Bjornn | I'm patiently waiting for 3.0 | 02:23 |
---|---|---|
ebisu | the beta version works perfect | 02:28 |
ebisu | other than the time not working out of the bo | 02:28 |
ebisu | *box | 02:28 |
Bjornn | oh, nice. haha | 02:31 |
ejr | hi! everytime i start devuan I am told that the group kvm does not exist. I don't use any virtualisation so that's fine, yet I would like to get rid of that warning. How can I do that? | 14:52 |
ullet | i don't know 'the right way' ejr but maybe you could just add the group kvm and be done with it | 14:59 |
ullet | groupadd is the command | 15:00 |
Bjornn | Guest79103 any success? | 15:19 |
GyrosGeier | Guest79103, I agree with ullet here -- just add the group | 15:43 |
GyrosGeier | the warning comes because the virtualization drivers are loaded, because your CPU is virtualization capable | 15:43 |
GyrosGeier | and loading the drivers causes the device node for them to appear, which needs to be given the appropriate group | 15:44 |
GyrosGeier | the three ways to get rid of that warning are a) create group, b) blacklist kvm driver, c) disable virtualization extensions in BIOS | 15:44 |
GyrosGeier | the latter two will require you to remember what you did when later on you do want to use virtualization | 15:45 |
Joril | It looks like there's a 4th way: modify /lib/udev/rules.d/50-udev-default.rules, it contains this line "KERNEL=="kvm", GROUP="kvm", MODE="0666", OPTIONS+="static_node=kvm" | 15:51 |
Joril | There's an issue on the systemd github too! https://github.com/systemd/systemd/issues/6360 | 15:53 |
Joril | Of course a WONTFIX :D | 15:54 |
MinceR | :> | 15:57 |
MinceR | wontfixd | 15:57 |
GyrosGeier | even my PPC box has /dev/kvm | 16:04 |
mason | If it's something we see in Devuan, how could it be something that could be reasonably fixed in systemd? | 16:25 |
slvr | there are usually many ways to fix something. | 16:45 |
GyrosGeier | yes | 17:27 |
GyrosGeier | Lennart is sometimes right about things not being systemd problems | 17:27 |
GyrosGeier | "this should be fixed properly, systemd is not there to work around your bugs" is a good stance, because it leads to bugs actually being fixed | 17:28 |
slvr | I don't agree with that as a rule but sure it is one way to do it. | 17:31 |
slvr | Pushing bugs out to end users to let them suffer until someone else fixes it is the modern way of the web, but it's not very nice to end users. | 17:32 |
GyrosGeier | right, but working around a bug means that it won't get fixed because it works for systemd users and the only people complaining are the usual suspects | 17:33 |
tmberg | apt-get install php7.4-fpm :-( | 17:47 |
mason | The point brought up is interesting and relevant. If the device node exists, that's independent of software using it. Maybe the user wants something utterly third-party. The error message itself is the issue. | 17:47 |
MinceR | strange, this means poettering believes there are things that do not belong in systemd | 17:51 |
MinceR | i wonder what those things are | 17:51 |
MinceR | as for making end users suffer, i'm pretty sure that's one of the main objectives | 17:51 |
ejr | hi. for some reason I cannot login to X anymore since today. I am hoping it's not related to the upgrade of my linux image. Running on devuan 10 | 20:25 |
gnarface | well, it's almost certainly related to that, but you're going to need to dig up a error message or something | 20:27 |
ejr | https://termbin.com/39xh | 20:28 |
ejr | that's the Xorg log | 20:28 |
gnarface | can you use paste.debian.net instead please? | 20:28 |
gnarface | i'll look at it if you do, i promise | 20:28 |
ejr | uh, not really because i cannot start x. lol | 20:28 |
gnarface | heh | 20:28 |
ejr | wait, i will check if paste.debian.net works with w3m | 20:29 |
gnarface | if it comes to it, you can get a cursor on the system terminal with gpm, but i think you can alter a config file for pastebinit to use paste.debian.net instead | 20:30 |
ejr | nah sorry, i have no way to paste it in there wiht w3m | 20:30 |
gnarface | can you just /msg it to me? | 20:30 |
gnarface | it's fine if it takes a long time to paste in private | 20:30 |
ejr | ok i will try pastebinit | 20:30 |
gnarface | this is a freshly upgraded image that login immediately failed on, or was it working for some days first? | 20:32 |
ejr | http://paste.debian.net/1143796/ | 20:32 |
gnarface | awesome | 20:32 |
gnarface | thank you for being cooperative about that | 20:33 |
ejr | i only installed the image today so i dont know. However I also wrote on a script that hacks on environment variables and permissions, so that might also be the reason, not sure... | 20:33 |
ejr | np | 20:33 |
gnarface | hmmm, is this an upgrade from a debian install? | 20:33 |
ejr | nope | 20:33 |
ejr | relatively recent beowulf install | 20:33 |
ejr | but i am running libreboot and I think there it's recognised as debian (also shows as Debian in grub, for example) | 20:34 |
gnarface | hmmm, maybe unrelatd | 20:34 |
gnarface | what does this command return for you? dpkg -l |grep logind | 20:35 |
gnarface | ejr: ^ | 20:36 |
ejr | installed are: elogind,libelogind0, libpam-elogind and two related polkit libs | 20:38 |
gnarface | alright, i'm running dry on ideas. you sure your capslock key isn't on? | 20:40 |
ejr | haha yes | 20:40 |
filipdevuan_ | hello | 20:41 |
filipdevuan_ | how can i delete the keyboard shortcut for switching desktops?? | 20:41 |
gnarface | ejr: the only thing suspicious i see in the log is this one: [ 361.860] (EE) Error systemd-logind returned paused fd for drm node | 20:41 |
ejr | i also had simultaneously weird issues with e.g. XDG_CONFIG_HOME not being recognised as an environment variable... and earlier on I got an error message that Xauthority couldnt be locked | 20:41 |
gnarface | ejr: (doesn't seem to be the issue though) | 20:41 |
gnarface | ejr: oooh. hmmm. i wonder... | 20:42 |
ejr | after i manually set the $XAUTHORITY without using $XDG_CONFIG_HOME (absolute path instead) it worked.. but now it doesnt anymore | 20:42 |
gnarface | ejr: XDG_RUNTIME_DIR, is that getting set? | 20:42 |
gnarface | ejr: i think they did change some XDG stuff and it caused me problems with enlightenment too, i had to start setting XDG_RUNTIME_DIR in my .bash_profile | 20:43 |
gnarface | ejr: (not sure if that's fixed or not) | 20:43 |
gnarface | filipdevuan_: it's gonna be different for every window manager. you'll have to check your's particular docs | 20:44 |
ejr | strangely enough, in the tty that i used startx in, it is not set, but in another one it is (e.g. in tmux). lol | 20:44 |
ejr | it should be set to /run/user/1000 right? | 20:44 |
gnarface | interesting | 20:44 |
gnarface | uh, it actually almost doesn't matter what it is set to. i'm not sure where it's even supposed to go, i pointed mine to /tmp | 20:45 |
gnarface | all that matters is it's unique for every user. there could be security issues otherwise (and obvious massive breakage) | 20:45 |
ejr | hmm, so if i add export XDG_RUNTIME_DIR="/some/path" to /etc/environment as well, it might work, you think? | 20:46 |
gnarface | hmm, not sure, but i assume so if that file is parsed like a shell script and has access to the user's environment variables | 20:47 |
gnarface | i just don't know (mine is empty here) | 20:47 |
gnarface | i put this in my ~/.bash_profile and it worked but it's fucking ghetto: http://paste.debian.net/1143800/ | 20:48 |
gnarface | (use at your own risk) | 20:48 |
ejr | the even stranger thing is that it worked earlier on, after I had the Xauthority issue fixed... but now it doesnt anymore... | 20:48 |
filipdevuan_ | on xfce | 20:49 |
filipdevuan_ | ok nvm cheers | 20:49 |
gnarface | ejr: yea, so if it worked after a full upgrade and a clean reboot, that suggests it's environmental and not package related, but just to be sure can you make sure you're fully upgraded? "apt-get update && apt-get upgrade && apt-get dist-upgrade" | 20:50 |
gnarface | ejr: is it possible you applied some fix manually and simply forgot to make sure it would persist through a reboot? | 20:50 |
ejr | ok, i will check upgrades, and I also just put your little script in my bashrc. will see if it works then, and return to let you know in some mins, brb | 20:50 |
gnarface | ejr: hmm, but if this was the issue i'd also expect some better error, maybe permissions related. did anything show up in dmesg or syslog about this perhaps? | 20:51 |
gnarface | drat, i'm too slow | 20:51 |
ejr | negative... | 20:55 |
ejr | It hung at startx for about a minute, then it read "Timeout in locking .config/Xauthority" and then after some seconds I just got a black screen on the tty that persists...lol | 20:55 |
gnarface | ejr: i forgot to be clear, that variable needs to be unique per user, not unique per ~/.bashrc, but the error you got suggests it was already being set right before that change, so yea, back it out. | 20:56 |
gnarface | *not unique per SHELL as ~/.bashrc would make it, i mean | 20:56 |
ejr | not sure what you mean. you mean the xdg script has to be in .bash_profile instead of .bashrc? | 20:57 |
gnarface | the error you got was the one i would expect for it being set wrong in the first place, so if you get that error setting it this way, but don't get the error otherwise, then it was probably already correct and this isn't the issue. | 20:57 |
gnarface | but yea, i explicitly specified ~/.bash_profile instead of ~/.bashrc on purpose. they aren't quite used the same. | 20:58 |
gnarface | (in ~/.bashrc what i would expect is it would work for the first login but not subsequent logins) | 20:59 |
ejr | ok, well, i dont even have a .bash_profile but i can try putting it in there. | 21:00 |
ejr | also, the blank screen on the startx tty just went away and now it reads "xauth: /home/user/.config/Xauthority not writable, changes will be ignored" | 21:00 |
gnarface | hmmm | 21:01 |
gnarface | that's after backing out my change then relogging? | 21:01 |
fsmithred | what's it doing in ~/.config? | 21:01 |
gnarface | i wonder if you have been getting this error all along or this is new? | 21:01 |
ejr | nope, not yet | 21:01 |
ejr | will do that now, brb | 21:01 |
ejr | ahh, i had that once, yeah, but not all the time | 21:02 |
ejr | brb | 21:02 |
gnarface | hmm, fresh install... i wonder what could have gone wrong? | 21:07 |
ejr | ok nguys, it's getting stranger and stranger.lol | 21:08 |
ejr | when I tried the xdg script in .bash_profile I could not run any commands, as if PATH was suddenly unset. Also aliases from .bashrc were not recognized anymore. X didnt work either ofc. I deleted .bash_profile again and your script too, but x does not start still, claiming that it has no authority to write to .config/Xauthority | 21:10 |
ejr | but when I run startx there first is a blackscreen on that tty for about 5 mins, and at least once that even transfers to my OTHER tty on which I am in tmux right now, running irssi | 21:10 |
gnarface | ejr: well, first of all, Xauthority shouldn't be in ~/.config to begin with. so something is already wrong with that, but secondarily it should be readable by your user, so that suggests two problems | 21:11 |
ejr | so then I have to manually reload that tty with CTRL+ALT+F2 to get back to irssi lol | 21:11 |
gnarface | ejr: should be *writable* by your user too, to be clear | 21:11 |
ejr | I set Xauthority to .config becuase I wanted it there, it can easily be set to that with the $XAUTHORITY var, and this has worked for me for months without any problems | 21:11 |
ejr | ah also, I just noticed that it now created new Xauthority files in .config, like Xauthority-l, Xauthority-n, ... | 21:13 |
gnarface | i'm just not sure if setting XAUTHORITY means you'll need to set other variables too to match | 21:13 |
gnarface | i'm pretty sure ~/.config/ isn't where they're supposed to be but that might not be what is breaking this unless the issue is they need to be in the same directory with something else | 21:14 |
ejr | and for some reason permissions on Xauthority were set root:root, and to user:user on the other two. But I deleted the Xauthority earlier on, so that must have happened afterwards | 21:14 |
ejr | i can temporarily unset $XAUTHORITY just to exclude that that's the problem ofc | 21:14 |
gnarface | it should be easy to fix the permissions on it | 21:15 |
gnarface | i don't know how they could have got set wrong though unless you launched it as root but used the user's home directory still | 21:15 |
ejr | just to be sure: I can delete Xauthority files with no worries, right? | 21:15 |
gnarface | if X isn't currently running, yes | 21:15 |
ejr | well tbh I might have done something stupid earlier on that might have caused some trouble: i ran chown -R user:user on /home/user | 21:15 |
bgstack15 | that shouldn't have caused an issue. | 21:16 |
gnarface | hmm, yea, but what lead to needing to do that... because that could have | 21:17 |
ejr | ok..i was worried, but not sure | 21:17 |
ejr | brb, restarting without $XAUTHORITY set | 21:17 |
ejr | now it does not linger for 5 mins after startx, but terminates x right away. | 21:20 |
ejr | http://paste.debian.net/1143809/ is the current Xorg log, but I dont think much has changed to earlier | 21:20 |
ejr | also, i can launch xserver with sudo startx, but ofc i dont want to do that with sudo privileges | 21:24 |
ejr | (but this means its most likely not a graphics issue i guess) | 21:24 |
gnarface | hmm, does suggest a permissions issue on some files or a device node | 21:26 |
ejr | permissions on .Xauthority are "0 -rw-------" | 21:26 |
gnarface | needs to be owned by your user too | 21:26 |
gnarface | i can't tell that from this^ | 21:27 |
ejr | so just chown user:user .Xauthority? | 21:27 |
gnarface | yea, but if it wasn't already set to that by default then something is already wrong | 21:27 |
ejr | ah well, it's "0 -rw------ 1 user user", sorry | 21:27 |
gnarface | should be fine then | 21:28 |
gnarface | did you change video cards recently too? | 21:28 |
gnarface | or drivers? | 21:28 |
gnarface | permissions requirements are different for nvidia than others | 21:28 |
gnarface | (even different than nouveau, and different between releases) | 21:28 |
ejr | hm, well, in some mins my other fresh install is finished, then I'll see whether it works on a similar laptop. if not, i will just use ceres | 21:28 |
ejr | no, i didnt change anything with respect to vidoe cards or drivers | 21:29 |
gnarface | which laptop is this? | 21:29 |
gnarface | make and model? | 21:29 |
gnarface | just for my notes | 21:29 |
ejr | X60 | 21:29 |
ejr | one librebooted (this one), the other one (fresh-installing now) isnt | 21:29 |
ejr | thinkpad x60 that is | 21:29 |
gnarface | that's a lenovo thinkpad, or a IBM one? | 21:29 |
ejr | technically lenovo, officially still ibm | 21:30 |
gnarface | so it is from before they sold the brand to them? | 21:30 |
gnarface | i don't remember the timeline on that exactly | 21:30 |
slvr | after sale, before brand change | 21:30 |
slvr | I have a similar machine | 21:31 |
slvr | need to crack the svp to use it though | 21:31 |
ejr | it was right around that time, 2006 | 21:31 |
ejr | i basically exclusively use my 3 X60 machines, lol | 21:31 |
ejr | or one is actually an X61 | 21:31 |
slvr | I need to find a t42 with a broken screen and bad keyboard I think | 21:32 |
gnarface | hmm, i forget if the intel driver in beowulf requires Xorg to be run through the suid root wrapper or not | 21:32 |
gnarface | at some point i think they changed it so it did not anymore, but it could still cause issues if you have a configuration halfway between the two states | 21:33 |
ejr | well, if it does, it only does so since today's update, cause it worked until today | 21:34 |
fsmithred | should work ok since elogind is installed | 21:34 |
fsmithred | as user, I mean | 21:34 |
ejr | i have just tried on the other x60 with a minimal beowulf install with only the Xserver packages and i3-wm, there i also cannot get into x unless i am root. | 21:34 |
MinceR | my X60s has both IBM and Lenovo logos on it | 21:35 |
ejr | if i install ceres, i will not get today's kernel upgrade, right? | 21:36 |
ejr | even if i run apt-get upgrade with ceres packaging? | 21:37 |
fsmithred | if you choose a network mirror, you should get whatever is in the repo today | 21:37 |
fsmithred | oh, not a new install. Upgrade should also give you whatver is current. | 21:38 |
tom | I am unable to mount network shares in beowulf | 21:38 |
Guest87495 | # mount -t nfs -v viridi:/ /mnt | 21:38 |
Guest87495 | mount: /mnt: bad option; for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount.<type> helper program. | 21:38 |
ejr | hmm... | 21:38 |
Guest87495 | despite never having to do it before, i specified type | 21:38 |
ejr | i will purge the 0-8 kernel from today and try with 0-6, brb | 21:39 |
Guest87495 | and it still did not work | 21:39 |
Guest87495 | nevermind | 21:41 |
Guest87495 | looks like beowulf requires nfs-common | 21:41 |
Guest87495 | i feel like that really should be part of the base system | 21:41 |
fsmithred | I thought it was part of the base system | 21:41 |
ejr | still no luck. | 21:45 |
Guest87495 | fsmithred, apparently not | 21:45 |
Guest87495 | at least not when using the beowulf install disc | 21:45 |
ejr | should I file a bug report for my issue? | 21:48 |
fsmithred | nfs-common in the desktop-live, which has the same packages as a default desktop install from installer isos (with network mirror) | 21:49 |
fsmithred | Guest87495, did you install with a mirror? | 21:49 |
Guest87495 | yes | 21:49 |
fsmithred | what did you install? | 21:50 |
gnarface | ejr: since it's recreateable with a fresh install, i think so yes. the fact it works with sudo suggests it is a mundane permissions or path issue that can be trivially fixed by hand if you can find it though. any chance adding yourself to the "video" group changes anything by the way? | 21:53 |
gnarface | ejr: sorry i don't have any further advice really. i haven't run into this one yet. | 21:54 |
ejr | that would be "adduser myusername video", right? | 21:56 |
slvr | usermod $(whoami) -aG video | 21:56 |
ejr | ^ gnarface | 21:56 |
fsmithred | ejr, elogind and libpam-elogind are installed? | 21:57 |
ejr | ok, done that. gonna reboot and try one last time :) | 21:57 |
ejr | fsmithred: yes | 21:57 |
ejr | still doesnt work. | 21:59 |
ejr | so if i want to keep using devuan, which version should i install to avoid this problem? ceres, and never running upgrade/dist-upgrade? | 22:00 |
gnarface | ejr: adding yourself to the video group didn't work either? at this point we don't even know what actually caused this. we need to figure that out first. | 22:01 |
slvr | ejr: are you even using a desktop? | 22:01 |
slvr | looks like x starts fine and then stops | 22:02 |
fsmithred | ejr, I'm about to test this in ceres. So far, it's working, but I need to upgrade and see if it breaks. | 22:02 |
ullet_ | x marks the stop | 22:02 |
slvr | something like the startxfce4 might work for you | 22:02 |
fsmithred | still working here after upgrade | 22:05 |
ejr | slvr: no, just i3-wm, and dwm | 22:06 |
slvr | ejr: did X leave you at a black screen or did it start and terminate? | 22:06 |
ejr | and actually this issue has brought me to the brink of going completely without Xserver, just using plain tty+tmux, lol | 22:06 |
slvr | X & sleep 2; DISPLAY=:0 xterm | 22:07 |
ejr | slvr: see above; originally it gave me a black screen for some mins, now it just terminates | 22:07 |
slvr | yeah you need an xprofile | 22:07 |
ejr | now after i installed dwm on the other machine on a fresh install, it works without any problems. | 22:07 |
slvr | imho reinstall your desktop env and display manager | 22:07 |
fsmithred | no dm | 22:07 |
slvr | nodm is pretty great | 22:07 |
fsmithred | ejr, what libpolkit packages are installed? | 22:08 |
fsmithred | libpolkit-{gobject|backend}-elogind-1-0 might be needed | 22:08 |
gnarface | for just running startx from a virtual terminal though? shouldn't be required, and i wonder if it could be causing issues...? | 22:09 |
fsmithred | right, it really shouldn't be needed | 22:09 |
gnarface | ejr: ^ i already asked you this but can you paste the exact literal package names for fsmithred here? | 22:10 |
ejr | fsmithred: theyre both installed | 22:10 |
ejr | ok | 22:10 |
fsmithred | don't past 5 lines at once | 22:11 |
fsmithred | paste | 22:11 |
Guest87495 | How is it that USB DAC/ADCs are used for production? | 22:12 |
ejr | http://paste.debian.net/1143827/ | 22:12 |
Guest87495 | Wouldn't the added jitter and latency of USB be a show stopper (literally) compared to a sound card on a PCIE bus? | 22:12 |
gnarface | Guest87495: that might be a better question for #alsa | 22:13 |
gnarface | Guest87495: though my assumption is that the answer is a combination of software and workflow mitigations | 22:14 |
Guest87495 | but buying a dedicated dac is all about latency | 22:14 |
slvr | usb and pcie are both packet based so who cares | 22:15 |
ejr | the more i think about it, the more i enjoy the idea of not using x anymore. do any of you guys do this? running everything on the tty (e.g. with tmux)? if i really needed a graphical browser i could set up another x60 with mate-desktop or something and use that once in a while if people require me to use stupid javascript websites | 22:15 |
fsmithred | people who are recording live music care | 22:15 |
gnarface | Guest87495: it isn't my area of expertise, but you will have to be very careful which USB device you buy. support for them is spotty on linux, so do your research first | 22:15 |
slvr | ejr: I do most of my work in a screen session | 22:15 |
fsmithred | ejr, I like X, but there are plenty of folks in the devuan community who are cli-only | 22:15 |
slvr | X is how I launch my xterms. :p | 22:16 |
fsmithred | me too, slvr | 22:16 |
fsmithred | lol | 22:16 |
Guest87495 | slvr, why screen instead of tmux? | 22:16 |
slvr | because I know screen and tmux users are too preachy | 22:16 |
slvr | congrats on finding salvation though | 22:17 |
ejr | slvr: there are two syntactic ways to make sense of your statement about preachyness :P | 22:17 |
ejr | (could be said about emacs and vim users as well) | 22:17 |
slvr | hmmm. not sure if vim users are more or less preachy than emacs users. | 22:18 |
fsmithred | who uses emac and vim? :P | 22:18 |
slvr | I don't actually like vim. prefer lighter vi clones. | 22:18 |
* onefang suggests the holy warriors head over to #debianfork. | 22:18 | |
slvr | i refuse to push my religion on others. :) I will help fix X though. | 22:19 |
ejr | fsmithred: s/and/when there is/ | 22:20 |
fsmithred | I've been trying to reproduce your problem, but qemu is giving me problems of my own | 22:20 |
slvr | I have a sneaking suspicion that X is starting and then failing to start the not-desktop-environment that the user is expecting. | 22:21 |
ejr | fsmithred: talking about qemu, i also had a weird issue with kvm, complaints that I the kvm group did not exist. Not sure if that error was there before. I dont use qemu/kvm but I just created the kvm group to get rid of the error... | 22:21 |
ejr | maybe THAT caused all my trouble. lol | 22:21 |
slvr | nah | 22:21 |
slvr | unrelated | 22:21 |
fsmithred | I think the general consensus is that you can ignore that warning | 22:21 |
ejr | btw are you guys devuan developers? | 22:21 |
fsmithred | I am - I make the live isos | 22:22 |
* gnarface is not | 22:22 | |
ejr | nice :) | 22:22 |
ejr | yes, fsmithred, i was told i can ignore the warning, but still such warnings trigger my autism so that i want to get rid of them | 22:22 |
fsmithred | gnarface is unofficial but essential member of the support team | 22:22 |
gnarface | thanks fsmithred :) | 22:23 |
ejr | just like lost of .files in my homefolder do, which makes me want to have as many of them as possible in .config or other suitable folders | 22:23 |
fsmithred | keep in mind that linux checks for everything when you boot, and it likes to tell you what it didn't find | 22:23 |
slvr | you could build without kvm support | 22:23 |
onefang | I'm a Devuan developer, mirror herder. | 22:23 |
slvr | but most of us would prefer to have it by default... | 22:23 |
* slvr runs kvm on devuan for some web services I prefer to isolate | 22:24 | |
fsmithred | I found a ceres VM with no X. I'm adding it now. | 22:30 |
bgstack15 | Is it so terrible a thing to say, that if stuff works the way you want, you can ignore any warnings you investigate to some degree with no conclusive results? | 22:30 |
fsmithred | bgstack15, I sure hope so - been doing that for 20 years | 22:30 |
bgstack15 | I've removed 7.5 mattress tags, with no adverse effects! (Minus becoming a GNU/Linux nerd) | 22:32 |
fsmithred | I added libpam-elogind and elogind, and X still is not starting. (installed without Recommends) | 22:32 |
gnarface | ejr: all those tests i walked you through, they were on a bare-metal devuan install, not one running in a guest like qemu, right? | 22:32 |
ejr | a while ago i wanted to get rid of the lingering "sda5_crypt ...busy" message on luks-encrypted installations. i remember someone here or in the forum pointed me to some config file that can be downloaded somewhere and just used to replace the old one on the system. does anyone remember which file that is? | 22:33 |
ejr | gnarface: yes, bare-metal with only basic system utilities installed (netinstall) | 22:34 |
aitor_ | hi | 22:36 |
fsmithred | hi aitor_ | 22:37 |
aitor_ | :) | 22:37 |
fsmithred | ejr, the other way to start X wtihout display manager is to install xserver-xorg-legacy and add a line to Xwrapper.config. That still works in ceres. | 22:37 |
fsmithred | aitor_, I tried packaging simple-netaid in beowulf, but it failed. Error message wasn't helpful. | 22:38 |
aitor_ | no worries, maybe my fault. I've just uploaded the packages of simple-netaid-cdk for beowulf: | 22:39 |
aitor_ | http://gnuinos.org/simple-netaid-cdk/?dir=beowulf | 22:39 |
fsmithred | it mentioned lines 15 and 18 in debian/rules | 22:39 |
fsmithred | cool | 22:39 |
aitor_ | i tested it and worked | 22:39 |
aitor_ | there are a lot of features pending to be added, but first things first | 22:40 |
aitor_ | the most important part is libnetaid | 22:40 |
fsmithred | what's that do? | 22:41 |
aitor_ | the shared library? | 22:41 |
fsmithred | yeah | 22:41 |
fsmithred | oh | 22:41 |
fsmithred | nm | 22:41 |
fsmithred | I thought you meant that libnetaid was part of the pending features | 22:42 |
onefang | I'm still wondering what this netaid thing is. | 22:42 |
fsmithred | I think I need more coffee | 22:42 |
fsmithred | it's a simple network manager | 22:42 |
fsmithred | that doesn't require dbus | 22:43 |
aitor_ | in part | 22:44 |
aitor_ | it's not difficult to cause failures in wicd | 22:44 |
aitor_ | on the other hand, simple-netaid will have a non Xorg interface (ncurses) and also a Gtk and Qt interfaces | 22:47 |
onefang | The only problem I have with wicd is that in keeps rewriting it's config file, even though nothing actually changed, it's just reordering bits. Annoys etckeeper. | 22:47 |
aitor_ | onefang: if you try to connect to a network device from the command line without flushing the ip addresses, the status icon of wicd will fail | 22:50 |
aitor_ | those attempts cause intermittences in the network connection | 22:51 |
onefang | I don't do that, nor use the status icon. I just use the GUI to change WiFi connections. | 22:51 |
aitor_ | my phone... | 22:52 |
onefang | Which I had to do a lot during Easter, when the home Internet connection was horrid. It's much more stable now we have changed ISP. | 22:52 |
onefang | With stable Internet, my only interactions with wicd are if I need to reboot, which makes it re-order a couple of lines in it's config file, which makes etckeeper complain next time I apt update or apt install. | 22:55 |
onefang | I usually only boot for new kernels or power failures. | 22:58 |
onefang | But I'm sufficiently annoyed at my /etc git logs filling with "Wicd is being wicked." that' I'd try out simple-netaid if it get into Devuan. | 23:00 |
openbsdtai123 | devuan has a serious problem with amd64 while copying on sd pendrives or disk, using fdisk and mkfs.... and more. amd64 has a big bug with FSs, especially ext and fat. | 23:11 |
golinux | https://bugs.devuan.org/ | 23:13 |
openbsdtai123 | no i have other duties than posting bugs :( I am not a developer or programmer. | 23:14 |
fsmithred | I see no problem | 23:14 |
slvr | sometimes dd just failes to right jack diddly for me | 23:14 |
slvr | conv=sync sometimes helps | 23:14 |
openbsdtai123 | I just told you guys... you will see. I am sure about it. | 23:15 |
slvr | seeing it claim to write 800Mb/s to usb is lol | 23:15 |
openbsdtai123 | I am running 4 partitions and various devuan. the issue comes from ascii amd64 <-. | 23:15 |
* golinux bites her tongue | 23:16 | |
slvr | *MB/s | 23:16 |
fsmithred | ok, I just formatted an sd card in ascii, made four partitions, mounted one and copied a file to it. All worked fine. What next? | 23:23 |
openbsdtai123 | done know it does that on devuan ascii kernel, two amd64 kernels and the classic from repo. | 23:25 |
gnarface | openbsdtai123: most likely just the writeback cache sabotaging you. try running "sync" before unplugging the drive from the usb port | 23:27 |
gnarface | openbsdtai123: (alternately you can run the "eject" command on the device) | 23:27 |
openbsdtai123 | I do alwys sync. mkfs.XXX fails to perform, it can sometimes hang. it does that with two kernels on several intel notebooks. | 23:28 |
gnarface | hmm, might be the USB key itself. this is a problem someone else here would have noticed. | 23:28 |
openbsdtai123 | I have 4 to 6 scandisk 16 gb | 23:28 |
openbsdtai123 | On hdd it does same (500gb wd). | 23:29 |
gnarface | maybe you just didn't wait long enough? i mean, usb is really slow. | 23:29 |
openbsdtai123 | if I boot x32 ascii, it just format. | 23:29 |
gnarface | flash storage is often even slower | 23:29 |
openbsdtai123 | it works on same hardware with 32bits so it should. | 23:29 |
openbsdtai123 | I waited 1 hour about. | 23:29 |
gnarface | for how much data? | 23:30 |
openbsdtai123 | mkfs.vfat /dev/sdb1 | 23:30 |
openbsdtai123 | 16 gb | 23:30 |
gnarface | try it with xfs instead | 23:31 |
gnarface | ext4 does a really slow format. i forget if vfat does too, but i know xfs doesn't. | 23:31 |
fsmithred | mkfs.ext2 and mkfs.vfat both work here | 23:31 |
openbsdtai123 | ok, so it is fine.... i just mentioned about it. no worries - maybe it will be fixed one day.- | 23:32 |
fsmithred | not if nobody knows what to fix | 23:32 |
gnarface | well the issue is that you've probably inflicted this on yourself | 23:32 |
gnarface | since none of us can recreate the issue | 23:32 |
golinux | <gnarface> well the issue is that you've probably inflicted this on yourself | 23:33 |
golinux | This ^^^ | 23:33 |
gnarface | and to be absolutely clear here, 1 hour isn't even close to enough time to write 16GB of data to a cheap flash drive | 23:35 |
gnarface | (that's why i asked for a xfs format for comparison) | 23:36 |
openbsdtai123 | ahhh a good idea would be to try with ubuntu kernel ... i will test it this weekend. | 23:36 |
golinux | Why is anyone still playing this fruitless conversation? | 23:37 |
gnarface | i just gave up right then | 23:37 |
gnarface | when i realized he won't execute any tests he knows might prove him wrong. he just wants to assert his false reality for the sake of wasting other people's time. | 23:37 |
openbsdtai123 | sure... I understand. | 23:38 |
golinux | gnarface: You just figured that out? | 23:38 |
gnarface | well, i give everyone two chances | 23:39 |
golinux | There have been more that that, I'm sure. :) | 23:40 |
gnarface | uh, maybe but i think i only remember this guy from a couple days ago | 23:40 |
* golinux needs a break | 23:40 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!