| opendevreview | Merged openstack/ironic-tempest-plugin master: Secure RBAC Test https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/842427 | 00:35 |
|---|---|---|
| rpittau | good morning ironic! o/ | 07:13 |
| dtantsur | rpittau: morning, could you check https://review.opendev.org/c/openstack/ironic-python-agent/+/879156 please? | 07:18 |
| dtantsur | I think this is the last bit that metal3 needs from us. | 07:18 |
| rpittau | dtantsur: I'll have a look at it now | 07:18 |
| rpittau | approved | 07:20 |
| dtantsur | \o/ thx | 07:20 |
| dtantsur | TheJulia: bifrost also uses md5, we broke it as well... | 07:22 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic-python-agent master: Revert disabling MD5 checksums https://review.opendev.org/c/openstack/ironic-python-agent/+/882236 | 07:28 |
| dtantsur | TheJulia, JayF, I'm sorry, but this ^^^ is the only reasonable path forward. | 07:28 |
| dtantsur | If Red Hat has issues with it, I'll be happy to clarify in an internal discussion. | 07:28 |
| dtantsur | TheJulia: re your question about raw conversion. I've realized it's only the case for image_download_source=local. | 07:36 |
| dtantsur | Neither bifrost nor metal3 set that. | 07:38 |
| opendevreview | Merged openstack/sushy master: Retry on ilo state error https://review.opendev.org/c/openstack/sushy/+/880542 | 08:49 |
| opendevreview | Merged openstack/ironic-python-agent master: Add network interface speed to the inventory https://review.opendev.org/c/openstack/ironic-python-agent/+/879156 | 09:04 |
| opendevreview | Verification of a change to openstack/ironic master failed: Support longer checksums for redfish firmware upgrade https://review.opendev.org/c/openstack/ironic/+/882163 | 09:22 |
| opendevreview | Verification of a change to openstack/ironic master failed: Support sha256/sha512 with the ilo firmware upgrade logic https://review.opendev.org/c/openstack/ironic/+/882164 | 09:22 |
| zigo | In my nova-compute.log, I get: | 09:31 |
| zigo | nova.compute.manager nova.exception.NoResourceClass: Resource class not found for Ironic node <node-uuid>. | 09:31 |
| zigo | What should I check to fix? | 09:31 |
| dtantsur | zigo: does the node have its resource_class field populated? | 09:35 |
| zigo | dtantsur: How do I check? The node is still in status "enroll" ... | 09:41 |
| zigo | s/status/Provisioning State/ | 09:41 |
| dtantsur | then you can probably ignore the error | 09:43 |
| dtantsur | but it's just a field, you can see it in $ baremetal node show | 09:43 |
| zigo | dtantsur: No, I can't. | 09:51 |
| zigo | FYI, that's under Zed. | 09:51 |
| zigo | Why isn't my node switching to Provisioning State: available ? | 09:52 |
| kaloyank | zigo: you have to move it to manageable with $ baremeatl node manage <node-name> | 11:37 |
| zigo | Ah, thanks. | 11:37 |
| kaloyank | then move it to available with $ baremetal node provide <node-name> | 11:37 |
| kaloyank | with the first operation Ironic will verify that it can control the machine via its management interface (IPMI, Redfish) | 11:37 |
| zigo | kaloyank: The IPMI password isn't set, I don't think Ironic has it. Will it make a login / pass itself? | 11:38 |
| kaloyank | and with the second it will try to "clean" it, that is prepare it to have instances scheduled on it | 11:38 |
| zigo | It's currently the default Dell thingy ... ( root / calvin ...). | 11:38 |
| kaloyank | zigo: no, it's required that you configure it, Ironic can't guess it | 11:38 |
| zigo | kaloyank: Isn't there is a way to make it set the login / pass randomly for me? | 11:39 |
| zigo | The node is then back to enroll, probably because of this. | 11:39 |
| dtantsur | How will it work with random credentials? | 11:40 |
| zigo | dtantsur: That's what I did on *MY* custom baremetal provisionner: simply, my controller ssh the node to provision, and runs "ipmitool user set 2 password <BLA> | 11:41 |
| dtantsur | At the very least, you need to somehow SSH into the node. | 11:41 |
| zigo | Yeah, which is why I have dropped a public key in my live-build Debian image ... | 11:42 |
| dtantsur | You can, of course, extend IPA to do it for you. Some time ago, I wrote a whole blog post on why we don't do it in Ironic https://owlet.today/posts/setting-ipmi-credentials-the-history/ | 11:42 |
| kaloyank | zigo: I don't get what you're frustrated about, it's perfectly fine to tell Ironic how to authenticate to the machines' BMC | 11:43 |
| zigo | dtantsur: IMO, this would be a job for IPA, indeed, and then it could report the login / pass when doing the hardware introspection. | 11:43 |
| zigo | kaloyank: Well, it's not automated, it's a manual procedure which I would like to avoid. | 11:43 |
| dtantsur | zigo: see the blog post for how you can do it via a site plugin (and why we won't do it) | 11:43 |
| * zigo reads | 11:44 | |
| zigo | Thanks for the link. | 11:44 |
| zigo | dtantsur: The "Channels and Numbers:" is IMO not a valid point, as mostly, user 2 and channel 1 works in most hardware, and we could have some kind of exceptions depending on vendors. | 11:47 |
| zigo | I have setup literally thousands of servers and it was never an issue. | 11:47 |
| zigo | "Error Handling" never was an issue for me either. As the PXE server ssh into the server after populating its db, if it fails, then the old server remains, just the PXE server db contains the wrong info, and later on, a "ipmi apply" can fix things. | 11:49 |
| zigo | Then for "Security", well, what's the issue if something pretends to be a server to provision, and we give it a random password? | 11:51 |
| zigo | You're doomed anyway if this happens... | 11:51 |
| kaloyank | zigo: I still don't get what's bothering you regarding Ironic | 11:51 |
| zigo | kaloyank: Let's say I have 36 servers to setup (real life experience: I already did this...), I don't want to manually log into each of them one by one to setup IPMI credentials, and inform Ironic about it. I need this to be automated. | 11:52 |
| dtantsur | zigo: that's a YMMV case. We've seen issues I describe there in practice. | 11:54 |
| kaloyank | IMO, it's basically a couple of for loops in bash but again, new contributions are always welcome, so feel free to hack around and submit your work :) | 11:55 |
| zigo | dtantsur: IMO, if it's controvertial, it's easy to fix: make the feature available, but disabled by default. | 11:55 |
| zigo | dtantsur: You probably will see me opening such a PR in the near future. | 11:55 |
| dtantsur | zigo: you can give it a go, I'd suggest presenting the design to the community first | 11:56 |
| zigo | dtantsur: I'm not there yet, I need my PoC to be fully functional first ! :) | 11:56 |
| zigo | Knowing there's no consensus, I'll make sure to write a spec for approval first then. | 11:57 |
| * zigo jumps into a (borring) meeting now... | 11:57 | |
| iurygregory | good morning Ironic | 12:02 |
| rpittau | mmmm seems the rbac enforced job is broken bad | 12:19 |
| rpittau | yep, very broken | 12:21 |
| iurygregory | may the force be with you rpittau =) | 12:22 |
| rpittau | iurygregory: may the Force be with us all! | 12:22 |
| iurygregory | yes! | 12:22 |
| rpittau | I see the most recent change in our tempest plugin is very recent https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/842427 | 12:25 |
| rpittau | TheJulia: probably need help from you for this | 12:26 |
| rpittau | yeah, looks like one of the new tests is failing hard test_reader_cannot_create_node | 12:27 |
| rpittau | I think the expected exception is not correct, shouldn't that be 403 ? | 12:34 |
| * iurygregory look | 12:34 | |
| rpittau | there are other 2 tests failing, most likely same problem, wrong exception, let's see if I can put together a quick patch | 12:35 |
| iurygregory | rpittau, humm yeah to me it makes sense that is 403 - forbidden | 12:35 |
| rpittau | I have a doubt on test_reader_cannot_get_indicator_state, the other two are clear | 12:41 |
| iurygregory | if it should be 403 also? | 12:44 |
| rpittau | it gives ServerFault but I'm not sure that's the correct exception, it may be the function that is wrong | 12:44 |
| opendevreview | Lana Kaleif proposed openstack/ironic master: Fix anaconda stage2_id loading from image properties https://review.opendev.org/c/openstack/ironic/+/882285 | 12:45 |
| iurygregory | we need to look at our policy code I think, but from the description of the test "Reader cannot get indicator state" sounds like a 403 instead of 404 | 12:45 |
| rpittau | mmm not sure, it should not be tehre, so 404 sounds correct | 12:47 |
| iurygregory | to me it sounds like the indicator might be there, but the reader can't get the information | 12:49 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-tempest-plugin master: Fix rbac tests https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882288 | 13:01 |
| rpittau | TheJulia and others please have a look ^ | 13:01 |
| dtantsur | rpittau: do you have an idea why it was not caught in the CI initially? | 13:05 |
| iurygregory | I was wondering the same thing ^ | 13:14 |
| rpittau | I don't see the tests running in the tempest plugin CI with the rbac enforced, that's why | 13:19 |
| rpittau | at leat looking at the change that enabled the tests https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/842427 | 13:20 |
| rpittau | am I wrong ? | 13:21 |
| * TheJulia is just starting to wake up | 13:23 | |
| TheJulia | ... I could have sworn those tests were getting run previously | 13:57 |
| TheJulia | but it looks like we only have scenario jobs on tempest now | 13:58 |
| TheJulia | no general testsing... | 13:58 |
| TheJulia | dtantsur: ack | 14:02 |
| TheJulia | w/r/t md5 | 14:02 |
| TheJulia | rpittau: got a link to the job where we noticed the tempest chagne breaking? The only guess I have is that somewhere between the change being started and now, we dropped the test from tempest itself?! | 14:02 |
| rpittau | TheJulia: here's one https://zuul.opendev.org/t/openstack/build/9eee93c64bc44a7d81fd390b470e4f1f | 14:03 |
| TheJulia | sweeet, thanks | 14:04 |
| TheJulia | ahh, it is just the flag | 14:13 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Advance tempest plugin tests to Zed https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882311 | 14:16 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Add RBAC specific tempest jobs to gate plugin https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882312 | 14:16 |
| TheJulia | if zed is good, I'll add 2023.1 | 14:17 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Advance tempest plugin tests to Zed https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882311 | 14:36 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Add RBAC specific tempest jobs to gate plugin https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882312 | 14:37 |
| TheJulia | hmmmmm | 15:31 |
| dtantsur | any objections if I approve https://review.opendev.org/c/openstack/ironic-python-agent/+/882236 then? I have no problems with phasing out md5, but I'm afraid we've rushed it a bit. | 15:32 |
| TheJulia | Approved, lets please actually do actually change that and not forget it next year | 15:33 |
| JayF | https://review.opendev.org/c/openstack/ironic-specs/+/881509 trivial, needs 1 more +2A | 15:36 |
| dtantsur | TheJulia: I'll drive the conversation on the Metal3 side if someone can take Bifrost | 15:36 |
| JayF | TheJulia: https://review.opendev.org/c/openstack/ironic-specs/+/874189 you wanna resolve nits here, or land it and you'll fix in a followup? | 15:37 |
| dtantsur | TheJulia, JayF, we also need to make some significant noise about the MD5 deprecation. I don't think we even had an ML announcement? | 15:44 |
| dtantsur | I'm afraid a lot of people keep using md5 just because it has always been there | 15:45 |
| JayF | dtantsur: TBH; I'm not sure I realized it was likely to be breaky, beyond upgrade level instructions, until recently | 15:45 |
| JayF | dtantsur: so I'm not certain what the best way to communicate this is | 15:45 |
| dtantsur | We cannot do much beyond release notes (already in place, but probably worth repeating in Ironic) and the ML | 15:46 |
| opendevreview | Merged openstack/ironic-specs master: Update spec template to reflect launchpad move https://review.opendev.org/c/openstack/ironic-specs/+/881509 | 15:56 |
| rpittau | good night! o/ | 16:11 |
| TheJulia | I guess the challenge is openstack as a whole moved past md5 checksums a long time ago | 16:19 |
| TheJulia | we lagged by continuing to support md5 explicitly | 16:19 |
| TheJulia | so to the openstack community as a whole, it seems like a moot topic since the enhanced checksum stuffs have been around for years | 16:20 |
| TheJulia | I guess I feel like we're just punting a can down the road and without a forced means of change/improvement we're going to be revisiting it again, and again, since we're allowing others to not have to change/evolve/improve their own situations and they can inherently just kick the can down the road | 16:27 |
| TheJulia | I think it was also first viewed as "an option to re-enable is acceptable" and then realizing metal3 has it as part of their public api, then we hit a "oh, well, crap" sort of situation | 16:28 |
| TheJulia | it has been a topic for a couple years, and now we're likely not going to remove it for another ?2? Granted, I don't care about removal as much as I do about disabling by default. At the end of the day I'll just shrug though. | 16:36 |
| JayF | dtantsur: I do wanna specifically point out: the problem here is not our deprecation (md5 can be re-enabled by operators or downstream packagers) but metal3 wanting to reuse our IPA deliverable | 16:38 |
| JayF | dtantsur: that to me makes it less of a clear deliniation of "Ironic broke an API" vs just two projects that intersect making different decisions w/r/t timing | 16:38 |
| dtantsur | s/metal3/anyone/ and s/our IPA deliverable/any IPA deliverable/ | 16:39 |
| dtantsur | now it's fair | 16:39 |
| TheJulia | the inherent challenge is we do control bifrost's code, we can fix stuff there fairly quickly | 16:39 |
| dtantsur | we *expect* people to use our IPA deliverables or someone else's like RDO | 16:39 |
| JayF | Metal3 and Bifrost are the only Ironic consumers that require MD5 right now; we can fix Bifrost. | 16:39 |
| dtantsur | [citation needed] | 16:39 |
| TheJulia | no we don't, the deliverables are published for testing onlly | 16:39 |
| JayF | I can't prove a negative. | 16:39 |
| dtantsur | TheJulia: that's a very different take from what I had in mind when I created these images | 16:40 |
| TheJulia | you can, but there is no warranty to the artifacts | 16:40 |
| JayF | This is not the best path to solving the problem; the images exist, many people use them regardless of what we think about them. | 16:40 |
| dtantsur | I would expect a lot of people to just casually set image_checksum because it's there | 16:40 |
| dtantsur | and because Red Hat's business around FIPS has nothing to do with them | 16:40 |
| JayF | but if you use the prebuilt images, you lose quite a bit of customization -- that's the tradeoff | 16:40 |
| JayF | that would include the ability to re-enable md5 | 16:40 |
| JayF | I'm not saying we need to flip the switch back today or tomorrow; I'm saying we need an actual plan and a timeline to do ito | 16:41 |
| dtantsur | We could just follow our standard and agreed deprecation process | 16:41 |
| JayF | and to consider that ironic and metal3 are not joined tightly enough to do many coordinated deprecations like this without significantly slowing up both projects | 16:41 |
| TheJulia | dtantsur: complicated by the tick/tock impact now | 16:41 |
| dtantsur | Hold on please. It's not just metal3. It's possiblity THE MAJORITY of our consumers | 16:41 |
| dtantsur | You keep ignoring the fact that absolutely nothing has motivated people to migrate away from MD5 before this Monday. | 16:42 |
| TheJulia | Anyone using a recent glance would have just worked since the extra checksums have been around for quite a long time | 16:42 |
| dtantsur | Okay, the majority of non-Glance customers :) | 16:42 |
| JayF | dtantsur: So I put metal3 in a different bucket here; bceause AIUI there's an API tie-in there. The usual operator experience would involve needing to update the checksums in their images in OpenStack's API, which is hefty but not out of the norm for an upgrade requirement | 16:42 |
| TheJulia | And we've provided the pass-through, just nobody adopted it | 16:42 |
| TheJulia | the underlying issue is, we never added support for the extended use of the checksum field until now | 16:43 |
| TheJulia | so, there is an easier path now | 16:43 |
| dtantsur | Yep, metal3 is kinda special, but we cannot assume how much effort it is for the operators to change | 16:43 |
| TheJulia | so... maybe that will help | 16:43 |
| JayF | dtantsur: I like the lens you're using; but I struggle with how to quantify that | 16:43 |
| JayF | dtantsur: and making decisions for a spooky mysterious operator I can't quantify is really difficult without just defaulting to 'brakes on at all times' | 16:43 |
| dtantsur | I routinely talk to operators who are overworked, underpaid, and have to deal with a whole cloud alone | 16:43 |
| dtantsur | Your exposure may be to people of Yahoo scale | 16:44 |
| TheJulia | dtantsur: that is an unfortunate but common theme | 16:44 |
| dtantsur | yep | 16:44 |
| JayF | This would be a pain the the rear for any environment I've worked in dtantsur :-( | 16:44 |
| TheJulia | things cannot be painless forever | 16:44 |
| dtantsur | So when you start with "Hey, we broke you.." they may smash the monitor before you get to ".. but it's easy to fix" :) | 16:45 |
| JayF | dtantsur: FWIW; I've never worked on an openstack cloud, regardless of scale, that had enough people to maintain it properly | 16:45 |
| TheJulia | pain needs to be in logical chunks which can be navigated | 16:45 |
| TheJulia | otherwise techdebt just gets piled on | 16:45 |
| dtantsur | Right. I'm by no means advocating to never remove MD5 or anything. FWIW, I've removed the iSCSI deploy :) | 16:45 |
| JayF | So how would you all feel about me marking this as a topic for Vancouver PTG? | 16:45 |
| dtantsur | But if our own removal has caught ourselves by surprise.. it's a bad sign. | 16:45 |
| * JayF suspects dtantsur would not be there and it might not be super useful | 16:46 | |
| TheJulia | we also gave iscsi deploy an exceptionally long time on a boat before being pushed into the water and lit ablaze :) | 16:46 |
| dtantsur | JayF: I won't be there, but if you have an ops feedback session, it's good to mention the MD5 situation there | 16:46 |
| dtantsur | TheJulia: yep, I'm not advocating for exercising so much care as with iSCSI. | 16:46 |
| TheJulia | can we get a graphic of me being an assassin chasing after md5 ?!? | 16:46 |
| dtantsur | But even we ourselves need some graceful period. | 16:46 |
| dtantsur | OH | 16:46 |
| * TheJulia wants artwork! | 16:46 | |
| dtantsur | We SO MUCH need an artist here | 16:47 |
| JayF | It sounds like, if we are going this route, we might want to implement the configure-it-via-conductor path | 16:47 |
| JayF | because it sounds like we're going to want to give people a more easy freedom to flip it on/off | 16:47 |
| TheJulia | JayF: already proposed, needs to be revised now since it forces a setting | 16:47 |
| JayF | ack | 16:47 |
| JayF | that to me makes this all... less interesting | 16:47 |
| JayF | because regardless of which way we fall; it's a single conductor config value to change behavior | 16:47 |
| dtantsur | It's temporary. We have lived with MD5 for a long time, we can live with it for 6 months more. | 16:48 |
| TheJulia | well, API lookup pass reply, but yeah | 16:48 |
| TheJulia | dtantsur: year+ | 16:48 |
| TheJulia | we can't remove it until 2024 under tick/tock | 16:48 |
| dtantsur | Depends on whether you want to respect the tick-tock thingy. | 16:48 |
| TheJulia | yeah | 16:48 |
| TheJulia | true | 16:48 |
| TheJulia | I don't really want to | 16:48 |
| dtantsur | *I* personally don't have any stakes there as long as I have a few months to adapt Metal3 and someone fixes Bifrost. | 16:48 |
| TheJulia | but I'm fine having an option and actually not explicitly removing as long as there is a knob that ends up with default false one day | 16:49 |
| dtantsur | once it's false, we can plan eventual removal | 16:49 |
| TheJulia | but first... why does grenade hate us | 16:49 |
| * TheJulia files this next to "why does god need a starship" | 16:49 | |
| dtantsur | yep. and the inspector grenade seems borked too | 16:49 |
| TheJulia | so, in general, it *looks* like we start the network validation, but then things go sideways | 16:50 |
| TheJulia | we don't get back to re-querying the image ref data so we definitely die somewhere in there | 16:50 |
| TheJulia | just no debugging and no exception | 16:50 |
| dtantsur | To the summit topic, I don't expect budget considerations to allow a lot of traveling for me in the near future. | 16:50 |
| dtantsur | So unless it's somewhere close... we won't have a chance to discuss md5 over a shot of whiskey any soon | 16:50 |
| TheJulia | Well, first we need glasses, neat or on the rocks | 16:51 |
| TheJulia | :) | 16:51 |
| TheJulia | It is looking like my travel is also going to be highly constrained | 16:52 |
| dtantsur | Prosperous times are over in the cloud world | 16:52 |
| JayF | TheJulia: dtantsur: We do not have the choice to ignore SLURP. First of all; how can we complain about operators not having enough time AND want to skip something designed to make their life easier in the same convo? | 16:54 |
| dtantsur | Well, we started with "let's switch it off effective immediately" | 16:54 |
| JayF | TheJulia: dtantsur: but on top of that; we are an OpenStack project, TC resolutions say we need to respect it, so we should regardless of personal opinions. | 16:54 |
| JayF | dtantsur: well, if we're going to go the route of slow we should be properly-slow. I didn't realize such a change was as disruptive/controversial as it seems to be so my opinion is softening :) | 16:55 |
| TheJulia | The US is starting to look like https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fabdc3765-35e2-4382-8a27-143484c06ba1_1610x998.png and | 16:55 |
| TheJulia | https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5ed1d68f-3552-41b8-b726-1b72d8285b71_1582x984.png for me | 16:55 |
| TheJulia | so travel moving forward is going to be... difficult even if there is budget | 16:56 |
| dtantsur | oh yeah. sorry to hear :( | 16:56 |
| TheJulia | c'est la vie | 16:56 |
| TheJulia | interesting, there seems to be a multi-second lag between the logs | 17:02 |
| TheJulia | maybe that might be why they are so confusing | 17:02 |
| opendevreview | Verification of a change to openstack/ironic-lib master failed: Add jsonrpc client port capability https://review.opendev.org/c/openstack/ironic-lib/+/879211 | 17:16 |
| opendevreview | Julia Kreger proposed openstack/ironic master: DNM: Attempting to torubleshoot grenade https://review.opendev.org/c/openstack/ironic/+/882347 | 17:40 |
| TheJulia | Worth a try.... next step is adding debug logging to nova | 17:41 |
| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Advance tempest plugin tests to Zed (mostly) https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/882311 | 18:00 |
| JayF | Borderline-off-topic: I'm working on tshirt/sticker ideas around the theme of "Ironic Bear Metal Antelope Tour 2023.1" -- does this fit the bill? https://cdn.discordapp.com/attachments/306957680734371840/1103752466538975262/TastyMcRib_a_bear_playing_guitar_in_front_of_a_crowd_of_antelop_8516d96f-e297-4e0d-a488-801e2e945e04.png | 18:39 |
| JayF | Oh, we need to add a grenade-skip-level job | 18:58 |
| JayF | I think iurygregory had a PR up for that a cycle early | 18:58 |
| JayF | it'd be interesting to see if it fails similarly to the grenade job now | 18:58 |
| sschmitt | Anyone here try running the networking-baremetal ml2 driver along with another ml2 driver and vlan-transparency? Seems like since networking-baremetal doesnt specifically support this, it prevents any network from being created with vlan-transparency | 19:33 |
| TheJulia | JayF: dunno, I'm not a fan of big prints | 20:17 |
| JayF | to get the colors right, it'd likely have to be stickerified | 20:18 |
| TheJulia | yeah | 20:18 |
| JayF | WDYT about something along those lines (I've been trying, not so successfully, to refine it) for a sticker? | 20:26 |
| TheJulia | could work. Anyway we could get a more metaly guitar? ;) | 20:32 |
| JayF | do you have discord? I can invite you to my discord and you can see all the cursed things I've generated trying to improve that LOL | 20:32 |
| TheJulia | I do not | 20:33 |
| TheJulia | (have discord that is) | 20:33 |
| JayF | https://media.discordapp.net/attachments/306957680734371840/1103757136057610260/TastyMcRib_a_bear_and_an_antelope_playing_on_dueling_electric_g_7f1e1ef7-c1ba-4dcb-be64-51ec6e1205c6.png being an example lol | 20:33 |
| JayF | also https://media.discordapp.net/attachments/306957680734371840/1103755810955341946/TastyMcRib_a_bear_playing_guitar_in_front_of_a_crowd_of_antelop_bbe1693f-dbf2-4fdb-b468-dfb2427ec8e4.png if I could touch up the eyes | 20:33 |
| * TheJulia wonders where a third arm came from... | 20:33 | |
| JayF | midjourney doing midjourney things lol | 20:34 |
| TheJulia | heh | 20:34 |
| NobodyCam | hahahahaha why is inspect wait in this list??? | 20:44 |
| NobodyCam | `Node <UUID> can not have management_interface updated unless it is in one of allowed (enroll, inspecting, inspect wait, manageable, available) states or in maintenance mode. (HTTP 409)` | 20:44 |
| NobodyCam | and inspecting | 20:44 |
| NobodyCam | seem odd to allow update in those states | 20:45 |
| NobodyCam | hehehe any who Good AfterNoon Ironic Folks | 20:46 |
| TheJulia | because you can have introspection rules to change settings | 20:48 |
| NobodyCam | ahh | 20:49 |
| iurygregory | yay, I'm getting "We are sorry! Your provided token seems to be void. Please check your email inbox for a newer one." | 22:30 |
| iurygregory | .-. | 22:30 |
| JayF | From where? :( | 22:31 |
| iurygregory | the forum submission was accepted, so in the email in step one, they ask me to confirm .-. | 22:31 |
| JayF | ah | 22:31 |
| JayF | #openinfra-events might be able to help | 22:32 |
| iurygregory | since now I have my flight tickets, I was going to confirm things :D | 22:32 |
| JayF | woo freakin hoo! | 22:32 |
| iurygregory | no hotel yet, but not worried about it :D | 22:32 |
| iurygregory | I just sent an email to speakersupport | 22:32 |
| clarkb | iurygregory: I've been told the see the email and will be in touch (probably tomorrow?) | 22:37 |
| iurygregory | clarkb, tks! | 22:38 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
