| *** hoonetorg has joined #openstack-ironic | 00:01 | |
| *** MentalSiege has quit IRC | 00:03 | |
| iurygregory | morning janders o/ | 00:24 |
|---|---|---|
| openstackgerrit | Julia Kreger proposed openstack/ironic master: WIP Implement secure RBAC for port groups https://review.opendev.org/c/openstack/ironic/+/763268 | 00:44 |
| *** jdandrea has quit IRC | 01:22 | |
| *** jamesdenton has quit IRC | 01:33 | |
| *** jamesden_ has joined #openstack-ironic | 01:34 | |
| *** eagereagle1 has joined #openstack-ironic | 01:54 | |
| eagereagle1 | hi ironic - does anyone have good experience/knowledge of ironic python agent builder? I'm using it to generate a centos 8 ramdisk and it has started failing for some unknown reason: gcc: error trying to exec 'cc1plus': execvp: No such file or directory error: command 'gcc' failed with exit status 1 | 01:56 |
| *** ricolin_ has joined #openstack-ironic | 02:14 | |
| *** lbragstad_ is now known as lbragstad | 02:15 | |
| *** paras333 has joined #openstack-ironic | 02:16 | |
| *** paras333 has quit IRC | 02:21 | |
| *** cp- has quit IRC | 02:33 | |
| *** cp- has joined #openstack-ironic | 02:38 | |
| *** cp- has quit IRC | 02:43 | |
| *** cp- has joined #openstack-ironic | 02:45 | |
| *** xinliang has joined #openstack-ironic | 02:55 | |
| *** mkrai has joined #openstack-ironic | 03:25 | |
| xinliang | Hello Ironic, please take a look at this RFE. And see if we need to add a spec. | 03:41 |
| xinliang | https://storyboard.openstack.org/#!/story/2008543 | 03:41 |
| xinliang | thank you | 03:41 |
| janders | xinliang does Primehpc FX700 support Fujitsu's existing iRMC Ironic driver? | 04:11 |
| janders | ( https://docs.openstack.org/ironic/latest/admin/drivers/irmc.html ) | 04:11 |
| xinliang | No, I checked that, it's different from iRMC | 04:12 |
| xinliang | janders: FX700's BMC use IPMI protocol with some OEM cmds. | 04:13 |
| janders | ACK. | 04:15 |
| janders | what problems are you currently having with the non-OEM IPMI boot setting commands? | 04:15 |
| janders | are those ignored, or do they cause problems? | 04:16 |
| janders | (depending on the use case I might be able to suggest an existing workaround) | 04:16 |
| xinliang | It just igored and says invalid cmd | 04:16 |
| janders | if you want, you can have a look at "noop" management driver | 04:17 |
| janders | with this bootdev commands won't be issued at all | 04:17 |
| janders | and the hardware can be configured to boot off local storage first and off the network if local storage has no bootable image | 04:17 |
| janders | this may allow you to work around this problem | 04:18 |
| janders | before a better solution is available | 04:18 |
| xinliang | janders: yes, I arm now doing it by so. But that can not handle things automatically | 04:20 |
| *** ricolin_ has quit IRC | 04:20 | |
| janders | xinliang good move! | 04:21 |
| xinliang | janders: we need to mannually set boot device in this way right? | 04:22 |
| xinliang | If we want to deploy automatically, we need to write a driver for it, right? | 04:23 |
| janders | xinliang yes, but if you put in the right configuration, it should be a one-off task (and this could be automated if Fujitsu provides suitable BIOS management tools) | 04:24 |
| *** ricolin has joined #openstack-ironic | 04:24 | |
| janders | xinliang regarding the driver it's hard to tell without looking deep into details | 04:25 |
| xinliang | janders: understand, is there any way to inject a set boot device step with noop management driver? | 04:27 |
| janders | no, I do not think so - the idea behind the noop driver is not interacting with the BIOS settings for boot management (for hardware that has issues with supporting that) | 04:28 |
| janders | one of the use cases for this driver is just having a static boot sequence configured once | 04:29 |
| xinliang | ok | 04:29 |
| janders | then when the image is written to disk, the node boots of disk and when the node is cleaned, it can no longer boot from disk | 04:29 |
| janders | so it will PXE | 04:29 |
| janders | that's one way how some people use it | 04:30 |
| xinliang | understand, if we always want netboot the machine we can do this | 04:31 |
| janders | or if as I said you configure bootorder 1) local-disk 2) netboot | 04:31 |
| janders | then if there is no image on the disk, it will netboot | 04:31 |
| janders | and if there is it will boot from disk | 04:31 |
| janders | and removing ironic instance will remove the image, so next time node is provisioned, it will netboot | 04:32 |
| janders | so the effective boot order still changes, just without changing it in the BIOS | 04:32 |
| janders | does this make sense to you? | 04:32 |
| janders | it's no golden bullet but it is quite powerful for what it is | 04:33 |
| xinliang | haha, it seems work for normal server in this way. Not sure it works for FX700, will check soon | 04:34 |
| xinliang | thank you janders | 04:35 |
| janders | no worries xinliang let us know how you go | 04:35 |
| janders | happy to help | 04:35 |
| xinliang | Yes, it help much :-) | 04:36 |
| *** pmannidi_ has joined #openstack-ironic | 04:50 | |
| *** pmannidi has quit IRC | 04:52 | |
| *** xinliang has quit IRC | 04:53 | |
| *** mkrai has quit IRC | 05:01 | |
| *** mkrai has joined #openstack-ironic | 05:01 | |
| *** ricolin has quit IRC | 05:11 | |
| *** ricolin has joined #openstack-ironic | 05:24 | |
| *** zzzeek has quit IRC | 05:41 | |
| *** zzzeek has joined #openstack-ironic | 05:42 | |
| *** tzumainn has quit IRC | 05:48 | |
| *** rcernin has quit IRC | 06:07 | |
| openstackgerrit | ankit proposed openstack/ironic master: Fixes issue of redfish firmware update https://review.opendev.org/c/openstack/ironic/+/764914 | 06:24 |
| *** uzumaki has joined #openstack-ironic | 06:38 | |
| *** anuradha1904 has joined #openstack-ironic | 06:41 | |
| *** gyee has quit IRC | 06:46 | |
| *** xinliang has joined #openstack-ironic | 07:01 | |
| *** mkrai has quit IRC | 07:02 | |
| *** moshiur has joined #openstack-ironic | 07:06 | |
| *** tkajinam_ has joined #openstack-ironic | 07:19 | |
| uzumaki | good morning ironic! | 07:20 |
| *** tkajinam has quit IRC | 07:20 | |
| uzumaki | Can the inspection data provide me storage device identifiers like Disk.Bay.0:Enclosure.Internal.0-1:RAID.Integrated.1-1? Which is sort of a UUID idrac uses to label storage devices. | 07:24 |
| arne_wiebalck | Good morning, ironic! | 07:50 |
| uzumaki | Morning arne_wiebalck comment ca va? | 07:53 |
| *** Qianbiao has joined #openstack-ironic | 07:56 | |
| uzumaki | Morning Qianbiao :) | 07:59 |
| *** ociuhandu has joined #openstack-ironic | 08:00 | |
| Qianbiao | morning uzumaki :) | 08:00 |
| arne_wiebalck | Hey uzumaki Qianbiao o/ | 08:01 |
| Qianbiao | hey arne, morning | 08:02 |
| uzumaki | How're we doing today? arne_wiebalck Qianbiao | 08:04 |
| *** mkrai has joined #openstack-ironic | 08:04 | |
| *** ociuhandu has quit IRC | 08:05 | |
| Qianbiao | I am idle recently. looking for new project :) | 08:05 |
| *** jamesden_ has quit IRC | 08:05 | |
| *** jamesdenton has joined #openstack-ironic | 08:05 | |
| arne_wiebalck | All good over here, preparing for the first meetings today ... :) | 08:06 |
| arne_wiebalck | And you? | 08:06 |
| *** lucasagomes has joined #openstack-ironic | 08:08 | |
| *** fgofurov has joined #openstack-ironic | 08:08 | |
| *** rpittau|afk is now known as rpittau | 08:11 | |
| rpittau | good morning ironic! o/ | 08:11 |
| *** tosky has joined #openstack-ironic | 08:16 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Replace retrying with tenacity https://review.opendev.org/c/openstack/ironic/+/376574 | 08:38 |
| openstackgerrit | Merged openstack/ironic master: Update iDRAC doc with missing interfaces https://review.opendev.org/c/openstack/ironic/+/771653 | 08:53 |
| *** dougsz has joined #openstack-ironic | 08:54 | |
| *** ociuhandu has joined #openstack-ironic | 08:58 | |
| *** ociuhandu has quit IRC | 09:04 | |
| *** ociuhandu has joined #openstack-ironic | 09:10 | |
| *** ociuhandu has quit IRC | 09:14 | |
| *** xinliang has quit IRC | 09:26 | |
| openstackgerrit | Riccardo Pittau proposed openstack/bifrost stable/ussuri: Fix CI issues https://review.opendev.org/c/openstack/bifrost/+/766742 | 09:32 |
| *** Rahman has joined #openstack-ironic | 09:35 | |
| *** QianbiaoNG has joined #openstack-ironic | 09:35 | |
| *** tosky has quit IRC | 09:36 | |
| *** tosky_ has joined #openstack-ironic | 09:36 | |
| QianbiaoNG | morning rpittau o/ | 09:37 |
| rpittau | hey QianbiaoNG :) | 09:37 |
| *** moshiur has quit IRC | 09:37 | |
| *** Qianbiao has quit IRC | 09:38 | |
| *** tosky_ is now known as tosky | 09:43 | |
| *** derekh has joined #openstack-ironic | 09:46 | |
| *** jamesdenton has quit IRC | 10:00 | |
| *** jamesdenton has joined #openstack-ironic | 10:01 | |
| arne_wiebalck | it seems like pip on my centos8 host does not like the comments in https://github.com/openstack/ironic-lib/blob/master/tox.ini#L39 ff. ... rpittau to the rescue :) | 10:04 |
| rpittau | lol | 10:04 |
| rpittau | arne_wiebalck: what version of centos8/pip ? | 10:07 |
| arne_wiebalck | rpittau: centos 8.2.2004 | 10:07 |
| arne_wiebalck | rpittau: pip 20.3.3 | 10:08 |
| arne_wiebalck | rpittau: removing the comments helps | 10:09 |
| rpittau | oh wow, what's complaining about? | 10:11 |
| arne_wiebalck | rpittau: http://paste.openstack.org/show/801811/ | 10:11 |
| arne_wiebalck | rpittau: it seems pip considers the comment to be part of the requirement | 10:12 |
| rpittau | yeah | 10:13 |
| rpittau | arne_wiebalck: can you downgrade pip to 20.3.1 for a test ? | 10:13 |
| arne_wiebalck | rpittau: I can try ... give me a sec ... | 10:14 |
| rpittau | I'm not sure it's pip, might be tox | 10:14 |
| arne_wiebalck | rpittau: same issue with 20.3.1 | 10:17 |
| rpittau | ooook | 10:17 |
| rpittau | alright, which version of tox ? | 10:17 |
| *** ociuhandu has joined #openstack-ironic | 10:18 | |
| rpittau | cause for me it works fine with pip 20.3.1 and 20.3.3 | 10:19 |
| arne_wiebalck | tox 3.4.0 | 10:19 |
| arne_wiebalck | I don't think it is tox. | 10:19 |
| rpittau | it's tox | 10:20 |
| rpittau | just tested | 10:20 |
| rpittau | can you upgrade it please ? | 10:20 |
| arne_wiebalck | http://paste.openstack.org/show/801812/ | 10:20 |
| openstackgerrit | ankit proposed openstack/sushy stable/victoria: Fixes issue of redfish firmware update https://review.opendev.org/c/openstack/sushy/+/771604 | 10:21 |
| *** mkrai has quit IRC | 10:21 | |
| rpittau | I tested with tox 3.4.0 and it gives me the same error, it was fixed in 3.9.0 | 10:22 |
| rpittau | I guess we need to update the min version of tox in tox.ini | 10:23 |
| rpittau | arne_wiebalck: https://tox.readthedocs.io/en/latest/changelog.html#v3-9-0-2019-04-17 | 10:25 |
| rpittau | https://github.com/tox-dev/tox/pull/1262 | 10:25 |
| arne_wiebalck | rpittau: nice! | 10:26 |
| arne_wiebalck | rpittau: thanks! | 10:26 |
| rpittau | anytime :) | 10:26 |
| arne_wiebalck | rpittau: I still do not see how this comes into the game: http://paste.openstack.org/show/801812/ | 10:26 |
| rpittau | remove the '' | 10:27 |
| arne_wiebalck | well, that makes it a shell comment | 10:28 |
| arne_wiebalck | ok, let me upgrade tox :) | 10:29 |
| rpittau | yeah, pip doesn't really understand comments in that context, only in requirements files and only if they're nmot surrounded by quotes | 10:30 |
| rpittau | you can test building a constraint file and adding a requirement with a comment, pip will fail if it's surrounded by quotes | 10:31 |
| arne_wiebalck | ah ok, I see | 10:33 |
| arne_wiebalck | tox 3.4.0 is the most recent in epel, btw | 10:33 |
| arne_wiebalck | (which is where I got it from) | 10:33 |
| rpittau | if you're using a private user to test, you can locally install a new version with pip maybe? | 10:34 |
| rpittau | using --user option | 10:34 |
| *** Dhuldev_Valekar has joined #openstack-ironic | 10:37 | |
| arne_wiebalck | tox 3.21.2 works | 10:38 |
| rpittau | oh yeah, that for sure :) | 10:38 |
| arne_wiebalck | as I said in the beginnig: must be tox | 10:38 |
| rpittau | :D | 10:38 |
| arne_wiebalck | :-D | 10:38 |
| arne_wiebalck | thanks for your help, rpittau ! | 10:38 |
| rpittau | glad to see it works :) | 10:38 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic-lib master: Handle EBRs and tiny partitions when removing metadata https://review.opendev.org/c/openstack/ironic-lib/+/771788 | 10:41 |
| *** QianbiaoNG has quit IRC | 10:41 | |
| arne_wiebalck | TheJulia: dtantsur|afk: from the discussion yesterday ^^ | 10:42 |
| *** mkrai has joined #openstack-ironic | 10:50 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-lib master: Update minversion of tox https://review.opendev.org/c/openstack/ironic-lib/+/771790 | 10:52 |
| rpittau | ^ let's formalize that | 10:53 |
| *** dtantsur|afk is now known as dtantsur | 10:58 | |
| dtantsur | good morning ironic | 11:03 |
| janders | good morning dtantsur | 11:03 |
| dtantsur | folks, especially ajya, rpioso and TheJulia, could you please check this proposal http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019946.html ? | 11:19 |
| *** rcernin has joined #openstack-ironic | 11:23 | |
| iurygregory | good morning Ironic o/ | 11:29 |
| openstackgerrit | Merged openstack/ironic master: Write stub ACL test for every existing API call https://review.opendev.org/c/openstack/ironic/+/767445 | 11:29 |
| *** k_mouza has joined #openstack-ironic | 11:29 | |
| ajya | dtantsur: will take a look, in short, I personally would prefer separating vendor specific code from generic. I can see how current situation can be confusing as ipmi can be used for many vendors, and redfish is supposed to be that way too, but for now when see Dell server, use 'idrac' hardware type that will handle valid interfaces. | 11:31 |
| dtantsur | ajya: I can see you point, but you need to carefully consider its implications. What the end customer will hear is "iDRAC is not Redfish compliant", because the nuances will certainly be lost on the way to them. | 11:32 |
| openstackgerrit | Merged openstack/ironic master: Attempt to slim down protection test base class https://review.opendev.org/c/openstack/ironic/+/770673 | 11:33 |
| *** mkrai has quit IRC | 11:38 | |
| ajya | dtantsur: sure, the end goal is to become compliant with generic driver, but it will take time. If considering moving vendor specific code to generic code, then have to think about what happens with idrac hardware type. Does it start supporting redfish interfaces too, what's in idrac-redfish interfaces - pure OEM, no workarounds, etc.. | 11:49 |
| dtantsur | ajya: I don't think anything should change for the idrac hardware type | 11:55 |
| dtantsur | I also don't suggest adding purely vendor features to the generic hardware type | 11:55 |
| dtantsur | (e.g. the reset iDRAC actions should stay in iDRAC) | 11:55 |
| *** rcernin has quit IRC | 11:57 | |
| ajya | dtantsur: me neither, just thinking about the case where user has set up redfish for Dell, then realize that need some OEM stuff, and need to reconfigure. | 12:00 |
| dtantsur | we have exactly this problem now: they start with Redfish for Dell, then vmedia suddenly does not work | 12:01 |
| *** ricolin has quit IRC | 12:06 | |
| *** QianbiaoNG has joined #openstack-ironic | 12:07 | |
| *** Dhuldev_Valekar has quit IRC | 12:14 | |
| *** jamesdenton has quit IRC | 12:18 | |
| *** jamesdenton has joined #openstack-ironic | 12:19 | |
| *** Dhuldev_Valekar has joined #openstack-ironic | 12:26 | |
| *** Dhuldev_Valekar has quit IRC | 12:40 | |
| *** ociuhandu has quit IRC | 12:49 | |
| *** rh-jelabarre has joined #openstack-ironic | 13:02 | |
| *** tosky has quit IRC | 13:06 | |
| *** tosky has joined #openstack-ironic | 13:06 | |
| *** ricolin has joined #openstack-ironic | 13:08 | |
| *** ociuhandu has joined #openstack-ironic | 13:23 | |
| *** Dhuldev_Valekar has joined #openstack-ironic | 13:31 | |
| *** paras333_ has joined #openstack-ironic | 13:31 | |
| *** rloo has joined #openstack-ironic | 13:53 | |
| *** paras333_ has quit IRC | 13:56 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Update minversion of tox https://review.opendev.org/c/openstack/ironic/+/771815 | 13:57 |
| *** uzumaki has quit IRC | 14:02 | |
| openstackgerrit | Merged openstack/ironic master: Follow-up for ramdisk deploy configdrive support https://review.opendev.org/c/openstack/ironic/+/770172 | 14:06 |
| iurygregory | rpittau, let us know if you are going to do some spam with the update minversion of tox =D | 14:09 |
| *** jdandrea has joined #openstack-ironic | 14:09 | |
| rpittau | iurygregory: I might :) | 14:09 |
| iurygregory | ok =) | 14:09 |
| rpittau | I'll try to distribute it, it's not a super urgent thing | 14:09 |
| iurygregory | no worries, I just because I will get email notifications so I want to disable if you are going to do hehehe | 14:10 |
| rpittau | I'm going to spam rdo first :P | 14:15 |
| fgofurov | Hello ironic 0/ | 14:21 |
| fgofurov | We were testing ironic scalability in metal3 and were wondering if you know/used any tools that can be used for metrics collections, i.e CPU, RAM, network output to see general the impact of the metrics in the environment | 14:23 |
| fgofurov | @arne wiebalck if you have used any? | 14:24 |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: Update requirements for diskimage-builder https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/771820 | 14:29 |
| *** Dhuldev_Valekar has quit IRC | 14:33 | |
| TheJulia | good morning | 14:33 |
| *** Rahman has quit IRC | 14:33 | |
| TheJulia | fgofurov: fwiw, irc client's key off of the irc handle, so in Arne's case it is arne_wiebalck, typically no @ required, you just mention his nickname. | 14:34 |
| fgofurov | Hi Julia, thanks for letting me know:) | 14:35 |
| dtantsur | morning TheJulia | 14:36 |
| arne_wiebalck | fgofurov: o/ | 14:41 |
| arne_wiebalck | fgofurov: these metrics are collected on our controllers via collectd and sent into our monitoring system | 14:43 |
| *** tzumainn has joined #openstack-ironic | 14:44 | |
| arne_wiebalck | fgofurov: apart from the RAM issue caused by image conversions during parallel deployments, like 100 nodes, I never needed to look at these | 14:44 |
| arne_wiebalck | fgofurov: when scaling our Ironic infra, there were always other issues so far :) | 14:45 |
| openstackgerrit | Dmitry Tantsur proposed openstack/sushy master: Add release version to release notes https://review.opendev.org/c/openstack/sushy/+/771823 | 14:46 |
| * arne_wiebalck goes and checks the metrics for the controllers ... | 14:46 | |
| fgofurov | Hi arne wiebalck! | 14:48 |
| fgofurov | Thanks. I see, would you mind sharing what other issues have you faced when scaling? | 14:49 |
| *** sshnaidm|ruck is now known as sshnaidm|afk | 14:49 | |
| arne_wiebalck | fgofurov: CPU on the controllers: 85% idle (on avg), each controller uses ~1.5G of RAM (this is sensitive to the number of API processes), network is mostly idle as well (each controller is managing ~500 nodes) | 14:53 |
| arne_wiebalck | fgofurov: https://techblog.web.cern.ch/techblog/post/conductor-groups/ | 14:53 |
| arne_wiebalck | fgofurov: this ^ was one of the main issues | 14:54 |
| arne_wiebalck | fgofurov: other scaling issues we saw included: DB accesses, image conversion, flavor management, and handling the 1% of nodes with issues | 14:56 |
| arne_wiebalck | fgofurov: power sync with nova | 14:57 |
| fgofurov | Thank you very much for detailed info and links, much appreciated. | 14:59 |
| arne_wiebalck | np :) | 15:00 |
| *** MentalSiege has joined #openstack-ironic | 15:06 | |
| *** uzumaki has joined #openstack-ironic | 15:07 | |
| *** MentalSiege has quit IRC | 15:10 | |
| *** dougsz has quit IRC | 15:15 | |
| *** mkowalski has quit IRC | 15:19 | |
| *** irclogbot_0 has quit IRC | 15:21 | |
| *** irclogbot_1 has joined #openstack-ironic | 15:22 | |
| *** rh-jelabarre has quit IRC | 15:27 | |
| TheJulia | fgofurov: fwiw, their deployment is built a little differently than the way metal3 works... as far as I'm really aware. They will have more restful API activity in their deployment which also consumes a portion of CPU resources, I think metal3 would be less, but much more serialized since it is a reconciliation loop. | 15:29 |
| *** QianbiaoNG has quit IRC | 15:29 | |
| *** rh-jelabarre has joined #openstack-ironic | 15:31 | |
| *** mkowalski has joined #openstack-ironic | 15:31 | |
| fgofurov | Julia: true, agree with that. And also we are not trying to have that much of nodes deployed in metal3, but was a good practice to do and test ironic scalability in metal3 context | 15:31 |
| TheJulia | ahh, yeah | 15:33 |
| TheJulia | Network will include, somewhere... the amount of bytes for the disk image and ramdisk(s) to get shipped across the wire depending on how deployments are occuring | 15:34 |
| TheJulia | but that is also kind of implied | 15:34 |
| TheJulia | we don't have magic "teleport deployment into $machine in $rack in $row in $data_center"... yet. | 15:34 |
| TheJulia | virtual media booting is the closest to that since it can shift some of that to OOB | 15:35 |
| *** dougsz has joined #openstack-ironic | 15:55 | |
| *** dougsz has quit IRC | 15:55 | |
| *** dougsz has joined #openstack-ironic | 15:55 | |
| *** dougsz has quit IRC | 15:55 | |
| *** dougsz has joined #openstack-ironic | 15:55 | |
| *** sshnaidm|afk is now known as sshnaidm|ruck | 15:56 | |
| trandles | Quick question: I have a bifrost install on a server. A colleague wants to use Ironic. He sourced the openrc file (only OS_ENDPOINT, OS_CLOUD, OS_AUTH_TYPE defined) and he gets "cloud bifrost not found" when trying 'openstack baremetal node list'...but it works for me... | 15:56 |
| dtantsur | trandles: do they have clouds.yaml in their profile? | 16:00 |
| dtantsur | IIRC bifrost installs this file for the calling user, not globally (I wonder if we should change that)\ | 16:00 |
| trandles | That's it, thanks dtantsur | 16:01 |
| trandles | We're a little torn at LANL. We don't allow shared "service" accounts on our systems, even if they can't be logged into remotely. Instead of having a shared "stack" account we do everything as ourselves. | 16:02 |
| trandles | I think we'll probably end up changing our account creation ansible play to automatically copy in .config/openstack/clouds.yaml for the people who should have access to it. | 16:03 |
| *** juanoterocas has joined #openstack-ironic | 16:03 | |
| dtantsur | yep. the problem with putting it to /etc/openstack is that all accounts will read the passwords | 16:05 |
| dtantsur | we don't feel safe defaulting to that | 16:05 |
| trandles | I would not default to that | 16:05 |
| *** anuradha1904 has quit IRC | 16:10 | |
| *** uzumaki has quit IRC | 16:21 | |
| *** tosin has joined #openstack-ironic | 16:29 | |
| *** gyee has joined #openstack-ironic | 16:32 | |
| openstackgerrit | Merged openstack/ironic stable/victoria: Add a delay/retry is vmedia insert fails https://review.opendev.org/c/openstack/ironic/+/771243 | 16:32 |
| bfournie | Hi! We're having an issue with Supermicro nodes booting to PXE instead of disk when the reboot is initiated outside of Ironic, I've described it here - https://storyboard.openstack.org/#!/story/2008547. It would work fine if the BootSourceOverrideEnabled': 'Continuous' is used but the final Redfish PATCH request removes it. I know there has been a lot of discussion about the use of BootSourceOverrideEnabled (http:/ | 16:38 |
| bfournie | nstack.org/pipermail/openstack-discuss/2020-April/014543.html) but its not clear the best way to solve this. dtantsur recommended using pxe.enable_netboot_fallback. I'll try this, but the fact that the Supermicro switches the BootMode from UEFI to Legacy when BootSourceOverrideEnabled': 'Continuous is not used makes me wonder if that will work | 16:38 |
| *** uzumaki has joined #openstack-ironic | 16:47 | |
| *** lucasagomes has quit IRC | 16:47 | |
| *** ociuhandu_ has joined #openstack-ironic | 16:49 | |
| *** ociuhandu has quit IRC | 16:53 | |
| *** ociuhandu_ has quit IRC | 16:54 | |
| *** uzumaki has quit IRC | 17:21 | |
| *** dougsz has quit IRC | 17:32 | |
| openstackgerrit | Yogesh proposed openstack/ironic master: idrac ipmi support intial version https://review.opendev.org/c/openstack/ironic/+/771860 | 17:33 |
| openstackgerrit | Yogesh proposed openstack/ironic master: WIP Add idrac HW type IPMI interface support https://review.opendev.org/c/openstack/ironic/+/771862 | 17:37 |
| rpittau | good night! o/ | 17:42 |
| *** rpittau is now known as rpittau|afk | 17:42 | |
| *** juanoterocas has quit IRC | 17:47 | |
| *** k_mouza has quit IRC | 17:48 | |
| openstackgerrit | Yogesh proposed openstack/ironic master: WIP Add idrac HW type IPMI interface support https://review.opendev.org/c/openstack/ironic/+/771862 | 17:52 |
| *** rcernin has joined #openstack-ironic | 17:54 | |
| *** k_mouza has joined #openstack-ironic | 17:57 | |
| *** rcernin has quit IRC | 17:59 | |
| *** cp- has quit IRC | 17:59 | |
| *** cp- has joined #openstack-ironic | 18:08 | |
| arne_wiebalck | bye everyone o/ | 18:09 |
| *** cp- has quit IRC | 18:14 | |
| *** cp- has joined #openstack-ironic | 18:30 | |
| *** dtantsur is now known as dtantsur|afk | 18:32 | |
| dtantsur|afk | o/ | 18:32 |
| *** tosin has quit IRC | 18:39 | |
| iurygregory | ajya, hey by any chance you can check if FW < 4.40.00.00 will return a list with the items? | 18:43 |
| *** jamesdenton has quit IRC | 18:43 | |
| *** jamesdenton has joined #openstack-ironic | 18:43 | |
| ajya | iurygregory: I can check, but it could be that older FW don't have the error :) | 18:45 |
| iurygregory | oh yeah (makes sense) | 18:45 |
| ajya | iurygregory: anyway 4.40.00.00 does not have InsertMedia working, I'm checking that with firmware team; about EjectMedia I still have to check if it's broken too and in what way | 18:46 |
| iurygregory | ajya, ack | 18:46 |
| iurygregory | so the possible workaround if a machine comes with FW 4.40.00.00 would be downgrade correct? | 18:46 |
| ajya | iurygregory: yes, even if EjectMedia could be fixed/worked around in sushy (it works for me using Redfish directly), InsertMedia is definitely broken and I couldn't bypass that. | 18:48 |
| iurygregory | ajya, ack tks for the info! | 18:48 |
| *** k_mouza has quit IRC | 19:03 | |
| TheJulia | stevebaker: w/r/t prints, I'm semi-conflicted since in other api tests we do print the data in the output to make them more troubleshootable. | 19:04 |
| *** bbezak has quit IRC | 19:28 | |
| *** bbezak has joined #openstack-ironic | 19:28 | |
| *** fgofurov has quit IRC | 19:32 | |
| *** dougsz has joined #openstack-ironic | 20:02 | |
| stevebaker | TheJulia: If they serve a purpose I think its fine to leave them in | 20:05 |
| *** jamesdenton has quit IRC | 20:06 | |
| *** jamesdenton has joined #openstack-ironic | 20:06 | |
| *** irclogbot_1 has quit IRC | 20:14 | |
| *** tosky has quit IRC | 20:14 | |
| *** tosky has joined #openstack-ironic | 20:15 | |
| *** k_mouza has joined #openstack-ironic | 20:16 | |
| *** irclogbot_1 has joined #openstack-ironic | 20:16 | |
| *** k_mouza has quit IRC | 20:20 | |
| *** dougsz has quit IRC | 20:31 | |
| TheJulia | ack, regardless I'm going to leave them until I'm likely later along, started chassis right before my last call | 21:18 |
| eagereagle1 | hi ironic. during deployment (wait callback) how would I verify what IP address is assigned to the baremetal node currently running IPA? | 21:24 |
| *** derekh has quit IRC | 21:37 | |
| *** tosin has joined #openstack-ironic | 21:40 | |
| TheJulia | define verify | 21:44 |
| TheJulia | you mean like ping it? | 21:44 |
| TheJulia | You can get the neutron port id by looking at the port internal_info field | 21:45 |
| TheJulia | depending on network interface the behavior changes a little | 21:46 |
| TheJulia | if the network_interface on the node is set to "flat", then it just re-uses the tenant VIF | 21:46 |
| *** jdandrea has quit IRC | 21:52 | |
| *** rcernin has joined #openstack-ironic | 21:54 | |
| *** rcernin has quit IRC | 21:59 | |
| *** derekh has joined #openstack-ironic | 22:06 | |
| *** derekh has quit IRC | 22:06 | |
| JayF | you can also look at driver_internal_info/callback_url for the IP, right? | 22:24 |
| TheJulia | JayF: only if it has successfully heartbeated once | 22:27 |
| JayF | ah, that's a good point | 22:27 |
| JayF | because if you're wondering "what IP?" you probably aren't heartbeating | 22:27 |
| TheJulia | exactly! | 22:27 |
| * TheJulia looks around for her spoons | 22:37 | |
| *** rcernin has joined #openstack-ironic | 22:38 | |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Implement secure RBAC for port groups https://review.opendev.org/c/openstack/ironic/+/763268 | 22:52 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Implement secure RBAC for chassis https://review.opendev.org/c/openstack/ironic/+/763269 | 22:53 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: Implement secure RBAC for baremetal drivers https://review.opendev.org/c/openstack/ironic/+/763270 | 22:53 |
| openstackgerrit | Julia Kreger proposed openstack/ironic master: WIP Implement secure RBAC for node and driver passthru https://review.opendev.org/c/openstack/ironic/+/763271 | 22:53 |
| * TheJulia thinks that is enough for today | 22:54 | |
| *** rcernin has quit IRC | 22:56 | |
| *** rcernin has joined #openstack-ironic | 22:57 | |
| *** uzumaki has joined #openstack-ironic | 22:58 | |
| janders | good morning Ironic o/ | 23:17 |
| *** ociuhandu has joined #openstack-ironic | 23:24 | |
| *** ociuhandu has quit IRC | 23:29 | |
| *** uzumaki has quit IRC | 23:54 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!