| *** ayoung has quit IRC | 00:01 | |
| *** gyee has quit IRC | 00:11 | |
| *** rh-jelabarre has quit IRC | 00:16 | |
| *** rh-jelabarre has joined #openstack-ironic | 00:19 | |
| *** gyee has joined #openstack-ironic | 00:25 | |
| *** penick has joined #openstack-ironic | 00:25 | |
| *** penick has quit IRC | 00:30 | |
| *** ChanServ changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/ironic/ | Bugs: https://storyboard.openstack.org/#!/project_group/75 | Contributors are generally present between 6 AM and 12 AM UTC, If we do not answer, please feel free to pose questions to openstack-discuss mailing list." | 00:37 | |
| -openstackstatus- NOTICE: The Gerrit service at review.opendev.org is back up and running; for outage details see analysis here: http://lists.opendev.org/pipermail/service-announce/2020-October/000011.html | 00:37 | |
| stevebaker | huzzah! | 00:41 |
|---|---|---|
| *** adrianc has quit IRC | 00:45 | |
| *** Goneri has quit IRC | 00:46 | |
| *** adrianc has joined #openstack-ironic | 00:47 | |
| janders | stevebaker: +1 :) | 00:50 |
| stevebaker | its probably a good time to update one's launchpad password | 01:01 |
| *** gyee has quit IRC | 01:12 | |
| *** zaneb has quit IRC | 01:19 | |
| *** zaneb has joined #openstack-ironic | 01:19 | |
| *** rh-jelabarre has quit IRC | 01:56 | |
| *** stevebaker has quit IRC | 02:15 | |
| *** Qianbiao has joined #openstack-ironic | 02:19 | |
| *** tzumainn has quit IRC | 02:51 | |
| *** dsneddon has joined #openstack-ironic | 03:04 | |
| *** dsneddon has quit IRC | 03:10 | |
| *** mkrai has joined #openstack-ironic | 03:10 | |
| *** sai438 has joined #openstack-ironic | 03:14 | |
| *** dsneddon has joined #openstack-ironic | 03:14 | |
| *** pmannidi has quit IRC | 03:18 | |
| janders | Good idea. Better safe than sorry! | 03:18 |
| *** dsneddon has quit IRC | 03:19 | |
| *** dsneddon has joined #openstack-ironic | 03:20 | |
| TheJulia | ++ | 03:24 |
| *** dsneddon has quit IRC | 03:25 | |
| *** pmannidi has joined #openstack-ironic | 03:26 | |
| *** dsneddon has joined #openstack-ironic | 03:28 | |
| *** sai438 has quit IRC | 03:29 | |
| *** dsneddon has quit IRC | 03:33 | |
| *** dsneddon has joined #openstack-ironic | 03:34 | |
| *** dsneddon has quit IRC | 03:39 | |
| *** dsneddon has joined #openstack-ironic | 03:40 | |
| *** dsneddon has quit IRC | 03:45 | |
| *** dsneddon has joined #openstack-ironic | 03:46 | |
| *** sai438 has joined #openstack-ironic | 03:49 | |
| *** pmannidi has quit IRC | 03:52 | |
| *** pmannidi has joined #openstack-ironic | 03:53 | |
| *** pmannidi_ has joined #openstack-ironic | 03:55 | |
| *** sai438 has quit IRC | 03:57 | |
| *** pmannidi has quit IRC | 03:58 | |
| *** dsneddon has quit IRC | 04:00 | |
| *** dsneddon has joined #openstack-ironic | 04:00 | |
| *** dsneddon has quit IRC | 04:06 | |
| TheJulia | All, re the gerrit outage/compromise, I've sent an email the mailing list and created a spreadsheet for us to track a change review from. Details can be found at http://lists.openstack.org/pipermail/openstack-discuss/2020-October/018148.html Please let me know if you have any questions, thanks. | 04:07 |
| *** dsneddon has joined #openstack-ironic | 04:12 | |
| *** Qianbiao has quit IRC | 04:22 | |
| *** mkrai has quit IRC | 05:02 | |
| *** mkrai_ has joined #openstack-ironic | 05:02 | |
| *** mkrai_ has quit IRC | 05:22 | |
| *** abdysn has joined #openstack-ironic | 05:23 | |
| *** bdodd_ has joined #openstack-ironic | 05:28 | |
| *** bdodd has quit IRC | 05:29 | |
| *** mkrai_ has joined #openstack-ironic | 05:30 | |
| *** Qianbiao has joined #openstack-ironic | 05:35 | |
| *** mkrai_ has quit IRC | 05:57 | |
| *** mkrai has joined #openstack-ironic | 06:05 | |
| *** anuradha1904 has joined #openstack-ironic | 06:12 | |
| arne_wiebalck | Good morning, ironic! | 06:42 |
| janders | good morning arne_wiebalck | 06:42 |
| *** rpittau|afk is now known as rpittau | 06:42 | |
| janders | congratulations on your talk, it was excellent, I really enjoyed watching | 06:42 |
| rpittau | good morning ironic! o/ | 06:42 |
| janders | good morning rpittau | 06:42 |
| rpittau | hey janders arne_wiebalck :) | 06:42 |
| rpittau | I will watch the recording arne_wiebalck :) | 06:43 |
| *** jtomasek has joined #openstack-ironic | 06:43 | |
| arne_wiebalck | hey janders rpittau o/ | 06:43 |
| arne_wiebalck | janders: thanks, good to hear we had some views :) | 06:43 |
| janders | arne_wiebalck I have a couple questions if that's okay? | 06:44 |
| arne_wiebalck | janders: the platform gives you the impression you're totally alone watching your own talk ... | 06:44 |
| arne_wiebalck | janders: sure, tell me | 06:44 |
| janders | 1) do you auto-scale ironic-driven nova-compute instances based on current demand? | 06:45 |
| janders | (in some shape or form)? | 06:45 |
| arne_wiebalck | ad 1) no, nova-compute does just fine; a single one can handle things; it even could before we added conductor groups | 06:46 |
| janders | I was more thinking in the context of dynamically creating and deleting KVM-nova-compute instances | 06:46 |
| arne_wiebalck | ad 1) well, except for the fact it tooks very long for the resource tracker | 06:46 |
| arne_wiebalck | no, everything is static | 06:47 |
| arne_wiebalck | it would make sense for Ironic, though | 06:47 |
| janders | I'd imagine you're not short of baremetal nodes for the users to use so this makes sense :) | 06:48 |
| arne_wiebalck | scale up conductors on demand | 06:48 |
| janders | yeah! | 06:48 |
| arne_wiebalck | the main issue is that the demand is very bursty | 06:48 |
| arne_wiebalck | so, by the time the requests hit Ironic, there is very little time to scale up | 06:48 |
| arne_wiebalck | this is why I scale up on what we call the leading group | 06:49 |
| arne_wiebalck | as we expect that is where the action will be | 06:49 |
| janders | ironic-running-virtnovacompute is one thing... ironic deploying more ironic is inception next level | 06:49 |
| janders | awesome! :) | 06:49 |
| arne_wiebalck | the whole control plane is virtual | 06:49 |
| janders | ah - so the inception is going both ways? | 06:50 |
| arne_wiebalck | yes | 06:50 |
| janders | that's awesome | 06:50 |
| janders | I love your architecture even more | 06:50 |
| janders | :) | 06:50 |
| janders | 2) do you use baremetal-capable SDN (and if so, which one)? | 06:50 |
| arne_wiebalck | ad 2) no, our network is (embarassingly) simple and flat | 06:51 |
| * rpittau breakfast time | 06:51 | |
| *** hoonetorg has joined #openstack-ironic | 06:51 | |
| janders | I don't think that's embarrassing, I think it's practical | 06:51 |
| arne_wiebalck | ad 2) we have a separate management network, but that is it | 06:52 |
| janders | I can't easily think of any SDN that could operate at your scale without an imaginable amount of tweaking | 06:52 |
| arne_wiebalck | ad 2) we are looking into getting more fancy for VMs, but no plans for bare metal atm | 06:52 |
| janders | s/imaginable/unimaginable | 06:52 |
| janders | right! | 06:52 |
| arne_wiebalck | ad 2) we carry some history in our data centre | 06:52 |
| arne_wiebalck | and all is in prod, so not that easy to change on the fly | 06:53 |
| janders | yeah I can imagine | 06:53 |
| arne_wiebalck | but we are working on it :) | 06:53 |
| janders | cool! :) | 06:53 |
| janders | and the last one | 06:53 |
| janders | 3) in one of the slides you mention complexity of stacked Ironic, nova, placement, ... | 06:53 |
| arne_wiebalck | yep | 06:53 |
| janders | have you considered using nova+ironic for user instances, but standalone Ironic for "infra" workloads? | 06:53 |
| janders | standalone may not be ideal for many user-facing applications, but for running OpenStack bits, hmm... | 06:54 |
| janders | the idea only bubbled up while I was watching your video so I haven't thought it through in fine detail, but I thought that could be kind of cool | 06:54 |
| arne_wiebalck | never considered this, no ... one of the reasons for Ironic is accounting: to see who is using what, that may be more difficult if we split it ... but it is an interesting idea | 06:55 |
| arne_wiebalck | some users regard Ironic as an additional unnecessary layer | 06:56 |
| janders | what's your split between user baremetal instances and "infra" baremetal instances, if you are happy to share this info? | 06:56 |
| arne_wiebalck | b/c from the user pov it is not always obvious why we have it | 06:56 |
| arne_wiebalck | "infra" you mean for hypervisors? | 06:56 |
| janders | yeah | 06:56 |
| arne_wiebalck | let me check ... | 06:57 |
| janders | and any other cloud components running on baremetal that you may have | 06:57 |
| iurygregory | good morning Ironic | 06:58 |
| janders | good morning iurygregory | 06:59 |
| iurygregory | hey janders o/ | 07:00 |
| arne_wiebalck | janders: it seems to be 3:2, so 3000 nodes are hypervisors and 2000 are user instances | 07:00 |
| janders | right! | 07:00 |
| arne_wiebalck | janders: more or less, at a quick galnce | 07:00 |
| arne_wiebalck | *glance | 07:00 |
| janders | I suppose in this case the baremetal approach for infra stuff could possibly make sense | 07:01 |
| janders | if it was significantly more user instances, it wouldn't make much sense in my opinion | 07:01 |
| janders | but if "infra" as I call them is a fair bit, simplicity could help optimise and troubleshoot these I suppose | 07:02 |
| arne_wiebalck | apart from the accounting part, there is also the provisioning workflows: it is the same for VMs and bare metal nodes | 07:02 |
| janders | ah that is a good point | 07:02 |
| arne_wiebalck | splitting this off, may break this | 07:02 |
| arne_wiebalck | but, yeah, main complaint from users is the additional layer where they do not really see a benefit | 07:03 |
| arne_wiebalck | they need to go through ironic to get the hardware they ordered | 07:03 |
| arne_wiebalck | it pays off when nodes go from one user to another | 07:03 |
| arne_wiebalck | or for occasional requests | 07:04 |
| arne_wiebalck | but not so much for the ones which expect 100 servers and just want to get them | 07:04 |
| arne_wiebalck | (and everything that happens behind the scenes like benchmarking, burn-in, health checking, instance creation, ... is not directly relevant to users) | 07:05 |
| janders | makes sense | 07:06 |
| janders | thank you for sharing this very interesting perspective :) | 07:06 |
| janders | beyond creating a baremetal-flavor instance.. what are the actions where the users need to interact with ironic directly? | 07:07 |
| arne_wiebalck | janders: thanks for asking! | 07:07 |
| arne_wiebalck | janders: nothing | 07:07 |
| arne_wiebalck | janders: but that is the problem | 07:08 |
| arne_wiebalck | janders: if sth goes wrong, they need to get in touch | 07:08 |
| janders | right! | 07:08 |
| arne_wiebalck | janders: while with direct access, they could do it themselves | 07:08 |
| arne_wiebalck | janders: the other main issue is that users cannot see the available resources | 07:09 |
| janders | true | 07:09 |
| arne_wiebalck | janders: I will give details on the ops feedback later today, but we have project with many different flavors | 07:09 |
| arne_wiebalck | janders: and the user cannot see how many of which are still available | 07:09 |
| arne_wiebalck | janders: this is a major pain point for our users | 07:10 |
| janders | arne_wiebalck yeah thinking about this it's quite a non-trivial problem to solve with ironic-nova | 07:10 |
| arne_wiebalck | janders: and one of the main complaint I hear | 07:10 |
| arne_wiebalck | janders: yeah :) | 07:10 |
| janders | arne_wiebalck and this likely may lead to unreasonable workarounds like trying to spin things up just to probe the Cloud | 07:10 |
| *** dsneddon has quit IRC | 07:12 | |
| *** dsneddon has joined #openstack-ironic | 07:13 | |
| arne_wiebalck | yeah | 07:14 |
| arne_wiebalck | and even if the user counts, there may be nodes in clean_failed and hence not available, but the user cannot see this | 07:14 |
| * arne_wiebalck has a meeting now ... | 07:14 | |
| janders | thank you arne_wiebalck | 07:16 |
| *** dsneddon has quit IRC | 07:18 | |
| *** dsneddon has joined #openstack-ironic | 07:19 | |
| *** openstackgerrit has quit IRC | 07:38 | |
| *** rcernin has quit IRC | 07:39 | |
| *** tosky has joined #openstack-ironic | 07:42 | |
| *** sshnaidm is now known as sshnaidm|afk | 07:43 | |
| arne_wiebalck | janders: my pleasure! | 07:44 |
| *** dsneddon has quit IRC | 07:45 | |
| *** dsneddon has joined #openstack-ironic | 07:46 | |
| *** jtomasek has quit IRC | 07:47 | |
| *** dougsz has joined #openstack-ironic | 07:50 | |
| *** priteau has joined #openstack-ironic | 07:54 | |
| *** dougsz has quit IRC | 07:55 | |
| *** pmannidi_ has quit IRC | 08:02 | |
| *** lucasagomes has joined #openstack-ironic | 08:07 | |
| *** rcernin has joined #openstack-ironic | 08:08 | |
| *** dtantsur|afk is now known as dtantsur | 08:10 | |
| dtantsur | morning ironic | 08:14 |
| *** rcernin has quit IRC | 08:14 | |
| rpittau | hey dtantsur :) | 08:15 |
| *** k_mouza has joined #openstack-ironic | 08:16 | |
| *** pmannidi has joined #openstack-ironic | 08:19 | |
| *** rcernin has joined #openstack-ironic | 08:28 | |
| iurygregory | morning dtantsur | 08:30 |
| *** derekh has joined #openstack-ironic | 08:34 | |
| *** sshnaidm|afk is now known as sshnaidm | 08:39 | |
| *** dsneddon has quit IRC | 08:45 | |
| *** dsneddon has joined #openstack-ironic | 08:46 | |
| *** ociuhandu has joined #openstack-ironic | 08:51 | |
| *** dsneddon has quit IRC | 08:52 | |
| *** dsneddon has joined #openstack-ironic | 08:52 | |
| *** rcernin has quit IRC | 08:56 | |
| *** dsneddon has quit IRC | 08:58 | |
| *** dsneddon has joined #openstack-ironic | 09:00 | |
| *** openstackgerrit has joined #openstack-ironic | 09:01 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allow passing rootfs_uuid for the standalone case https://review.opendev.org/758970 | 09:01 |
| dtantsur | arne_wiebalck: ^^ | 09:01 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: DNM try to inject the correct grub location https://review.opendev.org/758809 | 09:02 |
| *** dsneddon has quit IRC | 09:05 | |
| arne_wiebalck | dtantsur: thanks, lgtm! | 09:11 |
| rpittau | wondering if we should also include bifrost and sushy-oem-idrac to the change audit doc ? | 09:23 |
| dtantsur | definitely the former | 09:23 |
| rpittau | ack, done | 09:24 |
| *** pmannidi has quit IRC | 09:30 | |
| *** pmannidi has joined #openstack-ironic | 09:30 | |
| *** pmannidi has quit IRC | 09:35 | |
| *** pmannidi has joined #openstack-ironic | 09:39 | |
| iurygregory | I'm starting to look at the IPE | 09:40 |
| rpittau | iurygregory: I'm done already :0 | 09:41 |
| iurygregory | rpittau, oh the spreadsheet wasn't marked "done" | 09:41 |
| rpittau | iurygregory: no, I mean, I finished the first pass | 09:41 |
| rpittau | we need two anyway | 09:41 |
| rpittau | before signing off | 09:42 |
| iurygregory | ack | 09:42 |
| iurygregory | nvm just saw "NOT the originator of the change." since most commits are mine in IPE doesn't make sense =) | 09:43 |
| rpittau | ah yeah, that too | 09:43 |
| rpittau | I skipped one in inspector for the same reason | 09:44 |
| iurygregory | I will look at inspector =) | 09:44 |
| *** abdysn has quit IRC | 09:45 | |
| *** pmannidi has quit IRC | 09:47 | |
| *** pmannidi has joined #openstack-ironic | 09:47 | |
| *** sai438 has joined #openstack-ironic | 09:51 | |
| *** dsneddon has joined #openstack-ironic | 09:52 | |
| *** pmannidi has quit IRC | 09:55 | |
| *** dsneddon has quit IRC | 09:58 | |
| openstackgerrit | QianBiao Ng proposed openstack/ironic stable/stein: opt: Enhance old stable branches to use latest python-ibmcclient https://review.opendev.org/757607 | 09:58 |
| dtantsur | iurygregory, rpittau, I assumed the comment was about the review itself, not who checks the patches | 09:58 |
| dtantsur | I'm sure we can trust any core to check their patches in this regard | 09:58 |
| dtantsur | we just need to make sure we don't see strange approvals | 09:58 |
| dtantsur | like myself approving patches to metalsmith.. oh wait :) | 09:59 |
| rpittau | mmmm | 09:59 |
| *** pmannidi has joined #openstack-ironic | 09:59 | |
| iurygregory | *magic* | 09:59 |
| iurygregory | hahaha | 09:59 |
| *** dsneddon has joined #openstack-ironic | 10:00 | |
| *** sai438 has quit IRC | 10:02 | |
| *** sai438 has joined #openstack-ironic | 10:02 | |
| *** pmannidi has quit IRC | 10:05 | |
| *** dsneddon has quit IRC | 10:05 | |
| rpittau | ok so no single approval of the same person who proposed the patch | 10:05 |
| rpittau | as usual | 10:06 |
| *** pmannidi_ has joined #openstack-ironic | 10:06 | |
| dtantsur | and no suspicious approvals, e.g. from a person who's not an ironic-core | 10:06 |
| *** dsneddon has joined #openstack-ironic | 10:06 | |
| rpittau | oh yeah, ofc | 10:06 |
| dtantsur | or something you're sure couldn't have been approved | 10:07 |
| rpittau | I'm also looking for emails | 10:07 |
| *** sai438 has quit IRC | 10:07 | |
| *** dsneddon has quit IRC | 10:12 | |
| *** dsneddon has joined #openstack-ironic | 10:12 | |
| *** k_mouza has quit IRC | 10:16 | |
| *** dsneddon has quit IRC | 10:18 | |
| *** k_mouza has joined #openstack-ironic | 10:19 | |
| *** dsneddon has joined #openstack-ironic | 10:20 | |
| Qianbiao | aha, is it means all patch from none-core will not be reviewed ? | 10:21 |
| dtantsur | Qianbiao: this is not related to the normal reviewing process, please see the mailing list for the information on the security breach we had | 10:22 |
| Qianbiao | ok ty | 10:22 |
| dtantsur | your changes will be reviewed as normally as soon as we have time again (the summit and now this issue..) | 10:22 |
| janders | I'm back to https://storyboard.openstack.org/#!/story/2008038 (well overdue) and I'm about to push the first version of the inspector side patch. Given the change is spanning Ironic and Inspector, do I use the same commit message for both? | 10:23 |
| dtantsur | janders: whatever makes sense to you, there is no rule. Just don't forget to link to the story/task. | 10:24 |
| janders | dtantsur ACK - and thank you | 10:24 |
| *** dsneddon has quit IRC | 10:25 | |
| *** dsneddon has joined #openstack-ironic | 10:26 | |
| *** abdysn has joined #openstack-ironic | 10:27 | |
| *** dsneddon has quit IRC | 10:32 | |
| openstackgerrit | Jacob Anders proposed openstack/ironic-inspector master: Use OOB inspection to fetch MACs for IB inspection https://review.opendev.org/758994 | 10:33 |
| *** k_mouza_ has joined #openstack-ironic | 10:33 | |
| *** k_mouza has quit IRC | 10:37 | |
| *** priteau has quit IRC | 10:39 | |
| *** k_mouza has joined #openstack-ironic | 10:46 | |
| *** dsneddon has joined #openstack-ironic | 10:48 | |
| *** k_mouza_ has quit IRC | 10:49 | |
| *** dsneddon has quit IRC | 10:53 | |
| *** dsneddon has joined #openstack-ironic | 10:55 | |
| *** mkrai has quit IRC | 10:58 | |
| *** dsneddon has quit IRC | 11:01 | |
| *** abdysn has quit IRC | 11:02 | |
| *** dsneddon has joined #openstack-ironic | 11:02 | |
| *** dsneddon has quit IRC | 11:07 | |
| *** dsneddon has joined #openstack-ironic | 11:08 | |
| *** jtomasek has joined #openstack-ironic | 11:08 | |
| *** dsneddon has quit IRC | 11:13 | |
| *** priteau has joined #openstack-ironic | 11:15 | |
| *** priteau has quit IRC | 11:26 | |
| *** rh-jelabarre has joined #openstack-ironic | 11:55 | |
| *** jtomasek has quit IRC | 12:00 | |
| *** priteau has joined #openstack-ironic | 12:01 | |
| *** fgofurov has joined #openstack-ironic | 12:12 | |
| *** abdysn has joined #openstack-ironic | 12:17 | |
| *** k_mouza has quit IRC | 12:27 | |
| *** k_mouza has joined #openstack-ironic | 12:41 | |
| iurygregory | rpittau, since you reviewed the IPE I can just put DONE ? | 12:43 |
| *** Goneri has joined #openstack-ironic | 12:46 | |
| rpittau | iurygregory: we need a second core to check the same diffs | 12:49 |
| iurygregory | but you did for IPE no? | 12:49 |
| iurygregory | and I finished | 12:49 |
| rpittau | jsut the first pass | 12:49 |
| iurygregory | so it would be 2 .-. | 12:49 |
| rpittau | yeah, check the IPE tab in the spreadsheet | 12:49 |
| rpittau | or any other tab | 12:49 |
| iurygregory | ohhhhhhh | 12:50 |
| iurygregory | I see | 12:50 |
| rpittau | :) | 12:50 |
| TheJulia | brraaaaaiinnnsss | 12:50 |
| iurygregory | lol I haven't seen the tabs | 12:50 |
| iurygregory | <insert coffee to continue> | 12:50 |
| dtantsur | morning TheJulia | 12:50 |
| rpittau | mmm coffee brains | 12:50 |
| TheJulia | I <3 you guys | 12:50 |
| rpittau | :) | 12:50 |
| iurygregory | coffee brains sounds like something for TheJulia | 12:51 |
| iurygregory | good morning TheJulia =) | 12:51 |
| rpittau | iurygregory: remember to add your initials on each cell corresponding to a verified branch | 12:52 |
| iurygregory | in the Check by #* right? | 12:53 |
| *** k_mouza_ has joined #openstack-ironic | 12:55 | |
| rpittau | in the main tab, High Level Sign-Off | 12:55 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Simplify injecting network data into an ISO image https://review.opendev.org/757598 | 12:55 |
| rpittau | for each verified branch in IPE for example, since you marked it done | 12:55 |
| *** k_mouza has quit IRC | 12:56 | |
| iurygregory | ack | 12:56 |
| * iurygregory is wondering if IG is enough or IGMF is required... | 12:58 | |
| dtantsur | iurygregory: no way, we'll confuse you with instagram! | 13:00 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: DNM try to inject the correct grub location https://review.opendev.org/758809 | 13:00 |
| rpittau | lol | 13:00 |
| iurygregory | dtantsur, hahaha | 13:01 |
| iurygregory | that's a good reason! | 13:01 |
| iurygregory | it's suspicious that instagram verified something in openstack :D | 13:01 |
| dtantsur | right? | 13:01 |
| dtantsur | although apparently facebook has something to do with us (I didn't listen carefully) | 13:02 |
| iurygregory | yup! | 13:02 |
| Qianbiao | Hello, may someone has a minute help me look at this error log: https://zuul.opendev.org/t/openstack/build/3a6e0d4ada8c447ebea26322d7f3c635 | 13:07 |
| *** k_mouza_ has quit IRC | 13:07 | |
| *** rloo has joined #openstack-ironic | 13:07 | |
| Qianbiao | It seems tox doc does not install driver requirements. How should this be fixed. | 13:07 |
| openstackgerrit | Merged openstack/metalsmith master: Set safe version of hacking https://review.opendev.org/758769 | 13:07 |
| anuradha1904 | Hi TheJulia, iurygregory: If the review is taking some time, can I simultaneously start with another issue? | 13:08 |
| dtantsur | Qianbiao: you need to make sure all regular tox jobs pass without the dependency installed | 13:09 |
| iurygregory | anuradha1904, I would try to fix the problems in the tests that the CI is reporting in your patch before trying to start something new | 13:09 |
| iurygregory | it will take some time to review things because we need to verify all the repos in Ironic after the gerrit problem | 13:09 |
| anuradha1904 | okay iurygregory: Thank you very much. Sure, no problem. | 13:10 |
| rpittau | TheJulia, arne_wiebalck, is there a zoom instance for the ironic operator feedback session? or something similarly interactive :) | 13:10 |
| iurygregory | rpittau, I just noticed that some links in the ironic-inspector tab are for the same URL lol | 13:10 |
| rpittau | mmm like ? | 13:11 |
| arne_wiebalck | rpittau: https://zoom.us/j/99815255534?pwd=aDQvVDZLN3UvRGt6c1VzemFoR0crUT09 | 13:11 |
| rpittau | oh Isee | 13:11 |
| rpittau | arne_wiebalck: thanks! | 13:11 |
| iurygregory | like 0001 the link is to 0000 | 13:11 |
| rpittau | iurygregory: yeah, copy-paste mistake | 13:11 |
| iurygregory | I'm updating =) | 13:11 |
| rpittau | thanks | 13:12 |
| iurygregory | just in case =) | 13:12 |
| Qianbiao | dtantsur i see, it seems i need to verify whether ibmc is None first. | 13:15 |
| *** KurtB has joined #openstack-ironic | 13:16 | |
| Qianbiao | dtantsur it's wierd that same code on ussuri and train both pass the CI. | 13:18 |
| openstackgerrit | Aija Jaunteva proposed openstack/ironic-specs master: Pre-defined system hw configuration in single step https://review.opendev.org/740721 | 13:20 |
| openstackgerrit | QianBiao Ng proposed openstack/ironic stable/ussuri: opt: Enhance old stable branches to use latest python-ibmcclient https://review.opendev.org/752006 | 13:24 |
| *** k_mouza has joined #openstack-ironic | 13:26 | |
| *** tzumainn has joined #openstack-ironic | 13:33 | |
| *** abdysn has quit IRC | 13:39 | |
| *** pmannidi has joined #openstack-ironic | 13:47 | |
| *** pmannidi_ has quit IRC | 13:49 | |
| *** k_mouza has quit IRC | 13:54 | |
| *** k_mouza has joined #openstack-ironic | 13:54 | |
| dtantsur | arne_wiebalck: what's your rough node/conductor numbers today? I need to impress a person :) | 14:00 |
| arne_wiebalck | dtantsur: 5000 nodes in 10 conductor groups, most groups with 1 conductor, some with 3 conductors (14 conductors) | 14:02 |
| dtantsur | arne_wiebalck: thx! do you know if there is a public link for your yesterday's scaling talk? | 14:05 |
| arne_wiebalck | dtantsur: hmm, good question ... I have not seen any, but guess the talks will make it to youtube at some point ... I can share the slide deck if that helps? | 14:06 |
| dtantsur | arne_wiebalck: no worries, I'll link to your older blog post | 14:07 |
| Qianbiao | dtantsur arne_wiebalck will the amount of nodes being a big chanllenge? I will delivery a cloud contains almost 1400 nodes (1350+ ramdisk, 50+direct) next month. | 14:09 |
| Qianbiao | currently, on my test lab, I did not consider anything about scale for now. | 14:09 |
| Qianbiao | will setup stack with kolla in prod env. Not sure how many conductor is suitable. | 14:10 |
| arne_wiebalck | it is mostly about how many in parallel you would like to deploy | 14:10 |
| arne_wiebalck | from my setup, you need a conductor per 60 parallel installations | 14:11 |
| dtantsur | Qianbiao: I'd recommend 3 conductors for good reliability. Should be more than enough for you. | 14:11 |
| Qianbiao | Hmmm, not sure for now. | 14:11 |
| dtantsur | but yeah, if you want to deploy really many nodes in parallel, you may need more conductors. | 14:12 |
| arne_wiebalck | we have 3 conductors and can install ~200 nodes in parallel | 14:12 |
| arne_wiebalck | I did not try with more in parallel | 14:12 |
| Qianbiao | dtantsur arne_wiebalck ok, gotcha. thanks. | 14:13 |
| arne_wiebalck | the conductor groups is mostly for the resource tracker in nova | 14:13 |
| dtantsur | but may help with DHCP as well | 14:13 |
| arne_wiebalck | it seems one can simply scale out Ironic | 14:13 |
| openstackgerrit | Merged openstack/sushy-cli master: Set safe version of hacking https://review.opendev.org/758439 | 14:13 |
| Qianbiao | dtantsur the different part is we are mostly deploy ramdisk, it should be light weight than direct deploy. | 14:13 |
| openstackgerrit | Verification of a change to openstack/networking-baremetal failed: Set safe version of hacking https://review.opendev.org/758784 | 14:14 |
| openstackgerrit | Verification of a change to openstack/networking-generic-switch failed: Set safe version of hacking https://review.opendev.org/758785 | 14:15 |
| openstackgerrit | Merged openstack/ironic-ui master: Set safe version of hacking https://review.opendev.org/758788 | 14:15 |
| openstackgerrit | Merged openstack/virtualbmc master: Set safe version of hacking https://review.opendev.org/758768 | 14:15 |
| openstackgerrit | Merged openstack/sushy-tools master: Set safe version of hacking https://review.opendev.org/758437 | 14:18 |
| *** w14161 has quit IRC | 14:21 | |
| *** sshnaidm is now known as sshnaidm|rover | 14:22 | |
| *** w14161 has joined #openstack-ironic | 14:26 | |
| *** mkrai has joined #openstack-ironic | 14:29 | |
| *** Qianbiao has quit IRC | 14:37 | |
| *** Qianbiao has joined #openstack-ironic | 14:37 | |
| *** jtomasek has joined #openstack-ironic | 14:41 | |
| *** jawad_axd has joined #openstack-ironic | 14:42 | |
| *** mkrai has quit IRC | 14:44 | |
| *** Qianbiao has quit IRC | 14:44 | |
| openstackgerrit | Merged openstack/networking-baremetal master: Set safe version of hacking https://review.opendev.org/758784 | 14:44 |
| openstackgerrit | Merged openstack/networking-generic-switch master: Set safe version of hacking https://review.opendev.org/758785 | 14:49 |
| *** mkrai has joined #openstack-ironic | 14:52 | |
| NobodyCam | Good Morning Folks, | 14:56 |
| *** alexmcleod_ has quit IRC | 15:02 | |
| iurygregory | nice zoom doesn't let me listen to what people are saying .-. | 15:02 |
| *** alexmcleod has joined #openstack-ironic | 15:07 | |
| *** alexmcleod has quit IRC | 15:09 | |
| openstackgerrit | Merged openstack/ironic master: Update test requirements https://review.opendev.org/758432 | 15:16 |
| openstackgerrit | Merged openstack/python-ironicclient master: Set safe version of hacking https://review.opendev.org/758442 | 15:16 |
| openstackgerrit | Verification of a change to openstack/ironic-lib failed: Set safe version of hacking https://review.opendev.org/758433 | 15:16 |
| openstackgerrit | Merged openstack/ironic-inspector master: Set safe version of hacking https://review.opendev.org/758774 | 15:16 |
| JayF | Thank you to the folks who did the auditing before I even was getting started this morning | 15:19 |
| *** akankshab has joined #openstack-ironic | 15:21 | |
| openstackgerrit | Merged openstack/ironic-tempest-plugin master: Set safe version of hacking https://review.opendev.org/758790 | 15:24 |
| openstackgerrit | Riccardo Pittau proposed openstack/python-ironic-inspector-client master: Set safe version of hacking https://review.opendev.org/758764 | 15:30 |
| *** alexmcleod has joined #openstack-ironic | 15:39 | |
| openstackgerrit | Riccardo Pittau proposed openstack/bifrost master: Set safe version of hacking https://review.opendev.org/758771 | 15:40 |
| openstackgerrit | Merged openstack/ironic-lib master: Set safe version of hacking https://review.opendev.org/758433 | 15:41 |
| *** k_mouza has quit IRC | 15:43 | |
| TheJulia | JayF: +++++++ | 15:54 |
| openstackgerrit | Merged openstack/bifrost master: Fix hacking min version to 3.1.0 https://review.opendev.org/753235 | 15:55 |
| * dtantsur is tired and needs a break | 15:55 | |
| *** dtantsur is now known as dtantsur|afk | 15:55 | |
| dtantsur|afk | o/ | 15:55 |
| *** pmannidi has quit IRC | 15:58 | |
| *** pmannidi has joined #openstack-ironic | 16:01 | |
| rpittau | good night! o/ | 16:04 |
| *** rpittau is now known as rpittau|afk | 16:04 | |
| *** k_mouza has joined #openstack-ironic | 16:15 | |
| *** tosky has quit IRC | 16:16 | |
| *** portdirect has quit IRC | 16:19 | |
| *** k_mouza has quit IRC | 16:19 | |
| *** portdirect has joined #openstack-ironic | 16:20 | |
| *** portdirect has quit IRC | 16:21 | |
| *** portdirect has joined #openstack-ironic | 16:21 | |
| *** k_mouza has joined #openstack-ironic | 16:26 | |
| *** k_mouza has quit IRC | 16:28 | |
| *** ayoung has joined #openstack-ironic | 16:28 | |
| *** dsneddon has joined #openstack-ironic | 16:28 | |
| ayoung | The docs for openstack baremetal create ./nodes.ipmi.json don't seem to say whether you can pass driver specific values. I know in early versions, you could pass in pm_address and stuff. Can I do that all up front, or do I need to do a separate set command to update each nodes IPMI values? | 16:30 |
| *** lucasagomes has quit IRC | 16:34 | |
| *** ociuhandu_ has joined #openstack-ironic | 16:39 | |
| *** jtomasek has quit IRC | 16:42 | |
| *** ociuhandu has quit IRC | 16:42 | |
| *** ociuhandu_ has quit IRC | 16:44 | |
| iurygregory | sad they didn't uploaded the new version I sent -.-' | 16:48 |
| *** jtomasek has joined #openstack-ironic | 16:48 | |
| TheJulia | iurygregory: :( | 16:49 |
| TheJulia | ayoung: if your thining pm_address, then that is a tripleo command | 16:50 |
| TheJulia | thinking | 16:50 |
| iurygregory | TheJulia, yeah =( the demo will be a bit strange because my face will be in the top left of the video =( | 16:50 |
| iurygregory | I sent a new version a week ago and they told me it would upload =( | 16:50 |
| *** jtomasek has quit IRC | 16:58 | |
| *** mkrai has quit IRC | 17:08 | |
| arne_wiebalck | bye everyone o/ | 17:08 |
| iurygregory | bye arne_wiebalck o/ | 17:08 |
| *** derekh has quit IRC | 17:18 | |
| *** sshnaidm|rover is now known as sshnaidm|afk | 17:24 | |
| *** dsneddon has quit IRC | 17:24 | |
| *** dsneddon has joined #openstack-ironic | 17:31 | |
| *** dsneddon has quit IRC | 17:36 | |
| *** dsneddon has joined #openstack-ironic | 17:37 | |
| *** dsneddon has quit IRC | 17:42 | |
| *** dsneddon has joined #openstack-ironic | 17:43 | |
| *** dsneddon has quit IRC | 17:49 | |
| *** dsneddon has joined #openstack-ironic | 17:49 | |
| TheJulia | my brain is like "I can has nap... or maybe flight through mountains with a challenging cross-wind" | 17:54 |
| *** dsneddon has quit IRC | 17:54 | |
| *** dsneddon has joined #openstack-ironic | 17:57 | |
| *** priteau has quit IRC | 18:01 | |
| *** dsneddon has quit IRC | 18:02 | |
| *** akankshab has quit IRC | 18:03 | |
| *** k_mouza has joined #openstack-ironic | 18:04 | |
| *** dsneddon has joined #openstack-ironic | 18:06 | |
| *** k_mouza has quit IRC | 18:08 | |
| *** gyee has joined #openstack-ironic | 18:09 | |
| ayoung | I had has nap. TheJulia I was actually referring to an OSP7 post where it had that, which might even predate tripleo. | 18:14 |
| *** iurygregory has quit IRC | 18:18 | |
| ayoung | I'm still learning this. So node create allows you to pass through driver options like this: | 18:18 |
| ayoung | node create --driver ipmi --management-interface ipmitool --power-interface ipmitool --driver-info ipmi_address=$IPMI_ADDR --driver-info ipmi_username=root -f value -c uuid | 18:18 |
| ayoung | But I can't do the same thing with the json import? | 18:19 |
| *** iurygregory has joined #openstack-ironic | 18:20 | |
| *** iurygregory has quit IRC | 18:24 | |
| TheJulia | nap +++++++++++++++++++++++++++ | 18:25 |
| *** iurygregory has joined #openstack-ironic | 18:25 | |
| TheJulia | ayoung: there is a json imput if memory serves, but you have to feed it a file with field smatching the node object structure or something like that | 18:25 |
| TheJulia | ayoung: my memory on this is super fuzzy | 18:25 |
| ayoung | Twould be nice to have a document | 18:26 |
| ayoung | I like the word smatching | 18:26 |
| ayoung | its like half mathc, half smash | 18:26 |
| ayoung | match | 18:26 |
| TheJulia | i thought we did | 18:29 |
| * TheJulia is out of brains today | 18:29 | |
| TheJulia | ayoung: https://docs.openstack.org/python-ironicclient/latest/user/create_command.html | 18:29 |
| ayoung | I was uing the ipmi driver, but I'm guessin I should actually be using pxe_ipmitool | 18:29 |
| * TheJulia now raises ENOBRAINS | 18:30 | |
| ayoung | Yeah, I have the doc, but it says nothing about how to reach the node via IPMI, other than to set that as the driver | 18:30 |
| TheJulia | OH! | 18:32 |
| TheJulia | heh, you need a driver_info field | 18:32 |
| TheJulia | ayoung: if you can extrapolate into your file then ++, otherwise give me 10 minutes and i'll post a chagne to those docs | 18:32 |
| *** stevebaker has joined #openstack-ironic | 18:33 | |
| ayoung | TYVM. You, as alway, rock. | 18:36 |
| TheJulia | pep8 running | 18:39 |
| TheJulia | this should work, I've not tried/used that command before but if memory serves it is just supposed to pass the data through | 18:39 |
| * TheJulia gives the laptop a piece of chocolate | 18:39 | |
| ayoung | I think laptops are like dogs...not good to give them chocolate. They prefer meat. | 18:40 |
| ayoung | SHould I be using the pxe_ipmitool driver, then? I can go add that. | 18:41 |
| TheJulia | doh | 18:41 |
| TheJulia | detached head off of queens | 18:41 |
| TheJulia | nooooo | 18:41 |
| openstackgerrit | Julia Kreger proposed openstack/python-ironicclient master: Update create node from file example https://review.opendev.org/759124 | 18:46 |
| TheJulia | ayoung: ^^^ | 18:46 |
| *** jawad_axd has quit IRC | 18:49 | |
| *** tosky has joined #openstack-ironic | 19:12 | |
| *** fgofurov has quit IRC | 19:13 | |
| *** k_mouza has joined #openstack-ironic | 19:55 | |
| *** k_mouza has quit IRC | 20:00 | |
| *** priteau has joined #openstack-ironic | 20:02 | |
| *** dhellmann has quit IRC | 20:12 | |
| *** ddurst has quit IRC | 20:15 | |
| *** ddurst has joined #openstack-ironic | 20:17 | |
| ayoung | TheJulia, +1 from me. TYVM | 20:20 |
| *** k_mouza has joined #openstack-ironic | 20:26 | |
| *** k_mouza has quit IRC | 20:31 | |
| *** mrda has quit IRC | 22:03 | |
| *** mrda has joined #openstack-ironic | 22:05 | |
| *** tosky has quit IRC | 22:10 | |
| *** tosky has joined #openstack-ironic | 22:11 | |
| *** ociuhandu has joined #openstack-ironic | 22:34 | |
| *** k_mouza has joined #openstack-ironic | 22:35 | |
| *** k_mouza has quit IRC | 22:39 | |
| *** ociuhandu has quit IRC | 22:39 | |
| *** dhellmann has joined #openstack-ironic | 22:45 | |
| *** rcernin has joined #openstack-ironic | 22:51 | |
| *** tosky has quit IRC | 22:51 | |
| *** rloo has quit IRC | 22:57 | |
| *** rh-jelabarre has quit IRC | 23:01 | |
| *** rh-jelabarre has joined #openstack-ironic | 23:03 | |
| *** rh-jelabarre has quit IRC | 23:04 | |
| *** rh-jelabarre has joined #openstack-ironic | 23:04 | |
| janders | good morning Ironic o/ | 23:16 |
| *** sai438 has joined #openstack-ironic | 23:31 | |
| *** pmannidi has quit IRC | 23:34 | |
| *** anuradha1904 has quit IRC | 23:42 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!