| opendevreview | Jacob Anders proposed openstack/sushy stable/xena: Use Etag in precondition header when setting boot options https://review.opendev.org/c/openstack/sushy/+/839842 | 05:16 |
|---|---|---|
| opendevreview | Jacob Anders proposed openstack/sushy stable/xena: Attempt using Redfish settings resource to get URI https://review.opendev.org/c/openstack/sushy/+/839843 | 05:16 |
| opendevreview | Jacob Anders proposed openstack/sushy stable/xena: Follow up to Attempt using Redfish settings resource to get URI https://review.opendev.org/c/openstack/sushy/+/839844 | 05:16 |
| opendevreview | Jacob Anders proposed openstack/sushy stable/xena: Re-trying InsertMedia call with TransferProtocolType if required https://review.opendev.org/c/openstack/sushy/+/839845 | 05:21 |
| rpittau | good morning ironic! o/ | 06:31 |
| janders | good morning rpittau and Ironic o/ | 07:48 |
| janders | Happy Friday | 07:48 |
| rpittau | hey janders :) | 07:48 |
| rpittau | Happy Friday! \o/ | 07:48 |
| ebbex | dtantsur: https://github.com/openshift/ironic-image/pull/268 is this something I can leverage in 4.11 in install-config.yaml to get ttyS0 output during `openshift-baremetal-install create cluster`? | 08:07 |
| hjensas | looks like xena metalsmith-integration-glance-localboot-centos8-uefi is broken. Exception: Versioning for this project requires either an sdist tarball, or access to an upstream git repository ... | 08:08 |
| hjensas | see it on change 839675 and change 839739 | 08:09 |
| hjensas | sorry, s/xena/yoga | 08:10 |
| dtantsur | ebbex: nope, this is specifically for image builder implementations | 09:02 |
| dtantsur | ironic-image itself does have an ability to set the default kernel params, but I don't remember if it's exposed in the installer | 09:03 |
| dtantsur | hjensas: we've fixed that on master, I think. | 09:03 |
| ebbex | dtantsur: ok, thanks :) | 09:05 |
| dtantsur | ebbex: what's your use case? debugging something? | 09:06 |
| rpittau | hjensas: https://opendev.org/openstack/metalsmith/commit/018a19c1e89d91d0468ca9a59ca4cf011a90a2ab | 09:21 |
| ebbex | dtantsur: yup, I figured out i could just modify the files under /var/lib/containers/storage/volumes/ironic/_data/ on the bootstrap node. | 09:46 |
| ebbex | the network drops out at some point during install. probably the ipa doing something with the network. | 09:48 |
| dtantsur | only LLDP comes to mind | 09:50 |
| ebbex | it's detecting/probing vlans on all individual interfaces that are part of a bonded interface. | 09:53 |
| ebbex | https://github.com/openshift/okd/discussions/1200#discussioncomment-2646895 if you scroll down you can see the interfaces before and after. | 09:54 |
| dtantsur | ah, yeah, there is such a feature, I think | 10:13 |
| opendevreview | Harald Jensås proposed openstack/metalsmith stable/yoga: Add a workaround to fix the CI https://review.opendev.org/c/openstack/metalsmith/+/839910 | 11:12 |
| hjensas | dtantsur: rpittau: thanks for pointer, I proposed a backport ^ | 11:13 |
| *** iurygregory__ is now known as iurygregory | 11:14 | |
| iurygregory | morning ironic | 11:14 |
| opendevreview | Derek Higgins proposed openstack/ironic master: Ensure enable_netboot_fallback writes out pxe config on adopt. https://review.opendev.org/c/openstack/ironic/+/811989 | 11:16 |
| dtantsur | okay folks, we need a good way to detect partition names on a device | 12:07 |
| dtantsur | <device>p<number> is not cutting it on devicemapper devices | 12:07 |
| rpittau | can we consider the TYPE of the device ? | 12:15 |
| rpittau | if it's mpath we name the partitions differenntly | 12:15 |
| iurygregory | hummm I think it should be possible... | 12:22 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: Collect a full lsblk output in the ramdisk logs https://review.opendev.org/c/openstack/ironic-python-agent/+/839939 | 12:24 |
| dtantsur | rpittau: absolutely, the question is how to detect these names.. | 12:25 |
| iurygregory | dtantsur, rpittau, I'm wondering if Gorka would have some info .. | 12:25 |
| dtantsur | maybe? | 12:25 |
| iurygregory | not sure about his irc upstream, but I see him in the dev channel downstream | 12:27 |
| janders | random idea - would major/minor number of the devnode be of use? | 12:28 |
| dtantsur | probably? | 12:29 |
| dtantsur | hmm, no: NAME="sdb1" KNAME="sdb1" MAJ:MIN="8:17" | 12:29 |
| janders | https://paste.openstack.org/show/bmCeNotbqaMiI9MMyIjo/ | 12:29 |
| janders | IIRC major will be same for all devices, minor will be different for each partition | 12:29 |
| janders | which the sample seems to confirm | 12:30 |
| dtantsur | yeah, but it completely breaks in the presence of several devices | 12:30 |
| dtantsur | NAME="sda" MAJ:MIN="8:0" | 12:31 |
| dtantsur | NAME="sda1" MAJ:MIN="8:1" | 12:31 |
| dtantsur | NAME="sda2" MAJ:MIN="8:2" | 12:31 |
| dtantsur | NAME="sdb" MAJ:MIN="8:16" | 12:31 |
| dtantsur | NAME="sdb1" MAJ:MIN="8:17" | 12:31 |
| janders | dammit | 12:31 |
| janders | I was hoping major will be different | 12:31 |
| janders | (was just looking for a machine with few drives to play with) | 12:31 |
| rpittau | you can get the tree of the partitions with lsblk based on the device | 12:31 |
| rpittau | e.g. lsblk --tree /dev/dm-0 | 12:32 |
| rpittau | if the dedvice is mpath we can probalby get a list of the partitions from there | 12:32 |
| iurygregory | hummm | 12:32 |
| dtantsur | rpittau: but how do you get which device is number 5, for example? can we rely on the ordering there? | 12:32 |
| dtantsur | (maybe we can.. dunno) | 12:32 |
| dtantsur | (--tree is not compatible with --pairs, but you can use PKNAME to do the same) | 12:33 |
| iurygregory | crystal ball required to identify the right device so far =( | 12:33 |
| janders | would blkid be any use in finding duplicates? | 12:34 |
| dtantsur | a nuclear way is to compare the partition lists before and after creating the configdrive | 12:38 |
| iurygregory | we would get Failed to create config drive on disk in prepare_image no? | 12:41 |
| iurygregory | or it would be 1 step back before this? | 12:42 |
| dtantsur | okay, this is interesting: sgdisk allows setting a partition UUID when creating | 12:42 |
| dtantsur | let me place with it locally | 12:42 |
| rpittau | AFAIK the numbers assigned to the dm devices are allocated in natural order during bootup, so for a first configuration it should work | 12:43 |
| dtantsur | wut, my lab machine is not GPT?? | 12:44 |
| dtantsur | anyway: sudo sgdisk -n 0:-64MB:0 -u 0:$uuid $device | 12:45 |
| dtantsur | I think this will allow us to set PARTUUID on creation. and not rely on numbers | 12:45 |
| dtantsur | now I need to understand if a similar thing can be done for MBR | 12:46 |
| dtantsur | ... of you can it can not *crying* | 12:47 |
| dtantsur | well, PARTUUID is not a thing for MBR, so I must be delirious at this point | 12:47 |
| dtantsur | ACTUALLY | 12:48 |
| iurygregory | <eyes> | 12:49 |
| dtantsur | why are we even using numbers? | 12:49 |
| iurygregory | numbers are overrated :D | 12:50 |
| iurygregory | what is your idea? | 12:50 |
| dtantsur | this part is bullshit https://opendev.org/openstack/ironic-python-agent/src/branch/master/ironic_python_agent/partition_utils.py#L429-L439 | 12:50 |
| dtantsur | we literally know which partition is new, why are we even trying to do partition_index_to_path | 12:50 |
| dtantsur | we can use the output of list_partitions we already have to get the name | 12:50 |
| dtantsur | hold on, I'll prepare a patch rather than talking | 12:51 |
| dtantsur | ahh, let me guess, list_partitions doesn't give us a name | 12:52 |
| iurygregory | :returns: list of dictionaries (one per partition) with keys: | 12:53 |
| iurygregory | number, start, end, size (in MiB), filesystem, partition_name, | 12:53 |
| iurygregory | flags, path. | 12:53 |
| dtantsur | or rather: it does, but by using the same broken partition_index_to_path logic | 12:53 |
| TheJulia | rpittau: I... fixed the unit test failure https://review.opendev.org/c/openstack/networking-generic-switch/+/839645 | 12:58 |
| TheJulia | mgoddard already put a +2 on it, another would be appreciated since the shell scripting fix *should* help make ngs in our gate multinode job happier | 12:58 |
| rpittau | TheJulia: thanks, +A :) | 12:59 |
| rpittau | didn't have time to go deeper on that.... | 12:59 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: [PoC] Use a pre-defined partition UUID to detect configdrive on GPT https://review.opendev.org/c/openstack/ironic-python-agent/+/839946 | 13:02 |
| dtantsur | I'm not sure how much it helps now ^^ but that's what I meant re PARTUUID | 13:02 |
| dtantsur | (could not test it locally, hopefully we have CI coverage for this case) | 13:03 |
| iurygregory | interesting | 13:06 |
| rpittau | dtantsur: I think you can just generate a random guid wihtout generate_uuid directly with sgdisk, but the rest looks ok | 13:07 |
| dtantsur | yeah. I have a feeling that partition numbers are MBR legacy, and we should stop relying on them for GPT | 13:08 |
| dtantsur | rpittau: if I do it with sgdisk, I won't know it :) | 13:08 |
| dtantsur | the point is to know the UUID in advance, so that I can use it for lookup | 13:08 |
| rpittau | mmm ok I see what you're doing later | 13:10 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-lib stable/wallaby: [PoC] Use a pre-defined partition UUID to detect configdrive on GPT https://review.opendev.org/c/openstack/ironic-lib/+/839949 | 13:38 |
| dtantsur | rpittau, iurygregory, ^^^ | 13:38 |
| iurygregory | dtantsur, ok o/ | 13:39 |
| dtantsur | it's basically the same code, just in a different location and without get_partition | 13:40 |
| iurygregory | this will make things funnier downstream... | 13:42 |
| dtantsur | even funnier than now? :) | 13:42 |
| iurygregory | yes | 13:42 |
| dtantsur | other ideas are welcome, but I'm afraid we'll have to patch ironic-lib anyway: the logic is there | 13:43 |
| iurygregory | yeah | 13:43 |
| iurygregory | just wondering how complicated will be the pipeline now :D | 13:44 |
| opendevreview | Harald Jensås proposed openstack/ironic stable/yoga: Exclude current conductor from offline_conductors https://review.opendev.org/c/openstack/ironic/+/839739 | 13:50 |
| hjensas | TheJulia: re https://review.opendev.org/c/openstack/networking-generic-switch/+/839645 , are we sure the CONF.host thing is not cause by https://review.opendev.org/c/openstack/neutron/+/837392 ? | 13:54 |
| TheJulia | for sure | 13:55 |
| TheJulia | that created it | 13:55 |
| * dtantsur really hopes the ironic-lib wallaby CI is working | 13:56 | |
| TheJulia | at the same time, I looked at the loading/interaction and it is definitely is primarily impacting with how the unit tests run | 13:57 |
| TheJulia | dtantsur: it did the last time I looked ~4ish months ago | 13:57 |
| dtantsur | an eternity :D | 13:57 |
| hjensas | TheJulia: k, I'm just worried the host option is never registered. But I see both tempest-dlm and the multinode job passed. So it is probably good. | 14:02 |
| TheJulia | dtantsur: potentially yes | 14:04 |
| TheJulia | hjensas: as a plugin, it gets executed in the neutron-server context, so my thought was it would just have access to it from that standpoint where as the unit tests don't have implicit access | 14:04 |
| TheJulia | hjensas: but yeah, the two tests passing are a good sign | 14:05 |
| hjensas | TheJulia: ah yes, you are right something else in neutron will load the opt. Makes sense. My worries have left me. :) | 14:06 |
| TheJulia | \o/ | 14:06 |
| TheJulia | success! | 14:06 |
| TheJulia | now... what does my brain have tiemf ro before my enxt meeting | 14:07 |
| TheJulia | doh, it is now | 14:07 |
| * TheJulia looks at her spam email and shakes her head | 14:11 | |
| admiyo | If I am using just Ironic (bifrost install) is there any way to keep any of the end users from hogging all the resources? I quota management a Nova only thing? | 14:15 |
| TheJulia | nope unfortunately not, but patches welcome! | 14:16 |
| TheJulia | with increasing support for non-admin direct usage, it makes sense to implement, just $time/braincells | 14:16 |
| admiyo | Harumph | 14:16 |
| TheJulia | *and* I need to add very distinct high performance db queries soon | 14:17 |
| admiyo | I would actually be prone to add a project id to an instance and say you can only provision nodes you own | 14:17 |
| TheJulia | owner/lessee fields | 14:17 |
| TheJulia | we do have those | 14:18 |
| admiyo | So I can set the owner field to a project Id, and then you need to be in that proejct to use the node? | 14:18 |
| TheJulia | and if the user is in that project they would be restricted to that | 14:18 |
| TheJulia | yes | 14:18 |
| admiyo | Poifect | 14:18 |
| TheJulia | lessee *can* also be granted access | 14:18 |
| TheJulia | but it is a far more restricted access set | 14:18 |
| admiyo | spectabulous | 14:18 |
| TheJulia | okay, happy admiyo \o/ | 14:18 |
| admiyo | I'm enchanted with bifrost | 14:18 |
| TheJulia | Built with whiskey and hypnotoad | 14:19 |
| admiyo | what kind of whiskey? | 14:19 |
| TheJulia | I was drinking mostly irish wiskeys when I was doing initial bifrost work | 14:19 |
| admiyo | A drop of the craether e'ry morn | 14:20 |
| TheJulia | more so late night :) | 14:29 |
| opendevreview | Ameya Raut proposed openstack/ironic-tempest-plugin master: Add iDRAC management cleaning steps tests https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/826646 | 14:53 |
| opendevreview | Merged openstack/networking-generic-switch master: CI: Fix Multinode ssh key/hostname for unit tests https://review.opendev.org/c/openstack/networking-generic-switch/+/839645 | 14:53 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: Use a pre-defined partition UUID to detect configdrive on GPT https://review.opendev.org/c/openstack/ironic-python-agent/+/839946 | 14:57 |
| dtantsur | finalised ^^^ | 14:57 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-lib stable/wallaby: DNM add an UEFI job https://review.opendev.org/c/openstack/ironic-lib/+/839955 | 15:00 |
| dtantsur | please also review https://review.opendev.org/c/openstack/ironic-python-agent/+/839939, it's going to simplify debugging such cases | 15:06 |
| TheJulia | arne_wiebalck: maybe check the survey poll next week, the foundation re-shared one of the posts with it today | 15:14 |
| arne_wiebalck | TheJulia: Ok, I will. | 15:21 |
| arne_wiebalck | TheJulia: Just did now: there are some additional responses, and the reinforce the trends we have seen before. | 15:22 |
| TheJulia | arne_wiebalck: okay, well we should wait until next week since they shared it like 30 minuts ago | 15:23 |
| TheJulia | minutes | 15:23 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: v6/grenade multinode jobs https://review.opendev.org/c/openstack/ironic/+/839086 | 15:27 |
| opendevreview | Riccardo Pittau proposed openstack/sushy-tools master: Use python Zed tests https://review.opendev.org/c/openstack/sushy-tools/+/838674 | 15:28 |
| * TheJulia crosses her fingers for the multinode tempest job | 15:32 | |
| arne_wiebalck | TheJulia: the tool will delete some answers after 30 days ... FWIU this means the survey will stay but lose answers over time | 15:57 |
| TheJulia | wonderful :( | 15:57 |
| rpittau | bye everyone, have a great weekend! o/ | 16:21 |
| opendevreview | Vishal Manchanda proposed openstack/ironic-ui master: Address RemovedInDjango40Warning https://review.opendev.org/c/openstack/ironic-ui/+/839971 | 16:31 |
| opendevreview | Merged openstack/metalsmith stable/yoga: Add a workaround to fix the CI https://review.opendev.org/c/openstack/metalsmith/+/839910 | 16:35 |
| TheJulia | Okay, Finally booked. | 17:08 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for iface vlan plugging https://review.opendev.org/c/openstack/networking-baremetal/+/835157 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for switch vlans https://review.opendev.org/c/openstack/networking-baremetal/+/835158 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for network-instance https://review.opendev.org/c/openstack/networking-baremetal/+/835159 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for interface aggregate https://review.opendev.org/c/openstack/networking-baremetal/+/835160 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for LACP https://review.opendev.org/c/openstack/networking-baremetal/+/835161 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Device management driver iface https://review.opendev.org/c/openstack/networking-baremetal/+/835170 | 17:20 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add netconf-openconfig device driver https://review.opendev.org/c/openstack/networking-baremetal/+/835324 | 17:20 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: Trivial: fix a typo in the devstack plugin https://review.opendev.org/c/openstack/ironic/+/839982 | 17:36 |
| dtantsur | I cannot believe this ^^^ | 17:36 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/yoga: Trivial: fix a typo in the devstack plugin https://review.opendev.org/c/openstack/ironic/+/839983 | 17:36 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/xena: Trivial: fix a typo in the devstack plugin https://review.opendev.org/c/openstack/ironic/+/839984 | 17:36 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/wallaby: Trivial: fix a typo in the devstack plugin https://review.opendev.org/c/openstack/ironic/+/839985 | 17:37 |
| admiyo | bifrost is not very idempotent is it? I keep finding things that don't work on a second run.... | 17:37 |
| dtantsur | admiyo: it's supposed to be, but there may be bugs | 17:37 |
| admiyo | for example, I have had to chown the /ops/stack subdirs | 17:37 |
| dtantsur | fwiw I run it very often on my lab machines | 17:37 |
| admiyo | right now I have a show stopper with mysql which seems strange | 17:37 |
| admiyo | [bifrost-keystone-install : MySQL - Creating DB] fails due to ans_network_interface]['ipv4']['address'] does not have an ipv4 attribute... | 17:38 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-lib stable/wallaby: DNM add an UEFI job https://review.opendev.org/c/openstack/ironic-lib/+/839955 | 17:38 |
| admiyo | I ran the install, ran the tls, and then started messing with Keystone, and well it went pear shaped | 17:39 |
| dtantsur | is it master? I built an environment with TLS and keystone 2-3 days ago | 17:39 |
| dtantsur | (on CentOS Stream 9 if it matters) | 17:39 |
| admiyo | Yeah, although aarch64 | 17:39 |
| admiyo | Ubuntu 20.04 | 17:39 |
| admiyo | is there a way to tear down? | 17:40 |
| dtantsur | there is no tear down support, no | 17:40 |
| admiyo | I suspect I need to wipe the MySQL part | 17:40 |
| dtantsur | I guess you did double-check that there is an ipv4 address there? | 17:41 |
| admiyo | https://paste.openstack.org/show/bzlsyOyd6OSAYvPQ8JTZ/ | 17:41 |
| admiyo | its localhost | 17:41 |
| admiyo | its in the anisble variables: {{ hostvars[inventory_hostname]['ansible_' + ans_network_interface]['ipv4']['address'] }}: | 17:42 |
| admiyo | that is the only thing that references ipv4 | 17:42 |
| dtantsur | try debug output for hostvars[inventory_hostname]['ansible_' + ans_network_interface] | 17:42 |
| dtantsur | I can look with you on Monday, it's a bit late here | 17:43 |
| rpioso | Good Friday morning, ironic :) | 17:47 |
| admiyo | THanks. I have debug running. I just needed to vent. If it is still an issue on Monday I'll ask, but I assume that if it was something you regularly tripped over you would have shouted by now. | 17:47 |
| dtantsur | Friday \o/ hi rpioso | 17:47 |
| rpioso | dtantsur: Hey! | 17:47 |
| admiyo | "msg": "The conditional check 'keystone.database.host == 'localhost'' failed. The error was: error while evaluating conditional (keystone.database.host == 'localhost'): | 17:48 |
| admiyo | but I even commented out that conditional and it complained. strange | 17:49 |
| rpioso | Folks, I would appreciate core reviews of https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/826646. It's been ironic-week-prio for a ... while. TheJulia gave it a -1 back in February. I believe her comments have been addressed. | 17:50 |
| * rpioso and his teammates are tempest newbies :) | 17:52 | |
| dtantsur | will try to get to it on Monday, but I have an active escalation, sooo.. | 17:53 |
| rpioso | dtantsur: That would be much appreciated. Sorry about the escalation :( | 17:53 |
| dtantsur | :) | 18:05 |
| dtantsur | have a good weekend folks o/ | 18:05 |
| NobodyCam | welcome alingonv | 19:10 |
| alingonv | My favorite webcam brand! | 19:10 |
| NobodyCam | ehehe | 19:10 |
| opendevreview | Julia Kreger proposed openstack/networking-generic-switch stable/yoga: CI: Fix Multinode ssh key/hostname for unit tests https://review.opendev.org/c/openstack/networking-generic-switch/+/839919 | 19:33 |
| opendevreview | Julia Kreger proposed openstack/networking-generic-switch stable/xena: CI: Fix Multinode ssh key/hostname for unit tests https://review.opendev.org/c/openstack/networking-generic-switch/+/839920 | 19:34 |
| * TheJulia has never thought about it as "webcam brand" | 19:34 | |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: v6/grenade multinode jobs https://review.opendev.org/c/openstack/ironic/+/839086 | 19:36 |
| opendevreview | Alexander Lingo proposed openstack/ironic master: Add SNMP drivers for Raritan, ServerTech, and Vertiv PDUs https://review.opendev.org/c/openstack/ironic/+/839995 | 19:39 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: Fix ironic-lib from source and branch detection in DIB builds https://review.opendev.org/c/openstack/ironic/+/839982 | 19:39 |
| JayF | I just realized I read NobodyCam as "No Body Cam" when it's probably supposed to be "Nobody Cam" | 19:39 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/yoga: Fix ironic-lib from source and branch detection in DIB builds https://review.opendev.org/c/openstack/ironic/+/839983 | 19:39 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/xena: Fix ironic-lib from source and branch detection in DIB builds https://review.opendev.org/c/openstack/ironic/+/839984 | 19:40 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/wallaby: Fix ironic-lib from source and branch detection in DIB builds https://review.opendev.org/c/openstack/ironic/+/839985 | 19:40 |
| NobodyCam | hehehe it a legacy type from 100 years ago that just stuck | 19:40 |
| NobodyCam | I was trying for nobodyCan | 19:40 |
| NobodyCam | ;p | 19:41 |
| alingonv | Hello. I've submitted a review that adds support for new PDU models (ServerTech, Raritan, and Vertiv) -- https://review.opendev.org/c/openstack/ironic/+/839995. Looking for comments and/or reviews. | 19:41 |
| TheJulia | I *think* I heard the whole irc nick story once over dinner | 19:41 |
| TheJulia | but, I don't remember | 19:41 |
| NobodyCam | +++ | 19:41 |
| TheJulia | alingonv: looking at it now actually! I suspect the pep8 CI job is going to have an opinion. Also, do we not have unit tests on the other snmp drivers? | 19:42 |
| * TheJulia has not looked at the snmp driver code in.... Well... When did I work at HPE? | 19:44 | |
| alingonv | It appears I will need to add some unit tests, then. | 19:45 |
| TheJulia | So generally, we look for tests to help prevent things inadvertently getting broken and doc updates where applicable. So in this case, I would imagine he snmp driver stuff is documented, in this case https://opendev.org/openstack/ironic/src/branch/master/doc/source/admin/drivers/snmp.rst which would just be adding to adding driver names it looks like | 19:52 |
| TheJulia | also, w/r/t testing, I woudln't go overboard | 19:52 |
| alingonv | I was planning on going underboard, so there's likely a reasonable middleboard to be found. | 19:53 |
| TheJulia | lol | 19:59 |
| TheJulia | good answer :) | 19:59 |
| alingonv | I have to update driver code, unit tests, and documentation?!? I knew I should have started with contributing my one-line fix for https://storyboard.openstack.org/#!/story/2004956 | 20:01 |
| JayF | Ironic is quicksand | 20:07 |
| JayF | there are no one-line changes here | 20:07 |
| JayF | only things that look deceptively easy ;) | 20:07 |
| JayF | Honestly, s/Ironic/OpenStack/ really :) | 20:07 |
| alingonv | Yeah, I've come to see that over the last few years ;) | 20:12 |
| TheJulia | alingonv: https://zuul.opendev.org/t/openstack/status#839995 may be a useful link btw :) | 20:16 |
| TheJulia | JayF: don't scare the new contributors! Little do they know that we have tractor beams to prevent escape! | 20:17 |
| TheJulia | muahahahahahaha | 20:17 |
| JayF | If you don't want me to scare new contributors, release the tractor beam from my ship /s | 20:17 |
| TheJulia | alingonv: seriously though, It is a lot, you've got some great info in there in notes, but some cleanup and maybe somethinking about logging a little. I dunno... I'm also conflicted on that, personally. | 20:18 |
| TheJulia | JayF: well, to be fair, that is a shuttle craft dispatched from your own ship | 20:18 |
| TheJulia | :) | 20:19 |
| * TheJulia wonders, would we be sued if we put pixie boots inside of a spacedock | 20:21 | |
| JayF | pixie magnetic boots | 20:21 |
| TheJulia | Oh, those would need to be expanse style | 20:21 |
| * TheJulia happens to be wearing an MCRN t-shirt right now | 20:22 | |
| TheJulia | pixie boots in power armor? | 20:22 |
| * TheJulia needs to find an artist | 20:24 | |
| alingonv | I'm open to any suggestions that gets the code better, and more importantly, upstreamed and out of my 'weird internal hacks' repo. | 20:29 |
| TheJulia | ++ | 20:30 |
| TheJulia | Well, we're pretty good about helping with that | 20:30 |
| * TheJulia glares at CI hoping multinode grenade will live again | 20:30 | |
| JayF | I'd venture to say 30%+ of Ironic contribution is folks doing exactly that alingonv | 20:30 |
| TheJulia | or attempting to get $thing upstream first before it has to appear in a downstream repo | 20:31 |
| TheJulia | which reminds me.... I wanted to do the "concurrency" check stuffs | 20:31 |
| TheJulia | oh 4 ci "baremetal" instances on the subnode, why you no clean | 20:32 |
| TheJulia | (networking, rhetorical question of TheJulia's present insanity | 20:32 |
| TheJulia | ) | 20:32 |
| TheJulia | I wrote sqlalchemy query code which worked on the first try... | 21:05 |
| * TheJulia backs away from the comptuer | 21:05 | |
| JayF | No you didn't; you're just not far enough in the future to see how it's broken /s | 21:11 |
| JayF | I could've also gone for "I know people who would say all sqlalchemy code is broken" :P | 21:12 |
| TheJulia | JayF: I had to break the unit test on purpose to validate it was working | 21:12 |
| TheJulia | test passed first try | 21:12 |
| TheJulia | I was like "whiskey tango!" | 21:12 |
| JayF | I'm lucky if the test /compiles/ the first time | 21:12 |
| TheJulia | JayF: this too, is true | 21:12 |
| JayF | lol | 21:12 |
| TheJulia | lol | 21:12 |
| * TheJulia tries to make it more flexible and powerful without raw sql | 21:32 | |
| opendevreview | Merged openstack/networking-generic-switch stable/yoga: CI: Fix Multinode ssh key/hostname for unit tests https://review.opendev.org/c/openstack/networking-generic-switch/+/839919 | 22:18 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Trivial: Remove prints in db api layer code https://review.opendev.org/c/openstack/ironic/+/840005 | 23:49 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Concurrent Distructive/Intensive ops limits https://review.opendev.org/c/openstack/ironic/+/840006 | 23:49 |
| TheJulia | arne_wiebalck: https://review.opendev.org/c/openstack/ironic/+/840006 | 23:49 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!