minutes :) | 16:57 |
| sambetts | \o/ | 16:57 |
| JayF | sambetts: I'm wondering if we need jobs to do both? I don't want whole disk support missed in ci either (especially since I run it, haha) | 16:58 |
| krtaylor | jroll, yep, I remembered my action this morning :) | 16:58 |
| jroll | heh | 16:58 |
| *** [1]cdearborn has quit IRC | 16:58 | |
| jlvillal | thiagop: sim senhor! | 16:58 |
| krtaylor | jroll, but we still need to remind everyone to fulfill their infra requirements for third party test systems | 16:58 |
| krtaylor | but we can discuss in meeting... | 16:59 |
| sambetts | JayF: there will be, https://review.openstack.org/#/c/329625/ there are 2 agent_ssh jobs | 16:59 |
| thiagop | jlvillal: oh, not "senhor". I have enought white hair already | 16:59 |
| *** NobodyCam has quit IRC | 16:59 | |
| *** mgkwill has quit IRC | 16:59 | |
| jroll | krtaylor: indeed | 16:59 |
| sambetts | JayF: one whole disk and one partition | 16:59 |
| JayF | oh beautiful! | 16:59 |
| * jroll slides over to meeting room | 16:59 | |
| JayF | Thanks for linking me to that patch | 16:59 |
| *** dtantsur is now known as dtantsur|afk | 17:00 | |
| sambetts | JayF: thats where vdrok saw the error | 17:00 |
| JayF | that links all the pieces together | 17:00 |
| JayF | thats what I get for being too busy to read all the scrollback :( sorry for making you repeat it all | 17:00 |
| *** NobodyCam has joined #openstack-ironic | 17:00 | |
| *** mgkwill has joined #openstack-ironic | 17:00 | |
| sambetts | :D thats ok | 17:01 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Fix broken conductor unittest https://review.openstack.org/330087 | 17:01 |
| *** mtanino has joined #openstack-ironic | 17:01 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Properly set ephemeral size in agent drivers https://review.openstack.org/330089 | 17:01 |
| *** bcornec2 has quit IRC | 17:01 | |
| *** mtanin___ has quit IRC | 17:03 | |
| jroll | vdrok: hrm, how does that suddenly break? :/ | 17:04 |
| vdrok | jroll: it gives me attribute error | 17:04 |
| jroll | hrm, new mock version maybe? | 17:05 |
| * jroll checks | 17:05 | |
| vdrok | i see only assert_called_with and assert_called_once_with | 17:05 |
| *** cdearborn has quit IRC | 17:06 | |
| rloo | vdrok: https://bugs.launchpad.net/ironic/+bug/1592852 | 17:06 |
| openstack | Launchpad bug 1592852 in Ironic "test_cleanup_cleanwait_timeout AttributeError: assert_called_once" [Undecided,In progress] - Assigned to Corey Bryant (corey.bryant) | 17:06 |
| jroll | yeah, that does look wrong, just curious why it's suddenly broken | 17:06 |
| rloo | i'm also curious as to why it suddenly broke. | 17:06 |
| *** cdearborn has joined #openstack-ironic | 17:07 | |
| *** PollyZ_ has joined #openstack-ironic | 17:07 | |
| vdrok | heh, yep, 2.0.0. mock has this method :) | 17:07 |
| vdrok | rloo: jroll ^^ | 17:07 |
| vdrok | i'll abandon | 17:07 |
| jroll | hrm, tox -repy27 passes on master for me | 17:07 |
| rloo | vdrok: check your dev env. you should be using mock 2.0.0, it is in test-requirements.txt | 17:09 |
| vdrok | yes, thanks rloo, false alarm :) | 17:09 |
| rloo | vdrok: phew! | 17:09 |
| vdrok | need to recreate it more often | 17:09 |
| jroll | though I wonder if it isn't in python 3.5 or something | 17:10 |
| jroll | idk | 17:10 |
| *** PollyZ has quit IRC | 17:10 | |
| *** sacharya has quit IRC | 17:11 | |
| *** sacharya has joined #openstack-ironic | 17:12 | |
| *** bcornec2 has joined #openstack-ironic | 17:13 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Properly set ephemeral size in agent drivers https://review.openstack.org/330089 | 17:16 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Change the logic of selecting image for tests https://review.openstack.org/329625 | 17:16 |
| jlvillal | JayF: http://ci-watch.tintri.com/project?project=ironic&time=7+days looks pretty good. The link from krtaylor | 17:18 |
| JayF | nice | 17:18 |
| JayF | I can't see any situations where grenade failed that something else did | 17:18 |
| JayF | sambetts: vdrok: We need to add that partition image job to IPA as a -src job too, one each for coreos and tinyipa | 17:19 |
| JayF | sambetts: vdrok: before it can vote on ironic imo | 17:19 |
| sambetts | Totally | 17:19 |
| openstackgerrit | Sam Betts proposed openstack/ironic-inspector: Add generic LLDP processing hook https://review.openstack.org/321082 | 17:20 |
| *** PollyZ has joined #openstack-ironic | 17:21 | |
| *** [2]cdearborn has quit IRC | 17:21 | |
| *** pcaruana has joined #openstack-ironic | 17:21 | |
| vdrok | JayF: agree | 17:22 |
| jlvillal | TheJulia: Would you mind looking at: https://review.openstack.org/#/c/330001/ Thanks in advance :) | 17:23 |
| jlvillal | JayF: Yes. Failures seem to map close to other failures. I did see one of the 'apache restart' failures. But that isn't the Grenade job's issue | 17:23 |
| *** PollyZ_ has quit IRC | 17:24 | |
| JayF | Yeah; I mean I almost want to say I think that's compelling enough to make it voting | 17:24 |
| JayF | the only thing I'd be curious about would be a negative test | 17:24 |
| JayF | should we make sure something that does break upgrades fails grenade? | 17:24 |
| jroll | JayF: ++ | 17:26 |
| *** Sukhdev has quit IRC | 17:26 | |
| *** piet has quit IRC | 17:29 | |
| *** fragatina has quit IRC | 17:29 | |
| *** appu103_ has quit IRC | 17:30 | |
| *** aparnav has joined #openstack-ironic | 17:30 | |
| *** moshele has quit IRC | 17:33 | |
| *** aparnav has quit IRC | 17:34 | |
| *** electrofelix has quit IRC | 17:34 | |
| *** appu103_ has joined #openstack-ironic | 17:35 | |
| *** sacharya_ has joined #openstack-ironic | 17:40 | |
| *** sacharya has quit IRC | 17:43 | |
| TheJulia | jlvillal: sure | 17:44 |
| jlvillal | TheJulia: Thanks. vdrok pointed out that it was correct because of the version of mock we use. I guess that call is new in mock. | 17:45 |
| jlvillal | TheJulia: The existing code is correct, that is. | 17:45 |
| TheJulia | jlvillal: done | 17:46 |
| jlvillal | :) | 17:47 |
| TheJulia | Yeah, I glanced at the bug report and immediately thought something is odd, but it is all good | 17:47 |
| *** bcornec2 has quit IRC | 17:47 | |
| *** mtanino has quit IRC | 17:48 | |
| krtaylor | thiagop, sambetts - I removed the duplicate driver wiki table entries - sorry about that, I was searching on the driver name from stackalytics | 17:51 |
| *** trown|lunch is now known as trown | 17:51 | |
| thiagop | krtaylor: thanks for that | 17:52 |
| *** bcornec2 has joined #openstack-ironic | 17:52 | |
| vdrok | sambetts: success \o/ http://logs.openstack.org/25/329625/3/check/gate-tempest-dsvm-ironic-agent_ssh-partition-image-nv/68cbcd0/console.html | 17:54 |
| sambetts | vdrok: \o/ | 17:55 |
| vdrok | good night everyone! | 17:55 |
| sambetts | night vdrok! | 17:55 |
| jlvillal | vdrok: Ciao | 17:55 |
| *** bcornec1 has quit IRC | 17:55 | |
| vdrok | jlvillal: paka-paka :D | 17:55 |
| NobodyCam | night vdrok | 17:55 |
| jlvillal | vdrok: I can't help but think mock would have been better if they had done it as: mock.assert.what_we_want_to_assert() | 17:55 |
| jlvillal | vdrok: Dos ve dos | 17:55 |
| * jlvillal has no idea if that would have been easily done or not. Too late now. | 17:56 | |
| *** bcornec2 has quit IRC | 17:57 | |
| *** PollyZ has quit IRC | 17:57 | |
| *** daemontool has quit IRC | 17:59 | |
| *** PollyZ has joined #openstack-ironic | 18:02 | |
| *** jvgrant_ has joined #openstack-ironic | 18:02 | |
| *** appu103_ has quit IRC | 18:02 | |
| *** mtanino has joined #openstack-ironic | 18:02 | |
| *** mbound has quit IRC | 18:06 | |
| *** ccamacho is now known as ccamacho|out | 18:08 | |
| *** fragatina has joined #openstack-ironic | 18:08 | |
| *** jvgrant_ has quit IRC | 18:09 | |
| *** bcornec1 has joined #openstack-ironic | 18:12 | |
| *** ChubYann has joined #openstack-ironic | 18:15 | |
| devananda | hah! figured out why https://review.openstack.org/#/c/325599/ is failing some tests and not others | 18:16 |
| * jroll curious | 18:16 | |
| devananda | in gate-tempest-dsvm-ironic-pxe_ipa Nova is creating a neutron port and then Ironic is tyring to update it | 18:16 |
| devananda | and fails with PortNotFound, because I changed the service tenant | 18:16 |
| devananda | why is this NOT a problem for gate-tempest-dsvm-ironic-agent_ssh ??? | 18:17 |
| jroll | O_o | 18:18 |
| jroll | it should be | 18:18 |
| devananda | yea - it's not | 18:18 |
| devananda | http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-ir-cond.txt.gz isn't updating hte neutron port | 18:19 |
| devananda | created here - http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-n-cpu.txt.gz#_2016-06-15_03_51_35_563 | 18:19 |
| devananda | but it's never invoked from ironic | 18:19 |
| devananda | wtf? | 18:19 |
| devananda | here's the error in the pxe_ipa job: http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-pxe_ipa/dbe5abc/logs/screen-ir-cond.txt.gz?#_2016-06-15_03_53_17_366 | 18:20 |
| devananda | here's the matching port create in the pxe_ipa job: http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-pxe_ipa/dbe5abc/logs/screen-n-cpu.txt.gz#_2016-06-15_03_51_35_227 | 18:20 |
| JayF | what do the *_ipmitool jobs do? | 18:21 |
| * JayF looks | 18:21 | |
| JayF | it looks like pxe_* driver vs agent_* driver is the difference, at a glance the pxe_ipmitool driver fails too and the agent_ipmitool job passes | 18:21 |
| jroll | what is going on here | 18:22 |
| jroll | this is weird | 18:22 |
| jroll | like, we can't pxe if we don't update the port | 18:22 |
| devananda | JayF: yes | 18:22 |
| devananda | passing job uses agent, failing job uses pxe. passing job downloads image in IPA, failing job writes image successfully over iscsi | 18:23 |
| JayF | https://review.openstack.org/gitweb?p=openstack/ironic.git;a=blob;f=devstack/lib/ironic;h=5df3e09b6a1c9938d9f53772719b2c95e0b1404b;hb=ff0150dec26741756640c884b9a8c25dd25fb131#l513 | 18:23 |
| JayF | could that locally set tenant_id be leaking? but i don't think that explains the behavior | 18:23 |
| JayF | that's just the only place I could see devstack changing behavior based on driver | 18:24 |
| JayF | so I guess this has to be a bug in actual ironic code? | 18:24 |
| *** bcornec1 has quit IRC | 18:25 | |
| jroll | it just never plugs the vifs | 18:26 |
| jroll | here's where it gets into _plug_vifs http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-n-cpu.txt.gz#_2016-06-15_03_51_35_841 | 18:26 |
| jroll | https://github.com/openstack/nova/blob/master/nova/virt/ironic/driver.py#L992 | 18:26 |
| jroll | but it just doesn't do the rest? | 18:26 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: Canery test for ironic networking changes https://review.openstack.org/330138 | 18:27 |
| jroll | or maybe it does, hard to tell | 18:27 |
| devananda | context here is this change in my patch: https://review.openstack.org/#/c/325599/6/devstack/lib/ironic | 18:28 |
| devananda | I *think* that has exposed a difference in the behavior of these drivers? | 18:29 |
| jroll | I'm not going that far yet, something is weird | 18:30 |
| jroll | how did the agent pxe boot if we didn't pass dhcp opts to neutron? | 18:30 |
| devananda | in both jobs, nova creates the port | 18:30 |
| *** e0ne has quit IRC | 18:30 | |
| jroll | I think it is running _plug_vifs fine, I see lots of PATCH /v1/ports/foo in that timeframe | 18:30 |
| *** bcornec1 has joined #openstack-ironic | 18:31 | |
| * sambetts -> home | 18:31 | |
| sambetts | Night all | 18:31 |
| jroll | erm | 18:31 |
| jroll | look at this: http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-ir-cond.txt.gz#_2016-06-15_03_51_12_701 | 18:31 |
| jroll | down to http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-ir-cond.txt.gz#_2016-06-15_03_51_12_752 | 18:32 |
| jroll | clearly not a real deploy | 18:32 |
| *** sambetts is now known as sambetts|afk | 18:32 | |
| jroll | unless we also test that api in tempest with fake driver | 18:32 |
| jroll | and this is the wrong deploy | 18:32 |
| *** bcornec2 has joined #openstack-ironic | 18:33 | |
| thiagop | night sambetts|afk | 18:33 |
| jroll | this should be raising errors correctly if things are wacky (e.g. no port info) https://github.com/openstack/ironic/blob/master/ironic/dhcp/neutron.py#L146 | 18:33 |
| * jroll is super confused here | 18:35 | |
| jroll | devananda: hrm, it did get dhcp opts somewhere somehow http://logs.openstack.org/99/325599/6/check/gate-tempest-dsvm-ironic-agent_ssh/698df48/logs/screen-q-svc.txt.gz#_2016-06-15_03_51_36_749 | 18:37 |
| jroll | devananda: I wonder if one of two things are happening here | 18:37 |
| jroll | 1) the token expired and we re-auth'd with the neutron creds in ironic.conf | 18:38 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: Canery test for ironic networking changes https://review.openstack.org/330138 | 18:38 |
| devananda | 2016-06-15 03:53:17.364 12227 INFO neutron.wsgi [req-67765959-cc23-4507-8066-e7c3dd140d73 ironic -] 104.130.211.184 - - [15/Jun/2016 03:53:17] "PUT /v2.0/ports/8f9297de-3929-424e-9538-5ea20cf902de.json HTTP/1.1" 404 338 0.028829 | 18:38 |
| devananda | I think that's the problem | 18:38 |
| *** PollyZ_ has joined #openstack-ironic | 18:38 | |
| jroll | 2) somehow we're using different neutron tokens/creds in agent vs pxe | 18:38 |
| jroll | well | 18:38 |
| jroll | see my link? | 18:38 |
| jroll | timeline lines up correctly on my link too | 18:38 |
| jroll | 2016-06-15 03:51:36.749 12426 DEBUG neutron.api.v2.base [req-bff773b1-b506-4f0b-8e4b-5426fb6c9546 admin -] Request body: {u'port': {u'extra_dhcp_opts': [{u'opt_value': u'undionly.kpxe', | 18:39 |
| jroll | 2016-06-15 03:51:37.158 22480 DEBUG ironic.dhcp.neutron [req-d6ce37da-3cfb-460f-b58f-83a947b37e94 - - - - -] Waiting 15 seconds for Neutron. update_dhcp_opts /opt/stack/new/ironic/ironic/dhcp/neutron.py:215 | 18:39 |
| *** PollyZ has quit IRC | 18:39 | |
| jroll | (or is yours the pxe driver, mine is agent) | 18:39 |
| jroll | yeah, yours must be from the failure | 18:40 |
| *** joprovost1 has joined #openstack-ironic | 18:40 | |
| jroll | right, so that is the problem, question is if 404 is because it's actually not there, or because it belongs to another user | 18:40 |
| devananda | the 404 error is clearly coming from a request made by the 'ironic' user | 18:41 |
| devananda | whereas the port was created by the 'admin' user | 18:41 |
| jroll | why was the port created by the admin user? | 18:41 |
| jroll | it should be created by the 'demo' user | 18:42 |
| devananda | so, two problems I see | 18:42 |
| *** joprovost has quit IRC | 18:42 | |
| *** joprovost1 is now known as joprovost | 18:42 | |
| jroll | you're right thuogh, my port update came from admin user | 18:42 |
| devananda | 1. in the agent driver, there's only one PUT request to Neutron, which I believe is coming from Nova, not Ironic | 18:42 |
| devananda | 2. in the pxe driver, both Nova and Ironic need to be able to see and update the same neutron port (and my patch has run afoul of this) | 18:43 |
| jroll | 1. no, these both use the PxeBoot interface. it follows this code path: https://github.com/openstack/ironic/blob/master/ironic/dhcp/neutron.py#L146 | 18:44 |
| jroll | however | 18:45 |
| jroll | the neutron client will transparently re-authenticate if needed, using the creds defined in ironic.conf | 18:45 |
| jroll | also for 2. I'd clarify that to "ironic needs to be able to update ports on behalf of the user" | 18:45 |
| jroll | in both drivers | 18:45 |
| devananda | yea, well, it seems to do that as the admin user, on behalf of the end user, which is perhaps the problem | 18:46 |
| devananda | I need to run out for a bit ... this is going to be fun to unwind :-/ | 18:46 |
| devananda | wish I could stay and keep working on it, but I'll bbl | 18:46 |
| jroll | yeah, we update ports on behalf of the user, it's kinda required | 18:46 |
| jroll | unless we can pass that context all the way down, we'll need to be admin | 18:47 |
| jroll | see ya | 18:47 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/dhcp/neutron.py#L87 | 18:50 |
| devananda | https://review.openstack.org/#/c/325599/6/devstack/lib/ironic | 18:50 |
| devananda | I moved the ironic user from 'admin' to 'service' | 18:51 |
| devananda | but why DIDN"T that break all the drivers? | 18:51 |
| devananda | that's what I don't get | 18:51 |
| devananda | anyhow... bbl | 18:51 |
| jroll | neutron re-auths if it gets 401 | 18:51 |
| jroll | I assume you got lucky and the token expired just in case | 18:51 |
| jroll | but I'm not 100% sure what all is going on here | 18:51 |
| *** bcornec1 has quit IRC | 18:52 | |
| jroll | oh, I see now | 18:52 |
| jroll | wtf | 18:52 |
| devananda | how could ironic have had a valid admin-tenant token? | 18:53 |
| jroll | yeah I have nfi | 18:54 |
| jroll | oh you know what | 18:54 |
| *** SiRiuS__ has quit IRC | 18:54 | |
| jroll | I bet there's magic here https://github.com/openstack/ironic/blob/master/ironic/common/context.py#L47 | 18:55 |
| jroll | maybe not, ugh | 18:55 |
| *** Sukhdev has joined #openstack-ironic | 19:02 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize Config Options - patch merge, cleanup https://review.openstack.org/330160 | 19:04 |
| *** piet has joined #openstack-ironic | 19:05 | |
| *** mbound has joined #openstack-ironic | 19:06 | |
| *** mbound has quit IRC | 19:11 | |
| *** bcornec1 has joined #openstack-ironic | 19:15 | |
| *** dprince has quit IRC | 19:21 | |
| *** skramaja has quit IRC | 19:24 | |
| *** jvgrant_ has joined #openstack-ironic | 19:25 | |
| *** piet has quit IRC | 19:25 | |
| *** piet has joined #openstack-ironic | 19:26 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Add Ironic specs process to the code contribution guide https://review.openstack.org/330167 | 19:31 |
| *** MattMan has quit IRC | 19:32 | |
| *** MattMan has joined #openstack-ironic | 19:32 | |
| *** openstackgerrit has quit IRC | 19:33 | |
| *** openstackgerrit has joined #openstack-ironic | 19:33 | |
| *** piet has quit IRC | 19:33 | |
| *** sdake has joined #openstack-ironic | 19:36 | |
| *** skramaja has joined #openstack-ironic | 19:36 | |
| *** bcornec1 has quit IRC | 19:37 | |
| *** bcornec1 has joined #openstack-ironic | 19:40 | |
| *** mrda has quit IRC | 19:40 | |
| *** sdake_ has joined #openstack-ironic | 19:40 | |
| openstackgerrit | Merged openstack/ironic: Use assert_called_once_with in test_cleanup_cleanwait_timeout https://review.openstack.org/330001 | 19:41 |
| *** sdake has quit IRC | 19:42 | |
| *** mrda has joined #openstack-ironic | 19:42 | |
| *** SiRiuS__ has joined #openstack-ironic | 19:46 | |
| *** e0ne has joined #openstack-ironic | 19:48 | |
| *** bcornec1 has quit IRC | 19:51 | |
| *** abcxyz has quit IRC | 19:53 | |
| *** hoangcx has joined #openstack-ironic | 19:54 | |
| *** bcornec1 has joined #openstack-ironic | 19:58 | |
| *** jcoufal_ has joined #openstack-ironic | 19:59 | |
| *** krotscheck_dcm is now known as krotscheck | 20:00 | |
| *** jcoufal has quit IRC | 20:02 | |
| *** Sukhdev has quit IRC | 20:02 | |
| *** jvgrant_ has quit IRC | 20:02 | |
| *** joprovost has quit IRC | 20:03 | |
| *** dprince has joined #openstack-ironic | 20:07 | |
| *** sdake_ has quit IRC | 20:07 | |
| *** krsacme has joined #openstack-ironic | 20:07 | |
| *** sdake has joined #openstack-ironic | 20:07 | |
| jlvillal | Did keystone client break python-ironicclient stable/liberty? http://logs.openstack.org/51/234451/1/gate/gate-python-ironicclient-python27/c6c6b5b/console.html#_2016-06-14_20_06_51_972 | 20:09 |
| *** skramaja has quit IRC | 20:11 | |
| *** sdake_ has joined #openstack-ironic | 20:11 | |
| *** sdake has quit IRC | 20:12 | |
| *** jvgrant_ has joined #openstack-ironic | 20:17 | |
| jroll | jlvillal: looks like it, but a rebase should fix: https://review.openstack.org/#/c/237251/5/requirements.txt | 20:21 |
| jroll | that merged yesterday | 20:21 |
| jlvillal | jroll: Thanks. | 20:22 |
| jroll | np | 20:22 |
| jlvillal | jroll: The recheck worked, zuul is showing me. | 20:23 |
| *** bcornec1 has quit IRC | 20:23 | |
| *** ijw has quit IRC | 20:26 | |
| *** sdake_ has quit IRC | 20:27 | |
| *** mtanino has quit IRC | 20:43 | |
| rloo | jroll or anyone else. wrt the network patches, I don't think I was paying attention. This seems to be the start of the chain now, right? https://review.openstack.org/#/c/317390/ | 20:51 |
| jroll | rloo: yes, and the devstack code for it is a separate chain | 20:52 |
| rloo | and the next patch: https://review.openstack.org/#/c/206244/ portgroups API stuff, still has the -2 so we don't merge anything more until some number (the rest?) of the patches are good? | 20:52 |
| jroll | we should reconsider that -2 imo | 20:53 |
| rloo | jroll: we meaning deva et al. i think deva was against putting in the portgroup API if the rest doesn't work. | 20:53 |
| rloo | jroll: and we're ok with this chain getting merged, before the devstack code, or can the devstack chain be reviewed/merged in parallel or before this chain? | 20:54 |
| jroll | rloo: I'd say in parallel, the devstack code is an easier review so probably easy to land first | 20:54 |
| rloo | jroll: thx | 20:55 |
| jroll | np | 20:55 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: Canery test for ironic networking changes https://review.openstack.org/330138 | 20:59 |
| *** jvgrant_ has quit IRC | 20:59 | |
| krtaylor | cdearborn, do you want me to add DRAC to stackalytics? | 21:00 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: another test for ironic networking https://review.openstack.org/330210 | 21:02 |
| jlvillal | So two +2s on the Adopt state patch: https://review.openstack.org/#/c/275766/ I'm too scared to +A it without more reviews :) | 21:03 |
| krtaylor | cdearborn, I went ahead and added the DRAC record, feel free to review the patch for any corrections needed, patch getting close | 21:04 |
| TheJulia | I think 206244 may be fine, but I'm rather concerned regarding 285852 at the moemnt... hence the two test jobs I have spinning against bifrost at the moment pulling in patches on that change set for testing | 21:04 |
| jroll | TheJulia: https://review.openstack.org/#/c/285852/54/ironic/drivers/modules/network/none.py | 21:06 |
| jroll | I think that is what standalone folks are meant to use, right? | 21:06 |
| jroll | and then 'flat' covers the existing neutron use case | 21:07 |
| TheJulia | yeah, except if they use an older client, or the current client defaults... they can never ever create the node | 21:07 |
| TheJulia | at least on that specific patch and the database patch | 21:07 |
| jroll | hrm | 21:07 |
| * jroll looks | 21:07 | |
| jroll | why you gotta nerd snipe me when I'm about to leave :P | 21:07 |
| *** spandhe has joined #openstack-ironic | 21:08 | |
| TheJulia | I've spent the last two or so hours looking at it, so you might be looking for a while | 21:08 |
| TheJulia | jroll: look tomorrow! | 21:08 |
| NobodyCam | lol | 21:08 |
| jroll | TheJulia: heh | 21:08 |
| jroll | so | 21:08 |
| jroll | https://review.openstack.org/#/c/317391/6/ironic/objects/node.py | 21:08 |
| jroll | looks like it defaults to none? | 21:08 |
| jroll | I should stop talking without reading it all | 21:08 |
| jroll | ah I see it | 21:09 |
| jroll | TheJulia: I agree, default to 'none' if not using neutron dhcp | 21:09 |
| TheJulia | heh :) | 21:10 |
| TheJulia | go enjoy your evening | 21:10 |
| jroll | 'enjoy' | 21:10 |
| * jroll has car work to do | 21:10 | |
| TheJulia | okay, go get a beer, and work on the car then ;) | 21:11 |
| jroll | not the fun kind either, stripped bolts | 21:11 |
| jroll | hehe | 21:11 |
| jroll | have a good night :) | 21:11 |
| TheJulia | ouch :( | 21:11 |
| TheJulia | well, hopefully it goes well then ;) | 21:11 |
| NobodyCam | jroll: the new Audi? | 21:11 |
| *** nicodemos is now known as nico-afk | 21:11 | |
| *** hoangcx has quit IRC | 21:11 | |
| jroll | should be fine, buddy has every tool imaginable | 21:11 |
| jroll | NobodyCam: no, chevy cruze, oil drain plug is rounded off :/ | 21:12 |
| NobodyCam | oh | 21:12 |
| * NobodyCam offers two beers | 21:12 | |
| jroll | I also have audi work to do but that's part swapping for fun and horsepower :P | 21:12 |
| NobodyCam | lol | 21:12 |
| TheJulia | heh | 21:12 |
| *** dprince has quit IRC | 21:13 | |
| jroll | see y'all tomorrow \o | 21:13 |
| TheJulia | goodnight | 21:13 |
| *** nico-afk has quit IRC | 21:14 | |
| *** e0ne has quit IRC | 21:14 | |
| *** trown is now known as trown|outtypewww | 21:15 | |
| *** spandhe has quit IRC | 21:16 | |
| *** spandhe has joined #openstack-ironic | 21:16 | |
| rloo | jlvillal: no guts no glory. should be safe to +A the adopt one, it had two +2's before I intervened. | 21:16 |
| TheJulia | heh | 21:17 |
| *** joprovost has joined #openstack-ironic | 21:17 | |
| NobodyCam | ANC? | 21:17 |
| TheJulia | yup | 21:19 |
| *** Sukhdev has joined #openstack-ironic | 21:22 | |
| *** joprovost has quit IRC | 21:27 | |
| *** Sukhdev has quit IRC | 21:32 | |
| *** bcornec1 has joined #openstack-ironic | 21:32 | |
| *** ccamacho|out has quit IRC | 21:33 | |
| *** Sukhdev has joined #openstack-ironic | 21:33 | |
| *** Sukhdev has quit IRC | 21:36 | |
| *** bcornec1 has quit IRC | 21:37 | |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic: [WIP] Using Oneview drivers w/ IRONIC_IS_HARDWARE https://review.openstack.org/328451 | 21:40 |
| *** cdearborn has quit IRC | 21:47 | |
| *** PollyZ_ has quit IRC | 21:48 | |
| *** bcornec1 has joined #openstack-ironic | 21:49 | |
| *** lindycod_ has quit IRC | 21:53 | |
| *** openstackgerrit has quit IRC | 22:02 | |
| *** catintheroof has joined #openstack-ironic | 22:04 | |
| *** openstackgerrit has joined #openstack-ironic | 22:06 | |
| *** moshele has joined #openstack-ironic | 22:07 | |
| *** jcoufal_ has quit IRC | 22:09 | |
| *** Sukhdev has joined #openstack-ironic | 22:09 | |
| *** Sukhdev has quit IRC | 22:09 | |
| *** Sukhdev has joined #openstack-ironic | 22:10 | |
| *** mbound has joined #openstack-ironic | 22:10 | |
| *** bcornec1 has quit IRC | 22:11 | |
| *** thiagop has quit IRC | 22:12 | |
| *** Sukhdev has quit IRC | 22:15 | |
| *** bcornec1 has joined #openstack-ironic | 22:22 | |
| *** _vsaienko has quit IRC | 22:24 | |
| *** hoangcx has joined #openstack-ironic | 22:27 | |
| *** frontrunner has quit IRC | 22:31 | |
| JayF | So I'm fairly certain at this point the DIB image build support in devstack that's recently merged is broken. | 22:31 |
| JayF | I've had it break multiple different ways, most recently with 2016-06-15 22:30:57.541 | /opt/stack/ironic/devstack/lib/ironic: line 1058: disk-image-create: command not found | 22:32 |
| JayF | this is from a fresh install, with the config as listed in documentation + IRONIC_BUILD_DEPLOY_RAMDISK=True IRONIC_RAMDISK_TYPE=dib IRONIC_DIB_RAMDISK_OPTIONS=fedora | 22:32 |
| *** rpioso has quit IRC | 22:34 | |
| JayF | https://bugs.launchpad.net/ironic/+bug/1593005 filed | 22:38 |
| openstack | Launchpad bug 1593005 in Ironic "DIB Image building in Devstack nonworking" [Undecided,New] | 22:38 |
| *** catintheroof has quit IRC | 22:41 | |
| *** Sukhdev has joined #openstack-ironic | 22:42 | |
| cinerama | weird | 22:44 |
| *** Sukhdev has quit IRC | 22:44 | |
| *** Sukhdev has joined #openstack-ironic | 22:46 | |
| JayF | I put what I believe is the cause in the bug | 22:47 |
| JayF | for an ironic-only devstack configured stock, doesn't look like dib is ever pulled in | 22:48 |
| *** Sukhdev has quit IRC | 22:53 | |
| *** jcoufal has joined #openstack-ironic | 22:55 | |
| *** sacharya_ has quit IRC | 22:56 | |
| *** sacharya has joined #openstack-ironic | 23:00 | |
| *** rbudden has quit IRC | 23:01 | |
| *** sacharya_ has joined #openstack-ironic | 23:02 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
