iurygregory | good morning janders and Ironic o/ | 06:50 |
---|---|---|
*** rpittau|afk is now known as rpittau | 07:21 | |
rpittau | good morning ironic! o/ | 07:21 |
iurygregory | morning rpittau o/ | 07:21 |
rpittau | reading the "clock skew" thread reminds me of fun times with puppet | 07:21 |
rpittau | hey iurygregory :) | 07:21 |
iurygregory | puppet <3 | 07:26 |
cenne | Good morning ironic. :) | 07:41 |
iurygregory | morning cenne | 07:43 |
cenne | hey iurygregory o/ | 07:44 |
iurygregory | o/ | 07:47 |
rpittau | hey cenne :) | 07:50 |
cenne | hey rpittau o/ | 08:12 |
dtantsur | morning folks! | 08:41 |
dtantsur | arne_wiebalck: okay, so the patch seems to help? interesting | 08:42 |
iurygregory | morning dtantsur =) | 08:43 |
*** sshnaidm|afk is now known as sshnaidm | 08:45 | |
cenne | morning dtantsur. | 09:15 |
dtantsur | cenne: very good job with your patch! I've left some comments when you have time. Let me know if you have questions. | 09:39 |
cenne | Yes dtantsur. Thank you. Checking them now. | 09:52 |
opendevreview | cenne proposed openstack/ironic master: [WIP] Add `boot_mode` and `secure_boot` to node object and expose in api https://review.opendev.org/c/openstack/ironic/+/797055 | 09:54 |
opendevreview | cenne proposed openstack/ironic master: [WIP] Add `boot_mode` and `secure_boot` to node object and expose in api https://review.opendev.org/c/openstack/ironic/+/797870 | 10:21 |
opendevreview | cenne proposed openstack/ironic master: [WIP] Add `boot_mode` and `secure_boot` to node object and expose in api https://review.opendev.org/c/openstack/ironic/+/797870 | 10:40 |
opendevreview | cenne proposed openstack/ironic master: [WIP] Add `boot_mode` and `secure_boot` to node object and expose in api https://review.opendev.org/c/openstack/ironic/+/797055 | 10:43 |
cenne | Hmm. Wonder why it gave me a different changeId 🤔 | 10:44 |
opendevreview | Dmitry Tantsur proposed openstack/ironic master: Refactor: untie IloVendor from validate_image_properties https://review.opendev.org/c/openstack/ironic/+/797872 | 10:44 |
dtantsur | cenne: did you leave ChangeId in the trailer of the commit message? | 10:44 |
cenne | Yes. It appended an extra change id by itself. | 10:45 |
dtantsur | cenne: weird. maybe it didn't recognize the trailer, try separating Done: and Todo: with a newline | 10:45 |
cenne | Ah. Cool maybe it's bugging out while parsing that. I had added a wip in the Todo thing. It's fixed now though when i changed the commit message slightly anf removed the extra commitId | 10:47 |
cenne | dtantsur: I have made the fixes taking into consideration your comments. They are in the latest patch https://review.opendev.org/c/openstack/ironic/+/797055 | 10:50 |
dtantsur | thank you! I'll check them as early as I can | 10:50 |
cenne | Yet to do unit tests for the function I added. I'm doing the regular feild addition tests one by one first. Then will get to that. | 10:51 |
opendevreview | Dmitry Tantsur proposed openstack/ironic master: [WIP] Refactor validate_image_properties https://review.opendev.org/c/openstack/ironic/+/797875 | 11:00 |
*** viks_ is now known as viks | 11:23 | |
viks | hi, when i create baremetal instance, instance is getting 2 ip's, one pubic ip in first interface and the other one is the internal ip which is the cleaning network in the second interface... is there any wayi can avoid showing up of 2 ip which is of internal network? | 11:41 |
dtantsur | viks: hi! it sounds more like a neutron/nova question or a question related to your image | 11:45 |
dtantsur | ironic doesn't control DHCP for instances | 11:45 |
viks | dtantsur: ok... just to confirm, cleaning/management network ip should not show up in user instance right? | 11:50 |
arne_wiebalck | Good morning, ironic! | 11:50 |
arne_wiebalck | dtantsur: Correct, I have not seen the SSL error on the controller with your patch since I deployed it (a node is deployed/cleaned once an hour automatically). | 11:54 |
dtantsur | viks: it depends on whether networks are physically separated in your case. | 11:54 |
dtantsur | and how DHCP is organized | 11:54 |
arne_wiebalck | dtantsur: Unless you see major other issues with that patch, I would suggest to brush it up like we did with the heartbeat coalescing one and propose it. | 11:55 |
-opendevstatus- NOTICE: Our Zuul gating CI/CD services will be offline starting around 14:00 UTC (in roughly two hours from now) in order to apply some critical security updates, and is not expected to remain offline for more than 30 minutes. | 12:02 | |
viks | dtantsur: they are physically separated... since instance gets the ip from the cleaning network also, i can ping the ironic node ... not sure how to avoid this.. i mean not sure how an instance not to have visibility to cleaning network.. | 12:04 |
janders | see you tomorrow Ironic o/ | 12:08 |
janders | have a good day everyone | 12:09 |
viks | basically it's like this https://www.irccloud.com/pastebin/KKEMoNlF/ | 12:12 |
viks | internal is the one which i use as cleaning network | 12:12 |
viks | dtantsur: just wanted to know what i'm doing above is the correct approach? | 12:15 |
dtantsur | arne_wiebalck: you wanna do it or should I? | 12:44 |
dtantsur | viks: I suspect it's a nova question rather than ironic one | 12:45 |
opendevreview | cenne proposed openstack/ironic master: [WIP] Add `boot_mode` and `secure_boot` to node object and expose in api https://review.opendev.org/c/openstack/ironic/+/797055 | 12:45 |
viks | dtantsur: ok..thanks | 12:45 |
arne_wiebalck | dtantsur: I can give it a shot (add tests, reno and so on) ... feature-wise it seems complete, no? | 12:51 |
dtantsur | arne_wiebalck: I think so. Maybe add debug logging when we open a new client? | 12:51 |
arne_wiebalck | dtantsur: ok, will try to do it today or tomorrow, thanks again! | 12:52 |
TheJulia | good morning | 13:32 |
dtantsur | morning TheJulia | 13:32 |
TheJulia | viks: I concur with dtantsur. Nova appears to be somehow translating it into multiple networks. I don't think that would really be a result of configuration so it may be how the instance is being invoked... or maybe there is a new $thing we're just not aware about here | 13:34 |
viks | TheJulia: ok.. thanks | 13:35 |
TheJulia | nova network interface creation code is... a little weird too. I think I traced it down as five different paths that can be taken for an instance last time I dug at it | 13:36 |
viks | TheJulia: ok.. | 13:38 |
-opendevstatus- NOTICE: Our Zuul gating CI/CD services are being taken offline now in order to apply some critical security updates, and are not expected to remain offline for more than 30 minutes. | 13:57 | |
TheJulia | wheeeeee | 13:57 |
iurygregory | it's friday/monday for zuul... | 14:02 |
opendevreview | Dmitry Tantsur proposed openstack/ironic master: [WIP] Refactor validate_image_properties https://review.opendev.org/c/openstack/ironic/+/797875 | 14:07 |
*** tcr2 is now known as trandles | 14:46 | |
opendevreview | Arne Wiebalck proposed openstack/ironic master: Cache AgentClient on Task, not globally https://review.opendev.org/c/openstack/ironic/+/797674 | 14:53 |
opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Use get_service_clients framework with basic Secure RBAC https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/797521 | 15:07 |
rpittau | good night! o/ | 16:09 |
*** rpittau is now known as rpittau|afk | 16:09 | |
cenne | Gpod night rpittau | 16:14 |
cenne | *good =) | 16:15 |
opendevreview | Julia Kreger proposed openstack/ironic master: Change UEFI ipxe bootlaoder default https://review.opendev.org/c/openstack/ironic/+/797973 | 16:17 |
opendevreview | Julia Kreger proposed openstack/ironic master: Deprecate [pxe]ip_version parameter https://review.opendev.org/c/openstack/ironic/+/797984 | 16:57 |
TheJulia | \o/ my tempest change passed CI | 16:58 |
TheJulia | of course, at the same time, all of the multinode jobs failed... apparently because of a legacy client invocation inside of tempest. | 16:58 |
TheJulia | https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_3f5/797521/10/check/ironic-tempest-ipa-wholedisk-direct-tinyipa-multinode/3f5add7/testr_results.html | 16:58 |
TheJulia | gmann: ^^ | 16:58 |
arne_wiebalck | dtantsur: TheJulia: Preparing for our Victoria upgrade here downstream, you helped to identify two issues (stacked heartbeats, agent client caching) which resulted in upstream patches ... thanks again! I hope you're ready for the challenge of the last issue I see :) | 17:09 |
dtantsur | hehe | 17:09 |
arne_wiebalck | :-D | 17:09 |
arne_wiebalck | I will have a look at come back with details tomorrow. | 17:10 |
dtantsur | I probably need to call it a day already, so any challenges come tomorrow | 17:10 |
dtantsur | yep | 17:10 |
* arne_wiebalck will keep the suspense until tomorrow ;) | 17:11 | |
arne_wiebalck | bye everyone, have a good evening o/ | 17:11 |
dtantsur | o/ | 17:12 |
TheJulia | arne_wiebalck: okay, have a wonderful evening :) | 17:14 |
opendevreview | Julia Kreger proposed openstack/ironic master: Change UEFI ipxe bootloader default https://review.opendev.org/c/openstack/ironic/+/797973 | 17:33 |
opendevreview | Julia Kreger proposed openstack/ironic master: Deprecate [pxe]ip_version parameter https://review.opendev.org/c/openstack/ironic/+/797984 | 17:33 |
erbarr | how does bifrost apply the ironic code review change in the ironic CI? | 19:35 |
JayF | erbarr: I'm not sure I understand the question you're asking. | 19:46 |
erbarr | JayF, like on this console log, on tempest jobs I can see the review merged by devstack: https://zuul.opendev.org/t/openstack/build/94d00b7c8c2d44998c177c27ddfc931f/console if I open the devstack console and I cntl+F the review number, but on bifrost if i do the same on the run.yaml i don't see a match for the review number: https://zuul.opendev.org/t/openstack/build/50c864dc04b7456c91625ee2ec4c5394/console | 20:02 |
erbarr | so i wonder how does it get in there for the bifrost case | 20:02 |
JayF | I'm really confused, because on both of those links the review number and patchset are in the top left corner | 20:06 |
erbarr | yea, that's zuul display, I think that comes directly from the patch that triggers weather or not that gets used in whatever you run | 20:08 |
JayF | ooh, so you're curious about the devstack console itself | 20:08 |
JayF | hmm | 20:08 |
JayF | aha | 20:09 |
JayF | erbarr: if you go to the "logs" tab, and click "job-output.txt", you'll get an error, because that file is gone | 20:09 |
JayF | the job ran so long ago that the logs have cycled out | 20:09 |
JayF | which is likely why you're missing data from the 50c864dc04b7456c91625ee2ec4c5394 one | 20:09 |
erbarr | mmnn, weird, I can click on logs and see them | 20:11 |
JayF | Yeah, I'm not sure -- if you're seeing something different than me IDK if I'll be able to help :) | 20:12 |
erbarr | okay, thanks, what I'm seeing in the devstack console is the output of a "git show --oneline head -1" and that gives the review, so i think is zuul handling that in both cases but just bifrost not explicitly showing the state in the console, maybe somewhere else idk about. Sorry, thanks for the help :) | 20:18 |
JayF | I'm not a bifrost expert | 20:20 |
JayF | so I'm not sure, but I think bifrost jobs aren't based on devstack? | 20:20 |
JayF | but that may be wrong :D | 20:21 |
erbarr | they're not | 20:21 |
TheJulia | erbarr: what is the reivew number your looking for in that bifrost log? | 20:24 |
TheJulia | basically, the answer is zuul places everything for either job, and the jobs just pickup the state of the files since they have been explicitly provided | 20:25 |
erbarr | TheJulia, the review is 796432, but that doesn't matter, I just picked one at random, just trying to understand how that works, thanks! | 20:28 |
TheJulia | ahh, okay. Yeah, zuul has an integrated cloner which takes everything and lays it out for the jobs to pickup. | 20:28 |
TheJulia | That cloner groks the dependency chains that can result and clone appropriately | 20:28 |
TheJulia | I wrote something similar just in bifrost that understood the entire changes map, but we ended up ripping it out a long time ago | 20:29 |
erbarr | ahh i see, I think i get the picture now | 20:30 |
opendevreview | Verification of a change to openstack/ironic failed: Fix ramdisk boot option handling https://review.opendev.org/c/openstack/ironic/+/797517 | 21:07 |
stevebaker | good morning | 21:14 |
gmann | TheJulia: seems like we need to make create_network() in tempest also check for neutron enforce scope policy and then use the appropriate token. let me push a patch for that and then we can test. | 21:35 |
kkillsfirst | TheJulia and JayF, Could we set a day/time to have a meeting to talk about using kexec in Ironic?? | 21:45 |
TheJulia | kkillsfirst: yes! when would be good for you, Mondays and Tuesdays are typically not good for me | 22:58 |
TheJulia | gmann: I guess that is a possibility and that is the most re-usable | 22:59 |
TheJulia | gmann: What is interesting is that was even with scope enforcement off realistically, the test doesn't force it yet | 23:00 |
TheJulia | I guess the test could also be tenant aware, but it is doing the standard test newtork init in those scenarios | 23:00 |
opendevreview | Steve Baker proposed openstack/metalsmith master: Set image_disk_format from file extension for HTTP images https://review.opendev.org/c/openstack/metalsmith/+/798028 | 23:05 |
stevebaker | this is where I discover that overcloud raw images haven't been streamed to disk this whole time ^^ | 23:05 |
opendevreview | Merged openstack/ironic master: Remove redundant/legacy is_admin logic https://review.opendev.org/c/openstack/ironic/+/796552 | 23:09 |
kkillsfirst | TheJulia: Can we meet this Friday? Either before 11am MST or after 2 pm MST. | 23:22 |
TheJulia | stevebaker: doh! | 23:23 |
TheJulia | kkillsfirst: before would work best for me tomorrow. I don't know if JayF is available though | 23:24 |
kkillsfirst | TheJulia: How about 10 am MST tomorrow? | 23:28 |
TheJulia | kkillsfirst: that works | 23:39 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!