| openstackgerrit | Julia Kreger proposed openstack/ironic stable/train: Sync boot mode when changing the boot device via Redfish https://review.opendev.org/762439 | 00:06 |
|---|---|---|
| TheJulia | stevebaker: that redfish backport ^^^ I think it will be good, but I also can't unit test lcoally for train with what gets pulled down on my machine :\ | 00:07 |
| TheJulia | so I guess we shall see | 00:08 |
| ayoung | I think I forgot this step: | 00:12 |
| ayoung | for UUID in `openstack baremetal node list -f json | jq -r '.[] | .UUID' ` ; do openstack baremetal node set $UUID --driver-info ipmi_password=`cat ~/ipmi.password` --driver-info ipmi_username=root ; done | 00:12 |
| ayoung | Ah, now I hear them purring in the other room. | 00:13 |
| *** rloo has quit IRC | 00:21 | |
| TheJulia | \o/ | 00:23 |
| TheJulia | hopefully you have no cats sitting on actual servers | 00:23 |
| TheJulia | Because that can create a break in spacetime | 00:23 |
| stevebaker | TheJulia: ack, I'll have a look | 00:29 |
| TheJulia | oh wow i figured it out | 00:46 |
| *** jhesketh has joined #openstack-ironic | 00:49 | |
| *** tzumainn has quit IRC | 00:49 | |
| *** gyee has quit IRC | 01:12 | |
| stevebaker | TheJulia: oops I rewrote the module https://review.opendev.org/762446 | 01:14 |
| patchbot | patch 762446 - tripleo-ansible - Reimplement os_tripleo_baremetal_node_introspection - 1 patch set | 01:14 |
| openstackgerrit | likui proposed openstack/python-ironicclient master: Use TOX_CONSTRAINTS_FILE https://review.opendev.org/762451 | 01:27 |
| *** thiagop has quit IRC | 01:28 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: WIP: Fix default disk label with partition images https://review.opendev.org/762453 | 01:32 |
| TheJulia | stevebaker: doh ;) | 01:32 |
| TheJulia | dtantsur|afk: ^^^ I _think_ that is why my the centos8 metalsmith partition image job is not working with my chagnes.... it is getting a bios mbr partition table | 01:33 |
| TheJulia | anyway, I haven't figured out tonight how to block the wait_for_disks call (maybe I just need another argument *shrugs*) | 01:33 |
| TheJulia | anyway, need to go make dinner | 01:33 |
| *** Qianbiao has joined #openstack-ironic | 01:40 | |
| *** Goneri has quit IRC | 02:15 | |
| *** rcernin has quit IRC | 02:18 | |
| *** paras333 has joined #openstack-ironic | 02:49 | |
| *** rcernin has joined #openstack-ironic | 02:50 | |
| TheJulia | dtantsur|afk: yeah, that fix seems to do it https://35b627a3a7ec13a9f167-db6d669a6944592ec6491b3d692f1e09.ssl.cf5.rackcdn.com/762453/1/check/metalsmith-integration-ipa-src-uefi/1f02743/controller/ironic-bm-logs/node-0_no_ansi_2020-11-12-02%3A56%3A55.log <-- uses gpt for the partition image \o/ | 03:11 |
| ayoung | So... with ipmitool, should I be able to query the expected boot order? Like, if I do: | 03:15 |
| ayoung | ipmitool -I lanplus -H 192.168.123.10 -L ADMINISTRATOR -R 1 -N 1 -U root -f ~/ipmi.password chassis bootdev pxe | 03:15 |
| ayoung | that should set something that says "next time try pxe" but it doesn't. And I also don't see a way to query it | 03:15 |
| TheJulia | ayoung: chassis bootparam get 5 I think | 03:15 |
| TheJulia | BUT! it can be a complete lie! | 03:16 |
| TheJulia | at least to get | 03:16 |
| TheJulia | and that should be an override, you can't query the absolute boot order over ipmi | 03:16 |
| ayoung | $ ipmitool -I lanplus -H 192.168.123.10 -L ADMINISTRATOR -R 1 -N 1 -U root -f ~/ipmi.password chassis bootparam get 5 | 03:16 |
| ayoung | Boot parameter version: 1 | 03:16 |
| ayoung | Boot parameter 5 is valid/unlocked | 03:16 |
| ayoung | Boot parameter data: 0000000000 | 03:16 |
| ayoung | Boot Flags : | 03:16 |
| ayoung | - Boot Flag Invalid | 03:16 |
| ayoung | - Options apply to only next boot | 03:16 |
| ayoung | - BIOS PC Compatible (legacy) boot | 03:16 |
| ayoung | - Boot Device Selector : No override | 03:16 |
| ayoung | - Console Redirection control : System Default | 03:16 |
| ayoung | - BIOS verbosity : Console redirection occurs per BIOS configuration setting (default) | 03:16 |
| TheJulia | rofl, yeah.... all zeros is a bad sign | 03:16 |
| ayoung | - BIOS Mux Control Override : BIOS uses recommended setting of the mux at the end of POST | 03:16 |
| ayoung | [ayoung@nuzleaf nodes]$ | 03:16 |
| ayoung | sorry for the crap flood, | 03:16 |
| TheJulia | it means the bmc won't tell you basically or it can't translate it | 03:17 |
| ayoung | does that mean that that these ipmi devices are basically worthless for ironic management? | 03:17 |
| TheJulia | so is the machine on uefi or legacy bios mode | 03:17 |
| ayoung | Ohh, I don't klnow. Probably legacy | 03:17 |
| TheJulia | so, check if each interface is actually enabled in the bios settings | 03:18 |
| TheJulia | it may be the bmc can't enable it because the cards are set to an incompatible mode like... iscsi initiator | 03:18 |
| *** paras333 has quit IRC | 03:19 | |
| ayoung | rebooting to get tothe bios...lets see if I grab it in time... | 03:19 |
| TheJulia | the webui on the bmc, depending on version/firmware and all might let you get into those settings | 03:21 |
| *** mkrai has joined #openstack-ironic | 03:27 | |
| *** k_mouza has joined #openstack-ironic | 04:16 | |
| *** k_mouza has quit IRC | 04:21 | |
| openstackgerrit | likui proposed openstack/ironic-python-agent master: Remove the unused coding style modules https://review.opendev.org/759792 | 05:45 |
| *** hjensas has quit IRC | 05:57 | |
| *** hjensas has joined #openstack-ironic | 05:59 | |
| *** rcernin_ has joined #openstack-ironic | 05:59 | |
| *** rcernin has quit IRC | 06:01 | |
| *** fmuyassarov has joined #openstack-ironic | 06:47 | |
| *** rcernin_ has quit IRC | 07:26 | |
| *** mkrai has quit IRC | 07:42 | |
| iurygregory | good morning Ironic o/ | 07:50 |
| Qianbiao | Hi TheJulia, is it possible that enable console interface for ibmc in stable branch? | 07:50 |
| Qianbiao | Our customer is using stable/train now, and they are looking for serial console feature. | 07:51 |
| *** fmuyassarov_ has joined #openstack-ironic | 07:57 | |
| arne_wiebalck | Good morning iurygregory Qianbiao and ironic! | 07:59 |
| iurygregory | hey arne_wiebalck and Qianbiao o/ | 07:59 |
| Qianbiao | Morning arne | 07:59 |
| Qianbiao | Morning iurygregory o/ | 07:59 |
| *** fmuyassarov has quit IRC | 08:00 | |
| *** xinliang has joined #openstack-ironic | 08:08 | |
| *** rcernin_ has joined #openstack-ironic | 08:09 | |
| *** mkrai has joined #openstack-ironic | 08:16 | |
| *** rpittau|afk is now known as rpittau | 08:19 | |
| rpittau | good morning ironic! o/ | 08:19 |
| *** rcernin_ has quit IRC | 08:22 | |
| openstackgerrit | Xinliang Liu proposed openstack/ironic-specs master: Propose uefi pxe boot from volume https://review.opendev.org/762474 | 08:38 |
| *** rcernin_ has joined #openstack-ironic | 08:40 | |
| *** early has quit IRC | 08:45 | |
| iurygregory | hey rpittau o/ | 08:46 |
| *** early has joined #openstack-ironic | 08:47 | |
| *** dougsz has joined #openstack-ironic | 08:49 | |
| openstackgerrit | Feruzjon Muyassarov proposed openstack/ironic master: Allow disabling automated_clean per node https://review.opendev.org/762343 | 08:51 |
| *** rcernin_ has quit IRC | 08:57 | |
| *** rcernin_ has joined #openstack-ironic | 09:01 | |
| *** lucasagomes has joined #openstack-ironic | 09:03 | |
| *** rcernin_ has quit IRC | 09:06 | |
| *** afasano has joined #openstack-ironic | 09:11 | |
| *** yolanda has quit IRC | 09:29 | |
| *** yolanda has joined #openstack-ironic | 09:30 | |
| *** ociuhandu has joined #openstack-ironic | 09:32 | |
| *** mkrai has quit IRC | 09:45 | |
| *** zzzeek has quit IRC | 09:45 | |
| *** zzzeek has joined #openstack-ironic | 09:46 | |
| *** xinliang has quit IRC | 09:49 | |
| *** k_mouza has joined #openstack-ironic | 09:55 | |
| *** fmuyassarov__ has joined #openstack-ironic | 10:11 | |
| *** fmuyassarov_ has quit IRC | 10:13 | |
| *** ociuhandu has quit IRC | 10:21 | |
| *** zzzeek has quit IRC | 10:22 | |
| *** zzzeek has joined #openstack-ironic | 10:25 | |
| *** dtantsur|afk is now known as dtantsur | 10:29 | |
| dtantsur | good (?) morning (??) ironic | 10:30 |
| dtantsur | is it Friday yet? | 10:30 |
| rpittau | it will be soon, but never too soon | 10:32 |
| *** ociuhandu has joined #openstack-ironic | 10:34 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic stable/victoria: Fix DHCP-less operations with the noop network interface https://review.opendev.org/762494 | 10:39 |
| *** k_mouza has quit IRC | 10:39 | |
| *** tosin has joined #openstack-ironic | 10:41 | |
| *** k_mouza has joined #openstack-ironic | 10:48 | |
| tosin | Hello. I am currently trying to run emulators on my Ubuntu. When I try to install libvirt, I get this error 'E: Unable to locate package libvirt' | 10:53 |
| tosin | Can I please get some help with this | 10:53 |
| tosin | The same happens for libvirt-python and virt-install | 10:54 |
| dtantsur | tosin: not an ubuntu expert, but I think it's called libvirt-bin | 10:54 |
| dtantsur | and libvirt-dev | 10:54 |
| rpittau | tosin: and python-libvirt | 10:55 |
| tosin | I got this error with libvirt-bin 'E: Package 'libvirt-bin' has no installation candidate' | 10:56 |
| rpittau | tosin: actualyy that depends | 10:56 |
| rpittau | tosin: which version of ubuntu ? | 10:56 |
| tosin | Ubuntu 20.04 | 10:56 |
| rpittau | tosin: ok, so that would be python3-libvirt, libvirt-clients, libvirt-daemon, and libvirt-dev, if I'm not mistaken | 10:58 |
| rpittau | you probably also need virt-manager | 10:59 |
| *** rh-jelabarre has quit IRC | 11:02 | |
| *** rh-jelabarre has joined #openstack-ironic | 11:02 | |
| iurygregory | virt-manager would be necessary I would say | 11:02 |
| dtantsur | I'm actually not sure, it's a GUI after all | 11:03 |
| rpittau | yep, right, that's just a nice tool, but not strictly required | 11:06 |
| tosin | I have installed kvm and virt-manager and it still didn't work. When I ran the command kvm-ok, it says /dev/kvm does not exist. I think it's a problem with my hardware | 11:28 |
| *** tosin54 has joined #openstack-ironic | 11:29 | |
| *** tosin has quit IRC | 11:30 | |
| *** tosin has joined #openstack-ironic | 11:30 | |
| fmuyassarov__ | Hello, | 11:48 |
| fmuyassarov__ | When disabling automated cleaning globally, does Ironic still clean the root disk or is it also kept untouched ? | 11:48 |
| dtantsur | fmuyassarov__: the root disk is always cleaned during deployment right before writing to it | 11:50 |
| fmuyassarov__ | That's great. Thank you dtantsur | 11:50 |
| *** tosin has quit IRC | 11:51 | |
| *** k_mouza has quit IRC | 11:51 | |
| janders | I'm a little stuck with https://storyboard.openstack.org/#!/story/2008038, specifically around passing the lookup from Ironic to Inspector. I thought I'm doing it right but testing is showing otherwise :). dtantsur would you have some time to spare after the meeting to look into this with me? | 11:54 |
| dtantsur | janders: the schedule is a bit packed today, maybe we can do it via emails? if not, I probably have a 30 minutes slot at 13:30 UTC, but it may be too late for you. | 11:56 |
| janders | dtantsur no worries, let's try emails - and thank you! :) | 11:57 |
| *** thiagop has joined #openstack-ironic | 11:58 | |
| *** bfournie has quit IRC | 11:59 | |
| *** tosin has joined #openstack-ironic | 12:12 | |
| *** ociuhandu has quit IRC | 12:13 | |
| *** k_mouza has joined #openstack-ironic | 12:14 | |
| *** tosky has joined #openstack-ironic | 12:18 | |
| *** tosin has quit IRC | 12:23 | |
| *** ociuhandu has joined #openstack-ironic | 12:27 | |
| *** ociuhandu has quit IRC | 12:32 | |
| *** tosin has joined #openstack-ironic | 12:45 | |
| *** tosin has quit IRC | 12:53 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Fix incorrect network_data.json location https://review.opendev.org/762333 | 13:09 |
| *** tosin has joined #openstack-ironic | 13:20 | |
| *** paras333 has joined #openstack-ironic | 13:23 | |
| *** tosin has quit IRC | 13:32 | |
| *** rloo has joined #openstack-ironic | 13:51 | |
| *** Goneri has joined #openstack-ironic | 13:53 | |
| TheJulia | good morning | 14:02 |
| TheJulia | Qianbiao: that would basically be a feature addition. But if your just wanting to label update the hardware type so it is possible to use ipmitool-socat then I think that could be acceptable if it was an oversight.. maybe. Dmitry might have thoughts on this | 14:05 |
| TheJulia | dtantsur: tomorrow is friday! We shall SPUC as well. | 14:05 |
| Qianbiao | TheJulia Yes, I just want to enable console interface ipmi shellinabox for ibmc driver | 14:14 |
| TheJulia | i would have thought we would have had socat console on generichardware | 14:21 |
| TheJulia | that seems like a bug.... | 14:22 |
| TheJulia | dtantsur: If you have a thought regarding ^ | 14:22 |
| dtantsur | sec, on 1-1 | 14:22 |
| dtantsur | and good morning | 14:22 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-tempest-plugin master: Run redfish victoria job on focal https://review.opendev.org/760102 | 14:27 |
| dtantsur | TheJulia: yeah, hmm. maybe we need GenericIPMIHardware. | 14:27 |
| dtantsur | that most of hardware types inherit | 14:27 |
| dtantsur | (but not SNMP, manual management and so on) | 14:28 |
| TheJulia | Qianbiao's case is kind of in that featureish gray area, but it seems logical and seems like we should have kind of enabled in some way :\ | 14:32 |
| *** bfournie has joined #openstack-ironic | 14:37 | |
| rpittau | dtantsur: is this https://review.opendev.org/762326 related to the issue with dd ? | 14:38 |
| patchbot | patch 762326 - bifrost - Fix handling rootwrap.d filters - 1 patch set | 14:38 |
| dtantsur | rpittau: it is, yes | 14:38 |
| rpittau | ok, thanks, I will rebase the focal migration for vmedia on that then | 14:39 |
| rpittau | uh no wait | 14:39 |
| rpittau | that's bifrost | 14:39 |
| dtantsur | well, actually, I had "access denied" even with that patch | 14:39 |
| rpittau | ah | 14:39 |
| dtantsur | but my refactoring patch removes the code that causes it :) | 14:40 |
| rpittau | lol | 14:40 |
| dtantsur | (but we also have dd in the iscsi deploy) | 14:40 |
| rpittau | ok so it's similar to what I'm seeing in ironic vmedia jobs | 14:40 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Fix default disk label with partition images https://review.opendev.org/762453 | 14:40 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Change default ironic_lib invocation to flag local booting https://review.opendev.org/762528 | 14:40 |
| dtantsur | rpittau: see if https://review.opendev.org/757598 helps | 14:41 |
| patchbot | patch 757598 - ironic - Simplify injecting network data into an ISO image - 7 patch sets | 14:41 |
| rpittau | ack | 14:41 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Convert last bionic jobs to focal https://review.opendev.org/760326 | 14:43 |
| Qianbiao | TheJulia dtantsur by default ipmi serial console is not enabled ~ | 14:43 |
| Qianbiao | Yes i am agree with dtantsur, we need a GenericIpmiHardware | 14:44 |
| TheJulia | Yeah, I think that is the right path | 14:44 |
| TheJulia | tox -epy3 | 14:49 |
| dtantsur | TheJulia: 4532 tests failed, 19 passed | 14:49 |
| TheJulia | lol | 14:50 |
| iurygregory | dtantsur, lol | 14:50 |
| dtantsur | but at least that was quick, right? | 14:50 |
| iurygregory | very quick | 14:51 |
| TheJulia | well, this is in ipa | 14:51 |
| TheJulia | so more like 600 tests | 14:51 |
| dtantsur | heh, okay | 14:52 |
| *** tosin has joined #openstack-ironic | 14:52 | |
| *** tosin has quit IRC | 14:53 | |
| dtantsur | TheJulia: oh, yeah YEAH re SPUC. I'll need someone for the morning session though since it'll be my day off and I'll be out in the morning. | 14:53 |
| TheJulia | My shceduled sleep is during that time | 14:53 |
| dtantsur | yeah, I rather mean someone from EU or APAC time | 14:54 |
| dtantsur | *quick glance at iurygregory* | 14:54 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Fix boot mode detection for partition images https://review.opendev.org/762531 | 14:58 |
| iurygregory | I'm ok running | 14:58 |
| iurygregory | =) | 14:58 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Fix boot mode detection for partition images https://review.opendev.org/762531 | 15:00 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Support partition image efi contents https://review.opendev.org/748049 | 15:01 |
| TheJulia | and I bet that will work perfectly | 15:01 |
| TheJulia | I can see one issue with the uefi boot path, but only if someone manually changes UEFI boot to bios boot mode | 15:03 |
| TheJulia | which would be unsupported | 15:04 |
| *** fmuyassarov__ has quit IRC | 15:06 | |
| *** yolanda has quit IRC | 15:11 | |
| rpittau | dtantsur: I can't be there in the evening, but what time is the morning SPUC ? | 15:17 |
| TheJulia | I'll be there in the evening | 15:22 |
| *** dsneddon has quit IRC | 15:25 | |
| openstackgerrit | Merged openstack/bifrost master: Fix handling rootwrap.d filters https://review.opendev.org/762326 | 15:25 |
| TheJulia | that moment when you change something that should have broken unit tests, and nothing fails | 15:28 |
| *** ociuhandu has joined #openstack-ironic | 15:35 | |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic-python-agent master: Add clean step 'erase_pstore' https://review.opendev.org/762537 | 15:35 |
| arne_wiebalck | dtantsur: rpittau: ^^ this is for what we discussed the other day | 15:35 |
| rpittau | arne_wiebalck: yeah, thanks! | 15:36 |
| arne_wiebalck | tested on real hardware with a filled pstore :) | 15:36 |
| dtantsur | arne_wiebalck: awesome! could you update https://docs.openstack.org/ironic-python-agent/latest/admin/hardware_managers.html#clean-steps please? | 15:41 |
| arne_wiebalck | dtantsur: oh, es, I missed that ... thanks! | 15:42 |
| arne_wiebalck | *yes | 15:42 |
| *** ociuhandu has quit IRC | 15:47 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Fix disk label to account for UEFI https://review.opendev.org/762540 | 15:50 |
| *** Qianbiao has quit IRC | 15:50 | |
| *** Qianbiao has joined #openstack-ironic | 15:51 | |
| *** Qianbiao has quit IRC | 15:53 | |
| TheJulia | arne_wiebalck: well that is a new one... I had no idea that was a kernel feature | 15:54 |
| TheJulia | and long after I was building my own kernels on a regular basis | 15:55 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic-python-agent master: Add clean step 'erase_pstore' https://review.opendev.org/762537 | 15:57 |
| arne_wiebalck | dtantsur: for the reno, I was not sure if this should go into feature or security | 15:58 |
| TheJulia | both most likely | 15:59 |
| TheJulia | the security nature alone makes it compelling to backport | 15:59 |
| arne_wiebalck | TheJulia: ok, thanks, I can add a line | 15:59 |
| dtantsur | yeah, maybe with different emphasis | 16:00 |
| TheJulia | ++ | 16:06 |
| TheJulia | the instructions basically suggest that | 16:06 |
| *** k_mouza has quit IRC | 16:06 | |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic-python-agent master: Add clean step 'erase_pstore' https://review.opendev.org/762537 | 16:12 |
| dtantsur | arne_wiebalck: do you have any public links re your ipmitool cipher suite problems? | 16:17 |
| arne_wiebalck | dtantsur: only code :) | 16:18 |
| dtantsur | okay. I'm talking to ipmitool folks at RH now. | 16:19 |
| dtantsur | arne_wiebalck: but you had incorrect negotiation of a cipher suite, or? | 16:19 |
| arne_wiebalck | dtantsur: yes, the BMC was advertising, but could not deliver | 16:20 |
| arne_wiebalck | dtantsur: https://gitlab.cern.ch/hw/ipmiproxy | 16:20 |
| arne_wiebalck | (not sure you can see this?) | 16:20 |
| dtantsur | yeah, I've seen it, thanks! | 16:21 |
| * TheJulia updates all the merged things on the etherpad | 16:21 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost stable/victoria: Fix handling rootwrap.d filters https://review.opendev.org/762548 | 16:23 |
| *** ociuhandu has joined #openstack-ironic | 16:27 | |
| *** sdanni has joined #openstack-ironic | 16:38 | |
| JayF | I'm really confused about erase_pstore. Can that stuff persist across reboots? | 16:40 |
| TheJulia | apparently so | 16:40 |
| *** k_mouza has joined #openstack-ironic | 16:40 | |
| JayF | Hmm. | 16:40 |
| TheJulia | the last dying breath into nvram | 16:40 |
| JayF | My spidey sense is tingling. I wonder if that could break badly configured hardware... | 16:41 |
| TheJulia | The way I read the lwn post on it, that it was a separate area of memory that was moderately platform dependent that could house tens to hundreds of kilobytes | 16:41 |
| * dtantsur is immediately thinking of sinister ways to use this for ironic | 16:42 | |
| TheJulia | rutro | 16:42 |
| JayF | It's pretty brave to enable it as a default cleaning step unless we've tested with a lot of hardware... | 16:42 |
| JayF | let me put it this way: I would not enable that step in my downstream | 16:42 |
| JayF | at least not without additional testing and validation | 16:42 |
| TheJulia | I'd like to at least see how it shakes out in CI and maybe see if we can get some vendor testing of it, unless arne_wiebalck can give us more details of his use of it downstream | 16:43 |
| JayF | Like I' | 16:43 |
| JayF | *I'm not -1, but I'm -0 on it right now | 16:43 |
| JayF | I think our default clean steps being maximally simple is an advantage | 16:43 |
| dtantsur | if really need to finish https://review.opendev.org/744117 so that having stuff on by default would be a smaller priority | 16:43 |
| patchbot | patch 744117 - ironic - Generic way to configure clean step priorites - 17 patch sets | 16:43 |
| arne_wiebalck | JayF: yes, it survives reboots | 16:44 |
| openstackgerrit | Merged openstack/ironic stable/victoria: Fix redfish BIOS apply config error handling https://review.opendev.org/762437 | 16:44 |
| openstackgerrit | Merged openstack/ironic stable/victoria: Move the multinode grenade job to the experimental pipeline https://review.opendev.org/760118 | 16:44 |
| arne_wiebalck | JayF: I found some dmesg dumps from 2 years ago on a node that is re-instantiated every hour | 16:45 |
| TheJulia | nice | 16:46 |
| arne_wiebalck | TheJulia: there is no use of it here atm, I only learned about this 2 or 3 weeks ago :) | 16:47 |
| *** lmcgann has joined #openstack-ironic | 16:47 | |
| arne_wiebalck | TheJulia: a user was pointing me to these files that were not from his instance | 16:47 |
| openstackgerrit | Julia Kreger proposed openstack/ironic stable/train: Sync boot mode when changing the boot device via Redfish https://review.opendev.org/762439 | 16:48 |
| TheJulia | it occurs to me... pflash is a thing in UEFI vms | 16:51 |
| dtantsur | that's where we put the UEFI loader (aka OVMF) | 16:53 |
| TheJulia | And that can be variable size... Ohhh! UEFI standalong jobs should be fun to watch | 16:54 |
| dtantsur | will anyone cry if I kill https://opendev.org/openstack/bifrost/src/branch/master/tools/vagrant_dev_env with fire? I'm not sure how long it has been since anyone has tried that | 16:55 |
| TheJulia | dtantsur: I used it about 6 months ago | 16:55 |
| dtantsur | okay, not so bad | 16:55 |
| TheJulia | fwiw | 16:55 |
| dtantsur | TheJulia: do you remember wtf https://opendev.org/openstack/bifrost/src/branch/master/tools/virsh_dev_env is? | 16:55 |
| dtantsur | seems to duplicate bifrost-create-vms | 16:56 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Fix DHCP-less operations with the noop network interface https://review.opendev.org/762494 | 16:56 |
| dtantsur | oh of course | 16:56 |
| TheJulia | if I remember correctly it was to have something that walked a user through not using vagrant | 16:57 |
| TheJulia | so using only virsh commands to get a development VM | 16:57 |
| dtantsur | "CentOS-7" yeaaaah | 16:57 |
| TheJulia | I think the commands are in the centralized docs, because the last chagne to the folder appears to be me moving the docs | 16:57 |
| TheJulia | but 4 years ago... | 16:57 |
| dtantsur | TheJulia: just I bring https://github.com/dtantsur/config/blob/master/virt-install.sh update instead? :) | 16:57 |
| dtantsur | s/just/should/ | 16:58 |
| dtantsur | s/update/upstream/ | 16:58 |
| dtantsur | sigh, how do you still agree to talk to me... | 16:58 |
| TheJulia | because I read dtantsur | 16:58 |
| TheJulia | I think that is fine, we just need to update the docs | 16:58 |
| *** lucasagomes has quit IRC | 17:00 | |
| dtantsur | TheJulia: "that" = the old commands or my approach with virt-builder/virt-install? | 17:00 |
| TheJulia | dtantsur that being your approach | 17:01 |
| TheJulia | you actively use it, so it is better in my mind since I dont' think anyone is using the virsh commands, at least that I know of | 17:01 |
| TheJulia | and even then, they are just templates | 17:01 |
| dtantsur | okay, cool | 17:02 |
| * TheJulia pokes her web browser with a stick and wonders why it has frozen twice in the last few minutes | 17:05 | |
| TheJulia | is etherpad working for folks? | 17:06 |
| dtantsur | I can open the whiteboard | 17:07 |
| iurygregory | let me see | 17:07 |
| iurygregory | yup | 17:07 |
| iurygregory | it's fine for me | 17:07 |
| TheJulia | okay... web browser | 17:07 |
| * TheJulia restarts it | 17:07 | |
| iurygregory | bad web browser | 17:07 |
| iurygregory | no cookies! | 17:07 |
| TheJulia | mmmm coookies | 17:07 |
| TheJulia | I need to make cookies soon | 17:08 |
| iurygregory | cookies are from tue - fri / brains on monday? | 17:08 |
| *** ociuhandu_ has joined #openstack-ironic | 17:09 | |
| *** ociuhandu has quit IRC | 17:13 | |
| *** ociuhandu_ has quit IRC | 17:14 | |
| TheJulia | opposite, I think | 17:14 |
| lmcgann | Hello, working on the security interface spec we have received some comments about multitenant use case with Keylime, but we are currently confused as to how this could even effect our work. Could someone please elaborate? | 17:17 |
| TheJulia | bahahahahaha I see why tempest tests sometimes fail on multinode | 17:18 |
| TheJulia | the test looks for ping results with "64 bytes from $address" | 17:19 |
| TheJulia | but thats not always the case | 17:19 |
| * TheJulia goes into lolsob mode to fix the test | 17:19 | |
| *** k_mouza has quit IRC | 17:21 | |
| *** k_mouza has joined #openstack-ironic | 17:21 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Expand testenv documentation https://review.opendev.org/762553 | 17:22 |
| openstackgerrit | Dmitry Tantsur proposed openstack/bifrost master: Support UEFI in bifrost-cli and document it https://review.opendev.org/762555 | 17:24 |
| openstackgerrit | Verification of a change to openstack/ironic failed: Fix incorrect network_data.json location https://review.opendev.org/762333 | 17:27 |
| * arne_wiebalck has a day off tomorrow | 17:29 | |
| arne_wiebalck | see you all next week o/ | 17:29 |
| *** dougsz has quit IRC | 17:31 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-tempest-plugin master: Fix byte check in tempest plugin https://review.opendev.org/762556 | 17:31 |
| TheJulia | lmcgann: can you point me to the comment? | 17:32 |
| dtantsur | I also have a day off tomorrow, so see you on Monday | 17:34 |
| *** dtantsur is now known as dtantsur|afk | 17:34 | |
| * dtantsur|afk will try to be online for the evening SPUC | 17:35 | |
| lmcgann | TheJulia: its on line 62. There was a comment about multitenancy in the previous version of the spec as well I could dig up | 17:36 |
| lmcgann | https://review.opendev.org/#/c/576718/13/specs/approved/security-interface.rst@62 | 17:36 |
| patchbot | patch 576718 - ironic-specs - Add security interface spec - 13 patch sets | 17:36 |
| TheJulia | lmcgann: its not going to be a problem for the packets to get routed is it? | 17:37 |
| openstackgerrit | Merged openstack/ironic-python-agent master: Log configuration options on start-up https://review.opendev.org/762338 | 17:39 |
| lmcgann | I have no idea why it would or wouldnt be | 17:41 |
| *** k_mouza has quit IRC | 17:45 | |
| *** k_mouza has joined #openstack-ironic | 17:46 | |
| *** dking has quit IRC | 17:47 | |
| rpittau | good night! o/ | 17:51 |
| *** rpittau is now known as rpittau|afk | 17:51 | |
| *** k_mouza has quit IRC | 17:51 | |
| *** k_mouza has joined #openstack-ironic | 17:52 | |
| *** k_mouza has quit IRC | 17:52 | |
| TheJulia | lmcgann: I suspect it is perception of OpenCIT which requires it. | 18:01 |
| *** trandles has quit IRC | 18:12 | |
| sdanni | TheJulia: lmcgann is having lunch so I'm hijacking the conversation. Is OpenCIT relevant to multitenancy? I find that OpenCIT is deprecated now. | 18:13 |
| *** akiselev has joined #openstack-ironic | 18:28 | |
| openstackgerrit | Merged openstack/ironic master: Retrieve BIOS configuration when moving node to ``manageable`` https://review.opendev.org/761978 | 18:29 |
| openstackgerrit | Merged openstack/ironic stable/ussuri: Use assertCountEqual instead of assertItemsEqual https://review.opendev.org/762138 | 18:30 |
| *** trandles has joined #openstack-ironic | 18:30 | |
| *** afasano has quit IRC | 18:35 | |
| TheJulia | I don't think it is relevant at all | 18:41 |
| TheJulia | sdanni: lmcgann: ^^ | 18:41 |
| sdanni | TheJulia: I have another question about a different comment. https://review.opendev.org/#/c/576718/13/specs/approved/security-interface.rst@161 | 18:50 |
| patchbot | patch 576718 - ironic-specs - Add security interface spec - 13 patch sets | 18:50 |
| TheJulia | okay | 18:50 |
| sdanni | during node cleaning/deployment, we can create a keylime-agent element into the ramdisk. After the node is active, should we also install ipa and keylime-agent onto user image? | 18:51 |
| TheJulia | errr | 18:52 |
| TheJulia | The ramdisks are pre-built | 18:52 |
| TheJulia | so the element would create a ramdisk with the required agent/data that should be generally applicable | 18:53 |
| TheJulia | installing ipa or keylime intoa user image is optional, and ipa into a user instance is actually not supported and a risk | 18:53 |
| sdanni | TheJulia: so should we attest the node when it is in active state? | 18:56 |
| TheJulia | ironic wouldn't control that, that would be up to a user to decide they want to attest the node because it is the user's instance | 18:56 |
| TheJulia | it may be to a different attestation system that ironic has no clue about nor cares about | 18:56 |
| sdanni | TheJulia: got it. Thanks! | 18:57 |
| openstackgerrit | Merged openstack/ironic master: Fix idrac-wsman RAID step async error handling https://review.opendev.org/755731 | 19:09 |
| *** gyee has joined #openstack-ironic | 19:19 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: json-rpc: surround IPv6 address with [] in conductor URL https://review.opendev.org/762106 | 19:28 |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: Support partition image efi contents https://review.opendev.org/748049 | 19:38 |
| *** JayF has quit IRC | 20:04 | |
| *** JayF has joined #openstack-ironic | 20:04 | |
| iurygregory | rpioso, hey o/ by any chance you know if the Firmware Versions for idrac8 are always "2.A.B.C" and idrac9 are always "4.A.B.C" ? =) | 20:13 |
| rpioso | iurygregory: I believe so, but I'm not absolutely certain, especially about "always". I've reached out to a colleague. Let's see what they say. | 20:21 |
| iurygregory | rpioso, thanks! =) | 20:21 |
| iurygregory | if there is a casa where idrac8 would have firmware "4.A.B.C" for example =) | 20:22 |
| *** ociuhandu has joined #openstack-ironic | 20:22 | |
| *** ociuhandu has quit IRC | 20:25 | |
| *** ociuhandu_ has joined #openstack-ironic | 20:25 | |
| rpioso | iurygregory: Version numbers are not duplicated nor overlap like that across iDRAC generations. | 20:27 |
| iurygregory | cool, but would be a chance that idrac 8 would have "3.A.B.C" ? | 20:28 |
| iurygregory | or the first number would be always 2 and you only increment A B and C ? | 20:28 |
| iurygregory | asking because we will add some validation for the minimum firmware version that the hardware need to have in case they want to use virtual media on metal3 | 20:29 |
| *** ociuhandu has joined #openstack-ironic | 20:31 | |
| *** ociuhandu_ has quit IRC | 20:31 | |
| *** rcernin has joined #openstack-ironic | 20:32 | |
| *** rcernin has quit IRC | 20:35 | |
| rpioso | iurygregory: It's more gnarly than I suggested above. When iDRAC 8 came out, it could run iDRAC 7 firmware "1.A.B.C". It was "backwards compatible". That can still happen as new generations emerge. | 20:39 |
| rpioso | iurygregory: Couldn't the presence of the #VirtualMedia.InsertMedia action be checked, instead? | 20:39 |
| iurygregory | this would be something redfish can tell me? | 20:40 |
| rpioso | iurygregory: Yes, it can. | 20:42 |
| iurygregory | so if i have an dell with idrac8/9 that doesn't work with virtual media and I do some request to redfish asking for "#VirtualMedia.InsertMedia" I would probably get "Yes" or "No" ? | 20:43 |
| iurygregory | do you know how it would be that request in a normal curl so I can do some investigation on the hardware I have? | 20:44 |
| iurygregory | =) | 20:44 |
| rpioso | iurygregory: https://github.com/dell/iDRAC-Redfish-Scripting/blob/d3e674b05b7b699c4af185ebb9c4993a904e4128/Redfish%20Python/InsertEjectVirtualMediaREDFISH.py#L42-L55 | 20:50 |
| * rpioso hopes iurygregory is satisfied with Python requests code, instead of curl. | 20:50 | |
| iurygregory | rpioso, totally fine! =) | 20:50 |
| iurygregory | I will test in a few machines tomorrow =) | 20:51 |
| iurygregory | tyvm! | 20:51 |
| rpioso | iurygregory: I wouldn't suggest you hard code it like that, though. While the path may be valid for a long time, it could change. It would be better to walk the Redfish tree from the System Root to the Manager VirtualMedia subresource. | 20:52 |
| iurygregory | rpioso, yeah I agree | 20:53 |
| iurygregory | and I would probably need to find a generic way to check other hardware vendors to see how it goes =) | 20:54 |
| rpioso | iurygregory: That's how you'd do it in sushy. sushy doesn't publicly expose the vmedia actions. It's a private data member. :-( | 20:55 |
| iurygregory | gotcha | 20:55 |
| rpioso | iurygregory: Are you familiar with standing up Zuul v3? | 20:56 |
| iurygregory | I only know a little bit since we changed a few jobs from zuulv2 to zuulv3 | 20:56 |
| iurygregory | but if I can't help there is always #zuul channel =) | 20:57 |
| *** mrda has quit IRC | 20:58 | |
| rpioso | iurygregory: Did ironic stand up its own Zuul v3 or do we share/use #openstack-infra 's or #zuul 's? | 20:59 |
| iurygregory | rpioso, well we use the infra from openstack =) we only had to change a few templates for the jobs | 20:59 |
| iurygregory | #openstack-infra would have some information also | 21:00 |
| rpioso | iurygregory: Gotcha. That makes sense. Thank you! | 21:00 |
| iurygregory | both channels would be useful I would say | 21:00 |
| *** mrda has joined #openstack-ironic | 21:01 | |
| *** ociuhandu_ has joined #openstack-ironic | 21:03 | |
| *** ociuhandu has quit IRC | 21:06 | |
| *** ociuhandu has joined #openstack-ironic | 21:17 | |
| *** ociuhandu_ has quit IRC | 21:18 | |
| TheJulia | iurygregory: and idrac9's did have 3.x versions for a while if memory serves | 21:21 |
| * TheJulia sighs at grub | 21:24 | |
| *** ociuhandu has quit IRC | 21:34 | |
| openstackgerrit | Verification of a change to openstack/ironic failed: Also wipe agent token on manual power off or reboot https://review.opendev.org/752393 | 21:35 |
| *** kuku has joined #openstack-ironic | 21:35 | |
| janders | good morning Ironic o/ | 21:38 |
| stevebaker | hi | 21:40 |
| iurygregory | morning janders stevebaker o/ | 21:41 |
| kuku | Using this [https://0bin.net/paste/3O-CAgCi#MwkuFpCjQUoRVQAL7S3nMW5Sg5RFQVgiY1v+puJ7t21] as the payload for PUT /v1/nodes/....../states/provision, however cloud init doesn't set the ip to the nics | 21:41 |
| *** rcernin has joined #openstack-ironic | 21:47 | |
| *** paras333 has quit IRC | 21:51 | |
| *** k_mouza has joined #openstack-ironic | 21:53 | |
| *** k_mouza has quit IRC | 21:57 | |
| *** lmcgann has quit IRC | 22:00 | |
| *** rcernin has quit IRC | 22:01 | |
| *** Goneri has quit IRC | 22:07 | |
| *** sdanni has quit IRC | 22:10 | |
| *** rcernin has joined #openstack-ironic | 22:14 | |
| *** zbitter has joined #openstack-ironic | 22:29 | |
| *** persia has quit IRC | 22:31 | |
| *** zaneb has quit IRC | 22:31 | |
| *** persia has joined #openstack-ironic | 22:32 | |
| *** zzzeek has quit IRC | 22:34 | |
| *** zzzeek has joined #openstack-ironic | 22:34 | |
| *** tosky has quit IRC | 22:44 | |
| *** zbitter is now known as zaneb | 22:49 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic-python-agent master: WIP: Support partition image efi contents https://review.opendev.org/748049 | 22:55 |
| TheJulia | back to wip because yeah.. :( | 22:56 |
| TheJulia | Didn't update the unit tests | 22:56 |
| TheJulia | kuku: umm... as the config drive network metadata? | 22:57 |
| TheJulia | I wonder if youre version of cloud-init doesn't grok it | 22:57 |
| *** rloo has quit IRC | 23:13 | |
| *** Goneri has joined #openstack-ironic | 23:18 | |
| kuku | TheJulia: Yes, thats within configdriver and then network_data | 23:21 |
| kuku | Cloud init 0.7.5 ironic 1.58 | 23:22 |
| kuku | Python agent sets the right ip during setup, but when you the os starts, the interface doesn't have the ip | 23:24 |
| kuku | even the interfaces aren't up | 23:25 |
| *** akiselev has quit IRC | 23:30 | |
| openstackgerrit | Merged openstack/ironic stable/ussuri: Handle default_boot_mode during cleaning/inspection with PXE/iPXE https://review.opendev.org/751577 | 23:38 |
| openstackgerrit | Merged openstack/ironic master: Fixes the issue that instance bond port can't get IP address https://review.opendev.org/761161 | 23:41 |
| openstackgerrit | Merged openstack/ironic master: Remove root device hint after delete_configuration https://review.opendev.org/757346 | 23:41 |
| openstackgerrit | Merged openstack/ironic-inspector master: Add possibility to configure application root https://review.opendev.org/757536 | 23:41 |
| openstackgerrit | Merged openstack/ironic stable/victoria: Fix DHCP-less operations with the noop network interface https://review.opendev.org/762494 | 23:41 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!