| opendevreview | Verification of a change to openstack/ironic bugfix/18.1 failed: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/812310 | 00:04 |
|---|---|---|
| stevebaker | I'm back | 00:14 |
| TheJulia | welcome back | 00:28 |
| opendevreview | Verification of a change to openstack/ironic stable/wallaby failed: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/809552 | 00:52 |
| arne_wiebalck | Good morning, Ironic! | 05:23 |
| iurygregory | good morning arne_wiebalck and Ironic o/ | 05:54 |
| arne_wiebalck | hey iurygregory o/ | 05:54 |
| opendevreview | Verification of a change to openstack/ironic stable/wallaby failed: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/809552 | 06:19 |
| opendevreview | Verification of a change to openstack/ironic bugfix/18.1 failed: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/812310 | 06:19 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic stable/xena: [iRMC] Avoid repeatedly resuming clean after creating raid configuration https://review.opendev.org/c/openstack/ironic/+/812431 | 06:53 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic stable/xena: [iRMC] Set polling after RAID is built https://review.opendev.org/c/openstack/ironic/+/812432 | 06:55 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic bugfix/18.1: [iRMC] Set polling after RAID is built https://review.opendev.org/c/openstack/ironic/+/812433 | 06:56 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic bugfix/18.1: [iRMC] Avoid repeatedly resuming clean after creating raid configuration https://review.opendev.org/c/openstack/ironic/+/812434 | 06:57 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic bugfix/18.1: [iRMC] Set polling after RAID is built https://review.opendev.org/c/openstack/ironic/+/812433 | 06:57 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/bifrost stable/xena: Change git_branch in stable/xena https://review.opendev.org/c/openstack/bifrost/+/810817 | 06:59 |
| rpittau | good morning ironic! o/ | 07:36 |
| iurygregory | good morning rpittau o/ | 07:36 |
| rpittau | hey iurygregory :) | 07:36 |
| opendevreview | Merged openstack/ironic-inspector stable/wallaby: Ignored error state cache for new requests https://review.opendev.org/c/openstack/ironic-inspector/+/812374 | 07:38 |
| opendevreview | Merged openstack/ironic-inspector stable/victoria: Ignored error state cache for new requests https://review.opendev.org/c/openstack/ironic-inspector/+/812375 | 07:38 |
| opendevreview | Merged openstack/ironic-inspector stable/ussuri: Ignored error state cache for new requests https://review.opendev.org/c/openstack/ironic-inspector/+/812376 | 07:38 |
| opendevreview | Merged openstack/ironic-inspector stable/train: Ignored error state cache for new requests https://review.opendev.org/c/openstack/ironic-inspector/+/812377 | 07:38 |
| dtantsur | good morning | 07:55 |
| rpittau | morning dtantsur :) | 07:55 |
| iurygregory | morning dtantsur =) | 07:59 |
| opendevreview | Harald Jensås proposed openstack/metalsmith stable/xena: Write network_data.json metadata to config-dirve https://review.opendev.org/c/openstack/metalsmith/+/810480 | 08:16 |
| opendevreview | Harald Jensås proposed openstack/metalsmith stable/wallaby: Write network_data.json metadata to config-dirve https://review.opendev.org/c/openstack/metalsmith/+/810481 | 08:16 |
| arne_wiebalck | jcosmao: your udev settle patch works for me as well: I just set up a s/w RAIDs on NVMe drives on 88 servers and they all succeeded (compared to 50% failure rate before) | 09:23 |
| jcosmao | great ! | 09:31 |
| arne_wiebalck | jcosmao: I will submit a corresponding patch ... you have not done this yes, have you? | 09:32 |
| jcosmao | i didn't because i was on an old IPA version with some custom raid function (previous work done by raphael glon) | 09:34 |
| jcosmao | i can submit patch if you want | 09:34 |
| opendevreview | Merged openstack/ironic bugfix/18.1: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/812310 | 09:36 |
| opendevreview | Merged openstack/ironic stable/wallaby: Redfish RAID: Use identity instead of durable_name https://review.opendev.org/c/openstack/ironic/+/809552 | 09:36 |
| opendevreview | Arne Wiebalck proposed openstack/ironic-python-agent master: Software RAID: Call udev_settle before creation https://review.opendev.org/c/openstack/ironic-python-agent/+/812470 | 09:38 |
| arne_wiebalck | jcosmao: ^^ | 09:38 |
| arne_wiebalck | jcosmao: we are on Victoria, and I think this patch will need to be backported ... I should add a reno I realise now. | 09:39 |
| rpittau | arne_wiebalck: I was going to add a comment about the reno :) | 09:39 |
| * arne_wiebalck is faster than rpittau's shadow | 09:40 | |
| opendevreview | Arne Wiebalck proposed openstack/ironic-python-agent master: Software RAID: Call udev_settle before creation https://review.opendev.org/c/openstack/ironic-python-agent/+/812470 | 09:43 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic bugfix/18.1: Disable Neutron firewall https://review.opendev.org/c/openstack/ironic/+/812437 | 09:47 |
| jssfr | arne_wiebalck, re CentOS and udev and mdadm: we encountered that while building images for ironic. Switching the CI runner from CentOS 7 to 8 helped, before that the run would not recognize the partitions quickly enough even on the loopback device. | 10:00 |
| jssfr | thanks for the hint anyway, I saw above you were discussing patches, so we know where to look should that hit us again with CentOS8 | 10:00 |
| arne_wiebalck | jssfr: thanks for the follow-up! There might be other places where we hit the same issue. | 10:07 |
| opendevreview | Merged openstack/bifrost stable/xena: Change git_branch in stable/xena https://review.opendev.org/c/openstack/bifrost/+/810817 | 11:02 |
| janders_ | see you tomorrow Ironic o/ | 12:11 |
| dtantsur | folks, I'd be interested to hear your opinion about https://storyboard.openstack.org/#!/story/2009203 | 12:15 |
| dtantsur | especially ajya and TheJulia ^^^ | 12:16 |
| ajya | dtantsur: looks good, I was thinking about something like that at some point - that there is 'queue' with nodes in async processing which this proposal achieves by storing that into the new field. And I'm all in for removing boilerplate code as much as possible. | 12:26 |
| dtantsur | cool! I'll start looking in this direction | 12:26 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: WIP add a helper for node-based periodics https://review.opendev.org/c/openstack/ironic/+/812495 | 12:36 |
| iurygregory | dtantsur, I like the proposal ++ | 12:55 |
| * TheJulia makes coffeeeeeeee | 13:15 | |
| dtantsur | good morning TheJulia | 13:27 |
| iurygregory | good morning TheJulia | 13:34 |
| TheJulia | dtantsur: so, I think a new field for this on nodes is not a great idea, in part because every new field is a chunk of extra data we need to carry/convert/handle/provide to api consumers as part of any detailed list or single node get... I think a separate table since there seems to be more quantitative information that could also be stored that is transient for the node. A separate table could be beneficial in that we | 13:35 |
| TheJulia | could ultimately stack items or convey to a conductor that *you really* need to execute $other_thing as soon as the current task exits which is totally unrelated. Although, that means we're basically building a queuing system for "follow up" | 13:35 |
| dtantsur | TheJulia: if we think this field is not valuable to consumers, we don't have to return it in the API | 13:35 |
| dtantsur | I don't understand how a separate table can make it any better | 13:35 |
| dtantsur | (and a queueing system is vastly out of scope of this problem) | 13:36 |
| TheJulia | dtantsur: but we still have to handle it on the intermediate layers db object -> ironic object -> json dict -> api end user | 13:36 |
| TheJulia | dtantsur: even ignoring the queing system idea, a "get /v1/jobs" could provide insight into what is going on maybe, and that is something people have kind of expressed interest in | 13:37 |
| dtantsur | I don't necessarily object to what you're proposing (although we closed a related RFE just a week ago), but it has barely anything to do with my intention to reduce copy-paste in vendor drivers while reducing the number of database rows to fetch | 13:38 |
| dtantsur | all I need is something to filter on the database level rather than checking driver_internal_info/task.driver afterwards | 13:38 |
| TheJulia | That is fair, and I suspect it would definitely fit that bill | 13:39 |
| dtantsur | the context is your comment on https://review.opendev.org/c/openstack/ironic/+/807295 | 13:39 |
| TheJulia | i know, I'm just thinking the next logical step. Overall I like it | 13:40 |
| TheJulia | dtantsur: how would we navigate upgrades, because obviously with online upgrades we would have to support both patterns unless we dropped field/object model far in advance (or ya know... finally implement an internal self running online upgrade worker/mechanism | 13:42 |
| dtantsur | ugh | 13:42 |
| TheJulia | db/object could get queued up and we could drop it in far in advance and then retool the periodics | 13:43 |
| TheJulia | but that also elongates it unfortunately :( | 13:44 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: WIP add a helper for node-based periodics https://review.opendev.org/c/openstack/ironic/+/812495 | 13:44 |
| dtantsur | okay, I'll think about it. meanwhile ^^^ will at least reduce copy-paste (once finished) | 13:44 |
| * TheJulia is surprised moltenironis still used to this day | 13:46 | |
| dtantsur | and quite actively apparently | 14:12 |
| iurygregory | I'm wondering if we need to at least update the unit-tests template in molteniron... | 14:14 |
| iurygregory | we are still using ussuri :D | 14:14 |
| rpittau | iurygregory: yeah there are some other things that need to be updated, I was looking at it after the answers to TheJulia's mail :) | 14:31 |
| rpittau | iurygregory: I can take care of that, unless you (or someone else) want it | 14:31 |
| iurygregory | rpittau, up to you =) if you want to do I don't mind | 14:32 |
| rpittau | iurygregory: yup, I'll take care of it | 14:33 |
| iurygregory | rpittau, ack =) | 14:34 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove debian packages file for devstack https://review.opendev.org/c/openstack/ironic/+/810314 | 14:36 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove debian packages file for devstack https://review.opendev.org/c/openstack/ironic/+/810314 | 14:39 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove devstack bindep packages for building disk images https://review.opendev.org/c/openstack/ironic/+/812513 | 14:39 |
| arne_wiebalck | TheJulia: mail and tweet are sent out now. | 14:51 |
| arne_wiebalck | TheJulia: for the SIG meetings next week ... (context!) | 14:51 |
| TheJulia | arne_wiebalck: awesome | 14:52 |
| iurygregory | tks arne_wiebalck | 14:53 |
| iurygregory | :D | 14:53 |
| arne_wiebalck | iurygregory: :) | 14:53 |
| *** dtantsur_ is now known as dtantsur | 15:10 | |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove debian packages file for devstack https://review.opendev.org/c/openstack/ironic/+/810314 | 15:22 |
| rpittau | molteniron patches are not reported in the channel | 15:37 |
| iurygregory | funny :D | 15:37 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: WIP add a helper for node-based periodics https://review.opendev.org/c/openstack/ironic/+/812495 | 16:01 |
| arne_wiebalck | bye everyone o/ | 16:07 |
| rpittau | molteniron really needs some hugs :/ | 16:26 |
| rpittau | if anyone has a moment -> https://review.opendev.org/c/openstack/molteniron/+/812520 | 16:26 |
| rpittau | see ya tomorrow! bye! o/ | 16:26 |
| dtantsur | I think we should ask people who use molteniron to help maintain it | 16:35 |
| dtantsur | going as well, see you | 16:35 |
| dtantsur | o/ | 16:35 |
| TheJulia | rpittau: it is not a shipped/released repository, so I kind of concur with dtantsur. if we have a standard we want it at, they need to help maintain it. | 16:43 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove debian packages file for devstack https://review.opendev.org/c/openstack/ironic/+/810314 | 17:05 |
| janders_ | good morning Ironic o/ | 19:27 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Make iDRAC management steps verify steps https://review.opendev.org/c/openstack/ironic/+/804032 | 19:31 |
| iurygregory | morning janders_ o/ | 19:33 |
| ayoung | OK, I have a couple of newly repaired nodes stuck in the "error" provisioning state. How do I kick them into an acceptable state? | 19:34 |
| janders_ | hey iurygregory o/ | 19:34 |
| *** janders_ is now known as janders | 19:34 | |
| ayoung | The requested action "abort" can not be performed on node "2effcab6-dad0-4d55-b6a9-57faf2d8edd2" while it is in state "error". (HTTP 400) | 19:37 |
| iurygregory | ayoung, you can try to trigger rebuild I would say | 19:37 |
| iurygregory | this would be the only way based on the state machine we have in ironic | 19:37 |
| ayoung | what command is that? | 19:38 |
| ayoung | Like delete and reimport? | 19:39 |
| iurygregory | if I recall you can do via API using PUT /v1/nodes/{node_ident}/states/provision | 19:40 |
| ayoung | $ openstack baremetal node rebuild jade12-r097 | 19:40 |
| ayoung | The provisioning operation can't be performed on node 2effcab6-dad0-4d55-b6a9-57faf2d8edd2 because it's in maintenance mode. (HTTP 400) | 19:40 |
| ayoung | OK that is good... | 19:40 |
| ayoung | ailed to validate deploy or power info for node 2effcab6-dad0-4d55-b6a9-57faf2d8edd2. Error: Cannot validate image information for node 2effcab6-dad0-4d55-b6a9-57faf2d8edd2 because one or more parameters are missing from its instance_info and insufficent information is present to boot from a remote volume. Missing are: ['image_source', 'kernel', 'ramdisk'] (HTTP 400) | 19:40 |
| iurygregory | at least this would be the way based on the https://docs.openstack.org/ironic/latest/_images/states.svg | 19:40 |
| iurygregory | yeah, you probably need to set a few information on the node if they are not present .-. | 19:41 |
| ayoung | those look like instance values, not baremetal node values | 19:42 |
| ayoung | RESP: [406] Content-Length: 217 Content-Type: application/json Date: Tue, 05 Oct 2021 19:44:03 GMT X-Openstack-Ironic-Api-Maximum-Version: 1.58 X-Openstack-Ironic-Api-Minimum-Version: 1.1 | 19:44 |
| ayoung | RESP BODY: {"error_message": "{\"code\": 406, \"title\": \"Not Acceptable\", \"description\": \"Version 1.67 was requested but the minor version is not supported by this service. The supported version range is: [1.1, 1.58].\"}"} | 19:44 |
| ayoung | So I am running too old a version of Ironic to do that? | 19:44 |
| * ayoung promises to upgrade as soon as I can do so safely | 19:45 | |
| iurygregory | what version are you running? O.o | 19:45 |
| ayoung | I am not certain. It was installed when I joined the company, but that was less than a month ago | 19:46 |
| iurygregory | 1.58 I think is train | 19:47 |
| ayoung | So,basically, I need to delete and reimport the node? | 19:49 |
| iurygregory | well, delete the node would be the last thing you should try, are you running ironic standalone or with nova? | 19:50 |
| ayoung | with nova | 19:50 |
| iurygregory | oh god XD | 19:50 |
| ayoung | It is all legacy to me | 19:51 |
| iurygregory | you would need to delete the instance associated with the node also | 19:53 |
| iurygregory | and make sure there is no orphan vifs etc | 19:53 |
| iurygregory | otherwise you will be unable to create the node again (you will probably see some interesting errors) | 19:54 |
| ayoung | node has no instance on it | 19:55 |
| iurygregory | in this case I think you can delete without problems | 19:57 |
| ayoung | Error contacting Ironic server: Can not delete node "43123529-e839-40ce-95f8-38ba7485667c" while it is in provision state "error". Valid provision states to perform deletion are: "('manageable', 'enroll', 'adopt failed', 'available')", or set the node into maintenance mode (HTTP 409). Attempt 6 of 6 | 19:57 |
| ayoung | ok put in maint mode | 19:58 |
| ayoung | deleted | 19:58 |
| iurygregory | \o/ | 19:58 |
| ade_lee | TheJulia, stevebaker hey ya'll | 20:06 |
| ade_lee | sorry for the delay -- meetings meetings .. | 20:06 |
| ade_lee | TheJulia, stevebaker so how would you like me to change the fips jobs so that they can be merged? | 20:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Remove debian packages file for devstack https://review.opendev.org/c/openstack/ironic/+/810314 | 20:45 |
| TheJulia | ade_lee: so, I'd prefer to swap the base OS on on two existing scenario jobs | 21:44 |
| ade_lee | TheJulia, sure - we can do that -- which jobs? | 21:46 |
| TheJulia | ade_lee: well, you duped an integration job and a functional test job. The functional test job I wouldn't bother with as it is more api contract operation Integration scenario with the log name would be the thing to change where we run one of the tempest scenario tests | 21:48 |
| ade_lee | TheJulia, ironic-tempest-functional-python3? | 21:48 |
| ade_lee | TheJulia, ok -- so not bother with the functional job and change the baseOS on ironic-inspector-tempest ? | 21:50 |
| TheJulia | so ironic-inspector-tempest is on the ironic-inspector repo | 21:54 |
| ade_lee | TheJulia, I was just looking at the jobs we added here -- https://review.opendev.org/c/openstack/ironic/+/797739/18/zuul.d/ironic-jobs.yaml | 21:55 |
| TheJulia | on ironic, ironic-standalone-something and ironic-tempest-ipa-something | 21:55 |
| ade_lee | ok - lookslike there is an ironic-standalone-redfish | 21:56 |
| ade_lee | and ironic-standalone | 21:57 |
| TheJulia | ironic-standalone-redfish uses https, ironic-standalone uses ipmi | 21:57 |
| ade_lee | ack - we want to change both then? | 21:57 |
| ade_lee | or just one? | 21:57 |
| ade_lee | and on the ironic-tempest-ipa-something - which one(s) would you prefer? | 21:59 |
| TheJulia | ade_lee: I'd prefer only one of the two standalone jobs since they run identical tests, it is just transport to a bmc that is going to impact any fips restrictions. The ironic-tempest one... no preference. | 23:03 |
| TheJulia | from my point of view that is | 23:03 |
| janders | @TheJulia - I left a comment/reply to your review in https://review.opendev.org/c/openstack/ironic/+/812437. If my understanding of what is happening is correct, it may be worthwhile merging this patch. | 23:47 |
| janders | TheJulia: ^ (sorry, slack overdose hence the @ :) ) | 23:48 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
