| NobodyCam | however the default image DID deploy and is working :) nice! | 00:00 |
|---|---|---|
| *** jlvillal_ has left #openstack-ironic | 00:01 | |
| jroll | NobodyCam: as the demo user you can't hit ironic :P | 00:02 |
| NobodyCam | i'm not admin ...doh | 00:03 |
| NobodyCam | yeppers I'm running as demo | 00:03 |
| * NobodyCam notes difference from his devtest setup | 00:03 | |
| NobodyCam | :-p | 00:04 |
| *** jlvillal_ has joined #openstack-ironic | 00:04 | |
| NobodyCam | and now goes away... g'night all... | 00:04 |
| *** BadCub has quit IRC | 00:04 | |
| jroll | heh | 00:05 |
| jroll | night NobodyCam gl | 00:05 |
| *** jlvillal_ has quit IRC | 00:05 | |
| *** jlvillal has joined #openstack-ironic | 00:06 | |
| *** naohirot has joined #openstack-ironic | 00:07 | |
| *** dttocs has quit IRC | 00:07 | |
| *** kkoski has quit IRC | 00:10 | |
| *** r-daneel has quit IRC | 00:15 | |
| *** penick has quit IRC | 00:16 | |
| *** jlvillal has left #openstack-ironic | 00:16 | |
| *** romcheg has quit IRC | 00:22 | |
| *** Haomeng|2 has joined #openstack-ironic | 00:23 | |
| *** Haomeng has quit IRC | 00:24 | |
| *** Marga_ has quit IRC | 00:24 | |
| *** jlvillal has joined #openstack-ironic | 00:25 | |
| *** Marga_ has joined #openstack-ironic | 00:25 | |
| *** naohirot has quit IRC | 00:28 | |
| *** naohirot has joined #openstack-ironic | 00:29 | |
| *** devlaps has quit IRC | 00:41 | |
| *** penick has joined #openstack-ironic | 00:42 | |
| *** wshao has joined #openstack-ironic | 00:51 | |
| *** jerryz_ has joined #openstack-ironic | 00:52 | |
| *** chuckC_ has quit IRC | 00:58 | |
| *** ijw has quit IRC | 01:01 | |
| TheJulia | NobodyCam: devananda: deployments are still failing, but it looks like we can't use a config drive with the agent's image write out. qemu-img converts the raw disk on a write-out to /dev/sda, so when parted comes along to create the partition for the config drive, all of the space appears gone. | 01:08 |
| jroll | TheJulia: parted should deal with that... | 01:09 |
| jroll | TheJulia: our production uses agent + configdrive | 01:09 |
| TheJulia | I wonder if it is the image that your writing out then? | 01:10 |
| TheJulia | to be more specific, the partition table in the image? | 01:10 |
| jroll | maybe, though I don't thinhk they're special | 01:10 |
| * jroll looks at code again | 01:10 | |
| jroll | TheJulia: is your raw image the same size as the drive? | 01:11 |
| TheJulia | http://paste.openstack.org/show/188361/ Although I have rebooted to get this, but the command is exactly what the next step would have been on the config drive write out | 01:12 |
| TheJulia | jroll: The partition table in the image is something like 1.5GB, created as a sparse file, it only has ~500 MB of data in it | 01:13 |
| jroll | 'parted: invalid token: primary' seems weird | 01:14 |
| jroll | hmm | 01:14 |
| TheJulia | qemu-img's convert code makes my head hurt a little | 01:14 |
| jroll | lol | 01:14 |
| jroll | I'm not super familiar with images/partitions/etc | 01:14 |
| jroll | lucas or JayF would be better people to ask tomorrow | 01:15 |
| TheJulia | I kind of wonder if the image just has a bad partition table | 01:15 |
| jroll | could be | 01:15 |
| *** penick has quit IRC | 01:15 | |
| jroll | there's a cirros image you could test with | 01:15 |
| * jroll finds | 01:15 | |
| jroll | I think it's http://download.cirros-cloud.net/0.3.3/cirros-0.3.3-x86_64-disk.img | 01:18 |
| jroll | but there's more: http://download.cirros-cloud.net/0.3.3/ | 01:18 |
| naohirot | SpamapS: hi, good evening | 01:21 |
| * TheJulia fires that off and waits | 01:21 | |
| * TheJulia has a sudden craving for banana pudding and blames JayF for it :) | 01:22 | |
| naohirot | good evening jroll | 01:22 |
| jroll | lol | 01:22 |
| jroll | morning naohirot :) | 01:22 |
| naohirot | jroll: https://review.openstack.org/#/c/159888/2/elements/ramdisk-base/extra-data.d/01-inject-ramdisk-build-files | 01:23 |
| jroll | yes? | 01:23 |
| naohirot | jroll: can I ask a question regarding ^^^? | 01:23 |
| jroll | you can ask, we'll see if I can answer | 01:23 |
| naohirot | jroll: I still thinking what the contents of '. /etc/virtual-media-settings.sh' should be. | 01:24 |
| naohirot | jroll: even if we created '. /etc/virtual-media-settings.sh' , I couldn't figure out how to passing parameter to the busy box environment. | 01:25 |
| TheJulia | jroll: Yeah, looks like it may be the partition table. Just changing the image over to cirros and it worked like a chammp | 01:25 |
| * naohirot TheJulia, hi, sorry for interrupting you. | 01:26 | |
| TheJulia | naohirot: Hi, Perfectly fine. :) | 01:27 |
| naohirot | TheJulia: thank you for saying that :) | 01:27 |
| jroll | TheJulia: awesome :) | 01:28 |
| jroll | naohirot: something like: | 01:28 |
| jroll | export __SYSFS_VMEDIA_DEVICE_MODEL_CONTENTS ${SYSFS_VMEDIA_DEVICE_MODEL_CONTENTS:-\"(virtualmedia|Virtual Floppy0)\"} | 01:28 |
| jroll | export __WAIT_TIME_FOR_VFLOPPY_MOUNT ${WAIT_TIME_FOR_VFLOPPY_MOUNT:-15} | 01:28 |
| jroll | and then source that, and use e.g. __SYSFS_VMEDIA_DEVICE_MODEL_CONTENTS as needed | 01:28 |
| naohirot | jroll: but it doesn't pass my environment variable's value to busy box, assume if set WAIT_TIME_FOR_VFLOPPY_MOUNT=30 in my bash | 01:30 |
| *** mtanino has quit IRC | 01:30 | |
| naohirot | jroll: the value 30 is set my bash to execute disk-image-create, but not shell in the busy box. | 01:31 |
| jroll | naohirot: source it from elements/ramdisk-base/extra-data.d/scripts/d/init-func | 01:31 |
| jroll | hm | 01:31 |
| jroll | I think I see | 01:31 |
| jroll | idk how DIB works enough to answer this question, though | 01:31 |
| naohirot | jroll: we use virtual floppy to pass parameter to busy box environment. | 01:33 |
| naohirot | jroll: If there is a way to pass parameter via some other way, basically we don't have to use virtual floppy | 01:33 |
| naohirot | jroll: there is a paradox | 01:33 |
| jroll | naohirot: let me be clear, I have no idea how DIB works, at all. I don't think I can answer your question intelligently | 01:34 |
| naohirot | jroll: thank you for your help, I'll ask SpamapS via gerrit :) | 01:36 |
| jroll | naohirot: excellent, sorry I couldn't help more :) | 01:36 |
| *** ijw has joined #openstack-ironic | 01:36 | |
| naohirot | jroll: I really appreciate your help :) | 01:37 |
| jroll | :) | 01:37 |
| *** Marga_ has quit IRC | 01:37 | |
| *** ijw has quit IRC | 01:38 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Implement execute clean steps https://review.openstack.org/155561 | 01:56 |
| *** wshao has quit IRC | 02:08 | |
| *** wshao has joined #openstack-ironic | 02:08 | |
| *** wshao has quit IRC | 02:08 | |
| *** wshao has joined #openstack-ironic | 02:09 | |
| *** Marga_ has joined #openstack-ironic | 02:09 | |
| *** Marga_ has quit IRC | 02:09 | |
| *** Marga_ has joined #openstack-ironic | 02:10 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Implement execute clean steps https://review.openstack.org/155561 | 02:10 |
| *** enikanorov_ has joined #openstack-ironic | 02:16 | |
| *** penick has joined #openstack-ironic | 02:16 | |
| *** naohirot_ has joined #openstack-ironic | 02:19 | |
| *** Shalmazar has joined #openstack-ironic | 02:20 | |
| *** jgrimmn has joined #openstack-ironic | 02:20 | |
| *** cinerama` has joined #openstack-ironic | 02:20 | |
| *** rwsu_ has joined #openstack-ironic | 02:20 | |
| *** Madasi has quit IRC | 02:21 | |
| *** zz_jgrimm has quit IRC | 02:21 | |
| *** cinerama has quit IRC | 02:21 | |
| *** naohirot has quit IRC | 02:21 | |
| *** pcrews has quit IRC | 02:21 | |
| *** rwsu has quit IRC | 02:21 | |
| *** enikanorov__ has quit IRC | 02:21 | |
| *** penick has quit IRC | 02:21 | |
| *** penick has joined #openstack-ironic | 02:21 | |
| *** jgrimmn has quit IRC | 02:21 | |
| *** jgrimmn has joined #openstack-ironic | 02:21 | |
| *** Shalmazar is now known as Madasi | 02:21 | |
| *** coolsvap_ is now known as coolsvap | 02:22 | |
| *** ChuckC has joined #openstack-ironic | 02:26 | |
| *** pcrews has joined #openstack-ironic | 02:27 | |
| *** wshao has quit IRC | 02:34 | |
| *** rwsu_ is now known as rwsu-afk | 02:35 | |
| *** wshao has joined #openstack-ironic | 02:37 | |
| *** chenglch has joined #openstack-ironic | 02:39 | |
| *** coolsvap is now known as coolsvap_ | 02:41 | |
| *** cinerama` is now known as cinerama | 02:42 | |
| Haomeng|2 | naohirot_: got your message, dont worry:) | 02:49 |
| Haomeng|2 | Haomeng|2: welcome:) | 02:49 |
| naohirot_ | Haomeng|2: Hi okay, thanks! | 02:50 |
| Haomeng|2 | naohirot_: yw:) | 02:50 |
| naohirot_ | Haomeng|2: :) | 02:50 |
| Haomeng|2 | naohirot_: did you run with IPA and swift? | 02:50 |
| mrda | Hi Haomeng|2 and naohirot_ | 02:50 |
| Haomeng|2 | mrda: morning:) | 02:50 |
| naohirot_ | mrda: good morning :) | 02:51 |
| *** achanda has quit IRC | 02:55 | |
| *** yog_ has joined #openstack-ironic | 02:56 | |
| *** ChuckC has quit IRC | 03:01 | |
| openstackgerrit | Merged openstack/ironic: Fix nits for supporting non-glance images https://review.openstack.org/160732 | 03:04 |
| *** ChuckC has joined #openstack-ironic | 03:06 | |
| openstackgerrit | Michael Davies proposed openstack/ironic-specs: API Microversions https://review.openstack.org/161110 | 03:22 |
| *** rloo has quit IRC | 03:23 | |
| openstackgerrit | Michael Davies proposed openstack/ironic-specs: API Microversions https://review.openstack.org/161110 | 03:24 |
| *** wshao has quit IRC | 03:27 | |
| mrda | devananda: I've just created a bp for api microversions, and targeted it to K-3 (https://blueprints.launchpad.net/ironic/+spec/api-microversions) | 03:27 |
| *** achanda has joined #openstack-ironic | 03:29 | |
| *** spandhe has quit IRC | 03:30 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Ilo drivers sets capabilities:boot_mode in node https://review.openstack.org/155731 | 03:37 |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-python-agent: Add cleaning/zapping support to IPA https://review.openstack.org/161066 | 03:43 |
| *** penick has quit IRC | 03:47 | |
| *** yog_ has quit IRC | 03:55 | |
| *** yog_ has joined #openstack-ironic | 04:01 | |
| *** rameshg87 has joined #openstack-ironic | 04:12 | |
| rameshg87 | good morning ironic | 04:14 |
| mrda | hi rameshg87 | 04:20 |
| rameshg87 | hello mrda | 04:21 |
| naohirot_ | rameshg87: good morning, agent deploy worked, thanks! | 04:21 |
| rameshg87 | naohirot_, great :) | 04:21 |
| rameshg87 | naohirot_, was it proxy issue ? | 04:22 |
| *** achanda has quit IRC | 04:22 | |
| naohirot_ | rameshg87: yes, and if I didn't know about vm option, I would have encountered another problem. but I didn't. | 04:23 |
| rameshg87 | naohirot_, i used to face the same in my workplace as well | 04:23 |
| rameshg87 | naohirot_, are you using older ironic. i thought we fixed all those recently | 04:23 |
| rameshg87 | naohirot_, JoshNang made changes to capture connection errors in agent_client.py and they shouldn't be throwing json decode errors now | 04:24 |
| naohirot_ | rameshg87: Yes, i used a little bit old one. but I rebased yesterday. | 04:24 |
| rameshg87 | naohirot_, okay. i think it should throw a meaningful error now for such connection errors due to proxy | 04:25 |
| naohirot_ | rameshg87: I see, great. | 04:25 |
| rameshg87 | naohirot_, it's very hard to debug otherwise | 04:25 |
| naohirot_ | rameshg87: Yes, does IPA coreos have an option to get shell like DIB's type 't'? | 04:26 |
| rameshg87 | naohirot_, you can ssh to ipa ramdisk itself | 04:27 |
| naohirot_ | rameshg87: really, id/pw? | 04:28 |
| rameshg87 | naohirot_, https://github.com/openstack/ironic-python-agent/blame/master/imagebuild/coreos/README.rst#L85-L88 | 04:28 |
| rameshg87 | naohirot_, you have to import ssh keys into it | 04:28 |
| naohirot_ | rameshg87: via keystone and meta service? | 04:29 |
| rameshg87 | naohirot_, no just put the ssh keys in oem directory before building ramdisk | 04:29 |
| naohirot_ | rameshg87: I see | 04:29 |
| * naohirot_ let me read the url | 04:30 | |
| naohirot_ | rameshg87: Okay, I'll try it later, thanks! | 04:31 |
| *** Haomeng has joined #openstack-ironic | 04:33 | |
| *** Haomeng|2 has quit IRC | 04:34 | |
| naohirot_ | rameshg87: I'd like to hear your idea about generalizing the vfloppy parameter https://review.openstack.org/#/c/160813/ | 04:35 |
| naohirot_ | rameshg87: if you had some time. | 04:36 |
| rameshg87 | naohirot_, sure .. | 04:36 |
| rameshg87 | naohirot_, same thing as dib applies here as well, right ? | 04:37 |
| rameshg87 | naohirot_, some environment variable or a conf file for ipa using which we can mention the device filename at the time of building | 04:37 |
| rameshg87 | naohirot_, and defaults to the both of the devices that we know now | 04:37 |
| naohirot_ | rameshg87: Yes, but IPA seems quit complicated comparing with DIB. | 04:39 |
| rameshg87 | naohirot_, i will explore the same if we have some option and will comment on the review today itself | 04:39 |
| naohirot_ | rameshg87: basically does coreos have similar role with busybox in DIB? | 04:40 |
| naohirot_ | rameshg87: sorry, i wanted to say | 04:41 |
| naohirot_ | rameshg87: basically does docker have similar role with busybox in DIB? | 04:41 |
| *** chenglch has quit IRC | 04:42 | |
| rameshg87 | naohirot_, busybox as i know is a single-command-wrapper for all the basic commands | 04:42 |
| rameshg87 | naohirot_, but docker facilitates running something else as a container | 04:42 |
| rameshg87 | naohirot_, did you see this ? - https://github.com/openstack/ironic-python-agent/blob/master/etc/ironic_python_agent/ironic_python_agent.conf.sample | 04:43 |
| rameshg87 | naohirot_, i am wonder we could put it as a conf option, but i don't how we can get some custom conf options into the ramdisk at build time | 04:43 |
| rameshg87 | naohirot_, but i am sure rackspace folks would have left some provision for that :) | 04:43 |
| naohirot_ | rameshg87: Yeah, I'll dig into IPA from the sample file, and ask them. | 04:45 |
| rameshg87 | naohirot_, okay | 04:45 |
| naohirot_ | rameshg87: thank yo for giving me the hint. | 04:45 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/161389 | 04:50 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/161575 | 04:50 |
| *** wshao has joined #openstack-ironic | 04:51 | |
| *** wshao_ has joined #openstack-ironic | 04:55 | |
| *** wshao has quit IRC | 04:56 | |
| *** harlowja is now known as harlowja_away | 04:57 | |
| *** wshao has joined #openstack-ironic | 04:58 | |
| *** wshao_ has quit IRC | 04:59 | |
| *** chenglch has joined #openstack-ironic | 04:59 | |
| *** spandhe has joined #openstack-ironic | 05:07 | |
| *** stendulker has joined #openstack-ironic | 05:13 | |
| *** spandhe_ has joined #openstack-ironic | 05:17 | |
| *** spandhe has quit IRC | 05:18 | |
| *** spandhe_ is now known as spandhe | 05:18 | |
| *** wshao has quit IRC | 05:21 | |
| *** pradipta has joined #openstack-ironic | 05:21 | |
| *** wshao has joined #openstack-ironic | 05:22 | |
| *** achanda has joined #openstack-ironic | 05:23 | |
| *** achanda has quit IRC | 05:23 | |
| *** achanda has joined #openstack-ironic | 05:23 | |
| *** wshao has quit IRC | 05:26 | |
| *** ParsectiX has quit IRC | 05:26 | |
| *** ParsectiX has joined #openstack-ironic | 05:27 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: create and get should return physical disks info https://review.openstack.org/160307 | 05:34 |
| *** sandhya has joined #openstack-ironic | 05:35 | |
| *** chenglch has quit IRC | 05:49 | |
| *** chenglch has joined #openstack-ironic | 05:54 | |
| *** spandhe has quit IRC | 06:03 | |
| *** stendulker_ has joined #openstack-ironic | 06:04 | |
| *** stendulker has quit IRC | 06:06 | |
| *** lazy_prince has quit IRC | 06:08 | |
| *** ParsectiX has quit IRC | 06:08 | |
| *** killer_prince has joined #openstack-ironic | 06:08 | |
| *** killer_prince is now known as lazy_prince | 06:08 | |
| *** stendulker has joined #openstack-ironic | 06:19 | |
| *** spandhe has joined #openstack-ironic | 06:22 | |
| *** stendulker_ has quit IRC | 06:23 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: Add support for sharing physical disks https://review.openstack.org/160385 | 06:30 |
| *** gridinv has joined #openstack-ironic | 06:46 | |
| *** sabah has joined #openstack-ironic | 06:49 | |
| *** Krast has joined #openstack-ironic | 06:51 | |
| *** gridinv has quit IRC | 06:55 | |
| *** Marga_ has quit IRC | 07:20 | |
| *** spandhe has quit IRC | 07:32 | |
| *** openstackgerrit has quit IRC | 07:36 | |
| *** openstackgerrit has joined #openstack-ironic | 07:36 | |
| *** ifarkas has joined #openstack-ironic | 07:37 | |
| *** killer_prince has joined #openstack-ironic | 07:37 | |
| *** ifarkas_ has joined #openstack-ironic | 07:37 | |
| *** lazy_prince has quit IRC | 07:38 | |
| *** killer_prince is now known as lazy_prince | 07:38 | |
| *** ifarkas_ has quit IRC | 07:41 | |
| *** yog_ has quit IRC | 07:55 | |
| openstackgerrit | Merged openstack/ironic: Add a note to dev-quickstart https://review.openstack.org/145690 | 07:56 |
| *** athomas has joined #openstack-ironic | 08:06 | |
| *** dlpartain has joined #openstack-ironic | 08:10 | |
| *** achanda has quit IRC | 08:16 | |
| openstackgerrit | Merged openstack/ironic: Fix nits in cleaning https://review.openstack.org/160599 | 08:24 |
| *** sabah has quit IRC | 08:25 | |
| *** subah has joined #openstack-ironic | 08:25 | |
| *** pradipta has quit IRC | 08:29 | |
| *** takadayuiko has joined #openstack-ironic | 08:31 | |
| *** Krast has quit IRC | 08:33 | |
| *** achanda has joined #openstack-ironic | 08:33 | |
| *** dlpartain has left #openstack-ironic | 08:38 | |
| *** pradipta has joined #openstack-ironic | 08:42 | |
| *** jistr has joined #openstack-ironic | 08:49 | |
| *** jcoufal has joined #openstack-ironic | 08:54 | |
| *** ndipanov has joined #openstack-ironic | 08:57 | |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: Add functional tests for create https://review.openstack.org/158122 | 08:58 |
| *** dtantsur|afk is now known as dtantsur | 08:58 | |
| dtantsur | Morning | 08:58 |
| rameshg87 | dtantsur, o/ | 09:01 |
| *** erwan_taf has joined #openstack-ironic | 09:05 | |
| *** achanda has quit IRC | 09:12 | |
| *** lucasagomes has joined #openstack-ironic | 09:21 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: Add support for nested RAID levels https://review.openstack.org/158994 | 09:24 |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: Add support for nested RAID levels https://review.openstack.org/158994 | 09:28 |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: Add support for nested RAID levels https://review.openstack.org/158994 | 09:28 |
| *** mgoddard1 has joined #openstack-ironic | 09:28 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: Handle all storage units https://review.openstack.org/159001 | 09:28 |
| *** mgoddard2 has joined #openstack-ironic | 09:30 | |
| *** mgoddard has quit IRC | 09:30 | |
| *** mgoddard1 has quit IRC | 09:32 | |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: Handle all storage units https://review.openstack.org/159001 | 09:33 |
| *** athomas has quit IRC | 09:33 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: create and get should return physical disks info https://review.openstack.org/160307 | 09:34 |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: Add support for sharing physical disks https://review.openstack.org/160385 | 09:34 |
| *** romcheg has joined #openstack-ironic | 09:35 | |
| *** mgoddard has joined #openstack-ironic | 09:37 | |
| *** romcheg has quit IRC | 09:37 | |
| *** romcheg1 has joined #openstack-ironic | 09:37 | |
| *** mgoddard2 has quit IRC | 09:39 | |
| *** stendulker has quit IRC | 09:44 | |
| *** igordcard has joined #openstack-ironic | 09:49 | |
| *** vdrok_afk is now known as vdrok | 09:52 | |
| vdrok | morning ironic | 09:52 |
| Kinnison | Morning | 09:56 |
| *** naohirot_ has quit IRC | 10:00 | |
| *** mgoddard1 has joined #openstack-ironic | 10:01 | |
| *** MattMan has joined #openstack-ironic | 10:02 | |
| *** mgoddard has quit IRC | 10:03 | |
| rameshg87 | vdrok, Kinnison, o/ | 10:04 |
| *** yuanying has quit IRC | 10:05 | |
| Kinnison | hi rameshg87 | 10:05 |
| *** wendar has quit IRC | 10:06 | |
| vdrok | morning rameshg87, Kinnison | 10:08 |
| *** wendar has joined #openstack-ironic | 10:16 | |
| *** pas-ha has joined #openstack-ironic | 10:16 | |
| *** agordeev has joined #openstack-ironic | 10:17 | |
| *** subah has quit IRC | 10:17 | |
| *** jlk has joined #openstack-ironic | 10:18 | |
| *** jlk has quit IRC | 10:18 | |
| *** jlk has joined #openstack-ironic | 10:18 | |
| *** sabah has joined #openstack-ironic | 10:19 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add AMT-PXE-Driver Power&Management&Vendor Interface https://review.openstack.org/152858 | 10:21 |
| agordeev | good morning, ironic :) | 10:23 |
| Kinnison | Morning agordeev | 10:23 |
| agordeev | Haomeng: hello, i replied to you message. thanks :) | 10:25 |
| agordeev | Kinnison: hello! | 10:25 |
| openstackgerrit | Merged openstack/ironic: Add tftp mapfile configuration in install-guide https://review.openstack.org/161074 | 10:29 |
| *** athomas has joined #openstack-ironic | 10:31 | |
| *** takadayuiko has quit IRC | 10:32 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add driver interface for RAID configuration https://review.openstack.org/155230 | 10:34 |
| Haomeng | agordeev: thank you:) | 10:35 |
| *** alexpilotti has joined #openstack-ironic | 10:35 | |
| Haomeng | Haomeng: got it, "-v -v -v -v -v" should be *very* verbose, but I try to with single '-v', it run with error message | 10:37 |
| *** chenglch has quit IRC | 10:38 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add driver interface for RAID configuration https://review.openstack.org/155230 | 10:38 |
| agordeev | Haomeng: did it work despite of that error message? it may be non-critical error | 10:38 |
| Haomeng | agordeev: ok, got, thanks for the confirmation:) | 10:39 |
| Haomeng | agordeev: :) | 10:39 |
| Haomeng | agordeev: nice day:) | 10:39 |
| *** pelix has joined #openstack-ironic | 10:42 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add server-supported min and max API version to HTTPNotAcceptable(406) https://review.openstack.org/160758 | 10:45 |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: HPSSA: Add support for sharing physical disks https://review.openstack.org/160385 | 10:49 |
| *** sandhya has quit IRC | 10:52 | |
| *** lazy_prince is now known as killer_prince | 10:57 | |
| *** mgoddard has joined #openstack-ironic | 10:57 | |
| *** sabah has quit IRC | 10:58 | |
| *** mgoddard1 has quit IRC | 11:00 | |
| *** Haomeng|2 has joined #openstack-ironic | 11:15 | |
| *** Haomeng has quit IRC | 11:16 | |
| *** pradipta has quit IRC | 11:16 | |
| *** rameshg87 has quit IRC | 11:32 | |
| *** erwan_taf has quit IRC | 11:33 | |
| *** erwan_taf has joined #openstack-ironic | 11:33 | |
| openstackgerrit | Merged openstack/ironic: Updated from global requirements https://review.openstack.org/161389 | 11:36 |
| *** yog_ has joined #openstack-ironic | 11:44 | |
| *** athomas has quit IRC | 12:01 | |
| *** EmilienM|afk is now known as EmilienM | 12:04 | |
| *** dtantsur is now known as dtantsur|brb | 12:13 | |
| *** athomas has joined #openstack-ironic | 12:16 | |
| *** Marga_ has joined #openstack-ironic | 12:33 | |
| *** jcoufal_ has joined #openstack-ironic | 12:34 | |
| *** jcoufal has quit IRC | 12:37 | |
| *** coolsvap_ has quit IRC | 12:38 | |
| *** mgoddard has quit IRC | 12:40 | |
| *** Marga_ has quit IRC | 12:40 | |
| *** coolsvap_ has joined #openstack-ironic | 12:46 | |
| *** dmellado has quit IRC | 12:52 | |
| *** dprince has joined #openstack-ironic | 13:01 | |
| *** dmellado has joined #openstack-ironic | 13:02 | |
| *** athomas has quit IRC | 13:06 | |
| *** rameshg87 has joined #openstack-ironic | 13:07 | |
| *** athomas has joined #openstack-ironic | 13:13 | |
| jroll | morning all :) | 13:27 |
| *** lucasagomes is now known as lucas-hungry | 13:27 | |
| *** Marga_ has joined #openstack-ironic | 13:28 | |
| jroll | rameshg87: what we've done for build-time configs is just modify oem/cloud-config.yml and pass as command line args | 13:29 |
| jroll | but I'm +1 to working on actually reading the config file (that might even work today, haven't tried it) | 13:30 |
| jroll | and actually, we only pass --debug this way, the rest we do via pxe config | 13:30 |
| rameshg87 | jroll, yeah | 13:31 |
| rameshg87 | jroll, i was suggesting to naohirot to put it as a config option | 13:31 |
| rameshg87 | jroll, because the virtual media device file will vary by vendor | 13:31 |
| rameshg87 | jroll, the model string varies hence | 13:31 |
| rameshg87 | jroll, so better use a conf file if it's there | 13:32 |
| *** dtantsur|brb is now known as dtantsur | 13:32 | |
| dtantsur | jroll, morning! | 13:32 |
| jroll | rameshg87: yeah, agree, I've just never tried the config file myself :) | 13:32 |
| jroll | rameshg87: and good morning to you and dtantsur :) | 13:32 |
| rameshg87 | jroll, okay .. and good morning btw :) | 13:32 |
| jroll | rameshg87: I think you would have to change cloud-config.yml to point at the config file, at lease | 13:33 |
| * rameshg87 looks cloud-config.yml | 13:33 | |
| rameshg87 | jroll, okay, will let naohirot know, he is not around now i think | 13:34 |
| jroll | right, he is not | 13:34 |
| jroll | not enough time zone overlap, we should us the mailing list more | 13:34 |
| rameshg87 | yeah and irc bouncers too :) | 13:36 |
| jroll | lol, that too | 13:36 |
| * jroll considers making a quick bouncer-as-a-service | 13:36 | |
| rameshg87 | jroll, +100 | 13:36 |
| rameshg87 | jroll, that will be useful for most openstack developers :) | 13:36 |
| jroll | https://github.com/jimeh/docker-znc | 13:37 |
| *** yog__ has joined #openstack-ironic | 13:38 | |
| jroll | this was a good read, too: http://tech.paulcz.net/i-built-a-saas/ | 13:38 |
| *** dmellado_ has joined #openstack-ironic | 13:40 | |
| *** dmellado has quit IRC | 13:40 | |
| *** dmellado_ is now known as dmellado | 13:40 | |
| rameshg87 | jroll, but why would we want to run znc on a docker container | 13:41 |
| rameshg87 | jroll, i thought it was a harmless piece of software :) | 13:41 |
| * rameshg87 reads | 13:42 | |
| *** yog_ has quit IRC | 13:42 | |
| jroll | rameshg87: makes it easy to spin up a ton of them | 13:43 |
| jroll | rather than making a kludge for multiple instances | 13:43 |
| rameshg87 | jroll, ah okay .. | 13:43 |
| * rameshg87 understands that's the purpose of docker :) | 13:43 | |
| *** Marga_ has quit IRC | 13:47 | |
| *** mgoddard has joined #openstack-ironic | 13:49 | |
| dtantsur | ZNC rules btw :) | 13:49 |
| *** rloo has joined #openstack-ironic | 13:51 | |
| *** Marga_ has joined #openstack-ironic | 13:57 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Fix take over for agent driver https://review.openstack.org/161717 | 13:59 |
| rameshg87 | jroll, ^^^ | 14:10 |
| rameshg87 | jroll, do we allow take over for nodes that are in wait-call-back ? i think we don't | 14:10 |
| jroll | rameshg87: yeah, I saw that. that was my next question :) | 14:10 |
| rameshg87 | jroll, i mean for nodes with agent* drivers | 14:10 |
| jroll | actually it's more if... | 14:11 |
| rameshg87 | if <what> ? | 14:11 |
| jroll | the dhcp config gets to neutron and the conductor dies before the node boots | 14:11 |
| jroll | I guess that's in wait-callback | 14:11 |
| rameshg87 | yeah | 14:11 |
| jroll | yeah I have no idea | 14:11 |
| rameshg87 | but conductor seems to allow take_over for only active nodes | 14:11 |
| jroll | oh, awesome | 14:11 |
| rameshg87 | looking at code from conductor | 14:11 |
| rameshg87 | https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L977-L980 | 14:11 |
| jroll | so we should just delete that code | 14:12 |
| vdrok | jroll, rameshg87, ok, will change, thanks :) | 14:12 |
| vdrok | and morning jroll | 14:12 |
| rameshg87 | yeah | 14:12 |
| jroll | morning vdrok :) | 14:12 |
| rameshg87 | and no point in updating just bootfile-name :) | 14:12 |
| jroll | vdrok: now that I'm thinking about it... | 14:12 |
| jroll | I wonder if it's fine to fix this and leave it, as we'll have pxe boot for agent deployed instances eventually | 14:13 |
| jroll | maybe should just add it back then, though | 14:13 |
| rameshg87 | yeah, that's wip right | 14:13 |
| jroll | vdrok: sorry, just thinking out loud, let's delete the code :) | 14:13 |
| jroll | supposedly | 14:13 |
| jroll | but I haven't seen any work on it | 14:13 |
| rameshg87 | i think there is a patch on it | 14:14 |
| jroll | http://specs.openstack.org/openstack/ironic-specs/specs/kilo/partition-image-support-for-agent-driver.html | 14:14 |
| jroll | oh? | 14:14 |
| jroll | https://blueprints.launchpad.net/ironic/+spec/partition-image-support-for-agent-driver | 14:14 |
| jroll | bumped to L for lack of progress | 14:14 |
| vdrok | jroll, yup, I see, if it's the case maybe indeed better to delete it | 14:14 |
| jroll | yeah +1 | 14:15 |
| rameshg87 | and this as well: https://review.openstack.org/#/c/160224/ | 14:15 |
| jroll | yeah :( | 14:15 |
| * jroll wants to make a todo list before the summit | 14:15 | |
| *** kkoski has joined #openstack-ironic | 14:16 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Fix take over for agent driver https://review.openstack.org/161717 | 14:18 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: None is a valid default value for base_url and auth_token in the client https://review.openstack.org/161728 | 14:19 |
| *** rameshg87 is now known as rameshg87-dinner | 14:24 | |
| *** Marga_ has quit IRC | 14:25 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Fix take over for agent driver https://review.openstack.org/161717 | 14:25 |
| *** rameshg87-dinner has quit IRC | 14:30 | |
| *** BadCub has joined #openstack-ironic | 14:36 | |
| BadCub | Morning Iroinc | 14:36 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Add python-openstackclient plugin for ironic-discoverd https://review.openstack.org/157448 | 14:37 |
| openstackgerrit | Sirushti Murugesan proposed openstack/ironic: Adds support for deploying whole disk images https://review.openstack.org/150142 | 14:38 |
| *** jgrimmn has quit IRC | 14:38 | |
| *** maurosr has quit IRC | 14:38 | |
| *** BadCub has quit IRC | 14:39 | |
| *** absubram has joined #openstack-ironic | 14:39 | |
| *** jgrimm has joined #openstack-ironic | 14:40 | |
| *** maurosr has joined #openstack-ironic | 14:41 | |
| *** BadCub has joined #openstack-ironic | 14:41 | |
| *** fuzzyrubble has joined #openstack-ironic | 14:42 | |
| *** lucas-hungry is now known as lucasagomes | 14:43 | |
| *** jgrimm has quit IRC | 14:44 | |
| *** maurosr has quit IRC | 14:44 | |
| lucasagomes | BadCub, jroll good morning | 14:44 |
| jroll | morning lucasagomes :) | 14:45 |
| jroll | and BadCub! | 14:45 |
| *** jgrimm has joined #openstack-ironic | 14:45 | |
| *** maurosr has joined #openstack-ironic | 14:46 | |
| *** r-daneel has joined #openstack-ironic | 14:46 | |
| *** chuckC_ has joined #openstack-ironic | 14:46 | |
| *** chuckC_ has quit IRC | 14:48 | |
| *** kkoski has quit IRC | 14:48 | |
| *** kkoski has joined #openstack-ironic | 14:49 | |
| *** yog__ has quit IRC | 14:55 | |
| *** krtaylor has quit IRC | 14:59 | |
| lucasagomes | jroll, yo, you've created the gate job for IPA right? | 15:02 |
| lucasagomes | I've never done it before, you know which project I have to change to create a new job? | 15:03 |
| jroll | lucasagomes: openstack-infra/project-config | 15:03 |
| lucasagomes | a-ha | 15:04 |
| jroll | it was a different project at the time or else I'd link you the commit | 15:04 |
| lucasagomes | will take a look | 15:04 |
| lucasagomes | jroll, ta much :) | 15:04 |
| NobodyCam | good morning Ironicers :) | 15:04 |
| jroll | np lucasagomes :) | 15:04 |
| jroll | morning NobodyCam :) | 15:04 |
| NobodyCam | morning jroll and lucasagomes :) | 15:05 |
| lucasagomes | NobodyCam, hey buddy, morning | 15:05 |
| NobodyCam | morning . | 15:06 |
| NobodyCam | :) | 15:06 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Add python-openstackclient plugin for ironic-discoverd https://review.openstack.org/157448 | 15:06 |
| dtantsur | NobodyCam, morning | 15:10 |
| BadCub | Morning lucasagomes jroll and NobodyCam :-p | 15:10 |
| jroll | morning BadCub :) | 15:10 |
| dtantsur | BadCub, morning you too :) | 15:10 |
| BadCub | morning there dtantsur :-) | 15:11 |
| NobodyCam | mornign dtantsur BadCub | 15:12 |
| *** mtanino has joined #openstack-ironic | 15:13 | |
| *** romcheg1 has quit IRC | 15:18 | |
| *** ChuckC has quit IRC | 15:19 | |
| *** romcheg has joined #openstack-ironic | 15:23 | |
| *** rameshg87 has joined #openstack-ironic | 15:25 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Check UUID correctness for Glance images https://review.openstack.org/151951 | 15:25 |
| *** krtaylor has joined #openstack-ironic | 15:26 | |
| *** spandhe has joined #openstack-ironic | 15:27 | |
| *** anderbubble has joined #openstack-ironic | 15:29 | |
| *** hj-hp has joined #openstack-ironic | 15:31 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add module for in-band inspection using ironic-discoverd https://review.openstack.org/156562 | 15:32 |
| dtantsur | NobodyCam, rameshg87 ^^^ I hope I fixed everything you spotted :) | 15:33 |
| NobodyCam | dtantsur: :) will look in a few min :) | 15:33 |
| rameshg87 | dtantsur, me tooo | 15:33 |
| dtantsur | thx | 15:33 |
| rameshg87 | NobodyCam, have a few mins to talk about https://review.openstack.org/#/c/159855/ | 15:35 |
| NobodyCam | let me bring that one up | 15:38 |
| NobodyCam | rameshg87: ya I do see your point about taking the space from root not being a good thing | 15:39 |
| rameshg87 | NobodyCam, yeah, i just thought we might end by proving a few hundereds of Mbs less root partition | 15:40 |
| rameshg87 | NobodyCam, and if it's growing to grow towards end of disk, then taking from root partition right now would be same (beacuse root partition will be expanded anyway) | 15:40 |
| rameshg87 | NobodyCam, another point lucasagomes brought was we might end up in a failure if user request root exactly == size of disk | 15:41 |
| rameshg87 | NobodyCam, but we have same issue with config drive as well, so i think should be fixed later | 15:41 |
| NobodyCam | rameshg87: the reason I left that comment was I have worked with several folk to use the entire phyical disk and didn't see where they would account for the uefi parition | 15:42 |
| rameshg87 | NobodyCam, okay | 15:43 |
| NobodyCam | rameshg87: are you planing any out updates to goalong with that? | 15:43 |
| rameshg87 | NobodyCam, so is there way to request that from nova ? | 15:43 |
| NobodyCam | * go along* | 15:43 |
| rameshg87 | NobodyCam, i mean use up the entire physical disk ? | 15:43 |
| rameshg87 | NobodyCam, i assume they can give only root_mb, swap_mb, etc | 15:43 |
| NobodyCam | rameshg87: not that I know | 15:43 |
| rameshg87 | NobodyCam, so how would they request for entire physical disk ? (because i assume they can request for only particular size of root_mb) | 15:44 |
| rameshg87 | NobodyCam, i thought it will good to revisit in L | 15:44 |
| NobodyCam | rameshg87: ack :) | 15:46 |
| NobodyCam | I will revisit in a bit... | 15:46 |
| NobodyCam | thank under standing helps ... thank you | 15:46 |
| rameshg87 | NobodyCam, okay, let me know your thoughts in the review | 15:47 |
| NobodyCam | s/thank under/that under/ | 15:47 |
| rameshg87 | lucasagomes, jroll, please have a look at https://review.openstack.org/#/c/159855/ when you get some time | 15:47 |
| lucasagomes | +1 | 15:48 |
| lucasagomes | ack | 15:48 |
| NobodyCam | rameshg87: lucasagomes: are there ufei enabling docs | 15:48 |
| rameshg87 | lucasagomes, yes | 15:49 |
| rameshg87 | i mean NobodyCam, yes :) | 15:49 |
| lucasagomes | :) ^ | 15:49 |
| *** stendulker has joined #openstack-ironic | 15:49 | |
| NobodyCam | heheh | 15:49 |
| rameshg87 | NobodyCam, http://docs.openstack.org/developer/ironic/deploy/install-guide.html#pxe-uefi-setup | 15:49 |
| rameshg87 | NobodyCam, since ilo* drivers boot from virtualmedia, they virtual media itself is dualboot capable | 15:50 |
| *** krtaylor has quit IRC | 15:50 | |
| NobodyCam | :) | 15:50 |
| rameshg87 | JoshNang, you around ? | 15:54 |
| *** kozhukalov has joined #openstack-ironic | 15:57 | |
| *** hj-hp has quit IRC | 15:58 | |
| *** fuzzyrubble has quit IRC | 15:59 | |
| *** fuzzyrubble has joined #openstack-ironic | 15:59 | |
| *** spandhe has quit IRC | 16:01 | |
| *** krtaylor has joined #openstack-ironic | 16:02 | |
| *** PaulCzar has joined #openstack-ironic | 16:05 | |
| *** fuzzyrubble has quit IRC | 16:05 | |
| *** fuzzyrubble has joined #openstack-ironic | 16:06 | |
| *** dprince has quit IRC | 16:07 | |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: create and get should return physical disks info https://review.openstack.org/160307 | 16:08 |
| *** fuzzyrubble has quit IRC | 16:08 | |
| *** fuzzyrubble has joined #openstack-ironic | 16:09 | |
| jlvillal | Good morning Ironic | 16:09 |
| kkoski | Good Morning | 16:09 |
| victor_lowther | Good morning Ironic | 16:09 |
| NobodyCam | mornign jlvillal kkoski and victor_lowther | 16:09 |
| *** fuzzyrubble has quit IRC | 16:12 | |
| *** hj-hp has joined #openstack-ironic | 16:15 | |
| *** ChuckC has joined #openstack-ironic | 16:17 | |
| *** ChuckC has quit IRC | 16:18 | |
| *** ChuckC has joined #openstack-ironic | 16:18 | |
| openstackgerrit | Merged stackforge/proliantutils: HPSSA: Add support for sharing physical disks https://review.openstack.org/160385 | 16:20 |
| *** stendulker has quit IRC | 16:23 | |
| *** romcheg has quit IRC | 16:25 | |
| *** Marga_ has joined #openstack-ironic | 16:27 | |
| *** chuckC_ has joined #openstack-ironic | 16:29 | |
| *** chuckC_ has quit IRC | 16:29 | |
| *** gridinv has joined #openstack-ironic | 16:36 | |
| devananda | Haomeng|2: around? | 16:37 |
| JoshNang | rameshg87: o/ | 16:38 |
| rameshg87 | JoshNang, o/ | 16:39 |
| rameshg87 | JoshNang, have some time to discuss regarding inband zapping | 16:39 |
| JoshNang | rameshg87: sure! | 16:39 |
| jlvillal | rameshg87: Let me know if I'm being dumb on wanting to delete that mkfs() function in the UEFI boot mode patch | 16:39 |
| *** spandhe has joined #openstack-ironic | 16:40 | |
| rameshg87 | jlvillal, will check the comment in a few mins .. | 16:40 |
| rameshg87 | jlvillal, thanks for the review :) | 16:41 |
| jlvillal | rameshg87: You're welcome. Happy to help. | 16:41 |
| *** gridinv has quit IRC | 16:41 | |
| *** rameshg87_ has joined #openstack-ironic | 16:42 | |
| rameshg87_ | JoshNang, sorry, got disconnected | 16:42 |
| JoshNang | rameshg87_: no worries | 16:42 |
| rameshg87_ | JoshNang, yeah i would like to use node.raid for agent inband raid configuration | 16:42 |
| jlvillal | NobodyCam: On the ipmitool/tempfile thing. How do we know that error is because of a tempfile? | 16:42 |
| *** anderbubble has quit IRC | 16:42 | |
| jlvillal | NobodyCam: Never mind. Finally saw the error! | 16:43 |
| rameshg87_ | JoshNang, basically i can use the same get_clean_steps and execute_clean_step that you will be using for agent.deploy | 16:43 |
| *** rwsu-afk is now known as rwsu | 16:43 | |
| JoshNang | rameshg87_: works for me. i can put it all in a common file then | 16:43 |
| rameshg87_ | JoshNang, but get_clean_steps is really a static thing - it will have only create_configuration and delete_configuration | 16:43 |
| rameshg87_ | JoshNang, and important thing is it will always have them | 16:43 |
| rameshg87_ | JoshNang, as i understand zapping is an operator invoked thing | 16:43 |
| rameshg87_ | JoshNang, we can check only if the zap step exists on the agent at run-time, is that correct ? | 16:44 |
| JoshNang | rameshg87_: correct | 16:44 |
| rameshg87_ | JoshNang, because ramdisk is not booted on the node always :) | 16:44 |
| *** spandhe has quit IRC | 16:44 | |
| rameshg87_ | (just like you have) | 16:44 |
| rameshg87_ | :) | 16:44 |
| openstackgerrit | Victor Lowther proposed openstack/ironic: Add DRAC BIOS config vendor passthru API https://review.openstack.org/150488 | 16:44 |
| JoshNang | rameshg87_: heh yes. | 16:44 |
| JoshNang | some of the code is going to look a bit...bad due to that. | 16:45 |
| rameshg87_ | JoshNang, okay so i can create an implementation of RaidInterface for agent | 16:45 |
| JoshNang | yup! | 16:45 |
| NobodyCam | jlvillal: just fyi I am going to pushing up an new rev shortly | 16:45 |
| openstackgerrit | Merged openstack/ironic: Fix take over for agent driver https://review.openstack.org/161717 | 16:45 |
| jlvillal | NobodyCam: The puzzling part to me is the error message | 16:45 |
| rameshg87_ | JoshNang, i have create_configuration and delete_configuration as abstract methods here: https://review.openstack.org/#/c/155230/19/ironic/drivers/base.py | 16:45 |
| *** rameshg87 has quit IRC | 16:45 | |
| jlvillal | NobodyCam: "Unexpected error while running command" | 16:45 |
| rameshg87_ | JoshNang, just wondering if i should make it not abstract for agent case | 16:46 |
| rameshg87_ | JoshNang, because agent doesn't need them | 16:46 |
| jlvillal | NobodyCam: Since the except block catches all exceptions | 16:46 |
| rameshg87_ | JoshNang, i mean not abstract because of agent case | 16:46 |
| JoshNang | rameshg87_: hmmmm | 16:46 |
| rameshg87_ | JoshNang, the other way round | 16:46 |
| rameshg87_ | JoshNang, i can have create_configuration and delete_configuration of AgentRaidInterface | 16:47 |
| rameshg87_ | JoshNang, which will send the commands to the agent directly | 16:47 |
| rameshg87_ | JoshNang, ^^ (initially i was having this in mind) | 16:47 |
| rameshg87_ | JoshNang, what's your thought on that ? | 16:47 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Start using in-band inspection in pxe_{ssh,ipmitool,ipminative} https://review.openstack.org/161132 | 16:48 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Write documentation for hardware inspection https://review.openstack.org/161775 | 16:48 |
| JoshNang | rameshg87_: having two interfaces both trying to talk to the agent is going to break things | 16:49 |
| JoshNang | the agent triggers a lot of things on heartbeats. the raid interface wouldn't be able to take advantage of that | 16:49 |
| *** krtaylor has quit IRC | 16:49 | |
| rameshg87_ | JoshNang, okay so do you prefer the first approach ? | 16:50 |
| rameshg87_ | JoshNang, and one more thing. as part of raid configuration i will need to pass in input to the raid.create_configuration method and get the output from it back | 16:50 |
| JoshNang | rameshg87_: that'll already be supported | 16:51 |
| rameshg87_ | JoshNang, i need to call update_raid_info() (in https://review.openstack.org/#/c/155230/19/ironic/common/raid.py) with what raid.create_configuration returns | 16:51 |
| JoshNang | rameshg87_: actually, every clean/zap step gets the whole node and ports object (minus sensitive data), so if its stored on the node object, you'll have it | 16:51 |
| JoshNang | err, every clean/zap step in the agent | 16:52 |
| rameshg87_ | JoshNang, yeah i saw that in your review | 16:52 |
| rameshg87_ | JoshNang, so input part is taken care | 16:52 |
| rameshg87_ | JoshNang, so after the step is complete, i will need to take it's return value | 16:52 |
| *** achanda has joined #openstack-ironic | 16:52 | |
| rameshg87_ | JoshNang, and call update_raid_info() with it | 16:52 |
| JoshNang | and commands already return results, i'll pass that up through zapping. i'll need to change things a bit for that i think | 16:52 |
| JoshNang | ahh | 16:52 |
| rameshg87_ | JoshNang, update_raid_info() will take care of updating the node properties like local_gb, root_device_hint, etc after raid configuration | 16:53 |
| *** romcheg has joined #openstack-ironic | 16:53 | |
| rameshg87_ | JoshNang, can i capture the result of raid.create_configuration in a heartbeat and then do it ? | 16:54 |
| JoshNang | rameshg87_: the heartbeats are all going to the agent driver | 16:54 |
| rameshg87_ | JoshNang, yeah that's fine. even AgentRaidInterface will be part of agent driver | 16:55 |
| * rameshg87_ wonders if i am missing something here :) | 16:55 | |
| rameshg87_ | JoshNang, do you have the next patch in the series for clean task. basically it's all some assumptions with unconnected dots for me now :) | 16:56 |
| JoshNang | ha sorry, agent deploy interface is what i mean. though its actually the agent vendor passthru interface now that i think about it | 16:56 |
| JoshNang | (we called it the agent driver for so long that's stuck in my head) | 16:56 |
| rameshg87_ | :) | 16:57 |
| JoshNang | so, the agent heartbeats to the vendor passthru, the node sees zapping, it would go to the agent deploy interface, because in every other case that's where it would go | 16:58 |
| rameshg87_ | JoshNang, only if the step looks like deploy.<something> right ? | 16:59 |
| JoshNang | hmm | 16:59 |
| rameshg87_ | JoshNang, what if the step name is raid.<something-else> ? can't it be redirected somewhere else ? | 16:59 |
| JoshNang | it could | 17:00 |
| JoshNang | i'm not a fan of that though | 17:00 |
| JoshNang | but i don't know that there's much of another option | 17:00 |
| rameshg87_ | the idea of a common-api and a common approach for raid was that the operator could configure raid in the same way and invoke the raid zap task in the same way | 17:01 |
| jlvillal | NobodyCam: When you have a moment I would like to chat about the tempfile issue, hopefully before you post new patch :) | 17:01 |
| rameshg87_ | for both inband as well as out-of-band, operator would call the step as "raid.create_configuration" and "raid.delete_configuration" | 17:01 |
| *** krtaylor has joined #openstack-ironic | 17:01 | |
| rameshg87_ | so we would need to that be in sync with out-of-band approaches | 17:01 |
| JoshNang | i'm aware, i reviewed the spec | 17:02 |
| rameshg87_ | so are there any hooks for processing after a clean/zap task ? | 17:02 |
| rameshg87_ | i mean do some deploy clean/zap operations need such a hook ? | 17:02 |
| JoshNang | there aren't currently | 17:02 |
| JoshNang | i would hope not | 17:03 |
| rameshg87_ | okay | 17:03 |
| JoshNang | i think what you've proposed should work. though i haven't had coffee yet ;) | 17:03 |
| rameshg87_ | hehe :) | 17:03 |
| rameshg87_ | can we put a small hook on the heartbeat for zapping ? | 17:04 |
| rameshg87_ | may be a workaround for now. to be bettered in L release :) | 17:04 |
| rameshg87_ | i just need to get a lock on the node and call update_raid_info. that's it :) | 17:05 |
| JoshNang | yeah i forsee some refactoring in the L release | 17:05 |
| rameshg87_ | okay | 17:05 |
| *** achanda has quit IRC | 17:05 | |
| JoshNang | yeah, i should be able to inspect the zap step on completed commands | 17:05 |
| rameshg87_ | okay | 17:06 |
| rameshg87_ | is your deploy part ready ? i mean your next patch ? | 17:07 |
| rameshg87_ | JoshNang, with whatever we knew about 1 month back about hardware manager | 17:07 |
| rameshg87_ | JoshNang, we created one here for proliant hardware: https://github.com/stackforge/proliantutils/blob/master/proliantutils/ipa_hw_manager/hardware_manager.py#L21-L46 | 17:08 |
| *** jistr has quit IRC | 17:08 | |
| JoshNang | its not. the execute clean steps patch too much longer than i thought it would yesterday...trying to get it so get_clean_steps would work for the agent with a node that was likely powered off :/ | 17:08 |
| JoshNang | rameshg87_: that looks pretty spot on | 17:08 |
| JoshNang | the function args will be a bit different, but otherwise, i think that'll work | 17:09 |
| rameshg87_ | yeah | 17:09 |
| rameshg87_ | JoshNang, will change after your ironic-python-agent patch gets merged | 17:09 |
| JoshNang | nice! :D | 17:09 |
| rameshg87_ | JoshNang, will change accordingly | 17:09 |
| JoshNang | rameshg87_: sounds good. that should be in a mergeable state today | 17:09 |
| JoshNang | i'm off tomorrow and monday (my parents are visiting), but i'll probably poke my head in to do fixups both days since we're close to the deadline | 17:10 |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Check temp dir is writable https://review.openstack.org/160383 | 17:10 |
| *** romcheg has quit IRC | 17:10 | |
| rameshg87_ | JoshNang, oh okay | 17:10 |
| rameshg87_ | JoshNang, really waiting for your next patch on clean steps :) | 17:10 |
| * JoshNang gets writing | 17:11 | |
| rameshg87_ | JoshNang, just a quick question on node powering on | 17:11 |
| rameshg87_ | JoshNang, so it would setup the dhcp options for the node and then power it on, right ? i am wondering how would i make it work for agent_ilo thing (which boots from virtual media) | 17:12 |
| JoshNang | rameshg87_: i call utils.node_power_action to power the node on | 17:14 |
| rameshg87_ | JoshNang, and neutron config ? | 17:15 |
| JoshNang | heh that's removed in tear_down isn't it? | 17:15 |
| rameshg87_ | JoshNang, yeah | 17:15 |
| JoshNang | ugh | 17:15 |
| NobodyCam | jlvillal: take a look at the latest rev of 160383 | 17:16 |
| dtantsur | g'night folks! | 17:16 |
| *** dtantsur is now known as dtantsur|afk | 17:16 | |
| NobodyCam | night dtantsur | 17:16 |
| jlvillal | NobodyCam: Okay, but my worry is that is really isn't a tempfile space issue. | 17:16 |
| JoshNang | rameshg87_: i might just have to move tear_down to after cleaning | 17:16 |
| jlvillal | NobodyCam: Okay, but my worry is that it really isn't a tempfile space issue. Not sure though | 17:17 |
| JoshNang | hmm but that's not good either | 17:17 |
| rameshg87_ | JoshNang, and the pxe config setup as well | 17:17 |
| NobodyCam | jlvillal: I'm about to file another bug | 17:17 |
| jlvillal | NobodyCam: I'm looking at the function for the password writing. | 17:17 |
| jlvillal | NobodyCam: My worry is that error message. | 17:17 |
| NobodyCam | jlvillal: ack | 17:19 |
| *** igordcard has quit IRC | 17:20 | |
| jlvillal | NobodyCam: ./ironic.130228/.tox/py27/lib/python2.7/site-packages/oslo_concurrency/processutils.py: description = _("Unexpected error while running command.") | 17:22 |
| jlvillal | NobodyCam: WARNING ironic.drivers.modules.ipmitool [-] IPMI power status failed for node a8cb6624-0d9f-c882-affc-046ebb96ec01 with error: Failed to create the password file. Unexpected error while running command. | 17:22 |
| NobodyCam | jlvillal: yep the temp dir was not writtable | 17:24 |
| jlvillal | NobodyCam: Wouldn't that same error occur if the command returned a non-zero value? 'ipmitool' | 17:26 |
| *** dprince has joined #openstack-ironic | 17:26 | |
| jlvillal | NobodyCam: But as you can tell I'm confused on the underlying source of the error | 17:27 |
| jlvillal | NobodyCam: the _make_password_file is a decorator | 17:28 |
| jlvillal | So it seems to catch the process execution error | 17:28 |
| *** erwan_taf has quit IRC | 17:29 | |
| jlvillal | I think the try loop in _make_password_file should be smaller. For sure the 'yield' statement should be out of the try loop. | 17:29 |
| lucasagomes | IPA folks, when you get some time: https://review.openstack.org/#/c/157469/ | 17:31 |
| *** krtaylor has quit IRC | 17:31 | |
| *** vdrok is now known as vdrok_afk | 17:32 | |
| *** spandhe has joined #openstack-ironic | 17:32 | |
| *** BadCub is now known as BadCub_Breakfast | 17:34 | |
| JoshNang | rameshg87_: i think we're going to have to add a function called pre_cleaning or something, so each driver has a chance to reboot from the tenant configs to a ramdisk (if necessary) | 17:36 |
| rameshg87_ | JoshNang, yeah that makes sense .. | 17:37 |
| *** krtaylor has joined #openstack-ironic | 17:37 | |
| rameshg87_ | thanks | 17:37 |
| JoshNang | it'll get refactored to the boot interface next cycle i think | 17:38 |
| NobodyCam | jlvillal: http://bugs.launchpad.net/ironic/+bug/1428763 | 17:40 |
| openstack | Launchpad bug 1428763 in Ironic "many ironic functions appear to not be honoring tempdir conf file setting" [Undecided,New] | 17:40 |
| jlvillal | NobodyCam: I am not sure. | 17:40 |
| jlvillal | NobodyCam: About to push a patch ... | 17:41 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Make try block shorter for _make_password_file https://review.openstack.org/161803 | 17:41 |
| jlvillal | NobodyCam: Can you take a look at ^^^^ | 17:41 |
| jlvillal | Please tell me if I'm crazy :) | 17:41 |
| NobodyCam | no thats a good patch. thou I would but the close in a finally: block | 17:43 |
| NobodyCam | but thats just me | 17:44 |
| jlvillal | NobodyCam: Good idea :) | 17:45 |
| rameshg87_ | jlvillal, i just replied to your comments | 17:45 |
| rameshg87_ | jlvillal, please have a look at it when you get some time | 17:46 |
| jlvillal | rameshg87_: Will do! | 17:46 |
| rameshg87_ | jlvillal, regarding deleting deploy_utils.mkfs(), i think the main purpose of having methods to call inline ones in utils is for mocking | 17:46 |
| jlvillal | rameshg87_: I assume a new patch is coming. | 17:46 |
| rameshg87_ | jlvillal, that's the only reason i can think of why it is done os | 17:46 |
| rameshg87_ | *soo | 17:46 |
| jlvillal | rameshg87_: Isn't it relatively easy to mock the utils.mkfs one though? | 17:47 |
| rameshg87_ | jlvillal, if i change them to utils.mkfs then it would require a lot of unrelated changes in tests | 17:47 |
| jlvillal | rameshg87_: Before the function was mkswap() | 17:47 |
| jlvillal | I thought | 17:47 |
| rameshg87_ | jlvillal, wouldn't it be better if it was in a follow-up patch may be sometime later ? | 17:47 |
| * jlvillal goes back and looks at patch... | 17:47 | |
| jlvillal | rameshg87_: I still think the order of the arguments should match with utils.mkfs | 17:48 |
| rameshg87_ | jlvillal, yeah i agree, will change that | 17:48 |
| rameshg87_ | jlvillal, pushing a new patch now | 17:48 |
| jlvillal | Since function was renamed from mkswap() to mkfs() | 17:48 |
| jlvillal | rameshg87_: I'd be fine if want to leave mkfs in there and not use utils.mkfs | 17:49 |
| rameshg87_ | jlvillal, i will fix others and put a new patch soon | 17:49 |
| jlvillal | rameshg87_: But would prefer it was removed at some point. e.g. a later patch | 17:49 |
| rameshg87_ | jlvillal, yeah i will give a try .. there are other methods also | 17:50 |
| rameshg87_ | jlvillal, def dd() | 17:50 |
| jlvillal | rameshg87_: Oh :( | 17:50 |
| rameshg87_ | jlvillal, so may be a single patch can remove them and may be modify the tests accordingly | 17:50 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/161811 | 17:52 |
| NobodyCam | rameshg87_: also if you have a chance can you take a look at the latest changes to 160383 | 17:56 |
| rameshg87_ | NobodyCam, sure .. | 17:57 |
| jlvillal | NobodyCam: I think it makes it much more complicated to try to do the os.close() in the finally clause. :( If you have ideas, it would be appreciated. I'm thinking I would need to do it in finally with a try/except OSError, and probably also in the current except block. | 17:57 |
| *** kkoski has quit IRC | 17:57 | |
| *** kkoski has joined #openstack-ironic | 17:57 | |
| NobodyCam | jlvillal: you make be correct I just looked quickly | 17:57 |
| jlvillal | jlvillal: I'm thinking best idea would be to go back to the 'with os.fdopen()' | 17:58 |
| NobodyCam | make == maybe | 17:58 |
| *** harlowja_away is now known as harlowja | 17:58 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add localboot support for uefi boot mode https://review.openstack.org/159855 | 17:59 |
| *** lazy_prif has joined #openstack-ironic | 17:59 | |
| *** lazy_prif is now known as lazy_prince | 17:59 | |
| *** killer_prince has quit IRC | 17:59 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Make try block shorter for _make_password_file https://review.openstack.org/161803 | 17:59 |
| jlvillal | NobodyCam: I went back to the with statement :) I think that is most elegant/simple solution | 18:00 |
| rameshg87_ | NobodyCam, in 160383, we went back to the previous way of not-returning the tempdir ? | 18:05 |
| rameshg87_ | NobodyCam, i see changes made in ps9 are not there in ps10 - https://review.openstack.org/#/c/160383/10/ironic/common/utils.py | 18:06 |
| *** achanda has joined #openstack-ironic | 18:08 | |
| *** penick has joined #openstack-ironic | 18:09 | |
| *** lucasagomes is now known as lucas-dinner | 18:09 | |
| *** jlk has left #openstack-ironic | 18:10 | |
| * rameshg87_ is too sleepy to sit in front of computer | 18:10 | |
| *** rameshg87_ is now known as rameshg87-sleepy | 18:10 | |
| rameshg87-sleepy | JoshNang, please post any incomplete patches that you might have by the end of the day | 18:11 |
| JoshNang | rameshg87-sleepy: i'll do what i can | 18:11 |
| rameshg87-sleepy | JoshNang, i will try to build something on top of it tomorrow and check if everything works fine | 18:11 |
| rameshg87-sleepy | JoshNang, <i will try> :) | 18:11 |
| rameshg87-sleepy | JoshNang, thanks a lot | 18:11 |
| * rameshg87-sleepy leaves for the day | 18:13 | |
| rameshg87-sleepy | goodnight all | 18:13 |
| *** rameshg87-sleepy has quit IRC | 18:13 | |
| *** anderbubble has joined #openstack-ironic | 18:15 | |
| *** romcheg has joined #openstack-ironic | 18:16 | |
| *** BadCub_Breakfast is now known as BadCub | 18:16 | |
| *** jcoufal_ has quit IRC | 18:18 | |
| *** romcheg has quit IRC | 18:19 | |
| *** romcheg has joined #openstack-ironic | 18:22 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-python-agent: Add cleaning/zapping support to IPA https://review.openstack.org/161066 | 18:22 |
| *** Marga_ has quit IRC | 18:22 | |
| devananda | BadCub: hi there! still up for continuing our kilo3 status review? | 18:28 |
| BadCub | devananda yep, just doing some final housekeeping on internal thing. :-) | 18:29 |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-python-agent: Add cleaning/zapping support to IPA https://review.openstack.org/161066 | 18:29 |
| devananda | BadCub: np. I just got to the office. happy to start a bit late... | 18:30 |
| BadCub | Coolness. That works for me too! Had my head down the rabbit hole for a few minutes lol | 18:31 |
| NobodyCam | morning devananda :) | 18:31 |
| *** hj-hp has quit IRC | 18:38 | |
| jlvillal | NobodyCam: So on my patch. Does it make sense that maybe that ipmitool isn't an issue with the tempdir being out of space? Maybe it is out of space, but the error I see is that the impitool command failed. So I am unsure if it is really a space issue. | 18:38 |
| jlvillal | NobodyCam: Wondering what you think. | 18:39 |
| NobodyCam | jlvillal: thats why I added a space check | 18:39 |
| *** wshao has joined #openstack-ironic | 18:39 | |
| NobodyCam | tho in the latest rev of my patch it is only checked at startup | 18:40 |
| *** hj-hp has joined #openstack-ironic | 18:40 | |
| jlvillal | NobodyCam: Okay. I just wasn't sure if the thought it was a space issue was a red herring because we had seen the PasswordFileFailedToCreate exception. And I believe that was raised erroneously. | 18:41 |
| jlvillal | NobodyCam: Space check is still a good idea. But not sure if that was the underlying cause of that error seen. | 18:41 |
| *** dtantsur|afk has quit IRC | 18:43 | |
| *** athomas has quit IRC | 18:43 | |
| openstackgerrit | Clif Houck proposed openstack/ironic: Added support for image caching for the agent driver https://review.openstack.org/161832 | 18:43 |
| *** dtantsur has joined #openstack-ironic | 18:45 | |
| *** krtaylor has quit IRC | 18:45 | |
| *** BB_ has joined #openstack-ironic | 18:46 | |
| BB_ | Hello | 18:46 |
| NobodyCam | jlvillal: I was not the one who hit that error, so I've been going off the error it self | 18:47 |
| BB_ | I am trying to understand the docker support with ironic | 18:47 |
| BB_ | on openstack.. | 18:47 |
| *** jrist has joined #openstack-ironic | 18:47 | |
| *** penick has quit IRC | 18:47 | |
| BB_ | can someone point to right links ? | 18:47 |
| BB_ | for dockers using ironic | 18:47 |
| NobodyCam | hi there BB_ | 18:48 |
| NobodyCam | I believe the IPA ramdisk is where you'll want to look | 18:48 |
| jroll | uhh | 18:48 |
| jroll | BB_: what are you trying to do? use ironic to provision bare metal + docker? | 18:49 |
| BB_ | Nobodycam:Hi | 18:49 |
| BB_ | Jroll: Yes | 18:49 |
| NobodyCam | oh I miss read that | 18:49 |
| jroll | ok, that's totally different. I think there may be a thing that does that, but I don't know what it is. maybe magnum? | 18:50 |
| BB_ | jroll: okay. | 18:51 |
| *** sambetts has quit IRC | 18:51 | |
| BB_ | jroll: is there a link explaining about ironic brtter ? | 18:51 |
| jroll | BB_: http://docs.openstack.org/developer/ironic/deploy/user-guide.html | 18:52 |
| jroll | BB_: basically, ironic deploys bare metal machines. nothing more. | 18:52 |
| NobodyCam | :) jroll is quicker then /me | 18:52 |
| *** sambetts has joined #openstack-ironic | 18:52 | |
| *** DCWilliams_VA has joined #openstack-ironic | 18:53 | |
| *** jerryz_ has quit IRC | 18:53 | |
| *** ndipanov has quit IRC | 18:55 | |
| *** hj-hp has quit IRC | 18:55 | |
| *** jmccrory has joined #openstack-ironic | 18:58 | |
| *** krtaylor has joined #openstack-ironic | 18:59 | |
| NobodyCam | brb | 19:03 |
| *** athomas has joined #openstack-ironic | 19:04 | |
| *** achanda has quit IRC | 19:04 | |
| *** penick has joined #openstack-ironic | 19:05 | |
| devananda | BadCub: let's see if my headphones work | 19:06 |
| devananda | BB_: you could, in theory, deploy any OS onto said bare metal machines. It could contain docker, or windows, or what ever | 19:07 |
| BB_ | jroll: thanks for the link | 19:08 |
| jroll | np :) | 19:08 |
| BB_ | devanada & jroll: | 19:08 |
| jroll | devananda: I presume you're going to want a spec for something like: https://review.openstack.org/161832 | 19:08 |
| BB_ | was looking at http://www.symantec.com/connect/blogs/lxc-and-docker-containers-nova-openstack | 19:08 |
| jroll | devananda: or is that one of those vendor passthru things that we just completely turn a blind eye to | 19:08 |
| BB_ | this did not have ironic in picture for baremetal | 19:09 |
| *** hj-hp has joined #openstack-ironic | 19:09 | |
| *** kozhukalov has quit IRC | 19:10 | |
| devananda | jroll: /me looks | 19:11 |
| *** Marga_ has joined #openstack-ironic | 19:11 | |
| devananda | jroll: so not a blind eye -- it should still get sanity reviewed. it may or may not warrant a BP to describe the design. but also it's kinda late for kilo, given our other priorities | 19:12 |
| devananda | jroll: my reaction to skimming the code is that it doesn't need a BP. it's limited to the driver's vendor inteface, and I grok it right away | 19:13 |
| * BadCub beats google hangouts into submission | 19:15 | |
| openstackgerrit | Adam Gandelman proposed openstack/ironic: Check-in Ironic's devstack code as in-tree plugin https://review.openstack.org/161485 | 19:20 |
| *** pelix has quit IRC | 19:23 | |
| *** achanda has joined #openstack-ironic | 19:24 | |
| *** ijw has joined #openstack-ironic | 19:24 | |
| *** DCWilliams_VA has quit IRC | 19:25 | |
| *** DCWilliams_VA has joined #openstack-ironic | 19:26 | |
| *** DCWilliams_VA has quit IRC | 19:26 | |
| *** DCWilliams_VA has joined #openstack-ironic | 19:26 | |
| *** jmccrory has quit IRC | 19:27 | |
| *** jmccrory has joined #openstack-ironic | 19:31 | |
| NobodyCam | devananda: you bumped the ceilomemter stuff? | 19:31 |
| *** hj-hp has quit IRC | 19:32 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Image building support for fedora hosts https://review.openstack.org/154559 | 19:39 |
| BadCub | NobodyCam It is still on the list. I am reading the notes on that now | 19:42 |
| *** erwan_taf has joined #openstack-ironic | 19:42 | |
| *** pas-ha has quit IRC | 19:44 | |
| *** gridinv has joined #openstack-ironic | 19:47 | |
| NobodyCam | BadCub: you should have been cc on a emails between Jim and myself about his ablity to complete that work | 19:47 |
| NobodyCam | brb | 19:47 |
| *** anderbubble has quit IRC | 19:49 | |
| *** anderbubble has joined #openstack-ironic | 19:50 | |
| openstackgerrit | Merged openstack/ironic-python-agent: Add the image extension (for local boot) https://review.openstack.org/157469 | 19:51 |
| *** DCWilliams_VA has quit IRC | 19:52 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: ironic port deletion fails even if node is locked by same process https://review.openstack.org/161861 | 19:52 |
| *** hj-hp has joined #openstack-ironic | 20:01 | |
| * NobodyCam is back | 20:03 | |
| *** wshao has quit IRC | 20:11 | |
| *** hj-hp has quit IRC | 20:20 | |
| *** Marga_ has quit IRC | 20:20 | |
| jroll | devananda: right on, just curious what you thought; not necessarily trying to target to kilo. thanks :) | 20:20 |
| *** Marga_ has joined #openstack-ironic | 20:21 | |
| *** coolsvap_ is now known as coolsvap | 20:24 | |
| *** openstackgerrit has quit IRC | 20:25 | |
| *** openstackgerrit has joined #openstack-ironic | 20:26 | |
| *** penick has quit IRC | 20:31 | |
| *** gridinv has quit IRC | 20:33 | |
| *** achanda has quit IRC | 20:38 | |
| *** achanda has joined #openstack-ironic | 20:41 | |
| *** penick has joined #openstack-ironic | 20:44 | |
| *** david-lyle_afk has joined #openstack-ironic | 20:45 | |
| *** david-lyle_afk has quit IRC | 20:47 | |
| *** david-lyle_afk has joined #openstack-ironic | 20:47 | |
| *** coolsvap is now known as coolsvap_ | 20:52 | |
| *** penick has quit IRC | 20:53 | |
| *** hj-hp has joined #openstack-ironic | 20:59 | |
| *** david-lyle_afk is now known as david-lyle | 21:03 | |
| *** romcheg1 has joined #openstack-ironic | 21:04 | |
| jlvillal | NobodyCam: Thanks for the patch review :) | 21:05 |
| *** romcheg has quit IRC | 21:05 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Implement execute clean steps https://review.openstack.org/155561 | 21:08 |
| *** achanda has quit IRC | 21:09 | |
| *** ifarkas has quit IRC | 21:10 | |
| NobodyCam | huh | 21:10 |
| NobodyCam | :-p | 21:10 |
| *** Marga_ has quit IRC | 21:10 | |
| *** Marga_ has joined #openstack-ironic | 21:11 | |
| NobodyCam | JoshNang: I'm looking at https://review.openstack.org/#/c/161474/1/nova/virt/ironic/driver.py what do you think about a special log message for cleaning .. Ironic node %(node)s is now unprovisioned, and will become available after cleaning .... or something like that? | 21:14 |
| NobodyCam | as the commit message points this could take hours | 21:15 |
| mjturek1 | heyy is this the best place to ask triple-o questions? | 21:16 |
| NobodyCam | I would ask in #TripleO | 21:16 |
| mjturek1 | NobodyCam, thanks! | 21:16 |
| NobodyCam | hehe NP mjturek1 :) | 21:16 |
| JoshNang | NobodyCam: seems like something that shouldn't concern the user | 21:17 |
| JoshNang | though, i guess its just in the logs. /me adds | 21:18 |
| NobodyCam | I was thinking as a operator, if I had only a limited # of nodes and I was to to them becoming available right away might be somehting good to note | 21:18 |
| NobodyCam | :) | 21:18 |
| NobodyCam | s/to to/use to/ | 21:19 |
| JoshNang | that makes sense | 21:19 |
| JoshNang | NobodyCam: though, this makes it a race condition on which log message they get. even if the node is going to be cleaned, it'll be in deleting for a short time. might be more confusing. how about i mention cleaning in the log message as a possibility? | 21:21 |
| *** hj-hp has quit IRC | 21:24 | |
| NobodyCam | JoshNang: that would work. I was just looking for a way a operator could tell that even thou the was unprovisioned it may be a while before its back in the pool | 21:28 |
| jlvillal | JoshNang: I'm confused: https://review.openstack.org/#/c/155561/13/ironic/conductor/manager.py | 21:28 |
| jlvillal | JoshNang: def _do_next_clean_step | 21:28 |
| JoshNang | NobodyCam: hopefully that will at least give them a hint | 21:28 |
| jlvillal | JoshNang: I'm confused on what is value in passing 'last_step' around. | 21:29 |
| jlvillal | Seems like last_step is node.clean_step | 21:29 |
| JoshNang | jlvillal: so i can use the same function for cleaning and zapping | 21:29 |
| jlvillal | So in that function, it uses last_step as an argument. | 21:30 |
| jlvillal | But then it modifies: node.clean_step. | 21:30 |
| *** penick has joined #openstack-ironic | 21:30 | |
| jlvillal | Which seems like modifying the same thing two ways. Thus, why I am confused. | 21:30 |
| *** achanda has joined #openstack-ironic | 21:30 | |
| jlvillal | JoshNang: So I'm still confused, even after what you just said. | 21:31 |
| NobodyCam | JoshNang: +1'd :) thank you | 21:32 |
| *** achanda has quit IRC | 21:32 | |
| JoshNang | jlvillal: right, so i don't have states.ZAPPING yet, so i can't add the if provision_state==zapping yet. | 21:32 |
| JoshNang | though i do spy one spot where i should use last_step instead of node.clean_step in a log message. the only time it should be clean_step or zap_step is when dealing with the database, because i can't get around that | 21:33 |
| JoshNang | it's really just a stylistic choice to avoid "if clean... if zapping..." all over the place | 21:34 |
| *** achanda has joined #openstack-ironic | 21:34 | |
| jlvillal | JoshNang: So looking at line 916, I see node.clean_step = step | 21:35 |
| jlvillal | JoshNang: Okay, I'm getting more confused. Is 'last_step' always going to equal node.clean_step? | 21:35 |
| jlvillal | or when zapping happens it will be different? | 21:35 |
| jlvillal | Is that what you mean? | 21:35 |
| * jlvillal thinks he is getting less confused... | 21:36 | |
| JoshNang | right. last_step will be last cleaning step in cleaning, last zapping step in zapping | 21:36 |
| jlvillal | JoshNang: And it is okay on line 916 that it is basically modifying 'last_step' if it is a cleaning step? | 21:36 |
| *** achanda has quit IRC | 21:37 | |
| JoshNang | jlvillal: last_step isn't used again at that point, so i think it should be fine | 21:38 |
| JoshNang | NobodyCam: thanks for the review! | 21:39 |
| *** achanda has joined #openstack-ironic | 21:39 | |
| jlvillal | JoshNang: Okay. I'm still confused but I will read the code more and hopefully become un-confused :) | 21:39 |
| JoshNang | jlvillal: thanks for reviewing :) | 21:40 |
| jlvillal | JoshNang: You're welcome | 21:40 |
| mrda | rloo: Thanks again for your thorough reviews :) | 21:41 |
| *** erwan_taf has quit IRC | 21:42 | |
| *** dprince has quit IRC | 21:42 | |
| *** Marga_ has quit IRC | 21:50 | |
| *** Marga_ has joined #openstack-ironic | 21:50 | |
| *** anderbubble_ has joined #openstack-ironic | 21:51 | |
| *** anderbubble has quit IRC | 21:52 | |
| *** anderbubble_ is now known as anderbubble | 21:52 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Make try block shorter for _make_password_file https://review.openstack.org/161803 | 21:58 |
| *** ijw has quit IRC | 22:00 | |
| *** hj-hp has joined #openstack-ironic | 22:10 | |
| *** ijw has joined #openstack-ironic | 22:12 | |
| *** oomichi has joined #openstack-ironic | 22:16 | |
| NobodyCam | JoshNang: get_clean_steps in https://review.openstack.org/#/c/161066/5/ironic_python_agent/hardware.py just looks odd to me | 22:17 |
| JoshNang | NobodyCam: how so? it definitely looks better when you return a bigger list of clean_steps | 22:19 |
| NobodyCam | if your buildnig a static list why pass in node and port? | 22:20 |
| JoshNang | my thought was to allow the hardware manager to change the list based on node properties or something of that sort | 22:21 |
| jroll | capabilities~ | 22:22 |
| jroll | (or something, idk) | 22:22 |
| NobodyCam | :) | 22:23 |
| JoshNang | ^ yeah. and its way easier to not use those props than to add them later (as i found with executing steps downstream) | 22:24 |
| *** penick has quit IRC | 22:25 | |
| jlvillal | JoshNang: Feel free to overrule me on my request for a unittest :) | 22:25 |
| JoshNang | jlvillal: i'll look deeper in a bit but it looks reasonable. thanks for the review :) | 22:26 |
| *** penick has joined #openstack-ironic | 22:36 | |
| *** achanda has quit IRC | 22:36 | |
| rloo | mrda: yw. sorry, it wasn't that thorough. I kind of peter'd out, but I gave enough comments to keep you busy for a bit ;) | 22:36 |
| *** penick has quit IRC | 22:37 | |
| *** achanda has joined #openstack-ironic | 22:40 | |
| *** kkoski has quit IRC | 22:41 | |
| *** Marga_ has quit IRC | 22:45 | |
| *** Marga_ has joined #openstack-ironic | 22:45 | |
| *** achanda has quit IRC | 23:05 | |
| *** achanda has joined #openstack-ironic | 23:08 | |
| *** yuanying has joined #openstack-ironic | 23:11 | |
| *** ChuckC is now known as ChuckC_ | 23:19 | |
| NobodyCam | hey hey JoshNang on https://review.openstack.org/#/c/160515 I fully get the stick with dict thing. :) why the cheange from driver to interface? | 23:22 |
| *** mjturek1 has quit IRC | 23:23 | |
| JoshNang | NobodyCam: woo! mostly from confusing terminology. cleaning is implemented by the deploy, power, etc interfaces, rather than the driver itself (it was called out a few times in other reviews) | 23:25 |
| devananda | does IPA yet support pulling down a partition image, partitioning the disks, and all that? | 23:25 |
| devananda | I think not, but we could really use some docs on this ... | 23:26 |
| NobodyCam | JoshNang: mind adding that to the comit mesage? | 23:26 |
| JoshNang | NobodyCam: can do! | 23:26 |
| NobodyCam | devananda: there was a patch for that | 23:26 |
| *** ChuckC_ is now known as ChuckC | 23:26 | |
| * NobodyCam looks | 23:26 | |
| JoshNang | devananda: in progress it looks like https://review.openstack.org/#/c/160224/ | 23:26 |
| devananda | NobodyCam: IPA can expose the disks and let the pxe driver (really iscsi deploy code) hande them | 23:26 |
| devananda | ahhh there it is | 23:27 |
| devananda | the lib code | 23:27 |
| jroll | devananda: only with pxe driver | 23:27 |
| *** anderbubble has quit IRC | 23:27 | |
| devananda | jroll: ok, that's what I thought. thanks | 23:27 |
| jroll | why are we pulling the lib code in like that | 23:27 |
| jroll | I hope it's just for testing | 23:27 |
| devananda | jroll: I'm guessing at this point that lib-ironic isn't going to be a think this cycle | 23:27 |
| devananda | also, yea, wtf? | 23:28 |
| jroll | yeah | 23:28 |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-specs: Change clean_steps to a dict https://review.openstack.org/160515 | 23:30 |
| *** hj-hp has quit IRC | 23:31 | |
| jlvillal | JoshNang: https://review.openstack.org/#/c/160515/2/specs/kilo/implement-cleaning-states.rst line 100ish. Is the 'overlapping priorities' check implemented? | 23:39 |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Implement execute clean steps https://review.openstack.org/155561 | 23:40 |
| JoshNang | jlvillal: not yet | 23:40 |
| jlvillal | JoshNang: Thanks | 23:40 |
| JoshNang | np! | 23:41 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/161575 | 23:42 |
| jlvillal | JoshNang: https://review.openstack.org/#/c/155561/13..14/ironic/drivers/base.py Why is task being added? | 23:44 |
| jlvillal | Oh, for override. | 23:44 |
| *** wshao has joined #openstack-ironic | 23:44 | |
| jlvillal | NVM! | 23:44 |
| NobodyCam | devananda: I see logical name support under client on the priorities google doc. Mind if I pick up: https://review.openstack.org/#/c/158520 [ tomorrow ofc ;-P ] | 23:44 |
| devananda | NobodyCam: go for it. the code is, I think, basically good. it needs unit tests, maybe abit of polish, but really it depends on naohirot's change to land first | 23:45 |
| devananda | oh, did thta land? | 23:46 |
| devananda | no... | 23:46 |
| devananda | oh, odd | 23:46 |
| JoshNang | jlvillal: heh i knew adding that comment was a good idea :) | 23:47 |
| devananda | NobodyCam: so my logical names client patch should be based on 155624 | 23:48 |
| devananda | NobodyCam: because it's impossible to use it without sending the X-OpenStack-Ironic-API-Version header | 23:49 |
| devananda | gah. s/naohirot/tanlin/ .... I dont know why my brain swapped their names just a minute ago | 23:49 |
| devananda | aaaand that is now stalled on the spec | 23:50 |
| jlvillal | JoshNang: +1! :) | 23:52 |
| *** openstack has joined #openstack-ironic | 23:53 | |
| devananda | NobodyCam: no | 23:54 |
| NobodyCam | :) | 23:55 |
| devananda | NobodyCam: https://review.openstack.org/#/c/161110/1/specs/kilo/api-microversions.rst,cm | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!