KatolaZ | shnaps: ok but then gnarface is right in saying that you must have something else in that .xinitrc | 00:00 |
---|---|---|
shnaps | Hm, I can try to copy xinitrc from /X11 folder | 00:01 |
shnaps | I seen example somewhere | 00:01 |
KatolaZ | shnaps: what do you want to run in .xinitrc? | 00:01 |
KatolaZ | and, more importantly, are you using any login manager? | 00:02 |
KatolaZ | (like xdm, wdm, slim, etc.?) | 00:02 |
shnaps | Nope, not using right now. Going to use later. Slim or lightdm | 00:02 |
KatolaZ | then just simplify your life | 00:02 |
KatolaZ | get lightdm or slim installed | 00:02 |
shnaps | It is not so interesting! | 00:03 |
KatolaZ | oh boy :) | 00:03 |
gnarface | shnaps: just fyi, this might work as a way to finish an automated install, in principle, but there is actually something called preseeding that's actually for this purpose. you're sortof misusing the ~/.xinitrc here. | 00:03 |
gnarface | that's not why it's not working though | 00:03 |
shnaps | Basically, I want to put in .xinitrc stuff that supposed be in statusbar: wicd, volumeicon, some other icons | 00:03 |
gnarface | but if this did work, your Xorg startup would be really heavy | 00:03 |
KatolaZ | shnaps: then if you want to learn, you need to study | 00:03 |
KatolaZ | shnaps: start with a simple example in .xinitrc | 00:04 |
KatolaZ | like | 00:04 |
KatolaZ | exec xterm | 00:04 |
KatolaZ | and try to see what happens | 00:04 |
KatolaZ | then add stuff later | 00:04 |
shnaps | I have like 15 tabs with different examples and wiki pages rn | 00:04 |
KatolaZ | you can't start with a complex example and expect to make it work | 00:04 |
KatolaZ | walk first, run later | 00:04 |
KatolaZ | :) | 00:04 |
gnarface | shnaps: a bunch of the lines in this ~/.xinitrc, you wouldn't even want to be running more than once, let alone every time Xorg starts up | 00:05 |
gnarface | shnaps: i agree, try to get it working with a simpler version first :) | 00:05 |
shnaps | Rn my .xinitrc is empty | 00:05 |
gnarface | well, not THAT simple, obviously | 00:05 |
shnaps | And I wanted to have volumeicon everywhere | 00:06 |
KatolaZ | shnaps: minimal viable program | 00:06 |
KatolaZ | put in your .xinirc: | 00:06 |
KatolaZ | exec /usr/bin/xterm | 00:06 |
shnaps | What's next? | 00:07 |
KatolaZ | have you tried? | 00:07 |
shnaps | Not yet. So, do I need to exit i3? | 00:08 |
shnaps | Restart it? | 00:08 |
KatolaZ | obviously | 00:08 |
KatolaZ | how did you start i3? | 00:08 |
shnaps | (I'm just using irc on my laptop rn) | 00:08 |
shnaps | Well, it is a good question | 00:08 |
shnaps | startx and it works | 00:09 |
shnaps | I tried to find place where it is supposed to choose dm but no luck | 00:09 |
KatolaZ | mmmhhh | 00:09 |
KatolaZ | do you have other wms installed? | 00:09 |
shnaps | Nope | 00:09 |
shnaps | Fresh install, nothing here | 00:09 |
gnarface | wouldn't it just be /usr/bin/i3wm? | 00:09 |
shnaps | Also, restarting i3 | 00:10 |
KatolaZ | ok | 00:10 |
KatolaZ | then | 00:10 |
KatolaZ | ls -l /usr/bin/x-window-manager | 00:10 |
KatolaZ | sorry | 00:11 |
KatolaZ | ls -l /etc/alternatives/x-window-manager | 00:11 |
shnaps_ | Yeah, I got idea | 00:11 |
KatolaZ | it will point to i3 | 00:11 |
KatolaZ | actually, that choice is made in /etc/X11/Xsession | 00:12 |
KatolaZ | which is launched by X at startup | 00:12 |
KatolaZ | to find out if it should launch a login manager, a window manager, or just xterm | 00:12 |
shnaps_ | KatolaZ, yeah, it is linked to i3! | 00:13 |
KatolaZ | X is launched by xinit | 00:13 |
KatolaZ | actually | 00:13 |
KatolaZ | xinit will first launch the X server | 00:13 |
KatolaZ | then, it will execute ~/.xinitrc to start launching clients | 00:13 |
shnaps_ | So, I have 2 ways here: put code before i3 in .xinitrc or put code in i3config? | 00:14 |
KatolaZ | no shnaps_ | 00:14 |
KatolaZ | you should not put i3 in .xinitrc | 00:15 |
KatolaZ | there is no need | 00:15 |
KatolaZ | startx will start it atm | 00:15 |
KatolaZ | since Xsession will look for any login manager installed in yout system, and find none | 00:15 |
KatolaZ | then it will look for a window manager, and will find /usr/bin/x-window-manager | 00:16 |
shnaps_ | Which is pointed to i3 | 00:16 |
KatolaZ | (which points to /etc/alternatives/x-window-manager which points to i3) | 00:16 |
KatolaZ | so i3 will be started in any case | 00:16 |
shnaps_ | i3config then? | 00:17 |
KatolaZ | dunno i3 | 00:17 |
KatolaZ | but you can use ~/.xinitrc if you like | 00:18 |
shnaps_ | Got it! | 00:19 |
Latrina | KatolaZ: got it installed with openwrt | 00:24 |
Latrina | meant openrc lol | 00:25 |
KatolaZ | :) | 00:25 |
Latrina | you needed a tricky eye to spot it, thanks ^^ | 00:25 |
KatolaZ | you just need to know what to look for :) | 00:25 |
KatolaZ | it will be in expert-install by default in beowulf | 00:25 |
Latrina | yeah, indeed I didnt | 00:25 |
KatolaZ | hopefully with more choices ;) | 00:25 |
Latrina | thats grand. I am going to setup TDE on it now | 00:26 |
ejr | does anybody know how to turn off the laptop display in a devuan tty entirely, without X? i am trying sudo vbetool dpms off but this only echoes "Real mode call failed" and then does nothing. | 13:05 |
ejr | fwiw, setterm also doesnt work | 13:08 |
buZz | setterm doesnt work? | 13:13 |
buZz | why not | 13:13 |
buZz | do you get a actual error? | 13:13 |
r3boot | unload the console driver? | 13:14 |
ejr | buZz: no error | 13:15 |
ejr | r3boot? how do i do that? | 13:15 |
buZz | setterm --powersave off --powerdown 1 | 13:16 |
buZz | that doesnt make the display turn off after 1 minute? | 13:16 |
r3boot | ejr: recompile the kernel w/o the driver, make sure you set your console to something else then your display (serial for instance). What is it you're trying to achieve btw? | 13:17 |
buZz | oh maybe --powersave powerdown | 13:17 |
ejr | r3boot: i sometimes have a bigger screen attached to my laptop, so i want the laptop screen backlight to be off completely as I dont need it | 13:20 |
buZz | hehe | 13:20 |
buZz | thats quite a advanced topic yeah | 13:20 |
ejr | should be pretty basic imo though | 13:20 |
buZz | why? | 13:20 |
r3boot | With X, that's trivial; With a console, not so much | 13:21 |
buZz | i've never seen textmode to be able to do any fancy >1 monitor stuff | 13:21 |
ejr | yeah, with X I've got my keybindings for that | 13:21 |
r3boot | no, Xrandr | 13:22 |
ejr | r3boot: i will recompile the kernel later on, though I've never done that before and hope I won't break anything. Can I set the console during the process, or do i have to pass it as an argument? | 13:22 |
ejr | yeah, the keybindings are based on xrandr | 13:22 |
r3boot | nah, dont do that :) I was working under the assumption that you want to get rid of your console | 13:22 |
ejr | (i use i3wm) | 13:22 |
ejr | ahh ok | 13:22 |
rrq | sudo tee /sys/class/backlight/*/brightness <<< "100" | 13:23 |
rrq | sudo tee /sys/class/backlight/*/brightness <<< "0" | 13:23 |
buZz | hehe | 13:23 |
buZz | rrq: that would turn off the second monitor aswell | 13:23 |
buZz | but , nice option perhaps | 13:23 |
rrq | hmm it didn't for me | 13:24 |
ejr | i have two screen folders in /backlight, will check which one is which screen :) one sec | 13:24 |
KatolaZ | ejr: what's your plan to restore the backlight *after* you succeed? | 13:24 |
ejr | KatolaZ: i'd have backlight on my second screen still, so basically just setting it back to on/100 | 13:24 |
KatolaZ | fair enough :) | 13:25 |
ejr | or maybe a small script with entr or read | 13:25 |
r3boot | /sys/class/brightness is hooked up to a specific video driver, and afaik you cannot set the brightness value per monitor from there (checked with 4.20.2, intel gpu) | 13:25 |
ejr | yeah, also didnt work here | 13:25 |
ejr | well, i would also like to use this for my other laptop which is currently used as a server, just for energy saving and hardware wearout reasons | 13:26 |
ejr | and i remember once having done it under (i think) debian | 13:27 |
jyri | anything from: sudo find /sys -name brightness ? | 13:39 |
jyri | or sudo find /sys -name "*backlight*" ? | 13:42 |
ejr | jyri: yes, many lines pointing to particular files, but i would not know which one to modify | 13:46 |
ejr | the latter command gives as files backlight, intel_backlight, disable_backlight_sysfs_if, enable_dpcd_backlight | 13:47 |
ejr | the first command even more | 13:47 |
jyri | on the first you can ignore the lines with usb in them, anything else in there? | 13:48 |
KatolaZ | ejr: which make and model is your laptop? | 13:50 |
jyri | ejr: this page might give some hints https://wiki.archlinux.org/index.php/Backlight | 13:51 |
ejr | jyri: this is the entire output: https://pastebin.com/5kwxW2qb | 13:51 |
ejr | KatolaZ: thinkpad x60 (librebooted) | 13:52 |
jyri | ejr: ok, lines 8 and 9 | 13:53 |
KatolaZ | then ejr you need to use /sys/class/backlight/intel_backlight/brightness | 13:54 |
KatolaZ | ejr: the value you put in there must be smaller than max_brightness | 13:55 |
KatolaZ | (a file in the same folder) | 13:55 |
ejr | ok, the current value of that file is 4426545. i will echo a 0 to it now. if i don't reply within the next minute, i guess that means darkness forever on my side, haha | 13:55 |
KatolaZ | :D | 13:55 |
KatolaZ | just get an ssh console | 13:55 |
ejr | yay, it works! | 13:56 |
ejr | thanks a lot | 13:56 |
rrq | ejr: intel_backlight/max_brightness tells maximum for brightness | 13:57 |
ejr | yes, just looked at it. it's the same value as brightness had before | 13:57 |
KatolaZ | ok | 13:57 |
KatolaZ | good | 13:57 |
ejr | just tried, it also works on my server laptop (which is also an old x60) :) | 13:58 |
ejr | now the only question is how i can bind this to a tmux binding that works, because i first need to be root (sudo echo ... doesn't suffice apparently) | 14:10 |
rrq | sudo sh -c "echo 100 > /sys/class/backlight/intel_backlight/brightness" .. it's the writing to the file that needs root | 14:14 |
rrq | s/10// | 14:14 |
_stephen_ | You can make a udev rule to make it so anyone can write the file. | 14:15 |
_stephen_ | Or switch that redirect to a sudo pipe... | 14:15 |
_stephen_ | echo 100 | sudo tee /sys/class/backlight/intel_backlight/brightness >/dev/null | 14:16 |
Tuor | sudo might need password... Maybe write the `echo ...>...` line into a file owner:root;group:root; 755 and then add a sudo rule, so you can run the file with out any password. | 14:16 |
ejr | rrq's command works fine, thanks | 14:17 |
ejr | i think i'll stick with that solution | 14:17 |
ejr | (i have the sudo PW deactivated anyway) | 14:17 |
Tuor | ejr: ah then my stuff is not required. :) | 14:18 |
Tuor | Ah co-worker is running devuan on his laptop and most stuff works. I might be one of the next... Comming from Ubuntu 16.04 (can't upgrade to 18.04 for obvious reasons). | 14:20 |
sxpert | what sort of obvious reasons that don't appear obvious to me ? | 14:21 |
Tuor | ubuntu 18.04 has systemd only. no other init available. | 14:22 |
sxpert | ah yeah | 14:22 |
sxpert | of course | 14:22 |
sxpert | the antichrist | 14:22 |
Tuor | ^^ | 14:25 |
Tuor | my distribution was very great (KDE Neon) until they decided to switch to Ubuntu 18.04 and not support 16.04 any more... | 14:25 |
James1138 | General question about Devuan. Kodi 18 just came out Jan 28th. Will it be available for Devuan as it is now available for Ubuntu? | 14:42 |
KatolaZ | James1138: it might be available in beowulf | 14:42 |
KatolaZ | https://tracker.debian.org/pkg/kodi <- James1138 | 14:43 |
KatolaZ | it looks like only 17 is in unstable | 14:43 |
KatolaZ | (man Jan 28th is TWO days ago...) | 14:43 |
ejr | is devuan's testing/unstable soon going to be as up-to-date as that of debian? | 14:46 |
sxpert | there's "stale" and "usable" ;) | 14:46 |
KatolaZ | ejr: they are more or less on the same page | 14:47 |
ejr | hmm, still thinking about upgrading to testing... but on the other hand i don't want to break anything. | 14:50 |
ejr | weÄll see | 14:50 |
ejr | *we'll | 14:50 |
KatolaZ | ejr: maybe wait for a couple of days more | 14:50 |
KatolaZ | there is an issue with policykit which will be fixed soon | 14:51 |
ejr | good to know, thanks | 14:52 |
_abc_ | The latest irssi for devuan ascii per apt-get update + later install irssi is 1.0.7-1~deb9u1 but irssi.org has irssi 1.1.2 out and a security fix from 2019 already. Is there a way to upgrade other than manually? | 15:03 |
KatolaZ | _abc_: the security fix seems to be already in 1.0.7-1~deb9u1, AFAIK | 15:04 |
djph | _abc_: you are aware how debian "stable" behaves with versions, right? | 15:04 |
KatolaZ | _abc_: 1.1.2 is in testing and unstable | 15:05 |
KatolaZ | https://tracker.debian.org/pkg/irssi | 15:05 |
_abc_ | djph: I did not ask whether manual upgrade is the only option by hazard | 15:05 |
_abc_ | ok so the latest 1.1.2 bug was only affecting 1.1.x not 1.0.x | 15:05 |
KatolaZ | _abc_: I don't know | 15:05 |
KatolaZ | look at the CVE please | 15:05 |
_abc_ | ascii is jessie-backports level right? | 15:06 |
KatolaZ | nope | 15:06 |
djph | ascii is stable ... | 15:06 |
KatolaZ | ascii is ascii | 15:06 |
buZz | ascii-backports exists | 15:06 |
KatolaZ | and merges stretch | 15:06 |
KatolaZ | https://irssi.org/security/ | 15:06 |
djph | jessie is oldstable | 15:06 |
_abc_ | stretch does not appear on the page KatolaZ linked above at debian | 15:06 |
KatolaZ | o_O | 15:07 |
buZz | _abc_: fyi, irssi github master has a native OTR plugin ;) | 15:07 |
_abc_ | KatolaZ: search for stretch the word? | 15:07 |
KatolaZ | _abc_: which one? | 15:07 |
KatolaZ | stable | 15:07 |
_abc_ | https://tracker.debian.org/pkg/irssi | 15:07 |
KatolaZ | stretch is the codename | 15:07 |
_abc_ | Yeah so I hate hate hate debian's codename release names. | 15:08 |
* KatolaZ shrugs | 15:08 | |
_abc_ | Thanks for the quick answer. | 15:09 |
buZz | we aim to please | 15:09 |
djph | KatolaZ: devuan is also sticking to 'oldstable' / 'stable' / 'unstable' / 'testing' to describe the various versions, right? | 15:09 |
KatolaZ | nope | 15:09 |
KatolaZ | devuan uses codenames | 15:09 |
KatolaZ | mainly | 15:09 |
_abc_ | djph: it's a conspiration. | 15:09 |
KatolaZ | because the releases are not always aligned with the corresponding debian one | 15:09 |
KatolaZ | _abc_: read above ^^ :) | 15:10 |
KatolaZ | this is just to avoid confusion | 15:10 |
djph | KatolaZ: OK, so there's not necessarily a concept that "jessie is oldstable" for example | 15:10 |
_abc_ | It "avoids" is "just fine" | 15:10 |
KatolaZ | djph: nope | 15:10 |
KatolaZ | djph: the main issue is that a devun version might be still unreleased (testing) while the corresponding debian version is stable | 15:11 |
KatolaZ | we are currently in a time of grace | 15:11 |
KatolaZ | where this is not the case | 15:11 |
KatolaZ | (everything is in sync) | 15:11 |
KatolaZ | but it has not always been like that | 15:11 |
djph | KatolaZ: er, wait, I think we're saying the same thing but from different points of view. | 15:12 |
KatolaZ | we had devuan jessie still to be released, and ascii incomplete, when stretch became stable | 15:12 |
djph | KatolaZ: what I meant was that "within the confines of the Devuan project, there are four active versions mapped to the descriptors 'oldstable' / 'stable' / 'unstable' / 'testing' ... " | 15:13 |
KatolaZ | devuan jessie was stable for about one year while debian stretch was stable | 15:13 |
KatolaZ | yes djph | 15:13 |
KatolaZ | but I would not put that much attention on those | 15:13 |
KatolaZ | it's better to use codenames | 15:13 |
KatolaZ | in order to avoid confusion | 15:13 |
djph | KatolaZ: okay, perfect. I don't have to re-learn new words to think of the codenames. | 15:13 |
KatolaZ | you actually might need to, djph | 15:14 |
djph | KatolaZ: err... | 15:14 |
KatolaZ | if debian buster becomes stable tomorrow and devuan beowulf takes three more months, then what is stable in debian is testing in devuan | 15:14 |
djph | KatolaZ: "so"? | 15:14 |
KatolaZ | that's totally clear for me | 15:14 |
djph | I don't care about Debian. | 15:14 |
KatolaZ | and maybe for you as well | 15:14 |
KatolaZ | but apparently this causes "distress" | 15:15 |
KatolaZ | :) | 15:15 |
djph | Short version is that as long as "I" can still think of Devuan Jessie as "oldstable" and Devuan Ascii as "stable", I can keep things straight. | 15:15 |
KatolaZ | yep | 15:16 |
djph | While still talking as "Devuan Jessie" to others. | 15:16 |
KatolaZ | issues arise if you make comparisons between a devuan version and the corresponding (merged) debian one | 15:16 |
djph | in which case, it'd be codenames (and some googling) so I was accurate. | 15:16 |
KatolaZ | that's why it's better to say "Devuan ASCII" and "Devuan Beowulf" instead than "Devuan stable" and "Devuan testing" | 15:17 |
KatolaZ | yep | 15:17 |
djph | like, I didn't know Beowulf was the upcoming version. | 15:17 |
KatolaZ | djph: www.devuan.org | 15:17 |
KatolaZ | :) | 15:17 |
djph | KatolaZ: As if I have time to read upcoming things (much less test them) with a 2 year old running around | 15:18 |
KatolaZ | djph: don't mention that to me :P | 15:22 |
djph | KatolaZ: have kiddos of your own? | 15:23 |
KatolaZ | :) | 15:23 |
djph | well, this is #1, so every day is a new adventure | 15:24 |
KatolaZ | it will get better | 15:24 |
djph | Oh, I'm not saying it's not great. Just, well, time-consuming :) | 15:24 |
Liebach | 2-year olds are endless fun, and a lot of work. :-) Enjoy it while it lasts. | 15:25 |
djph | haha, yeah, it's definitely that | 15:26 |
djph | (uh, and sorry for going way ot, before the ops tell me off) | 15:26 |
KatolaZ | djph: I am not saying it's not great either | 15:28 |
KatolaZ | :) | 15:28 |
KatolaZ | a lot of fun, and it gets better | 15:29 |
djph | and then worse, when she discovers boys. | 15:29 |
djph | (aaaa) | 15:29 |
fsmithred | or girls | 15:29 |
djph | at least with girls there's a pretty low chance of getting knocked up ... :| | 15:31 |
fsmithred | good point | 15:31 |
buZz | you'd think that | 15:31 |
KatolaZ | well... | 15:31 |
buZz | but nowadays there girls with a outtie | 15:31 |
buZz | and boys with a innie | 15:31 |
Liebach | If your kids do stupid things, they have it from you. So, be your very best. When my son become a man, he will be somewhat like me. When my daughter drags her first boyfriends home and present them, I will see part of myself in them. I better like what I see. | 15:32 |
buZz | welcome to teh futurrrrre | 15:32 |
djph | buZz: I hate you. | 15:32 |
buZz | :( | 15:32 |
buZz | i dont dislike you | 15:32 |
buZz | full disclosure, i'm dating a girl with a daughter ;) | 15:32 |
djph | buZz: more for ruining my little walled garden that girls don't have outies... | 15:32 |
djph | Liebach: quite. | 15:33 |
buZz | djph: ppl didnt like me when i made gender a longint either | 15:33 |
buZz | now they're all happy | 15:34 |
djph | ugh | 15:34 |
buZz | adam's apple removal surgery is supposedly one of the hardest, so you might like that cheat to check | 15:35 |
buZz | meh nvidia-driver 390.87-6~bpo9+1 in ascii-backports still :( | 15:37 |
buZz | i want 4xx already | 15:37 |
buZz | and totally dont want to install it manually :P | 15:37 |
Liebach | Any guesses when beowulf become stable? | 15:40 |
KatolaZ | Liebach: when it's ready? | 15:40 |
fsmithred | certainly will happen after buster freezes | 15:40 |
Liebach | KatolaZ: Obviously. Just asking because if it could happen within the next 2 weeks I'd rather wait, instead of installing ASCII now. | 15:43 |
KatolaZ | Liebach: it seems unrealistic for it to happen in the next two weeks | 15:43 |
Liebach | KatolaZ: Thanks. | 15:43 |
KatolaZ | yw | 15:43 |
_abc_ | If devuan development cycles will follow debian's in timing in any kind of way, we'd have a few years between releases I think. | 16:52 |
furrywolf | we should do the web browser thing, and immediately release devuan 10, followed by a new version every time we fix a typo in a man page. | 16:52 |
KatolaZ | furrywolf: that would be called ubuntuan | 17:05 |
golinux | Advice to all here reveling in parenthood: http://vhemt.org | 17:24 |
furrywolf | golinux: that's offtopic, take it to #debianfork. :) | 17:27 |
* furrywolf hides | 17:27 | |
* FlibberTGibbet peers under the carpet in case furrywolf is hiding there | 17:46 | |
furrywolf | nope, but there might be a wolf-sized pile of shed fur there. :) | 17:47 |
MinceR | nakedwolf? | 17:50 |
furrywolf | heh, you've never seen how much fur a canine can shed? :P | 17:52 |
fsmithred | I have. Friends of mine have a rake for their living room. Not kidding. | 17:53 |
golinux | furrywolf: Touche!! | 17:55 |
PinkBellyNagger | syslog is reporting some hard disks in my mdadm raid5 are over 100c | 20:11 |
PinkBellyNagger | My bet this is a bug by pottering | 20:11 |
gnarface | still, maybe you should check the case fans | 20:14 |
PinkBellyNagger | gnarface I've checked everything is fine. hddtemp reports normal temps | 20:14 |
PinkBellyNagger | ilo looks fine too | 20:15 |
PinkBellyNagger | It's a bug in debian | 20:15 |
furrymcgee | there is recurring sytemctl not found with different packages, could devuan provide its own systemctl? | 20:55 |
amesser | why? | 20:56 |
amesser | the point is to not depend on systemctl(systemd) | 20:57 |
furrymcgee | some packages use it for installation, maybe a dummy or replacement is possible | 20:59 |
amesser | whell, systemctl is used in most cases to manage systemd services. So if a package does not ship with e.g. init script, then it wont help to emulate systemctl. And if a package ship with init and systemd.service files, it should detect automatically if systemd or sysvinit is required | 21:06 |
amesser | Or even better, user the command "service" instead. it should work on systemd and sysvinit systems. | 21:06 |
gnarface | seems like a logical argument. one has to wonder why Debian didn't do it that way to begin with. | 21:10 |
gnarface | it's almost as if they're purposefully making choices that will increase compatibility issues even when it's clearly more work to do so | 21:10 |
amesser | For me its more the packagers responsibility. there now rule against supporting sysvinit. | 21:12 |
amesser | One could try to file a bug against apackage which doessn't work properly on sysvinit | 21:12 |
furrymcgee | yes I think of a common script named /bin/systemctl that executes /usr/sbin/service with the same arguments | 21:12 |
amesser | urgs, my today's typing is even more worse than usually | 21:13 |
amesser | it wont help, if the package doesn't install /etc/init.d/<package> script. | 21:14 |
amesser | also, in many cases, the systemd service files have different names than the sysvinit scripts | 21:14 |
gnarface | of course they do. why would they bother being even a little orderly. | 21:15 |
* gnarface sighs | 21:15 | |
amesser | actually, the post install hook of a package could decide. But this would require some manual handcrafting by the packagers, it wont be handled by deb-builder magic | 21:15 |
furrymcgee | lets do this ln -s /bin/true /bin/systemctl | 21:34 |
KatolaZ | furrymcgee: which packages? | 21:35 |
KatolaZ | please report them to bugs.devuan.org | 21:35 |
furrymcgee | one is third party repository the other is apache in a devuan chroot that checks pid1 of debian and calls systemctl | 21:38 |
rwp | It would be good to collect up at least an initial list of packages that are offenders in the systemctl/service issue. To be able to understand the general problem. | 22:31 |
rwp | And then I think it would be beneficial to have a systemctl script in Devuan that calls service after checking that the packge has installed a sysvinit compatible script. | 22:31 |
rwp | If not then it can complain. | 22:31 |
rwp | This could facilitate in automated finding of problems by installing and removing every package in devuan. | 22:31 |
rwp | Did I say "service" in the above? I meant to say "invoke-rc.d" for postinst scripts instead. | 22:32 |
rwp | Actually... (brain? Where brain?) I meant to say "update-rc.d" since that is really what I meant to say. | 22:34 |
ttkv | for your edification, application or mirth: | 23:39 |
ttkv | https://capsule8.com/blog/exploiting-systemd-journald-part-1/ | 23:39 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!