| *** tosky has quit IRC | 00:06 | |
| *** shadower has quit IRC | 00:26 | |
| *** shadower has joined #openstack-ironic | 00:29 | |
| *** anuradha1904 has quit IRC | 00:33 | |
| *** zzzeek has quit IRC | 00:42 | |
| *** uzumaki has quit IRC | 00:45 | |
| *** zzzeek has joined #openstack-ironic | 00:45 | |
| *** dsneddon has quit IRC | 00:51 | |
| *** trandles has quit IRC | 00:56 | |
| openstackgerrit | likui proposed openstack/ironic-python-agent master: Remove the unused coding style modules https://review.opendev.org/759792 | 01:02 |
|---|---|---|
| *** dsneddon has joined #openstack-ironic | 01:10 | |
| *** dsneddon has quit IRC | 01:16 | |
| *** ayoung has quit IRC | 01:41 | |
| *** xinliang has joined #openstack-ironic | 01:45 | |
| *** johnsom has quit IRC | 02:00 | |
| *** buhman has quit IRC | 02:00 | |
| *** rpittau|afk has quit IRC | 02:00 | |
| *** TheJulia has quit IRC | 02:00 | |
| *** rpittau|afk has joined #openstack-ironic | 02:00 | |
| *** TheJulia has joined #openstack-ironic | 02:00 | |
| *** buhman has joined #openstack-ironic | 02:00 | |
| *** johnsom has joined #openstack-ironic | 02:01 | |
| *** vdrok has quit IRC | 02:02 | |
| *** gyee has quit IRC | 02:02 | |
| *** vdrok has joined #openstack-ironic | 02:03 | |
| *** k_mouza has joined #openstack-ironic | 02:06 | |
| *** dsneddon has joined #openstack-ironic | 02:13 | |
| *** k_mouza has quit IRC | 02:14 | |
| *** dsneddon has quit IRC | 02:19 | |
| *** dsneddon has joined #openstack-ironic | 02:20 | |
| *** xinliang has quit IRC | 02:22 | |
| *** xinliang has joined #openstack-ironic | 02:22 | |
| openstackgerrit | Armstrong Liu proposed openstack/ironic-python-agent master: Add an option about collecting network information https://review.opendev.org/758950 | 02:38 |
| *** dsneddon has quit IRC | 02:43 | |
| *** dsneddon has joined #openstack-ironic | 02:45 | |
| *** dsneddon has quit IRC | 02:50 | |
| *** dsneddon has joined #openstack-ironic | 02:52 | |
| *** paras333 has joined #openstack-ironic | 02:57 | |
| *** dsneddon has quit IRC | 02:57 | |
| *** mkrai has joined #openstack-ironic | 02:59 | |
| *** tzumainn has quit IRC | 03:01 | |
| *** dsneddon has joined #openstack-ironic | 03:02 | |
| *** dsneddon has quit IRC | 03:07 | |
| *** mkrai has quit IRC | 03:07 | |
| *** mkrai_ has joined #openstack-ironic | 03:07 | |
| *** paras333_ has joined #openstack-ironic | 03:07 | |
| *** dsneddon has joined #openstack-ironic | 03:08 | |
| *** paras333_ has quit IRC | 03:12 | |
| *** dsneddon has quit IRC | 03:13 | |
| *** dsneddon has joined #openstack-ironic | 03:14 | |
| *** Qianbiao has joined #openstack-ironic | 03:35 | |
| *** dking has quit IRC | 03:35 | |
| *** mkrai_ has quit IRC | 03:40 | |
| *** larsks has quit IRC | 03:45 | |
| *** larsks has joined #openstack-ironic | 03:46 | |
| *** mkrai has joined #openstack-ironic | 03:57 | |
| *** k_mouza has joined #openstack-ironic | 04:15 | |
| *** xinliang has quit IRC | 04:17 | |
| *** k_mouza has quit IRC | 04:19 | |
| *** dsneddon has quit IRC | 05:12 | |
| *** dsneddon has joined #openstack-ironic | 05:13 | |
| *** dsneddon has quit IRC | 05:28 | |
| *** dsneddon has joined #openstack-ironic | 05:29 | |
| *** dsneddon has quit IRC | 05:34 | |
| *** dsneddon has joined #openstack-ironic | 05:37 | |
| *** mushrushu has quit IRC | 06:07 | |
| *** mushrushu has joined #openstack-ironic | 06:08 | |
| *** mkrai has quit IRC | 06:33 | |
| *** uzumaki has joined #openstack-ironic | 06:42 | |
| *** fgofurov has joined #openstack-ironic | 06:58 | |
| *** sshnaidm|afk is now known as sshnaidm|rover | 07:01 | |
| *** xinliang has joined #openstack-ironic | 07:05 | |
| *** rcernin has quit IRC | 07:31 | |
| *** bfournie has quit IRC | 07:37 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Explicitly open ports 68 and 69 on Ubuntu https://review.opendev.org/760292 | 07:43 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Add a CI job with Redfish and UEFI https://review.opendev.org/744434 | 07:46 |
| *** pmannidi has quit IRC | 07:47 | |
| openstackgerrit | Merged openstack/ironic master: json-rpc: surround IPv6 address with [] in conductor URL https://review.opendev.org/759877 | 07:49 |
| *** pmannidi has joined #openstack-ironic | 07:52 | |
| *** bfournie has joined #openstack-ironic | 07:53 | |
| *** mkrai has joined #openstack-ironic | 07:55 | |
| *** ociuhandu has joined #openstack-ironic | 08:00 | |
| arne_wiebalck | Good morning, ironic! | 08:06 |
| fgofurov | Good morning ironic! Hi arne o/ | 08:16 |
| *** yolanda has quit IRC | 08:24 | |
| *** yolanda has joined #openstack-ironic | 08:25 | |
| Qianbiao | Morning Ironic. | 08:40 |
| Qianbiao | Morning arne_wiebalck and fgofurov. | 08:40 |
| *** rpittau|afk is now known as rpittau | 08:40 | |
| rpittau | good morning ironic! o/ | 08:40 |
| Qianbiao | morning rpittau | 08:42 |
| Qianbiao | is your time switch to winter time? | 08:42 |
| rpittau | hey Qianbiao :) | 08:43 |
| rpittau | yep | 08:43 |
| rpittau | hopefully they will stop this non sense soon | 08:43 |
| Qianbiao | :), need to adjust your internal clock and habits. | 08:44 |
| rpittau | yeah | 08:44 |
| rpittau | suddenly it's night 1 hour earlier, it's terrible | 08:45 |
| Qianbiao | I can imagination that. :) | 08:46 |
| fgofurov | morning rpittau, Qianbiao. | 08:46 |
| rpittau | hey fgofurov :) | 08:47 |
| rpittau | major disruptions in mirrors from yesterday, going to recheck some changes | 08:47 |
| fgofurov | rpittau: is CI got better since https://review.opendev.org/759910 got merged? | 08:48 |
| patchbot | patch 759910 - ironic - Mark standalone job non-voting/remove from gate (MERGED) - 2 patch sets | 08:48 |
| rpittau | fgofurov: yes | 08:48 |
| fgofurov | hmm, awesome! how can we move forward with this patch then: https://review.opendev.org/#/c/759245/ | 08:49 |
| patchbot | patch 759245 - ironic - Add timeout to image operations in the direct deploy - 1 patch set | 08:49 |
| fgofurov | I was waiting for this to get merged soon if CI is in a good shape :) | 08:50 |
| rpittau | I issued a recheck there too | 08:50 |
| *** tosky has joined #openstack-ironic | 08:51 | |
| fgofurov | yes, thanks rpittau o/ | 08:51 |
| rpittau | np | 08:51 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/victoria: [DNM] Test victoria ci https://review.opendev.org/760300 | 08:56 |
| *** dtantsur|afk is now known as dtantsur | 08:58 | |
| dtantsur | morning ironic | 08:58 |
| rpittau | hey dtantsur | 08:58 |
| fgofurov | morning dtantsur | 08:59 |
| iurygregory | morning everyone o/ | 08:59 |
| rpittau | hey iurygregory :) | 08:59 |
| fgofurov | hey iurygregory | 08:59 |
| iurygregory | o/ | 09:00 |
| dtantsur | I couldn't make Ubuntu or CentOS work, so https://review.opendev.org/744434 only has Fedora now | 09:01 |
| patchbot | patch 744434 - bifrost - Add a CI job with Redfish and UEFI - 24 patch sets | 09:01 |
| dtantsur | I guess it's enough for me, any help welcome | 09:01 |
| dtantsur | (especially with Ubuntu, it just refuses to TFTP boot) | 09:01 |
| rpittau | hehhhh | 09:01 |
| dtantsur | logs from the last run: https://zuul.opendev.org/t/openstack/build/772426927e9a49e198f93689c7dce6bb/log/logs/testvm1_console.log#8 | 09:04 |
| * iurygregory looks | 09:04 | |
| *** anuradha1904 has joined #openstack-ironic | 09:06 | |
| rpittau | dtantsur: I don't see port 68 in netstat | 09:11 |
| *** lucasagomes has joined #openstack-ironic | 09:14 | |
| rpittau | could it be an selinux or permissions problem as well | 09:15 |
| Qianbiao | I hate selinux.. So, first thing after os boot is disable it :) | 09:16 |
| rpittau | Qianbiao: that's very bad actually :/ | 09:16 |
| Qianbiao | yeah i know | 09:16 |
| rpittau | but I totally understand, it gives some major headaches | 09:17 |
| *** rcernin has joined #openstack-ironic | 09:18 | |
| Qianbiao | yeah, just in test lab, make life easy. | 09:18 |
| dtantsur | there's no selinux on ubuntu | 09:18 |
| dtantsur | fedora actually works | 09:18 |
| rpittau | right | 09:18 |
| rpittau | I got the logs swapped :D | 09:18 |
| Qianbiao | I got an issue when using cloud-init with metadate service to init ramdisk os. | 09:19 |
| dtantsur | also tftp is port 69 | 09:19 |
| Qianbiao | cloud-init will only try dhcp on first port, if pxe port is not the first port, then cloud-init can not connect to the metadate service. | 09:19 |
| Qianbiao | any experience to fix this issue? | 09:20 |
| rpittau | from what I can see in the console logs, the PXE-E99 error is a generic network issue, what puzzles me is that the size of the NBP file is reported as 0 bytes, so it can't really access it | 09:22 |
| rpittau | that's all I have :/ | 09:22 |
| Qianbiao | * metadata | 09:23 |
| *** dsneddon has quit IRC | 09:25 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/victoria: Use centos as base element for dib images https://review.opendev.org/760309 | 09:28 |
| arne_wiebalck | Hey Qianbiao rpittau fgofurov dtantsur iurygregory o/ | 09:30 |
| rpittau | hey arne_wiebalck :) | 09:30 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent stable/victoria: Run dib ipa src jobs on ubuntu focal https://review.opendev.org/760310 | 09:30 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent stable/victoria: Run dib ipa src jobs on ubuntu focal https://review.opendev.org/760310 | 09:32 |
| Qianbiao | hey arne_wiebalck | 09:33 |
| Qianbiao | Or is there a method to let dhcp-all-interfaces run before cloud-init fetch metadata | 09:34 |
| iurygregory | arne_wiebalck, o/ | 09:36 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: Re-enable ironic-inspector-tempest-managed-non-standalone job https://review.opendev.org/759711 | 09:40 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector master: Re-enable ironic-inspector-tempest-managed-non-standalone job https://review.opendev.org/759711 | 09:40 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/victoria: Install gunicorn from pip https://review.opendev.org/760150 | 09:41 |
| *** dtantsur is now known as dtantsur|brb | 09:44 | |
| *** k_mouza has joined #openstack-ironic | 09:49 | |
| *** derekh has joined #openstack-ironic | 09:52 | |
| *** dsneddon has joined #openstack-ironic | 10:03 | |
| *** rcernin has quit IRC | 10:07 | |
| *** dsneddon has quit IRC | 10:18 | |
| *** dsneddon has joined #openstack-ironic | 10:19 | |
| *** dhellmann has quit IRC | 10:23 | |
| *** zzzeek has quit IRC | 10:28 | |
| *** zzzeek has joined #openstack-ironic | 10:29 | |
| *** mkrai has quit IRC | 10:40 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic stable/victoria: migrate testing to ubuntu focal https://review.opendev.org/760324 | 10:41 |
| *** xinliang has quit IRC | 10:48 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Convert last bionic jobs to focal https://review.opendev.org/760326 | 10:50 |
| *** mkrai has joined #openstack-ironic | 10:55 | |
| *** dhellmann has joined #openstack-ironic | 10:55 | |
| openstackgerrit | Riccardo Pittau proposed x/sushy-oem-idrac master: Update lower-constraints.txt https://review.opendev.org/727260 | 11:08 |
| openstackgerrit | Riccardo Pittau proposed x/sushy-oem-idrac master: Use safe version of hacking https://review.opendev.org/727260 | 11:14 |
| openstackgerrit | Riccardo Pittau proposed x/sushy-oem-idrac master: Use safe version of hacking https://review.opendev.org/727260 | 11:15 |
| *** rcernin has joined #openstack-ironic | 11:15 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-inspector stable/victoria: [DNM] Test victoria ci https://review.opendev.org/760330 | 11:17 |
| *** kafilat has joined #openstack-ironic | 11:50 | |
| *** rh-jelabarre has joined #openstack-ironic | 11:52 | |
| *** thiagop has joined #openstack-ironic | 11:54 | |
| *** dtantsur|brb is now known as dtantsur | 11:57 | |
| jroll | :w | 11:59 |
| jroll | oops :) | 11:59 |
| dtantsur | jroll: you know you cannot leave :) | 12:00 |
| jroll | I would never! | 12:00 |
| jroll | that would be :q not :w | 12:01 |
| dtantsur | oh, right | 12:01 |
| * dtantsur needs to switch back in work mode | 12:01 | |
| openstackgerrit | Merged openstack/ironic master: Add timeout to image operations in the direct deploy https://review.opendev.org/759245 | 12:20 |
| *** mkrai has quit IRC | 12:30 | |
| *** ianychoi__ is now known as ianychoi | 12:34 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Handle agent still doing the prior command https://review.opendev.org/759906 | 12:36 |
| TheJulia | ^^^ just a commit message update. Should pass just fine and should result in our CI failures reducing and will allow us to get the standalone job failure fix w/r/t raid configuration in as well since the failure rate is high there. | 12:38 |
| *** anuradha1904 has quit IRC | 12:39 | |
| iurygregory | good morning TheJulia =) | 12:44 |
| openstackgerrit | Julia Kreger proposed openstack/metalsmith master: Add a Centos8 UEFI based job https://review.opendev.org/748315 | 12:45 |
| dtantsur | morning TheJulia | 12:52 |
| * dtantsur has sent the CfSPUC! | 12:52 | |
| TheJulia | ooohh ahhhhhh | 12:52 |
| TheJulia | ++ | 12:52 |
| dtantsur | for those who want to skip my moderately humorous text, the doodle is https://doodle.com/poll/tw72utagidz2nac3 | 12:53 |
| *** anuradha1904 has joined #openstack-ironic | 12:54 | |
| TheJulia | where did the humorous text go to? | 12:55 |
| dtantsur | to the ML around 1 minute ago | 12:56 |
| dtantsur | (so may not appear in your inbox just yet) | 12:56 |
| * dtantsur distributes downstream as well | 12:56 | |
| * TheJulia is laughing | 12:56 | |
| dtantsur | I'm glad it has positive effect already :) | 12:57 |
| TheJulia | dmitry wins at generating laughs first thing | 12:57 |
| dtantsur | \o/ | 12:57 |
| *** rloo has joined #openstack-ironic | 12:57 | |
| iurygregory | I loved the Boo in the begin of the email XD | 12:58 |
| dtantsur | :) | 12:58 |
| TheJulia | dtantsur: w/r/t detecting supermicro bmcs... any thoughts on if it is checked/done in the get power state check so funcitonally we end up taking a little longer but then we're guaranteed to get it for everything at some point if not already populated | 13:03 |
| TheJulia | I was actually kind of hoping to avoid doing that but it is kind of an inevitable thing I guess | 13:06 |
| TheJulia | or maybe we need to consider a learn() method on interfaces that defaults to a noop except in that early state transition | 13:08 |
| *** paras333 has joined #openstack-ironic | 13:09 | |
| dtantsur | TheJulia: I'm fine with power state check. we need to only do it once, right? | 13:11 |
| TheJulia | yeah | 13:11 |
| TheJulia | And on a plus side, we'll start getting a little more insight when people paste node show to us | 13:11 |
| dtantsur | mmmmm! | 13:12 |
| dtantsur | that's a wonderful idea. | 13:12 |
| dtantsur | honestly, we need a get_vendor driver call. | 13:12 |
| TheJulia | ++ | 13:12 |
| dtantsur | we already have use cases for detecting a vendor | 13:12 |
| dtantsur | normalization of vendors names can pose an interesting issue | 13:13 |
| TheJulia | a bit of an issue, but we'll never know until we start storing that data | 13:13 |
| dtantsur | yeah | 13:13 |
| dtantsur | I'm worried about stuff like "Dell" vs "Dell EMC" or so | 13:13 |
| TheJulia | I'll hack it together, not expose it for now | 13:13 |
| dtantsur | but in the end, both are usable if you do 'dell' in vendor.lower() :) | 13:13 |
| TheJulia | yup | 13:14 |
| dtantsur | but yeah, we could totally use a vendor field in metal3 | 13:14 |
| dtantsur | maybe make it a top-level ironic field instead of shoving into properties? or is it an overkill? | 13:14 |
| TheJulia | I think if not set in a driver, it would make sense to go ahead to colelct and store it which means a redfish get_vendor would be fairly easy as well | 13:15 |
| TheJulia | I think it likely is | 13:15 |
| TheJulia | it is a detail about a bmc and we do kind of want/need humans to be able to override it if they learn of things we do not know | 13:15 |
| TheJulia | "if you need to change this setting, and your oem is actually what it is, please let us know" blah blah blah | 13:15 |
| dtantsur | so, properties/vendor then? | 13:16 |
| TheJulia | I was thinking driver_info/<driver>_vendor | 13:16 |
| TheJulia | since it also governs behavior for ipmi | 13:16 |
| TheJulia | since we have to send different things for supermicro | 13:17 |
| dtantsur | mmmmm | 13:17 |
| dtantsur | my thinking of it is as of a property, similar to local_gb and such | 13:17 |
| dtantsur | while driver_info stuff is usually input for a driver | 13:17 |
| TheJulia | hmm | 13:17 |
| TheJulia | this is true | 13:17 |
| dtantsur | okay, it's both in this case. but I think from UX perspective it's more of a property | 13:17 |
| TheJulia | I thought we had consensus on the properties data governing out driver behavior? | 13:17 |
| TheJulia | (of course, we have the one exception to that rule) | 13:18 |
| * TheJulia doesn't remember that exception, just knows it is there somewhere | 13:18 | |
| dtantsur | properties/capabilities? ;) | 13:18 |
| dtantsur | I'm fine with your proposal as well, just trying to think how to make it more widely useful | 13:19 |
| TheJulia | so with lenovo, I think the dance has to change | 13:19 |
| TheJulia | We don't ship driver_internal_info to the agent do we? | 13:19 |
| *** paras333 has quit IRC | 13:19 | |
| TheJulia | so maybe properties is far better | 13:19 |
| *** paras333_ has joined #openstack-ironic | 13:20 | |
| TheJulia | I'm thinking because setting uefi boot to disk up on lenovo is a different higher level workflow, we basically need to short circuit our attempt to externally set it and just let efibootmgr do $magic | 13:20 |
| TheJulia | which reminds me | 13:20 |
| TheJulia | I was supposed to talk to someone about lenovo hardware this week | 13:20 |
| TheJulia | jungleboyj: yo, we should like talk as well at some point soon. Maybe not this week though ;) | 13:21 |
| *** paras333_ has quit IRC | 13:21 | |
| *** paras333 has joined #openstack-ironic | 13:22 | |
| jungleboyj | TheJulia: Ok. Let me know when you want to chat. Next week or something. | 13:22 |
| *** paras333 has joined #openstack-ironic | 13:22 | |
| *** akiselev has quit IRC | 13:22 | |
| TheJulia | I'm trying *not* to work next week :) | 13:27 |
| jungleboyj | Lucky! So, whenever works for you. | 13:28 |
| TheJulia | I'm thinking clean my solar panels and paint my roof with the super reflective paint.... | 13:28 |
| jungleboyj | At this point I am hoping to take a good portion of December off. I have had all of like 7 days off this year. | 13:29 |
| jungleboyj | Do you need to clean Solar Panels? | 13:29 |
| TheJulia | ++ | 13:29 |
| dtantsur | TheJulia: honestly, with agent tokens and TLS in place, we should just send the whole node (minus passwords) to the agent | 13:29 |
| TheJulia | jungleboyj: we get lots of dust here | 13:29 |
| TheJulia | dtantsur: perhaps yeah | 13:29 |
| jungleboyj | Oh, gotcha. I really don't want to think about trying to clean all of my panels. | 13:29 |
| *** rcernin has quit IRC | 13:43 | |
| *** akiselev has joined #openstack-ironic | 13:44 | |
| TheJulia | jungleboyj: did you do roof or ground frame mount? | 13:47 |
| *** tzumainn has joined #openstack-ironic | 13:52 | |
| *** kafilat has quit IRC | 13:58 | |
| *** sdanni has joined #openstack-ironic | 14:10 | |
| *** uzumaki has quit IRC | 14:20 | |
| *** fgofurov has quit IRC | 14:34 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Remove nodeset option https://review.opendev.org/760370 | 14:40 |
| jungleboyj | TheJulia: Roof. | 14:42 |
| jungleboyj | https://usercontent.irccloud-cdn.com/file/nm8DK5Gw/20191122_150409.jpg | 14:45 |
| TheJulia | ugh | 14:46 |
| TheJulia | wow | 14:46 |
| rpittau | that's interesting | 14:46 |
| TheJulia | rpittau: it is kind of common | 14:46 |
| jungleboyj | https://usercontent.irccloud-cdn.com/file/aIPiP633/20191122_150346.jpg | 14:46 |
| rpittau | here's becoming more common, but not at that level | 14:47 |
| jungleboyj | Yeah, doing ground mount would have taken up my whole back yard. | 14:47 |
| jungleboyj | And my roof is scary. | 14:47 |
| rpittau | I'd just be worried about hte weight tbh | 14:47 |
| jungleboyj | 44 Panels. 13.79 KW system | 14:47 |
| rpittau | not bad :) | 14:48 |
| jungleboyj | Yeah, 16.03 MWh produced this year. | 14:48 |
| TheJulia | I've got 6kW of panels, can't add more, but they are changing the laws here and I'll likely add more panels then | 14:49 |
| jungleboyj | I have them spread across two 6 KW inverters. | 14:50 |
| jungleboyj | https://usercontent.irccloud-cdn.com/file/snrUHfIo/20191122_135223.jpg | 14:50 |
| jungleboyj | They are tied into my Sense so I can monitor consumption and production. :-) | 14:50 |
| TheJulia | so, anyone remember where we got people to sign up for short presentations? | 14:53 |
| *** lmcgann has joined #openstack-ironic | 15:00 | |
| *** tkajinam has quit IRC | 15:05 | |
| TheJulia | arne_wiebalck: when were you thinking the next baremetal sig meeting would be? | 15:05 |
| arne_wiebalck | TheJulia: Tue Nov 10 at 2pm UTC if that works for you? | 15:06 |
| TheJulia | yes | 15:06 |
| arne_wiebalck | TheJulia: you can prepare the ML2 topic we discussed? | 15:07 |
| TheJulia | yes | 15:07 |
| TheJulia | one link inbound! | 15:07 |
| TheJulia | https://etherpad.opendev.org/p/os-recordings | 15:12 |
| TheJulia | thoughts? ideas? | 15:12 |
| TheJulia | The idea is one central clearinghouse for cross-coordination | 15:12 |
| *** Qianbiao has quit IRC | 15:32 | |
| dtantsur | sounds good | 15:35 |
| rpittau | yeah looks good | 15:40 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: DNM try to inject the correct grub location https://review.opendev.org/758809 | 15:48 |
| * TheJulia looks at priorities document and the vide on the other monitor and signs | 16:07 | |
| * dtantsur looks how ironic is trying to boot an isolinux CD with UEFI and sighs as well | 16:09 | |
| TheJulia | wut? | 16:10 |
| dtantsur | yeah.. I guess I've found another place where default_boot_mode is not respected. | 16:10 |
| TheJulia | yup | 16:10 |
| dtantsur | we have get_boot_mode and get_boot_mode_for_deploy, what are the chances of using them incorrectly? | 16:13 |
| dtantsur | wtf do we even have get_boot_mode_for_deploy if all it does is ignoring default_boot_mode? | 16:13 |
| TheJulia | I'd need to see the code in front of me and brain is on human words, not computer words right now | 16:14 |
| dtantsur | mmm, yeah, the ilo drivers have a use case of ignoring the default in favour of their own default | 16:18 |
| TheJulia | wheeeeeee | 16:18 |
| rpittau | heh why not :/ | 16:20 |
| *** trandles has joined #openstack-ironic | 16:21 | |
| dtantsur | okay, a one line fix and I can finally debug vmedia+uefi locally \o/ | 16:26 |
| TheJulia | post the fix! | 16:26 |
| rpittau | last 2 test-reqs changes if someone has a minute to donate to the cause https://review.opendev.org/758764 https://review.opendev.org/758436 :) | 16:26 |
| patchbot | patch 758764 - python-ironic-inspector-client - Set safe version of hacking - 7 patch sets | 16:26 |
| patchbot | patch 758436 - sushy - Set safe version of hacking - 2 patch sets | 16:26 |
| dtantsur | working on unit tests now | 16:26 |
| lmcgann | Hello, I am looking into ways of passing in certain keylime config options to the ipa, but being that we might want to pass these parameters before the first heartbeat, I dont see how we could include these parameters in the image, and I dont see any existing apis for the ipa, how does Ironic typically communicate with the ipa? | 16:28 |
| dtantsur | lmcgann: kernel params is the most universal way | 16:28 |
| *** fgofurov has joined #openstack-ironic | 16:28 | |
| JayF | lmcgann: The typical way of passing short configuration, that you need before heartbeat, is kernel params. Configuration can also be baked into the image; but this precludes changing those configs for operators who use prebuilt images. | 16:28 |
| sdanni | Hi! I work with lmcgann on keylime. we are trying to setup keylime-agent on the node and let it run at boot time. we wonder what's a proper way to install it on the ramdisk, like create a tarball and put it onto the image? and is it feasible to start the keylime-agent service by a bash script somewhere inside ipa? | 16:34 |
| TheJulia | so I think lmcgann your thinking lookup, because config is passed through and back that way into the agent once it has started | 16:34 |
| TheJulia | sdanni: I'd create a diskimage-builder element | 16:34 |
| TheJulia | starting, I think depends on mechanics and where in the sequence of events. | 16:35 |
| rpittau | sdanni: if you want to do something clean, maybe having a systemd unit file would also be nice | 16:36 |
| TheJulia | it may not even need to be a systemd unit file, it sounds like the flow is required | 16:37 |
| rpittau | it depends on how the keylime-agent runs, is it a daemon? a one-shot? | 16:37 |
| lmcgann | TheJulia perhaps I am misunderstanding the purpose of the 'track_machine()' method of the security interface you suggested. If we can just pass the config through kernel parameters, what purpose does this method serve? And if we want to allow the agent to do a lookup in the same vein that it does a lookup of its uuid, wouldnt this exist outside of the security_interface? | 16:37 |
| TheJulia | sorry, context switching between multiple discussions, a few minutes | 16:39 |
| * dtantsur very calmly and politely reminds himself that ISO filesystem is, in fact, case-sensitive | 16:40 | |
| *** lucasagomes has quit IRC | 16:44 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Make redfish-virtual-media respect default_boot_mode https://review.opendev.org/760403 | 16:44 |
| sdanni | rpittau: keylime-agent is a daemon I think. So it needs a systemd unit file? TheJulia, what kind of flow is required? | 16:44 |
| TheJulia | a daemon doesn't necessarily require a systemd unit file in an ephemeral ramdisk | 16:45 |
| dtantsur | let's go with a unit file to avoid chaos | 16:46 |
| TheJulia | I'm not understanding the overall process start interaction we're talking about fwiw | 16:46 |
| dtantsur | sdanni: check https://opendev.org/openstack/ironic-python-agent-builder for how IPA itself is installed in the two types of our ramdisks (you're mostly interested in the dib one, which is used in production) | 16:46 |
| TheJulia | ipa itself does make a good example :) | 16:46 |
| dtantsur | if keylime-agent has to be just running, we can make it a systemd service and make IPA depend on it | 16:47 |
| dtantsur | I suggest creating a new element for installing it, so that people can opt-in | 16:47 |
| rpittau | so we're talking about running keylime-agent *before* ipa, I guess that | 16:47 |
| rpittau | 's a good solution | 16:47 |
| TheJulia | dtantsur: making it depend on it is worrisome | 16:49 |
| TheJulia | opt in ++ | 16:49 |
| dtantsur | depends on what they need? | 16:49 |
| dtantsur | systemd has reversed dependencies, so keylime-agent can claim it must come before IPA | 16:50 |
| TheJulia | make ipa depending on keylime a hard requirement | 16:50 |
| dtantsur | but IPA won't depend on it | 16:50 |
| TheJulia | but yeah, all depends on what people need | 16:50 |
| sdanni | so if I understand it correctly, I should create a systemd unit file like ironic-python-agent.service inside dib install.d | 16:50 |
| sdanni | and make ipa and keylime-agent independent with each other? | 16:51 |
| dtantsur | a new element (similar to ironic-python-agent-ramdisk directory) | 16:52 |
| dtantsur | there a new service that declares that it comes before IPA (I don't remember the exact syntax, maybe just Before:) | 16:52 |
| rpittau | sdanni: sometihing similar to what you can find in https://opendev.org/openstack/ironic-python-agent-builder/src/branch/master/dib/ironic-python-agent-ramdisk | 16:52 |
| sdanni | thanks! | 16:56 |
| rpittau | np | 16:56 |
| rpittau | need to run, literally, talk tomorrow! o/ | 16:57 |
| *** rpittau is now known as rpittau|afk | 16:57 | |
| iurygregory | dtantsur, you still around? | 17:05 |
| dtantsur | iurygregory: seemingly :) | 17:05 |
| iurygregory | quick discussion re https://review.opendev.org/#/c/757907/ that was converted in https://review.opendev.org/#/c/759834/ | 17:06 |
| patchbot | patch 757907 - sushy - Adds sushy library overview (ABANDONED) - 7 patch sets | 17:06 |
| patchbot | patch 759834 - sushy - Adds sushy library features - 2 patch sets | 17:06 |
| iurygregory | looking at the current docs for sushy https://docs.openstack.org/sushy/latest/ maybe we just want to change "About Sushy" to "Overview" ? | 17:07 |
| dtantsur | ++ | 17:07 |
| iurygregory | and the features we keep in "Sushy Library Reference" https://docs.openstack.org/sushy/latest/reference/index.html | 17:07 |
| *** ociuhandu_ has joined #openstack-ironic | 17:07 | |
| iurygregory | and maybe "Using Sushy" should become a new section by itself ? | 17:08 |
| dtantsur | dunno, it feels like something to advertise upfront | 17:08 |
| iurygregory | humm | 17:08 |
| iurygregory | ok so we would have the Features section in the main page (makes sense) | 17:08 |
| iurygregory | and just remove from the Sushy Library Reference | 17:08 |
| TheJulia | iurygregory: section by itself ++ | 17:10 |
| *** ociuhandu has quit IRC | 17:11 | |
| *** ociuhandu_ has quit IRC | 17:11 | |
| *** fgofurov has quit IRC | 17:11 | |
| TheJulia | lmcgann: I think it all depends on the mechanics of how sdanni's work and plays out and how the agent needs to be started or if the configuration needs to be supplied explicitly or can be passed inband. | 17:13 |
| lmcgann | well the current idea is the controller has to go to the verifier to learn if everything checks out. And the controller knows not to go to the verifier until it gets the first heartbeat | 17:15 |
| lmcgann | So I think kernel parameters is the simplest way to do things | 17:15 |
| lmcgann | for that initial setup at least | 17:15 |
| lmcgann | Basically, I am not sure we need the 'track_machine()' method at this point. | 17:17 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Redfish virtual media boot in UEFI https://review.opendev.org/760423 | 17:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Redfish virtual media boot in UEFI https://review.opendev.org/760423 | 17:34 |
| dtantsur | okay, this was fun | 17:34 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Redfish virtual media boot in UEFI https://review.opendev.org/760423 | 17:39 |
| *** k_mouza has quit IRC | 17:41 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Redfish virtual media boot in UEFI https://review.opendev.org/760423 | 17:47 |
| *** fgofurov has joined #openstack-ironic | 17:51 | |
| openstackgerrit | Merged openstack/ironic-inspector master: Re-enable ironic-inspector-tempest-managed-non-standalone job https://review.opendev.org/759711 | 17:52 |
| *** derekh has quit IRC | 18:07 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Document how to build an ESP image for redfish-virtual-media https://review.opendev.org/760434 | 18:09 |
| * dtantsur celebrates | 18:09 | |
| TheJulia | lmcgann: it very well could be. There is likely some internal mechanics to work out but that would likely be in review. I will try to look at the spec later today. | 18:11 |
| TheJulia | dtantsur: omg thank you | 18:11 |
| TheJulia | re the doc | 18:11 |
| dtantsur | np :) | 18:11 |
| dtantsur | the great discovery: we don't really need to rebuild the grub image | 18:11 |
| TheJulia | I was talking to chrisj a few days ago and he was asking questions about that and I was like "uhhhh" | 18:11 |
| dtantsur | we only need to provide the correct grub_config_path | 18:11 |
| TheJulia | that is awwesome! | 18:11 |
| TheJulia | nice | 18:11 |
| dtantsur | (and it took me weeks to realize it's case-sensitive, which probably says a lot about me) | 18:11 |
| TheJulia | that is so awesome I can't even put into words after the past two weeks | 18:11 |
| TheJulia | no it doesn't... | 18:12 |
| dtantsur | it does, it's on an ISO file system | 18:12 |
| TheJulia | thats obscure iso9660 knowledge | 18:12 |
| dtantsur | NOT on the ESP VFAT | 18:12 |
| TheJulia | yup | 18:12 |
| dtantsur | ah, I didn't get what you're objecting to probably | 18:12 |
| TheJulia | heh | 18:12 |
| dtantsur | (which also says something about my condition) | 18:12 |
| TheJulia | gah, should you be like... having a good cup of tea and maybe opening a book by now? | 18:12 |
| dtantsur | I have an excuse: I have a device attached to me that measures my blood pressure every 15 minutes | 18:12 |
| dtantsur | it's annoying AS HELL | 18:13 |
| TheJulia | oh jeeze | 18:13 |
| TheJulia | yeah! | 18:13 |
| dtantsur | what is worse, I'm going to have it during the night.... | 18:13 |
| dtantsur | yes, a good cup of tea and dinner it is :) | 18:13 |
| TheJulia | ugh, yeah. That is reminiscent of the last couple times I was in a hospital | 18:13 |
| TheJulia | Last ER visit I took the cuff off because it was just a stupidly bad UTI or something there and it was really clear but they needed to check BP anyway. | 18:14 |
| dtantsur | heh | 18:14 |
| dtantsur | sorry, I wanted to finish on a more positive note :) | 18:14 |
| dtantsur | a positive note: I've just observed a virtual media deployment in UEFI in bifrost :) | 18:15 |
| TheJulia | heh | 18:16 |
| TheJulia | That is super positive, goodnight! | 18:16 |
| *** dtantsur is now known as dtantsur|afk | 18:18 | |
| dtantsur|afk | g'night! | 18:18 |
| openstackgerrit | Kafilat Adeleke proposed openstack/sushy master: Adds sushy library overview https://review.opendev.org/759834 | 18:22 |
| *** happy-contrib-43 has joined #openstack-ironic | 18:27 | |
| *** rh-jelabarre has quit IRC | 18:28 | |
| happy-contrib-43 | Hello, I was wondering if you guys ever tried to get the media insert working with openstack for the sushy-tools emulator. | 18:29 |
| *** tosky has quit IRC | 18:29 | |
| *** rh-jelabarre has joined #openstack-ironic | 18:34 | |
| *** uzumaki has joined #openstack-ironic | 18:38 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Remove the support to create port in API https://review.opendev.org/759546 | 18:40 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Remove the support to create port in API https://review.opendev.org/759546 | 18:40 |
| *** thiagop has quit IRC | 18:42 | |
| openstackgerrit | Bill Dodd proposed openstack/ironic master: WIP: Add Redfish RAID management to Ironic https://review.opendev.org/730000 | 18:44 |
| *** dsneddon has quit IRC | 19:01 | |
| *** dsneddon has joined #openstack-ironic | 19:03 | |
| *** vesper11 has quit IRC | 19:35 | |
| *** k_mouza has joined #openstack-ironic | 19:42 | |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/sushy master: Avoid running functional jobs for doc changes https://review.opendev.org/760445 | 19:44 |
| *** k_mouza has quit IRC | 19:46 | |
| stevebaker | good morning | 19:50 |
| *** rh-jelabarre has quit IRC | 19:57 | |
| *** gyee has joined #openstack-ironic | 19:57 | |
| *** uzumaki has quit IRC | 20:01 | |
| *** uzumaki has joined #openstack-ironic | 20:04 | |
| *** happy-contrib-43 has quit IRC | 20:11 | |
| stevebaker | brb upgrading fedora | 20:12 |
| *** stevebaker has quit IRC | 20:12 | |
| openstackgerrit | Leo McGann proposed openstack/ironic-specs master: Add security interface spec https://review.opendev.org/576718 | 20:30 |
| *** zzzeek has quit IRC | 20:32 | |
| *** zzzeek has joined #openstack-ironic | 20:33 | |
| akiselev | TheJulia: I was about to make a change and open a review on https://storyboard.openstack.org/#!/story/2008281 but then realized somebody else took it already. I don't mind since I have a bunch of other things to do, even though looks like a good first issue to learn the process. Is it safe to assume the person who took it will actually work on it in the near future? | 20:36 |
| akiselev | I can probably start from https://storyboard.openstack.org/#!/story/2008280 which has no takers, but for that one I will need some guidance. | 20:39 |
| *** stevebaker has joined #openstack-ironic | 20:40 | |
| *** jamesdenton has quit IRC | 20:47 | |
| *** jamesdenton has joined #openstack-ironic | 20:54 | |
| *** jamesdenton has quit IRC | 20:54 | |
| *** lmcgann has quit IRC | 21:02 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-specs master: Add Wallaby priorities https://review.opendev.org/760460 | 21:09 |
| TheJulia | akiselev: unfortunately, we don't have a good way to know if they will or will not before Monday. 2008280, as you pointed out does look like it should be fairly easy. Your absolutely welcome to give it a try. | 21:11 |
| *** vesper11 has joined #openstack-ironic | 21:14 | |
| *** thiagop has joined #openstack-ironic | 21:16 | |
| *** thiagop has quit IRC | 21:27 | |
| *** thiagop has joined #openstack-ironic | 21:31 | |
| *** vesper has joined #openstack-ironic | 21:35 | |
| *** vesper11 has quit IRC | 21:36 | |
| *** tosky has joined #openstack-ironic | 21:58 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Handle agent still doing the prior command https://review.opendev.org/759906 | 21:58 |
| *** thiagop has quit IRC | 22:08 | |
| *** anuradha1904 has quit IRC | 22:29 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: WIP: IPMI: Handle vendor set boot device differences https://review.opendev.org/757198 | 22:41 |
| TheJulia | dtantsur|afk: ^^ lmk if that is along the lines of what you were thinking with your comment and I'll try to finish redoing the tests and do the higher level equivelent that wouldn't be backportable tomorrow | 22:41 |
| * TheJulia goes and makes dinner | 22:42 | |
| *** rloo has quit IRC | 22:51 | |
| *** rcernin has joined #openstack-ironic | 22:58 | |
| *** rcernin has quit IRC | 23:00 | |
| *** rcernin has joined #openstack-ironic | 23:00 | |
| *** ociuhandu has joined #openstack-ironic | 23:09 | |
| *** ociuhandu has quit IRC | 23:13 | |
| *** akiselev_ has joined #openstack-ironic | 23:30 | |
| *** akiselev has quit IRC | 23:32 | |
| *** uzumaki has quit IRC | 23:44 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
