| rpioso | TheJulia: Thank you! | 00:03 |
|---|---|---|
| *** k_mouza has joined #openstack-ironic | 00:08 | |
| *** k_mouza has quit IRC | 00:10 | |
| *** rloo has quit IRC | 00:15 | |
| *** bfournie has joined #openstack-ironic | 00:16 | |
| *** gyee has quit IRC | 00:58 | |
| *** ociuhandu has joined #openstack-ironic | 01:04 | |
| *** ociuhandu has quit IRC | 01:09 | |
| *** cdearborn has quit IRC | 01:43 | |
| *** jdandrea has quit IRC | 01:51 | |
| *** Goneri has quit IRC | 02:00 | |
| *** k_mouza has joined #openstack-ironic | 02:10 | |
| *** k_mouza has quit IRC | 02:15 | |
| *** igordc has joined #openstack-ironic | 02:33 | |
| *** ociuhandu has joined #openstack-ironic | 02:34 | |
| *** ociuhandu has quit IRC | 02:39 | |
| *** hamzy has quit IRC | 02:46 | |
| *** hamzy has joined #openstack-ironic | 02:46 | |
| *** igordc has quit IRC | 03:16 | |
| openstackgerrit | Merged openstack/ironic master: Use auth values from neutron conf when managing Neutron ports https://review.opendev.org/708678 | 03:16 |
| *** mkrai has joined #openstack-ironic | 03:23 | |
| *** mkrai has joined #openstack-ironic | 03:27 | |
| *** tzumainn has quit IRC | 03:53 | |
| openstackgerrit | Merged openstack/ironic-python-agent master: Use unittest mock from standard library https://review.opendev.org/717709 | 04:21 |
| *** ricolin has joined #openstack-ironic | 05:30 | |
| iurygregory | good morning Ironic o/ | 06:23 |
| arne_wiebalck | Good morning, iurygregory and ironic! | 06:44 |
| iurygregory | arne_wiebalck, o/ | 06:51 |
| *** rpittau|afk is now known as rpittau | 06:56 | |
| rpittau | good morning ironic! o/ | 06:56 |
| iurygregory | morning rpittau o/ | 06:57 |
| rpittau | hey iurygregory :) | 06:57 |
| *** uzumaki has joined #openstack-ironic | 07:07 | |
| *** hjensas has quit IRC | 07:07 | |
| uzumaki | good morning ironic! o/ | 07:07 |
| iurygregory | morning uzumaki | 07:08 |
| uzumaki | how you doing iurygregory ? | 07:08 |
| iurygregory | doing good and you? | 07:08 |
| uzumaki | i'm great | 07:08 |
| *** hjensas has joined #openstack-ironic | 07:21 | |
| *** uzumaki has quit IRC | 07:29 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-lib master: Convert jobs to dib https://review.opendev.org/716998 | 07:34 |
| *** diga has joined #openstack-ironic | 07:40 | |
| *** ociuhandu has joined #openstack-ironic | 07:45 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-prometheus-exporter master: Use hacking for pep8 tests https://review.opendev.org/717689 | 07:50 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-prometheus-exporter master: Use hacking for pep8 tests https://review.opendev.org/717689 | 07:51 |
| *** alexmcleod has joined #openstack-ironic | 08:00 | |
| *** mkrai has quit IRC | 08:01 | |
| *** hjensas is now known as hjensas|afk | 08:04 | |
| *** uzumaki has joined #openstack-ironic | 08:07 | |
| kaifeng | morning everyone | 08:08 |
| kaifeng | have anyone paid attention to the memory consumption of ironic conductor? it seems consuming more than expected | 08:10 |
| kaifeng | system shows ~277M mem usage even in the idle state | 08:10 |
| arne_wiebalck | kaifeng: yes | 08:10 |
| arne_wiebalck | kaifeng: I have raised this a year ago or so | 08:11 |
| arne_wiebalck | kaifeng: I see the exact same thing | 08:11 |
| *** k_mouza has joined #openstack-ironic | 08:11 | |
| kaifeng | arne_wiebalck: i have removed some drivers but it seems doesn't help | 08:11 |
| arne_wiebalck | kaifeng: at the time I discussed this briefly with dtantsur|afk , but we never got to the bottom of this | 08:12 |
| arne_wiebalck | kaifeng: I am not sure there are any traces of our discussion from back then, let me check ... | 08:13 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Collect tftpd info from journald https://review.opendev.org/717745 | 08:14 |
| rpittau | hey kaifeng :) | 08:14 |
| kaifeng | arne_wiebalck: wow, thanks! | 08:14 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Collect tftpd info from journald https://review.opendev.org/717745 | 08:14 |
| kaifeng | hi rpittau | 08:14 |
| *** k_mouza has quit IRC | 08:15 | |
| *** tkajinam has quit IRC | 08:17 | |
| *** mkrai has joined #openstack-ironic | 08:20 | |
| arne_wiebalck | kaifeng: here's what I saw at the time when changing parameters: | 08:21 |
| arne_wiebalck | 16 processes/1 thread each --> 300MB/process --> 4.8GB total | 08:21 |
| arne_wiebalck | 4 processes/4 threads each --> 900MB/process --> 3.6GB total | 08:21 |
| arne_wiebalck | 1 process / 16 threads --> 2.1GB/process --> 2.1GB total | 08:21 |
| *** lucasagomes has joined #openstack-ironic | 08:22 | |
| arne_wiebalck | kaifeng: the DB size was around 50MB | 08:22 |
| kaifeng | arne_wiebalck: how do you get the consumption of each part, say, DB? | 08:23 |
| kaifeng | we are evaluating an edge case and running conductor in a VM which has only two vcpu | 08:24 |
| openstackgerrit | Merged openstack/networking-baremetal master: Cleanup py27 support https://review.opendev.org/717615 | 08:25 |
| arne_wiebalck | kaifeng: for the DB I probably ran some command to sum up the contents; this was mostly to show that the one would need to DB to lay out in memory *many* times to explain the RSS foot print | 08:25 |
| arne_wiebalck | kaifeng: in our qa env, the DB was 25 times smaller, but the RAM consumption only a factor of 3 | 08:26 |
| arne_wiebalck | kaifeng: let me check what the consumption is right now in our prod nodes ... | 08:27 |
| kaifeng | arne_wiebalck: in my case ironic doesn't manage any node at the time, just installed, configured and running idle | 08:27 |
| arne_wiebalck | kaifeng: yes ... we also saw it is jumping: it starts low, then all of a sudden it jumps to 300MB RSS and stays there | 08:28 |
| arne_wiebalck | kaifeng: probably when some periodic tasks kicks in and it loads some lib | 08:28 |
| kaifeng | arne_wiebalck: i see this jumping after service restart, but it will drop after a while | 08:29 |
| kaifeng | arne_wiebalck: periodic task consumes cpu (though it's also a bit higher than other services), i guess most of our periodic tasks are just doing db query | 08:31 |
| * arne_wiebalck just realised it is almost 2 years ago we noticed this | 08:32 | |
| * kaifeng feels like a profiling is required :( | 08:33 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: Switch to unittest mock https://review.opendev.org/717979 | 08:34 |
| *** uzumaki has quit IRC | 08:36 | |
| *** derekh has joined #openstack-ironic | 08:38 | |
| *** mkrai has quit IRC | 08:38 | |
| *** mkrai_ has joined #openstack-ironic | 08:38 | |
| *** ociuhandu has quit IRC | 08:50 | |
| *** ociuhandu has joined #openstack-ironic | 08:57 | |
| *** dtantsur|afk is now known as dtantsur | 09:01 | |
| *** priteau has joined #openstack-ironic | 09:01 | |
| dtantsur | morning ironic | 09:01 |
| arne_wiebalck | kaifeng: I made some attempts with pyrasite at the time but did not get very far | 09:03 |
| arne_wiebalck | kaifeng: in any case, let me know in case you find something: I just checked and our 16 worker API process use 3.4GB RSS | 09:05 |
| kaifeng | arne_wiebalck: np, i will let you know if there is any finding | 09:07 |
| iurygregory | morning dtantsur | 09:25 |
| *** k_mouza has joined #openstack-ironic | 09:37 | |
| *** k_mouza has quit IRC | 09:49 | |
| *** k_mouza has joined #openstack-ironic | 09:57 | |
| *** k_mouza has quit IRC | 09:57 | |
| *** k_mouza has joined #openstack-ironic | 09:57 | |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `network_data` field to ironic node object https://review.opendev.org/687910 | 10:04 |
| openstackgerrit | Pavlo Shchelokovskyy proposed openstack/ironic master: Allow INSPECTWAIT state for lookup https://review.opendev.org/718017 | 10:12 |
| *** ricolin has quit IRC | 10:16 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: DRAC: Fix a failure to create virtual disk bug https://review.opendev.org/710852 | 10:25 |
| *** rpittau is now known as rpittau|bbl | 10:25 | |
| *** k_mouza has quit IRC | 10:42 | |
| *** k_mouza has joined #openstack-ironic | 10:46 | |
| *** rachit7 has joined #openstack-ironic | 10:49 | |
| *** rachit7 has quit IRC | 11:03 | |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Validate node `network_data` field contents https://review.opendev.org/714614 | 11:12 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/ironic master: Switch to unittest mock https://review.opendev.org/717979 | 11:15 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `get_node_network_data` method to NetworkInterface https://review.opendev.org/688184 | 11:16 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `get_node_network_data` to non-Neutron NetworkInterface https://review.opendev.org/688266 | 11:16 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Make redfish boot interface to use `network_data` https://review.opendev.org/688272 | 11:17 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient master: Standalone CLI: optional support for ironic-inspector https://review.opendev.org/718067 | 11:27 |
| dtantsur | TheJulia: ^^^ | 11:33 |
| *** ociuhandu has quit IRC | 11:41 | |
| *** ociuhandu has joined #openstack-ironic | 11:42 | |
| *** ociuhandu has quit IRC | 11:47 | |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Support burning configdrive into boot ISO https://review.opendev.org/687250 | 11:48 |
| *** rh-jelabarre has joined #openstack-ironic | 11:48 | |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `network_data` field to ironic node object https://review.opendev.org/687910 | 11:48 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Validate node `network_data` field contents https://review.opendev.org/714614 | 11:48 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `get_node_network_data` method to NetworkInterface https://review.opendev.org/688184 | 11:49 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `get_node_network_data` to non-Neutron NetworkInterface https://review.opendev.org/688266 | 11:49 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Make redfish boot interface to use `network_data` https://review.opendev.org/688272 | 11:49 |
| * etingof has been bitten by a tick - Spring! | 11:50 | |
| iurygregory | etingof, ouch =( I hope you are ok | 11:52 |
| dtantsur | etingof: oh. where did you manage to get it? | 11:53 |
| * etingof wishes that the tick has died in a good health | 11:56 | |
| etingof | I went to the woods the other day, might be from there | 11:57 |
| *** dougsz has joined #openstack-ironic | 12:03 | |
| *** rpittau|bbl is now known as rpittau | 12:06 | |
| *** tkajinam has joined #openstack-ironic | 12:08 | |
| *** yaawang has quit IRC | 12:08 | |
| *** hjensas|afk is now known as hjensas | 12:10 | |
| * arne_wiebalck removed the first this year from his cat as well some days ago | 12:11 | |
| dtantsur | sigh | 12:11 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: [WIP] Use mock from unittest https://review.opendev.org/718080 | 12:17 |
| openstackgerrit | Merged openstack/ironic master: Do not autoescape all Jinja2 templates https://review.opendev.org/716845 | 12:23 |
| openstackgerrit | Verification of a change to openstack/ironic-inspector failed: Move Babel to test-requirements https://review.opendev.org/717743 | 12:23 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Replace netaddr with ipaddress https://review.opendev.org/717852 | 12:24 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient master: [WIP] Get rid of oslo.config requirements https://review.opendev.org/717724 | 12:30 |
| arne_wiebalck | dtantsur: unless I am missing sth https://review.opendev.org/#/c/701926/ fell through the cracks I think | 12:33 |
| patchbot | patch 701926 - ironic - [doc] Images need some metadata for software RAID - 2 patch sets | 12:33 |
| *** igordc has joined #openstack-ironic | 12:34 | |
| dtantsur | I'll check, thanks | 12:34 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add `get_node_network_data` to Neutron NetworkInterface https://review.opendev.org/688896 | 12:34 |
| arne_wiebalck | dtantsur: would be good to have this (or something improved if needed) in before the release since it describes the requirements on images to be used with software RAID | 12:35 |
| arne_wiebalck | dtantsur: thanks | 12:35 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Add L3 boot section to the docs https://review.opendev.org/689844 | 12:35 |
| dtantsur | arne_wiebalck: first comment: I think we have a fallback to the old approach, no? | 12:35 |
| dtantsur | second comment: does it work with standalone ironic? | 12:36 |
| arne_wiebalck | dtantsur: do partition images work with standalone ironic | 12:36 |
| arne_wiebalck | ? | 12:36 |
| dtantsur | arne_wiebalck: perfectly fine | 12:37 |
| arne_wiebalck | dtantsur: then this should also work, I believe | 12:37 |
| dtantsur | arne_wiebalck: where will rootfs_uuid come from in this case? | 12:37 |
| arne_wiebalck | dtantsur: where does it come from for partition images> | 12:37 |
| arne_wiebalck | ? | 12:37 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-prometheus-exporter master: Use hacking for pep8 tests https://review.opendev.org/717689 | 12:38 |
| dtantsur | arne_wiebalck: we know it because we create partitioning. Which makes me realize, do we even support partition images with sw RAID? | 12:38 |
| arne_wiebalck | dtantsur: I think we explicitly say, no support | 12:39 |
| dtantsur | okay, so rootfs_uuid is needed for whole disk images? | 12:39 |
| arne_wiebalck | dtantsur: but this mostly since noone ever tried | 12:39 |
| dtantsur | I don't think we require it for non-RAID case | 12:39 |
| arne_wiebalck | mind you, this is only the documentation :) | 12:40 |
| dtantsur | arne_wiebalck: yeah, but I'm trying to understand if rootfs_uuid has to be provided somehow for the standalone case (and document it) | 12:40 |
| arne_wiebalck | dtantsur: IIRC, the code will try to find the UUID first from glance, then ... let me check ... | 12:41 |
| dtantsur | yep, and if there's no glance, we only have the fallback? | 12:42 |
| *** rloo has joined #openstack-ironic | 12:44 | |
| arne_wiebalck | yes, we fall back to internal_info | 12:45 |
| *** ociuhandu has joined #openstack-ironic | 12:45 | |
| dtantsur | arne_wiebalck: well, internal_info cannot be modified, so we expect the first partition? | 12:45 |
| arne_wiebalck | dtantsur: on the conductor side, I think internal_info was where we looked before | 12:46 |
| arne_wiebalck | dtantsur: so, this is what the code does when glance fails | 12:46 |
| arne_wiebalck | dtantsur: on the IPA side I think we fall back to the first partition, let me check ... | 12:47 |
| arne_wiebalck | dtantsur: yes, if we do not manage to find the partition by uuid, we use the first one | 12:51 |
| openstackgerrit | Merged openstack/networking-baremetal master: Use mock from unittest https://review.opendev.org/717674 | 12:51 |
| dtantsur | okay, let's make it clear in the docuemtnation | 12:51 |
| *** ociuhandu has quit IRC | 12:52 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: Use mock from unittest https://review.opendev.org/718080 | 12:53 |
| arne_wiebalck | dtantsur: so we will keep this fallback? | 12:54 |
| arne_wiebalck | dtantsur: I mean longterm :) | 12:54 |
| dtantsur | arne_wiebalck: we probably should. definitely until we have a solution for the standalone case. | 12:55 |
| arne_wiebalck | dtantsur: so, using the 1st partition will help with the standalone case | 12:55 |
| arne_wiebalck | dtantsur: probably we should also mark this in the code as such, to avoid it gets removed by some cleanup later | 12:56 |
| arne_wiebalck | dtantsur: did you have a chance to try standalone with s/w RAID? | 12:56 |
| arne_wiebalck | dtantsur: or know someone who did? | 12:56 |
| *** ociuhandu has joined #openstack-ironic | 12:59 | |
| dtantsur | no and no, unfortunately | 13:00 |
| *** ociuhandu has quit IRC | 13:00 | |
| *** ociuhandu has joined #openstack-ironic | 13:01 | |
| *** jcoufal has joined #openstack-ironic | 13:03 | |
| *** mkrai_ has quit IRC | 13:06 | |
| TheJulia | good morning | 13:07 |
| TheJulia | This whole tick business is why I live in the desert.... | 13:10 |
| iurygregory | good morning TheJulia | 13:10 |
| dtantsur | morning TheJulia | 13:11 |
| rpittau | hey TheJulia :) | 13:12 |
| dtantsur | I just cover myself in horribly-smelling repellents every time I venture out | 13:12 |
| openstackgerrit | Merged openstack/ironic-python-agent master: Editing follow-up for UEFI Software RAID support https://review.opendev.org/717787 | 13:18 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient master: Get rid of the oslo.config requirement https://review.opendev.org/717724 | 13:19 |
| dtantsur | another big dependency cut ^^ | 13:19 |
| *** jdandrea has joined #openstack-ironic | 13:20 | |
| *** Goneri has joined #openstack-ironic | 13:23 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Replace netaddr with ipaddress https://review.opendev.org/717852 | 13:25 |
| dtantsur | if anybody has a minute, this is probably an easy fix: https://storyboard.openstack.org/#!/story/2007526 | 13:30 |
| dtantsur | any opinions if https://review.opendev.org/#/c/717724/ needs a release note? | 13:31 |
| patchbot | patch 717724 - python-ironicclient - Get rid of the oslo.config requirement - 3 patch sets | 13:31 |
| dtantsur | there may be discrepancies in what the new approach supports.. | 13:32 |
| TheJulia | that bug looks easy | 13:32 |
| *** ociuhandu has quit IRC | 13:32 | |
| *** ociuhandu has joined #openstack-ironic | 13:33 | |
| rpittau | dtantsur: I was going to add a comment, I'm for a release note | 13:33 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient master: Get rid of the oslo.config requirement https://review.opendev.org/717724 | 13:33 |
| dtantsur | rpittau: something alone these lines ^^^? | 13:33 |
| TheJulia | I'm for, just a minor one should be fine | 13:33 |
| dtantsur | * along | 13:33 |
| dtantsur | wording suggestions welcome | 13:33 |
| TheJulia | "This version of ironic no longer requires oslo.config. Configuration loading logic was migrated to the already utilized openstacksdk library." | 13:35 |
| dtantsur | hjensas: can https://bugs.launchpad.net/tripleo/+bug/1871346 be an effect of your recent patch? | 13:35 |
| openstack | Launchpad bug 1871346 in tripleo "Ironic nodes registration failing with error - ironicclient.common.apiclient.exceptions.InternalServerError: 'NoneType' object has no attribute 'keys'" [Critical,Triaged] | 13:35 |
| dtantsur | TheJulia: "configuration loading" may be a big vague | 13:35 |
| dtantsur | reviews of https://review.opendev.org/#/c/718067/ are also appreciated | 13:38 |
| patchbot | patch 718067 - python-ironicclient - Standalone CLI: optional support for ironic-inspector - 1 patch set | 13:38 |
| hjensas | dtantsur: It looks like the local_link_connection changes I did is a suspect. yes. | 13:39 |
| dtantsur | mind taking a look? | 13:40 |
| hjensas | dtantsur: I will push the patch to drop netaddr, then look at it. | 13:40 |
| TheJulia | hjensas: already did | 13:44 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: Drop netaddr - use netutils.is_valid_ipv6() https://review.opendev.org/718113 | 13:44 |
| * TheJulia shrugs | 13:44 | |
| hjensas | TheJulia: that one ^^, or the use in the dual-stack patch? | 13:45 |
| TheJulia | hjensas: https://review.opendev.org/717852 | 13:45 |
| patchbot | patch 717852 - ironic - Replace netaddr with ipaddress - 2 patch sets | 13:45 |
| dtantsur | TheJulia: the ipaddress change is not backportable, unfortunately | 13:46 |
| TheJulia | ahh | 13:46 |
| *** tzumainn has joined #openstack-ironic | 13:46 | |
| dtantsur | not without adding a dependency on the python 2 backport | 13:46 |
| TheJulia | true | 13:46 |
| TheJulia | yeah | 13:46 |
| TheJulia | okay | 13:46 |
| * TheJulia goes back to inspector bug | 13:46 | |
| dtantsur | so we may need to make it two-staged | 13:46 |
| hjensas | I think we can take both, first min then TheJulia's. :) | 13:46 |
| *** yaawang has joined #openstack-ironic | 13:47 | |
| dtantsur | right | 13:48 |
| * dtantsur will never stop dreaming of dropping oslo.utils :) | 13:48 | |
| * TheJulia points at code reviews | 13:49 | |
| TheJulia | I feel like the evil monkey from https://www.youtube.com/watch?v=fxynvzg03kM when I point out code reviews | 13:50 |
| dtantsur | lol | 13:54 |
| iurygregory | dtantsur, https://www.mememaker.net/api/bucket?path=static/img/memes/full/2020/Apr/7/13/drop-all-oslo-36622.png | 13:54 |
| dtantsur | :D | 13:54 |
| *** jcoufal has quit IRC | 13:54 | |
| dtantsur | to be clear, I don't suggest literally that | 13:54 |
| iurygregory | I know | 13:54 |
| iurygregory | XD | 13:54 |
| dtantsur | I'm trying to make people mindful about dependencies | 13:55 |
| iurygregory | ++ | 13:55 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-inspector master: Catch NotFound properly in the explicit inspect path https://review.opendev.org/718118 | 13:59 |
| TheJulia | dtantsur: ^^^ I think that is what you were hoping for? Looks like the testing is fine around that and searches the entire string in the stack trace, but since we log the exception and pass that the test still passes. | 14:00 |
| dtantsur | TheJulia: I don't think that's the right direction to take (although I only have gut feelings at this point) | 14:00 |
| dtantsur | NotFound inherits Error, so this change shouldn't change much | 14:01 |
| dtantsur | I suspect we don't process known exceptions correctly at the RPC boundary | 14:01 |
| *** tkajinam has quit IRC | 14:03 | |
| TheJulia | so utils.Error is all that is expected for messaging | 14:03 |
| *** tkajinam has joined #openstack-ironic | 14:03 | |
| TheJulia | so... I guess it should be good.. maybe? | 14:03 |
| dtantsur | TheJulia: NotFound inherits Error though? | 14:05 |
| dtantsur | I'm confused by the RemoteError bit in the traceback | 14:05 |
| *** etingof has quit IRC | 14:06 | |
| *** yaawang_ has joined #openstack-ironic | 14:11 | |
| TheJulia | dtantsur: NotFound is a bare declaration I think from Exception in the local ironic client | 14:11 |
| TheJulia | so it would result in a nice error since that is not expected | 14:12 |
| TheJulia | I thinik | 14:12 |
| *** yaawang has quit IRC | 14:14 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Debug logging for failed disk folder mount with _install_grub2 https://review.opendev.org/716603 | 14:15 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic master: [doc] Images need some metadata for software RAID https://review.opendev.org/701926 | 14:16 |
| arne_wiebalck | dtantsur: I tried to transform our discussion earlier into the doc patch ^. | 14:17 |
| TheJulia | yay rebasing patches | 14:22 |
| dtantsur | TheJulia: it's not a client side problem though. it comes as HTTP 500 to the client, I think because utils.Error gets translated into oslo's RemoteError. | 14:25 |
| dtantsur | but I can only speculate based on the error message | 14:25 |
| TheJulia | gah, my brain is not functioning today | 14:26 |
| TheJulia | tzumainn's patch redid the dhcp unit tests quite a bit :( | 14:26 |
| TheJulia | dtantsur: but utils.Error is classified as remote expected error | 14:27 |
| TheJulia | so something is not working right there :\ | 14:27 |
| *** etingof has joined #openstack-ironic | 14:30 | |
| tzumainn | TheJulia, did my patch break something? | 14:31 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: Fix AttributeError in check allowed port fields https://review.opendev.org/718127 | 14:31 |
| hjensas | dtantsur: I think ^^ should fix it. But I tried to write a test for it. create or a patch replacing llc with 'None' did'nt fail in unit test. | 14:33 |
| TheJulia | tzumainn: just the client mockings on unit testing for my make the pxe code friendly to dual stack configs | 14:36 |
| tzumainn | TheJulia, ah, let me know if I can help with anything! | 14:37 |
| *** priteau has quit IRC | 14:37 | |
| TheJulia | tzumainn: will do, I think I'll just have to have two mocks | 14:37 |
| TheJulia | because I actually have to look at the port's cconfiguration to determine what to send | 14:38 |
| *** ricolin has joined #openstack-ironic | 14:39 | |
| *** tkajinam has quit IRC | 14:42 | |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Revert "Cleanup py27 support" https://review.opendev.org/718136 | 14:46 |
| *** yaawang has joined #openstack-ironic | 14:53 | |
| *** yaawang__ has joined #openstack-ironic | 14:54 | |
| *** yaawang_ has quit IRC | 14:56 | |
| *** mkrai has joined #openstack-ironic | 14:57 | |
| *** yaawang has quit IRC | 14:58 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Refactor AgentBase.heartbeat and process_next_step https://review.opendev.org/718142 | 15:07 |
| * TheJulia sighs | 15:08 | |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: Fix AttributeError in check allowed port fields https://review.opendev.org/718127 | 15:10 |
| *** rcernin has quit IRC | 15:13 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: In-band deploy steps https://review.opendev.org/698768 | 15:14 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: "dual stack" support for PXE/iPXE https://review.opendev.org/714212 | 15:15 |
| openstackgerrit | Merged openstack/ironic master: Nodes in maintenance didn't fail, when they should have https://review.opendev.org/702927 | 15:17 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: Decompose core deploy step https://review.opendev.org/698309 | 15:18 |
| *** dking_desktop has joined #openstack-ironic | 15:23 | |
| *** gyee has joined #openstack-ironic | 15:31 | |
| openstackgerrit | Tzu-Mainn Chen proposed openstack/ironic master: Add node lessee field https://review.opendev.org/706864 | 15:34 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic master: Software RAID: Pass the boot mode to the IPA https://review.opendev.org/696169 | 15:39 |
| arne_wiebalck | TheJulia: ^^ is meant to address your point on the remaining UEFI RAID patch set | 15:40 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: Support executing in-band deploy steps https://review.opendev.org/698768 | 15:54 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: Decompose core deploy step https://review.opendev.org/698309 | 15:54 |
| dtantsur | this deploy steps work is an infinite rabbit hole >_< | 15:55 |
| rpittau | rabbitmq hole ? | 15:56 |
| dtantsur | :D | 15:56 |
| TheJulia | arne_wiebalck: thanks | 16:06 |
| markguz_ | HI there, anyone here know anything about building ironic-python-agent with the ipa-builder tool? I'm having a 'mare of a time getting it to work | 16:08 |
| rpittau | good night! o/ | 16:08 |
| *** rpittau is now known as rpittau|afk | 16:08 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: Support executing in-band deploy steps https://review.opendev.org/698768 | 16:08 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: WIP: Decompose core deploy step https://review.opendev.org/698309 | 16:08 |
| *** lucasagomes has quit IRC | 16:08 | |
| *** dtantsur is now known as dtantsur|afk | 16:08 | |
| dtantsur|afk | o/ | 16:08 |
| TheJulia | markguz_: what are you running into? | 16:08 |
| markguz_ | it keeps bombing out with and error about "python -c "import requests; print requests.certs.where()"): ambiguous redirect" | 16:08 |
| TheJulia | markguz_: what are the arguments your passing and env vars your using for it if any? | 16:09 |
| markguz_ | ironic-python-agent-builder -o my-ipa ubuntu | 16:09 |
| markguz_ | http://paste.openstack.org/show/791751/ | 16:10 |
| markguz_ | TheJulia: i tried the pip version of ipab, with no luck, so now i tried the git HEAD version.. same result. | 16:12 |
| TheJulia | markguz_: would you try centos8 and see if that works? | 16:12 |
| markguz_ | already did | 16:12 |
| markguz_ | nope | 16:12 |
| markguz_ | centos or centos-minimal 8 | 16:13 |
| TheJulia | I think centos defaults to centos7 unless you explicitly use centos8 | 16:13 |
| * TheJulia installs the latest builder to see what is going on | 16:14 | |
| markguz_ | yeah. i was supplying the --release 8 param | 16:14 |
| TheJulia | try the centos8 element instead, just to be on the safe side | 16:14 |
| markguz_ | k | 16:15 |
| *** diga has quit IRC | 16:15 | |
| *** ociuhandu has quit IRC | 16:18 | |
| *** ociuhandu has joined #openstack-ironic | 16:19 | |
| markguz_ | TheJulia: "Element 'centos8' not found" | 16:19 |
| markguz_ | it also failed with "--release 8 centos" | 16:19 |
| TheJulia | :( | 16:19 |
| * TheJulia tries to figure out why she is running out of disk space trying to install it in ~ | 16:20 | |
| markguz_ | http://paste.openstack.org/show/791753/ | 16:20 |
| *** mkrai has quit IRC | 16:20 | |
| markguz_ | i'm building this in a virtualenv on a CentOS 8 vm | 16:21 |
| markguz_ | if that makes any difference | 16:21 |
| markguz_ | python 3.6 | 16:21 |
| TheJulia | markguz_: it is using python2 internally :\ | 16:37 |
| markguz_ | yeah i was getting errors about python versions earlier, "different version required 2.17.17" something | 16:38 |
| *** ociuhandu has quit IRC | 16:42 | |
| TheJulia | have you tried forcing it with the dib env var for python3? | 16:43 |
| *** ociuhandu has joined #openstack-ironic | 16:43 | |
| markguz_ | hmmm. no | 16:45 |
| arne_wiebalck | we use DIB_PYTHON_VERSION=3 | 16:45 |
| rpittau|afk | markguz_: that is really weird, I'm running it on virtualenv on vms (ubuntu bionic and centos8) and it works fine for me using: ironic-python-agent-builder -o my-ipa --release 8 centos | 16:45 |
| markguz_ | rpittaulafk: TheJulia: i removed my other env vars and it worked for me, so one of my env vars seems to be breaking it | 16:46 |
| markguz_ | i just tried that just now | 16:46 |
| rpittau|afk | good :) | 16:46 |
| TheJulia | I'm guessing the pem bit | 16:47 |
| * rpittau|afk goes back in the shadows.... | 16:47 | |
| * arne_wiebalck leaves as well | 16:47 | |
| arne_wiebalck | bye everyone o/ | 16:47 |
| markguz_ | TheJulia: yep.. my guess too. | 16:48 |
| *** ociuhandu has quit IRC | 16:48 | |
| markguz_ | hmmph. i wanted that in there... i wonder if it's because i have the root and intermediate in the same file | 16:48 |
| markguz_ | yep. confirmed. b0rking on the pem file... | 16:55 |
| TheJulia | :( | 16:55 |
| markguz_ | thanks for your help.. I can live without the pem for now. I'll fiddle with it, see if it's a "bug" or user error | 16:57 |
| openstackgerrit | Ilya Etingof proposed openstack/ironic master: Fix gunicorn name on Py3@CentOS7 in devstack https://review.opendev.org/718173 | 16:57 |
| openstackgerrit | Merged openstack/python-ironicclient master: Get rid of the oslo.config requirement https://review.opendev.org/717724 | 16:59 |
| openstackgerrit | Tzu-Mainn Chen proposed openstack/ironic master: Added node multitenancy doc https://review.opendev.org/718174 | 17:02 |
| *** dougsz has quit IRC | 17:05 | |
| *** alexmcleod has quit IRC | 17:11 | |
| TheJulia | arne_wiebalck: thanks for the try/except catch add | 17:11 |
| arne_wiebalck | TheJulia: Thanks for the blueprint ;) | 17:13 |
| openstackgerrit | Harald Jensås proposed openstack/ironic master: Fix AttributeError in check allowed port fields https://review.opendev.org/718127 | 17:14 |
| TheJulia | :) | 17:17 |
| *** dougsz has joined #openstack-ironic | 17:21 | |
| *** dougsz has quit IRC | 17:22 | |
| *** k_mouza has quit IRC | 17:31 | |
| markguz_ | TheJulia: different question. Is it possible to set a tftp_server per network? i.e. one for the cleaning and one for the provisioning network? | 17:32 |
| *** dking_desktop is now known as dking | 17:51 | |
| *** dking has quit IRC | 17:56 | |
| *** dking_desktop has joined #openstack-ironic | 17:57 | |
| *** dking_desktop has quit IRC | 17:58 | |
| *** ricolin has quit IRC | 18:02 | |
| *** k_mouza has joined #openstack-ironic | 18:13 | |
| TheJulia | markguz_: not really, it is intended to be the conductor itself so firewalling rules . Is there a specific reason why that won't work? | 18:13 |
| *** k_mouza has quit IRC | 18:14 | |
| markguz_ | if you have separate cleaning and provisioning networks then you have to allow traffic from one to flow to the other in order for the tftp server to be reachable | 18:14 |
| markguz_ | which is not in and of itself a problem, but to my mind that kind of negates the two isolated networks thing... | 18:17 |
| *** cdearborn has joined #openstack-ironic | 18:19 | |
| TheJulia | markguz_: not between the networks if the conductor is on a third separate address external from the two | 18:36 |
| TheJulia | two networks that is | 18:36 |
| markguz_ | hmm. then i need a third network that can route to both. ideally there should be no routing... at least IMHO. anyway, i'll just stick with one clean/provide network and keep it simple | 18:38 |
| TheJulia | etingof: so I'm looking at https://storyboard.openstack.org/#!/story/2007527 and wondering if we shouldn't put logic in to handle if the system is in UEFI mode, to not send a continious option at all? | 18:40 |
| etingof | TheJulia, does UEFI mode correlate with Continues value not being accepted? | 18:43 |
| etingof | we can see if 'BootSourceOverrideEnabled@Redfish.AllowableValues' has Continues listed and apply it if it's there | 18:45 |
| *** dougsz has joined #openstack-ironic | 18:45 | |
| etingof | however this is confusing... at the sushy level we accept 'continues' flag, then at the redfish level we won't always pass it to the BMC... | 18:46 |
| etingof | *Continuous | 18:46 |
| rpioso | etingof: I believe 'Continuous' can be used with UEFI. Should we ask mraineri? | 18:47 |
| etingof | Continuous and UEFI/Legacy boot modes - are they anyhow exclusive? | 18:48 |
| etingof | to a layman it appears that all the combinations make sense... | 18:49 |
| * rpioso pinged mraineri out-of-band | 18:51 | |
| mraineri | Continuous has some restrictions | 18:51 |
| mraineri | It depends on the pairing of the boot override target | 18:52 |
| mraineri | Let me pull up the schema... It has the full restriction list... | 18:52 |
| etingof | considering all these complications, would it be easier in the end to operate on the boot order (as opposed to BootoverrideTarget)? | 18:53 |
| mraineri | If you're trying to set up a permanent boot order, yes | 18:53 |
| TheJulia | etingof: basically yes. Continious is not a possible option, and it seems the only non-override way to set it is via an oem-specific labeled attribute option under bios | 18:55 |
| mraineri | This is one of the pieces of text in the schema: "The `Continuous` value is not supported for this property for UEFI boot source override because UEFI supports one-time boot only." | 18:55 |
| rpioso | etingof, mraineri: Might the boot order require the use of vendor-specific names for the boot sources? | 18:55 |
| mraineri | And the same rule applies to UefiBootNext | 18:55 |
| TheJulia | lenovo's example code has 4 different attribute names it checks :( | 18:55 |
| mraineri | It might, but there should also be a more well-defined enum in there to classify the type of boot option | 18:56 |
| mraineri | Let me look at that now... | 18:56 |
| TheJulia | And the conundrum is we're trying to set a permanent override and it looks like the alternative options are all vendor specific without support for the field. :( | 18:56 |
| mraineri | The "Alias" property will tell you if a given boot option is a CD, USB, drive, etc | 18:57 |
| mraineri | It's the same enum used in the boot override property | 18:57 |
| TheJulia | except that all seems one-time based on the 1.9 version document I've got on my other screen | 19:00 |
| etingof | if we go the boot-order way, my understanding is that if we tell sushy to boot from Hdd, under the hood sushy will reorder all boot devices so that those mapping to Hdd alias will appear first in the list | 19:03 |
| etingof | I am saying that we probably can't set the node to boot straight from the disk we have deployed the image | 19:04 |
| etingof | we can only operate on "kinds" of boot media (Cd, Hdd), not concrete devices | 19:05 |
| mraineri | Then that might be a reason to use the AliasBootOrder property | 19:05 |
| etingof | yeah | 19:06 |
| rpioso | etingof: How's that different from override? | 19:06 |
| rpioso | And doesn't UEFI find the boot sources available on the drives? | 19:06 |
| etingof | rpioso, I think this imprecise boot media addressing problem is the same in both boot target override and boot order cases | 19:07 |
| rpioso | etingof: Me, too. It's typically good enough. | 19:09 |
| TheJulia | Yeah, AlaisBootOrder seems to be the right way to go since we don't know the internal uefi names for the targets | 19:09 |
| etingof | that's ^ what I mean | 19:09 |
| TheJulia | rpioso: one example from one vendor, with UefiTarget was just hexadecimle numbers, another vendor was the UEFI boot path explicitly set as if it was just queried from nvram | 19:10 |
| TheJulia | so alias should provide us a nice mapping to keep... some semblance of sanity | 19:10 |
| TheJulia | I joke, we can be insane all we want! | 19:10 |
| rpioso | TheJulia: Not surprised :-) | 19:10 |
| rpioso | mraineri: Did the standard change or was there an interpretation issue? | 19:12 |
| TheJulia | rpioso: I think the answer is the standard evolved | 19:12 |
| etingof | can we rely on the omnipresence of AliasBootOrder? | 19:13 |
| * etingof hides | 19:13 | |
| *** Lucas_Gray has joined #openstack-ironic | 19:14 | |
| TheJulia | I suspect we need support read/grok/handle AliasBootOrder, and then maybe we could.... possibly | 19:14 |
| etingof | are you saying that we should have both mechanisms in sushy i.e. bootoverridetarget and bootorder? | 19:15 |
| TheJulia | yes | 19:15 |
| TheJulia | and ironic will need to grok them | 19:16 |
| etingof | why can't it all be hidden in sushy? | 19:16 |
| TheJulia | It won't fix the ?somewhere between redfish 1.5 and 1.8? (TBH, I haven't figured it out yet) that the thinksystem in the bug seems to be, but something at a higher level is going to need to have enough operational context to grok and navigate it | 19:17 |
| TheJulia | It could potentially be hidable in sushy completely | 19:17 |
| TheJulia | potentially | 19:17 |
| etingof | I do not see what kind of extra information sushy consumer (ironic) would possess that sushy does not... | 19:18 |
| TheJulia | Moving to support the alias as a first class citizen and then falling back to the override should work | 19:19 |
| TheJulia | That would enable it to handle pre-1.6 and post-1.6+ for the most part | 19:20 |
| TheJulia | I don't think the ThinkSystem in the storyboard link I pasted earlier will benefit from it though :( | 19:20 |
| etingof | I get the overall idea, I am just wondering why can't we have all these bits isolated inside sushy? | 19:21 |
| jungleboyj | :-( | 19:21 |
| etingof | can't ThinkSystem be tackled via boot-order? | 19:21 |
| *** Lucas_Gray has quit IRC | 19:22 | |
| TheJulia | jungleboyj: maybe you can convince a bmc engineer to come hang out and chat with us one day? https://media1.giphy.com/media/Q94xQWspTUkShljj8P/giphy.gif?cid=ecf05e47a8afb1143f15bdfc2ea511ef61456beb2ec25985&rid=giphy.gif | 19:22 |
| *** Lucas_Gray has joined #openstack-ironic | 19:22 | |
| TheJulia | etingof: you'd have to know internal names I think | 19:23 |
| TheJulia | think... looking at their examples | 19:23 |
| rpioso | mraineri: In the quote above from the schema, what does "UEFI boot source override" mean? | 19:23 |
| etingof | TheJulia, what about AliasBootOrder in that ThinkSystem? | 19:24 |
| jungleboyj | AHHHH The Cuteness! | 19:24 |
| TheJulia | etingof: https://github.com/lenovo/python-redfish-lenovo/blob/master/examples/lenovo_set_bios_boot_order.py#L68 | 19:24 |
| TheJulia | etingof: no mention at least example wise of alias support | 19:24 |
| jungleboyj | That would assume that I could figure out how to find the right developer. :-) | 19:24 |
| TheJulia | jungleboyj: we promise more cuteness | 19:24 |
| TheJulia | jungleboyj: no memo-list like thing there? :) | 19:25 |
| TheJulia | https://media1.giphy.com/media/TkBoNth0Ps3Vm/giphy.gif?cid=ecf05e47a8afb1143f15bdfc2ea511ef61456beb2ec25985&rid=giphy.gif <-- similar gif of a corgi doing the same thing | 19:25 |
| jungleboyj | NOOOOOOO! A Corgi....ahhh | 19:26 |
| TheJulia | is this a corgiphobia? | 19:26 |
| larsks | You folks are incorgi-able. | 19:26 |
| jungleboyj | Not that ... They are just too cute! | 19:26 |
| etingof | let's place these royal animals into github issue for python-redfish-lenovo | 19:26 |
| jungleboyj | I actually used to know who was leading up RedFish support for Lenovo but he left. Let me check here. | 19:27 |
| TheJulia | gremlin seeks redfish happiness https://usercontent.irccloud-cdn.com/file/wXGctIut/IMG_20200407_122945.jpg | 19:31 |
| jungleboyj | You are killing me. | 19:32 |
| TheJulia | heh :) | 19:32 |
| jungleboyj | Our dog is a Longhaired Dachshund. Similar body build, less log, more sausage like. | 19:33 |
| jungleboyj | We have a couple cute corgis in the neighborhood. | 19:33 |
| TheJulia | He is just starting to do the independent ear tracking of sounds, I was kind of hoping for a "Corgi speaking with the Corgi mothership!" sort photo, but those will come soon enough | 19:34 |
| *** derekh has quit IRC | 19:39 | |
| TheJulia | etingof: we also need to do the bios attribute BootMode too, not just the override | 19:41 |
| etingof | yeah, fun | 19:42 |
| etingof | but how do we identify the right bios attribute? | 19:42 |
| etingof | is it always BootMode across the fleet? | 19:43 |
| openstackgerrit | Merged openstack/ironic master: Add node lessee field https://review.opendev.org/706864 | 19:44 |
| TheJulia | yeah, looks like it is oem-definable | 19:45 |
| TheJulia | lenovo's example has a few different various ones | 19:45 |
| etingof | so in that light, may be we should abandon all hope for uniform boot management and go the vendor-specific sushy extensions way? | 19:47 |
| TheJulia | Well, maybe not because I think that might be too heavy wait and to heavy handed if it is uniformly in the attributes field | 19:47 |
| TheJulia | If someone else has stuff in Oem, there is nothing we can do | 19:48 |
| jungleboyj | TheJulia: Does this help? | 19:48 |
| jungleboyj | for reference, in our cluster solution we just use efibootmgr in linux local for persistent changes, and only ever use the 'Once' bootoverride for getting to that point | 19:48 |
| TheJulia | jungleboyj: maybe, that kind of ties in with the direction we're already heading. If UEFI boot mode, perhaps we just never try to set persistency with ironic ever | 19:49 |
| jungleboyj | So, Jarrod is a font of knowledge and has already been down this path. | 19:50 |
| TheJulia | Ohhhh idrac9 supports secure boot key setting via redfish | 19:52 |
| TheJulia | I'm not seeing an equivelent setting to the attribute value in Dell's docs.... still looking though | 19:53 |
| *** ociuhandu has joined #openstack-ironic | 19:54 | |
| TheJulia | yeah, it might be treated as an explicit bios setting without the override :\ Dell folks llikely know way better than my quick glance over their doc | 19:55 |
| * TheJulia looks up ilo | 19:56 | |
| rpioso | TheJulia: There's a BIOS attribute for setting the boot mode. | 19:56 |
| TheJulia | rpioso: an explicit attribute under the attributes field? | 19:56 |
| * rpioso looks | 19:57 | |
| openstackgerrit | Tzu-Mainn Chen proposed openstack/ironic master: Added node multitenancy doc https://review.opendev.org/718174 | 19:57 |
| rpioso | TheJulia: Yes, an explicit attribute. It's creatively named BootMode. | 19:59 |
| TheJulia | rpioso: \o/ for two vendors doing it the same way! | 19:59 |
| * TheJulia dances | 19:59 | |
| jungleboyj | Madness. | 20:00 |
| TheJulia | It _looks like_ HPE has a DefaultBootOrder array like the alias target setting | 20:00 |
| *** ociuhandu has quit IRC | 20:00 | |
| TheJulia | \o/ three vendors! | 20:01 |
| TheJulia | Bios attribute BootMode is in the ilo reference | 20:01 |
| TheJulia | \o/ | 20:01 |
| rpioso | TheJulia: How about removing the syncing of boot mode from preparing to boot the ramdisk and instance? Could it be a separate deploy step. | 20:01 |
| jungleboyj | TheJulia: I think you have talked to Jarrod before. Going to send an introductory e-mail for you, him and the BMC developer. | 20:02 |
| TheJulia | jungleboyj: thanks | 20:02 |
| jungleboyj | Anytime. I am going to have to get to meet the Corgi some time though. :-) | 20:02 |
| TheJulia | rpioso: I think it is more an issue than just syncing boot mode, and eventual deploy step makes sense, but I'm wondering if there is a backportable way we can make this better | 20:03 |
| TheJulia | And I think in part because we're seeing a whole blend of issues around the same area. Setting the boot mode, and setting what should be the preferred boot device because people seem to not want to be able to start with anything and end up in a consistent desirable end state, and that is kind of where these differences seem to start causing issues. | 20:05 |
| *** derekh has joined #openstack-ironic | 20:13 | |
| *** derekh has quit IRC | 20:14 | |
| *** Lucas_Gray has quit IRC | 20:15 | |
| *** Lucas_Gray has joined #openstack-ironic | 20:21 | |
| openstackgerrit | OpenStack Release Bot proposed openstack/tenks stable/1.0: Update .gitreview for stable/1.0 https://review.opendev.org/718220 | 20:37 |
| openstackgerrit | OpenStack Release Bot proposed openstack/tenks stable/1.0: Update TOX/UPPER_CONSTRAINTS_FILE for stable/1.0 https://review.opendev.org/718221 | 20:37 |
| openstackgerrit | OpenStack Release Bot proposed openstack/tenks master: Update master for stable/1.0 https://review.opendev.org/718222 | 20:37 |
| *** Lucas_Gray has quit IRC | 20:42 | |
| *** Lucas_Gray has joined #openstack-ironic | 20:44 | |
| *** dougsz has quit IRC | 20:45 | |
| *** igordc has quit IRC | 20:47 | |
| *** igordc has joined #openstack-ironic | 20:48 | |
| *** stevebaker has quit IRC | 21:01 | |
| *** stevebaker has joined #openstack-ironic | 21:09 | |
| rpioso | TheJulia: The first thing we do is set the boot mode. That's before the server is given to ironic. The boot mode affects 1) which BIOS attributes are available, 2) how a NIC port is configured to PXE boot, and 3) how the boot order is expressed. | 21:11 |
| rpioso | Kinda key. | 21:12 |
| openstackgerrit | Merged x/sushy-oem-idrac master: Bump hacking to 3.0.0 https://review.opendev.org/715900 | 21:27 |
| *** ociuhandu has joined #openstack-ironic | 21:34 | |
| TheJulia | rpioso: yeah, I wish I had a nice chart detailing variations across vendors though. | 21:34 |
| *** ociuhandu has quit IRC | 21:39 | |
| *** rcernin has joined #openstack-ironic | 22:11 | |
| *** Lucas_Gray has quit IRC | 22:36 | |
| openstackgerrit | Julia Kreger proposed openstack/sushy master: WIP: direct boot order setting support https://review.opendev.org/718275 | 22:45 |
| openstackgerrit | Julia Kreger proposed openstack/sushy master: WIP: Add an option for setting the http boot uri https://review.opendev.org/718276 | 22:45 |
| TheJulia | etingof: idea ^^ | 22:45 |
| TheJulia | err, ideas | 22:46 |
| *** tkajinam has joined #openstack-ironic | 22:56 | |
| *** yaawang__ has quit IRC | 23:16 | |
| *** yaawang__ has joined #openstack-ironic | 23:17 | |
| *** tzumainn has quit IRC | 23:37 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
