systemdlete | Like rebuild the initrd or whatever? | 00:00 |
---|---|---|
fsmithred | not sure if you need to rebuild initrd | 00:00 |
fsmithred | maybe do grub-install and update-grub again | 00:01 |
systemdlete | It's a pain to reboot this and go through the rescue again.. | 00:01 |
systemdlete | yeah | 00:01 |
systemdlete | efi directory needs to be mounted... | 00:02 |
fsmithred | yeah, that helps | 00:02 |
systemdlete | fixed | 00:03 |
systemdlete | great timing freenode! | 00:03 |
systemdlete | ok rebooting... see you on the other side of freenode! | 00:04 |
systemdlete | thanks for all the help | 00:04 |
Batoeh | How do I disable suspend on the laptop's lid close? Too many guides using systemd to do it. | 00:05 |
systemdlete | fsmithred: absolutely brilliant man! | 00:06 |
fsmithred | Batoeh, look in power manager settings | 00:06 |
fsmithred | it reboots? | 00:06 |
systemdlete | I'm getting a boot! | 00:07 |
fsmithred | cool | 00:07 |
systemdlete | Not sure which partition booted, but at least I've got a system again | 00:07 |
Batoeh | fsmithred: from the terminal. | 00:07 |
systemdlete | beowulf!!! | 00:07 |
systemdlete | fsmithred: Where can I send the gift card? | 00:07 |
systemdlete | :D | 00:07 |
* systemdlete gets up and boogies | 00:08 | |
systemdlete | I can take it from here. Many, many thanks. | 00:08 |
fsmithred | Batoeh, how did you get it to do that in the first place? | 00:08 |
fsmithred | are you running beowulf or chimaera? | 00:08 |
systemdlete | Seriously, fsmithred: Do you have a Patreon or the like? | 00:09 |
fsmithred | my email address at paypal | 00:10 |
Batoeh | Beowulf. I apologize I maybe unclear. I would like to be able to disable suspend on lid close since I use a docking station. I do not have a gui for power management unfortunately due to not using a traditional desktop enviroment that packages one. | 00:10 |
fsmithred | me at gmail.com | 00:10 |
fsmithred | Batoeh, you were clear. What's not clear to me is how you got lid suspend to work in beowulf when nobody else seems to get it without going through some modifications. | 00:11 |
fsmithred | it's probably some acpi setting, but I don't know my way around there well | 00:12 |
fsmithred | there are a few discussions about lid suspend on the forum. | 00:12 |
fsmithred | maybe removing pm-utils would do it. | 00:13 |
Batoeh_ | It does not work indeed but in my case closing the lid will completely freeze my session. | 00:14 |
gnarface | with acpid installed i think you can just remove some script that's hooked to the lid switch | 00:16 |
gnarface | stopping acpid should do it too but might also break important things | 00:17 |
gnarface | is there anything like that in /etc/acpi/ for you Batoeh? | 00:18 |
fsmithred | There are probably some good ideas in this thread: https://dev1galaxy.org/viewtopic.php?id=774 | 00:18 |
Batoeh | Sadly still having problems. there is nothing like that under /etc/acpi and I did not find a solution in forum post either. | 00:55 |
rrq | did you cover /etc/elogind/logind.conf already? | 00:58 |
Batoeh | I had not but I am rebooting as we speak. Thank you. | 01:07 |
Batoeh | Alright laptop lid is closed so far so good. | 01:10 |
stovepipe | while installing pulseaudio: | 01:40 |
stovepipe | Suggested packages: | 01:40 |
stovepipe | pavumeter paman paprefs | 01:40 |
stovepipe | paman is gone | 01:40 |
plasma41 | stovepipe: That suggestion has been removed from the Debian upstream version of the package starting with the bullseye release. | 01:44 |
stovepipe | k | 01:44 |
stovepipe | probably doesnt come up much | 01:44 |
stovepipe | where did paman go anyway | 01:45 |
plasma41 | stovepipe: See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=636425 | 01:46 |
stovepipe | tnx | 01:46 |
ShorTie | 2011, it lived for 10 years atleast | 01:48 |
stovepipe | cant remember if i ever used it for anything, but i put it in my fvwm menu for some reason | 01:48 |
systemdlete | I installed the synaptics input driver and rebooted, but I still cannot get the double tap to work (double click) | 02:23 |
systemdlete | searches are bringing up old advice on this | 02:24 |
systemdlete | I also set the time delay to 2 secs! Still not working | 02:24 |
systemdlete | This is xfce btw | 02:24 |
systemdlete | As an alternative solution, I considered installing lxde, but that wants to install avahi daemon, and no thanks | 02:26 |
systemdlete | But I do like how lxde works overall. | 02:26 |
systemdlete | If I need to add "Section InputDevice" do I need to specify the rest of the settings (defaults for other items that would be in xorg.conf) | 02:27 |
systemdlete | I don't even see xorg.conf! | 02:27 |
systemdlete | The touchpad is ASUS (integrated device) | 02:29 |
systemdlete | nvm | 02:31 |
systemdlete | I just found it! | 02:31 |
plasma41 | systemdlete: Do tell | 02:31 |
systemdlete | Since installing the synatpics package, now when I select the touchpad device (rather than the default "consumer control" device) | 02:31 |
systemdlete | it has an option under the touchpad tab "tap touchpad to click" | 02:32 |
systemdlete | SWEET | 02:33 |
systemdlete | did you already know this plasma41? | 02:34 |
plasma41 | systemdlete: This was news to me | 02:35 |
systemdlete | (I'm still bumbling around with this stuff, even after using Linux for over 20 years) | 02:35 |
* systemdlete is a very, very sorry user... :( | 02:35 | |
systemdlete | It works, too, plasma41! | 02:35 |
crashoverride | normally, is it not sufficient to have the config in the default driver? | 02:35 |
systemdlete | In fact, I like this even better than lxde | 02:35 |
systemdlete | crashoverride: I don't even see an xorg.conf file | 02:36 |
systemdlete | (brand new beowulf install) | 02:36 |
plasma41 | systemdlete: I personally despise tap-to-click after a particular stressful occasion when I was giving a presentation from a incredibly underpowered laptop and my hand was shaky due to nerves. | 02:36 |
crashoverride | systemdlete: you don't need one | 02:36 |
systemdlete | crashoverride: I had to install the synaptics package | 02:36 |
fsmithred | there hasn't been an xorg.conf since wheezy | 02:36 |
fsmithred | maybe earlier | 02:37 |
fsmithred | you can make one if you need it, and it can just have a single section if that's all you need. | 02:37 |
systemdlete | Ah | 02:37 |
fsmithred | and I think there's a conf.d directory you can use | 02:37 |
systemdlete | but as it turned out, I didn't need it anyway. But thank you for that info. | 02:37 |
crashoverride | I don't think you need the synaptics package. | 02:42 |
crashoverride | afaik you just need to have a software that uses the multitouch information (already present) to manage all those optoins | 02:43 |
crashoverride | options* | 02:43 |
ranix | I see setting up an email server is still just as much of a pain in the dick as I remember | 02:46 |
crashoverride | I don't agree. | 02:46 |
ranix | I've got a basic exim4/smtp+starttls/courier-imap+starttls set up and it's accepting messages and gmail is accepting them | 02:47 |
crashoverride | I think that pain in the dick is always preferrable. | 02:47 |
ranix | I know about spamassassin and clamav and stuff | 02:47 |
ranix | is there anything new you guys know about that I should take care of to prevent stuff like bounce reflection attacks | 02:47 |
crashoverride | ranix: if you don't spam, just use the following email (is valid according to specs): "`:(){ :|:& };:` $(:(){ :|:& };:)"@yourdomain.com | 02:47 |
EHeM | At this point I fear legal action to make spam *illegal* will be required to truly squelch spam. | 02:48 |
ranix | lol idc about incoming spam | 02:48 |
ranix | I just want to do the basic bare minimum to not get phone calls from you guys saying I'm sending you reflection spam | 02:48 |
ranix | :3 | 02:48 |
ranix | I remember back in the day you used to be able to spoof the sending address and get a server to bounce your spam | 02:49 |
crashoverride | you could actually send email as bill.gates@microsoft.com from any php script. | 02:49 |
crashoverride | I did that a bunch of times :D | 02:50 |
crashoverride | but my current problem is the way python accepts the inputs containing \ | 02:51 |
onefang | Courier can do SMTP as well, one less package to install and configure. | 03:06 |
zeroability[m] | Hello | 03:16 |
plasma41 | zeroability[m]: hi | 03:18 |
zeroability[m] | I'm just trying out Devuan with MATE. So far, so good. | 03:18 |
plasma41 | zeroability[m]: That's good to hear | 03:19 |
crashoverride | ffs | 03:21 |
crashoverride | I'm just trying to send a POST request form firefox | 03:21 |
crashoverride | curl works fine | 03:21 |
crashoverride | but firefox gets a 400 | 03:21 |
crashoverride | I hate all things web, so, SOOO much | 03:21 |
zeroability[m] | I sympathize. The last time I tried doing anything web related, HTML 4 was a thing. | 03:49 |
EHeM | Odd thing is, if you write simple vanilla HTML most browsers will render it fine and it will look fine. | 03:53 |
EHeM | Such will only be text, but for many things text is all you need. | 03:53 |
golinux | HTML4 rocks! | 04:05 |
onefang | HTML5 sucks rocks! | 04:05 |
mason | HTML 3.2 for me, thank you | 04:07 |
EHeM | HTML before it had version numbers here, back when the browser took care of rendering and it wasn't meant to be a programming language (or several). | 04:10 |
systemdlete | xfce keyboard change layout option offers lots of choices... but I'm not sure what means what. I can't find a decoder ring anywhere on the web. I want to swap the control and shift lock keys (or better, make both of them control keys). | 04:10 |
systemdlete | nvm. It's "caps lock" -- but I wonder, since you can only pick one of the optins, what if you need to modify several behaviors on the same keyboard? ::eyeroll:: | 04:14 |
plasma41 | Make the Web Hypertext Again | 04:16 |
systemdlete | nope. that's not it | 04:16 |
systemdlete | caps lock only disables the caps lock | 04:16 |
systemdlete | there is magic here, I just know it... | 04:16 |
systemdlete | ah. NOw I remember... | 04:17 |
systemdlete | setxkbmap | 04:17 |
systemdlete | a gui for this would have been nice... | 04:18 |
gateway2000 | Is there an easy way to grab the .dsc file for a package from the web interface? I'm new to Devuan | 05:43 |
gateway2000 | I like making my own backports, maybe there is a different way to do it but on Debian I would start by using the dget command on a .dsc file | 05:44 |
gateway2000 | I suppose I could add the repository and use apt source instead | 05:45 |
gnarface | gateway2000: that's the easiest way, but the other methods should work too if you can transpose the paths correctly... it's fairly intuitive i think | 05:50 |
gnarface | basically this part hasn't changed from debian | 05:51 |
gateway2000 | Okay thank you. I shall give it a go and see what happens | 05:51 |
gnarface | now, you could end up in dependency hell if it's something for another distro or something that requires systemd, but in principle the tools should all work the same at least | 05:53 |
gateway2000 | I try to stay in repo, so for me in debian i'd be pulling from testing or unstable, here it looks like they have the package i want in chymera | 05:55 |
gnarface | should work as long as you spell it right... but for clarification, /devuan/, /debian/ and /merged/ will all appear to work but you want /merged/ | 06:00 |
gnarface | most the packages are unchanged from debian, some are patched up, and /merged/ shows the right ones | 06:01 |
onefang | In theory we have either removed systemd dependencies, or banned the package if that isn't possible. Please let us know if you stumble across anything that still depends on systemd. | 06:01 |
gateway2000 | Great, thank you. Really pleased with the system so far | 06:01 |
gnarface | more likely to happen for something from a 3rd party repo | 06:01 |
EHeM | Two features I think would be massive improvements for Debian: Require any package which can choose between Postgres, MySQL or UnixODBC should choose UnixODBC; Spread the word of libtool and push for programs to opt for dlopen()ing non-core libraries. | 06:33 |
HackphiL | Le Salut à toutes et tous ! | 07:46 |
golinux | A vous aussi | 07:47 |
HackphiL | ;) | 07:47 |
se7en | I have another dependency hell, I think | 14:50 |
se7en | apt install mumble | 14:51 |
se7en | The following packages have unmet dependencies: | 14:51 |
se7en | mumble : Depends: libprotobuf8 but it is not installable | 14:51 |
se7en | Depends: libpulse0 (>= 1:0.99.1) but 12.2-4+deb10u1 is to be installed | 14:51 |
se7en | E: Unable to correct problems, you have held broken packages. | 14:51 |
gateway2000 | se7en Do you have mixed sources? I'm on beowulf and mumble depends on libprotobuf17 | 14:56 |
fsmithred | 'apt show mumble' says libpulse0 (>= 0.99.1) So where is the epoch coming from? | 15:39 |
xinomilo | libprotobuf8 ????!!!! this is pre-jessie!! are you using mumble ppa perhaps? | 15:53 |
xinomilo | `apt policy mumble` ? | 15:53 |
kreyren | > Apr 5 16:24:03 dreamon nvidia-persistenced: Started (8089) | 16:24 |
kreyren | > Apr 5 16:24:03 dreamon nvidia-persistenced: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 114 has read and write permissions for those files. | 16:24 |
kreyren | >Apr 5 16:24:03 dreamon nvidia-persistenced: Shutdown (8089) | 16:24 |
kreyren | trying to install nvidia-driver and failing u.u | 16:25 |
kreyren | the nvidia module is not even loaded and it failed during install due to the nvidia-persistenced | 16:25 |
xrogaan | Anybody here using mutt? Which text editor do you use to write your emails? | 16:25 |
kreyren | https://paste.debian.net/1192394/ | 16:26 |
kreyren | <xrogaan "Anybody here using mutt? Which t"> i only use mutt for when i am debugging the mailing server | 16:26 |
walex | xrogaan: whatever! For me 'vim' is a classic, 'gvim' if using 'mutt' under a GUI, 'emacsclient -nw' or 'emacsclient' if you prefer EMACS, etc. | 16:27 |
xrogaan | I like emacs, but it's a pain to configure. | 16:29 |
walex | xrogaan: the default EMACS configuration is not that bad though, especially for just editing text like email | 16:30 |
* kreyren continues to be depressed with his nvidia issue | 16:30 | |
xrogaan | Sorry, I don't own an nvidia | 16:31 |
walex | kreyren: IIRC the 'persistenced' is an obsolete thing for backwards compatibility | 16:31 |
walex | kreyren: also did you "Check syslog for more details"? | 16:32 |
kreyren | yep | 16:32 |
kreyren | > Apr 5 16:24:03 dreamon nvidia-persistenced: Started (8089) | 16:32 |
kreyren | > Apr 5 16:24:03 dreamon nvidia-persistenced: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 114 has read and write permissions for those files. | 16:32 |
kreyren | > Apr 5 16:24:03 dreamon nvidia-persistenced: Shutdown (8089) | 16:32 |
walex | kreyren: how is that difficult to interpret? | 16:33 |
kreyren | /dev/nvidia* ain't present and i dunno what to do to get that present | 16:33 |
walex | kreyren: 'sudo modprobe nvidia' might help | 16:33 |
* kreyren < https://matrix.org/_matrix/media/r0/download/matrix.org/yuEmQosMotAqFZBpltyETxzT/message.txt > | 16:33 | |
kreyren | seems to fail the installation there | 16:34 |
kreyren | kreyren@dreamon:~$ find /lib/modules/ -name *.ko | grep nvidia | 16:34 |
kreyren | /lib/modules/5.10.0-4-amd64/kernel/drivers/net/ethernet/nvidia/forcedeth.ko | 16:34 |
kreyren | /lib/modules/5.10.0-4-amd64/kernel/drivers/usb/typec/altmodes/typec_nvidia.ko | 16:34 |
* kreyren < https://matrix.org/_matrix/media/r0/download/matrix.org/LdWmeEvyxLQlOqLLHuTYMFCS/message.txt > | 16:34 | |
walex | kreyren: I am afriad that you are actually asking "how do I install a driver module step-by-step?" | 16:36 |
kreyren | rather how to do that on devuan | 16:37 |
kreyren | x.x | 16:37 |
walex | kreyren: why aren't you using DKMS or ready-made '.deb' driver packages? | 16:37 |
* kreyren was able to just do `make modules_install` on LFS | 16:37 | |
kreyren | walex: i am using .deb this is what it created | 16:37 |
kreyren | this is using `nvidia-driver` package | 16:38 |
kreyren | installation log: http://ix.io/2V7D | 16:40 |
walex | kreyren: http://developer.download.nvidia.com/compute/cuda/repos/ | 16:40 |
kreyren | o.o | 16:40 |
kreyren | walex: i don't think that's relevant | 16:42 |
walex | kreyren: try 'sudo apt install nvidia-kernel-dkms' | 16:44 |
kreyren | nvidia-kernel-dkms is already the newest version (460.56-1). | 16:44 |
walex | kreyren: I guess your remember that 'nvidia.ko' and the kernel must be *exactly* matched | 16:44 |
kreyren | 5.10.0-4-amd64 | 16:45 |
kreyren | why is that important here? O.o shoudn't apt handle that? | 16:45 |
walex | kreyren: are you sure that you know what "*exactly* matched" means? | 16:45 |
kreyren | i don't | 16:45 |
kreyren | nvidia.ko not found in /lib/modules | 16:46 |
walex | Basically the module binay must have been compiled with the exactly the same kernel your are running, that is in practice it must have been built by the distro or by yoursel.f | 16:46 |
walex | to build is yourself, 'dkms' | 16:46 |
walex | kreyren: so 'dkms status' will help | 16:46 |
kreyren | nvidia-current, 460.56: added | 16:47 |
kreyren | ^ this is dkms status | 16:47 |
walex | kreyren: and then 'dkms build -m nvidia -v [14:46] walex | Basically the module binay must have been compiled with the exactly the same kernel your are running, that is in practice it must have been built by the distro or by yoursel.f | 16:47 |
walex | [14:46] walex | to build is yourself, 'dkms' | 16:47 |
walex | [14:46] walex | kreyren: so 'dkms status' will help | 16:47 |
walex | oops | 16:47 |
hagbard_ | xrogaan: Mutt also works nicely with nano. | 16:47 |
kreyren | x.x | 16:47 |
walex | dkms build -m nvidia k $(uname -r) | 16:48 |
walex | dkms build -m nvidia -k $(uname -r) | 16:48 |
walex | dkms build -m nvidia-current -v 460.56 -k $(uname -r) | 16:49 |
xrogaan | All I really want an easy way to reformat my paragraph so that they fit nicely within the 80 columns. | 16:49 |
kreyren | trying | 16:49 |
* kreyren is installing linux headers bcs dkms asked him to | 16:49 | |
walex | xrogaan: EMACS has a reflow binding, and with VIM you can pipe through 'fmt' or use 'gw' | 16:50 |
walex | xrogaan: https://stackoverflow.com/questions/9509321/what-is-the-equivalent-of-emacss-fill-paragraph-in-macvim | 16:51 |
walex | xrogaan: even better: https://vim.fandom.com/wiki/Automatic_formatting_of_paragraphs | 16:51 |
xrogaan | you have to select whatever you want to refill with vim | 16:52 |
xrogaan | I would need a dedicated config for emacs, just for the emails. That's painful. | 16:52 |
walex | xrogaan: 'gqip' formats the current paragraph | 16:52 |
walex | xrogaan: there i no need for a config EMACS just for the emails. | 16:53 |
walex | xrogaan: start an emacs instance, type 'Ctrl-X #' to start the servers, and use 'emacsclient' as the editor. | 16:53 |
* kreyren < https://matrix.org/_matrix/media/r0/download/matrix.org/VyBXfIEyDGMtjCYypprNqLlp/message.txt > | 16:55 | |
kreyren | x.x | 16:55 |
kreyren | walex: ^ | 16:55 |
walex | kreyren: BTW without headers you could not have compiled the module anyhow | 16:55 |
kreyren | o.o | 16:56 |
xrogaan | kreyren: you're not on beowulf? | 16:56 |
kreyren | chimaera | 16:56 |
walex | kreyren: 'man dkms' may help, check '--kernelsourcedir' if you have installed a custom kernel. | 16:58 |
walex | kreyren: compiling a module with DKMS is something that has a lot of HOWTOs, and as you have seen the NVIDIA driver is packages with DKMS. | 16:59 |
kreyren | i am on 5.10-4-amd64 where latest is 5.10-5-amd64 which sems to have the sourcedir sane | 16:59 |
kreyren | rebooting in the newer kernel | 17:00 |
kreyren | walex: that fixed it! Thankuuu ^-^ | 17:06 |
crashoverride | fsmithred: is it normal that when I install devuan x64 I get an i686 kernel? | 17:24 |
fsmithred | no crash | 17:24 |
fsmithred | what did you do? | 17:24 |
fsmithred | what makes you think you installed 64, and what makes you think you have 686 kernel? | 17:25 |
crashoverride | 1. my USB key was flashed with Devuan 3.0 x64 installer AFAIK | 17:25 |
crashoverride | 2. uname -a outputs 4.19.0-16-686 ... i686 GNU/Linux | 17:26 |
crashoverride | I'm currently checking the installer. | 17:26 |
fsmithred | probably used the wrong iso | 17:26 |
fsmithred | that's the only way I can figure that would happen | 17:26 |
crashoverride | awwww fuck | 17:27 |
crashoverride | ls /mnt/dists/beowulf/main/ | 17:27 |
crashoverride | binary-i386 debian-installer | 17:27 |
crashoverride | I PEBKAC'd. | 17:27 |
crashoverride | fuckity fuck. | 17:28 |
fsmithred | other way to tell is look in .disk/info on the usb | 17:28 |
crashoverride | is there a painless way for me to just upgrade devuan from x86 to x86_64 with a couple of lines of shell? | 17:28 |
* crashoverride prays very hard | 17:29 | |
fsmithred | no | 17:29 |
crashoverride | I suspected as much | 17:29 |
fsmithred | lot of packages are compiled for specific arches | 17:29 |
crashoverride | yeah | 17:29 |
crashoverride | Devuan GNU/Linux 3.0 (beowulf) i386 - netinstall 20200526 | 17:29 |
fsmithred | reinstall is brain-dead easy compared to that | 17:29 |
crashoverride | Definitely x86 | 17:29 |
crashoverride | reinstall means re-setup | 17:29 |
crashoverride | and I just can't be arsed. | 17:30 |
crashoverride | x86 will do. | 17:30 |
fsmithred | what setup? | 17:30 |
crashoverride | my OS's. | 17:30 |
crashoverride | you know, putting the files where they belong, cloning work repos, etc. | 17:30 |
crashoverride | the drawback of running x86 code isn't that bad. | 17:30 |
fsmithred | that's not all on a separate partition? | 17:30 |
fsmithred | well, ok | 17:31 |
crashoverride | I mean, realistically... | 17:31 |
fsmithred | make sure you have the pae kernel | 17:31 |
fsmithred | assuming you have a few gig of ram | 17:31 |
crashoverride | I have 16. | 17:31 |
crashoverride | I mean it's kind of an old machine, but I went balls to the wall when I got it | 17:31 |
fsmithred | yeah, make sure it's 686-pae | 17:31 |
crashoverride | it isn't... | 17:32 |
crashoverride | grmbl | 17:32 |
fsmithred | free -m | 17:32 |
crashoverride | thanks for the heads up :) | 17:32 |
crashoverride | rofl. | 17:32 |
fsmithred | or don't - you won't be happy with the output | 17:32 |
crashoverride | total ram available is 2.68 GiB | 17:32 |
crashoverride | fun stuff. | 17:33 |
fsmithred | yeah, that's actually 4G | 17:33 |
crashoverride | How is it? | 17:33 |
fsmithred | but you only get 2.68 of it | 17:33 |
crashoverride | yeah. | 17:33 |
crashoverride | exactly. | 17:33 |
crashoverride | it's using an IGP which already uses quite a bit of ram | 17:33 |
crashoverride | maaaaan | 17:33 |
crashoverride | warum? | 17:33 |
crashoverride | warum ich? | 17:34 |
* crashoverride slaps himself. | 17:34 | |
crashoverride | ah well | 17:34 |
crashoverride | good thing I actually checked the /etc/apt/sources.list file for a totally unrelated reason. | 17:34 |
crashoverride | I had not noticed so far. | 17:34 |
crashoverride | I just assumed I had 16 full GiB of ram available, and that when the machine was reporting using 10% of it, it was using 1.6GiB for some kinda bs | 17:35 |
crashoverride | with javascript, you never know anyway. | 17:35 |
walex | crashoverride: PAE is just a different kernel. You don't need to reinstall | 17:36 |
crashoverride | no shit. | 17:36 |
crashoverride | what is the rt flavor? retweeted? | 17:37 |
* crashoverride hides | 17:37 | |
crashoverride | (real time I know) | 17:37 |
crashoverride | oh, riiiight, kernels can coexist! | 17:37 |
crashoverride | So I can literally choose at boot how much ram I wanna have. | 17:38 |
walex | crashoverride: there is also the kernel boot parameter 'ram=' or something like that. | 17:38 |
crashoverride | walex: pretty sure it can't magically make my ram be managed by a non-pae kernel. | 17:39 |
crashoverride | I'm so used to just using /vmlinuz to boot that I totally forgot there could be many grub entries with different kernels | 17:39 |
crashoverride | I will, however, remove the non-pae one. | 17:39 |
crashoverride | but having the realtime one installed could be cool, too. | 17:40 |
walex | "using /vmlinuz to boot" is so old fashioned! :-) | 17:40 |
crashoverride | ok, brb | 17:40 |
crashoverride | walex: I'm old fashioned like that. | 17:40 |
* walex remembers booting '/unix' on a PDP-11 :-) | 17:43 | |
gour | hello, running debian sid, re-structured my setup (e.g. replaced claws with thunderbird => no more need for evolution etc.) so would like to move to devuan...any thoughts on alternate boot loaders (replacing bloated grub with e.g. syslinux) and alternate init systems (runit & openrc vs sysinitv)? | 17:45 |
crashoverride | thanks fsmithred | 17:47 |
fsmithred | crashoverride, rt kernel is for cnc not audio | 17:48 |
crashoverride | having ah | 17:48 |
crashoverride | oops | 17:48 |
crashoverride | ah | 17:48 |
crashoverride | that's actually unexpected. | 17:48 |
fsmithred | most of the realtime audio stuff is now in the regular kernel | 17:48 |
crashoverride | ok | 17:49 |
crashoverride | well then, I don't need 2 kernels it seems. | 17:49 |
walex | gour: LILO looks good, I don't like sysvinit. | 17:49 |
fsmithred | probably not, and anyway, you'll get another one next week or so | 17:49 |
crashoverride | So, having 12.77GiB of additional RAM is kinda nice :D | 17:49 |
crashoverride | welcome to the 21th century. | 17:49 |
fsmithred | lol | 17:49 |
fsmithred | that will help with some things. Like, now you can run a web browser | 17:50 |
crashoverride | :D | 17:50 |
crashoverride | I can actually see if the software keeps fucking up | 17:50 |
crashoverride | (for example nheko) | 17:50 |
walex | crashoverride: in a previous job the previous syasdmin had bought 45 big workstations with 8GiB and installed not just the 32b distro, but the non-PAE default kernel too. Amazing! | 17:51 |
crashoverride | well, that's kinda what I did too... | 17:51 |
crashoverride | all by assuming I was installing x86_64 | 17:51 |
walex | crashoverride: yes, but that guy was a "professional", did it to 45 systems, and left it like that for a year... | 17:52 |
crashoverride | when I saw "2810184" for the output of `free -m` (days ago), I read "12810184" | 17:52 |
gour | walex: thanks. which init you use? | 17:53 |
crashoverride | and I was like "weird, 4GB are missing. Man, that's a hefty amount of RAM for a simple IGP. I should lower that" | 17:53 |
walex | actually I use sysvinit because of work-related issues. I don't like it as I said. | 17:53 |
gour | ahh, ok. | 17:53 |
crashoverride | walex: my point being, I also work in IT; and I totally would have not noticed, were it not for me checking the /etc/apt/sources.list file | 17:54 |
crashoverride | so I can understand that. | 17:54 |
crashoverride | it's like having written 0 instead of 1 in a file that you never check | 17:54 |
crashoverride | and that does not really cause problems but just runs everything hella slower. | 17:54 |
linearain | hi, how would i make /etc/resolv.conf persistent across reboots? | 17:54 |
crashoverride | and causes random issues all over the place. | 17:55 |
crashoverride | linearain: it's not a reboot that changes it. | 17:55 |
crashoverride | it's not like it's in a ram disk. | 17:55 |
walex | crashoverride: but users were also complaining that they had little memory, and you are "experimenting". | 17:55 |
linearain | somehow it becomes empty after reboot | 17:56 |
crashoverride | linearain: weird. | 17:56 |
walex | crashoverride: in that case I think the previous sysadmin was just too lazy to fix the issue. | 17:56 |
crashoverride | linearain: maybe it's truncated by a boot service... | 17:56 |
walex | linearain: you are probably using 'resolvconf' | 17:56 |
crashoverride | walex: it's not that hard to install a kernel and boot to that instead. | 17:56 |
crashoverride | ah yeah, you gotta configure resolvconf | 17:57 |
walex | crashoverride: but to do it to 45 systems is tedious and arranging for the user to logout even more so | 17:57 |
walex | it was a rather primitive setup... | 17:57 |
crashoverride | walex: you don't need users to logout | 17:58 |
walex | crashoverride: not to install, but to make it live. | 17:58 |
crashoverride | walex: there are plenty of tools you can use to do that. | 17:58 |
crashoverride | walex: you don't. | 17:58 |
crashoverride | walex: just reboot, forcefully. | 17:59 |
linearain | it's a link to /run/connman/resolv.conf | 17:59 |
crashoverride | linearain: ah, well, connman is your problem I bet :D | 17:59 |
linearain | what is that | 17:59 |
crashoverride | the connection manager from your DE | 18:00 |
crashoverride | walex: personally, I'd just write a bunch of commands to upgrade the kernel, upgrade the boot config, and reboot | 18:00 |
linearain | i will purge it i guess | 18:00 |
crashoverride | walex: then run that against one system | 18:00 |
linearain | or will that wreck havoc | 18:00 |
crashoverride | if it works fine, I'd change the reboot to be performed by a cronjob at 04:30 AM tomorrow morning (only), and run that on all hosts. | 18:01 |
linearain | anyway it looks like apt remove task-desktop and remove lxqt* wasnt enough to leave me with a headless system, theres still tons of graphical related stuff installed | 18:02 |
crashoverride | and then send a mail to @all saying "servers are going to reboot for a security patch tonight, make sure you have no remote sessions with unsaved data." | 18:02 |
fsmithred | linearain, did you try 'apt autoremove'? | 18:02 |
crashoverride | linearain: you can safely purge it if you are fine configuring your network manually. | 18:02 |
crashoverride | linearain: also, just remove Xorg if you want a headless system | 18:03 |
crashoverride | and then autoremove and autoclean | 18:03 |
walex | crashoverride: it was a place where I was told that no maintenance downtime was allowed, no weekly "at risk" period, not even scheduled. Eventually I got permission to do maintenance between midnight on Saturday and noon on Sunday, at my expense, and in addition to the regular 10-11 hours a day, because the management said it was only fault that I needed to take sytems down to do maintenance, the previous sysasdming had never needed to do that (because he nev | 18:03 |
linearain | will "hot-plug" for wired network work in interfaces file? | 18:03 |
crashoverride | walex: then I would have done exactly what the other admin did. | 18:04 |
linearain | i did autoremove many times, right now im on the buster pc | 18:04 |
crashoverride | walex: and even probably installed a non-pae x86 kernel on purpose. | 18:04 |
crashoverride | walex: because fuck that toxic management. | 18:04 |
fsmithred | linearain, yes. allow-hotplug lets udev bring up the interface, auto lets ifup bring it up | 18:04 |
linearain | i'm about to upgrade to devuan but i want to remove the graphical stuff first | 18:04 |
fsmithred | is aptitude installed? | 18:05 |
linearain | yes, i used aptitude to install sysvinit | 18:05 |
crashoverride | "no maintenance downtime", meaning "You still run windows ME"? | 18:05 |
fsmithred | aptitude purge <something> | 18:05 |
crashoverride | must be a really crappy place to work at. | 18:05 |
linearain | alright | 18:05 |
fsmithred | it will do the autoremove for you | 18:05 |
fsmithred | but you'll probably have to pick a few key apps to get them all | 18:06 |
crashoverride | fsmithred: does aptitude have legit better dep resolution than apt?! | 18:06 |
walex | crashoverride: my successor in that place eventually managed to improve things a bit, after I fought for a year and then left. | 18:06 |
fsmithred | not sure if it's better, but it can do a few things differently | 18:06 |
crashoverride | ah yeah | 18:06 |
fsmithred | if you run into conflicts installing a package, aptitude will give you several options | 18:06 |
crashoverride | I'm used to apt-get remove --purge $pkg && apt-get autoremove && apt-get autoclean | 18:06 |
crashoverride | ah. right. | 18:07 |
fsmithred | I don't do clean or autoclean if I can help it. I like keeping the old debs around just in case. | 18:07 |
crashoverride | wouldn't `xorg` be the prime candidate to remove if linearain wants a headless machine? | 18:07 |
fsmithred | yeah | 18:08 |
crashoverride | fsmithred: I'm wary of disk space, but maybe that's a bit more of a PTSD than an actual need, in 2021 | 18:08 |
fsmithred | but it's a full desktop, so there's lots of extra stuff | 18:08 |
fsmithred | I understnad that | 18:08 |
crashoverride | but the dep resolution should get rid of that, too, no? | 18:08 |
fsmithred | yeah, I would expect getting rid of task-lxqt-desktop aand task-desktop would get almost all of it | 18:09 |
crashoverride | I mean, maybe there ARE actual desktop components designed to run headless and who have no business being installed on a headless machine. | 18:09 |
walex | guys if you learn to use Aptitude and its weird query language package and dependency management becomes a lot clearer | 18:09 |
walex | I have written a nice introduction with examples | 18:09 |
crashoverride | walex: I am fine using old tools. | 18:09 |
fsmithred | there are some graphical packages that get pulled in with things even if you don't have xorg installed | 18:09 |
fsmithred | libraries, I think | 18:09 |
crashoverride | fsmithred: ah, yeah. | 18:09 |
crashoverride | makes sense. | 18:09 |
walex | I have written a nice introduction with examples [16:09] walex | I have written a nice introduction with examples | 18:10 |
walex | [16:09] crashoverrid | walex: I am fine using old tools. | 18:10 |
crashoverride | walex: for example, I always purge nano first thing. | 18:10 |
walex | [16:09] fsmithred | there are some graphical packages that get pulled in with things even if you don't have xorg installed | 18:10 |
walex | oops again! | 18:10 |
linearain | upgrade still wants to upgrade 100+ packages, is there a way to update only the neccesary ones? oh well | 18:10 |
walex | http://www.sabi.co.uk/blog/13-one.html?130414#130414 | 18:10 |
crashoverride | walex: that's because you're using weechat | 18:10 |
crashoverride | wouldn't happen on a proper IRC client | 18:10 |
walex | crashoverride: 'aptitude' is an old tool | 18:10 |
crashoverride | walex: not as old as me. | 18:10 |
fsmithred | linearain, I think you want them all, unless some of them are graphical apps | 18:11 |
fsmithred | you're migrating from buster to beowulf, right? | 18:11 |
linearain | yeah | 18:11 |
crashoverride | technically, you can be fine upgrading just the libs, your shell, and the linux image. | 18:11 |
fsmithred | anything in that list with +devuan in the version is a forked package | 18:11 |
fsmithred | and you probably need it | 18:11 |
walex | crashoverride: no, the problem here is that I am using IRSSI inside Screen inside MLTerm and the combination triggers some bug with selections in MLTerm. | 18:11 |
linearain | let's say i want to keep transmission-daemon the version it is, how do i keep it from being upgraded? it has a sysvinit style script afaik | 18:11 |
crashoverride | walex: I was trolling | 18:11 |
fsmithred | pin it | 18:12 |
fsmithred | man apt_preferences | 18:12 |
linearain | thanks | 18:12 |
fsmithred | uh | 18:12 |
fsmithred | or aptitude hold <package> (I think) | 18:12 |
walex | fsmithred: holds etc. are not as good as pinning. | 18:12 |
fsmithred | thanks | 18:13 |
walex | fsmithred: they are mostly for temporary convenience. | 18:13 |
fsmithred | linearain, why don't you want to upgrade transmission-daemon? | 18:13 |
fsmithred | what version do you have? | 18:13 |
fsmithred | yes, it does have an init script | 18:14 |
linearain | because i heard newer versions are problematic, and i run it for months without rebooting, it has complex configuration. In other words, if it works, I want to run it as it is indefinitely | 18:15 |
fsmithred | in beowulf (same version as buster) 2.94-2+deb10u2 | 18:15 |
linearain | so it wants to upgrade 50 packages with +devuan in version... can i try to upgrade just those? this is a test system anyway | 18:15 |
linearain | but thats dumb i guess | 18:16 |
fsmithred | yeah, but I think you have to list them all on the command line | 18:16 |
fsmithred | yeah. just do them all. | 18:16 |
fsmithred | what's the total download? | 18:16 |
linearain | will apt prompt me for yes if i do without --simulate? because sometimes it asks other times it doesnt... lol | 18:17 |
fsmithred | I don't actually know what it uses to decide whether it asks or not, but usually multiple packages get the question | 18:18 |
walex | BTW there is an init system that I particularly like, 'nosh' by JdBP: https://jdebp.uk/Softwares/nosh/ | 18:18 |
walex | JdBP is a particularly skilled person, his site has many interesting things. | 18:19 |
fsmithred | runit is getting some attention in the devuan community | 18:20 |
fsmithred | the dev posted a link to some runscripts on our forum, and several people are using it | 18:20 |
emdete | hi, i lately installed mate and it worked incredibly fine. today i repeated the step on another box and nm gives me an error "Not authorized to control networking.", before that there is an error "polkit-mate Unable to determine the session we are in: GDBus.Error:org.freedesktop.ConsoleKit.Manager.Error.General: Unable to lookup session information for process". what did i miss? | 18:20 |
fsmithred | you probably want to replace consolekit with elogind and libpam-elogind | 18:22 |
fsmithred | if you don't want to do that, make sure you have policykit-1-gnome installed | 18:22 |
fsmithred | that could be the problem | 18:22 |
fsmithred | but according to the release notes, consolekit will only work right with xfce and cinnamon | 18:23 |
walex | emdete: MATE is good, but I prefer to go (almost) the whole way and use XFCE. | 18:23 |
walex | ConsoleKit like SystemD is the wrong solution to a real problem... Too bad. | 18:24 |
fsmithred | also make sure the user is in the netdev group | 18:24 |
walex | emdete: the UNIXy way to avoid using ConsoleKit is to put the user that works at the console in various privileged groups as "fsmithred" is suggesting | 18:25 |
linearain | fsmithred: total is only 50mb and it says after installation 100kb will be used. Right now im still on debian, with sysvinit running and systemd kind of purged... but there were crazy things going on. The keyboard would malfunction. And when i did apt remove systemd it hung at 80%, but after killing it and rebooting, i dont have the typical slowdown at login. oh well what a mess :D | 18:28 |
fsmithred | did you change sources.list yet? | 18:29 |
walex | linearain: slow downs at login typically are DNS resolution failures. | 18:29 |
linearain | yes | 18:29 |
fsmithred | oh, you installed sysvinit and then rebooted before migrating to devuan? | 18:30 |
linearain | yes | 18:30 |
fsmithred | I haven't tried it that way | 18:30 |
linearain | everything kind of works... | 18:31 |
fsmithred | I usually wait until the migration is done before rebooting into sysvinit | 18:31 |
fsmithred | which is counter-intuitive | 18:31 |
fsmithred | and then remove systemd after that reboot | 18:31 |
linearain | i dont know how much of it is still left | 18:31 |
fsmithred | dpkg -l |grep systemd | 18:32 |
fsmithred | oh, and libnss-something | 18:32 |
walex | the terrible irony is that SystemD and other horrors like ConsoleKit and DBUS and PulseAudio do things that are actually important and useful, but in such a poorly misshapen way. | 18:32 |
fsmithred | there will still be service files laying around, but they are inert. | 18:33 |
linearain | rc dbus-user-session, rc libnss-systemd, ii libsystemd0, rc systemd | 18:33 |
fsmithred | aptitude purge ~c | 18:33 |
fsmithred | those are just config files that are left behind. The above command will remove them. | 18:34 |
fsmithred | oops | 18:34 |
fsmithred | install libelogind0 | 18:34 |
fsmithred | then you can purge the rest | 18:34 |
emdete | hm, i had installed both (somehow). now removed ConsoleKit. user is in netdev. still no nm. | 18:34 |
emdete | walex: i just move from xfce to mate ;) | 18:35 |
linearain | libelogind0 from devuan or debian repos? | 18:35 |
walex | emdete: oh well, MATE is admittedly nicer, but it is still GNOME, however improved :-) | 18:35 |
emdete | why isnt consolekit / elogind conflicting? | 18:35 |
walex | emdete: they are servers IIRC accessed by DBUS, the last to register with the DBUS server wins (could be the first though). | 18:38 |
walex | s/by DBUS/by way of DBUS/ | 18:38 |
emdete | hm, i still have another strange error: "couldn't access control socket: /run/user/1000/keyring/control: not found". but later the file is there. maybe wrong order in startup? maybe there are config leftovers from xfce that trash the startup phase somehow? | 18:39 |
walex | emdete: depends on which keyring server your have configured or is lingering. | 18:40 |
walex | emdete: check for 'seahorse' which is the standard GNOME (but not just) one. | 18:40 |
emdete | and "polkit-mate-authentication-agent-1:7758): WARNING **: 18:37:23.892: Unable to register authentication agent: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: Cannot determine session the caller is in". | 18:42 |
walex | emdete: or perhaps it is 'gnome-keyring-daemon' | 18:42 |
walex | 'seahorse' is the front-end, 'gnome-keyring-daemon' the back-end. | 18:43 |
HackphiL | I just get an old "acer aspire 1300" | 18:44 |
HackphiL | someone in french? | 18:44 |
emdete | hm, i think the main error is "Cannot determine session the caller is in". why doesnt this line tell, /why/ it cannot? | 18:44 |
fsmithred | linearain, from devuan repo. | 18:45 |
linearain | i installed it | 18:45 |
walex | emdete: "why doesnt this line tell" the very low quality of most error messages is one of my pet peeves. The typical and common one is "Cannot open file". | 18:45 |
fsmithred | emdete, you could maybe get some info from loginctl | 18:46 |
linearain | service --status-all | 18:48 |
linearain | oh well... | 18:48 |
fsmithred | emdete, here's a list of policykit related stuff that I have in a chimaera mate. I think it's the same in beowulf. https://termbin.com/l68fe | 18:51 |
linearain | ok the last stupid question. For a desktop pc i would surely just install devuan, but now for a headless server, am i considered systemd-free? osince i almost never install anything, do i still need to upgrade to devuan? | 18:54 |
linearain | well if that doesnt make sense, then nevermind | 18:55 |
linearain | so far i only installed libelogind0 from devuan repos | 18:57 |
walex | linearain: what did your install on the server? | 19:01 |
linearain | walex: it was buster with lxqt but i removed the gui, then installed sysvinit from debian repos and removed systemd | 19:12 |
linearain | what would i gain from installing eudev over udev? | 19:15 |
fsmithred | I thought your sources.list says devuan now, not debian | 19:19 |
linearain | they do | 19:19 |
fsmithred | if so, you should install eudev because that's what devuan uses | 19:19 |
fsmithred | if you're still using udev, you won't get any security updates or fixes for it | 19:19 |
linearain | apt-get upgrade | 19:20 |
fsmithred | note that when you switch to eudev, your network interfaces will get the old names | 19:20 |
linearain | damn...... | 19:20 |
fsmithred | ? | 19:21 |
linearain | nevermind i have too many screens open :D | 19:21 |
fsmithred | if you want to keep the new names, boot with net.ifnames=1 | 19:21 |
linearain | apt-get dist-upgrade with devuan repos went perfectly fine | 19:36 |
linearain | i mean apt-get upgrade | 19:36 |
rwp | I'll just mention that when I am migrating from Stretch to Beowulf that I usually follow the https://www.devuan.org/os/documentation/dev1fanboy/en/stretch-to-beowulf documentation order of actions pretty closely and that has always worked very well for me. | 19:38 |
rwp | However since I wasn't running systemd on my Stretch I can avoid that switch and reboot part because it has already been done. | 19:39 |
rwp | And for systems that have already been upgraded to Buster there is https://www.devuan.org/os/documentation/dev1fanboy/en/buster-to-beowulf too. | 19:39 |
linearain | it seems like now the system much faster | 19:40 |
linearain | this is on an old pentium 4 machine | 19:40 |
linearain | boot and login seems to be much faster than with systemd for some reason | 19:41 |
emdete | fsmithred: thnx for the list, i had libpolkit-gobject-consolekit-1-0 installed instead of libpolkit-gobject-elogind-1-0, which i now fixed. but still the same :/ | 19:42 |
fsmithred | did you reboot? | 19:42 |
fsmithred | also, maybe a command if I can remember it | 19:43 |
fsmithred | pam-auth-update | 19:43 |
fsmithred | I have unix and elogind checked | 19:44 |
fsmithred | emdete, what display manager are you using? | 19:45 |
fsmithred | oh, this looks different in mate | 19:47 |
emdete | lightdm | 19:48 |
fsmithred | ok, that should work | 19:49 |
emdete | is a reboot needed? | 19:49 |
fsmithred | current lightdm from devuan? | 19:49 |
emdete | i wonder why the last box just worked.. :/ | 19:49 |
emdete | yes, full devuan system | 19:49 |
emdete | (not migrated, i trashed that try ;) ) | 19:50 |
fsmithred | yeah reboot or 'init 1' | 19:50 |
fsmithred | all the kit-shit needs to start over | 19:50 |
emdete | yes, a reboot is helpful :( it works now | 19:51 |
emdete | thank you all! :) | 19:57 |
se7en | Would it be in deb-multimedia? | 21:27 |
se7en | gour: | 21:27 |
se7en | gateway2000: | 21:27 |
se7en | I think I'm still using that | 21:28 |
fsmithred | se7en, the version of libpulse0 with the epoch is from ubuntu | 21:46 |
fsmithred | or miint | 21:47 |
se7en | hmm | 21:47 |
fsmithred | what's in sources.list and sources.list.d? | 21:47 |
se7en | I have quite a few | 21:47 |
se7en | Playonlinux | 21:47 |
se7en | Wine | 21:47 |
se7en | Deb-Multimedia | 21:47 |
se7en | Mumble too | 21:47 |
se7en | deb http://ppa.launchpad.net/mumble/release/ubuntu trusty main | 21:48 |
fsmithred | there are some nifty aptitude commands that might tell where different packages came from | 21:48 |
se7en | deb-src http://ppa.launchpad.net/mumble/release/ubuntu trusty main | 21:48 |
se7en | This is the "official" mumble repo | 21:48 |
se7en | Why would they include libpulse? Don't they know it'll cause conflicts like this | 21:48 |
fsmithred | trusty is pretty old. That might account for the other dep that was pre-jessie | 21:48 |
se7en | I'll check it out. Thanks | 21:49 |
fsmithred | https://dev1galaxy.org/viewtopic.php?id=511 | 21:49 |
crashoverride | fsmithred: wrt more ram, have you seen the alleged DDR5 samsung dimms? | 22:01 |
fsmithred | nope | 22:01 |
crashoverride | 512GiB per dimm. | 22:01 |
crashoverride | ... | 22:01 |
crashoverride | 2TiB ram consumer rig. | 22:02 |
fsmithred | I just recently upgraded from a 2GHz Athlon X2 to a 2011 i5 | 22:02 |
crashoverride | puts things in perspective, right? | 22:02 |
fsmithred | I say recent, but it was pre-pandemic | 22:02 |
crashoverride | the present sounds more and more like a futuristic horror B movie. | 22:03 |
fsmithred | science fiction precedes reality | 22:04 |
crashoverride | yeah, but it's a bit more worrisome when it's science fiction HORROR that precludes reality | 22:06 |
Centurion_Dan | Hi. | 22:53 |
Centurion_Dan | I just upgraded to Chimaera on my main PC. Any thing I should watch out for?? | 22:53 |
hagbard_ | Three headed apes behind you, mostly. | 22:57 |
hagbard_ | Chimera seems to run just fine. | 22:57 |
rwp | fsmithred, My main desktop is also "new" and is an Core i5-750 first seen in 2009 but still doing well. | 23:26 |
rwp | At some point I will swap it with an Intel Xeon E3-1240 from 2011 as an upgrade since it is 2x faster but just haven't been pressed for it yet. | 23:28 |
fsmithred | run that one in the winter | 23:30 |
gnu_srs | fsmithred: I have some recent results on how to migrate from buster to beowulf. I'll try one more method before publishing on the dng ML. | 23:32 |
fsmithred | thanks | 23:32 |
fsmithred | it's not working like it used to | 23:32 |
gnu_srs | Hopefully, the buster-to-beowulf pages can be corrected. | 23:32 |
fsmithred | yeah, two methods that worked before are not working now | 23:33 |
fsmithred | someone here just did it, but that was without a desktop | 23:34 |
rwp | fsmithred, Actually we all are surprised to learn that the faster E3-1240 is lower power usage than the slower i5-750! | 23:45 |
rwp | When I put a meter on the E3-1240 it idles around 36 watts. At some point I will measure the i5-750 for a comparison. | 23:46 |
rwp | fsmithred, I also had XFCE DE upgrade issues that I need to really figure out and submit too. All related to the use of the debian-alternatives in the packaging. :-( | 23:48 |
systemdlete | I'm running into a peculiar problem with wireless on my cherry trail 2-in-1. | 23:48 |
systemdlete | Sometimes, I am able to reach the laptop from a lan host, but most of the time not. | 23:49 |
fsmithred | rwp, which alternatives? | 23:49 |
systemdlete | This is without making any changes to configs on the openwrt router, the laptop, or the host whence the request comes. | 23:49 |
fsmithred | I've upgraded a few xfce systems to chimaera, but they did not have task-* packages installed | 23:50 |
systemdlete | I've scoped both ends of the wireless portion of the connection (between laptop and openwrt router). Here is what I saw. | 23:50 |
systemdlete | You know that yesterday I installed beowulf on the laptop. I have not changed much on the laptop. The laptop can connect to the Internet, via the said router. | 23:50 |
systemdlete | When I run tcpdump on both the laptop and the router (from its console), I see arping and traffic to the laptop from the router. But on the laptop, sometimes it sees traffic intended for it, and responds accordingly, but not at other times. | 23:52 |
systemdlete | e.g., I could connect from the lan to the laptop once today. But that was in between two failed tries. | 23:52 |
fsmithred | realtek? | 23:53 |
systemdlete | yep | 23:53 |
fsmithred | 8168/9? | 23:53 |
* systemdlete ducks to avoid being smacked hard | 23:53 | |
systemdlete | no, 8812au | 23:53 |
systemdlete | no! | 23:53 |
systemdlete | sorry | 23:53 |
systemdlete | hold on | 23:53 |
systemdlete | (I'm confused due to just trying the 8812au, sorry) | 23:53 |
systemdlete | it is a Belkin | 23:53 |
fsmithred | on the laptop? | 23:54 |
systemdlete | no | 23:54 |
fsmithred | lspci | 23:54 |
systemdlete | it is whatever the cherry trail has | 23:54 |
fsmithred | oh, maybe intel | 23:54 |
systemdlete | ah, | 23:54 |
systemdlete | qualcomm atheros qca9377 ac | 23:54 |
systemdlete | belkin is on the wireless router | 23:55 |
systemdlete | just for comparison, neither my laptop nor my android phone have issue reaching the router (most of the time, except when the router is having issues; usually a router reboot clears that) | 23:56 |
fsmithred | well, you are not alone: https://forums.linuxmint.com/viewtopic.php?t=314231 | 23:57 |
systemdlete | sorry... I should have web searched the wireless info for the laptop. | 23:59 |
systemdlete | for some reason, it just didn't dawn on me. | 23:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!