| guiverc | the -proposed package made it to ISO; my testing checklist assumed something different.. | 00:34 |
|---|---|---|
| guiverc | @lynorian, did you see anything helpful if you tried `xscreensaver-demo --debug` to launch ? | 00:38 |
| * guiverc will amend testing checklist as required, once I've confirmed boot from media (writing now). (ps. I've got headache (3rd day in a row) so not working fast) | 00:47 | |
| guiverc | thanks arraybolt3[m] for all the testing I now see :) (having finally opened iso.qa.ubu...)... I'll copy to testing checklist (when I get to that) | 00:58 |
| arraybolt3[m] | guiverc: Thanks, it's been fun! | 00:58 |
| guiverc | :) | 00:59 |
| arraybolt3[m] | guiverc: (I'm juuust about to add another Ubuntu Studio test to the list, so don't copy things over quite yet.) | 00:59 |
| guiverc | I'm likely to massage https://phab.lubuntu.me/w/release-team/testing-checklist/ given the proposed calamares is on ISO (my checklist edit assumed it needed to be added; needs wording change)... but just started booting thumb-drive with ISO on it for confirmation of what I saw on manifest noted earlier | 01:01 |
| guiverc | ubuntu studio adds would be another page of iso.qa.ubuntu.com anyway; not the lubuntu page | 01:02 |
| arraybolt3[m] | guiverc: The new Calamares is on the ISO, but the new calamares-settings-ubuntu is not, so if you're doing btrfs testing, you will need to install calamares-settings-lubuntu (or calamares-settings-ubuntustudio depending on the distro you're testing). | 01:02 |
| guiverc | thank you for telling me.. I'll skip BTRFS today then (will hopefully be there tomorrow..) | 01:04 |
| guiverc | i see many entries in Ubuntu Studio now too; well done Aaron / arraybolt3[m] | 01:04 |
| arraybolt3[m] | (I already did BTRFS testing so it's probably not necessary that we do any more of that.) | 01:04 |
| guiverc | it is once the calamares-settings-lubuntu hits the ISO and no manual fudging is required.. even if only a 'smoke-test', it still needs test & to pass with NO package CHANGES made prior to install | 01:05 |
| arraybolt3[m] | Ah, that makes sense. But the package got verification-done on it, so it should show up in the ISO. | 01:06 |
| arraybolt3[m] | I mean, eventually it should show up. | 01:06 |
| guiverc | yeah i can confirm older calamares-settings-lubuntu & older calamares-settings-ubuntu-common on the ISO i'm testing | 01:07 |
| guiverc | I've dropped calamares version column (it's on ISO so no need in my opinion) ; still listed as a second table though (easier if we need to revert though that would required retesting everything anyway so maybe pointless..) | 01:25 |
| * guiverc still not added yours arraybolt3[m], but in my to-do | 01:25 | |
| * guiverc might reverse tables; so proposed is first as that's what will currently be edited/changed.. | 01:27 | |
| lubot | [telegram] <lynorian> guiverc it still segmenation faults should I comment on the bug | 01:27 |
| guiverc | I'd have expected segfault, but I hoped it might provide more clues/ideas as to issue (even if only where to go next)... you really need a dev, not me | 01:29 |
| guiverc | arraybolt3[m], fyi: I could kick of another ISO build; we're already using 20220728.1 so we'd likely get 22020728.2, but ISO respins take awhile to build | 01:57 |
| guiverc | (rebuild so as to hopefully get the missing -proposed packages on the iso..) | 01:59 |
| arraybolt3[m] | guiverc: I don't see any reason to - the only change in the missing package is the BTRFS install bug fix (one option dropped in two files), and it's already been verified to work. | 02:01 |
| arraybolt3[m] | On both Lubuntu and Ubuntu Studio. | 02:01 |
| arraybolt3[m] | So the one test case it affects has already been entirely done. | 02:01 |
| guiverc | :) I don't currently either, why I didn't kick one before... we've got tomorows being my thinking; I've still got plenty to test for today (more than time will allow!) | 02:02 |
| arraybolt3[m] | 👍️ | 02:02 |
| guiverc | :) & thank you for your testing (again) | 02:02 |
| Eickmeyer[m] | arraybolt3: I'm doing a standard installation test of Ubuntu Studio in a VM now. | 02:46 |
| arraybolt3[m] | Nice! | 02:47 |
| arraybolt3[m] | I should probably get back to testing pretty soon. | 02:47 |
| Eickmeyer[m] | I especially like the new log view. I don't remember that, and it's super handy. | 02:47 |
| arraybolt3[m] | Nice, I've not noticed it yet. | 02:47 |
| Eickmeyer[m] | So, I'm not sure if it's because my computer is insanely fast or what, but my test is already done without issues. | 02:50 |
| arraybolt3[m] | ?! Wow, that's gotta be that you have a good computer. Each test is taking me 20-30 minutes. | 02:51 |
| Eickmeyer[m] | It's a 6-core i7 gen 9 with 2 threads per core. | 02:51 |
| Eickmeyer[m] | 64 MB RAM | 02:51 |
| arraybolt3[m] | *GB | 02:52 |
| Eickmeyer[m] | GB yes | 02:52 |
| Eickmeyer[m] | All SSD | 02:52 |
| arraybolt3[m] | I was gonna say, if you only have 64 MB of RAM, you need a new computer. | 02:52 |
| arraybolt3[m] | Mines a 3rd gen i5, 16 GB, with a SATA SSD. | 02:52 |
| arraybolt3[m] | The other one is a 3rd or 4th gen Xeon, 32 GB, SATA SSD. | 02:53 |
| Eickmeyer[m] | Yeah, there's the differentiating factor. | 02:53 |
| arraybolt3[m] | Hey, it works, that's what matters. | 02:53 |
| Eickmeyer[m] | Yep. | 02:53 |
| Eickmeyer[m] | I'm going to run another one with some partitioning for kicks and giggles. | 02:54 |
| lubot | [telegram] <lynorian> I think actually write speed matters a lot for installs | 02:54 |
| arraybolt3[m] | There's a lot of uncompressing work involved too. | 02:54 |
| Eickmeyer[m] | @lynorian Absolutely. My SSDs are both NVMe with insane write speeds, but yes, there's decompression involved and some processing with VMs, so CPU does have some factor. | 02:55 |
| arraybolt3[m] | My bargain-basement SATA SSDs probably aren't helping then. (I've got one NVMe but I don't use it for write-heavy stuff since I paid good money for that and don't want to destroy it.) | 02:55 |
| Eickmeyer[m] | Ok, installation #2 (install alongside) running now. I expect this one to go slower as it'll have to repartition. | 02:58 |
| Eickmeyer[m] | Nevermind, repartitioning done and unpacking. 👀 | 02:58 |
| arraybolt3[m] | Got another two tests started. | 03:01 |
| arraybolt3[m] | You're using a 1st gen Kubuntu Focus, right? | 03:01 |
| Eickmeyer[m] | Yep. | 03:01 |
| Eickmeyer[m] | You're gonna hate me. It's already done installing and it's at the login screen. 👀 | 03:03 |
| arraybolt3[m] | Hey, we're getting stuff done! | 03:03 |
| Eickmeyer[m] | True! | 03:04 |
| arraybolt3[m] | My tests are just finishing up. | 03:04 |
| Eickmeyer[m] | Yeah. So, here's my verdict, and I'll post this on the SRU bug: I don't see any real installation difference whatsoever and everything is working exactly as it did before, perhaps better. If anything, I see substantial performance improvements, which I consider a win. | 03:07 |
| arraybolt3[m] | I also notice nothing different so far. | 03:08 |
| Eickmeyer[m] | But, arraybolt3 , I'll leave it to you to change verification-needed-jammy and verification-needed to verification-done-jammy and verification-done respectively. | 03:08 |
| arraybolt3[m] | OK. That may take a while since I'm attacking the whole checklist... twice. | 03:08 |
| Eickmeyer[m] | Yeah, that's fine. You're being much more thorough than I am. Mine was more of a smoke test, but I'm pleasantly surprised with the performance improvements I'm seeing. | 03:09 |
| guiverc | i noted 1904266 no longer occurs with it (of no importance though) | 03:09 |
| guiverc | (cursor jumps to end of field.. http://launchpad.net/bugs/1904266) | 03:10 |
| ubot93 | Launchpad bug 1904266 in calamares (Ubuntu) "calamares data entry; cursor jumps to end of field making editing difficult" [Undecided, Confirmed] | 03:10 |
| arraybolt3[m] | Woot! | 03:10 |
| Eickmeyer[m] | We can definitely mark that as fix released when it hits updates. | 03:11 |
| Eickmeyer[m] | In fact, we can already go "Fix Committed" on Jammy and "Fix Released" in Kinetic. | 03:12 |
| * arraybolt3[m] does that real quick | 03:12 | |
| Eickmeyer[m] | Already beat you. :) | 03:12 |
| arraybolt3[m] | Ah, OK. | 03:13 |
| arraybolt3[m] | Man, you're fast at everything. | 03:13 |
| Eickmeyer[m] | Not the first time I've heard that. It gets me in trouble sometimes, because I do stuff thoroughly, but I just work faster than others. | 03:14 |
| arraybolt3[m] | I think it's good. | 03:14 |
| arraybolt3[m] | I try to work fast, but I a lot of times end up multitasking in the process, which makes me slow down for each individual task. | 03:15 |
| arraybolt3[m] | Like, right now I'm wrangling three computers at once. | 03:15 |
| Eickmeyer[m] | People often assume that because I work faster than they can comprehend sometimes that I'm taking shortcuts when, in fact, I'm not. | 03:15 |
| arraybolt3[m] | You just got fast hardware, fast Internet, and good focus. | 03:16 |
| arraybolt3[m] | I have slower hardware, slower Internet, and am scatterbrained a lot of times. | 03:16 |
| Eickmeyer[m] | Dude, I have ADHD like you wouldn't believe. | 03:16 |
| Eickmeyer[m] | Adult-onset, mind you. | 03:17 |
| Eickmeyer[m] | guiverc: Good catch. | 03:17 |
| arraybolt3[m] | Oy. Doesn't seem to slow you down, thankfully. | 03:17 |
| tsimonq2 | Eickmeyer[m]: I've been given quite a few labels in my day and they're just that... labels. I find them limiting | 03:17 |
| tsimonq2 | You were created as you, not ADHD. It's why I don't even mention what I have anymore, cause if you pay attention close enough you can figure it out | 03:19 |
| Eickmeyer[m] | tsimonq2: Tjat | 03:19 |
| Eickmeyer[m] | typo | 03:19 |
| Eickmeyer[m] | That's a very good point. | 03:19 |
| arraybolt3[m] | Oh lovely. Somehow turning on Secure Boot on my desktop just made even the BIOS not want to use the nVidia card. | 03:20 |
| tsimonq2 | nvidia lol | 03:20 |
| Eickmeyer[m] | Ouch. | 03:20 |
| tsimonq2 | Eickmeyer[m]: yeah cause that's Nvidia | 03:20 |
| tsimonq2 | lmao | 03:20 |
| Eickmeyer[m] | Well, to be fair, the Kubuntu Focus M-series come with Nvidia, but we have secure boot disabled by default for that very reason. But, we also make it STUPID SIMPLE to get Nvidia working. | 03:21 |
| arraybolt3[m] | You'd think it would at least let the BIOS display something though. | 03:22 |
| arraybolt3[m] | Wow. Even the default Intel graphics aren't doing anything. | 03:22 |
| arraybolt3[m] | So now I have to remove the nVidia card entirely - and if you saw what I had to do to get it in, you'd know why I do not want to do that... | 03:23 |
| Eickmeyer[m] | That sounds more serious. We found out that if you have integrated graphics and you disable the Nvidia chipset altogether, the external video ports don't even work because the Nvidia chipset "owns" the ports. | 03:23 |
| arraybolt3[m] | (Got a full-size card for a small form factor desktop. Long story short, wire cutters = card fits.) | 03:23 |
| arraybolt3[m] | That's what I was afraid of. Ah well, I'll take the card out and hopefully everything will just work. | 03:24 |
| Eickmeyer[m] | We did also find out that the integrated graphics "own" the internal display via the framebuffer, so the Nvidia graphics actually project to the internal display via the framebuffer via integrated graphics. Interesting, eh? | 03:25 |
| arraybolt3[m] | There we go, monitor now works. | 03:26 |
| arraybolt3[m] | Whew, that was a bit distressing. | 03:26 |
| arraybolt3[m] | (On my desktop, plugging an nVidia card in makes the whole integrated graphics part stop working entirely. Won't even work in Linux itself.) | 03:27 |
| tsimonq2 | Kernel bug? | 03:27 |
| arraybolt3[m] | I figured the BIOS probably just kaiboshed it. | 03:27 |
| arraybolt3[m] | (Or however you spell that word.) | 03:28 |
| arraybolt3[m] | It's an HP workstation that specifiically supports certain Linux distros, so I'd be surprised if it was a Linux incompatibility with the system itself. | 03:28 |
| tsimonq2 | arraybolt3[m]: I can hear the word in my head I get it lol :) | 03:28 |
| tsimonq2 | arraybolt3[m]: Or perhaps just a kernel bug | 03:29 |
| Eickmeyer[m] | Sometimes one just needs to rebuild the initramfs. | 03:29 |
| arraybolt3[m] | Hmm. I guess I could boot up a distro from the age it supports (like 12.04 or something silly like that). | 03:30 |
| Eickmeyer[m] | arraybolt3[m]: 🤮 | 03:30 |
| tsimonq2 | I mean, I use the Ubuntu Kernel Team's mainline-crack repo for lean, current kernels on my local machine... | 03:30 |
| arraybolt3[m] | I could test both. If it works in <insert ancient distro> and doesn't work in the latest 22.04 kernel, that's a regression. | 03:31 |
| arraybolt3[m] | (And then I'll get the lovely experience of doing a git bisect - oh... joy...) | 03:31 |
| tsimonq2 | arraybolt3[m]: Time for you to git bisect the Linux kernel, ready for some fun?!?!? | 03:31 |
| Eickmeyer[m] | arraybolt3[m]: I did that once under the direct supervision of Linus Torvalds and Greg K-H themselves. | 03:31 |
| arraybolt3[m] | I wonder how hard it will be to bisect the kernel when you've got the distance between 12.04 to 22.04. | 03:32 |
| arraybolt3[m] | (I'd probably have to install the 22.04 kernel on 12.04 since part of me really doubts that 12.04's kernel will boot 22.04.) | 03:32 |
| tsimonq2 | arraybolt3[m]: Very difficult. You'd have to try every LTS in between | 03:32 |
| arraybolt3[m] | Oh that's a smart idea. | 03:33 |
| arraybolt3[m] | (Would testing regular releases also be helpful in there, or would it have to be strictly LTS-to-LTS?) | 03:33 |
| arraybolt3[m] | (And is this even a bug worth chasing?) | 03:33 |
| tsimonq2 | Eickmeyer[m]: That's def cap bro | 03:34 |
| tsimonq2 | No way | 03:34 |
| tsimonq2 | Pics or it didn't happen | 03:34 |
| Eickmeyer[m] | Well, I'd have to go find the emails... | 03:34 |
| tsimonq2 | arraybolt3[m]: LTS to LTS should be ok | 03:34 |
| tsimonq2 | arraybolt3[m]: Up to you man :) | 03:35 |
| Eickmeyer[m] | But it was in my openSUSE days when a kernel update borked my computer at the time. Each bisect took 90 minutes on that single kernel machine. | 03:35 |
| Eickmeyer[m] | *core | 03:35 |
| Eickmeyer[m] | s/kernel/core | 03:35 |
| Eickmeyer[m] | It's back when Greg K-H was still the maintainer of openSUSE Tumbleweed. | 03:36 |
| tsimonq2 | Interesting | 03:36 |
| Eickmeyer[m] | Eventually we found the commit, and it ended up being a regression caused by some dudes at Intel. | 03:36 |
| arraybolt3[m] | Anyone got any clue how to enable Secure Boot in an EFI VM? I selected the Secure Boot OVMF image, but I can't seem to figure out how to tell it to turn on. | 03:38 |
| arraybolt3[m] | (And if you're wondering why on earth I want Secure Boot, it's part of the testcases.) | 03:38 |
| tsimonq2 | Ancient tribal method. Second cave to the left, you shall pass four councils who will determine your fate | 03:39 |
| Eickmeyer[m] | arraybolt3[m]: Yeah, I don't think you can. Unfortunately, Secure Boot keys are proprietary. | 03:41 |
| arraybolt3[m] | They told me to run rm -rf /. | 03:41 |
| arraybolt3[m] | I see there's instructions on how to online, Debian instructions nonetheless. | 03:41 |
| Eickmeyer[m] | Oh! TIL! | 03:42 |
| arraybolt3[m] | I just had the wrong firmware selected. | 03:42 |
| arraybolt3[m] | Yep, that worked. Use OVMF_CODE.ms.fd or something like that. | 03:43 |
| arraybolt3[m] | (The ms part was the important part.) | 03:43 |
| arraybolt3[m] | And crud. I just got a failed Studio installation with Secure Boot on the desktop. | 03:44 |
| arraybolt3[m] | Let's hope it was a transient error. | 03:44 |
| arraybolt3[m] | Nope. Crud! Eickmeyer Installs of Ubuntu Studio are failing repeatedly with a simple no encryption, internet-enabled Secure Boot installation. | 03:56 |
| arraybolt3[m] | I'll get the error message in a bit... | 03:57 |
| Eickmeyer[m] | On real hardware? | 03:57 |
| arraybolt3[m] | (I'm waiting to see if Lubuntu is going to do the same thing or not.) | 03:57 |
| arraybolt3[m] | Yes, on real hardware. | 03:57 |
| arraybolt3[m] | HP Z220 SFF Workstation, baremetal install with Secure Boot enabled. | 03:57 |
| Eickmeyer[m] | Yikes, I've got its cousin sitting in back of me (a Z220 CMT) acting as a server. | 03:58 |
| Eickmeyer[m] | I've had anecdotal reports of secure boot failing before. | 03:59 |
| arraybolt3[m] | Looks like Lubuntu is behaving the same way in a VM. | 03:59 |
| arraybolt3[m] | It hangs forever on a contextual processes job, then crashes. | 03:59 |
| arraybolt3[m] | Oh wait, Lubuntu just got past it. | 04:00 |
| arraybolt3[m] | OK, that will definitely be something to test. | 04:00 |
| Eickmeyer[m] | Weird. Is this with cala-settings-ubuntu updated? | 04:00 |
| arraybolt3[m] | No. | 04:00 |
| Eickmeyer[m] | There's a huge factor to look at. | 04:00 |
| arraybolt3[m] | Aaaaand - I just accidentally closed the error message in Studio. 🤦♂️ | 04:01 |
| * arraybolt3[m] digs up logs | 04:01 | |
| arraybolt3[m] | The whole log is: | 04:02 |
| arraybolt3[m] | Oh lovely. The new Calamares hardly generates any installation logs. | 04:02 |
| arraybolt3[m] | === START CALAMARES 3.2.60 | 04:03 |
| arraybolt3[m] | 2022-07-28 - 22:56:13 [1]: void Calamares::ViewManager::onInstallationFailed(const QString&, const QString&) | 04:03 |
| arraybolt3[m] | ERROR: Installation failed: "External command failed to finish." | 04:03 |
| arraybolt3[m] | That's the entire log file. Eickmeyer ^ | 04:03 |
| Eickmeyer[m] | External command??? | 04:03 |
| arraybolt3[m] | Lubuntu on the other hand seems to have installed successfully. | 04:03 |
| arraybolt3[m] | It had something to do with gpgv or something - I was going to type it across but I set my Chromebook on the keyboard and poof! | 04:03 |
| arraybolt3[m] | So I guess I'll try to reinstall and get the message again. | 04:04 |
| arraybolt3[m] | The only difference in the new cala-settings is the BTRFS install bug fix - nothing else was changed, so I don't think it will matter much. | 04:04 |
| arraybolt3[m] | I'll also run Calamares in insane verbose mode so I can get the logs it decided to not give me. | 04:06 |
| arraybolt3[m] | Simon Quigley (Developer) Eickmeyer Just the absence of logs by default seems like a regression to me, though. I'm wondering if that alone is enough to block the new Calamares? | 04:08 |
| Eickmeyer[m] | It seems odd. There's literally no difference. | 04:08 |
| arraybolt3[m] | (I mean, it will make troubleshooting future problems for users rather tricky.) | 04:09 |
| arraybolt3[m] | Actually, looks like it's not Cala's fault - the command line in the Studio desktop file doesn't include any logging. | 04:09 |
| arraybolt3[m] | (The logs aren't Cala's fault I mean.) | 04:10 |
| Eickmeyer[m] | That makes sense. I didn't know to add anything there, it was literally a copy from the one from two years ago. | 04:10 |
| Eickmeyer[m] | This means another SRU, and there's no time left. | 04:10 |
| arraybolt3[m] | OK, I'm getting logs now. | 04:11 |
| arraybolt3[m] | (Yeah, the logs fix won't happen, and sadly it looks like Cala might not happen either if this keeps up, but maybe at least BTRFS will be fixed.) | 04:11 |
| arraybolt3[m] | Eickmeyer: Can you test on your end with a Secure Boot EFI VM and see if you can reproduce the problem? If you're on virt-manager you just select the right firmware file and it works. | 04:12 |
| Eickmeyer[m] | I use virtualbox, which doesn't have secure boot. | 04:12 |
| arraybolt3[m] | Ah. | 04:13 |
| arraybolt3[m] | I learned to really dislike VBox after fighting with it like crazy on a Win7 laptop - it really did not want to do anything I wanted it to. | 04:13 |
| Eickmeyer[m] | I never have a problem with it except on Windows. | 04:14 |
| arraybolt3[m] | It's also slower and less feature-rich than anything else. (Yes, this is a blatant attempt to convert you to my hypervisor. :P) | 04:14 |
| Eickmeyer[m] | I can test tomorrow morning on real hardware (a Kubuntu Focus XE with Secure Boot enabled) but I'm out of time for the night. | 04:14 |
| arraybolt3[m] | (Er, slower and less feature-rich than QEMU/KVM.) | 04:14 |
| arraybolt3[m] | No problem. I'll have that error message in a bit most likely. | 04:15 |
| arraybolt3[m] | I'll also see if the regular Calamares fails on Studio with Secure Boot - if so, this isn't a regression after all and we can just ignore it :P \o/ | 04:17 |
| Eickmeyer[m] | That's a good plan too. | 04:17 |
| arraybolt3[m] | Command: <i>apt install -y --no-upgrade -o | 04:20 |
| arraybolt3[m] | Acquire::gpgv::Options::=--ignore-time-conflict shim-signed</i> failed to finish in 300 seconds. | 04:20 |
| arraybolt3[m] | There was no output from the command. | 04:20 |
| arraybolt3[m] | That's the full error message. | 04:21 |
| Eickmeyer[m] | That's very odd. I wonder why it's OK on Lubuntu but it's failing on Ubuntu Studio. | 04:22 |
| Eickmeyer[m] | Sounds to me like there's a config diff of some sort. | 04:22 |
| arraybolt3[m] | And somehow I lost the log this time. | 04:23 |
| * arraybolt3[m] stinks at this | 04:23 | |
| Eickmeyer[m] | Ouch | 04:24 |
| arraybolt3[m] | I told it to log at level 8, and it all printed to the terminal, but somehow it got corrupted so I just closed it. I wonder if /dev/stdout will have anything. | 04:24 |
| arraybolt3[m] | Nope. Oh well. | 04:25 |
| * arraybolt3[m] needs to take a break | 04:25 | |
| Eickmeyer[m] | arraybolt3: When was calamares-settings-common implemented? | 04:27 |
| arraybolt3[m] | Idk, it's always been there AFAICT. | 04:28 |
| Eickmeyer[m] | nvm | 04:30 |
| Eickmeyer[m] | So, in my analysis, Lubuntu and Ubuntu Studio are running the exact same command to install the shim. | 04:31 |
| Eickmeyer[m] | There's literally no difference between the two config files. | 04:31 |
| arraybolt3[m] | Hmm. Yet Studio fails three times, and Lubuntu works. Now I really wish I had that log. | 04:32 |
| arraybolt3[m] | (I should also try installing Lubuntu on baremetal to see if it's just this one finicky desktop.) | 04:32 |
| Eickmeyer[m] | I wonder if the shim is on the Lubuntu seed and somehow not on the Ubuntu Studio seed, in which case I'll promptly throw a predecessor under the bus. for no reason. | 04:34 |
| arraybolt3[m] | The funny thing is, the ISO boots just fine in Secure Boot. I guess that's not the flavor's doing though. | 04:35 |
| Eickmeyer[m] | Nope, we're shipping shim-signed. | 04:36 |
| * Eickmeyer[m] is even more puzzled | 04:37 | |
| * arraybolt3[m] tried a Secure Boot VM to see what happens | 04:37 | |
| arraybolt3[m] | s/tried/tries/ | 04:37 |
| Eickmeyer[m] | arraybolt3[m]: No, that's the ISO build system. | 04:38 |
| arraybolt3[m] | OK, installation in progress. | 04:41 |
| Eickmeyer[m] | Alright, I'm heading to bed. Have a good night! | 04:52 |
| arraybolt3[m] | 👋 | 04:52 |
| arraybolt3[m] | OK, I think I found the problem. I finally managed to catch the logs this time, and found this in and amongst the contents: | 05:12 |
| arraybolt3[m] | Your system has UEFI Secure Boot enabled.... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/06db2e18023f0875e96a7750abfe8747ea9d8202) | 05:15 |
| arraybolt3[m] | So that's the problem - I think apt is getting hung up installing the shim on Studio because somehow it's gotten the idea that it needs to mess with the Secure Boot keys. | 05:16 |
| arraybolt3[m] | And now thanks to a crummy problem with GNOME Boxes and VM save-to-disk mode with Ubuntu, I've been locked out of doing anything with the log file. | 05:19 |
| arraybolt3[m] | I can't even get to a TTY in the VM. | 05:20 |
| arraybolt3[m] | Hey, I got in! | 05:23 |
| lubot | [telegram] <Leokolb> I will be testing uefi+secure boot on bare metal boxes in about 2 hours time | 05:23 |
| lubot | [telegram] <Leokolb> off to the gym.. | 05:24 |
| guiverc | :) & thanks @Leokolb, I've got a install.alongside on BIOS box currently running, doing a replace.partition next | 05:24 |
| arraybolt3[m] | Calamares logs: https://termbin.com/xmdw | 05:24 |
| guiverc | have fun @ gym | 05:24 |
| arraybolt3[m] | @Leokolb: Thanks! Have a great day! | 05:25 |
| arraybolt3[m] | Eickmeyer Simon Quigley (Developer) Calamares logs from failed Secure Boot install of Studio: https://termbin.com/xmdw | 05:25 |
| arraybolt3[m] | (That way they'll see it tomorrow.) | 05:25 |
| Eickmeyer[m] | It's not calamari, that's 1the good news. The bad news is it's v4l2loopback and I'll have to remove it from the seed. | 05:32 |
| arraybolt3[m] | Eickmeyer: Sorry if my ping woke you up - I was trying to make it so you'd see it tomorrow, not call you back now. | 05:32 |
| Eickmeyer[m] | *calamares | 05:32 |
| Eickmeyer[m] | Nah, wasn't asleep yet. | 05:33 |
| Eickmeyer[m] | So, the SRU can proceed since Calamares isn't the problem. It's a dkms module, and those are notoriously problematic with secure boot. | 05:34 |
| arraybolt3[m] | Aha. OK, that's good to know! | 05:35 |
| arraybolt3[m] | (That also means I'll be doing more installs tonight.) | 05:35 |
| arraybolt3[m] | What does v4l2loopback do anyway? | 05:35 |
| Eickmeyer[m] | Good luck! | 05:35 |
| Eickmeyer[m] | It provides a way to make a video loop back device, like the output from OBS can be a virtual Webcam for Google Meet, Skype, and the like. | 05:36 |
| arraybolt3[m] | Ooo, nice. | 05:37 |
| guiverc | grr :( I put dates/ISO in wrong columns in testing.checklist earlier sorry; I'll ignore (for now) as the 28.1 makes it rather obvious | 05:58 |
| guiverc | ^ corrected (no birbs outside yet & box beside me busy installing still on 'slow' spinning rust) | 06:05 |
| arraybolt3[m] | I thought I was going to do more tests, but I'm dog tired and probably should be done for the night. Thank you guys for all your help and for letting me help, and see you all tomorrow! | 06:05 |
| arraybolt3[m] | guiverc: Can I trust the setting of verification-done and verification-done-jammy to you once testing is done? The bug report is here: https://bugs.launchpad.net/ubuntu/+source/calamares/+bug/1980180 | 06:10 |
| ubot93 | Launchpad bug 1980180 in calamares (Ubuntu Jammy) "[SRU] Calamares 3.2 Series for 22.04 LTS" [Medium, Fix Committed] | 06:10 |
| guiverc | get sleep, and well done with your work Aaron / arraybolt3[m]. I don't see a rush to mark verification-done (yet) but we'll get it done... | 06:25 |
| * guiverc was out with birbs... my no-birbs-outside was a count 60+ & at least 6 species, but missing the flock I look really for | 06:26 | |
| lubot | [telegram] <Leokolb> Back in action..re testing (re @lubuntu_bot: (irc) <guiverc> was out with birbs... my no-birbs-outside was a count 60+ & at least 6 species, but missing the flock I look really for) | 08:45 |
| guiverc | :) (i'm doing a combined XFS / & separate /home install, in between laundry & other household stuff..) | 09:28 |
| lubot | [telegram] <Leokolb> Saw that ..very good - checklist now complete -minus BTRFS as all are aware - will test it as soon as it is added (cala settings update )in next ISO (re @lubuntu_bot: (irc) <guiverc> :) (i'm doing a combined XFS / & separate /home install, in between laundry & other household stuff..)) | 09:54 |
| guiverc | :) thank you Leó / @Leokolb | 09:56 |
| lubot | [telegram] <kc2bez> Just waking up and trying to catch up. | 09:59 |
| guiverc | :) I was writing a ping to you Dan ! | 09:59 |
| lubot | [telegram] <kc2bez> :) | 10:00 |
| lubot | [telegram] <kc2bez> Testing is going good? | 10:00 |
| guiverc | all done excluding BTRFS that's not been tested (expected failure as needs another change)... yep all good | 10:00 |
| lubot | [telegram] <kc2bez> Nice | 10:00 |
| guiverc | do you need/want anything else before we mark in bug report testing completed | 10:00 |
| guiverc | verification-done-jammy etc.. | 10:01 |
| lubot | [telegram] <kc2bez> I looked at the test results and I think we are good from a Lubuntu standpoint | 10:01 |
| guiverc | on https://bugs.launchpad.net/ubuntu/+source/calamares/+bug/1980180 Erich writes he's happy & leaves verification for us to mark | 10:02 |
| ubot93 | Launchpad bug 1980180 in calamares (Ubuntu Jammy) "[SRU] Calamares 3.2 Series for 22.04 LTS" [Medium, Fix Committed] | 10:02 |
| lubot | [telegram] <kc2bez> OK. I don't think we need anything beyond the testing we have. I am good if you mark it verified | 10:03 |
| guiverc | thanks, I'm busy with dinner :) | 10:03 |
| lubot | [telegram] <kc2bez> No worries, when you get a chance. | 10:03 |
| guiverc | okay will do. | 10:05 |
| * guiverc misread what you typed; distracted by dinner.. I'll do it asap | 10:06 | |
| lubot | [telegram] <kc2bez> Thanks everyone for all of their work on this. Sorry the day job tied me up so much yesterday. | 10:07 |
| lubot | [telegram] <Leokolb> Just checked manifest for todays Jammy build and the updates are all on board ie. 3.2.60 .. settings 1,22,04.4.1 - you devs have done a fantastic job! | 17:12 |
| tsimonq2 | <Eickmeyer[m]> "It's not calamari, that's 1the..." <- What exactly does that package do? | 17:22 |
| Eickmeyer[m] | tsimonq2: As I told Aaron last night (if you scrolled down just a little further): It provides a way to make a video loop back device, like the output from OBS can be a virtual Webcam for Google Meet, Skype, and the like. | 17:23 |
| tsimonq2 | Sorry /me partied too hard last night - change made in Kinetic first yet? | 17:24 |
| Eickmeyer[m] | No, but that's easy. I don't want to do it unless I get a preliminary go-ahead for Jammy. | 17:25 |
| tsimonq2 | You're the RL and the release hasn't been frozen frozen yet so just make the change and file the SRU for the metapackage imo | 17:25 |
| tsimonq2 | That's what I'd do if I were in your shoes | 17:25 |
| Eickmeyer[m] | Well, technically it'd require an FFe as well. | 17:25 |
| tsimonq2 | See the SRU page again. No it doesn't :) | 17:26 |
| Eickmeyer[m] | If that's the case, I was given bad advice. | 17:26 |
| tsimonq2 | Anyway /me finishes walking the good dog | 17:26 |
| tsimonq2 | Eickmeyer[m]: Iirc it's all about intent | 17:27 |
| tsimonq2 | Be nice, ask the question, but make sure all the technical bits are ready to go too | 17:27 |
| Eickmeyer[m] | Of course, especially in the interest of time. | 17:27 |
| Eickmeyer[m] | Of course, this one's getting marked as critical since, well, secure boot is utterly broken. | 17:28 |
| tsimonq2 | Is this a regression from 22.04? If not, why didn't anyone catch it then? | 17:44 |
| tsimonq2 | I don't mean to roast you, I mean this out of love, but come on dude, one week before GE is cutting it really really close | 17:45 |
| tsimonq2 | There's a reason the SRU team has the process they do, and if you spend the time to nail that and nail the upload I don't think they will have an issue with your request | 17:46 |
| Eickmeyer[m] | I failed to test on secure boot, that's why it never got caught. You guys aren't seeding a dkms module. | 18:17 |
| Eickmeyer[m] | I have no problem with combining testing in the future. :) | 18:17 |
| Eickmeyer[m] | This has been a great experience. | 18:17 |
| Eickmeyer[m] | Simon Quigley (Developer): I thought my machine had secure boot enabled until recently. | 18:21 |
| -queuebot:#lubuntu-devel- Builds: Lubuntu Desktop amd64 [Jammy 22.04.1] (20220729.1) has been added | 19:33 | |
| lubot | [telegram] <Leokolb> I am willing to help test UbuntuStudio with secure boot anytime ..just give a shout. (re @lubuntu_bot: (irc) <Eickmeyer[m]> Simon Quigley (Developer): I thought my machine had secure boot enabled until recently.) | 20:54 |
| Eickmeyer | @leokolb Yeah, don't worry about it, it *will fail*. dkms modules and secure boot don't mix. I'm going to be working on removing it over the weekend, Lukasz has the heads up. | 21:12 |
| lubot | [telegram] <Leokolb> 👌 (re @lubuntu_bot: (irc) <Eickmeyer> @leokolb Yeah, don't worry about it, it *will fail*. dkms modules and secure boot don't mix. I'm going to be working on removing it over the weekend, Lukasz has the heads up.) | 21:16 |
| arraybolt3[m] | Eickmeyer: Do you think a good project for Studio might be to try to get the v4l2loopback module signed by Canonical? | 21:21 |
| arraybolt3[m] | (I don't know how or if that would work, but it would really be handy for Secure Boot users, and the module is obviously somewhat trusted since it's in the archives.) | 21:24 |
| Eickmeyer | No clue how that would work. | 21:26 |
| arraybolt3[m] | (Plus GNOME is going to start alerting users if they don't have Secure Boot on, according to this news article: https://www.phoronix.com/news/GNOME-Secure-Boot-Warning So users of the Ubuntu Studio Installer on GNOME may be using Secure Boot more often if and when this trickles into Ubuntu. | 21:26 |
| lubot | [telegram] <teward001> i dont think canonical will sign Universe stuff. is v4l2loopback in main yet? (also this is Lubuntu devel channels not Studio devel heh) | 21:42 |
| * arraybolt3[m] banishes teward001 into the firmware of Simon Quigley (Developer)'s rice cooker | 21:44 | |
| tsimonq2 | Rice cooker? Air fryer. :P | 21:46 |
| Eickmeyer[m] | It feels like an air fryer outside. | 23:14 |
| Eickmeyer[m] | @teward001: For v4l2loopback to be in main, there would have to be a good use case for including it. I can't imagine one that would make sense from a main inclusion standpoint. | 23:15 |
| lubot | [telegram] <teward001> *already has a presence in the firmware of every system planetwide* | 23:19 |
| lubot | [telegram] <teward001> Eickmeyer: agreed. you could ask the kernel team if they'd be willing to sign it anyways | 23:20 |
| lubot | [telegram] <teward001> but theres a reason most of this stuff is universe at the moment :P | 23:20 |
| Eickmeyer[m] | Well, yes. Answer: It's not needed for Server or Ubuntu Desktop. It's that simple. | 23:21 |
| Eickmeyer[m] | I could talk to them, but that doesn't fix the immediate need. | 23:22 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!