| joerg | #eudev | 03:23 |
|---|---|---|
| hightower2 | On daedalus, in console, why doesnt shift+pg/up down scroll the buffer? | 17:49 |
| gnarface | hmm, it should, really | 17:50 |
| gnarface | i suppose i could imagine a couple reasons why it wouldn't | 17:51 |
| gnarface | maybe a framebuffer driver issue, or maybe it just loses history when you switch consoles... | 17:51 |
| gnarface | i think i've seen that happen a couple times anwyay | 17:51 |
| gnarface | *anyway | 17:51 |
| fsmithred | shift-pgup is gone. I was reminded of this recently. The workaround is supposed to be to use tmux. | 17:54 |
| debdog | (or screen (just to be complete)) | 17:57 |
| debdog | but yes, I missed shift-pgup, too | 17:58 |
| gnarface | damn, really? | 18:04 |
| gnarface | wtf | 18:04 |
| gnarface | more fucking vandlism | 18:04 |
| gnarface | *vandalism | 18:04 |
| gnarface | any way to re-enable that by a kernel rebuild or something? | 18:05 |
| debdog | gnarface: https://www.linuxquestions.org/questions/showthread.php?p=6167719 | 18:07 |
| debdog | .....The one thing that does show up in the diffstat is the softscroll removal (both fbcon and vgacon), and there are people who want to save that, but we'll see if some maintainer steps up. I'm not willing to resurrect it in the broken form it was in, so I doubt that will happen in 5.9, but we'll see what happens..... | 18:09 |
| gnarface | :( | 18:11 |
| rustyaxe | fbcons/vgacons have been broken for decades really | 18:13 |
| rustyaxe | at least a good decade | 18:13 |
| fsmithred | ok, so how do I get shift-pgup to work in screen? | 18:14 |
| fsmithred | is there some ctrl-? to do it? | 18:14 |
| debdog | .oO( mayhap I should add screen/tmux to start upon login for ttys. so I am able to scroll back if needed witout having to remember to start screen/tmux each time) | 18:14 |
| rustyaxe | ctrl-A pgup/dwn, fsmithred? | 18:14 |
| debdog | fsmithred: ctrl-a enters ... | 18:15 |
| debdog | what rustyaxe said | 18:15 |
| fsmithred | I'l ltry it | 18:15 |
| debdog | ctrl-a ESC exits the edit mode | 18:15 |
| rustyaxe | ctra-[ then pgup/pgdwn sorry | 18:15 |
| debdog | erm, just ESC | 18:15 |
| rustyaxe | i use tmux so its a bit different here | 18:15 |
| debdog | just tried ctrl-a and ctrl-[ seem to be the same (but on a non qwery keyb ctrl-a ESC is easier to type) | 18:16 |
| debdog | erm ctrl-a ESC | 18:17 |
| debdog | just tried "ctrl-a ESC" and "ctrl-[" seem to be the same (but on a non qwery keyb ctrl-a ESC is easier to type) | 18:17 |
| debdog | dang, too tired, sorry | 18:18 |
| debdog | ctrl-a ESC vs. ctrl-a [ | 18:18 |
| * debdog shuts up no | 18:18 | |
| buZz | debdog: ctrl-a [ | 18:18 |
| debdog | w | 18:18 |
| buZz | you dont need the second ctrl | 18:18 |
| debdog | right | 18:18 |
| debdog | sorry | 18:18 |
| buZz | np | 18:18 |
| buZz | thought you didnt know, was just aiding :) | 18:19 |
| fsmithred | ok, I got it to work once. Can someone tell me what sequence of keystrokes I used? | 18:22 |
| debdog | either ctrl-a ESC | 18:23 |
| debdog | or ctrl-a [ | 18:23 |
| debdog | this enters copy mode one thing that is capable of is scrolling | 18:24 |
| debdog | to exit this mode hit ESC | 18:24 |
| fsmithred | ok ctrl-a [ | 18:24 |
| fsmithred | thanks | 18:25 |
| buZz | its copy mode, you can put cursor on first postition, and press enter, put cursor on last position , press enter | 18:25 |
| buZz | you exit the mode then | 18:25 |
| buZz | ctrl-a ] is 'paste' then | 18:25 |
| buZz | :) | 18:25 |
| buZz | -sometimes- really handy | 18:25 |
| debdog | fsmithred: in ~/.screenrc I have added "defscrollback 5000" to have a longer history (default is 100) | 18:27 |
| debdog | lines that is, IIRC | 18:28 |
| debdog | esp. when compiling things this is useful | 18:29 |
| fsmithred | yeah, thanks | 18:29 |
| fsmithred | for years, I used to hit the scroll lock key to pause long output and then I leared about shift-pgup. | 18:30 |
| debdog | hehe, try scroll-lock with a 3 GHz CPU | 18:31 |
| debdog | I understand | 18:31 |
| buZz | just set your tty to 9600 baud | 18:35 |
| buZz | :D | 18:35 |
| rwp | I find that loss of vt console scroll back most annoying at boot time when I want to scroll back to read the boot messages which scrolled off. Can't use screen/tmux for those messages. So it is just a loss of functionality. | 18:46 |
| buZz | rwp: you're making a good point for hooking up old matrix printers for /dev/lp0 and redirect console to there ;) | 18:47 |
| buZz | feels very retro too :D *boot computer* *prrrrt prrrrrt prrrrrrt* paper rolls out :D | 18:47 |
| rwp | I presume that some government office still has a contract to buy tractor feed zig-zag computer paper but I would have no idea where to find it myself now. | 18:49 |
| buZz | :) | 18:50 |
| buZz | the building we were donated for the hackerspace, came with a couple of boxes of it , and a boxed Star LC-20 | 18:50 |
| rwp | A hackerspace would be a good place for a retro computer museum setup! | 18:51 |
| hightower2 | jezus christ... when will debilization of unix stop | 18:58 |
| hightower2 | or well, 'sane' linux anyways | 18:58 |
| buZz | rwp: we have some microVAX, a VT420 , 1 or 2 decwriters , a PiDP 11 , loads of stuff | 18:58 |
| hightower2 | I'm tired of stupid defaults being introduced, or things like this shift_pgup removed | 18:59 |
| hightower2 | like what the hell | 18:59 |
| buZz | rwp: feel welcome to come lick them or whatever ;) | 18:59 |
| buZz | haha | 18:59 |
| buZz | hightower2: are you talking about xterm? | 18:59 |
| hightower2 | it doesn't matter that we have devuan when things change for the worse in underlying aps | 18:59 |
| buZz | i doubt anyone remembers where in sourcecode shift+pgup/dn is defined :D | 18:59 |
| hightower2 | buZz, no the console (my question from some pages above) | 18:59 |
| buZz | ah, the textmode console without kms? | 19:00 |
| buZz | because, if you switch modes the buffer gets wiped | 19:00 |
| buZz | and sometimes the 'first displayed prompt' clears the buffer too , regardless of modes | 19:00 |
| hightower2 | no I am on the same term, didn't change console or anything | 19:00 |
| hightower2 | see responses I got for clarification | 19:01 |
| buZz | quite sure debian has been clearing that buffer for 'security' since debian 9 or 10 | 19:01 |
| buZz | weird, would my bookworm here do the same? /me tries | 19:01 |
| buZz | gee, indeed | 19:02 |
| buZz | TIL :) | 19:02 |
| buZz | i guess some linux 6.x change? | 19:02 |
| dan9er | Hello everyone! I'm going to attempt to rescue some data from a laptop with an inflating non-removable battery. It's running x64 Chimaera. My plan is to install `openssh-server` on it (with pre-downloaded .debs) and SSH in from afar to move files to a USB. How do I quickly set it up so I can SSH in with a password? | 19:04 |
| gnarface | dan9er: i think user passwords access is still enabled by default, just not root. you can "su -" after login and do whatever you need to do | 19:06 |
| gnarface | (i would be more sure of that if i hadn't just discovered that console scrollbacks are no longer available, but that's an unrelated issue) | 19:07 |
| dan9er | I recall reading somewhere I had to add users to the `sshusers` group | 19:07 |
| * dan9er realizes the PC he's typing on is also Chimera and he can just test it with his laptop | 19:08 | |
| gnarface | i haven't had to do any such thing on any of my installs here, but they're almost all upgrades from previous releases | 19:11 |
| gnarface | perhaps you're thinking of the "sudo" group for sudo? | 19:12 |
| gnarface | you do have to add yourself to the sudo group to get sudo access, but you don't need sudo to get root as long as your install enabled a root password (and you know what it is) | 19:12 |
| gnarface | and depending on what you're actually doing and how your install was setup you might not need root at all to backup what you want | 19:13 |
| dan9er | Now I'm thinking that `sshusers` group came from that "How To Secure a Linux Server" guide, that's an explicitly configured thing | 19:16 |
| gnarface | seems plausible | 19:16 |
| gnarface | if you're sharing the system with multiple users it certainly would be a good idea to make a special group for the ones who get ssh access | 19:17 |
| gnarface | and i don't know for sure that there aren't other distros that enforce that by default | 19:17 |
| dan9er | gnarface: huh, it indeed just works out-of-the-box | 19:25 |
| dan9er | Well that's both a relief and a bit concerning | 19:26 |
| dan9er | Right, thank you! | 19:27 |
| AlexLikeRock | O/ | 19:39 |
| rwp | buZz, "i doubt anyone remembers where in sourcecode shift+pgup/dn is defined :D" ahem... https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=20782abbbdfe922496a28f9cc0c3c0030f7dfb8f | 19:45 |
| buZz | <3 | 19:46 |
| buZz | rwp: but that comment was about xterm's source being archaic and ancient and barely maintainable ;) | 19:46 |
| buZz | lol the guy that removed it is called 'nopnop nop' ? :D | 19:47 |
| rwp | xterm? If we mention that three more times then Thomas Dickey will appear and answer our questions about it. (Seriously! He has a web search where he is notified of mentions of it.) | 19:48 |
| rwp | I don't see where we were talking about xterm as I thought we were talking about the Linux vt console. | 19:48 |
| buZz | sweet! | 19:48 |
| buZz | rwp: yeah sry, i assumed it was about gui before reading full backlog ;) | 19:48 |
| rwp | I don't see Thomas on IRC channels but in mailing lists that is definitely true. :-) | 19:49 |
| buZz | maybe he wants to update osso-xterm for Maemo Leste ;) | 19:49 |
| buZz | otoh its fine as-is | 19:49 |
| rwp | And then was there something about getty clearing the terminal? I just skimmed and was not sure. That changed a decade ago and I have been adding --noclear to my getty options ever since. | 19:49 |
| rwp | buZz, I used a DEC Writer terminal quite a bit in my university days! That brings back the memories. | 19:50 |
| buZz | we brought one to a makerfaire lastyear, and let children interface with our uncensored chatgpt instance hosted at the hackerspace :P | 19:52 |
| buZz | we have nice long printouts of 6yos trying to outsmart a chatbot , its great fun | 19:52 |
| rwp | Ha! | 19:52 |
| buZz | the kids usually won once someone helped them with spelling :P | 19:53 |
| buZz | </offtopic> | 19:53 |
| joerg | >>So rather than try to maintain a ***likely*** unused set of code, I'll just ***aggressively remove it***, and see if anybody even notices<< committer Greg Kroah-Hartman | 21:22 |
| joerg | :-S | 21:23 |
| golinux | The INFAMOUS Greg Kroah-Hartman! | 21:26 |
| AlexLikeRock | jejej | 21:26 |
| golinux | No surprise there. He's been at it for years | 21:26 |
| golinux | Hi AlexLikeRock! | 21:27 |
| mrnhmath | didn't he try to remove usb tethering | 21:27 |
| mrnhmath | or was it someone else | 21:27 |
| AlexLikeRock | hi | 21:27 |
| AlexLikeRock | jejeje | 21:27 |
| AlexLikeRock | 2 weeks ago we read each other, but you were drowsy : golinux | 21:29 |
| golinux | Tea time now which should wake me up. :D | 21:31 |
| rwp | The committer was GKH but the author was listed as Linus Torvalds so as the old saying goes the fish rots from the head to the tail. | 21:37 |
| mrnhmath | those people are nuts | 21:52 |
| * AlexLikeRock go to play Final Fantasy X | 21:54 | |
| buZz | joerg: ah that was GKH , not 'nopnop nop' :D | 21:56 |
| buZz | much better | 21:56 |
| joerg | anyway, how hard would it be to roll back that commit? O had no problems with the scrollback when it was still alive, so I don't exactly bother a lot about some "bugs / bitrot" they claim they found in there | 22:00 |
| buZz | it seems to be mostly contained to just those two files , shouldnt be very hard? | 22:02 |
| buZz | you could try plonking 1:1 copies from kernel-prior over it | 22:02 |
| buZz | there's likely 'a proper git way to do this' but i dont know it | 22:03 |
| debdog | why is this suddenly an issue? this has been done since at least chimaera? not one of you guys realised it since? | 22:03 |
| buZz | i dont really care about it :P | 22:04 |
| debdog | where was the outcry years back?! | 22:04 |
| buZz | we didnt start the fire | 22:04 |
| buZz | ;) | 22:04 |
| debdog | just curious | 22:04 |
| buZz | just topic du jour | 22:04 |
| joerg | debdog: it's a feature I need rarely but _when_ I do, it's a pain to not have it available. That's the reason it went "unnoticed" for quite a while, until rwp(?) mentioned it not being me and my senile memory but actually a kernel regression. Ever since this is an annoyance lingering on | 22:08 |
| debdog | <hightower2> On daedalus, in console, why doesnt shift+pg/up down scroll the buffer? – hightower started it | 22:09 |
| buZz | thanks hi<tab> ! | 22:09 |
| joerg | and honestly >>${mess_with_core_features_of_a_system_used_in_millions_of_installations} and see if anybody even notices<< is simply not tolerable approach, not even for FOSS | 22:10 |
| buZz | well | 22:10 |
| buZz | i do agree with 'we should remove it if nobody cares enough to maintain it for newer interfaces' | 22:10 |
| buZz | as seems to be mentioned in that commit removing it | 22:11 |
| buZz | i mean, you could always just run a older kernel, i guess | 22:11 |
| debdog | that was four(!) years ago! and the feature did not just got removed because, but because there were issues beforehand https://lkml.iu.edu/hypermail/linux/kernel/2009.2/05870.html | 22:13 |
| joerg | on that rationale, we SHOULD remove that liblzma(?), no? | 22:13 |
| debdog | "The one thing that does show up in the diffstat is the softscroll | 22:13 |
| debdog | removal (both fbcon and vgacon), and there are people who want to save | 22:13 |
| debdog | that, but we'll see if some maintainer steps up. I'm not willing to | 22:13 |
| debdog | resurrect it in the broken form it was in, so I doubt that will happen | 22:13 |
| debdog | in 5.9, but we'll see what happens." | 22:13 |
| Xenguy | Late to the conversation but in both tmux and screen I use 'Ctrl-a [' to enter 'copy mode', then use PgUp/Down to scroll (as someone mentioned a nice big scrollback can be configured in screen with: defscrollback 5000 )... | 22:13 |
| debdog | joerg: do not get me wrong, I am missing that feature, too. but the outcry is too late now | 22:13 |
| Xenguy | The only difference between tmux and screen that I've seen is exiting 'copy mode': in screen I think I use 'Esc', and in tmux it wants 'Enter' instead | 22:14 |
| Xenguy | As for pre-screen/tmux use cases, I think I have used Ctrl-s to freeze scrolling during bootup, and Ctrl-q to resume scrolling, but it's been awhile so YMMV | 22:15 |
| debdog | I was complaining about that missing feature since chimaera. but no one cared back then | 22:15 |
| debdog | why care now? | 22:15 |
| buZz | gee 5.9 already | 22:16 |
| buZz | guess it was ages ago yeah :D | 22:16 |
| Xenguy | It's annoying but we have a workaround, so I won't die on that hill : -) | 22:16 |
| buZz | the workaround is 'stop using a dec vt as workstation' :P | 22:17 |
| joerg | please explain "softscroll" to me | 22:17 |
| debdog | joerg: shift-pgup == softscroll | 22:19 |
| joerg | is that the same like a simple scrolling or even paging up/down? or is that a silly fancy "we don't want jerky line by line / page by page scroll, it has be be *smooth*, i.e. sub-line pixel-wise scrolling"? | 22:20 |
| buZz | nah :) | 22:21 |
| joerg | good | 22:21 |
| buZz | its 'the buffer doesnt exist in hw , vgacon is emulating it' | 22:21 |
| joerg | ta | 22:21 |
| joerg | >>I think I have used Ctrl-s to freeze scrolling during bootup<< which basicaly froze the complete system, or at very least the syslog process ;-D | 22:24 |
| debdog | ctrl-s stands for silent, right? exit that with ctrl-q, right? | 22:26 |
| joerg | stop | 22:26 |
| debdog | hmm | 22:26 |
| joerg | and qontinue ;-) | 22:26 |
| debdog | ok | 22:26 |
| joerg | it's software flow control | 22:27 |
| joerg | debdog: https://en.wikipedia.org/wiki/Software_flow_control#Representation | 22:39 |
| debdog | ahh, XON/XOFF got it | 22:42 |
| rwp | For both tmux & screen it is 'q' that exits copy scrollback mode. | 22:53 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!