| *** alexpilotti has quit IRC | 00:00 | |
| *** rbudden has quit IRC | 00:04 | |
| *** alexpilotti has joined #openstack-ironic | 00:08 | |
| *** alexpilotti has quit IRC | 00:12 | |
| *** garthb has quit IRC | 00:14 | |
| betherly | jroll: come online to look at pings lol. Will look properly when I'm up tomorrow | 00:16 |
|---|---|---|
| *** keedya has joined #openstack-ironic | 00:17 | |
| *** jaypipes has quit IRC | 00:21 | |
| *** harshs has quit IRC | 00:27 | |
| *** harshs has joined #openstack-ironic | 00:28 | |
| *** suro-patz has joined #openstack-ironic | 00:30 | |
| devananda | jroll: bug in the FlatNetwork class. __init__ is not called at proces start time now, but rather when ever a node is instantiated -- meaning that if the config is wrong, it becomes impossible to use the driver, but the conductor starts anyway | 00:31 |
| *** thrash is now known as thrash|g0ne | 00:33 | |
| *** hoangcx has joined #openstack-ironic | 00:39 | |
| *** smoriya_ has joined #openstack-ironic | 00:41 | |
| *** suro-patz has quit IRC | 00:54 | |
| *** cdearborn has quit IRC | 00:59 | |
| *** baoli has joined #openstack-ironic | 00:59 | |
| *** suro-patz has joined #openstack-ironic | 01:00 | |
| *** Sukhdev has quit IRC | 01:13 | |
| *** cfarquhar has quit IRC | 01:18 | |
| *** spandhe has quit IRC | 01:20 | |
| *** cfarquhar has joined #openstack-ironic | 01:21 | |
| *** cfarquhar has quit IRC | 01:21 | |
| *** cfarquhar has joined #openstack-ironic | 01:21 | |
| *** baoli has quit IRC | 01:23 | |
| *** baoli has joined #openstack-ironic | 01:28 | |
| *** harshs has quit IRC | 01:36 | |
| *** suro-patz has quit IRC | 01:41 | |
| *** praneshp__ has quit IRC | 01:43 | |
| *** jaybeale has quit IRC | 01:51 | |
| *** JieLi has joined #openstack-ironic | 02:02 | |
| *** JieLi has quit IRC | 02:03 | |
| *** Marga_ has quit IRC | 02:11 | |
| *** baoli has quit IRC | 02:11 | |
| *** baoli has joined #openstack-ironic | 02:12 | |
| *** alex_xu has quit IRC | 02:13 | |
| *** smoriya_ has quit IRC | 02:14 | |
| *** alex_xu has joined #openstack-ironic | 02:16 | |
| *** achanda has joined #openstack-ironic | 02:21 | |
| *** baoli has quit IRC | 02:22 | |
| *** suro-patz has joined #openstack-ironic | 02:35 | |
| *** suro-patz has quit IRC | 02:39 | |
| *** mfisch has joined #openstack-ironic | 02:45 | |
| *** spandhe has joined #openstack-ironic | 02:53 | |
| *** Marga_ has joined #openstack-ironic | 03:01 | |
| *** Marga_ has quit IRC | 03:06 | |
| *** sekrit has joined #openstack-ironic | 03:17 | |
| *** alex_xu has quit IRC | 03:18 | |
| *** Marga_ has joined #openstack-ironic | 03:19 | |
| *** alex_xu has joined #openstack-ironic | 03:20 | |
| *** jaybeale has joined #openstack-ironic | 03:20 | |
| *** spandhe_ has joined #openstack-ironic | 03:30 | |
| *** spandhe has quit IRC | 03:31 | |
| *** spandhe_ is now known as spandhe | 03:31 | |
| *** links has joined #openstack-ironic | 03:39 | |
| mrda | So, just a quick question, how do you get ironic's properties for disk size to align with nova's flavors? I would have thought that setting the ironic node's properties/disk to 250 and the flavor disk to 250 would have sufficed, but no :-( | 03:40 |
| *** Nisha has joined #openstack-ironic | 03:45 | |
| *** jaybeale has quit IRC | 03:47 | |
| *** Nisha has quit IRC | 03:51 | |
| *** alex_xu has quit IRC | 03:58 | |
| *** alex_xu has joined #openstack-ironic | 04:01 | |
| *** Nisha has joined #openstack-ironic | 04:04 | |
| *** vmud213 has joined #openstack-ironic | 04:05 | |
| *** garthb has joined #openstack-ironic | 04:09 | |
| *** moshele has joined #openstack-ironic | 04:15 | |
| *** keedya has quit IRC | 04:20 | |
| *** alexpilotti has joined #openstack-ironic | 04:32 | |
| Nisha | morning ironic | 04:33 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Enhance ilo drivers to do inband inspection https://review.openstack.org/201904 | 04:35 |
| *** alexpilotti has quit IRC | 04:37 | |
| *** kalpase has joined #openstack-ironic | 04:38 | |
| *** smoriya_ has joined #openstack-ironic | 04:38 | |
| *** spandhe has quit IRC | 04:41 | |
| *** spandhe has joined #openstack-ironic | 04:42 | |
| *** kalpase has quit IRC | 04:43 | |
| *** ChubYann has quit IRC | 04:43 | |
| *** hoangcx has quit IRC | 04:46 | |
| *** garthb has quit IRC | 04:46 | |
| *** Sukhdev has joined #openstack-ironic | 04:50 | |
| *** hoangcx_ has joined #openstack-ironic | 04:52 | |
| *** hoangcx__ has joined #openstack-ironic | 04:52 | |
| *** hoangcx_ is now known as hoangcx | 04:52 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Correct api version check conditional for node.name https://review.openstack.org/299264 | 04:53 |
| *** mtanino has quit IRC | 04:53 | |
| *** hoangcx__ has quit IRC | 04:53 | |
| *** hoangcx has quit IRC | 04:54 | |
| *** moshele has quit IRC | 04:54 | |
| *** hoangcx has joined #openstack-ironic | 04:54 | |
| *** hoangcx has quit IRC | 04:57 | |
| *** hoangcx has joined #openstack-ironic | 04:58 | |
| *** Sukhdev has quit IRC | 05:06 | |
| *** alexpilotti has joined #openstack-ironic | 05:06 | |
| *** Sukhdev has joined #openstack-ironic | 05:06 | |
| *** alexpilotti has quit IRC | 05:11 | |
| *** Nisha_away has joined #openstack-ironic | 05:21 | |
| *** Nisha has quit IRC | 05:24 | |
| *** praneshp_ has joined #openstack-ironic | 05:26 | |
| *** Nisha_away is now known as Nisha | 05:28 | |
| *** jcoufal has joined #openstack-ironic | 05:45 | |
| *** rcernin has joined #openstack-ironic | 05:51 | |
| *** Sukhdev has quit IRC | 05:51 | |
| *** dougs has quit IRC | 05:57 | |
| *** praneshp_ has quit IRC | 06:00 | |
| *** kalpase has joined #openstack-ironic | 06:01 | |
| *** alexpilotti has joined #openstack-ironic | 06:07 | |
| *** alexpilotti has quit IRC | 06:12 | |
| *** spandhe has quit IRC | 06:12 | |
| *** JieLi has joined #openstack-ironic | 06:15 | |
| *** achanda has quit IRC | 06:40 | |
| *** moshele has joined #openstack-ironic | 07:02 | |
| *** suro-patz has joined #openstack-ironic | 07:03 | |
| openstackgerrit | Sofiia Andriichenko proposed openstack/python-ironicclient: Add test for 'chassis-show --field' command https://review.openstack.org/301479 | 07:05 |
| *** JieLi has quit IRC | 07:08 | |
| *** karimb has joined #openstack-ironic | 07:13 | |
| *** suro-patz has quit IRC | 07:25 | |
| *** ndipanov has joined #openstack-ironic | 07:27 | |
| *** Marga_ has quit IRC | 07:29 | |
| *** pcaruana has joined #openstack-ironic | 07:37 | |
| *** Nisha has quit IRC | 07:40 | |
| *** achanda has joined #openstack-ironic | 07:40 | |
| Haomeng | lucasagomes: would you help to review my spec if you have time today - https://review.openstack.org/#/c/266688/ , got 1x+2, thanks, nice day! | 07:44 |
| *** ifarkas has joined #openstack-ironic | 07:44 | |
| *** jistr has joined #openstack-ironic | 07:44 | |
| *** achanda has quit IRC | 07:44 | |
| *** mgoddard has joined #openstack-ironic | 07:45 | |
| *** JieLi has joined #openstack-ironic | 07:46 | |
| *** Nisha has joined #openstack-ironic | 07:48 | |
| *** dtantsur|afk is now known as dtantsur | 07:49 | |
| dtantsur | Morning Ironic | 07:49 |
| Haomeng | dtantsur: morning:) | 07:50 |
| dtantsur | mrda, you meant properties/local_gb? | 07:50 |
| *** mbound has joined #openstack-ironic | 07:54 | |
| *** ohamada has joined #openstack-ironic | 07:54 | |
| mrda | dtantsur: yes | 07:54 |
| dtantsur | then it should work.. | 07:55 |
| mrda | Worked it out, it was reserved_host_disk_mb that was causing me grief | 07:55 |
| openstackgerrit | Haomeng,Wang proposed openstack/ironic-specs: Cloning bare metal disk https://review.openstack.org/301498 | 07:55 |
| mrda | by default, set to 2, causing a misalignment between flavor and the node's properties | 07:55 |
| * dtantsur has never heard about it | 07:56 | |
| mrda | Hard to find, easy to correct :) | 07:56 |
| openstackgerrit | Haomeng,Wang proposed openstack/ironic-specs: WIP: Cloning bare metal disk https://review.openstack.org/301498 | 07:56 |
| mrda | but thankyou dtantsur for answering! | 07:56 |
| dtantsur | :) | 07:56 |
| * mrda now fights swift_tmp_urls :) | 07:57 | |
| * mrda leaves that fight until tomorrow. | 07:58 | |
| *** mbound has quit IRC | 08:00 | |
| *** jistr has quit IRC | 08:01 | |
| *** jistr has joined #openstack-ironic | 08:01 | |
| *** alexpilotti has joined #openstack-ironic | 08:02 | |
| *** mbound has joined #openstack-ironic | 08:06 | |
| openstackgerrit | Anton Arefiev proposed openstack/ironic-inspector: Tempest plugin initial commit https://review.openstack.org/293925 | 08:15 |
| openstackgerrit | Anton Arefiev proposed openstack/ironic-inspector: Move unit tests to "unit" directory https://review.openstack.org/301507 | 08:15 |
| dtantsur | aarefiev, zaqar has zaqar.tests.tempest_plugin: https://github.com/openstack/zaqar/tree/master/zaqar/tests/tempest_plugin | 08:19 |
| dtantsur | maybe we should do something like that as well... | 08:19 |
| openstackgerrit | Haomeng,Wang proposed openstack/ironic-specs: WIP: Cloning bare metal disk https://review.openstack.org/301498 | 08:19 |
| dtantsur | also g'morning aarefiev :) | 08:19 |
| *** dkehn has quit IRC | 08:20 | |
| alineb | Morning all | 08:22 |
| aarefiev | dtantsur: morning, yeah tempest_plugin is fine, I don't like it because test will be like tempest_plugin.tests.scenario.basic_test | 08:22 |
| aarefiev | but maybe it's fine | 08:23 |
| *** dkehn has joined #openstack-ironic | 08:27 | |
| dtantsur | aarefiev, so, does tempest strip the prefix ironic_inspector.test? | 08:27 |
| * dtantsur has no clues how it works | 08:27 | |
| dtantsur | morning alineb | 08:27 |
| aarefiev | dtantsur: no, in your example path was test.tempest.*, so it will try to import module tempest | 08:29 |
| *** Marga_ has joined #openstack-ironic | 08:29 | |
| aarefiev | test.tempest_plugin.* will work | 08:29 |
| *** kromanenko has joined #openstack-ironic | 08:30 | |
| *** kromanenko_ has joined #openstack-ironic | 08:30 | |
| *** daemontool has joined #openstack-ironic | 08:31 | |
| aarefiev | names overlap, no more | 08:31 |
| dtantsur | aarefiev, so has does the import work there? do they use kind of a relative import | 08:33 |
| dtantsur | I wonder how ironic_inspector.test.tempest may overlap with just tempest | 08:33 |
| *** Marga_ has quit IRC | 08:34 | |
| *** chlong has quit IRC | 08:35 | |
| aarefiev | dtantsur: I guess they import plugin module itself, which is tempest in this case. I have tried this case few patches ago | 08:36 |
| *** e0ne has joined #openstack-ironic | 08:37 | |
| aarefiev | or as path for plugin: tempest.plugin | 08:37 |
| *** derekh has joined #openstack-ironic | 08:39 | |
| *** dkehn has quit IRC | 08:40 | |
| dtantsur | meh.... | 08:40 |
| *** electrofelix has joined #openstack-ironic | 08:40 | |
| aarefiev | dtantsur: yeah, so you vote for tempest_plugin? | 08:41 |
| dtantsur | aarefiev, ok, feel free to leave it as it is then. I don't want to spend weeks on naming :) | 08:41 |
| dtantsur | whatever, I'm fine with inspector_tempest_plugin then | 08:41 |
| aarefiev | dtantsur: ok, lets keep, looks like it would not big problem to rename, it affect only plugin code | 08:42 |
| *** dkehn has joined #openstack-ironic | 08:46 | |
| aarefiev | looks like it is possible conflicts between plugins | 08:46 |
| *** dtantsur is now known as dtantsur|brb | 08:49 | |
| *** Marga_ has joined #openstack-ironic | 08:56 | |
| *** JieLi has quit IRC | 09:04 | |
| *** Marga_ has quit IRC | 09:12 | |
| openstackgerrit | Aline Bousquet proposed openstack/ironic-lib: Add support for metrics https://review.openstack.org/301526 | 09:12 |
| *** achanda has joined #openstack-ironic | 09:13 | |
| *** mkovacik has joined #openstack-ironic | 09:16 | |
| *** achanda has quit IRC | 09:18 | |
| *** kromanenko_ has quit IRC | 09:21 | |
| *** kromanenko has quit IRC | 09:22 | |
| *** smoriya_ has quit IRC | 09:24 | |
| sambetts|afk | Morning all | 09:33 |
| *** sambetts|afk is now known as sambetts | 09:34 | |
| *** Lisa_ has joined #openstack-ironic | 09:53 | |
| Lisa_ | Ironic conductor is not started in devstack environment | 09:55 |
| Lisa_ | DriverNotFound: Could not find the following driver(s): flat. | 09:56 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add nodes tagging support - objs https://review.openstack.org/233357 | 09:57 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add nodes tagging support - API https://review.openstack.org/250478 | 09:57 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Joined 'tags' column while getting node https://review.openstack.org/253065 | 09:57 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add DB layer support for node tags filter https://review.openstack.org/272624 | 09:57 |
| openstackgerrit | Sofiia Andriichenko proposed openstack/python-ironicclient: Add test for 'node-show --field' command https://review.openstack.org/300064 | 09:57 |
| sambetts | Lisa_: are you testing the Ironic neutron intergration ? | 09:58 |
| Lisa_ | yes. i am new bie to ironic. | 09:58 |
| Lisa_ | gonna test ironic intergaration with neutron | 09:59 |
| sambetts | Lisa_: have you downloaded the latest in the intergration tree? https://review.openstack.org/#/c/269157 | 10:00 |
| Lisa_ | let me check, sam | 10:01 |
| sambetts | Lisa_: also once you've downloaded the intergration code did you make sure that you reinstalled Ironic so that the changes are all picked up | 10:01 |
| *** mgould has joined #openstack-ironic | 10:01 | |
| Lisa_ | i install a week back | 10:04 |
| mgould | morning Ironicers! | 10:04 |
| Lisa_ | so pulled all the changes using git pull from the link https://blueprints.launchpad.net/ironic/+spec/ironic-ml2-integration | 10:04 |
| sambetts | ohai mgould | 10:04 |
| mgould | hi sambetts | 10:04 |
| sambetts | Lisa_: then you reinstalled the package, e.g. pip install --upgrade /opt/stack/ironic | 10:05 |
| *** karimb has quit IRC | 10:11 | |
| *** karimb_ has joined #openstack-ironic | 10:11 | |
| Lisa_ | thanks sam. reinstall the package and come back | 10:12 |
| *** electrofelix has quit IRC | 10:14 | |
| *** electrofelix has joined #openstack-ironic | 10:14 | |
| mkovacik | sambetts, hi, could you please review the reapply patch on inspector if you have some spare time? https://review.openstack.org/#/c/267637/ thanks! | 10:15 |
| dtantsur|brb | sambetts, Lisa_, better use -e flag with pip install | 10:15 |
| sambetts | mkovacik: I'll add to my list :) | 10:15 |
| dtantsur|brb | then you won't have to reinstall it each time | 10:16 |
| dtantsur|brb | ofc don't forget to restart ironic | 10:16 |
| *** dtantsur|brb is now known as dtantsur | 10:16 | |
| mkovacik | sambetts, ok cool | 10:16 |
| sambetts | dtantsur|brb: new setup.cfg endpoints don't get detected by -e :( | 10:16 |
| dtantsur | sambetts, yep. still it's better to reinstall with -e flag | 10:16 |
| dtantsur | cause code changes are detected | 10:17 |
| sambetts | ++ | 10:17 |
| *** hoangcx has quit IRC | 10:18 | |
| *** hoangcx has joined #openstack-ironic | 10:19 | |
| vdrok | morning ironic, Haomeng dtantsur sambetts and mkovacik | 10:21 |
| dtantsur | o/ | 10:21 |
| sambetts | morning vdrok | 10:21 |
| mkovacik | vdrok, o/ | 10:21 |
| mgould | morning mkovacik dtantsur vdrok Lisa_ | 10:23 |
| mkovacik | mgould, o/ | 10:23 |
| vdrok | morning mgould | 10:23 |
| mkovacik | morning all! :) | 10:23 |
| * mgould is confused about the logic of Depends-On | 10:27 | |
| mgould | neither https://review.openstack.org/#/c/301140/ nor https://review.openstack.org/#/c/272206/ will pass CI without the other, AFAICT | 10:27 |
| mgould | can I make them both Depend-On each other? | 10:28 |
| sambetts | mgould: No the depends-on is a one way relationship, and in the gate it will test them together | 10:29 |
| mgould | OK | 10:29 |
| mgould | so currently I have a D-O relationship from 301140 -> 272206 | 10:30 |
| sambetts | yup | 10:30 |
| mgould | does that mean that when testing 272206 the gate won't check out 301140, but when testing 301140 it will check out 272206? | 10:30 |
| sambetts | I believe so | 10:30 |
| dtantsur | mgould, it's just a bug in one of the patches, lemme leave a comment :) | 10:31 |
| mgould | oh, excellent | 10:31 |
| *** karimb_ has quit IRC | 10:32 | |
| *** arturas has joined #openstack-ironic | 10:34 | |
| dtantsur | mgould, commented | 10:34 |
| vmud213 | morning ironicers | 10:34 |
| sambetts | Morning vmud213 | 10:37 |
| *** cameron_C has quit IRC | 10:37 | |
| vmud213 | sambetts o/ | 10:38 |
| *** cameron_C has joined #openstack-ironic | 10:39 | |
| mgould | dtantsur, thanks | 10:41 |
| openstackgerrit | Merged openstack/ironic: Fix NamedTemporaryFile() OSError Exception https://review.openstack.org/293762 | 10:41 |
| mgould | morning vmud213 | 10:41 |
| *** karimb has joined #openstack-ironic | 10:41 | |
| *** electrofelix has quit IRC | 10:42 | |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Tests for testing chassis-create command https://review.openstack.org/293634 | 10:42 |
| vmud213 | morning mgould | 10:43 |
| *** wagiel has joined #openstack-ironic | 10:45 | |
| wagiel | Hello I have been googling around but could not find the right path trying to automate ESXi deployment with ironic | 10:46 |
| wagiel | is it possible? | 10:46 |
| *** electrofelix has joined #openstack-ironic | 10:47 | |
| sambetts | wagiel: Ironic is for baremetal provisioning of whole baremetal servers, OpenStack Nova is for virtual machines | 10:47 |
| wagiel | I'd like to deploy ESXi on baremetal | 10:47 |
| sambetts | wagiel: Ohh got it, Ironic works more or less the same as deploying a virtual machine, you need to create an image containing the required software then nova boot it | 10:49 |
| sambetts | See http://docs.openstack.org/developer/ironic/deploy/install-guide.html for how to install Ironic into an OpenStack environment | 10:50 |
| wagiel | I have the Ironic up and running, have successfully deployed some images already | 10:51 |
| wagiel | now wondering how to get to the ESXi hypervisor | 10:51 |
| wagiel | i have the iso image of course | 10:51 |
| wagiel | I can surely extract iso image and have it boot it the standard iPXE way, but this is all manual work, now how to automate that with Ironic | 10:53 |
| wagiel | maybe some dummy drivers or something | 10:53 |
| sambetts | wagiel: So one way is to create an image of an already installed ESXi, one way you can do this is to use something like virtualbox to boot the iso and install it, then you need to take that virtual machine image you've created and upload it to glance then nova boot using that image | 10:55 |
| wagiel | that is the theory :) I was googling for the ESXi qcow2 or whatever image couldnt find anything interesting | 10:56 |
| wagiel | wondering if this is possible | 10:57 |
| sambetts | wagiel: I'm guessing people can't share them if they've made them for license reasons perhaps, I found this though https://www.linkedin.com/pulse/install-vmware-esxi-oracle-virtualbox-learn-chase-hammock | 10:58 |
| wagiel | and also I would rather deploy ESXi iso with iPXE with kickstart and then adjust the config with ansible if needed | 10:58 |
| sambetts | wagiel: unfortunatly thats not how Ironic operates | 10:59 |
| wagiel | installing ESXi on virtualbox seems little crazy to me but maybe I am mistaken | 10:59 |
| wagiel | Ok at least now I now this patch is closed | 10:59 |
| wagiel | Ok so I need image | 11:00 |
| wagiel | would that image be customizable during the deployment? | 11:00 |
| wagiel | simple-init | 11:00 |
| wagiel | cloud-init | 11:01 |
| dtantsur | wagiel, iirc tripleo is using puppet triggered by cloud-init to install everything | 11:01 |
| sambetts | Yes, ironic supports the normal nova cloud-init customisation through the config-drive | 11:01 |
| dtantsur | yeah. and cloud-init can work without config drive, if you have a metadata server running (both heat and nova provide one iirc) | 11:03 |
| sambetts | wagiel: virtualbox is just one way to create the image, there are other ways you could do it, such as deploying it on a real machine then DDing the harddrive to a image file e.g. dd if=/dev/sda of=/path/to/images/sda.img bs=1024 | 11:05 |
| wagiel | Ok i will try that | 11:09 |
| wagiel | thanks for sharing | 11:09 |
| sambetts | :) | 11:09 |
| *** Marga_ has joined #openstack-ironic | 11:12 | |
| *** achanda has joined #openstack-ironic | 11:13 | |
| TheJulia | good morning! | 11:13 |
| sambetts | Moring TheJulia | 11:13 |
| sambetts | Morning* | 11:13 |
| vdrok | good morning TheJulia | 11:14 |
| *** achanda has quit IRC | 11:15 | |
| mgould | morning TheJulia | 11:16 |
| *** Marga_ has quit IRC | 11:17 | |
| vmud213 | morning TheJulia | 11:19 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Forbid to remove node name for api versions < 1.5 https://review.openstack.org/300983 | 11:21 |
| *** b_bezak has joined #openstack-ironic | 11:22 | |
| *** mbound has quit IRC | 11:24 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic: Don't power off non-deploying iLO nodes in takeover https://review.openstack.org/299573 | 11:26 |
| Nisha | Hi All, I reproposed the spec https://review.openstack.org/201904 to Newton. however i am not sure how i can make docs gate to pass | 11:30 |
| mgould | dtantsur, just fixed the things you commented on (modulo one point of disagreement, detailed in comments) | 11:32 |
| jroll | devananda: I don't think the 'links' stuff is terribly useful, but I don't think it's extraneous either. re: FlatNetwork, this is with the latest code? it should be loaded at startup... https://review.openstack.org/#/c/285852/31/ironic/common/driver_factory.py | 11:33 |
| jroll | morning everyone | 11:33 |
| TheJulia | vdrok: your aise a very good question on popping the ilo_boot_iso var from instance info, but I don't think it should even be there in the first place... | 11:33 |
| TheJulia | good morning jroll | 11:33 |
| vdrok | morning jroll | 11:33 |
| mgould | morning jroll | 11:34 |
| vdrok | TheJulia: but what if a node is being migrated from another ironic deployment? | 11:34 |
| TheJulia | vdrok: I would think of it as driver_info | 11:34 |
| TheJulia | vdrok: the prepare_instance() call which I'm not touching bases everything on what the configured glance image has, not a user definable variable | 11:35 |
| *** trown|outtypewww is now known as trown | 11:35 | |
| TheJulia | vdrok: granted, it mgiht be good to move it down to after that, it would at least one extra guard when it comes to adoption because if configuration is not right, *kaboom* | 11:35 |
| vdrok | I mean that deployment constructed ilo_boot_iso from glance images, stored it on conductor and written that to instance_info | 11:35 |
| *** thrash|g0ne is now known as thrash | 11:36 | |
| vdrok | and then the node is being adopted | 11:36 |
| jroll | TheJulia: are you doing bifrost 1.0.1 or do you want me to handle it quickly? (http://lists.openstack.org/pipermail/openstack-dev/2016-April/091241.html) | 11:36 |
| vdrok | ilo_boot_iso is still present, node tries to boot and fails | 11:36 |
| TheJulia | vdrok: ahh, and what I'm saying is that is disregarded _when_ the other path that would be valid is taken | 11:36 |
| TheJulia | jroll: if you have it up in front of you, please, if not I can brian context switchin about 15 minutes | 11:37 |
| jroll | yeah np | 11:37 |
| TheJulia | vdrok: I'll move it, just seems like extranious data to me :) | 11:38 |
| *** mbound has joined #openstack-ironic | 11:39 | |
| vdrok | TheJulia: I think I didn't get what you meant by "and what I'm saying is that is disregarded _when_ the other path that would be valid is taken" :) | 11:39 |
| *** mbound has quit IRC | 11:39 | |
| *** mbound has joined #openstack-ironic | 11:40 | |
| *** mbound has quit IRC | 11:40 | |
| jroll | TheJulia: fyi https://review.openstack.org/301596 | 11:40 |
| TheJulia | jroll: thank you, i went looking for a reply on Sunday and didn't see it :( Maybe it is time I start using a real email client instead of the gmail web interface for the mailing list :( | 11:40 |
| jroll | heh, np | 11:41 |
| vdrok | TheJulia: I was referring to this possibility - https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/ilo/boot.py#L148-L192, it is a case when user image is in glance and has kernel and ramdisk, but not boot_iso properties. then it is constructed, stored on conductor and writtern to instance_info IIRC | 11:47 |
| TheJulia | vdrok: so, popping it would be bad then, since the path into that is prepare_instance not prepare_ramdisk, but both ultimately call it | 11:51 |
| TheJulia | at least I think | 11:51 |
| * TheJulia goes and pulls the bug up | 11:51 | |
| *** raildo-afk is now known as raildo | 11:54 | |
| TheJulia | so non agent drivers in non-active state will call prepare_ramdisk | 11:54 |
| *** dprince has joined #openstack-ironic | 11:57 | |
| *** jaypipes has joined #openstack-ironic | 11:58 | |
| dtantsur | morning jroll, TheJulia | 11:58 |
| jroll | morning dtantsur \o | 12:00 |
| * jroll just found https://github.com/guyzmo/vim-etherpad | 12:00 | |
| vdrok | TheJulia: right now looking at the code, it is popped out only when image_source is from glance, meaning that ilo_boot_iso either contains the value from boot_iso image property, or constructed one from kernel and ramdisk, which seems to be ok, but I might be missing something. I'll try to dig into this in a bit :) | 12:02 |
| TheJulia | vdrok: please, Ideally, in my mind, we don't want to do anything that modifies a running instance when it does not make sense, including records of | 12:04 |
| *** kalpase has left #openstack-ironic | 12:04 | |
| TheJulia | vdrok: in the case of prepare_instance, that kind of makes sense to leave it as-is... as much as I don't like it... because in both the ilo and irmc drivers, they expect that a user will be booting via virtual media if local boot is not defined | 12:04 |
| *** electrofelix has quit IRC | 12:08 | |
| *** baoli has joined #openstack-ironic | 12:09 | |
| *** mbound has joined #openstack-ironic | 12:11 | |
| *** cfarquhar has quit IRC | 12:30 | |
| *** cfarquhar has joined #openstack-ironic | 12:30 | |
| *** cfarquhar has quit IRC | 12:30 | |
| *** cfarquhar has joined #openstack-ironic | 12:30 | |
| *** mvpnitesh has joined #openstack-ironic | 12:31 | |
| mvpnitesh | Hi All | 12:31 |
| mvpnitesh | Good evening | 12:31 |
| TheJulia | good morning | 12:32 |
| jroll | is there anything written up on what we want to do in the future in terms of hardware metrics? I feel like there was but can't find it | 12:33 |
| jroll | dtantsur: ^ | 12:33 |
| sambetts | jroll: as in reading hw sensors etc? | 12:34 |
| * dtantsur does not remember | 12:34 | |
| *** mtanino has joined #openstack-ironic | 12:34 | |
| mvpnitesh | I've a query. I've provisioned a bare metal machine using ironic. The image with which we booted the bare meatal will be the /var/lib/ironic/master_images/ .When can be it deleted | 12:34 |
| jroll | sambetts: yeah, thought we were doing some general work there, I may be wrong | 12:34 |
| jroll | mvpnitesh: ironic will clean it up when it's no longer in use, I believe | 12:34 |
| sambetts | jroll: there is a spec somewhere | 12:35 |
| mvpnitesh | @jroll Thanks.I've a doubt should will the image still exist after delete the node? | 12:36 |
| sambetts | jroll: ironic-specs/specs/backlog/support-per-driver-sensor-meters.rst | 12:37 |
| jroll | mvpnitesh: yes, it will exist for some time, I believe it is a periodic task that will clean it up, but it may be based on free disk space, don't remember right away | 12:37 |
| dtantsur | it based on both time and free space | 12:38 |
| jroll | sambetts: nice, thank you | 12:38 |
| dtantsur | we do cache images for some time, cause deployment is likely to reuse the same images pretty often | 12:38 |
| dtantsur | e.g. the whole tripleo uses only the same deploy and the same instance image for everything | 12:38 |
| dtantsur | this cache can be tweaked via the configuration file though | 12:38 |
| dtantsur | mvpnitesh, ^^ | 12:38 |
| mvpnitesh | dtantsur: Thanks for clarifying my query | 12:39 |
| *** thiagop has joined #openstack-ironic | 12:39 | |
| thiagop | Good morning, Ironic | 12:39 |
| mgould | thiagop, morning! | 12:39 |
| TheJulia | good morning thiagop | 12:40 |
| *** baoli_ has joined #openstack-ironic | 12:40 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Forbid to remove node name for api versions < 1.5 https://review.openstack.org/300983 | 12:41 |
| vdrok | morning thiagop | 12:42 |
| *** baoli has quit IRC | 12:43 | |
| thiagop | dtantsur: around? | 12:45 |
| dtantsur | thiagop, morning, I am | 12:45 |
| thiagop | dtantsur: hiya, I have a quick question about the futurist implementation. The decorator of periodic tasks deprecated that 'periodic_interval' value on the ironic.conf. Are supposed to create a variable for spacing on periodic tasks for each specific (group of) tasks or it should be a global parameter? | 12:47 |
| thiagop | dtantsur: Context: I saw that inspector have a spacing of its own. We are creating some tasks for the OneView driver and I would know if I'm supposed to create a configuration option inside [oneview] section | 12:49 |
| dtantsur | thiagop, each task should have its own spacing, so yeah, you need your option | 12:50 |
| *** wagiel has quit IRC | 12:50 | |
| thiagop | dtantsur: nice, ty | 12:50 |
| openstackgerrit | Thiago Paiva Brito proposed openstack/ironic-specs: Dynamic allocation of nodes on OneView drivers https://review.openstack.org/275726 | 12:53 |
| *** mat128 has joined #openstack-ironic | 12:55 | |
| thiagop | Feedbacks on ^^^ would be most welcome :) | 12:55 |
| *** cdearborn has joined #openstack-ironic | 12:57 | |
| mat128 | thiagop: doesnt it overlaps with https://bugs.launchpad.net/ironic/+bug/1563694 ? | 12:59 |
| openstack | Launchpad bug 1563694 in Ironic "Can create/update node with name "" in any api version" [High,In progress] - Assigned to Zhenguo Niu (niu-zglinux) | 12:59 |
| *** Lisa_ has quit IRC | 12:59 | |
| thiagop | mat128: don't see how. Do you see any problem? | 13:00 |
| *** igordcard has joined #openstack-ironic | 13:01 | |
| *** igordcard has quit IRC | 13:02 | |
| mat128 | first is "delete is possible while it shouldnt" and second is add/update | 13:02 |
| *** igordcard has joined #openstack-ironic | 13:03 | |
| mat128 | maybe we could kill two birds with one stone? | 13:03 |
| mat128 | in any case I dont disagree, just pointing out similarities | 13:03 |
| *** b_bezak has left #openstack-ironic | 13:03 | |
| *** b_bezak_ has joined #openstack-ironic | 13:03 | |
| *** moshele has quit IRC | 13:03 | |
| thiagop | mat128: are you talking about the Dynamic allocation spec or vdrok's patch? | 13:04 |
| mat128 | Completely messed up on the reviews :) | 13:04 |
| mat128 | the latter, sorry for the noise | 13:04 |
| *** moshele has joined #openstack-ironic | 13:04 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Wait for the interfaces to get IP addresses before inspection https://review.openstack.org/300548 | 13:05 |
| dtantsur | jroll, reworked ^^^ | 13:05 |
| thiagop | mat128: no problem, it happens. But it's just tuesday, get some coffee | 13:05 |
| thiagop | :D | 13:05 |
| *** Nisha has quit IRC | 13:05 | |
| jroll | dtantsur: +2 | 13:09 |
| jroll | (with nits) | 13:09 |
| jroll | curious why u no loopingcall, but /shrug | 13:09 |
| dtantsur | probably I just don't know it too well :) does it have any benefits over just looping? | 13:10 |
| dtantsur | also thanks for review, will fix the nits, as it's part of release notes | 13:10 |
| *** wendar_ has joined #openstack-ironic | 13:14 | |
| *** b_bezak_ is now known as b_bezak | 13:15 | |
| *** b_bezak has quit IRC | 13:19 | |
| *** dutsmoc has joined #openstack-ironic | 13:20 | |
| *** mgagne_ has joined #openstack-ironic | 13:20 | |
| *** cfarquhar has quit IRC | 13:21 | |
| *** arturas has quit IRC | 13:21 | |
| *** krtaylor has quit IRC | 13:21 | |
| *** zzzeek has quit IRC | 13:21 | |
| *** chihhsin has quit IRC | 13:21 | |
| *** mgagne has quit IRC | 13:21 | |
| *** wendar has quit IRC | 13:21 | |
| *** comstud has quit IRC | 13:21 | |
| *** eglute has quit IRC | 13:21 | |
| *** mtanino has quit IRC | 13:21 | |
| *** karimb has quit IRC | 13:21 | |
| *** jcoufal has quit IRC | 13:21 | |
| *** zer0c00l has quit IRC | 13:21 | |
| *** marios has quit IRC | 13:21 | |
| *** adreznec has quit IRC | 13:21 | |
| *** pleia2 has quit IRC | 13:21 | |
| *** rbradfor has quit IRC | 13:21 | |
| *** bcornec has quit IRC | 13:21 | |
| *** hparekh has quit IRC | 13:21 | |
| *** afaranha has quit IRC | 13:21 | |
| *** shu-mutou has quit IRC | 13:21 | |
| *** igordcard has quit IRC | 13:21 | |
| *** daemontool has quit IRC | 13:21 | |
| *** jayc has quit IRC | 13:21 | |
| *** amit213 has quit IRC | 13:21 | |
| *** mtreinish has quit IRC | 13:21 | |
| *** yuanying has quit IRC | 13:21 | |
| *** marlinc has quit IRC | 13:21 | |
| *** mkoderer has quit IRC | 13:21 | |
| *** NobodyCam has quit IRC | 13:21 | |
| *** persia has quit IRC | 13:21 | |
| *** alaski has quit IRC | 13:21 | |
| *** yuriyz has quit IRC | 13:21 | |
| *** sirushti has quit IRC | 13:21 | |
| *** mbound has quit IRC | 13:21 | |
| *** alexpilotti has quit IRC | 13:21 | |
| *** ohamada has quit IRC | 13:21 | |
| *** ifarkas has quit IRC | 13:21 | |
| *** rcernin has quit IRC | 13:21 | |
| *** vishwanathj has quit IRC | 13:21 | |
| *** harlowja has quit IRC | 13:21 | |
| *** thrash has quit IRC | 13:21 | |
| *** raginbajin has quit IRC | 13:21 | |
| *** sturivnyi has quit IRC | 13:21 | |
| *** odyssey4me has quit IRC | 13:21 | |
| *** smoriya has quit IRC | 13:21 | |
| *** tonyb has quit IRC | 13:21 | |
| *** konetzed has quit IRC | 13:21 | |
| *** moshele has quit IRC | 13:21 | |
| *** e0ne has quit IRC | 13:21 | |
| *** vsaienko has quit IRC | 13:21 | |
| *** sergek has quit IRC | 13:21 | |
| *** BadCub has quit IRC | 13:21 | |
| *** romcheg has quit IRC | 13:21 | |
| *** alineb has quit IRC | 13:21 | |
| *** mariojv has quit IRC | 13:21 | |
| *** yarkot has quit IRC | 13:21 | |
| *** mikal has quit IRC | 13:21 | |
| *** zhenguo_ has quit IRC | 13:21 | |
| *** bradjones has quit IRC | 13:21 | |
| *** johnthetubaguy has quit IRC | 13:21 | |
| *** jroll has quit IRC | 13:21 | |
| *** trown has quit IRC | 13:21 | |
| *** TheJulia has quit IRC | 13:21 | |
| *** EmilienM has quit IRC | 13:21 | |
| *** agordeev has quit IRC | 13:21 | |
| *** sylwesterB has quit IRC | 13:21 | |
| *** cppforlife_ has quit IRC | 13:21 | |
| *** clif_h has quit IRC | 13:21 | |
| *** mgoddard has quit IRC | 13:21 | |
| *** lintan has quit IRC | 13:21 | |
| *** ptx0 has quit IRC | 13:21 | |
| *** jtomasek has quit IRC | 13:21 | |
| *** bapalm has quit IRC | 13:21 | |
| *** maurosr has quit IRC | 13:21 | |
| *** logan- has quit IRC | 13:21 | |
| *** kbyrne has quit IRC | 13:21 | |
| *** krotscheck has quit IRC | 13:21 | |
| *** adam_g has quit IRC | 13:21 | |
| *** phschwartz has quit IRC | 13:21 | |
| *** Nakato has quit IRC | 13:21 | |
| *** kozhukalov has quit IRC | 13:21 | |
| *** intr1nsic has quit IRC | 13:21 | |
| *** mrda has quit IRC | 13:21 | |
| *** izaakk has quit IRC | 13:21 | |
| *** _fortis has quit IRC | 13:21 | |
| *** jmccrory has quit IRC | 13:21 | |
| *** natorious has quit IRC | 13:21 | |
| *** mvpnitesh has quit IRC | 13:21 | |
| *** dprince has quit IRC | 13:21 | |
| *** pcaruana has quit IRC | 13:21 | |
| *** ppiela has quit IRC | 13:21 | |
| *** sinval-afk has quit IRC | 13:21 | |
| *** openstackgerrit has quit IRC | 13:21 | |
| *** alex_xu has quit IRC | 13:21 | |
| *** Fdaisuke has quit IRC | 13:21 | |
| *** yolanda has quit IRC | 13:21 | |
| *** dims has quit IRC | 13:21 | |
| *** early has quit IRC | 13:21 | |
| *** sivarama- has quit IRC | 13:21 | |
| *** lazy_prince has quit IRC | 13:21 | |
| *** keekz has quit IRC | 13:21 | |
| *** mag009_ has quit IRC | 13:21 | |
| *** mordred has quit IRC | 13:21 | |
| *** dansmith has quit IRC | 13:21 | |
| *** kwazar has quit IRC | 13:21 | |
| *** HenryG has quit IRC | 13:21 | |
| *** cdearborn has quit IRC | 13:21 | |
| *** ndipanov has quit IRC | 13:21 | |
| *** mkoderer__ has quit IRC | 13:21 | |
| *** raildo has quit IRC | 13:21 | |
| *** stevebaker has quit IRC | 13:21 | |
| *** SpamapS has quit IRC | 13:21 | |
| *** aarefiev has quit IRC | 13:21 | |
| *** zigo has quit IRC | 13:21 | |
| *** purp has quit IRC | 13:21 | |
| *** sekrit has quit IRC | 13:21 | |
| *** baoli_ has quit IRC | 13:21 | |
| *** mgould has quit IRC | 13:21 | |
| *** mkovacik has quit IRC | 13:21 | |
| *** david-lyle has quit IRC | 13:21 | |
| *** lifeless has quit IRC | 13:22 | |
| *** lynxman has quit IRC | 13:22 | |
| *** gabriel-bezerra has quit IRC | 13:22 | |
| *** leifmadsen has quit IRC | 13:22 | |
| *** stevemar has quit IRC | 13:22 | |
| *** serverascode has quit IRC | 13:22 | |
| *** JoshNang has quit IRC | 13:22 | |
| *** Madasi has quit IRC | 13:22 | |
| *** yhvh has quit IRC | 13:22 | |
| *** fyxim_ has quit IRC | 13:22 | |
| *** ChrisMead has quit IRC | 13:22 | |
| *** jasondotstar has quit IRC | 13:22 | |
| *** mfisch has quit IRC | 13:22 | |
| *** dtantsur has quit IRC | 13:22 | |
| *** JayF has quit IRC | 13:22 | |
| *** vdrok has quit IRC | 13:22 | |
| *** joemonk has quit IRC | 13:22 | |
| *** boris-42 has quit IRC | 13:22 | |
| *** anush has quit IRC | 13:22 | |
| *** lekha has quit IRC | 13:22 | |
| *** cinerama has quit IRC | 13:22 | |
| *** jistr has quit IRC | 13:22 | |
| *** gmmaha has quit IRC | 13:22 | |
| *** lucasagomes has quit IRC | 13:22 | |
| *** bnemec has quit IRC | 13:22 | |
| *** m3m0 has quit IRC | 13:22 | |
| *** anteaya has quit IRC | 13:22 | |
| *** lennyb has quit IRC | 13:22 | |
| *** crinkle has quit IRC | 13:22 | |
| *** ionutbalutoiu has quit IRC | 13:22 | |
| *** morgan has quit IRC | 13:22 | |
| *** morgabra has quit IRC | 13:22 | |
| *** aignatov has quit IRC | 13:22 | |
| *** pshige______ has quit IRC | 13:22 | |
| *** robcresswell has quit IRC | 13:22 | |
| *** jaypipes has quit IRC | 13:22 | |
| *** hoangcx has quit IRC | 13:22 | |
| *** Goneri has quit IRC | 13:22 | |
| *** davideagnello has quit IRC | 13:22 | |
| *** greghaynes has quit IRC | 13:22 | |
| *** athomas has quit IRC | 13:22 | |
| *** bigjools has quit IRC | 13:22 | |
| *** xek has quit IRC | 13:22 | |
| *** devananda has quit IRC | 13:22 | |
| *** sinh has quit IRC | 13:22 | |
| *** sambetts has quit IRC | 13:22 | |
| *** slagle has quit IRC | 13:22 | |
| *** thiagop has quit IRC | 13:22 | |
| *** cameron_C has quit IRC | 13:22 | |
| *** dkehn has quit IRC | 13:22 | |
| *** derekh has quit IRC | 13:22 | |
| *** vmud213 has quit IRC | 13:22 | |
| *** jrist has quit IRC | 13:22 | |
| *** Haomeng has quit IRC | 13:22 | |
| *** jlvillal has quit IRC | 13:22 | |
| *** jcook has quit IRC | 13:22 | |
| *** ekarlso- has quit IRC | 13:22 | |
| *** kragniz has quit IRC | 13:22 | |
| *** eglute has joined #openstack-ironic | 13:23 | |
| *** cfarquhar_ has joined #openstack-ironic | 13:24 | |
| *** zzzeek_ has joined #openstack-ironic | 13:24 | |
| *** rbudden has joined #openstack-ironic | 13:24 | |
| *** chihhsin_ has joined #openstack-ironic | 13:24 | |
| *** arturas_ has joined #openstack-ironic | 13:24 | |
| *** Marga_ has joined #openstack-ironic | 13:24 | |
| *** moshele has joined #openstack-ironic | 13:24 | |
| *** igordcard has joined #openstack-ironic | 13:24 | |
| *** cdearborn has joined #openstack-ironic | 13:24 | |
| *** baoli_ has joined #openstack-ironic | 13:24 | |
| *** thiagop has joined #openstack-ironic | 13:24 | |
| *** mvpnitesh has joined #openstack-ironic | 13:24 | |
| *** mbound has joined #openstack-ironic | 13:24 | |
| *** jaypipes has joined #openstack-ironic | 13:24 | |
| *** dprince has joined #openstack-ironic | 13:24 | |
| *** karimb has joined #openstack-ironic | 13:24 | |
| *** cameron_C has joined #openstack-ironic | 13:24 | |
| *** hoangcx has joined #openstack-ironic | 13:24 | |
| *** mgould has joined #openstack-ironic | 13:24 | |
| *** mkovacik has joined #openstack-ironic | 13:24 | |
| *** dkehn has joined #openstack-ironic | 13:24 | |
| *** derekh has joined #openstack-ironic | 13:24 | |
| *** e0ne has joined #openstack-ironic | 13:24 | |
| *** daemontool has joined #openstack-ironic | 13:24 | |
| *** alexpilotti has joined #openstack-ironic | 13:24 | |
| *** jistr has joined #openstack-ironic | 13:24 | |
| *** ohamada has joined #openstack-ironic | 13:24 | |
| *** mgoddard has joined #openstack-ironic | 13:24 | |
| *** ifarkas has joined #openstack-ironic | 13:24 | |
| *** pcaruana has joined #openstack-ironic | 13:24 | |
| *** ndipanov has joined #openstack-ironic | 13:24 | |
| *** rcernin has joined #openstack-ironic | 13:24 | |
| *** jcoufal has joined #openstack-ironic | 13:24 | |
| *** vmud213 has joined #openstack-ironic | 13:24 | |
| *** alex_xu has joined #openstack-ironic | 13:24 | |
| *** sekrit has joined #openstack-ironic | 13:24 | |
| *** mfisch has joined #openstack-ironic | 13:24 | |
| *** david-lyle has joined #openstack-ironic | 13:24 | |
| *** Goneri has joined #openstack-ironic | 13:24 | |
| *** jayc has joined #openstack-ironic | 13:24 | |
| *** lintan has joined #openstack-ironic | 13:24 | |
| *** ptx0 has joined #openstack-ironic | 13:24 | |
| *** zer0c00l has joined #openstack-ironic | 13:24 | |
| *** davideagnello has joined #openstack-ironic | 13:24 | |
| *** afaranha has joined #openstack-ironic | 13:24 | |
| *** greghaynes has joined #openstack-ironic | 13:24 | |
| *** anush has joined #openstack-ironic | 13:24 | |
| *** vishwanathj has joined #openstack-ironic | 13:24 | |
| *** ppiela has joined #openstack-ironic | 13:24 | |
| *** sinval-afk has joined #openstack-ironic | 13:24 | |
| *** openstackgerrit has joined #openstack-ironic | 13:24 | |
| *** marios has joined #openstack-ironic | 13:24 | |
| *** athomas has joined #openstack-ironic | 13:24 | |
| *** vsaienko has joined #openstack-ironic | 13:24 | |
| *** bigjools has joined #openstack-ironic | 13:24 | |
| *** adreznec has joined #openstack-ironic | 13:24 | |
| *** jtomasek has joined #openstack-ironic | 13:24 | |
| *** xek has joined #openstack-ironic | 13:24 | |
| *** shu-mutou has joined #openstack-ironic | 13:24 | |
| *** bapalm has joined #openstack-ironic | 13:24 | |
| *** Fdaisuke has joined #openstack-ironic | 13:24 | |
| *** amit213 has joined #openstack-ironic | 13:24 | |
| *** yolanda has joined #openstack-ironic | 13:24 | |
| *** stevemar has joined #openstack-ironic | 13:24 | |
| *** maurosr has joined #openstack-ironic | 13:24 | |
| *** jrist has joined #openstack-ironic | 13:24 | |
| *** logan- has joined #openstack-ironic | 13:24 | |
| *** morgan has joined #openstack-ironic | 13:24 | |
| *** harlowja has joined #openstack-ironic | 13:24 | |
| *** devananda has joined #openstack-ironic | 13:24 | |
| *** dims has joined #openstack-ironic | 13:24 | |
| *** kbyrne has joined #openstack-ironic | 13:24 | |
| *** thrash has joined #openstack-ironic | 13:24 | |
| *** mtreinish has joined #openstack-ironic | 13:24 | |
| *** yuanying has joined #openstack-ironic | 13:24 | |
| *** gmmaha has joined #openstack-ironic | 13:24 | |
| *** early has joined #openstack-ironic | 13:24 | |
| *** Haomeng has joined #openstack-ironic | 13:24 | |
| *** jlvillal has joined #openstack-ironic | 13:24 | |
| *** stevebaker has joined #openstack-ironic | 13:24 | |
| *** sivarama- has joined #openstack-ironic | 13:24 | |
| *** lazy_prince has joined #openstack-ironic | 13:24 | |
| *** serverascode has joined #openstack-ironic | 13:24 | |
| *** sergek has joined #openstack-ironic | 13:24 | |
| *** JoshNang has joined #openstack-ironic | 13:24 | |
| *** BadCub has joined #openstack-ironic | 13:24 | |
| *** krotscheck has joined #openstack-ironic | 13:24 | |
| *** keekz has joined #openstack-ironic | 13:24 | |
| *** marlinc has joined #openstack-ironic | 13:24 | |
| *** pleia2 has joined #openstack-ironic | 13:24 | |
| *** Madasi has joined #openstack-ironic | 13:24 | |
| *** lucasagomes has joined #openstack-ironic | 13:24 | |
| *** mkoderer__ has joined #openstack-ironic | 13:24 | |
| *** raginbajin has joined #openstack-ironic | 13:24 | |
| *** yhvh has joined #openstack-ironic | 13:24 | |
| *** fyxim_ has joined #openstack-ironic | 13:24 | |
| *** ChrisMead has joined #openstack-ironic | 13:24 | |
| *** romcheg has joined #openstack-ironic | 13:24 | |
| *** alineb has joined #openstack-ironic | 13:24 | |
| *** jasondotstar has joined #openstack-ironic | 13:24 | |
| *** rbradfor has joined #openstack-ironic | 13:24 | |
| *** sturivnyi has joined #openstack-ironic | 13:24 | |
| *** adam_g has joined #openstack-ironic | 13:24 | |
| *** sinh has joined #openstack-ironic | 13:24 | |
| *** phschwartz has joined #openstack-ironic | 13:24 | |
| *** Nakato has joined #openstack-ironic | 13:24 | |
| *** kozhukalov has joined #openstack-ironic | 13:24 | |
| *** sambetts has joined #openstack-ironic | 13:24 | |
| *** lifeless has joined #openstack-ironic | 13:24 | |
| *** mkoderer has joined #openstack-ironic | 13:24 | |
| *** NobodyCam has joined #openstack-ironic | 13:24 | |
| *** intr1nsic has joined #openstack-ironic | 13:24 | |
| *** bcornec has joined #openstack-ironic | 13:24 | |
| *** dtantsur has joined #openstack-ironic | 13:24 | |
| *** lynxman has joined #openstack-ironic | 13:24 | |
| *** persia has joined #openstack-ironic | 13:24 | |
| *** hparekh has joined #openstack-ironic | 13:24 | |
| *** mrda has joined #openstack-ironic | 13:24 | |
| *** odyssey4me has joined #openstack-ironic | 13:24 | |
| *** mag009_ has joined #openstack-ironic | 13:24 | |
| *** mariojv has joined #openstack-ironic | 13:24 | |
| *** alaski has joined #openstack-ironic | 13:24 | |
| *** JayF has joined #openstack-ironic | 13:24 | |
| *** vdrok has joined #openstack-ironic | 13:24 | |
| *** gabriel-bezerra has joined #openstack-ironic | 13:24 | |
| *** yuriyz has joined #openstack-ironic | 13:24 | |
| *** izaakk has joined #openstack-ironic | 13:24 | |
| *** mordred has joined #openstack-ironic | 13:24 | |
| *** jcook has joined #openstack-ironic | 13:24 | |
| *** dansmith has joined #openstack-ironic | 13:24 | |
| *** joemonk has joined #openstack-ironic | 13:24 | |
| *** m3m0 has joined #openstack-ironic | 13:24 | |
| *** ekarlso- has joined #openstack-ironic | 13:24 | |
| *** boris-42 has joined #openstack-ironic | 13:24 | |
| *** bnemec has joined #openstack-ironic | 13:24 | |
| *** yarkot has joined #openstack-ironic | 13:24 | |
| *** _fortis has joined #openstack-ironic | 13:24 | |
| *** jmccrory has joined #openstack-ironic | 13:24 | |
| *** mikal has joined #openstack-ironic | 13:24 | |
| *** zhenguo_ has joined #openstack-ironic | 13:24 | |
| *** raildo has joined #openstack-ironic | 13:24 | |
| *** bradjones has joined #openstack-ironic | 13:24 | |
| *** anteaya has joined #openstack-ironic | 13:24 | |
| *** johnthetubaguy has joined #openstack-ironic | 13:24 | |
| *** jroll has joined #openstack-ironic | 13:24 | |
| *** trown has joined #openstack-ironic | 13:24 | |
| *** TheJulia has joined #openstack-ironic | 13:24 | |
| *** sirushti has joined #openstack-ironic | 13:24 | |
| *** EmilienM has joined #openstack-ironic | 13:24 | |
| *** natorious has joined #openstack-ironic | 13:24 | |
| *** agordeev has joined #openstack-ironic | 13:24 | |
| *** sylwesterB has joined #openstack-ironic | 13:24 | |
| *** cppforlife_ has joined #openstack-ironic | 13:24 | |
| *** kragniz has joined #openstack-ironic | 13:24 | |
| *** kwazar has joined #openstack-ironic | 13:24 | |
| *** lekha has joined #openstack-ironic | 13:24 | |
| *** smoriya has joined #openstack-ironic | 13:24 | |
| *** slagle has joined #openstack-ironic | 13:24 | |
| *** HenryG has joined #openstack-ironic | 13:24 | |
| *** tonyb has joined #openstack-ironic | 13:24 | |
| *** lennyb has joined #openstack-ironic | 13:24 | |
| *** leifmadsen has joined #openstack-ironic | 13:24 | |
| *** konetzed has joined #openstack-ironic | 13:24 | |
| *** clif_h has joined #openstack-ironic | 13:24 | |
| *** SpamapS has joined #openstack-ironic | 13:24 | |
| *** aarefiev has joined #openstack-ironic | 13:24 | |
| *** crinkle has joined #openstack-ironic | 13:24 | |
| *** ionutbalutoiu has joined #openstack-ironic | 13:24 | |
| *** zigo has joined #openstack-ironic | 13:24 | |
| *** morgabra has joined #openstack-ironic | 13:24 | |
| *** aignatov has joined #openstack-ironic | 13:24 | |
| *** purp has joined #openstack-ironic | 13:24 | |
| *** pshige______ has joined #openstack-ironic | 13:24 | |
| *** cinerama has joined #openstack-ironic | 13:24 | |
| *** robcresswell has joined #openstack-ironic | 13:24 | |
| *** openstackstatus has quit IRC | 13:24 | |
| *** mgagne_ has quit IRC | 13:24 | |
| *** mgagne_ has joined #openstack-ironic | 13:24 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-specs: Add Port and Chassis logical name https://review.openstack.org/295082 | 13:24 |
| *** openstackstatus has joined #openstack-ironic | 13:26 | |
| *** ChanServ sets mode: +v openstackstatus | 13:26 | |
| *** krtaylor has joined #openstack-ironic | 13:28 | |
| *** xavierr has joined #openstack-ironic | 13:28 | |
| *** mgould has quit IRC | 13:29 | |
| *** xavierr has left #openstack-ironic | 13:30 | |
| openstackgerrit | Brad P. Crochet proposed openstack/python-ironicclient: Implementation of baremetal power state commands https://review.openstack.org/172517 | 13:31 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-ironicclient: Add maintenance mode commands https://review.openstack.org/216727 | 13:31 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-ironicclient: Implementation of baremetal port create https://review.openstack.org/172461 | 13:31 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-ironicclient: Bring OSC plugin inline with approved spec https://review.openstack.org/284160 | 13:31 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-ironicclient: Add provision state commands https://review.openstack.org/206119 | 13:31 |
| *** mgould has joined #openstack-ironic | 13:32 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-specs: Add Port and Chassis logical name https://review.openstack.org/295082 | 13:33 |
| *** links has quit IRC | 13:35 | |
| *** xavierr has joined #openstack-ironic | 13:35 | |
| xavierr | good morning | 13:37 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/301657 | 13:37 |
| *** ametts has joined #openstack-ironic | 13:42 | |
| thiagop | morning again xavierr | 13:42 |
| openstackgerrit | Gabriel Bezerra proposed openstack/ironic-specs: Dynamic allocation of nodes on OneView drivers https://review.openstack.org/275726 | 13:43 |
| *** mgoddard_ has joined #openstack-ironic | 13:45 | |
| xavierr | thiagop, morning | 13:47 |
| xavierr | guys, I'm implementing a periodic task for a driver, but I'm not sure the right place to put the periodic task | 13:48 |
| *** mgoddard has quit IRC | 13:48 | |
| TheJulia | xavierr: Could you elaborate a little more? | 13:49 |
| xavierr | I'm thinking the code could be on the driver implementation (that implements base.BaseDriver) but I think it could fit on some interface implementation like vendor interface | 13:49 |
| TheJulia | oh, there you go :) lag | 13:49 |
| xavierr | TheJulia, :) | 13:49 |
| sambetts | xavierr: they are normally places into which ever interface they align most with | 13:49 |
| openstackgerrit | Michael Krotscheck proposed openstack/ironic-webclient: Added maintenance and power filter dropdowns to node list https://review.openstack.org/301679 | 13:50 |
| sambetts | xavierr: An example is the inspector periodic https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/inspector.py#L125 | 13:52 |
| openstackgerrit | Michael Krotscheck proposed openstack/ironic-webclient: Added selection hover state to node list https://review.openstack.org/301681 | 13:53 |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Remove backward compatibility code with old Nova drivers https://review.openstack.org/301682 | 13:54 |
| *** mbound has quit IRC | 13:55 | |
| *** krtaylor has quit IRC | 13:56 | |
| *** mgoddard_ has quit IRC | 13:56 | |
| *** b_bezak has joined #openstack-ironic | 13:56 | |
| *** mgoddard has joined #openstack-ironic | 13:57 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for baremetal subcommands in openstackclient https://review.openstack.org/301687 | 13:57 |
| xavierr | sambetts, but in my case is really specific for the oneview driver | 13:59 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Correct api version check conditional for node.name https://review.openstack.org/299264 | 13:59 |
| sambetts | xavierr: right, so it needs to live in one of the interfaces you implement, and you need to decide which it fits into best, power, management, boot, deploy or vendor passthru | 14:01 |
| NobodyCam | good morning Ironicers :) | 14:01 |
| sambetts | Morning NobodyCam | 14:01 |
| NobodyCam | hey there sambetts :) | 14:02 |
| xavierr | sambetts, what about common.py, would be nice to all those who needs to run the periodic?! | 14:02 |
| *** mgoddard_ has joined #openstack-ironic | 14:02 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Forbid to remove node name for api versions < 1.5 https://review.openstack.org/300983 | 14:02 |
| xavierr | NobodyCam, morning | 14:03 |
| NobodyCam | morning xavierr :) | 14:03 |
| NobodyCam | moving slow this morning :) | 14:04 |
| *** mgoddard has quit IRC | 14:06 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Update resources only for specific node during deletion. https://review.openstack.org/287498 | 14:07 |
| * thiagop hands NobodyCam a double espresso | 14:09 | |
| NobodyCam | thiagop: Thank you :) | 14:10 |
| sambetts | xavierr: I think you may be able to do that now we're using the futurist periodics module | 14:11 |
| *** mbound has joined #openstack-ironic | 14:15 | |
| xavierr | sambetts, yes I'm going to use de periodic module, even I had some bad experiences with futurist for threading some weeks ago :) | 14:16 |
| xavierr | sambetts, I you give futurist a second chance | 14:16 |
| xavierr | sambetts, will* | 14:16 |
| jroll | xavierr: sambetts: seems like the task at hand fits in the deploy driver, as we don't need the task if ironic isn't deploying oneview machines, right? :) | 14:18 |
| *** arturas_ has quit IRC | 14:19 | |
| *** rbrud has joined #openstack-ironic | 14:19 | |
| sambetts | jroll: I think they meant their oneview common.py which would only get loaded if their driver is loaded right? | 14:20 |
| *** Marga_ has quit IRC | 14:20 | |
| jroll | sambetts: right, I guess I'm wondering if there's an e.g. oneview power + ilo deploy driver or something | 14:23 |
| jroll | in which case it probably isn't needed | 14:23 |
| openstackgerrit | Marcellin Fom Tchassem proposed openstack/ironic-inspector: Make tox respect upper-constraints.txt https://review.openstack.org/301700 | 14:24 |
| *** mtanino has joined #openstack-ironic | 14:25 | |
| jroll | then again, probably nobody would ever use such a thing, so /shrug | 14:26 |
| NobodyCam | morning jroll JayF jlvillal sambetts mgould sinval-afk dtantsur lucasagomes thiagop TheJulia devananda vdrok mat128 and all others not listed here | 14:26 |
| jroll | good morning PingCam | 14:27 |
| jroll | :P | 14:27 |
| thiagop | good morning NobodyCam | 14:27 |
| thiagop | lol | 14:27 |
| NobodyCam | lol | 14:27 |
| dtantsur | morning NobodyCam | 14:27 |
| NobodyCam | :) | 14:27 |
| TheJulia | good morning NobodyCam | 14:27 |
| sambetts | jroll: when do function decorators get processed when they are inside a class? On first class use or the module load? | 14:28 |
| jroll | wow, product WG has us high priority for newton http://lists.openstack.org/pipermail/product-wg/2016-April/001064.html | 14:29 |
| *** achanda has joined #openstack-ironic | 14:29 | |
| jroll | sambetts: fairly certain decorators run at import time, but not 100% sure, easy to test :) | 14:29 |
| NobodyCam | jroll: what part of us? | 14:30 |
| jroll | NobodyCam: this, apparently: https://review.openstack.org/#/c/289627/ | 14:30 |
| sambetts | jroll: thats what I thought too | 14:30 |
| jroll | I guess this is what the product WG wants out of a baremetal service | 14:31 |
| jroll | aaaaaaaand we're working on all but one of those :) | 14:31 |
| jroll | the first one being the exception, I think | 14:31 |
| jroll | but maybe not, it isn't clear to me | 14:31 |
| *** vmud213 has quit IRC | 14:31 | |
| sambetts | jroll: sounds like the fuzzy flavor matching stuff I've talked about in the past | 14:32 |
| jroll | maybe? | 14:32 |
| * jroll asks for clarification | 14:32 | |
| sambetts | I'd still like to see flavors be able to configure BM to meeting flavor requirements/specs | 14:33 |
| NobodyCam | not sure I get the continue my operation immediately one | 14:33 |
| dtantsur | jroll, we have a spec for infiniband, other things might involve inspection | 14:33 |
| jroll | NobodyCam: which one? | 14:34 |
| NobodyCam | line 70/71 | 14:34 |
| sambetts | NobodyCam: I think that means they want live migraation | 14:34 |
| NobodyCam | ahh | 14:34 |
| NobodyCam | so I see two then the backup and migration ones | 14:35 |
| jroll | oh yeah, I guess I was looking at the section above | 14:35 |
| NobodyCam | brb more coffee required | 14:35 |
| *** mvpnitesh has quit IRC | 14:35 | |
| jroll | ++ | 14:36 |
| sambetts | backup one == snapshoting | 14:36 |
| xavierr | jroll, sambetts, thanks, I test with the periodic on deploy driver :) | 14:36 |
| jroll | yep | 14:36 |
| thiagop | This WG is starting this work from now? It's good to see some governance... | 14:36 |
| TheJulia | jroll: I recognize Carol's name, some of the UX/UI interactions have involved people that afaik report up to Carol, so I think that may be mainly from a user interaction with the service | 14:36 |
| thiagop | This tends to set direction | 14:36 |
| dtantsur | honestly I have no clues what this workgroup is about :) | 14:36 |
| dtantsur | thiagop, well, in ideal world... in more practical world our priorities are also set by our employers | 14:37 |
| dtantsur | and the degree to which an employer decided it varies highly between companies | 14:37 |
| jroll | TheJulia: it's from a "people that want to sell ironic deployments and/or bare metal cloud" perspective, afaik | 14:38 |
| jroll | most of the people in this group are product people from our employers | 14:38 |
| thiagop | dtantsur: right you are, but I think that without some direction, we tend to deliver incomplete things. Mostly things that need cross-project effort. | 14:38 |
| dtantsur | completing things is about discipline, not about management :) | 14:39 |
| *** achanda has quit IRC | 14:39 | |
| TheJulia | jroll: yeah, looking at what they are tyring to list as the user story now | 14:39 |
| dtantsur | nobody here can be explicitly told what to do by any wg. we can only be convinced, which is hopefully what these folks plan to do | 14:39 |
| jroll | dtantsur: yep | 14:39 |
| thiagop | dtantsur: I disagree with that, but it's a long out of context talk... | 14:39 |
| dtantsur | it might be different for us :) | 14:40 |
| jroll | I for one welcome the PWG, as we've struggled to get useful feedback from ops/users in the past | 14:40 |
| dtantsur | jroll++ | 14:40 |
| jroll | thiagop: fwiw, this workgroup does not have any sort of method to make us do things or whatever, they're simply meant to feed in input from people outside of the dev community | 14:40 |
| * jroll runs to a meeting | 14:40 | |
| TheJulia | fwiw, some of what is on there is... just.. not possible short of laying down a thin hypervisor and calling it baremetal, which kind of defeats the prupose | 14:41 |
| TheJulia | at least, not easilly with ease and performance | 14:41 |
| jroll | right :P | 14:41 |
| thiagop | jroll I well aware of it, but as dtantsur said, it's good to have an overview of what users might want | 14:41 |
| jroll | there needs to be some education/syncup | 14:41 |
| TheJulia | indeed | 14:41 |
| jroll | thiagop: yeah, totally, I welcome that :) | 14:41 |
| * jroll wonders if we have summit session room for a "what do you need from ironic" thing | 14:42 | |
| dtantsur | jroll, we used to have such sessions during the ops summit | 14:42 |
| thiagop | they can indeed convince us (and nova, and cinder, and horizon) that some things need to be completely worked on to deliver value to OS users | 14:42 |
| jroll | dtantsur: right, and they weren't very useful | 14:42 |
| dtantsur | not too much feedback unfortunately | 14:42 |
| dtantsur | yep, exactly this :) | 14:42 |
| jroll | however, folks want to meet up apparently http://lists.openstack.org/pipermail/product-wg/2016-April/001067.html | 14:42 |
| jroll | not sure if worth a session though | 14:42 |
| dtantsur | hmm, I'd love these folks to involve us *before* they settle on user stories | 14:43 |
| dtantsur | I understand we can't provide them, but we can always say "well, that's barely possible" | 14:44 |
| jroll | yeah, there's a disconnect | 14:44 |
| jroll | I may go talk with them before summit | 14:44 |
| dtantsur | ++ | 14:44 |
| TheJulia | sounds like the best thing that we can do | 14:44 |
| *** spandhe has joined #openstack-ironic | 14:45 | |
| thiagop | I think that's what they want us to do reviewing these US's. Sanity check + align expectations from other vendors | 14:46 |
| *** rebase has joined #openstack-ironic | 14:46 | |
| * thiagop tends to look at this with good eyes | 14:46 | |
| *** mgoddard has joined #openstack-ironic | 14:47 | |
| *** mgoddard_ has quit IRC | 14:47 | |
| *** b_bezak has quit IRC | 14:48 | |
| JayF | How does it work when things depend on ironic-lib? Like say I have a new widget in ironic-lib and I want to use it in an IPA patch; is setting depends-on enough to use the new ironic-lib code/ | 14:50 |
| JayF | cc: alineb ^ | 14:50 |
| vdrok | morning NobodyCam and JayF :) | 14:50 |
| JayF | I'm 99.99% certain we install that from pip directly, don't we ;( | 14:50 |
| TheJulia | JayF: afaik we do | 14:53 |
| jroll | JayF: yeah, we do install it from pip | 14:53 |
| TheJulia | JayF: and as such, depends-on doesn't help :( | 14:53 |
| JayF | So is there any trick to doing testing like that then? Assuming would probably have to be some kind of devstack patch in the mix | 14:53 |
| jroll | JayF: you can hack it to install from a given patchset, lemme find an example | 14:53 |
| JayF | perfect, thanks jroll | 14:53 |
| *** mgould has quit IRC | 14:53 | |
| jroll | I've thought about putting ironic-lib/client src tests in the experimental queue for this reason though | 14:53 |
| jroll | JayF: https://review.openstack.org/#/c/296432/24/devstack/lib/ironic | 14:54 |
| jroll | something like that in the chain | 14:54 |
| JayF | well just annoying b/c who wants to write the lib, wait a week or two for it to merge, before then using it again | 14:54 |
| jroll | sure, I get it | 14:54 |
| jroll | also a release in the middle | 14:55 |
| jlvillal | Good morning Ironic :) | 14:56 |
| *** mgould has joined #openstack-ironic | 14:56 | |
| jroll | JayF: feel free to propose a project-config patch to add the -src job to the experimental queue, if you'd prefer that route | 14:56 |
| jroll | then it's just "check experimental" to change it | 14:56 |
| JayF | jroll: that's not actually going to work either | 14:56 |
| jroll | s/change/test/ | 14:56 |
| JayF | jroll: for an *IPA* patch | 14:56 |
| jroll | why | 14:56 |
| * jlvillal imagining jroll driving 65 MPH down the Interstate in the passenger seat while on IRC :) | 14:56 | |
| jroll | JayF: ironic-lib -> ironic patch to adjust the source -> ipa patch | 14:57 |
| JayF | jroll: because IPA builds check out shit without any deference whatseover to upper-constraints or anything else | 14:57 |
| thiagop | good morning jlvillal | 14:57 |
| jroll | jlvillal: I'm not on the road | 14:57 |
| JayF | jroll: ipa doesn't build out of devstack git dirs | 14:57 |
| * TheJulia wonders who doesn't irc while going down the highway at 65mph in the passenger seat.... | 14:57 | |
| jroll | JayF: ooooo fun | 14:57 |
| JayF | yeah, exactly | 14:57 |
| jroll | JayF: git+https://git.o.o/openstack/ironic-lib@refs/.... | 14:57 |
| jroll | in reqs.txt | 14:57 |
| JayF | yeah, that's whow I do it | 14:58 |
| JayF | where do you get the full ref link in new gerrit? | 14:58 |
| dtantsur | JayF, "download" link in the right corner | 14:58 |
| *** derekh is now known as derekh_brb | 14:59 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: DoNotMerge: Use ironic-lib w/metrics for devstack https://review.openstack.org/301740 | 15:00 |
| *** ametts has quit IRC | 15:00 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: DoNotMerge: Use ironic-lib w/metrics for devstack https://review.openstack.org/301740 | 15:01 |
| *** mkoderer__ has quit IRC | 15:02 | |
| *** dtantsur is now known as dtantsur|brb | 15:03 | |
| JayF | jroll: any idea how to build that link properly? clone https://git.openstack.org/openstack/ironic-lib + git checkout refs/changes/26/301256/1 throws an error | 15:06 |
| *** pcaruana has quit IRC | 15:06 | |
| jroll | JayF: origin/refs/...? | 15:06 |
| jroll | or as the download link says, git fetch ironic-lib refs/... && git checkout refs... | 15:06 |
| *** derekh_brb is now known as derekh | 15:06 | |
| JayF | it wants me to fetch via review.openstack | 15:07 |
| JayF | with creds | 15:07 |
| jroll | JayF: alineb: we should probably configure devstack to use this stuff | 15:07 |
| jroll | well | 15:07 |
| jroll | OH | 15:07 |
| jroll | yeah git.o.o probably won't have that | 15:07 |
| jroll | but https://review.o.o should work anonymously | 15:07 |
| jroll | git fetch https://git.openstack.org/openstack/ironic-lib refs/changes/26/301526/1 && git checkout FETCH_HEAD | 15:07 |
| jroll | should work without creds | 15:07 |
| *** garthb has joined #openstack-ironic | 15:08 | |
| JayF | aha perfect | 15:08 |
| jroll | straight from the download link :P | 15:08 |
| jroll | so that said | 15:08 |
| jroll | requirements.txt: git+https://git.openstack.org/openstack/ironic-lib@refs/changes/26/301526/1 should work for IPA | 15:09 |
| jroll | I've gotta bounce for now, movers coming soon | 15:09 |
| jroll | may or may not be back this afternoon, if not have a good day y'all | 15:09 |
| JayF | jroll: that does not appear to work locally | 15:10 |
| NobodyCam | drave safe jroll | 15:10 |
| JayF | jroll: I can' | 15:10 |
| JayF | I can't clone that then checkout that branch | 15:10 |
| NobodyCam | drive even | 15:10 |
| sambetts | the problem is with gerrit that 'branch' doesn't really exist | 15:10 |
| jroll | JayF: clone, then fetch, then checkout | 15:11 |
| jroll | like | 15:11 |
| jroll | in an existing clone: run the git fetch ... && git checkout | 15:11 |
| sambetts | Doesn't work in requirements.txt though ;( I've hit that so many times | 15:12 |
| jroll | just confirmed the fetch/checkout command works for me in an existing clone. | 15:12 |
| *** ametts has joined #openstack-ironic | 15:13 | |
| *** links has joined #openstack-ironic | 15:14 | |
| *** ChubYann has joined #openstack-ironic | 15:15 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Add command metrics to IPA API https://review.openstack.org/119981 | 15:15 |
| *** mkoderer__ has joined #openstack-ironic | 15:16 | |
| *** harshs has joined #openstack-ironic | 15:17 | |
| *** krtaylor has joined #openstack-ironic | 15:17 | |
| *** jaybeale has joined #openstack-ironic | 15:19 | |
| *** mdorman has joined #openstack-ironic | 15:23 | |
| openstackgerrit | Sergii Turivnyi proposed openstack/python-ironicclient: Negative tests for testing actions with Chassis https://review.openstack.org/301757 | 15:24 |
| *** links has quit IRC | 15:25 | |
| *** mkovacik has quit IRC | 15:26 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Add metrics support to IPA https://review.openstack.org/119981 | 15:30 |
| *** alexpilotti has quit IRC | 15:34 | |
| JayF | Could not find a tag or branch 'refs/changes/26/301526/1', assuming commit. | 15:35 |
| JayF | error: pathspec 'refs/changes/26/301526/1' did not match any file(s) known to git. | 15:35 |
| JayF | giving that git path to ipa did /not/ work :( | 15:35 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-ironicclient: Add tests for baremetal subcommands in openstackclient https://review.openstack.org/301687 | 15:35 |
| sambetts | JayF: are you still trying with requirements.txt? | 15:36 |
| JayF | yep | 15:37 |
| JayF | So I can run tests sanely on the IPA metrics patch alineb and I are working on | 15:37 |
| sambetts | Yeah gerrit doesn't work with requirements.txt because the gerrit ref isn't a "real" branch | 15:37 |
| *** Marga_ has joined #openstack-ironic | 15:38 | |
| JayF | There's no way to make that ref fetchable from somewhere? | 15:38 |
| *** Marga_ has quit IRC | 15:38 | |
| JayF | Unless like, I push it to my own fork or something? | 15:38 |
| *** Marga_ has joined #openstack-ironic | 15:38 | |
| *** Marga_ has quit IRC | 15:39 | |
| *** Marga_ has joined #openstack-ironic | 15:39 | |
| *** dtantsur|brb is now known as dtantsur | 15:39 | |
| sambetts | what I've done before is add the git.openstack.org for ironic-lib to the requirements.txt then once the .tox environment is setup doing (source .tox/toxenv/bin/activate && cd .tox/toxenv/src/ironic-lib && git fetch .... && git checkout FETCH_HEAD && pip install -e .) | 15:40 |
| JayF | yeah, I'm hoping for something more automated | 15:40 |
| JayF | lol | 15:40 |
| JayF | asking in -infra | 15:40 |
| JayF | you can help by reviewing https://review.openstack.org/#/c/301526/ "{ | 15:41 |
| JayF | :P | 15:41 |
| TheJulia | seems like we need something that groks $ZUUL_REFS | 15:41 |
| TheJulia | or is it ZUUL_CHANGES | 15:41 |
| * TheJulia doesn't remember | 15:41 | |
| sambetts | Sounds like we need something like what I have in networking-cisco for neutron client and neutron | 15:42 |
| sambetts | http://git.openstack.org/cgit/openstack/networking-cisco/tree/tools/tox_install.sh | 15:42 |
| sambetts | JayF: ^ | 15:42 |
| *** moshele has quit IRC | 15:42 | |
| sambetts | except of course you can replace the logic with the right pip install + git fetch + pip install again logic | 15:43 |
| sambetts | and for it to work upstream and understand the zuul refs environment variables you need to use the zuul cloner application | 15:43 |
| TheJulia | zuul_cloner is supposed to stage everything on disk based on current state | 15:44 |
| TheJulia | so basically pip install from /opt/git/blah/blah/blah | 15:44 |
| *** moshele has joined #openstack-ironic | 15:44 | |
| TheJulia | but zuul cloner has to explicitly run the checkout | 15:45 |
| *** daemontool has quit IRC | 15:45 | |
| TheJulia | krotscheck: are the npm jobs okay? | 15:50 |
| krotscheck | TheJulia: Eh? | 15:50 |
| * krotscheck checks | 15:50 | |
| TheJulia | just noticed on two separate projects the jobs have failed in rather awesomely not good ways | 15:51 |
| *** spandhe has quit IRC | 15:52 | |
| jlvillal | JayF: Did you get it figured out? | 15:52 |
| JayF | jlvillal: tl;dr: the only Right(tm) way to do it is like jroll said, an experimental job that builds all the reqs from source | 15:52 |
| jlvillal | I don't think you can do the fetch from git.openstack.org. I thought it had to be from review.openstack.org. I could be wrong of course. | 15:52 |
| jlvillal | For patches, that is. | 15:53 |
| JayF | jlvillal: I've heard 0 reports of anyone saying what I'm doing is possible, except jroll's suggestion which didn't work | 15:53 |
| JayF | so I'm open for additional suggestions, but for now I mostly have given up | 15:53 |
| JayF | the Best Fix(tm) is to get https://review.openstack.org/#/c/301526/ reviewed, merged, and relased ASAP | 15:53 |
| jlvillal | JayF: Okay. | 15:53 |
| JayF | and I bet alineb would appreciate the reviews | 15:53 |
| *** praneshp_ has joined #openstack-ironic | 15:54 | |
| krotscheck | TheJulia: Looks like the registry is a bit borked at the moment. | 15:55 |
| TheJulia | krotscheck: ok | 15:55 |
| jlvillal | JayF: alineb I posted some initial nits | 15:56 |
| JayF | jlvillal: your style nits ... if pep8 is passing I thought we didn't vote on style stuff? | 15:56 |
| JayF | jlvillal: sounds like pep is misconfigured in that repo though, if it's letting that stuff through | 15:56 |
| jlvillal | I don't believe so. flake8 doesn't know which libraries are built-in, 3rd party, or intree | 15:57 |
| alineb | jlvillal: thanks, i will change that | 15:57 |
| JayF | hm, okay. I've def gotten pep8 failures from that, but it seems like the sort of thing that'd be iffy in the gate | 15:58 |
| JayF | so they probably turned it off sometime ago and I just never noticed, lol | 15:58 |
| jlvillal | JayF: It will complain about sorting and spacing. | 15:58 |
| JayF | aha | 15:58 |
| JayF | the sorting is what I've been ding'd for | 15:58 |
| jlvillal | alineb: JayF So when it says adding support for Metrics. | 15:58 |
| jlvillal | Does it actually use metrics? | 15:58 |
| JayF | What do you mean, does it actually use metrics? | 15:58 |
| jlvillal | Or is it just adding metric support so that it can be used by projects | 15:59 |
| JayF | it's adding support to be used by IPA and Ironic | 15:59 |
| jlvillal | Basically if it was completely broken and merged would anything break? | 15:59 |
| jlvillal | :) | 15:59 |
| JayF | we have the Ironic side patches incoming, and the IPA patch is up (but depends on an Ironic patch that's not up yet) | 15:59 |
| JayF | jlvillal: heh, no, nothing would break | 15:59 |
| openstackgerrit | Merged openstack/ironic-inspector: Move unit tests to "unit" directory https://review.openstack.org/301507 | 16:00 |
| *** harshs has quit IRC | 16:01 | |
| openstackgerrit | Aline Bousquet proposed openstack/ironic-lib: Add support for metrics https://review.openstack.org/301526 | 16:01 |
| alineb | jlvillal: just submitted a new patch set | 16:01 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Add metrics support to IPA https://review.openstack.org/119981 | 16:01 |
| JayF | alineb: https://review.openstack.org/#/c/119981 I just W-1'd this one, as it depends on the Ironic patch to send over the configs ;) | 16:02 |
| *** alexpilotti has joined #openstack-ironic | 16:03 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: [WIP] Wait for the interfaces to get IP addresses before inspection https://review.openstack.org/300548 | 16:04 |
| JayF | dtantsur: ^ you willing to consider an alternative approach to that? | 16:04 |
| dtantsur | easily :) | 16:04 |
| JayF | dtantsur: we have downstream code that waits in coreos to start IPA until the network is up | 16:04 |
| JayF | dtantsur: but I guess that wouldn't support the other images, so is probably not good enough for you | 16:05 |
| dtantsur | JayF, coreos image doesn't have this problem afaik :) so it's more about weird DIB DHCP magic... | 16:05 |
| JayF | well, CoreOS image has systemd. If IPA dies coming up b/c no network, it just relaunches it 30s later :) | 16:05 |
| JayF | but IDK if it even dies on startup in the inspectino case | 16:05 |
| JayF | in the provisioning-agent case, if it starts with no network and can't lookup, it does | 16:06 |
| dtantsur | JayF, now, it does not, it fails the inspection | 16:06 |
| dtantsur | that's the problem here | 16:06 |
| JayF | ah | 16:06 |
| sambetts | JayF: theoretically the systemd service file we define for the DIB image for IPA should prevent it from starting before the network is ready for it, however the dhcp-all-interfaces element means that the network ready target isn't reliable | 16:07 |
| dtantsur | yep :( | 16:07 |
| JayF | well the prbolem is | 16:07 |
| JayF | when dhcping | 16:07 |
| JayF | systemd reports network-online.target when dhcp client daemon is /started/ | 16:07 |
| JayF | not when it gets a lease | 16:07 |
| dtantsur | daaaaaaaamn | 16:07 |
| * dtantsur didn't know that | 16:08 | |
| JayF | I spent like a week with a CoreOS dev working on getting the OnMetal image for CoreOS working, back before we launched originally | 16:08 |
| JayF | learned lots of black magic about systemd then | 16:08 |
| sambetts | well we need this logic anyways then I guess | 16:08 |
| JayF | dtantsur: I see it's WIP'd, you want a review anyway? | 16:08 |
| JayF | if not now, lmk when you do and I will :) | 16:09 |
| dtantsur | JayF, yes please. it's wip because of missing unit tests for one part | 16:09 |
| dtantsur | the code itself should be ready | 16:09 |
| JayF | Why did you decide to make wait_for_ip_addresses a HWM method? | 16:10 |
| openstackgerrit | Merged openstack/bifrost: Explicitly set language for DIB builds https://review.openstack.org/298784 | 16:10 |
| dtantsur | JayF, instead of...? | 16:10 |
| openstackgerrit | Merged openstack/bifrost: Fix DIB conditionals for CLI override https://review.openstack.org/298785 | 16:10 |
| JayF | well, things in hardwaremanager are generally for thigns other folks would want to change | 16:10 |
| JayF | imo at least | 16:10 |
| dtantsur | previously it was in inspector.py and someone asked it to be more generic | 16:10 |
| JayF | well, I was thinking more like utils.py | 16:10 |
| dtantsur | I think jroll said he like it | 16:10 |
| JayF | and make the whole damn IPA block on has_carrier, since it's a provisioning agent problem too | 16:11 |
| dtantsur | JayF, utils.py does not depends on hardware.py (actually, the opposite) | 16:11 |
| dtantsur | anyway, feel free to leave a review | 16:11 |
| JayF | It's fine to be in the hwm, but you'll have to refactor it a little | 16:11 |
| JayF | I'll commentin the patch | 16:11 |
| dtantsur | sure, will fix (tomorrow, I guess) | 16:11 |
| JayF | dtantsur: put those comments on for ya | 16:13 |
| *** moshele has quit IRC | 16:13 | |
| *** jcoufal has quit IRC | 16:17 | |
| jlvillal | alineb: I left a few more comments. | 16:17 |
| *** derekh has quit IRC | 16:19 | |
| *** jayc has quit IRC | 16:20 | |
| mat128 | anyone down for a review of https://review.openstack.org/#/c/264767/ ? | 16:28 |
| *** spandhe has joined #openstack-ironic | 16:28 | |
| alineb | jlvillal: thanks! i will look at them | 16:29 |
| * sambetts is calling it a day | 16:29 | |
| sambetts | See everyone tomorrow! | 16:29 |
| *** sambetts is now known as sambetts|afk | 16:29 | |
| JayF | mat128: landing it | 16:29 |
| JayF | mat128: ty for the patch | 16:29 |
| mat128 | :D | 16:29 |
| sambetts|afk | :d | 16:29 |
| sambetts|afk | :D | 16:29 |
| mat128 | Built a black box test downstream to prove resolution works, still need to find a way to contribute this back | 16:30 |
| mat128 | Where would you place such a test? | 16:30 |
| mat128 | It launches qemu with the kernel and ramdisk | 16:30 |
| JayF | I mean, that's essentailly what devstack does | 16:30 |
| JayF | just you did it with a need for dns resolution | 16:30 |
| JayF | whereas devstack plays in all IPs | 16:30 |
| JayF | I'm not sure something environmental like this is the best place for a test -- like for instance, we still have 2 other ramdisks (tinyipa, dib) that may still or may not still have this problem | 16:31 |
| mat128 | tinyipa works oob | 16:31 |
| mat128 | found that out because it instantly passed my test | 16:31 |
| mat128 | do we have tests in devstack around the ipa images themselves? | 16:31 |
| JayF | We have a couple, I can show you | 16:31 |
| *** jayc has joined #openstack-ironic | 16:32 | |
| mag009_ | hey guys | 16:32 |
| *** piet has joined #openstack-ironic | 16:33 | |
| JayF | mat128: I can't find it, but the thing to look for is -src on ajob in an ipa review | 16:33 |
| mat128 | ok, I'll find it | 16:33 |
| mat128 | thanks for your help | 16:33 |
| JayF | mat128: those -src jobs build a new ramdisk from source as part of the test. Today gates all run with the CoreOS image, but we have that tinyipa image about to take it over in the gate | 16:33 |
| mat128 | brb 30m | 16:33 |
| JayF | for obvious reasons (it's tiny, lol) | 16:33 |
| mat128 | yeah | 16:33 |
| mat128 | are we building it with BUILD_AND_INSTALL_TINYIPA on? | 16:34 |
| mat128 | (makes it even smaller) | 16:34 |
| JayF | the tinyipa + src job does | 16:34 |
| JayF | which is non voting today | 16:34 |
| *** mat128 is now known as mat128|afk | 16:35 | |
| mat128|afk | ok, I'll look around the jobs and the tests we have and try to submit a review for something similar to what I did here :) | 16:36 |
| JayF | I mean, you can, but I'm telling you I'm not sure that's a case we care deeply enough about to have a specific job for :) | 16:36 |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [database] https://review.openstack.org/301815 | 16:37 |
| *** suro-patz has joined #openstack-ironic | 16:38 | |
| *** Nisha_away has joined #openstack-ironic | 16:41 | |
| *** trown is now known as trown|lunch | 16:43 | |
| *** jayc has quit IRC | 16:44 | |
| *** wendar_ is now known as wendar | 16:45 | |
| *** mbound has quit IRC | 16:48 | |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Add user and project domains to ironic context https://review.openstack.org/295371 | 16:50 |
| openstackgerrit | Vladyslav Drok proposed openstack/ironic: Use get_admin_context() to create the context object https://review.openstack.org/246036 | 16:50 |
| *** jayc has joined #openstack-ironic | 16:56 | |
| *** ifarkas has quit IRC | 17:00 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [dhcp] https://review.openstack.org/301830 | 17:08 |
| *** alexpilotti has quit IRC | 17:09 | |
| dtantsur | JayF, thanks! | 17:13 |
| *** harshs has joined #openstack-ironic | 17:16 | |
| dtantsur | JayF, utils.py won't work, as it will create a mutual dependency between hardware.py and utils.py. I can put it to hardware.py as a function, wdyt? | 17:16 |
| JayF | Let me put it this way: | 17:17 |
| JayF | I care less about the filesystem location | 17:17 |
| JayF | and more about it being either being done "right" in the hardware manager (i.e. with a stub on HardwareManager + the real implementaiton on GHM) or it being out of the hardware manager classes altogether | 17:17 |
| JayF | dtantsur: actally found something else, going to comment | 17:18 |
| JayF | dtantsur: tl;dr you wait on all interfaces, is that intended vs just waiting for a single IP? | 17:18 |
| dtantsur | JayF, yes! cause we never know which interface we really need | 17:18 |
| dtantsur | that's the tricky part of this whole patch | 17:19 |
| JayF | dtantsur: Hmm. In our case, we have 2 interfaces, 1 will never dhcp | 17:19 |
| dtantsur | and that's why I was somewhat unsure about doing it unconditionally on start up | 17:19 |
| dtantsur | JayF, I understand it, that's the price to pay for not knowing anything before inspection... | 17:19 |
| JayF | maybe make it an option on the method? On every boot, you could look for any IP | 17:19 |
| JayF | for inspection | 17:20 |
| dtantsur | JayF, sorry, I don't quite get it.. what should be optional? | 17:20 |
| JayF | and for startup, look for any ip at all | 17:20 |
| JayF | Hmm. This is interesting, right | 17:20 |
| JayF | so here's my question: | 17:20 |
| JayF | Is your use case "my agent need to have eth0 and eth1 up because eth0 goes to ironic and eth1 goes to swift" or something similar | 17:21 |
| *** mgould has quit IRC | 17:21 | |
| *** morgan is now known as notnotmyname | 17:21 | |
| *** notnotmyname is now known as morgan | 17:21 | |
| JayF | or is it "I want all possible IPs up so that the inspector returns the maximum amount of valid information" | 17:21 |
| mat128|afk | From our previous discussions, it's the latter | 17:21 |
| dtantsur | JayF, the latter | 17:21 |
| *** mat128|afk is now known as mat128 | 17:21 | |
| JayF | so what I'm saying is | 17:22 |
| *** Marga_ has quit IRC | 17:22 | |
| JayF | you can add an option to wait_for_ip_addresses | 17:22 |
| JayF | that either waits for ANY ip | 17:22 |
| JayF | or waits for ALL potential ips | 17:22 |
| JayF | and make IPA wait on startup for ANY ip if in provisioning mode | 17:22 |
| dtantsur | JayF, what's the use case for waiting for any IP? | 17:22 |
| dtantsur | ah | 17:22 |
| JayF | and make inspector wait for all | 17:22 |
| dtantsur | JayF, well, we already wait for a route to ironic, isn't it enough? | 17:22 |
| dtantsur | (I mean, for generic case) | 17:22 |
| JayF | I think you just exposed a knowledge gap I gained from not working upstream enough | 17:23 |
| JayF | because I didn't know that existed, lol | 17:23 |
| JayF | nevermind me then! Wheee... :( | 17:23 |
| dtantsur | JayF, I mean this: https://github.com/openstack/ironic-python-agent/commit/6829d34c150fa1cd41064786e76d41dfccef3ef3 | 17:23 |
| dtantsur | it has a retry logic | 17:24 |
| JayF | Yep, and I didn't know/remember that existed | 17:24 |
| JayF | which completely handles any case I was worrying about | 17:24 |
| openstackgerrit | Merged openstack/ironic-python-agent: Provide /etc/resolv.conf to chroot https://review.openstack.org/264767 | 17:24 |
| dtantsur | it's a recent thing, nevermind :) | 17:24 |
| dtantsur | so | 17:24 |
| *** jistr has quit IRC | 17:24 | |
| dtantsur | JayF, I think I'll just move the function back to inspector.py.. objections? | 17:25 |
| *** moshele has joined #openstack-ironic | 17:25 | |
| JayF | none | 17:25 |
| JayF | I just think of HardwareManager as kinda a driver interface | 17:25 |
| * dtantsur hacks | 17:25 | |
| JayF | so if you put things in there I might push back if they aren't something folks would want to plug in their own implementation of :) | 17:25 |
| *** arturas has joined #openstack-ironic | 17:28 | |
| JayF | TheJulia: around? wanted to talk about https://review.openstack.org/#/c/270902 since it's basically done except for me and you disagreeing over the default :) | 17:36 |
| *** davidlenwell has quit IRC | 17:42 | |
| mat128 | JayF: would it be acceptable for you to default to not erase drives at all or just not using shred by default? | 17:45 |
| JayF | I put my comment in the gerrit; basically I'm against defaulting to fallback to shred IF the drive supports ATA erase | 17:45 |
| JayF | because the tenant can perform actions (like a security lock via hdparm) that the deployer could be 100% unaware of | 17:45 |
| JayF | and cause the machine to shred everytime, without so much as the deployer even knowing | 17:46 |
| mat128 | so if we rephrase it, if a drive supports ATA erase and is unlocked, use that without any fallback | 17:46 |
| mat128 | if it doesnt, dont even bother and use other means? | 17:46 |
| JayF | Let me state it this way: | 17:46 |
| JayF | A deployer should have to explicitly opt-in to having SHRED run on a disk that supports instant secure erase. | 17:47 |
| mat128 | Ooh | 17:47 |
| JayF | That patch, as it is, makes it where the deployer has to opt-out | 17:47 |
| mat128 | yes | 17:47 |
| JayF | and given it's a scenario that a tenant can cause -- even a non-malicious tenant -- it's important to have a safe default | 17:47 |
| *** garthb has quit IRC | 17:47 | |
| JayF | and shredding ssds is not safe, it's like shredding dollar bills in terms of what it does to the life of the drive | 17:47 |
| mat128 | agreed | 17:48 |
| mat128 | so opt-in to the unsafe (wrt the drive) behavior | 17:48 |
| mat128 | but safer from a data leak perspective | 17:48 |
| JayF | I had some drives in production that advertised ISE but just did a shred under the hood. Hence my passion about this specific problem :P | 17:48 |
| JayF | I think it should go to clean failed in that case | 17:48 |
| mat128 | we've had problems getting the drive to unlock, having to request custom BIOSes all the time | 17:48 |
| JayF | which is what it does today | 17:48 |
| JayF | mat128: same, we have custom firmwares for our drives / bios that prevents locking the drive unless you know the right tool/password/etc | 17:49 |
| *** davidlenwell has joined #openstack-ironic | 17:49 | |
| mat128 | JayF: that would be a dream here :) | 17:49 |
| JayF | heh, I mean, you probably don't run a public bare metal cloud. | 17:50 |
| mat128 | ordering a custom bios meant we probably lost bios upgrades in the future, and we had to request those for the plethora of different boards we have to support :( | 17:50 |
| JayF | For us things like that are must-haves not nice-to-haves | 17:50 |
| mat128 | fyi: http://www.internap.com/bare-metal/ | 17:50 |
| mat128 | we just end up shredding dollar bills, to use that analogy :( | 17:51 |
| JayF | welcome to the public ironic cloud club | 17:52 |
| JayF | I would wish you luck, but, I like getting paid too :P | 17:52 |
| JayF | may the person who commits the most upstream win, lol | 17:52 |
| mat128 | been a while now, but thanks :) | 17:52 |
| mat128 | it's all about upstream work IMO | 17:52 |
| JayF | how long have you all been using Ironic with that? | 17:52 |
| *** moshele has quit IRC | 17:53 | |
| *** Marga_ has joined #openstack-ironic | 17:53 | |
| *** mkovacik has joined #openstack-ironic | 17:54 | |
| mat128 | we launched a beta during the Vancouver summit (iirc may 2015), full launch october 2015 | 17:54 |
| JayF | Do you all have network switching support implemented downstream as well? | 17:55 |
| mat128 | yup | 17:55 |
| mat128 | we currently support both flat and lacp | 17:55 |
| mat128 | network attach/detach | 17:55 |
| JayF | Ah, do you not share Ironic installations across tenants? | 17:56 |
| mat128 | what do you mean by Ironic installations? | 17:56 |
| mat128 | instances of Ironic? | 17:56 |
| *** arturas has quit IRC | 17:59 | |
| mat128 | JayF: we run one ironic installation per region | 18:00 |
| JayF | was just curious how you had multiple customers on the same ironic install w/flat network | 18:00 |
| mat128 | each tenant has their own networks | 18:00 |
| *** Mr_T has joined #openstack-ironic | 18:01 | |
| JayF | Hm, okay. | 18:01 |
| mat128 | my bad, must be the "flat" that confused you | 18:01 |
| mat128 | by flat network model I mean WAN on one NIC, LAN on the other | 18:01 |
| JayF | Yeah, so you probably have a similar downstream patch to us | 18:01 |
| JayF | to make the networks flip on provision/deprovision | 18:01 |
| mat128 | yeah, essentially | 18:02 |
| mat128 | now with the latest reviews coming in, we will be able to simply extend that to do what we want and remove part of our downstream patches | 18:02 |
| *** trown|lunch is now known as trown | 18:02 | |
| *** ohamada has quit IRC | 18:03 | |
| JayF | Yep, same here. The upstream support, once the remaining patches get merged, we should be able to switch over to that | 18:03 |
| *** mgoddard_ has joined #openstack-ironic | 18:03 | |
| *** arturas has joined #openstack-ironic | 18:04 | |
| JayF | mat128: back to the original issue at hand; to be crystal clear: IMO inability to ISE a drive that supports it should default cleanfail, but I'm OK with it optionally shredding and continuing. I'd never +1 and probably would -2 a patch that would skip erasing the drives altogether if it couldn't ISE :) | 18:04 |
| mat128 | JayF: I'm fine with that behavior, would +1 | 18:05 |
| mat128 | Maybe we can summarize this for TheJulia in the review? | 18:05 |
| JayF | I think I did, and we've talked about it in IRC before as well :) | 18:06 |
| JayF | the thing is, before her most recent patchset, IPA could do this to itself | 18:06 |
| JayF | lock the drive because the first step of the ISE (locking the drive) would work, but the second step (erasing it) failed for $reasons | 18:06 |
| JayF | but now IPA tries to unlock it itself if it's locked, which makes sense | 18:07 |
| TheJulia | JayF: w/r/t https://review.openstack.org/#/c/270902, you want default disabled? which would actually be the same behavior now right? | 18:07 |
| *** mgoddard has quit IRC | 18:07 | |
| JayF | TheJulia: yep, exactly | 18:07 |
| TheJulia | JayF: honestly, I think the only thing that would really be left is to wire an info pass-through from the configuration to allow operators to toggle it, I guess | 18:07 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: Wait for the interfaces to get IP addresses before inspection https://review.openstack.org/300548 | 18:07 |
| JayF | TheJulia: and mat128 and I have been talking about this for like 30m or so :) | 18:07 |
| dtantsur | JayF, this should be it ^^^ | 18:07 |
| TheJulia | JayF: sorry for the delay, been on a call for a while | 18:07 |
| JayF | TheJulia: nbd, yeah and I agree re: config passthrough | 18:08 |
| JayF | TheJulia: in fact, we're using a pattern kinda like that for metrics | 18:08 |
| JayF | TheJulia: on lookup, conductor ships back some configs to the agent, agent applies it and uses it for metrics | 18:08 |
| dtantsur | and as my internet provider hates me, I have to call it a day (hopefully nobody pinged me while my connection was dropping) | 18:08 |
| JayF | TheJulia: you could probably use the same pattern if you wanted | 18:08 |
| TheJulia | JayF: okay, works for me | 18:08 |
| *** dtantsur is now known as dtantsur|afk | 18:08 | |
| mat128 | dtantsur|afk: didnt see anything | 18:09 |
| *** keedya has joined #openstack-ironic | 18:09 | |
| keedya | Hi All 0/ | 18:09 |
| *** Nisha_brb has joined #openstack-ironic | 18:09 | |
| mat128 | TheJulia: ping me once your changes are in | 18:10 |
| *** e0ne has quit IRC | 18:10 | |
| JayF | alineb: I'm going to upstream the ironic metrics patch today so I can show that pattern to TheJulia, but feel free to update it if it has comments and I'm not around | 18:10 |
| TheJulia | mat128: w/r/t what I was just talking to jayf about? | 18:10 |
| TheJulia | sorry, too much context switching :) | 18:10 |
| keedya | @here how can I enable nde_not_found hook in inspector? | 18:10 |
| mat128 | TheJulia: yes, JayF and I were talking about it while you were away | 18:11 |
| mat128 | seems like all of us agree now | 18:11 |
| JayF | mat128: you can add yourself as a reviewer, just hit "Add..." beside reviewers, and you'll get an email when it updates | 18:11 |
| TheJulia | ahh, cool | 18:11 |
| mat128 | JayF: good idea | 18:11 |
| mat128 | thx | 18:11 |
| mat128 | Just noticed I already am :) | 18:11 |
| JayF | heh, perfect | 18:11 |
| TheJulia | mat128: likely will be one of the things I do first thing in the morning, or later tonight if I my signifigant other lets me have more coffee :) | 18:11 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: Add support for building ISO for deploy ramdisk https://review.openstack.org/264590 | 18:12 |
| *** Nisha_away has quit IRC | 18:13 | |
| *** Sukhdev has joined #openstack-ironic | 18:13 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [iboot] https://review.openstack.org/301856 | 18:13 |
| keedya | iS example plugin in Ironic-inspector enabled by default? | 18:14 |
| *** Nisha_away has joined #openstack-ironic | 18:16 | |
| *** Nisha_brb has quit IRC | 18:19 | |
| *** piet has quit IRC | 18:21 | |
| keedya | nvm got it!! | 18:23 |
| *** piet has joined #openstack-ironic | 18:27 | |
| *** arturas has quit IRC | 18:30 | |
| *** Sukhdev has quit IRC | 18:31 | |
| *** rebase has quit IRC | 18:40 | |
| *** rcernin has quit IRC | 18:41 | |
| *** rebase has joined #openstack-ironic | 18:41 | |
| *** e0ne has joined #openstack-ironic | 18:55 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Add connections to Server Profile https://review.openstack.org/299579 | 18:58 |
| *** jaybeale has quit IRC | 19:03 | |
| *** alexpilotti has joined #openstack-ironic | 19:04 | |
| *** e0ne has quit IRC | 19:05 | |
| *** alexpilotti has quit IRC | 19:05 | |
| *** Sukhdev has joined #openstack-ironic | 19:06 | |
| *** mgoddard_ has quit IRC | 19:10 | |
| *** ametts has quit IRC | 19:17 | |
| *** e0ne has joined #openstack-ironic | 19:18 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic: Follow-up to I244c3f31d0ad26194887cfb9b79f96b5111296c6 https://review.openstack.org/301876 | 19:26 |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Added functions to translate between uuid and uri https://review.openstack.org/301877 | 19:30 |
| *** alexpilotti has joined #openstack-ironic | 19:31 | |
| *** ametts has joined #openstack-ironic | 19:31 | |
| *** Goneri has quit IRC | 19:35 | |
| *** e0ne has quit IRC | 19:35 | |
| *** alexpilotti has quit IRC | 19:36 | |
| openstackgerrit | Merged openstack/ironic-lib: Clean up test-requirements https://review.openstack.org/289294 | 19:37 |
| *** Nisha_away has quit IRC | 19:41 | |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: Allow to set neutron port timeout from config https://review.openstack.org/293876 | 19:44 |
| openstackgerrit | Merged openstack/ironic: Updated from global requirements https://review.openstack.org/301657 | 19:45 |
| openstackgerrit | Merged openstack/ironic: Don't power off non-deploying iLO nodes in takeover https://review.openstack.org/299573 | 19:46 |
| openstackgerrit | Marcellin Fom Tchassem proposed openstack/ironic-inspector: Make tox respect upper-constraints.txt https://review.openstack.org/301700 | 19:49 |
| *** jaybeale has joined #openstack-ironic | 19:49 | |
| *** Goneri has joined #openstack-ironic | 19:55 | |
| *** dprince has quit IRC | 19:57 | |
| *** MattMan has joined #openstack-ironic | 20:05 | |
| *** MattMan has quit IRC | 20:08 | |
| *** MattMan has joined #openstack-ironic | 20:12 | |
| openstackgerrit | Andre Aranha proposed openstack/python-oneviewclient: Refactor of unit Tests https://review.openstack.org/301895 | 20:15 |
| *** Sukhdev has quit IRC | 20:25 | |
| JayF | So have a general question re: metrics spec. http://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/add-pluggable-metrics-backend-for-ironic-and-ipa.html | 20:28 |
| JayF | we have that currently setup that all the metrics config is under [metrics], including IPA-specific metrics | 20:28 |
| *** moshele has joined #openstack-ironic | 20:28 | |
| JayF | If we do it that way, that means all those config vars shuold be declared in ironic-lib, right? | 20:28 |
| *** david-lyle_ has joined #openstack-ironic | 20:28 | |
| JayF | And I was toying with the idea that maybe the agent metrics configs should be under [agent], but I think I've convinced myself against that since more than just agent deploy driver can use agents + metrics | 20:29 |
| *** david-lyle has quit IRC | 20:29 | |
| *** david-lyle_ is now known as david-lyle | 20:29 | |
| *** rolfb_ has joined #openstack-ironic | 20:32 | |
| TheJulia | JayF: I can agree with everything you've stated, and I think I'm starting to get a headache | 20:34 |
| TheJulia | I think the question really shouldn't be agent deploy driver as much as it is use of the agent in general, and the deployment method used, since we're trying to get rid of the bash ramdisk, eventually.... | 20:35 |
| JayF | yep | 20:36 |
| *** moshele has quit IRC | 20:36 | |
| *** Sukhdev has joined #openstack-ironic | 20:36 | |
| TheJulia | JayF: then the headache begins to set in, and yoda saying something about paint and corners | 20:37 |
| JayF | go lie down if you've got that bad of a headache :/ | 20:37 |
| TheJulia | :) | 20:38 |
| openstackgerrit | Julia Kreger proposed openstack/ironic: Active Node Creation via adopt state https://review.openstack.org/275766 | 20:45 |
| openstackgerrit | Thiago Paiva Brito proposed openstack/python-oneviewclient: OneView client with managers https://review.openstack.org/286803 | 20:47 |
| *** Goneri has quit IRC | 20:48 | |
| *** rolfb_ has quit IRC | 20:48 | |
| *** rolfb_ has joined #openstack-ironic | 20:48 | |
| *** baoli_ has quit IRC | 20:51 | |
| *** karimb has quit IRC | 20:51 | |
| mrda | Morning Ironic | 20:53 |
| *** raildo is now known as raildo-afk | 20:54 | |
| *** trown is now known as trown|outtypewww | 20:59 | |
| *** rolfb_ has quit IRC | 21:03 | |
| *** rolfb_ has joined #openstack-ironic | 21:04 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Pass agent metrics config via conductor https://review.openstack.org/301919 | 21:06 |
| *** rcrit has joined #openstack-ironic | 21:07 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Pass agent metrics config via conductor https://review.openstack.org/301919 | 21:07 |
| rcrit | I'm trying to trace how ssh keys are injected into instances. I can see where it is done in libvirt. In ironic does a key get into /root/.ssh directly or only via the config drive? | 21:09 |
| JayF | Only via the configdrive / metadata service. | 21:11 |
| rcrit | ok thanks, I didn't want to assume | 21:12 |
| JayF | no problem, smart move :) | 21:12 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Metric chassis, driver, node, and port API calls https://review.openstack.org/301923 | 21:13 |
| JayF | people who assume get instances without ssh keys | 21:13 |
| JayF | lol | 21:13 |
| rcrit | heh | 21:13 |
| rcrit | JayF, so followup. Does the image need to do anything special to mount the config drive and access the keys? | 21:13 |
| * rcrit is fine with RTFM | 21:13 | |
| JayF | well in that case ... | 21:13 |
| JayF | google://cloud-init | 21:13 |
| JayF | :P | 21:14 |
| rcrit | heh | 21:14 |
| JayF | Most people use cloud-init to provision instances | 21:14 |
| rcrit | sure ok | 21:14 |
| JayF | there are newer things which do less, and I think infra uses it | 21:14 |
| JayF | but I don't recall the name of it | 21:14 |
| rcrit | so nothing special is needed inside the image, I can use same image for both h/w and virt, that's what I was going for | 21:14 |
| *** rbrud has quit IRC | 21:14 | |
| rcrit | jsut cloud-init would be different to mount the drive and pull out what is needed | 21:14 |
| JayF | oh yeah absolutely | 21:14 |
| *** rbrud has joined #openstack-ironic | 21:14 | |
| JayF | configdrive layout is an api | 21:15 |
| JayF | same with metadata service | 21:15 |
| *** jaypipes has quit IRC | 21:15 | |
| JayF | and it's date-versioned and such as well | 21:15 |
| rcrit | yeah, I've seen those | 21:15 |
| JayF | Ironic doesn't even build the configdrive; nova does | 21:15 |
| rcrit | but libvirt, for example, automagically sticks the key into authorized_keys | 21:15 |
| JayF | oh yeah, no craziness like that here. In fact if you use whole disk images, we never even mount your FS | 21:15 |
| rcrit | ok | 21:15 |
| * JayF just got confirmation from $bosspeoples that he's dedicated to upstream Ironic, at least until I'm not | 21:16 | |
| *** rbrud has quit IRC | 21:16 | |
| JayF | \o/ | 21:16 |
| *** rolfb_ is now known as rbrud | 21:16 | |
| rcrit | woo? | 21:16 |
| JayF | I'm quite thrilled about it :) | 21:16 |
| *** rolfb_ has joined #openstack-ironic | 21:17 | |
| cinerama | yay JayF | 21:17 |
| *** piet has quit IRC | 21:17 | |
| *** cppforlife_ has quit IRC | 21:19 | |
| *** alexpilotti has joined #openstack-ironic | 21:19 | |
| *** baoli has joined #openstack-ironic | 21:21 | |
| *** cppforlife_ has joined #openstack-ironic | 21:21 | |
| *** baoli_ has joined #openstack-ironic | 21:23 | |
| *** alexpilotti has quit IRC | 21:23 | |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 21:26 |
| *** baoli has quit IRC | 21:27 | |
| NobodyCam | JayF: awesome | 21:27 |
| NobodyCam | want to see if I addressed you concerns on ^^^^ | 21:27 |
| NobodyCam | lol | 21:27 |
| * NobodyCam *ducks* | 21:27 | |
| JayF | I had a vote on that? | 21:28 |
| * JayF looks suspiciously at NobodyCam | 21:28 | |
| JayF | NobodyCam: you know my vote in ironic-proper won't get you anywhere :P | 21:28 |
| NobodyCam | -1 Jay Faulkner | 21:28 |
| NobodyCam | lol | 21:28 |
| *** rbrud has quit IRC | 21:30 | |
| *** rbrud has joined #openstack-ironic | 21:31 | |
| *** keedya has quit IRC | 21:33 | |
| JayF | NobodyCam: +1'd | 21:34 |
| NobodyCam | w00t :) lol tuff part was getting past the pep8 tests | 21:35 |
| JayF | I looked to try and solicit a payback review from you, but everything up that I care about is already -1'd for something | 21:35 |
| JayF | lol | 21:35 |
| NobodyCam | lol... I'm sure I owd you at least one drink at summit | 21:36 |
| NobodyCam | owe even | 21:36 |
| JayF | hah, I'll take reviews over drinks | 21:36 |
| JayF | I can buy my own drinks, I can't buy reviews :P | 21:36 |
| NobodyCam | lol | 21:36 |
| JayF | ...or can I? | 21:36 |
| NobodyCam | lol | 21:38 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Add metrics support to IPA https://review.openstack.org/119981 | 21:39 |
| rcrit | funny, I work on a different IPA | 21:41 |
| JayF | I mean, our acronym was even overloaded by our own company once IPA as Ironic Python Agent was a thing | 21:41 |
| JayF | /topic for an internal channel says something like "IPA will always mean beer and the agent" | 21:41 |
| rcrit | it is also an identity management server | 21:42 |
| NobodyCam | see and I thought IPA stood for: International Phonetic Alphabet | 21:43 |
| * rcrit prefers the beer | 21:43 | |
| NobodyCam | :) | 21:44 |
| * NobodyCam steps afk for a bit | 21:44 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Metric chassis, driver, node, and port API calls https://review.openstack.org/301923 | 21:45 |
| *** hparekh has quit IRC | 21:48 | |
| *** adreznec has quit IRC | 21:48 | |
| *** hparekh has joined #openstack-ironic | 21:52 | |
| *** adreznec has joined #openstack-ironic | 21:52 | |
| *** rebase_ has joined #openstack-ironic | 21:54 | |
| *** harshs has quit IRC | 21:55 | |
| *** harshs has joined #openstack-ironic | 21:56 | |
| *** rebase has quit IRC | 21:57 | |
| *** harshs has quit IRC | 21:58 | |
| *** harshs has joined #openstack-ironic | 21:59 | |
| *** harshs has quit IRC | 22:00 | |
| *** ndipanov has quit IRC | 22:05 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [ilo] https://review.openstack.org/301947 | 22:10 |
| *** rebase_ has quit IRC | 22:21 | |
| *** baoli_ has quit IRC | 22:22 | |
| *** rebase has joined #openstack-ironic | 22:22 | |
| JayF | Deva suggested on one of my patches that I use a modulevar for a deprecation message so that it only prints once per call. Any idea where I can find an example of this pattern? | 22:23 |
| JayF | I think I got it, but we'll see what folks think about it on review :) | 22:35 |
| devananda | JayF: ironic/db/api.py | 22:36 |
| JayF | I don't see anything like that in there | 22:36 |
| JayF | https://gist.github.com/jayofdoom/2f40498b78e158281aa3f8514091e7f5 is what I did | 22:37 |
| *** saripurigopi has joined #openstack-ironic | 22:37 | |
| JayF | but it certainly seems clunky so I'm welcome to better suggestions | 22:37 |
| *** cdearborn has quit IRC | 22:37 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [inspector] https://review.openstack.org/301955 | 22:38 |
| JayF | I will make that bool an array of drivers that I've printed the warning for | 22:40 |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: [WIP] Add notification base and example to ironic https://review.openstack.org/298461 | 22:54 |
| *** ametts has quit IRC | 22:58 | |
| *** jayc has quit IRC | 22:59 | |
| *** Sukhdev has quit IRC | 23:00 | |
| openstackgerrit | Ramamani Yeleswarapu proposed openstack/ironic: Centralize config options - [ipmi] https://review.openstack.org/301967 | 23:12 |
| *** mdorman has quit IRC | 23:14 | |
| *** Marga__ has joined #openstack-ironic | 23:15 | |
| *** Marga__ has quit IRC | 23:15 | |
| *** garthb has joined #openstack-ironic | 23:18 | |
| *** Marga_ has quit IRC | 23:18 | |
| *** Marga_ has joined #openstack-ironic | 23:25 | |
| *** Sukhdev has joined #openstack-ironic | 23:25 | |
| *** rebase has quit IRC | 23:25 | |
| *** Goneri has joined #openstack-ironic | 23:36 | |
| *** Marga_ has quit IRC | 23:37 | |
| *** Marga_ has joined #openstack-ironic | 23:44 | |
| *** davideagnello has quit IRC | 23:46 | |
| *** Marga_ has quit IRC | 23:46 | |
| *** Marga_ has joined #openstack-ironic | 23:46 | |
| *** Marga_ has quit IRC | 23:50 | |
| *** Marga_ has joined #openstack-ironic | 23:51 | |
| *** Marga_ has quit IRC | 23:56 | |
| *** smoriya_ has joined #openstack-ironic | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!