Hydragyrum | is there anything about switching inits on devuan? I'm considering switching from sysv to openrc | 02:30 |
---|---|---|
Afdal | :o | 02:32 |
Afdal | We just had this conversation... | 02:33 |
golinux | Hydragyrum: Look on the dev1galaxy.org forum. | 02:34 |
Hydragyrum | I did some quick googling and didn't see anything, will take another look though | 02:35 |
Afdal | My takeaway from my conversation about switching from sysvinit to runit was | 02:36 |
Hydragyrum | *looks up* oh yeah and I even was involved in the convo some lol | 02:36 |
Afdal | you should try just installing the other init, rebooting | 02:36 |
Hydragyrum | it was just before my last restart of my irc client | 02:36 |
Afdal | and then seeing if you're on the new init | 02:36 |
Afdal | if you are then you're good to go | 02:37 |
Afdal | Apparently init selection tends to select the most recent install | 02:37 |
Afdal | so I've heard, anyway | 02:38 |
Afdal | I'll be making some adjustments to gather my services together in the runit way | 02:39 |
Afdal | instead of relying on its compatibility with sysvinit's way | 02:40 |
Afdal | but I actually think openrc does things more similarly to sysvinit | 02:40 |
Afdal | rip | 02:42 |
Hydragyrum | well now openrc gang | 02:44 |
Afdal | It worked? | 02:44 |
Afdal | damn, that simple huh | 02:44 |
Afdal | Maybe I'll try switching to runit soon then | 02:44 |
Hydragyrum | however, my external drive got labeled as sda this time which messed me up a bit (interactively, I wouldn't automount by device name) | 02:45 |
Hydragyrum | but I don't know why a usb external drive would be first in the order, esp. since I'm pretty sure I'm using the same udev | 02:50 |
Afdal | that's annoying | 02:53 |
Afdal | that's why I usually use UUIDs though | 02:53 |
Hydragyrum | yeah it's something I handmount on every boot | 02:59 |
Hydragyrum | which usually means either like 3 times a day or once every few weeks | 02:59 |
Afdal | What desktop environment do you use? | 03:11 |
Hydragyrum | i3wm | 03:12 |
Hydragyrum | de-less | 03:12 |
Hydragyrum | next is switching to openrc-init (because openrc just replaces sysv-rc) | 03:15 |
Hydragyrum | so I need to manually swap that bit | 03:15 |
Afdal | openrc is so complicated and confusing to me | 03:16 |
Afdal | that's why I prefer runit | 03:16 |
Afdal | simple and comfy... | 03:16 |
Hydragyrum | haven't used runit myself, openrc is what I use on many of my other systemd-less systems (artix, gentoo, etc.) | 03:16 |
adhoc | Hydragyrum: why the switch away from sysv-rc ? | 03:17 |
Afdal | Apparently you can even use openrc with runit | 03:17 |
Afdal | since openrc isn't the init... @_@ | 03:17 |
Hydragyrum | adhoc, to get all my systems on the same system mostly | 03:18 |
Hydragyrum | also because I'm bored lol | 03:18 |
adhoc | ok, so long as devuan isn't moving away and dropping it | 03:18 |
adhoc | being bored, or because you can change can be fun | 03:18 |
adhoc | so long as you have options to go back, if it doesn't work out =) | 03:19 |
Hydragyrum | now where's my agetty rc script... | 03:19 |
Hydragyrum | apparently nothing installs such a script, but openrc-init doesn't use inittab apparently and so I need agetty initscript to start my login gettys | 03:20 |
Hydragyrum | well just stole the script from my gentoo box, boutta reboot again | 03:28 |
Hydragyrum | and it worked | 03:33 |
ukine | cool stuff Hydragyrum | 03:34 |
adhoc | Hydragyrum: so that is not in a package? | 03:34 |
Hydragyrum | apparently not | 03:34 |
Hydragyrum | apt-file doesn't find it, and it's not provided by util-linux which provides agetty, or by openrc | 03:35 |
adhoc | sounds like its worth logging a bug ? | 03:35 |
adhoc | given you have a clear understanding of what should be happened, and what should be provided | 03:36 |
Hydragyrum | lol true enough | 03:37 |
Hydragyrum | better yet, I could submit a patch to devuan's openrc package with it | 03:37 |
Hydragyrum | (seeing as the script is written for openrc, and wouldn't function properly with sysvinit most likely even with changing the #!/sbin/openrc-run) | 03:39 |
CAPTCHA_REQUIRED | hello | 06:22 |
CAPTCHA_REQUIRED | i'm trying to install beowulf via debootsrap | 06:23 |
CAPTCHA_REQUIRED | can someone please link me to the gpg keyring? | 06:23 |
CAPTCHA_REQUIRED | the gpg keyring for verifying release files | 06:23 |
CAPTCHA_REQUIRED | E: Release signed by unknown key (key id BB23C00C61FC752C) | 06:24 |
CAPTCHA_REQUIRED | I didn't see any BB23C00 on https://www.devuan.org/os/team | 06:24 |
CAPTCHA_REQUIRED | oh nvm | 06:28 |
CAPTCHA_REQUIRED | i found the keyring | 06:28 |
CAPTCHA_REQUIRED | but some more official instructions for setting up devuan manually via debootsrap would be appreciated | 06:28 |
rwp | CAPTCHA_REQUIRED, There isn't any *GOOD* documentation on setting up the system with debootstrap. | 06:44 |
rwp | But there is https://wiki.debian.org/Debootstrap | 06:44 |
rwp | Which I don't like because I don't agree with the instructions. | 06:45 |
rwp | CAPTCHA_REQUIRED, Ah! This is much better documentation. https://www.debian.org/releases/stable/amd64/apds03.en.html#idm4334 | 06:47 |
CAPTCHA_REQUIRED | oh thankyou | 06:47 |
CAPTCHA_REQUIRED | and it's been a while since i installed a devuan system from scratch, what's the package you need to install to make simpleterm work correctly with it? | 06:47 |
CAPTCHA_REQUIRED | installs control-code defitions and whatnot | 06:48 |
lts | CAPTCHA_REQUIRED: stterm? | 06:49 |
CAPTCHA_REQUIRED | oh | 06:49 |
CAPTCHA_REQUIRED | ncurses-term | 06:49 |
CAPTCHA_REQUIRED | lts thankyou for the link | 06:49 |
CAPTCHA_REQUIRED | it reminds me of the freebsd handbook | 06:49 |
rwp | It does have some of the same look and feel as the freebsd handbook. | 06:52 |
CAPTCHA_REQUIRED | sigh, still no modern zfs | 06:55 |
rwp | You can always install zfs and use it through module-assistant. Or you can combine it yourself. | 07:04 |
rwp | The license is incompatible with the GPL of the Linux kernel. That's why the distro hasn't combined them. | 07:04 |
rwp | Yet Canonical decided that if they did then regardless of it not being a legal combination that no one would have standing to bring suite. So they did it anyway. | 07:05 |
rwp | In the US system only parties with standing in the case can bring suite. Since no one will have standing there is no one to bring suite. | 07:05 |
rwp | However I am not sure that is the best way to go. Since it weakens Copyright. And Free Software depends upon a strong Copyright. | 07:05 |
rwp | Anyway... You can install zfs and keep it automatically compiled as new kernels arrive with module-assistant. | 07:06 |
rwp | That's what everyone tells me. I still have learning ZFS on my task list. Will get to it sometime. | 07:06 |
CAPTCHA_REQUIRED | what is a module-assistant | 07:06 |
rwp | https://wiki.debian.org/ModuleAssistant | 07:15 |
CAPTCHA_REQUIRED | huh | 07:16 |
CAPTCHA_REQUIRED | madwifi is still a thing? | 07:16 |
CAPTCHA_REQUIRED | (from the screenshots) | 07:17 |
rwp | Here is the basic wiki page on ZFS. https://wiki.debian.org/ZFS | 07:17 |
rwp | And as I said I haven't learned ZFS yet. I just react to what people I trust tell me about it. | 07:18 |
CAPTCHA_REQUIRED | what do thee decree? | 07:19 |
rwp | I guess I am a little outdated. It looks like m-a is now DKMS the Dynamic Kernel Module Support Framework. | 07:20 |
rwp | I don't decree anything. I have not learned ZFS myself. I just reference other documentation as I can that I think will help. | 07:21 |
CAPTCHA_REQUIRED | why doesn't devuan use debian's backport level? | 07:21 |
CAPTCHA_REQUIRED | https://packages.debian.org/buster-backports/zfs-dkms | 07:21 |
rwp | "backport level"?? | 07:21 |
CAPTCHA_REQUIRED | I should be installing 2.x | 07:22 |
CAPTCHA_REQUIRED | yet, i'm installing zfs 0.7.x from backports | 07:22 |
CAPTCHA_REQUIRED | beowulf-backport | 07:22 |
rwp | I have no idea. But Devuan is Debian except for a relatively small list of packages that are blocked due to systemd entanglements. | 07:22 |
CAPTCHA_REQUIRED | but backports should be backports, right? | 07:23 |
rwp | Here are the Devuan banned packages list: https://pkgmaster.devuan.org/bannedpackages.txt | 07:23 |
rwp | Since I don't see zfs in that list then it should be good to go for Devuan. | 07:24 |
rwp | I think backports might be devuan specific in this case. I do not know. | 07:24 |
CAPTCHA_REQUIRED | shows up here too https://pkginfo.devuan.org/cgi-bin/policy-query.html?c=package&q=zfs-dkms&x=submit | 07:24 |
CAPTCHA_REQUIRED | https://pkginfo.devuan.org/cgi-bin/package-query.html?c=package&q=zfs-dkms=2.0.3-9~bpo10+1 | 07:25 |
rwp | zfs-dkms in chimaera is version 2.0.3-9 and chimaera is very close to actual release right now. Very close. | 07:25 |
rwp | That shows that zfs-dkms in beowulf-backports is 2.0.3-9. | 07:26 |
rwp | So you shouldn't be getting 0.7 from backports but 2.0.3-9, right? | 07:26 |
CAPTCHA_REQUIRED | correct | 07:32 |
CAPTCHA_REQUIRED | but i am | 07:32 |
CAPTCHA_REQUIRED | i followed the instructions here | 07:34 |
CAPTCHA_REQUIRED | https://openzfs.github.io/openzfs-docs/Getting%20Started/Debian/index.html#installation | 07:34 |
CAPTCHA_REQUIRED | and changed buster to beowulf | 07:34 |
CAPTCHA_REQUIRED | if do it with apt -t beowulf-backports it installs the correct one, but that uninstalls openrc | 07:36 |
CAPTCHA_REQUIRED | nvm | 07:38 |
CAPTCHA_REQUIRED | i got it working, it's just concerning package pinning isn't working as expected | 07:39 |
CAPTCHA_REQUIRED | also this: https://dpaste.com/BYYG2Q2ZQ.txt | 07:39 |
CAPTCHA_REQUIRED | for some reason | 07:39 |
rwp | Hmm... Some package conflict there, huh. Well... I don't know the deep issue there. Something to debug and figure out. | 07:42 |
CAPTCHA_REQUIRED | i didn't see any direct conflict | 07:42 |
CAPTCHA_REQUIRED | but ok | 07:42 |
CAPTCHA_REQUIRED | bizzare | 07:42 |
rwp | Not so much bizarre as unfortunate. This whole init unpleasantness has been... Unpleasant! :-( | 07:43 |
CAPTCHA_REQUIRED | so just kind of casual question here | 07:46 |
CAPTCHA_REQUIRED | is there an official way to contibute alternative initscripts, say as for openrc native scripts | 07:47 |
CAPTCHA_REQUIRED | or is devuan just unning sysv scripts into openrc via compatibility mode | 07:47 |
CAPTCHA_REQUIRED | also rwp, thankyou a lot for the help tonight | 07:58 |
Successus | Hello, I was reading the wiki, but couldn't find direct information, could you please tell me if Devuan's kernel is completely free? (no non-free blobs, firmware, drivers). All I read is that you can chose during installation if you want to install non-free drivers | 08:54 |
CAPTCHA_REQUIRED | Successus, don't enable the non-free repository and it will remain free | 08:54 |
CAPTCHA_REQUIRED | don't install anything from the non-free repository | 08:54 |
CAPTCHA_REQUIRED | eg, drivers | 08:54 |
CAPTCHA_REQUIRED | firmware | 08:54 |
Successus | So that means the kernel doesn't have anything non-free integrated? | 08:55 |
CAPTCHA_REQUIRED | i don't think so | 08:55 |
Successus | Just like Debian says for its kernel | 08:55 |
Successus | The process of "cleaning" the kernel is the same? | 08:55 |
CAPTCHA_REQUIRED | devuan wouldn't put anything nonfree in there when the debian version doesn't | 08:55 |
Successus | I see | 08:55 |
Successus | and that goes for kernel too | 08:55 |
Successus | Alright then, thank you | 08:56 |
Successus | I want to ask one more thing, would I be able to use encrypted /home partition, that was encrypted during Debian installation, using Debian's installer? | 08:56 |
Successus | I think one time I was testing, installing Debian, and even Debian's installer formatted the partition, even though I didn't select that | 08:57 |
CAPTCHA_REQUIRED | i use full disk encryption with devuan on my laptop | 08:57 |
CAPTCHA_REQUIRED | though i installed back in jessie | 08:57 |
CAPTCHA_REQUIRED | and just kept upgrading that system | 08:57 |
CAPTCHA_REQUIRED | there was a option i the installer (i forgot which one probably netinst) to do it | 08:58 |
CAPTCHA_REQUIRED | automaticlly | 08:58 |
Successus | you switched from Debian to Devuan on that laptop, or just installed Devuan? | 08:58 |
CAPTCHA_REQUIRED | no i installed devuan | 08:58 |
Successus | ok | 08:58 |
CAPTCHA_REQUIRED | just devuan | 08:58 |
CAPTCHA_REQUIRED | and upgraded to later devuan releases | 08:58 |
Successus | ok | 08:59 |
Successus | you used netinst to install it? | 09:00 |
CAPTCHA_REQUIRED | yes | 09:00 |
Successus | do you perhaps know how much space Devuan occupies after installation? | 09:00 |
CAPTCHA_REQUIRED | at least as far as i remember | 09:00 |
CAPTCHA_REQUIRED | that was years ago | 09:00 |
Successus | I'm interested in minimal installation sizes but can't find information on that, except for trying it out | 09:00 |
CAPTCHA_REQUIRED | Successus, it depends on what features you select | 09:00 |
Successus | right | 09:00 |
CAPTCHA_REQUIRED | if your trying to build a minimal system | 09:00 |
Successus | could I test that in a VM, will it occupy the same space as it would on hardware? | 09:00 |
CAPTCHA_REQUIRED | add this to your apt.conf | 09:01 |
CAPTCHA_REQUIRED | APT::Install-Recommends "false"; | 09:01 |
CAPTCHA_REQUIRED | APT::Install-Suggests "false"; | 09:01 |
Successus | I see | 09:01 |
Successus | thank you | 09:01 |
CAPTCHA_REQUIRED | and just watch carefully for reccomends you actually want | 09:01 |
CAPTCHA_REQUIRED | it will cut the bloat down significantly | 09:01 |
CAPTCHA_REQUIRED | just make sure you know whatyour doing | 09:01 |
Successus | alright | 09:01 |
ham5urg | Does anyone know a small board with open firmware, some arm board maybe. | 14:22 |
humpelstilzchen[ | ham5urg: beaglebone might help? | 14:23 |
ham5urg | humpelstilzchen[ , never heard of, thanks, I will take a look. | 14:24 |
humpelstilzchen[ | ham5urg: oh wow, you really never heard of beagleboard/beaglebone? That makes me wonder a bit. | 14:30 |
humpelstilzchen[ | Anyway pine64 might also have an interesting board for you | 14:31 |
onefang | https://www.olimex.com/ may or may not help. #devuan-arm might be more helpful. There are probably other channels where such things are actually on topic. | 14:39 |
luser979 | #olimex | 14:49 |
luser979 | Also #alpine-linux on oftc | 14:49 |
humpelstilzchen[ | #beagle to join the club of channels to join | 14:51 |
tr-amarsh04 | base-files upgrade broke building a custom kernel | 15:34 |
asbesto | <3 | 16:24 |
mdt | i try to run pipewire but it doesnt show any soundcard(s). i understand that pw uses alsa. alsa alone shows the soundcard, pw doesnt. any idea? | 16:28 |
gnarface | just do some tests with aplay, arecord, and speaker-test | 16:32 |
gnarface | i don't know what's wrong with pipewire, but once you have enough proof alsa is working right that should be enough to get some help from them directly | 16:34 |
mdt | alsa allone works fine. only pw doesnt show any soundcard | 16:35 |
gnarface | well someone here might know but you should ask in pipewire's channel too | 16:36 |
mdt | i read the wiki, but didnt find much. i thought it's maybe missing systemd 😉 | 16:38 |
gnarface | well i've never used it so i don't know but perhaps that means it could be a permissions issue | 16:42 |
mdt | gnarface: may well be but alsamixer shows the devices just fine. pw doesnt log any hard errors. | 16:47 |
user282069 | i wonder; is /usr/sbin/bluetoothd compiled with --enable-expermental ? | 17:23 |
user282069 | this in the bluez package | 17:23 |
user282069 | chromium doesnt see the adapter https://googlechrome.github.io/samples/web-bluetooth/discover-services-and-characteristics.html | 17:30 |
user282069 | d needed root with bluetoothctl to power on and pair successfully | 17:32 |
user282069 | problem is: chromium not seeing the adapter. thank you if you have any thoughts. | 17:35 |
buZz | mdt: i'm quite sure i've seen ppl using pipewire without systemd | 17:41 |
rkta | rwp: The crash we talked about a day or two ago was due to restarting elogind as needrestart suggested. | 18:25 |
rwp | rkta, Ah... Yes. That would probably do it too. Since elogind somewhat sorta sits in the same relationship to processes as the X server does for X processes. | 18:45 |
rwp | rkta, Thanks for sending me that information. I appreciate it! :-) | 18:45 |
rkta | Maybe we should teach needrestart not to auto mark elogind for a restart. | 18:48 |
rkta | Not sure where to report this. AFAIK elogind is not in debian and needrestart is the original debian version in devuan. | 18:49 |
gnarface | bugs.devuan.org | 18:50 |
gnarface | yes, elogind is made here so it is a valid devuan issue | 18:50 |
CAPTCHA_REQUIRED | what is elogind for | 18:51 |
CAPTCHA_REQUIRED | is it even needed? | 18:51 |
gnarface | whether they'll decide to change elogind or needrestart i couldn't say | 18:51 |
gnarface | you only need it if you're using graphical logins, it's a drop-in replacement for systemd-logind | 18:51 |
gnarface | it would be safe to uninstall probably unless you don't know how to get X11 running without a GUI login already in place | 18:51 |
CAPTCHA_REQUIRED | but i could do that fine with policykit | 18:52 |
CAPTCHA_REQUIRED | i mean consolekit | 18:52 |
rkta | I don't even do graphical logins - not sure why it is installed. Trying to uninstall gave dependency problems. | 18:52 |
gnarface | policykit or consolekit would be tied to the graphical login but that's not the only component of it | 18:52 |
CAPTCHA_REQUIRED | why the sudden requirement of elogind to do the same thing i've been doing for years? | 18:52 |
gnarface | upstream software that changed to attempt to force you to use systemd | 18:53 |
gnarface | someone clever created elogind to save you from sophie's choice | 18:53 |
gnarface | i'd toss it all overboard myself but ymmv | 18:53 |
CAPTCHA_REQUIRED | ty | 18:54 |
rwp | Right. My desktop and laptop running i3 started from a vt text console login with xinit (startx calls xinit) does not have elogind installed as I don't need it. | 18:55 |
CAPTCHA_REQUIRED | rwp what i mean is i had graphical logins way back in linux mint before systemd or elogind were a thing | 18:55 |
rwp | But XFCE which is a standard Stable Desktop Environment (and the other DEs) all want (e)logind to be installed and to us it to create "user sessions". | 18:55 |
CAPTCHA_REQUIRED | or older unixes with xdm | 18:56 |
rwp | You don't need to preach to us the choir here. We will sing the hymns along with you already. :-) | 18:56 |
CAPTCHA_REQUIRED | oh i know that | 18:56 |
gnarface | yea it's to do with these redhat and freedesktop.org people making everything more complicated so they can get feature-parity with windows for their malware vendors | 18:56 |
CAPTCHA_REQUIRED | I'm just still not sure exactly what a 'user session' is | 18:57 |
rwp | Honestly I have been trying to understand what a user session is for some years too. So far it has escaped my grasp. | 18:57 |
rkta | Does this solution to uninstall elogin looks safe to do? https://vs2.rkta.de/files/elogind.aptitude | 18:57 |
* CAPTCHA_REQUIRED sigh | 18:57 | |
CAPTCHA_REQUIRED | ok so it's not just me | 18:57 |
rwp | But (e)logind does a few other things that used to be handled by policy kit, console kit, libpam and other things previously. It's not all just "user sessions". | 18:58 |
user____ | rwp: a user session is the set of data required to maintain the logged-on state of a user usually in a graphical session sense. | 18:58 |
rwp | rkta, I don't use aptitude so am unfamiliar with the format but that all looks okay to me. | 18:59 |
user____ | rwp: that being said, the enemy camp came and broke all conventions, which implied some admin would set up the session parameters and permissions, such that any x user can access "his" session and automagical things fix what the admin would have previously. | 18:59 |
CAPTCHA_REQUIRED | i'm still confused | 18:59 |
user____ | And all that automagical thing is bad magic, more smoke than mirrors, and breaks conventions ad nauseam. polkit, elogind etc. | 19:00 |
rwp | rkta, In relation to this topic it is pulling in libcgmanager0 which is a previous implementer of control groups that elogind is implementing. | 19:00 |
rwp | rkta, I am totally confused about why it would want ethtool and hdparm but those are excellent tools that I always install anyway. | 19:01 |
rwp | rkta, The pm-utils is also replacing functionality from elogind for suspend and shutdown. Which are likely what you want anyway. | 19:01 |
rwp | (pm-utils, Pavel Machek utils, Power Management utils) | 19:02 |
* rwp wishing for Nigel's Tux On Ice suspend code instead but that's another sad tale... | 19:02 | |
CAPTCHA_REQUIRED | hey, just want to make sure everyone is aware of this. Eventually redhat with stop ""supporting"" all these smoke and mirrors to getting an Xsession running after replacing all the simple stuff with complex things only they can maintain and then we will be screwed | 19:02 |
rwp | user____, I'll acknowledge that the other camp wants control group functionality. (FB uses it extensively.) But I haven't been absorbed into the body yet and don't feel it myself. | 19:04 |
rwp | user____, The thing that always seems too fuzzy is "a user session is the set of data required to maintain the logged-on state of a user". Okay. Like what? | 19:05 |
_ds_ | tr-amarsh04, base-files 11,1+devuan3? “make bindeb-pkg” works fine here (tested using 5.13.19). | 19:05 |
user____ | rwp: everything from who, credentials, logged on status (elevated privileges or not), host, type of access (local, remote, graphical), etc etc. | 19:08 |
user____ | rwp: paused session, logged out temporarily (locked session) etc etc | 19:09 |
user____ | All the nitty gritty. | 19:09 |
rwp | user____, And to think that I don't have any of those and have been blissfully ignorant of my lack of them all of these years! :-/ | 19:09 |
rwp | With those things one must mention D-Bus too since the implementation of those is deeply tied to dbus too. Which created yet another set of problems. | 19:10 |
tr-amarsh04 | _ds_ base-files 11.1+devuan2 works fine for me, base-files 12+devuan1 breaks things | 19:11 |
user____ | Indeed. Remember when windows nt was boasting it runs without an admin? User session automation is exactly that missing admin brought to your formerly free *nix. | 19:11 |
user____ | This is going off topic a bit, move it to -offtopic? | 19:11 |
rwp | rkta, Please send another update on how things went for you when you arrive at your next status point. I have popcorn and am following along closely. :-) | 19:11 |
rkta | rwp: It's not safe to do :) | 19:12 |
rkta | Atleast not doing it while under X. This will lead to a crash and you need to do that magic key stuff to reboot. | 19:13 |
rkta | And then you can startx but you can't move the mouse cursor and pressing keys does nothing. Again Magic keys to reboot... | 19:14 |
rwp | You should not need the Magic SysRq to reboot. That you needed it is still an unexplained odd thing. | 19:14 |
rkta | How would you reboot? | 19:14 |
rwp | Control-Alt-Delete is often a useful quick way to reboot when sitting at the Linux text console vts. | 19:14 |
rwp | Implemented by /etc/inittab with "ca:12345:ctrlaltdel:/sbin/shutdown -t1 -a -r now" | 19:15 |
rwp | If I have started X I will usually just reboot from an XTerm command line the normal way. But can also Control-Alt-F1 to vt1 and then Control-Alt-Delete to trigger the shutdown. | 19:16 |
rwp | That one is a reboot not a halt. But that's usually what I want. And if I want to halt I can then turn off the power when I get to the BIOS POST display. | 19:16 |
rkta | This doesn't help if I can't type. | 19:19 |
rwp | I also set "setxkbmap -option terminate:ctrl_alt_bksp" (along with a compose key) and so can also Control-Alt-Backspace to kill X and everything all at once. | 19:19 |
* rkta needs to take the dog for a walk | 19:19 | |
rwp | rkta, That the system is not taking keyboard or mouse input is one of those unexplained odd things about your system. You will need to debug it. Others don't see that problem. | 19:19 |
rwp | Re: dogs. When they gotta go then you gotta go! :-) | 19:19 |
rkta | rwp: I'll try to fix it tomorrow by reinstalling elogind. I don't know how I would debug such a thing. We'll see... | 19:44 |
gnarface | CAPTCHA_REQUIRED: yea we all know that. you're late to the party. | 20:44 |
gnarface | ranting about it goes to #devuan-offtopic | 20:44 |
gnarface | here we just try to mitigate damage | 20:44 |
gnarface | i can tell these revelations are new to you and just as disturbing as they should be, but it is old news now | 20:45 |
gnarface | (also, to be very clear, i think the "we're screwed" scenario occurs long before that) | 20:48 |
gnarface | (they don't have to actually stop supporting any of it to make it all too expensive for anyone else to integrate into their distro securely, which it very nearly already is) | 20:49 |
luser978 | Then a devuan port of Slackware will emerge from the ruins, and save linux | 20:51 |
rwp | Can someone else running Ceres Unstable run "wget -q -O/dev/null https://www.example.com/" and tell me if it emits a spurious "SSL_INIT" string as of today's update? | 20:57 |
brocashelm | rwp: can confirm | 20:59 |
brocashelm | i updated my wget before running this command | 20:59 |
brocashelm | that's strange. wonder what's causing this | 20:59 |
rwp | I am going to downgrade wget and see if that isolates the problem. | 21:01 |
brocashelm | the string doesn't show if i use http://google.com | 21:01 |
brocashelm | but for https yes | 21:01 |
rwp | Yep! Downgrading avoids it. The problem is in 1.21.2-1 only. | 21:02 |
brocashelm | i can confirm with 1.21-1+b1 it works again (from chimaera) | 21:03 |
rwp | I filed a Debian bug report. Waiting for the bug number. bugs.debian.org has been very slow the past few days. | 21:07 |
brocashelm | thanks | 21:07 |
brocashelm | it's also happening on arch, so likely an upstream bug | 21:13 |
rwp | https://bugs.debian.org/996174 wget: Spurious SSL_INIT in 1.21.2-1 | 21:33 |
rwp | I almost never am fast enough to be the first filer so we will see if it turns out to be a duplicate or not. :-) | 21:33 |
rwp | brocashelm, Fixed upstream already: https://git.savannah.gnu.org/cgit/wget.git/commit/?id=3ea9658c072fa1dc87aec7af45393b46904b143f | 21:37 |
rwp | And here is the commit with the "accident" https://git.savannah.gnu.org/cgit/wget.git/commit/src/gnutls.c?id=5fe8d26904880ab59b2343f31ff7399edbbc43f1 | 21:38 |
brocashelm | nice | 21:38 |
rwp | Fixed in the main trunk but Sept 7 v1.21.2 was tagged and released which included the bug. Fixed Sept 9. But no release tag made since. | 21:42 |
rwp | Which for something like this would be really good if they tagged it and made a release so that distros could and would just pass it through. | 21:42 |
rwp | Upstream bug report: https://savannah.gnu.org/bugs/?61125 | 21:44 |
rwp | I posted a gentle nudge there since the maintainer originally reacted that a new release would happen within a week and it has now been a month. | 21:49 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!