| arne_wiebalck | Good morning, Ironic! | 07:05 |
|---|---|---|
| hjensas | Good morning, Ironic. | 07:09 |
| hjensas | Any diskimage-builder dhcp-all-interfaces element users, can I ask for reviews on https://review.opendev.org/c/openstack/diskimage-builder/+/825983 | 07:10 |
| rpittau | good morning ironic! o/ | 08:02 |
| rpittau | dib fix has merged \o/ please have a look at https://review.opendev.org/c/openstack/bifrost/+/827293 when you have a minute | 08:41 |
| dtantsur | morning folks | 09:30 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/827882 | 09:34 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: set the tempest plugin's partition_netboot option https://review.opendev.org/c/openstack/ironic/+/827481 | 09:37 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/827882 | 10:26 |
| iurygregory | good morning Ironic o/ | 11:26 |
| MahnoorAsghar | o/ | 11:30 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/827882 | 11:42 |
| opendevreview | Merged openstack/ironic master: Set correct initrd_filename for iPXE when using Swift https://review.opendev.org/c/openstack/ironic/+/828770 | 12:00 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/827882 | 12:59 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/xena: Set correct initrd_filename for iPXE when using Swift https://review.opendev.org/c/openstack/ironic/+/829186 | 12:59 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic bugfix/19.0: Set correct initrd_filename for iPXE when using Swift https://review.opendev.org/c/openstack/ironic/+/829187 | 12:59 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic stable/wallaby: Set correct initrd_filename for iPXE when using Swift https://review.opendev.org/c/openstack/ironic/+/829188 | 13:00 |
| dtantsur | iurygregory: not sure if you've already requested a bifros release, but maybe we should add https://review.opendev.org/c/openstack/bifrost/+/826307 to finish with breaking changes | 13:50 |
| dtantsur | ideally https://review.opendev.org/c/openstack/bifrost/+/826465 too, it's trivial | 13:50 |
| TheJulia | good morning | 13:58 |
| MahnoorAsghar | Good morning | 13:58 |
| TheJulia | dtantsur: I just realized, that change won't be backwards compatible... I guess a release note item is needed | 13:59 |
| opendevreview | Dmitry Tantsur proposed openstack/bifrost master: Continue decomposing the "How to" documentation https://review.opendev.org/c/openstack/bifrost/+/829245 | 14:02 |
| dtantsur | TheJulia: you mean, the Notify change? the first one has a note. | 14:03 |
| dtantsur | TheJulia: could you elaborate why you think it won't work and before which ironic change? | 14:04 |
| dtantsur | oslo.service has been using sd_notify since forever | 14:04 |
| TheJulia | Then why were we dependent upon the ironic change? | 14:04 |
| dtantsur | and bifrost updates ironic on each installation, so I'm really puzzled | 14:04 |
| dtantsur | TheJulia: to be able to test it | 14:04 |
| TheJulia | and before hand it was still getting called by oslo.service? | 14:04 |
| dtantsur | TheJulia: yep. just too late to be useful for error detection | 14:05 |
| TheJulia | ahh! | 14:05 |
| TheJulia | okay | 14:05 |
| TheJulia | I misunderstood | 14:05 |
| TheJulia | I'm still waking up | 14:05 |
| dtantsur | ehhmm, I mean too early. before ironic actually finishes starting. | 14:05 |
| dtantsur | no worries ;) | 14:05 |
| TheJulia | review updatecd | 14:05 |
| TheJulia | updated | 14:05 |
| dtantsur | thx! and good morning | 14:06 |
| iurygregory | dtantsur, ack I haven't pushed yet let me review =) | 14:06 |
| iurygregory | good morning TheJulia =) | 14:07 |
| opendevreview | Arne Wiebalck proposed openstack/ironic master: [doc][trivial] Fixing typos in RBAC documentation https://review.opendev.org/c/openstack/ironic/+/829249 | 14:10 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: use a custom cirros partition image instead of the default https://review.opendev.org/c/openstack/ironic/+/827882 | 14:13 |
| dtantsur | I think this ^^ is getting somewhere finally | 14:13 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: CI: set the tempest plugin's partition_netboot option https://review.opendev.org/c/openstack/ironic/+/827481 | 14:13 |
| opendevreview | Arne Wiebalck proposed openstack/ironic master: [doc][trivial] Fixing typos in RBAC documentation https://review.opendev.org/c/openstack/ironic/+/829249 | 14:14 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: [DNM] Testing the CI https://review.opendev.org/c/openstack/ironic/+/827500 | 14:14 |
| opendevreview | Harald Jensås proposed openstack/ironic-lib master: Add platform:base-py[2|3] to bindep https://review.opendev.org/c/openstack/ironic-lib/+/829272 | 14:23 |
| iurygregory | dtantsur, re https://review.opendev.org/c/openstack/bifrost/+/826307 I will request a major release for bifrost, wdyt? | 14:35 |
| dtantsur | I think there are many changes that justify it, yes | 14:35 |
| iurygregory | cool =) | 14:35 |
| * iurygregory will wait for all CI runs =) | 14:35 | |
| opendevreview | Riccardo Pittau proposed openstack/bifrost master: Fix dib ipa jobs https://review.opendev.org/c/openstack/bifrost/+/827293 | 14:49 |
| TheJulia | my brain doesn't want to wake up it seems | 14:50 |
| iurygregory | my brain was sleeping when I thought that 19.1 was good for Ironic release | 15:11 |
| opendevreview | Iury Gregory Melo Ferreira proposed openstack/ironic master: Fix release mapping (19.1 -> 20.0) https://review.opendev.org/c/openstack/ironic/+/829308 | 15:14 |
| opendevreview | Merged openstack/bifrost master: Use Type=notify in systemd units for services https://review.opendev.org/c/openstack/bifrost/+/826465 | 15:19 |
| opendevreview | Arne Wiebalck proposed openstack/ironic master: [doc][trivial] Fixing typos in RBAC documentation https://review.opendev.org/c/openstack/ironic/+/829249 | 15:20 |
| opendevreview | Verification of a change to openstack/bifrost master failed: Stop using sub-sections of driver_info https://review.opendev.org/c/openstack/bifrost/+/826307 | 15:32 |
| arne_wiebalck | TheJulia: brief RBAC question if you have a moment: what is the reason the `list` policies are done in code rather than in the policy definition? What I mean is: the policy is `"baremetal:node:list": "role:reader` while effectively, i.e. by code, the rule is `"baremetal:node:list": "(role:reader and scope:system) or (role:reader and project_id:%(node.owner)s`, no? | 15:48 |
| opendevreview | Mahnoor Asghar proposed openstack/ironic master: [WIP] Create API documentation from docstrings https://review.opendev.org/c/openstack/ironic/+/827200 | 16:04 |
| TheJulia | arne_wiebalck: hmm | 16:14 |
| TheJulia | arne_wiebalck: because effectively anyone can ask for a list, the query if they are not system scope ends up being project based | 16:15 |
| TheJulia | and if project based, then nodes are either invisible or not based upon the account used | 16:16 |
| TheJulia | but also, we can't evaluate project_id matching until we're way down inside the request handling | 16:17 |
| TheJulia | because it is a filtered view | 16:18 |
| TheJulia | arne_wiebalck: I guess the tl;dr is that is how it worked based upon the spec | 16:20 |
| arne_wiebalck | I was wondering if there was a technical reason since the rule is short but requires some implicit knowledge | 16:22 |
| arne_wiebalck | but I guess list may be a special case | 16:22 |
| TheJulia | Yeah, initial list *is* a super special case | 16:23 |
| arne_wiebalck | since there is nothing upfront to check the project id / owner against | 16:23 |
| arne_wiebalck | ok, I was just wondering :) | 16:23 |
| arne_wiebalck | we're moving to RBAC and are adapting/transferring our rules ... | 16:24 |
| TheJulia | useful to override if you just want to lock all ironic down to one tenant | 16:24 |
| TheJulia | or system... | 16:25 |
| arne_wiebalck | yeah ... our nova is not system scope atm | 16:25 |
| arne_wiebalck | and I am kind of hesistant to change this | 16:25 |
| arne_wiebalck | anyway ... I can report once we have new rules on our experience :) | 16:26 |
| arne_wiebalck | *I can report on our experience once we have new rules | 16:26 |
| arne_wiebalck | (better to read) | 16:26 |
| TheJulia | most operators run with configurations which are functionally helping enable the original bug the entire effort seeks to fix | 16:28 |
| TheJulia | so migration is... obviously... tricky | 16:28 |
| TheJulia | which has also started to become this expectation of operation too, which is... depressing to say the least. | 16:30 |
| arne_wiebalck | for operators I think maintaining a comprehensible rule set is an important aspect | 16:33 |
| arne_wiebalck | if the default rules match what you do: perfect | 16:34 |
| arne_wiebalck | if not, try to minimizing the delta, keeping it comprehensible, and do the transition in a low-risk way is not obvious to balance | 16:35 |
| arne_wiebalck | *trying to minimize | 16:35 |
| rpittau | good night! o/ | 17:02 |
| dtantsur | hey-hey folks, I've got you a working cirros partition image https://review.opendev.org/c/openstack/ironic/+/827882 | 17:43 |
| dtantsur | (only on UEFI, as explained yesterday) | 17:43 |
| iurygregory | YAY | 17:48 |
| iurygregory | time to open a beer and celebrate \o/ | 17:49 |
| dtantsur | and review ;) | 17:49 |
| iurygregory | before the beer =P | 17:49 |
| dtantsur | ideally :D | 17:49 |
| dtantsur | I think my script to build the image may increase your desire for beer | 17:49 |
| dtantsur | folks, especially TheJulia, wdyt about https://storyboard.openstack.org/#!/story/2007664 ? we may submit it for outreachy | 18:20 |
| TheJulia | dtantsur: i like it | 18:21 |
| dtantsur | thx! | 18:24 |
| dtantsur | back to the CI topic: https://review.opendev.org/c/openstack/ironic/+/827481 and https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/827340 together finally make the jobs test what they should be testing | 18:24 |
| arne_wiebalck | bye everyone, see you on Friday o/ | 18:26 |
| iurygregory | https://review.opendev.org/c/openstack/ironic/+/829308 if anyone can do a quick review it would be good :) | 18:34 |
| dtantsur | iurygregory: I wonder which changes you consider to be breaking? | 18:41 |
| iurygregory | https://zuul.opendev.org/t/openstack/build/e5174d1d8d68491aa01d9a2e3e020a4b/log/tox/list-changes-results.log#1932-1943 | 18:50 |
| dtantsur | iurygregory: the 2nd is definitely just a bug fix | 18:50 |
| iurygregory | specially the first one I would say | 18:50 |
| dtantsur | questionable, but I can live with it | 18:51 |
| iurygregory | yeah, the second I was considering a bugfix =) | 18:51 |
| dtantsur | I'm not sure how useful it is to signal that we no longer support ancient IPA | 18:51 |
| iurygregory | Let us see what others think about it | 18:52 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Create random character devices so systemd does not crash https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/829391 | 18:53 |
| TheJulia | dtantsur: ^^ ouch | 18:53 |
| dtantsur | O__o | 18:54 |
| iurygregory | woot | 18:55 |
| TheJulia | manually created the devices, repacked the ramdisk, the stupid thing booted just fine | 18:56 |
| TheJulia | not in fips mode, boots just fine too | 18:56 |
| dtantsur | TheJulia: https://github.com/systemd/systemd/issues/4167 doesn't seem immediately related, it's about lack of entropy, not the devices being present | 18:56 |
| dtantsur | I wonder what the difference in FIPS mode is. /dev/urandom too insecure? | 18:57 |
| TheJulia | dtantsur: the discussion shifted, but that is the fundimental issue | 18:57 |
| dtantsur | TheJulia: I can't quite agree. "systemd: uninitialized urandom read" is just a warning, I see it in my cirros testing. | 18:57 |
| TheJulia | nah, the kernel basically gives you what would be /dev/random if you ask for urandom | 18:58 |
| TheJulia | dtantsur: on rhel, the machine quite literally crashes hard, I've got someone opening a bz against rhel | 18:58 |
| TheJulia | and found the same basic issue reported against rhel7.4 too :( | 18:58 |
| dtantsur | TheJulia: I totally believe you, just saying that the linked but is about warnings, not crashing | 18:59 |
| dtantsur | so maybe linking to it is confusing | 18:59 |
| TheJulia | perhaps | 18:59 |
| TheJulia | I need to lay down | 18:59 |
| TheJulia | the wind storm is not getting any better, my migraine is getting worse, and the meds are making my stomach hate me | 18:59 |
| dtantsur | ouch :( | 19:00 |
| dtantsur | I hope you get better soon | 19:00 |
| TheJulia | thanks | 19:01 |
| iurygregory | take care TheJulia =) | 19:04 |
| opendevreview | Merged openstack/ironic-python-agent bugfix/8.4: Update .gitreview for bugfix/8.4 https://review.opendev.org/c/openstack/ironic-python-agent/+/829120 | 19:18 |
| stevebaker[m] | good morning | 19:29 |
| dtantsur | morning stevebaker[m] | 19:30 |
| dtantsur | you're serving as my reminder that I'm working too late :) | 19:31 |
| stevebaker[m] | dtantsur: heh :) | 19:32 |
| dtantsur | see you tomorrow folks | 19:38 |
| iurygregory | be back later, bye everyone o/ | 19:53 |
| opendevreview | Verification of a change to openstack/bifrost master failed: Stop using sub-sections of driver_info https://review.opendev.org/c/openstack/bifrost/+/826307 | 21:12 |
| opendevreview | Harald Jensås proposed openstack/ironic-lib master: DNM - Test CI https://review.opendev.org/c/openstack/ironic-lib/+/829408 | 21:15 |
| opendevreview | Harald Jensås proposed openstack/ironic-lib master: DNM - Add platform:base-py[2|3] to bindep https://review.opendev.org/c/openstack/ironic-lib/+/829272 | 21:31 |
| opendevreview | Merged openstack/ironic master: [doc][trivial] Fixing typos in RBAC documentation https://review.opendev.org/c/openstack/ironic/+/829249 | 21:45 |
| opendevreview | Harald Jensås proposed openstack/ironic-lib master: Add platform:base-py[2|3] to bindep https://review.opendev.org/c/openstack/ironic-lib/+/829272 | 21:55 |
| opendevreview | Verification of a change to openstack/ironic-python-agent master failed: Move prepare_boot_partitions_for_softraid to raid_utils https://review.opendev.org/c/openstack/ironic-python-agent/+/826814 | 22:16 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!