--- Log opened Fri May 18 00:00:24 2018
00:08 < nekoseam> Hewwo UwU
00:19 < M6HZ> .
00:21 < blaap> Is there an alternative for fwrite() inside the kernel? I want to dump the content of a struct so I can use it in my own kernel module.
00:27 < ayecee> printk ?
00:30  * collins prints ayecee 
00:30 < acresearch> people i have a file in google colab that i want to send to myself, i am not sure how to do that, is there a way to upload it to a server or something through the terminal?
00:30 < nekoseam> sudo apt install dwm
00:30 < nekoseam> oops
00:30 < nekoseam> wrong window lol
00:35 < Kharma> Google Collab.. First I heard of it, Google always creating things and then I never find out about the until years later lol
00:36 < Kharma> acresearch: it looks like you can save to Google Drive, Github, etc.. are you headless?
00:37 < Brainspackle> Kharma: very few google products stick around long enough for me to want to get attached to them
00:37 < acresearch> Kharma: what do you mean headless? i just started using colab yesterday so quit new to it
00:37 < Kharma> I was bummed about their URL shortner being discontinued..
00:38 < Kharma> I hope their cloud compute projects stick around at least..
00:38 < acresearch> Kharma: i kind of hate google, being all a big company that spies of us, but sometimes i tolerate them when they provide free stuff, they are good in that
00:38 < acresearch> Kharma: yeh i hope so too
00:39 < Brainspackle> no they are working with the US military on machine learning drones
00:39 < Brainspackle> "don't be evil" my ass
00:39 < acresearch> anyway Kharma i am on a VM, and i want to get a file out of it, it is a terminal,
00:40 < Brainspackle> scp
00:41 < djph> Brainspackle: that ended as soon as they went public.
00:42 < Kharma> acresearch: nah, they only spy if we let them. it's quite commendable that they support open source and don't lock down their products.. like ChromeOS is open and also they encourage hacking your Chromebook :D  -- I think highly of them for that stuff, not saying others should :)
00:42 < acresearch> Kharma: i am sort of with yout as well
00:42 < xamithan> I like how they keep all our data secret so they can sell advertisements for millions.  Unlike other companies who just sell the data itself
00:43 < sadbox> Brainspackle: that's a bummer about the drones
00:43 < Kharma> and if not for them I would have never explored sysadmin .. the Always-Free a $300 credit for cloud services opened my life to a whole bunch more learning
00:43 < acresearch> Brainspackle: scp to what? is there a server i can scp to? i cannot scp to my computer though
00:43 < Kharma> Github scp to
00:43 < acresearch> Kharma: i can scp to github? really?   how?
00:44 < stevendale> Damn
00:45 < stevendale> I bet nekoseam was a furry
00:45 < Kharma> acresearch: do you have your Google Drive Mounted? using drive seems your only way via terminal options only..
00:45 < stevendale> I could've yiffed them
00:45 < M6HZ> Hi, I currently have an issue with a usb device (a camera). There is no device associated with it in /dev. Here is what I got in dmesg: "[367875.979174] usb 4-1.3: can't set config #1, error -32"
00:45 < acresearch> Kharma: dam i did not set that up, it seemed complicated
00:46 < Kharma> So, I would save to drive, mount it to your VM
00:46 < Kharma> Not at all
00:46 < acresearch> i usually get things into the VM through wget
00:46 < ayecee> M6HZ: error -32 is EPIPE, broken pipe. it seems like the device is not responding correctly to usb setup.
00:46 < M6HZ> ayecee, ok
00:46 < Kharma> I mount a bucket and my GDrive and takes me less than 5 minutes from install to mount
00:47 < M6HZ> great, what should I do then ?
00:47 < revel> acresearch: If it's Virtualbox/VMware, then you can use that "shared folder" thingy.
00:47 < acresearch> revel: Kharma is there a resource i can read to understand how to do these things?
00:47 < Kharma> acresearch: https://github.com/astrada/google-drive-ocamlfuse   -- seems complicated but uber easy.
00:48 < Kharma> using lynx might be another option, the print (P) option in Lynx can save to harddrive
00:48 < revel> Dunno. Look up "virtualbox shared folder" (that's what they call it, don't correct it to directory)
00:48 < ayecee> M6HZ: maybe connect it to another computer, see if it works there
00:49 < revel> Just install the guest additions and set it up from the VirtualBox menus and idk.
00:49 < acresearch> hmmm i am thinking or getting a temporary server and SCP to it
00:49 < acresearch> this is something i am familiar with, but i will have to pay for it
00:49 < M6HZ> ayecee, sadly I only have access to one computer currently
00:50 < acresearch> so i will keep this as my backup plan if i didn't understand your methods in time
00:50 < ayecee> that's unfortunate
00:50 < xamithan> Why can't you just scp to the computer you are using now?
00:52 < mawk> an L2 interface address is always 8 bytes ?
00:52 < ayecee> no
00:52 < Kharma> acresearch:  google-drive-ocamlfuse is handy for many things. Being able to mount Drive as a local drive has helped me in more ways than I expected -- I only expected to get additional storage space lol
00:52 < stevendale> Yiff.
00:52 < mawk> it's 8 bytes max sorry
00:52 < Kharma> I just wish Goog Drive could create a http stream for media.. would make my ideal music server.. still haven't figured out what to do with all my music
00:53 < mawk> according to my struct sockaddr_ll
00:53 < mawk> but how often it isn't 6 bytes like for ethernet ?
00:53 < ayecee> probably any time it's not ethernet
00:53 < mawk> :(
00:53 < xamithan> You could always install a http streamer software and use the mount
00:54 < stevendale> Up the bum.
00:54 < mawk> what's the best C++ type to represent such an address, std::vector<uint8_t> ?
00:54 < acresearch> Kharma: nice
00:55 < Kharma> I want something remote that I can use locally as well, without necessarily needing to mount the music to play it.
00:55 < Kharma> Confussing.. seee, I don't know exactly what I want regarding my media yet I guess lol
00:56 < xamithan> So you want it to work locally,  and remote,  and http,  without mounts ?
00:56 < xamithan> Goodluck =/
00:57 < xamithan> I thought gdrive could play media directly from the browser though
00:57 < xamithan> I watch movies sometimes that way
00:57 < sadbox> there's always plex
00:57 < sadbox> It can hook up to cloud services now
00:58 < b0bby__> hello
00:58 < b0bby__> can someone help me fix my audio?
00:58 < Kharma> I use moc, I don't use a browser much for my usual tasks anymore, I swapped GUI for CLI.
00:59 < xamithan> Maybe,  maybe not,  depends what is wrong with your audio
00:59 < Kharma> Mega can create webdavs but they only make stream for video as well.. not audio :(
01:01 < Kharma> You know what I think is a little shady.. those services that consolidate all your cloud storage into one.. all of them that I've found seem to charge.. you can get so much cloud storage for free via different service but want it all as 1 huge drive? Pay.
01:02 < ayecee> the free service is there to get you hooked
01:02 < Kharma> lol so true.. because, your storage will only grow.. so, you will eventually NEED more
01:04 < koala_man> Also, supermarket samples are free but if you want to assemble them into one big meal? Pay.
01:04 < Kharma> I tried to host a onedrive awhile ago but... I was too noob to get the step by step easy instructions to work :/
01:04 < ayecee> koala_man: nice.
01:04 < b0bby__> xamithan: I have a rear mic plugged into my motherboard and pulseaudio sees the device is plugged in but no audio can be inputted.
01:04 < mawk> qdics are mandatory for network interfaces ?
01:04 < Kharma> koala_man: nice one.
01:05 < ayecee> mawk: are you just saying random words?
01:05 < mawk> why ?
01:05 < mawk> there is a qdisc attribute for network interfaces
01:05 < xamithan> Try unmuting it,  far as I know those devices get muted by default on many distros
01:05 < mawk> and I'm wondering if it's mandatory
01:05 < mawk> see rtnetlink(7)
01:06 < Kharma> are you a bot mawk?
01:06 < Kharma> :p
01:06 < mawk> :(
01:07 < Kharma> Seriously tho I can't see the other side of our convo but am intrigued
01:07 < Kharma> your*
01:07 < b0bby__> xamithan: It is unmuted. Also
01:07 < b0bby__> When I boost the "input" it is only static
01:09 < Kharma> what are you using your mic for b0bby__ ?
01:09 < b0bby__> Kharma: just some standard condenser mic
01:10 < Kharma> is your recording stream set? have you tried rebooting?
01:11 < Kharma> I am always fighting with Pulse output, not so much input. Drives me insane though
01:11 < Loshki> mawk: https://www.linuxjournal.com/content/queueing-linux-network-stack, the section on Queueing Disciplines, implies there is a default
01:11 < mawk> yeah I just found that
01:11 < mawk> default seems to be pfifo_fast
01:12 < mawk> also there seem to be the special "noop", for loopback
01:12 < mawk> so the attribute is always sent, good
01:12 < b0bby__> Kharma: tried rebooting multiple times
01:16 < b0bby__> Kharma: should I try reinstalling alsa
01:18 < mawk> Documentation/networking/tls.txt
01:18 < mawk> I didn't know about this, that's cool
01:23 < stevendale> Spank me daddy~
01:35 < Spawndemonic> Hello everyone I could use some advice. I have a trojain i have the ipaddress and the port any recommendations on how to get rid of it?
01:35 < Spawndemonic> the proccess takes on different names and always occurs on boot
01:36 < Spawndemonic> I triend asking in sercurity channel but no one answered
01:36 < revel> Might as well format and reinstall.
01:36 < xamithan> Remove the source of the script.  Or just formatn and reinstall from backups
01:36 < Spawndemonic> if I can find the script I would love too but I don't know how
01:36 < xamithan> You don't know the delete command?  it is rm
01:37 < Spawndemonic> I know that lol but how to find the actual script itself
01:38 < xamithan> Well you know the process in netstat|ss right?
01:38 < jim> Spawndemonic, you think it "continues" because of the boot process?
01:38 < turkeyhand> I'm still looking for a solution which will restore the missing crap from the MBR on the first partition which is windows, and then fix grub which is for the 3rd partition
01:38 < xamithan> Track it down using lsof or something
01:38 < xamithan> check your crons too
01:39 < jim> Spawndemonic, are you pretty new to linux?
01:39 < Spawndemonic> jim: I see the same 3 proccess name  show up on boot it's eiter sleep,ibus-x11, or dameon-dbus
01:40 < Spawndemonic> jim I am with dealing with this
01:40 < Spawndemonic> I'll try cron I've nver used it but ill check it out
01:40 < jim> I get that, but I'm still curious: how long have you been running linux for?
01:40 < Spawndemonic> a year
01:41 < jim> ok, so that's a good amount of basic experience... what have you mostly been doing for that year?
01:42 < Spawndemonic> I've been learning to code
01:42 < jim> ok, good... linux is great at supporting coders... what language?
01:42 < Spawndemonic> ooh anything lately mostly python
01:42 < revel> BASIC
01:42 < Spawndemonic> I'm trying to figure out what I wanna be
01:43 < Spawndemonic> I've messed around with other languages
01:43 < jim> BAAASIC... slowly, I turned... step by step, inch by inch I crept...
01:44 < Spawndemonic> ya right now I'm actually trying to learn nodejs to make a webapp for a coworker
01:44 < Spawndemonic> I love linux so much better than win 10
01:44 < jim> about that, you might want to visit python coding shops to see what it's really like professionally...
01:45 < Sveta> Spawndemonic: how do you know the IP address and port of the trojain?
01:45 < Spawndemonic> no I can imagine people make some crazy stuff but most of my python experiance is from automate the boring stuff lol
01:45 < Spawndemonic> Sveta: netstat
01:45 < Spawndemonic> and conky
01:46 < Spawndemonic> my conky netwokr section so I can see both
01:46 < Sveta> Spawndemonic: what information do they show you?
01:47 < Spawndemonic> Sveta: proccess names and usage on cpu and ram and number of open ports and a list of ips with ports next to it
01:48 < mouses> turkeyhand: if the are valid partitions - boot-repair will fix them.
01:48 < Sveta> Spawndemonic: would you be able to pastebin it (with the IP and port redacted if so desired)?
01:48 < mouses> turkeyhand: I think you might have a bit of a hot mess going on with that drive
01:48 < Sveta> Spawndemonic: http://paste.linux.community
01:48 < Spawndemonic> Sveta: 142.44.242.100 port 14444
01:49 < Spawndemonic> the ip came from canada but that's all I could understand from whereis site
01:49 < mouses> Spawndemonic: are you mining crypto?
01:49 < Spawndemonic> mouses: no
01:50 < mouses> is 142.44.242.100 your public IP?
01:50 < Sveta> Spawndemonic: I am asking to pastebin, because there is other information (not the ip) which may help us figure out where this script comes from.
01:50 < mouses> Spawndemonic: have you ever tried to mine crypto?
01:50 < Spawndemonic> ooh what command do you want me to run and ill get it for you
01:51 < Spawndemonic> mouses: no never lol
01:51 < mouses> Spawndemonic: or perhaps you copy/pasted some random commands from a google search without understand what they were?
01:51 < jim> Spawndemonic, this might be handy: you can pastebin the output of an arbitrary command by running "anArbitraryCommand | nc termbin.com 9999", and to include error messages, "anArbitraryCommand 2>&1 | nc termbin.com 9999"
01:52 < LeoB> Hello! Is this the right place to ask for help about Linux Device Drivers?
01:52 < jim> Spawndemonic, to try it out, do this: echo hi there | nc termbin.com 9999
01:53 < mouses> Spawndemonic: I only see a few open ports on said machine
01:53 < Spawndemonic> that is actually very cool lol
01:53 < mouses> 22 (ssh)
01:53 < Spawndemonic> I've never see that
01:53 < mouses> 445 (some microsoft stuff?)
01:53 < mouses> 8888 and 9999 (something you did)?
01:53 < jim> LeoB, maybe... in any case, go ahead and ask (and throw a lot of informative details into your question) and we can figure out from there
01:54 < Spawndemonic> mouses: I'm sorry what do you mean?
01:55 < Spawndemonic> Sveta: which command would you like me to run that I can paste for you
01:55 < LeoB> I have a laptop computer with a backlight keyboard. It works on Windows, but not on Linux.
01:56 < mouses> LeoB: what is the model number of said computer
01:56 < jim> Spawndemonic, did you try it out?
01:56 < Spawndemonic> mouses: I do dual boot with windows
01:56 < Spawndemonic> jim: ya
01:56 < Spawndemonic> I thought it was cool
01:56 < Spawndemonic> didn't know I could do that
01:56 < mouses> Spawndemonic: nmap --top-ports 20 localhost | nc termbin.com 9999
01:57 < jim> Spawndemonic. ok... now, you can pastebin any command sveta names for you, when nc tells you the termbin url that came back, give it to sveta, or just post it here
01:58 < Spawndemonic> mouses: http://termbin.com/k3nh
01:58 < jim> be careful... nmap is a port scanner
01:59 < jim> and, he's using it for helpful purposes
01:59 < Spawndemonic> I hope so anyway lol
01:59 < jim> not everyone will do that, so just be careful
02:00 < Spawndemonic> Its not there this time
02:00 < mouses> jim: +++
02:01 < jim> NO CARRIER
02:01 < Spawndemonic> I just did "netstat -tulpan > netscan.txt && gnome-open netscan.txt" but I don't see that prot this time
02:01 < mouses> Spawndemonic: I see nothing out of the normal or worth worrying about there
02:01 < Spawndemonic> hold on
02:01 < Pentode> LeoB, try entering: "xset led 3" in terminal and see what happens.
02:01 < Spawndemonic> I took a screen shot gunna share that you'll see what I see right now
02:02 < LeoB> mouses, Avell G1513 MXTI
02:03 < LeoB> is there a way to find this out using the linux subsystem? /sys ?
02:04 < Spawndemonic> anyone have a recommended way to share images on here
02:05 < stevendale> Spawndemonic Imgur
02:05 < jim> imgur.com?
02:05 < djph> ^
02:05 < skatetensor> hello
02:05 < jim> hi
02:05 < stevendale> Hi skatetensor
02:06 < stevendale> You shouldn't be logged in as root
02:06 < acresearch> i cannot accept an SSH from jupyter notebook in google, i hate jupyter notebooks, how can i pass a YES to scp ID@IP ?
02:06 < skatetensor> ignore -channel #Linux QUITS JOINS PARTS
02:06 < stevendale> Using Windows XP is more secure than using IRC, web browser, etc as root on Linux
02:06 < skatetensor> oops
02:06 < Spawndemonic> https://imgur.com/a/gIpvjPX
02:06 < jim> hmm. i kinda like jupy
02:08 < Spawndemonic> So you can see this time its ibus-x11 and you can see the ip and that 14444 port at the bottom
02:09 < acresearch> how can i pass a password without typing it? you know from the executing command   i tried   password | scp FILE ID@IP   but that did not work,  any suggestions?
02:10 < skatetensor> acresearch: you can always put it in a script
02:10 < revel> acresearch: `man ssh_config`
02:11 < acresearch> skatetensor: revel not possible in a jupyter notebook,   unless there is somethingi don't know
02:11 < revel> It doesn't have openssh...?
02:11 < djph> use a ssh key?
02:11 < acresearch> revel: cannot interact with the terminal just submit commands to it
02:11 < djph> ... I'm missing something here ...
02:11 < revel> Oh, wait, yeah, that's what I meant, not ssh_config. Stupid me.
02:12 < jim> probably the safest way is to make an ssh key pair with a blank passphrase
02:12 < jim> OR
02:12 < djph> so how're you even sending the 'scp' command if you have a non-interactive shell?
02:12 < Henry151> hey guys
02:13 < jim> you can ssh-add and it will ask you for the passphrases of the keys it knows,,, later, when you try to ssh, it will use the key that's been passed to the server
02:13 < acresearch> djph: its half an interactive shell   it jupyter notebook where you can only submit python scripts, but send some shell commands, but cannot interact with the shell when it prompts  like when it asks you for a password
02:13 < Henry151> any builtins in Debian that can make a 47mb .mp4 video file smaller (less than 25mb) without hurting the video quality too much?
02:13 < revel> acresearch: Do those commands include, y'know, any that can make the directory ~/.ssh on the remote host and stick a file there (plus maybe change said file's permissions)? Since then you can use pubkey auth.
02:13 < djph> ... oh, it's a /notebook/, not a PC ...
02:13 < Spawndemonic> mouses: did you see the image I posted
02:14 < acresearch> djph: yes notebook
02:14 < djph> Henry151: doubt it.
02:14 < Henry151> like by changing the format or shrinking the size of the video or anything that won't hurt the clarity too much (because there is some relatively small text in the video that needs to still be readable)
02:14 < revel> Assuming it's openssh or something that checks authorized_keys
02:14 < Henry151> djph: or anything I can download with apt?
02:14 < Henry151> I think i can do it with VLC media player
02:14 < Henry151> I just don't know how
02:15 < acresearch> revel: i don't think so, it is very limited, for example i cannot call vim or nano, because i cannot interact with the shell when propmted
02:15 < djph> Henry151: I more meant your "don't hurt the clarity when you transcode" bit.  What's it been encoded with already?  I mean, if it's already H.264 ...
02:15 < revel> acresearch: You don't really need more than, idk, cat or echo for that.
02:15 < acresearch> revel: how would i construct the command?
02:15 < revel> acresearch: Actually, can't you just `ssh -t user@host sh`?
02:16 < acresearch> revel: sh would be the password?
02:16 < revel> acresearch: No.
02:16 < acresearch> revel: i cannot ssh because i cannot interact with the shell
02:16 < Speed2u> is it possible to show motd when logging in with ssh if using key based auth only, with usePAM = no? noticed a line in sshd_config "printmotd" but that did not have any effect
02:16 < revel> Anyways, `ssh user@host "mkdir ~/.ssh; cat ~/.ssh/id_*.pub | ssh user@host "cat > ~/.ssh/authorized_keys`
02:17 < djph> Speed2u: I just stuck a motd in my ~/.bash_profile
02:17 < revel> Err, missing some double quotes at the end.
02:17 < djph> (errr)
02:17 < jml2> revel, ssh-copy-id command can do this automatically
02:17 < jml2> revel, much easier to use that command
02:17 < revel> Hmm. Right.
02:17 < jml2> lol
02:17 < stevendale> Pound me daddy
02:17 < revel> acresearch: Should probably get a better terminal emulator.
02:17 < Speed2u> djph: and that will display to logged in users automatically? thx will try that
02:18 < stevendale> Euro me daddy xD
02:18 < acresearch> revel: it is a jupyter notebook it is not a terminal
02:18 < revel> acresearch: Wait, are you connecting *to* or *from* it?
02:18 < djph> Speed2u: well, it's *my* login anyway.  If you want a static MOTD, that goes in /etc/motd (IIRC)
02:18 < acresearch> revel: from it
02:18 < revel> acresearch: And what OS is it running?
02:18 < djph> Speed2u: I just like seeing jokes, so /usr/bin/fortune it is :)
02:19 < djph> /usr/games/fortune *sigh*
02:19 < acresearch> revel: some linux os on the VM on the web broswer  it is google colab
02:19 < revel> And what's the notebook *itself* running? Since there's probably better options for terminal emulators for it.
02:19 < stevendale> cmd.exe
02:19 < revel> Never that.
02:20 < meyou> any mail filter admins noticing a lot higher allow:block ratio than usual lately?
02:20 < djph> no
02:20 < meyou> wondering if it's just our filters turning bad or more eager spammers
02:21 < stevendale> You are part of a botnet meyou
02:21 < meyou> we usually are about .9:1, past couple weeks it's been like 4:1
02:21 < meyou> neat
02:21 < stevendale> You should turn off your computer at the next available convience
02:21 < meyou> hilarious
02:23 < Speed2u> djph not really understanding your jokes (time to sleep long time ago here) but it worked so thank you :) just put cat /etc/motd into .bash_profile
02:23 < stevendale> Spank me daddy
02:24 < Pentode> LeoB, did that work for you? it _should_ be that easy. You can add a minus in front of led to turn it off...
02:25 < skatetensor> So the eternal pen question
02:25 < skatetensor> kali or parrot?
02:26 < xamithan> blackarch
02:28 < skatetensor> meh
02:28 < skatetensor> i mean the 1500+ preinstalled tools are nice but
02:28 < skatetensor> nah
02:29 < stevendale> Linux doesn't work with my touchpad
02:29 < stevendale> Only Windoed XP does
02:29 < toeshred> On Linux, my Intel 2600K-i7 cpu now shows only 4 cores in /proc/cpuinfo, and on htop. it used to show 8 cores. Any idea what changed?
02:29 < skatetensor> maybe your touchpad doesn't work with linux :)
02:34 < djph> Speed2u: there is a program called "fortune" that can be installed.  When run, it prints random text - fortunes, jokes, and the like.
02:45 < skatetensor> Really hate the guys at all the support channels who act like being annoyed
02:45 < skatetensor> i mean you are in a ubuntu channel don't get annoyed if most of the awnswers are basic
02:46 < stevendale> I use TOR :3
02:46 < mawk> level up stevendale , put a transparent Tor on your router
02:47 < stevendale> :C
02:47 < stevendale> I don't have access to my router
02:47 < stevendale> And my dad is overseas
02:47 < stevendale> I could brute force it but I'd get into trouble
02:47 < mawk> get up and sit next to your router
02:47 < mawk> then you have access to it
02:48 < ayecee> clever
02:48 < mawk> isn't it
02:48 < revel> stevendale: You don't. I do.
02:48 < revel> Unless you've somehow repeatedly managed to find exit nodes that aren't banned.
02:49 < dannylee> 2~2~2~2~
02:50 < toeshred> so has the kernel made any changes in the way it sees CPU's? both htop and /proc/cpuinfo used to show 8 cores, but now show only 4.
02:50 < stevendale> My router uses RS232
02:50 < stevendale> Not Ethernet
02:51 < bls> toeshred: maybe something in the way hyperthreading is dealt with?
02:52 < djph> your dad turned off ssh?  good move.
02:52 < djph> too bad he forgot to revoke your access entirely.
02:53 < revel> toeshred: I'm betting it's always been 4 cores, just with 8 threads.
02:53 < revel> Though the kernel can be configured to see a maximum of N cores/threads as well, I think...
02:54 < revel> Dunno. What changed it?
02:54 < stevendale> So can Windows :D
02:54 < stevendale> NUMPROC=
02:54 < stevendale> In msconfig
02:54 < phil42> !ubottu language
02:54 < stevendale> BOOT.ini
02:55 < skatetensor> wait stevendale is not a bot?
02:55 < skatetensor> :3
02:55 < stevendale> I'm not a bot owo
02:55 < xamithan> I don't think the ubuntu channel bot is in here
02:55 < stevendale> I'm a furry skatetensor
02:56 < skatetensor> ah skatetensor was my nick when i was 15
02:56 < skatetensor> probably should change it
02:56 < Guest43073> wat fug
02:56 < toeshred> revel: yes it is 4 physical cores + hyperthreading. but htop and /proc/cpuinfo used to show all 8, but now they don't. it's just wierd. I am wondering if the kernel no longer counts virtual cores to be seen by these tools.
02:57 < stevendale> Hi nekUwU Are you here to troll or do ya wanna yiff? :3 Either of the two, /part is your best option
02:57 < revel> toeshred: Well, they both show me 8, for what that's worth. Could be something specific to your distro's kernel or something, idk
02:57 < ayecee> stevendale: shush
02:57 < nekUwU> OwO; h-huh
02:57 < revel> (4 physical cores, 8 threads, like you)
02:58 < nekUwU> n-no thats not right here
02:58 < nekUwU> UmU
02:58 < revel> nekUwU: stevendale: https://translate.google.com/#ig/en/Owo%20uwu%20owo
02:58 < xamithan> toeshred: If you got the HT flag it is activated
02:59 < xamithan> Might want to check your bios to see if it got disabled though
02:59 < revel> Could have disabled hyperthreading in the kernel too, for some reason... Would be easier to check if /proc/config.gz is there.
02:59 < toeshred> xamithan: yep i am getting ht flag in cpuinfo. also bios shows hyperthreading enabled.
03:00 < xamithan> That is strange then.  I'd test a liveusb myself
03:01 < nekUwU> http://0pointer.de/blog/projects/the-biggest-myths.html
03:26 < cxc99> where would i find selinux contexts like samba_share_t and nfs_t in the man pages?
03:28 < xamithan> Install the man pages
03:28 < stevendale> Spank me daddy <3
03:28 < xamithan> selinux-policy-devel,   mandb,  man -k _selinux
03:31 < xamithan> I think that is right.  They definitely aren't there by default
03:34 < dannylee> its a great day tooo be alive with my linux machine...
03:34 < dannylee> openSuse open m,y eyes
03:34 < cxc99> should i just memorize it?
03:35 < cxc99> i installed selinux-policy-devel and don't find samba_share_t in there
03:37 < stevendale> Help
03:37 < stevendale> How do I grow weed
03:37 < dannylee> under a light
03:37 < dannylee> in rogan
03:38 < Sveta> stevendale: this is the last time we want to see non-linux questions in this room
03:38 < dannylee> hydroponic
03:38 < bluezinc> can someone just kick him?
03:38 < Sveta> stevendale: the next time they are seen, you may be penalised immediately
03:39 < Sveta> stevendale: without a warning
03:39 < stevendale> Sveta Whatever
03:40 < Sveta> stevendale: it is recommended to find a social channel at freenode at which you feel comfortable
03:40 < Sveta> stevendale: it can be at another irc network or it can be a personal blog
03:40 < Sveta> stevendale: the best is having a friend who has the capability to listen
03:41 < xamithan> Did you run the mandb to regenerate the man pages?  Sometimes you need to do a sepolicy manpage -a -p /usr/share/man/man8 first
03:41 < xamithan> Depends on distro version
03:42 < Sveta> thanks xamithan
03:42 < stevendale> I never get thanks for helping people
03:42 < stevendale> Outrageous
03:43 < cxc99> xamithan, i didn't know i had to do that trying it now thanks
03:43 < Sveta> stevendale: thanks for advising someone to not irc as root 2 hours ago
03:44 < rascul> i used to have an identd setup to give a fake ident so that it looks like i'm root on irc
03:45 < rascul> had that for years, nobody ever said anything :(
03:46 < rascul> i guess i didn't even need the identd but still i was much disappointed, all that trouble and nobody noticed
03:46 < cxc99> lol rascul
03:47 < cxc99> i remember the scripts that would autokick people for running as root
03:48 < bluezinc> cxc99: those were such a PITA...
03:48 < bluezinc> if I want to IRC as root, I'm darn well going to.
03:49 < xamithan> It should kick people as ident kali
03:49 < xamithan> Er,  amnesia
03:49 < Sveta> cxc99: at some channels at efnet they still do that
03:49 < nekoseam> h-hewwo frens i'm back UwU
03:50 < nekoseam> fluxbox vs openbox, which do you prefer? UwU
03:50 < jml2> no need for selinux on generating manpages lol
03:50 < jml2> that's clownish
03:51 < Sveta> jml2: they want man pages about  samba_share_t and nfs_t
03:51 < jml2> usually root runs man-caching update via cron
03:52 < xamithan> Sure if you want to reboot instead of typing mandb =o
03:53 < jim> bluezinc, ok... but you could get pwned rather hard :) someone could potentially get access to the underlying shell and... well, format your hd for you... or do any of a number of things that are not quite as severe but could still have unexpected consequences
03:53 < cxc99> well i already did type mandb...
03:53 < cxc99> i was just seeing if instead of memorizing them i could use the man pages, but i don't think they're included
03:54 < nekoseam> c-can someone explain what the deal is with xmonad?
03:54 < xamithan> Yeah you got to memorize the package name
03:54 < nekoseam> I've heard good things about it
03:54 < nekoseam> But can't really seem to figure out what the fuss is about UmU
03:54 < bluezinc> jim: they couldn't do much worse than they could with just a normal user account.
03:55 < bluezinc> obligatory XKCD: https://xkcd.com/1200/
03:58 < Roserin> I knew which one that was without opening it
03:58 < jml2> full samba docs are in a separate package usually -- as is many other server things that take much configuration options
04:01 < jim> bluezinc, well if they got into a shell, they could do some serious damage if they wanted to
04:03 < dannylee> only in root..its really hard too crack your password...
04:03 < jim> if they got a root shell, they could just change it
04:03 < Dreaman> no chance :)
04:04 < Dreaman> name is ok but pass
04:05 < xamithan> What if I deleted the passwd command
04:05 < jim> well thre are ways
04:05 < fryguy> any recommendations for a music server daemon I can install on my server that will allow me to access/play music from remote places (my phone or laptop for example).  Seems like a lot of things like mopidy and mpd are servers but primarily for local playback.  I currently do mpd over nfs, but that ends up requiring a vpn to get the nfs access, and nfs over road-warrior type vpn isn't great.
04:05 < bluezinc> jim: the point is that they can already screw everything up with a normal user account...
04:06 < bluezinc> if I'm using a VPS as root, IDGAF if they wipe the whole disk.
04:06 < jim> well they couldn't format the hd
04:06 < ayecee> fryguy: what program do you want to play with? that will decide what you can use.
04:06 < bluezinc> jim: and you consider that an improvement?
04:07 < jim> yeah, maybe your whole home dir could be gone... but not the overall system
04:07 < bluezinc> honestly, if hackers make it into my systems, I *hope* that they wipe the disk...
04:07 < fryguy> ayecee: I'm pretty flexible in that regard, just looking to survey my options.  I'd probably prefer having an mpd client, but I don't know what other options there are to consider
04:07 < dannylee> i just have 3 or 4 user names
04:08 < bluezinc> jim: and the home dir represents every fucking thing that's useful on a computer.
04:08 < dannylee> they can only kill me once
04:08 < bluezinc> re-imaging the OS takes maybe an hour, tops.
04:08 < ayecee> fryguy: it will be easier to explore the options if you lock down some details, if only temporarily
04:08 < bluezinc> it's the homedir that we should really be trying to protect.
04:08 < dannylee> its thje online banking you have to protect
04:09 < fryguy> ayecee: ok, let's say I want to playback using ncmpcpp
04:11 < rascul> then you need to use mpd
04:11 < fryguy> rascul: right, but what is the recommended setup for remote streaming
04:12 < rascul> you run mpd at home, do the dynamic dns thing, connect to it from the world
04:12 < rascul> iirc mpd does unix socket by default so you have to configure it to listen to ip and port
04:12 < fryguy> rascul: ok, and controlling the mpd running at home will try to playback on pulse/etc on the server, and not on the location of the mpd client, right?
04:12 < rascul> oh
04:12 < fryguy> i want the music to playback on the client
04:13 < rascul> i haven't done the mpd in a lot of years, i guess i had forgot that part
04:13 < fryguy> I have a cache of music "somewhere" and I want an interface to play back that music
04:13 < fryguy> i'm open to alternatives to mpd if there's a better choice for that usecase
04:14 < ayecee> fryguy: to me, it seems like it'd be easier to expose it as a collection of files, say, by webdav
04:14 < ayecee> for that, nextcloud might be a handy thing
04:14 < rascul> yeah that's probably the better way
04:14 < rascul> there's ampache but i find that really weird
04:15 < rascul> doesn't nextcloud have a music player plugin or something?
04:15 < ayecee> probably yes
04:15 < fryguy> ayecee: does mpd have the ability to index/playback from webdav?
04:15 < rascul> https://apps.nextcloud.com/apps/music
04:15 < ayecee> i don't know
04:16 < rascul> yes https://www.musicpd.org/doc/user/storage_plugins.html
04:16 < fryguy> yep just found that
04:16 < toeshred> revel: xamithan:  it ended up being my kernel parameter (acpi=off) that was causing it. I changed it to "acpi=ht". Now I see all 8 cores.
04:28 < granttrec> is there now way to install mupdf via distribution repos, do I just install via source?
04:28 < Sveta> granttrec: what distribution?
04:29 < granttrec> ubuntu
04:29 < Sveta> granttrec: there is mupdf 1.12.0+ds1-1 in Ubuntu Bionic
04:30 < granttrec> my bad then, didn't come up in aptitude
04:31 < granttrec> i'm still on 16.04 tho
04:32 < meingtsla> Yes, which also has mupdf.. version 1.7a-1
04:32 < meingtsla> What repos do you have enabled?
04:32 < Sveta> granttrec: there is mupdf 1.7a-1 for Ubuntu Xenial (16.04)
04:32 < Sveta> meingtsla: it is in universe
04:33 < granttrec> Sveta: yeah I can find it with apt, just trying to get the hang of aptitude, doesn't show up for some resion
04:33 < meingtsla> Sveta: Right so maybe they have to enable that repo?
04:34 < granttrec> meingtsla: thats a good point
04:37 < nekoseam> UwU...
04:44 < dannylee> security is really important..i'm just a bit too carefull with my online banking...but its better to be safe than sorry..linux is ten time better than window,,but most ATM machine are windows base..but the main fame computer may be linux or unix...
04:48 < Sveta> dannylee: it is advised to keep a record of each transaction you make so that you are able to match it against the records in the bank once in a while.
04:49 < dannylee> ya that true
04:49 < dannylee> i just did my taxes and i'm not broke yet..but ill be carefull
04:49 < Sveta> dannylee: there are techniques which may help you with this, such as keeping a record of transactions on paper immediately, and add notes from paper into software such as gnucash once in a while.
04:50 < dannylee> ok c0000l
04:56 < comjf> nouveau is frustrating me, any suggestions on how to just completely disable it from ever loading?
04:57 < comjf> I've added a /etc/modprobe/blacklist.conf line with no affect... even rd.driver.blacklist=nouveau in grub.... still it plagues me
05:10 < stevendale> Hey
05:10 < stevendale> comjf What exactly are you trying to do / run?
05:11 < wadadli> anyone familair with getting dnf to skil checksum verification?
05:11 < comjf> stevendale: basically this: http://dpaste.com/29WHJ0H
05:12 < comjf> I'd expect nouveau to not be loaded with the above configured (on fedora 28 if that makes a difference... don't think it should)
05:12 < stevendale> Hmmm....
05:18 < stevendale> comjf: Have you looked at this page? https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#ixzz4rQODN0jy
05:21 < comjf> stevendale: no... has some useful info I can try.. though I'm ultimately using this for a pci-passthrough and not installing CUDA directly
05:21 < stevendale> Ah
05:22 < stevendale> comjf I sent you a PM with a command, did you remember to rebuild the initramfs?
05:22 < jml2> comjf, you should be sure you're using dracut and know if your driver is either a module or built into the kernel image
05:22 < jml2> comjf, if you don't know those 3 things, then you're bubbling in a cold bath :)
05:23 < jml2> and um why is he bother to try this on a live-cd ?
05:23 < jml2> LOL
05:23  * jml2 hides XD
05:25 < celmor> I'm trying to create a dump, my system uses coredumpctl and using gdb doesn't work (freezes when attaching)
05:30 < wadadli> I've got a version of a package that
05:31 < wadadli> isn't available in the repos anymore, does rpm have some magic for getting all of it's dependencies so that I can install the package on other machiens?
05:34 < topicali> what's good way to (from the CLI) get the system specs (regarding cpu/memory/disk space)?
05:34 < topicali> in a pretty format
05:34 < jml2> topicali, dunno, maybe lwsh vs dmidecode XD
05:35 < topicali> i know cat /proc/cpuinfo, etc.
05:35 < jml2> topicali, lshw  -- typo
05:35 < jml2> topicali, shows things in tree-leaf output
05:35 < jml2> topicali, can be output to html
05:37 < topicali> my bigger question is if someone could tell me if my server specs are suitable to handle a ~50+ clients simultaneously connecting to a wordpress install
05:37 < topicali> i think they are, but i never solely trust my own opinion
05:38 < jml2> topicali, that's nothing
05:38 < jml2> topicali, a common workstation can easily handle reading -- but simulatenously "writing" is another issue
05:38 < topicali> jml2: also running sandstorm (https://sandstorm.io/)
05:39 < topicali> the server handles it great in production, just want to make sure i'm not missing anything regarding specs
05:39 < topicali> it's a dedicated SSD
05:40 < topicali> i'd rather have overkill than me being worried :)
05:41 < topicali> here are the basic specs: https://pastebin.ca/4029415
05:41 < topicali> hopefully that's overkill for my purposes
05:45 < topicali>  hdparm -Tt: Timing cached reads:   31462 MB in  1.99 seconds = 15785.21 MB/sec ............ Timing buffered disk reads: 1590 MB in  3.00 seconds = 529.94 MB/sec
05:46 < topicali> err, that's read speeds
05:46 < topicali> also want to make sure i'm using relevant benchmarks
05:49 < jml2> topicali, ssd's aren't reliable unless they're in a reaid set
05:49 < jml2> (/raid./ set --)
05:50 < jml2> add some redundancy that's what I say
05:50 < topicali> gotcha
05:50 < topicali> there an easy way to determine if raid is in use on the server (via CLI)
05:50 < topicali> not too worried about data loss tho
05:51 < topicali> have it set to backup regularly to a remote server
05:51 < topicali> mostly concerned, for the purposes of this server, with performance
05:52 < topicali> ~$ dd if=/dev/zero of=/tmp/test1.img bs=5G count=1 oflag=dsync \ 2147479552 bytes (2.1 GB) copied, 4.83528 s, 444 MB/s
05:54 < lord|> bs=5G?
05:54 < lord|> doesn't that fill up the memory with 5GiB of zeroes
05:55 < topicali> lord|: probably (i'm not super privy to disk r/w benchmarking)
05:55 < topicali> what would be a more appropriate/useful command
05:55 < lord|> bs=512 count=10485760
05:56 < topicali> ah, guessing bs = blocksize?
05:56 < lord|> yeah
05:57 < lord|> count is # of blocks
05:57 < lord|> idk if on a modern SSD/hard drive bs=4K would be faster
05:57 < topicali> i entered that command (dd if=/dev/zero of=/tmp/test1.img bs=512 count=10485760 oflag=dsync) and it's still processing it
05:57 < lord|> kernel probably handles that the same
05:58 < lord|> status=progress btw
05:58 < lord|> to see throughput without having to use pv
05:58 < topicali> good to know
05:58 < topicali> hmm, it's at around 315 kB/s
05:59 < topicali> 313kB/s 60secs in
06:06 < francute> Suddenly, appeared a giant "Terminal" text in my terminal e.e
06:07 < stevendale> I am back
06:07 < francute> Hello, back
06:10 < EL3CTR0N> exit
06:20 < uzdw> welcome back
06:50 < Jonno_FTW> I have a problem
06:51 < Jonno_FTW> I keep getting this error: pyopencl/_cffi.abi3.so: undefined symbol: _ZTVN10__cxxabiv120__si_class_type_infoE
06:51 < Jonno_FTW> despite the fact that the .so file is linked to libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6
06:52 < Jonno_FTW> and that libstdc++.so.6 file has that name in it
06:53 < neveroptimized> Do you have more context to the problem? Did you enter in a command, is it related to a program, or is it just randomly popping up?
06:53 < Jonno_FTW> neveroptimized: it pops every time I try to import that cffi.abi3.so file
06:53 < pingfloyd> do you have the headers for that installed though?
06:54 < Jonno_FTW> the headers for what?
06:54 < pingfloyd> some dists split the headers from the runtime lib
06:54 < pingfloyd> debian and rh for instance
06:54 < Jonno_FTW> i'm on ubuntu
06:54 < pingfloyd> is this error coming up when compiling?
06:55 < Jonno_FTW> here's the output http://pastebin.ws/9d6v73
06:55 < Jonno_FTW> this comes up after compiling
06:55 < Jonno_FTW> the file was compiled with -lstdc++
06:56 < pingfloyd> do you have the libstdc++-6-dev package installed?
06:57 < pingfloyd> also a good idea to install build-essential if you haven't already
06:57 < Jonno_FTW> yes libstdc++ and build-essential are installed
06:57 < pingfloyd> that's not what I asked
06:57 < pingfloyd> libstdc++-6-dev
06:57 < jack_rip_vim> hi pingfloyd
06:58 < Jonno_FTW> no such package exists
06:58 < pingfloyd> libstdc++-6 is the runtime portion of the lib
06:58 < jack_rip_vim> apt-cache search libstdc*?
06:59 < pingfloyd> try apt-file search libstdc++ | grep '\-dev'
06:59 < pingfloyd> ubuntu may name it slightly different than in debian
06:59 < pingfloyd> but with ubuntu you need to matching -dev packages install in order to compile successfully.
07:00 < pingfloyd> just like with Debian
07:00 < Jonno_FTW> there's libstdc++6-4.7-dev
07:01 < pingfloyd> what's dpkg -l "libstdc++*" | grep ^ii return
07:01 < jack_rip_vim> Jonno_FTW: I think you can compile it from source.
07:02 < Jonno_FTW> jack_rip_vim: compile what?
07:02 < jack_rip_vim> libstdc++
07:02 < pingfloyd> don't do that
07:03 < jack_rip_vim> =_=
07:03 < pingfloyd> that would open up a can of worms in having to manage multiple versions of libstdc++ and any other software linked against it.
07:03 < pingfloyd> should only go there as a last resort
07:04 < Jonno_FTW> http://paste.ubuntu.com/p/HfCfwwqDC2/
07:04 < pingfloyd> since it is going to introduce a lot of extra work in maintenance
07:05 < pingfloyd> apt-get install libstdc++-dev
07:05 < jack_rip_vim> seem like there are already a lot of choices.
07:06 < pingfloyd> then try compiling again to see if you get past the point before
07:06 < jack_rip_vim> hi OtakuSenpai
07:06 < pingfloyd> usually there will be packages without version part of the name field itself, those will be the default version
07:07 < pingfloyd> typing
07:08  * jack_rip_vim is thinking pingfloyd may think compiling is the best chioce 
07:08 < jikistro> in alot of technical books netcat is considered as swiss army knife of troubleshooting network problems
07:08 < Jonno_FTW> pingfloyd: http://pastebin.ws/2npaq1
07:09 < pingfloyd> jack_rip_vim: his problem is he's having troubles with compiling because he's missing the headers for libstdc++ (version that matched the runtime package for the same lib).
07:09 < jikistro> but if i want to do something simple like testing port connection it can give some false info
07:09 < jikistro> ok if i do nc -v -u ip 123 to test ntpd connection then it reports being connected which is fine
07:10 < jack_rip_vim> pingfloyd: which C headers lost?
07:10 < jikistro> but change it to any other port that it is not open on that host say 124 again it reports as connected
07:10 < pingfloyd> Jonno_FTW: closest I see in your case is libstdc++6-4.7-dev package
07:10 < pingfloyd> try that one
07:10 < pingfloyd> debian is would be simple
07:11 < pingfloyd> since the package is named like libstdc++-6-dev
07:11 < Jonno_FTW> ok I fixed it
07:11 < jack_rip_vim> pingfloyd: you should let him print g++ -v.
07:11 < Jonno_FTW> yeah ubuntu is shite
07:11 < Jonno_FTW> I don't have a choice in the matter
07:11 < jikistro> am i missing something? maybe it's working fine but i just lack some knowledge of networking?
07:11 < pingfloyd> ubuntu thinks they have a better idea at naming packages than debian or something
07:11 < Jonno_FTW> pingfloyd: it may have been a problem of me not uninstalling the old version of pyopencl
07:12 < pingfloyd> even though they're based on debian
07:12 < jikistro> even if so then it makes it useless for testing connection to an UDP service
07:13 < jack_rip_vim> there is always one distro base on the other one.
07:14 < jack_rip_vim> to handle a distro from source code need a lot of time. I do want to do it, but I still need to work.
07:15  * jack_rip_vim wishes he has a full time to make a real distro.
07:16 < pingfloyd> would be nice to be able to develop a distro and make enough money you can do that full time
07:16 < Sveta> make a debian blend with proper documentation and bug submission mechanism
07:16 < Sveta> maybe that works
07:17 < Jonno_FTW> jack_rip_vim: linux from scratch then
07:17 < jack_rip_vim> pingfloyd: yeah, I didn't have enough money, so it is hard for me to keep it
07:17 < Sveta> a lot of linux distributions do not have adequate documentation, and they do not have a big fat text area where users can complain about it either. i'd ship with a help viewer in which there is a text area at the bottom where users can write any feedback and click submit
07:17 < pingfloyd> jack_rip_vim: yeah, that's always the conundrum unless you happen to already have enough money you don't have to work.
07:18 < Sveta> then i know what help pages they're reading and whether they're happy
07:18 < Jonno_FTW> Sveta: who is going to read this feedback?
07:18 < pingfloyd> building and supporting a dist would be a great full time job
07:18 < pingfloyd> dream job for me really
07:18 < Jonno_FTW> would be ridiculous once you have too many users
07:18 < jack_rip_vim> Jonno_FTW: A real a distro start up from source code, linux from scratch it is just a simple manual.
07:19 < pingfloyd> Jonno_FTW: ideally at that point it becomes big enough you can bring more help in
07:19 < Sveta> Jonno_FTW: it'll have a score of how displeased the user is, eventually pages with highest negativity scores will be prioritised for improving them. the volunteers who work on these pages will be able to view the submitted feedback
07:20 < Sveta> Jonno_FTW: there are other ways to prioritise improvements of documentation, but they involve forcing users to view it online, keeping track of how many visitors each page has, have simple ':)' or ':(' buttons next to each page, keeping track how many pages are visited by people whose browser language setting is set to a language the page is not translated to etc
07:20 < surrealpie> what's the best FOSS upnp server for linux
07:20  * jack_rip_vim wishes he has time, more time
07:20 < Sveta> Jonno_FTW: but in my opinion a distribution needs to have an offline help viewer and when people go online their feedback from earlier is queued for delivery
07:21 < neveroptimized> You'd need to have some form of spam protection I'd imagine, else you may have certain people spamming their pet issue to force greater visibility
07:21 < Jonno_FTW> offline help is in the man pages
07:21 < Jonno_FTW> the difficulty is that there's such an absurd amount of offline pages
07:21 < Sveta> neveroptimized: when it's a local LAN computer instead of a website, i don't think people would spam as much as they do on online wikis and websites
07:21 < Sveta> Jonno_FTW: the help viewer would perhaps have a search box at the top
07:21 < Jonno_FTW> there's so many tasks you can do and peolpe will want to compile things that use specific old versions of libraries
07:22 < Sveta> Jonno_FTW: and if a user is not able to find something, they write a couple swear words in the box at the bottom, the help viewer takes note of what search query the user was using, and adds it to a special place for search queries
07:22 < Sveta> Jonno_FTW: i think that a vast majority of people trying a linux distribution just want very basic things like email and web browsing and documentation is lacking even in some of that
07:23 < Jonno_FTW> or you could force people to pass a mooc before they can do the distro
07:23 < Sveta> Jonno_FTW: but i agree that such a system would be successfully used for creating programmer oriented documentation as well - a lot of existing help files lack a troubleshooting section
07:24 < Sveta> Jonno_FTW: i'd like it to be easy for people to use a distribution without passing anything, and to have access to high quality documentation that is all centered in a universal help viewer without forcing them to visit the terminal
07:24 < jim> do the distro twistro?
07:24 < jack_rip_vim> hi jim
07:24 < jim> heya jack
07:24 < neveroptimized> For some reason, I imagined some giant dichotomous key sort of tool to diagnose any given issue
07:24 < Sveta> jim: the plan is to use a debian blend, but i've only found gnustep elegant so far, and its help viewer is nonexistant, so i'm planning to do a nasty hack that converts all help pages to pdf and allows them to view that
07:24 < jack_rip_vim> jim, :)
07:25 < neveroptimized> It just keeps asking you yes/no questions until you've narrowed it down or something
07:25 < Sveta> jim: so man pages or /usr/share/doc/foobar/foobar.html, either of that can be converted to pdf and be available to people to view, because that's the only file format which has an adequate viewer for it for gnustep and supports text formatting and images
07:26 < Sveta> neveroptimized: that's what windows xp had a long time ago, you'd have no internet connection and it'd ask you 'do you have a modem, is it powered on, is the ethernet cable plugged in, is the computer switched on, pls reboot modem, pls disable ethernet and enable it again'
07:26 < jim> Sveta, oh, this is for you?
07:26 < Sveta> neveroptimized: it was not terribly bad
07:27 < Sveta> jim: i'm considering making such a debian blend which has a help viewer where users can add feedback, however i think it was pingfloyd who first remarked about making a new distro and making enough money to be able to do it full time
07:27 < stevendale> And here I was getting in trouble for mentioning Windows, but Sveta goes and mentions it too :(
07:27 < Sveta> it's relevant to the current discussion, sorry mate
07:27 < jim> like a wiki?
07:27 < Sveta> yes, but offline
07:28 < Sveta> their feedback would need to be able to be queued for delivery the next time they come online
07:28 < Sveta> lots of people use a computer without being connected to the internet continuously i think
07:28 < jack_rip_vim> Sveta: how about we create a distro from zero, and the distro just for developer with nice feedback system
07:28 < jim> what would happen to the comments they add? or, are they like taking notes?
07:28 < Sveta> jack_rip_vim: i think the feedback would work for any audience, so if the distro is developer oriented then the feedback could be used for improving developer oriented documentation too
07:29 < Sveta> jim: they'd be delivered to a bug tracker where they would say what help page or what search query they correspond to
07:30 < neveroptimized> You could even ask people to categorize their feedback, or even have different builds of the distro that automatically tagged feedback entries with relevant information
07:30 < jack_rip_vim> Sveta: the distro we make just want to sync all the open source program's developing time line, keep them move together
07:31 < Sveta> jack_rip_vim: that's what any distribution does i think, it includes opensource programs. arch is a bit more inclusive in AUR, but other distributions are more restrictive
07:34 < jack_rip_vim> Sveta: the current distro just use the current stable version for their distro, but if we can have a distro that follow with the source's change. like if some part of the kernel change, the program use this part of kernel also changes, and user can change at same day.
07:37 < jack_rip_vim> it may cause the system quite unstable, but all the developer can keep their developing at same time line
07:38 < jack_rip_vim> the developer no needs to wait the stable version coming out to develop their own
07:41 < Sveta> jack_rip_vim: a lot of AUR follows the latest changes in git of the upstream
07:41 < Sveta> jack_rip_vim: also I'm not familiar with LFS enough to know whether it does the same thing or not
07:42 < Sveta> jack_rip_vim: please also look at exherbo, it is meant to decentralise development, but I can't tell from memory whether they include only releases from upstream or they follow the git versions (I only heard of it tangentially)
07:43 < jack_rip_vim> Sveta: yeah, but that isn
07:43 < jack_rip_vim> 't
07:43 < jack_rip_vim> what I mean
07:44 < Sveta> jack_rip_vim: downloading directly from upstream's git is not what you're after?
07:44 < jack_rip_vim> Sveta: I mean we need a distro that keep all the source developing at same time line
07:45 < jack_rip_vim> A sync developing
07:45 < Sveta> not via git?
07:45 < Sveta> or you are suggesting that the distro users can develop things and easily publish them without being obligated to wait for the upstream to approve their changes?
07:46 < jack_rip_vim> Sveta: yes
07:46 < Sveta> jack_rip_vim: I think that's a part of the exherbo philosophy
07:48 < jack_rip_vim> Sveta: but it is still using some part of old packages, and the distro didn't have everyday source's change
07:58 < jack_rip_vim> Hi GG_Diglett,
08:35 < shhr> Hello :)
08:38 < shhr> I'm testing color schemes in centerim5. Could someone says something (anything will do :)
08:38 < [R]> anything
08:38 < shhr> ok, thanks :)
08:44 < sauvin> anything else
08:47 < shhr> That's good :), I'm done.
08:48 < Sveta> "centerim5" didn't start to make sense until I read "centerIM" somewhere
08:48 < promach_> For https://paste.ubuntu.com/p/52b3gqrvPc/ , why "vbx_lib_common.h" cannot be found ?
08:50 < pingfloyd> my guess is because you didn't install it
08:51 < Sveta> https://packages.debian.org/search?searchon=contents&keywords=vbx_lib_common.h&mode=exactfilename&suite=stable&arch=any does not appear to be enlightening, does it?
08:52 < promach_> pingfloyd : what ?
08:53 < Sveta> promach_: I think someone is being snarky and is suggesting to install you a new package, whose name you'd have to find out
08:53 < [R]> promach_: curious, what if you comopile with -O0?
08:59 < stevendale> Hey Sveta :)
09:02 < promach_> [R]: no more error when I compiled with -o0
09:02 < promach_> Why ?
09:02 < [R]> compiling with optimizations can sometimes make debugging difficult/impossible
09:03 < [R]> of course its a catch 22 if turning on the optimizations causes it to break
09:03 < promach_> what catch 22 ?
09:03 < [R]> that its fine without optimizations...
09:03 < [R]> but with the optimziations, its broken, but you cant' debug it to figure out why
09:04 < promach_> could I say that there is a bug within my native arm gcc compiler ?
09:04 < [R]> maybe
09:04 < stevendale> It could also be out of date promach_
09:05 < promach_> I see
09:05 < stevendale> Thx for helping [R]
09:06 < promach_> [R]: wait
09:06 < promach_> I just found out by running the output binary still give the same error, sorry
09:07 < promach_> :(
09:07 < [R]> ?
09:08 < rx7a> Hey all!... Which FTP_package do you suggest for a personal usage!
09:08 < rx7a> ?
09:09 < [R]> none
09:09 < [R]> because ftp was taken out back adn shot years ago
09:10 < geirha> http://mywiki.wooledge.org/FtpMustDie
09:10 < well_laid_lawn> there are better alternatives
09:11 < Dagmar> scp or gtfo
09:12 < well_laid_lawn> rsync works too
09:13 < rx7a> Hmmmm!
09:13 < jim> rx7a, I liked ncftp... having said that, I've had bad experiences with authenticated ftp, and I prefer ssh-based file copying
09:14 < pingfloyd> if you want it to be like ftp, why not use sftp instead?
09:14 < hellyeah> anyone knows about corrupted stack and detected inside scheduler
09:15 < jim> rsync can use ssh-based file copying too, so maybe a good idea to learn rsync
09:16 < promach_> [R]: I only "make" just now, forgotten to run "./quad_mxp"
09:16 < [R]> promach_: what?
09:16 < promach_> huh ?
09:16 < promach_> it means compiling with -o0 does not help solve the problem
09:17 < pingfloyd> I think he is saying he never ran make to begin with
09:17 < promach_> pingfloyd: I only compiled, forgotten to run
09:17 < promach_> but then the problem is still there
09:17 < promach_> file is still not found
09:18 < pingfloyd> it
09:18 < pingfloyd> it's not -o0 it is -O0
09:20 < promach_> $(CC) -O0 -g $(MXP_IDIR) $(MXP_FLAGS) quad_byte_mxp.c -o quad_mxp $(LDIR) $(LIBS)
09:21 < stevendale> Hey
09:22 < stevendale> Hey
09:28 < greenit> hi, is there a kernel-setting i can set so that the touchscreen of my HP Envy x360 15-bq102ng works? I have gentoo with the 4.16.9 kernel installed
09:44 < post-factum> greenit: xinput -list
09:45 < greenit> post-factum: i'll exec this command once i'm at home (at work right now). just wanted to know if there's a known kernel setting for it ;)
09:48 < post-factum> greenit: i'm not sure it is about kernel setting. you should check things step by step
09:48 < greenit> ok, i'll do. meanwhile i found this bug-thread and probably will try the workaround https://bugzilla.kernel.org/show_bug.cgi?id=198715#c14
09:49 < post-factum> greenit: also, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1752437
09:49 < post-factum> ok
09:57 < p3rL> i need help how to echo if p is not empty only !!! while read p; do echo $p; done
09:59 < hellyeah> what is the type of p
09:59 < p3rL> proccess
09:59 < p3rL> pids
10:00 < hellyeah> you try to check spesificpid exist or not right?
10:01 < p3rL> my cmd will check the process name and get the pids if they get pids it will echo if not empty.
10:01 < KnightsOfNi> How do you copy a file locally when you're logged in on a server via ssh?
10:02 < KnightsOfNi> when I use scp somefile /home/frank/Desktop - it says that the directory does not exist, it tries to put that on the server, not on the local machine
10:04 < vlt> KnightsOfNi: cp
10:05 < KnightsOfNi> How do you tell it to copy it locally? I am logged in on the server
10:05 < vlt> KnightsOfNi: To me "locally" means from some place on the server to some place on the server.
10:06 < KnightsOfNi> Yeah well I mean from the server to my computer where I log in from
10:06 < SoItBegins> KnightsOfNi: To copy from a server to your local machine, you need to log out and use SCP. (or use SCP from within the server).
10:06 < SoItBegins> Try:
10:06 < cluelessperson> could someone help me determine why some things can open content located on a CIFS mount, but other things cannot?
10:06 < cluelessperson> it prevents VLC from playing off the network storage. :/
10:06 < KnightsOfNi> I don't want to logout, that's the point
10:06 < KnightsOfNi> I wanna stay on ssh session and download the file
10:07 < SoItBegins> You can’t do that. That’s not how ssh works.
10:07 < vlt> KnightsOfNi: Open a terminal on your local computer and run `scp server:file localfile`.
10:07 < SoItBegins> scp  user[at]server.name:path  localpath
10:07 < SoItBegins> Yeah. What vlt said.
10:07 < KnightsOfNi> There is no way to do that?!
10:07 < vlt> KnightsOfNi: I just told you.
10:07 < SoItBegins> There is a way. It involves using scp.
10:07 < SoItBegins> There is no way to do that with ssh.
10:07 < KnightsOfNi> No way to do it with SSH
10:08 < SoItBegins> That is correct.
10:08 < vlt> KnightsOfNi: There is a way.
10:08 < vlt> KnightsOfNi: You could use ssh's portforwarding and then `tar | nc ...`
10:08 < luke-jr> you can actually
10:08 < SoItBegins> OK, true. But scp will be a lot less hassle.
10:09 < luke-jr> OpenSSH has a multiplexing thing
10:09 < luke-jr> but you'd have to set it up before your initial connection
10:09 < KnightsOfNi> How then luke-jr ?
10:09 < luke-jr> another possible option is to use something like the Z-Modem protocol
10:09 < luke-jr> for the multiplexing, see the Control{Master,Path,Persist} SSH otpions
10:09  * vlt mumbles something something butterflies
10:10 < lukey> KnightsOfNi: https://serverfault.com/questions/228545/how-to-suspend-active-ssh-connection
10:10 < stevendale> Heya
10:12 < lesshaste> is it possible to get nautilus to start from the directory you launch it in?
10:12 < lesshaste> when I type nautilus from /user/foo/tmp it always starts from the home directory
10:12 < lesshaste> which is annoying
10:13 < lukey> lesshaste: https://askubuntu.com/questions/226755/how-to-open-nautilus-at-current-command-line-directory
10:16 < phre4k> does Linus still use Fedora?
10:18 < k_sze[work]> I have an open lvm2 snapshot for my rootfs. How do I remove it?
10:18 < k_sze[work]> If I just try `sudo lvremove my_root_group/my_root_snapshot`, I get this message: Logical volume my_root_group/my_root_snapshot in use.
10:21 < cluelessperson> KnightsOfNi: hey there
10:21 < cluelessperson> KnightsOfNi: there is a way actually.
10:21 < cluelessperson> KnightsOfNi: can you repeat what you're tyring to do?
10:27 < iflema> k_sze[work]: lvchange -an /PATH/TO/LV   and   vgchange -an VG    no good?
10:30 < pingfloyd> k_sze[work]: did you unmount the LV's filesystem?
10:31 < Ulrar> Hey, quick question. Somehow the permissions of /lib/x86_64-linux-gnu/* got messed up, which prevents me from running chmod
10:31 < k_sze[work]> pingfloyd: you mean the source LV's filesystem?
10:31 < Ulrar> how can I work around that ?
10:31 < Ulrar> any way to tell chmod to use the libs from somewhere else ?
10:31 < pingfloyd> k_sze[work]: the snapshot LV's
10:32 < Ulrar> Even busybox won't run :(
10:33 < lukey> Ulrar: restore these files from you latest Backup
10:33 < Ulrar> I can't
10:33 < Ulrar> I can't bash, I can't scp, I can't rsync
10:33 < lukey> Via recue cd
10:34 < Ulrar> If I close the shell I have now, I have no way to get back on the server
10:34 < Sveta> put the hdd into a box, restore while being booted into another OS
10:34 < Ulrar> Well yeah, but that means taking it offline
10:34 < Sveta> well
10:34 < Sveta> on a vps you can boot into rescue mode
10:34 < Sveta> this has networking
10:34 < pingfloyd> Ulrar: level, that "somehow" was you huh
10:34 < Ulrar> I can boot in rescue, I'm trying to find a way to not take it offline
10:34 < Ulrar> pingfloyd: nope
10:34 < Ulrar> I'm assuming the client did something, but I don't know what yet
10:34 < k_sze[work]> pingfloyd: I don't see the snapshot LV in the output of `mount`. Is that ok?
10:34 < pingfloyd> oh, a client is involved
10:35 < Ulrar> Yeah, unfortunatly they do have root access, for some reason
10:35 < Ulrar> I'm guessing chmod 750 -R / was run
10:35 < pingfloyd> dun dun duun!
10:36 < pingfloyd> Ulrar: it may be quicker to boot off from an alternate environment and fix it
10:36 < pingfloyd> e.g., liveusb or something
10:37 < Ulrar> Yeah but the sites will be offline, I'm trying to find an alternative way
10:37 < Ulrar> Trying to use LD_LIBRARY_PATH now, but doesn't seem to be working
10:37 < Sveta> tell them that they did something stupid that requires their websites to be offline unfortunately
10:37 < pingfloyd> can you compile a static chmod in your /home?
10:37 < Ulrar> can't run gcc no
10:37 < shrdlu68> Ulrar: Can't run anything, even as root?
10:37 < pingfloyd> what about compile on a seperate host
10:37 < Ulrar> I only get permission denied
10:38 < pingfloyd> and then copy the executable over
10:38 < Ulrar> pingfloyd: I can't run scp
10:38 < Ulrar> or rsync
10:38 < Ulrar> or wget
10:38 < Ulrar> Everything gives permission denied
10:38 < mAniAk-_-> maybe with /dev/tcp ?
10:38 < shrdlu68> Ulrar: Perhaps it's selinux? How do you know it's DAC permissions being denied?
10:38 < Ulrar> I whish I had a busybox-static on there
10:38 < pingfloyd> I like Sveta's idea better
10:38 < Ulrar> No SELinux on this
10:38 < pingfloyd> it's the pragmatic one and holds them accountable as a bonus
10:38 < lukey> Ulrar: You mght find more infos here: http://www.lug.wsu.edu/node/414 its the only guide in your specific situation
10:39 < pingfloyd> sometimes clients have to eat a little shit to start understanding ramifications of their actions
10:39 < Ulrar> Sure, but if I can avoid it
10:39 < Ulrar> using /dev/tcp might work
10:40 < Ulrar> if I can remember how
10:40 < k_sze[work]> And sometimes sysadmins have to eat a little shit to understand that users can't be trusted.
10:40 < pingfloyd> sounds like it is more trouble than it is worth to avoid (impractical)
10:40 < pingfloyd> but if you get paid by the hour...
10:40 < pingfloyd> it's more worth the trouble
10:40 < k_sze[work]> Is there anything like macOS's system integrity protection in the Linux world?
10:41 < k_sze[work]> I think macOS's SIP would have prevented something catastrophic like this?
10:41 < pingfloyd> the irony is management understands that logic, not so much the logic of computers
10:41 < lukey> Ulrar: you could also try scp'ing a static busybox executable tu it https://busybox.net/downloads/binaries/1.28.1-defconfig-multiarch/
10:41 < Ulrar> lukey: I can't scp
10:41 < Sveta> lukey: that's a great story, thanks for sharing :3
10:41 < Ulrar> scp needs to work on both the source and the destination
10:42 < shrdlu68> Well, if you can't do _anything_...
10:42 < lukey> Ulrar: have you tried it?
10:42 < Ulrar> well /dev/tcp should work I assume, I'm trying to look that up
10:42 < Ulrar> lukey: yes
10:43 < Ulrar> Altough ftp might work
10:45 < milp_3> hi, anyone got recommendations for legal torrents that could use seeding?
10:45 < Ulrar> It does work, but busybox segfaults
10:45 < Ulrar> almost there
10:46 < lukey> milp_3: *distro isos, blender, gimp ..
10:46 < shrdlu68> milp_3: Distro installation media.
10:46 < Ulrar> Not sure why it sefaults
10:46 < Ulrar> that's strange
10:47 < lukey> Ulrar: cat /dev/kmsg
10:49 < Ulrar> Well I can't cat
10:50 < shrdlu68> milp_3: Wikipedia dumps?
10:50 < Ulrar> I don't understand why a static binary doesn't work there
10:50 < Ulrar> bash-static works fine
10:50 < pingfloyd> milp_3: any free software is a great candidate for that
10:50 < Ulrar> why not busybox
10:51 < pingfloyd> milp_3: could just pick your favorites to seed
10:51 < milp_3> pingfloyd: i already did that, ubuntu, raspian, lxle
10:51 < milp_3> shrdlu68: people download that stuff?
10:51 < pingfloyd> add debian?
10:51 < milp_3> pingfloyd: yeah thats an idea
10:52 < shrdlu68> milp_3: I'll be very well prepared post-apocalypse with my offline Wiki.
10:52 < lukey> Ulrar: try a different arch
10:52 < shrdlu68> milp_3: Gentoo.
10:53 < milp_3> shrdlu68: where do you even get wikipedia dump torrents, and how often do you need to replace them when they become old?
10:53 < Ulrar> lukey: Not sure what arch, it is an x86_64
10:54 < FreakingOut1987> what is the difference between opt (all) vs opt (--) in `iptables -L`?
10:54 < lukey> Ulrar: and different versions
10:54 < Ulrar> i486 doesn't work either
10:54 < Ulrar> lukey: yeah I did
10:54 < FreakingOut1987> all options?
10:54 < Ulrar> I'll try the oldest
10:54 < Ulrar> just in case
10:54 < shrdlu68> milp_3: https://meta.wikimedia.org/wiki/Data_dump_torrents
10:55 < FreakingOut1987> nvmd, i'm reading it incorrectly
10:55 < Ulrar> not better
10:55 < Ulrar> they all segfault, that's strange
10:55 < lukey> dun dun dun
10:56 < shrdlu68> Ulrar: The challenge is getting statically-linked binaries onto the system, right?
10:56 < Ulrar> I wonder what busybox is doing for that to happen
10:56 < Ulrar> shrdlu68: Not really, I figured that out : FTP works
10:56 < Ulrar> the challenge is getting statically linked binaries to not segfault
10:56 < Ulrar> bash-static works
10:56 < Ulrar> busybox doesn't
10:56 < shrdlu68> Ah, problem solved.
10:56 < Ulrar> not solved at all
10:56 < revel> Ulrar: What does `file busybox` say?
10:57 < Ulrar> busybox: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), statically linked, stripped
10:57 < Ulrar> ran that on my system, not the server of course
10:57 < Ulrar> since file won't run either on there :)
10:57 < revel> And it's an x86(_64)? server, right?
10:57 < sauvin> +OK G0/iK//PmKr/rPJCP.UBWIo0GX7Ul0C6z9q.PDsGk.7OM6i/Q19N5/f0olZ/hciZ91L4sqy1W8lGX090J4i0Q8IXA/L7Jo6.MpZ7m.70eWh/XGoF71ggsG//3flOD.37H01/adEEL0Idccn1
10:57 < Ulrar> I tried i686, x86-64 and i486
10:57 < Ulrar> they all segfault
10:57 < lukey> Ulrar: take the rescue-cd way of restoring from the latest backup and after that put HA in place
10:57 < Ulrar> yes, it's a 64
10:58 < revel> Are you completely sure?
10:58 < Ulrar> lukey: No HA there, again, I have no control over what the client will or won't accept unfortunatly
10:58 < Ulrar> revel: yes
10:58 < revel> How?
10:58 < Ulrar> That's the only thing we have
10:58 < pingfloyd> you sure it isn't IA-64?
10:58 < lukey> Ulrar: could as well be PPC or Arm
10:58 < Ulrar> No, it couldn't
10:59 < revel> i386 binaries should work on ia64 though, I think.
10:59 < revel> Ulrar: What version of glibc?
10:59 < Ulrar> Intel(R) Xeon(R) CPU E5-1630 v4 @ 3.70GHz
10:59 < pingfloyd> revel: I don't think so
10:59 < Ulrar> it's a regular debian 8
10:59 < revel> ¯\_(ツ)_/¯
10:59 < pingfloyd> amd64 yes
11:00 < revel> Ulrar: Maybe the glibc's too old. Binaries statically linked against a newer glibc may apparently not like working in an env with an older one.
11:00 < revel> Or something.
11:00 < lukey> But debian 8 is quite recent
11:01 < revel> Well, it's oldstable now.
11:01 < Ulrar> I have an idea
11:01 < Ulrar> I'll get the .deb
11:02 < Ulrar> Meh, not better
11:02 < revel> 2.19, I think.
11:03 < Ulrar> Okay well
11:03 < Ulrar> I did my best
11:03 < Ulrar> Let's just reboot in rescue
11:03 < revel> Stretch/stable has 2.24. Though I've been wrong about glibc static binaries before.
11:03 < Ulrar> thanks anyway
11:07 < milp_2> are there any other things i could do to support the opensource community with my upstream bandwidth and storage space? other than seeding torrents?
11:08 < revel> Gimme.
11:09 < lukey> milp_2: Provide a mirror
11:09 < milp_2> lukey: how do i go about that? (i know how to set up a file/webserver)
11:11 < pingfloyd> Ulrar: have you tried turning it off and on again?
11:11 < lukey> milp_2: https://www.debian.org/mirror/ftpmirror for example
11:12 < revel> Find (a) project(s) you particularly like first and then contact them, I think.
11:12 < Ulrar> pingfloyd: With wrong permissions in lib ? That sounds like a bad idea :D
11:12 < Ulrar> pretty sure that wouldn't boot
11:12 < milp_3> thanks
11:12 < revel> Some may have higher bandwidth and storage requirements than others.
11:13 < revel> i.e with Gentoo, that's terabytes upon terabytes of distfiles, I think.
11:13 < pingfloyd> just tell them it lost power on its own
11:13 < |MLT|> anyone know of any decent infosec-related channels on this network?
11:14 < lukey> milp_2: "<distro/project> mirror howto" + google is the way to go
11:14 < rendar> i have my raspberry always on, connected on the lan network. to the raspberry is connected an external usb drive which power off after N time of idle.
11:14 < mnemon> |MLT|: ##security
11:14 < |MLT|> thanks
11:14 < rendar> Now, my problem is, when i turn on my desktop pc, that hd connected to the raspberry turns on
11:14 < rendar> why?!
11:14 < rendar> there is a nfs server on the raspberry
11:14 < rendar> but the desktop doesn't mount that drive automatically
11:17 < djph> rendar: you plugged the rpi's hdd into a pc-controlled power source (e.g. one of the "slave(tm)" ports on your UPS)
11:17 < rendar> i don't have an ups
11:18 < rendar> i think this could be more a nfs server thing
11:18 < jhasse> Hi! How can I let `stat` print out the mtime in the format that `touch -t` accepts?
11:18 < sauvin> I think I'm going to blow a weekend or three learning to operate irssi or weechat or something under the hood. Sick to death of hexchat crashing.
11:18 < stevendale> Hey sauvin :)
11:18 < stevendale> What distro
11:19 < stevendale> I am on Debian Unstable and HexChat 2.14.1 doesn't crash for me owo
11:19 < djph> that's not what "unstable" means in the Debian world.
11:20 < TingPing> sauvin, how do you still have all these problems, i don't get it
11:20 < revel> I don't think he was implying it did.
11:20 < revel> He was just saying that he was running Debian Buster adn that Hexchat wasn't crashing for hm.
11:21 < revel> I think I some letters there.
11:21 < stevendale> Sid actually revel, not buster
11:21 < sauvin> Looks like a GTK problem. Problem is, ONLY hexchat has these problems, ever. I use GTK stuff on KDE all the time without problems.
11:21 < revel> Argh. Debian and their naming conventions.
11:21 < stevendale> You could reinstall gtk?
11:21 < revel> sauvin: What version of Hexchat?
11:22 < revel> Might've been fixed in 2.14.1
11:22 < sauvin> 2.10.2-1ubuntu3 , on Ubuntu 16.04.
11:22 < revel> Sounds old.
11:23 < revel> Or maybe they were just more active with new releases 2 years ago, who knows.
11:23 < |MLT|> hexchat has current overflow from somethin i stumbled across
11:23 < TingPing> ofc, everything on 16.04 is ~3 years old
11:23 < sauvin> It probably is. I've got a flatpak of hexchat and a snap of hexchat installed, and can't use either of them yet until I figure out how to get them access to my scripts.
11:23 < revel> |MLT|: Yeah, 2.14.1 was a couple of months ago, I think.
11:23 < TingPing> sauvin, cp yourscripts ~/.var/app/io.github.Hexchat/config/hexchat/addons
11:23 < sauvin> Tried that. It dumped core.
11:23 < TingPing> well give me a backtrace
11:24 < |MLT|> (((((((((( hexchat[30394]: segfault at 7f0ca6862199 ip 00007f0ca67a5993 sp 00007ffd83fab310 error 4 in libpcre.so.3.13.3[7f0ca678b000+70000] ))))))))))
11:24 < TingPing> |MLT|, what version?
11:24 < sauvin> How to do that?
11:24 < sauvin> Oh... wait... I remember reading the error message, it said something about being unable to find File::Spec or something.
11:25 < TingPing> ugh
11:25 < revel> Probably needed Perl or something then.
11:25 < |MLT|> im unsure on version
11:25 < TingPing> installing perl modules will probably be a pain
11:25 < |MLT|> was just watching it being used to crash people's clients
11:25 < sauvin> Yeah. It needs ALL the perl stuff, and it ain't in the snap.
11:25 < |MLT|> in another network
11:25 < TingPing> the snap is crap, don't even consider it
11:25 < sauvin> You tried telling me a few months ago to use the snap.
11:26 < TingPing> i just wanted to give you an option because you've been bitching about this for 3 years
11:26 < revel> lol
11:26 < TingPing> never found out anything useful in that entire time
11:26 < sauvin> In fact, a snap or a flatpak is the *only* way I can use hexchat at the moment. The "machine-native" one dumps as soon as I move the mouse.
11:26 < TingPing> you refused to add the PPA maybe?
11:26 < TingPing> oh welll the PPA is dead now anyway
11:26 < sauvin> I've had rotten luck with PPA.
11:27 < TingPing> that is a meaningless comment
11:27 < TingPing> repos are repos, their contents are what matters
11:27 < sauvin> No, it isn't. People report variable success with PPA.
11:27 < TingPing> well yea, because they are distinct repos with unique content
11:27 < TingPing> any can be shit or not shit
11:28 < revel> sauvin: Compile it yerself :D
11:28 < TingPing> anyway. just learn irssi, i'm honestly not going to figure out perl module building in the short term
11:28 < sauvin> I suppose I could fire up a Gentoo VM and use its hexchat :sigh:
11:28 < TingPing> 18.04 has latest hexchat
11:28 < revel> I haven't heard much good of 18.04
11:29 < sauvin> Honestly, yes, I plan on abandoning Hexchat as soon as I can nagivate under the hood with something TUI.
11:29 < sauvin> There's just something about Hexchat that hates GTK.
11:29 < sauvin> (or vice-versa)
11:29 < TingPing> there are literally tens of thousands of users without problems of hexchat on ubuntu
11:29 < revel> Works fine for me.
11:29 < revel> Though I have an actually recent version.
11:30 < TingPing> yea being on an ancient version does make support much harder
11:30 < sauvin> I have a *tendency* to stick with what's in repos, almost never venturing *outside* repos unless there's a driving need for something more recent.
11:30 < TingPing> ubuntus universe repo is just full of outdated and unsupported shit
11:30 < sauvin> This, precisely, is WHY I tend to stick with stable and LTS distros and their "dated" packages. I don't have time to be fooling with this crap.
11:31 < TingPing> but that logic doesn't make any actual sense, hexchat could have fixed this crash 3 years ago
11:31 < sauvin> And yet, and yet, only Hexchat gives me this grief.
11:31 < Zephranoid> If you want up todate repos you could swap to arch.
11:31 < sauvin> Tried Arch. I'm not Arch-friendly.
11:33 < Ulrar> Okay, I think I got it
11:33 < veridiam> i tend to find more desktop stability in bleeding edge distros than LTS ones
11:33 < Ulrar> took me a couple of tries but the server just rebooted
11:33 < Ulrar> and it seems okay
11:33 < Ulrar> There was a lot more than /usr/lib/ that got messed up
11:33 < Ulrar> basically every .so on the system was wrong
11:33 < Ulrar> I really, really wonder wtf they did
11:33 < Zephranoid> that doesn't sound good
11:35 < shrdlu68> find / -type f -name '*.so' -exec ...
11:35 < sauvin> !@#$%!@#$#@
11:36 < sauvin> This is really, really frustrating because I spent a LOT of time writing scripts, too. Gonna take me forever to figure out how to change them over to something else.
11:41 < Furai> I'm looking for some good software that works cross platform for screensharing or pair programming. Actually it has to work on Linux, MacOS and Windows...I've looked into a lot of them. None was working well enough.
11:41 < Furai> Any suggestions?
11:42 < sauvin> What's "pair progrmaming"?
11:43 < Zephranoid> one person codes while the other notes mistakes etc. The idea is the person coding doen't have to worry about errors etc while getting a quick version of the code done
11:44 < shrdlu68> Wherein two persons share an editor buffer.
11:44 < Sveta> do they need two computer mice and two keyboards to work simultaneously?
11:46 < FreeFull> Sometimes it's two laptops and some special software
11:47 < Cyber_Akuma> So it's not like on CSI when two people were on the same keyboard to try to stop a hacker? XD
11:47 < Zephranoid> haha no that doesn't work, but its a great experiment
11:51 < Cyber_Akuma> Zephranoid: Yeah, I know that doesn't work, it's a lot easier to stop the hacker just by unplugging the monitor XD
11:52 < Furai> Pair programming is great when you learn stuff.
11:53 < Furai> And yeah, we all live kind of far apart so I'm looking for something working on-line.
11:54 < Furai> Tried jitsi, their solution works well but the quality isn't the best. Maybe their website is limited, so I tried self-host it. Couldn't make it work and I couldn't tackle the issue. No idea if self-hosted would actually improve quality in any way
11:55 < shrdlu68> mmaped source files with a REST API to synchronize buffers? Sounds like a fun project.
11:57 < APic> *nod*
12:11 < Furai> Btw, if it was up to me I'd probably just use tmux + vim for pair programming + some voice chat.
12:12 < NGC3982> tmux is nice.
12:21 < ananke> Furai: screen sharing is an overkill for that purpose. there are plenty of collaborative web-based editors out there
12:22 < Furai> With edgy privacy.
12:22 < ananke> Furai: you can host them yourself if you so desire
12:22 < Furai> Haven't seen any self-hosted.
12:22 < Furai> Are there any?
12:23 < ananke> Furai: etherpad is one
12:23 < Furai> Uh, isn't it just single file? No disk sync and so on?
12:24 < Furai> But yeah, I haven't considered it at all. I'll have a look at those.
12:24 < ananke> not sure what 'disk sync' would even mean in this case, since it's editing a file in a central location
12:25 < Furai> I meant by that loading project from your HDD.
12:29 < Furai> Anyway, thanks. :)
12:39 < gr8> hi guys, I have a funny question: Do you know any linux distributions that have the program 'xsecurelock' in their official repository?
12:40 < gr8> asking because this would give an indication about the philosophy of that distro
12:40 < revel> What package is it generally bundled with?
12:41 < veridiam> it's on AUR (not official)
12:42 < epicmetal> gr8: ALT Linux Sisyphus, apparently
12:42 < epicmetal> gr8: https://pkgs.org/download/xsecurelock
12:43 < revel> Well, Debian and Gentoo don't seem to have it, at the very least, for all that that's worth.
12:43 < epicmetal> gr8: also https://repology.org/metapackage/xsecurelock/versions
12:43 < epicmetal> gr8: you have the choice of one distro :)
12:44 < pingfloyd> what does it indicate about their philosophy?
12:44 < revel> ^
12:44 < pingfloyd> Enquiring minds want to know
12:44 < epicmetal> "Sisyphus is not intended for end users and its usage may result in unpredictable behaviour." -- http://www.sisyphus.ru/en/
12:44 < veridiam> it's google authored?
12:45 < revel> Curious about that as well. It seems to be a Google thing, and they aren't usually too enthusiastic about developing things for more than a year or two, so it'd eventually just get deprecated anyway.
12:46 < pingfloyd> or is it more about how the dist prioritizes security?
12:46 < pingfloyd> security first, freedom second, or vice versa
12:48 < gr8> hmm, I've never heard of Alt Linux. xsecurelock tries to handle a fundamental security issue of X11, which should be a priority for security-aware desktop distributions
12:49 < pingfloyd> I'm fricken Sherlock Holmes huh?
12:49 < gr8> no you are Watson lmao :P
12:49 < pingfloyd> though even more secure option would be FDE combined with hibernation instead of screen lock
12:50 < epicmetal> pingfloyd: sometimes you want the network to stay up
12:50 < pingfloyd> should have it setup to fire a shotgun at the user after too many failed attempts too
12:50 < gr8> sometimes it is necessary to have a PC running while you are leaving the desk, i.e. for networking applications.
12:50 < epicmetal> s/sometimes/often/
12:51 < gr8> hehe you guys are great ^^ you already know what I am going to say :D
12:51 < epicmetal> gr8: you could always package it and be the hero that Gotham deserves
12:51 < veridiam> when it comes to physical access, locking the screen seems like more of a formality
12:51 < oiaohm> epicmetal: http://blog.martin-graesslin.com/blog/2015/01/why-screen-lockers-on-x11-cannot-be-secure/  somehow I wonder if it manages to address the problem or is just xsecurelock something with fancy name and just as flawed as everything else.
12:51 < epicmetal> oiaohm: it can't address all the problems
12:51 < epicmetal> oiaohm: https://github.com/google/xsecurelock#security-design
12:52 < epicmetal> I wonder if Wayland is any better at all this
12:52 < pingfloyd> will its makefile make a deb pkg for you (has a target)?
12:52 < gr8> epicmetal, right, unfortunately I don't have time for stuff like that. I need something that works and is sufficiently secure
12:52 < epicmetal> Not that Wayland is usable (for me) yet
12:52 < pingfloyd> some authors do that
12:52 < epicmetal> gr8: maybe look into Wayland
12:52 < epicmetal> gr8: probably Fedora/GNOME
12:53 < epicmetal> gr8: X11 is kind of a lost cause
12:53 < pingfloyd> X11 in general has some fundamental security issues
12:53 < gr8> unfortunately
12:53 < epicmetal> Still not sure why Linux didn't just switch to framebuffer graphics
12:53 < gr8> I really need a security focused desktop linux distro
12:54 < gr8> well, want
12:54 < dgurney> qubes?
12:54 < pingfloyd> X11 can't even do vsync without hackery ffs
12:54 < dgurney> can't think of many more secure distros
12:54 < pingfloyd> some drivers try to hack around the issue to varying degrees of success and failure.
12:55 < gr8> right, cubes ... I tried it, but I think they told me that it can't solve the inherent issues of X11. They don't have a separated X11 layer / VM
12:55 < pingfloyd> wayland was sort of born with the idea of addressing fundamental flaws like that
12:55 < epicmetal> gr8: it depends on your hardware, I think
12:55 < pingfloyd> iirc, lack of any proper vsync handling was a catalyst for the wayland project
12:55 < dgurney> you can't have everything gr8
12:56 < pingfloyd> sometimes you just have to be a bad ass like me and be a risk taker
12:57 < gr8> damn, please remind me to set up an IRC bouncer ;) I'll be back in an hour
12:57 < pingfloyd> running linux already a big step away from low hanging fruit territory
12:58 < epicmetal> I kinda miss Windows
12:58 < pingfloyd> I still have windows in a VM
12:58 < epicmetal> I have it on a VM at work that I RDP to
12:58 < revel> I did. Then I used it again. Then I stopped missing it.
12:58 < epicmetal> But not at home
12:58 < pingfloyd> I don't really do much with it though
12:58 < pingfloyd> I mainly keep it around in case I need to provide support
12:59 < epicmetal> revel: you're right, I'd hate it in a month
12:59 < pingfloyd> I just really don't enjoy being in its environment.  Macos is more hell though.
13:00 < revel> It was a bit less for me...
13:00 < oiaohm> epicmetal: qingy a long time ago show there was methods to fix the lock screen problem.   Of course for it to be really functional kms need to be properly supported.
13:00 < pingfloyd> they both feel like you're forced to wear training wheels
13:00 < oiaohm> epicmetal: note qingy had the lockscreen done by framebuffer not X11.
13:01 < oiaohm> epicmetal: of course using the framebuffer that way was a great way to have Nvidia and closed source ATI drivers at the time explode badly.
13:01 < seven-eleven> hi
13:01 < epicmetal> I should try a minimal Wayland compositor this weekend
13:01 < pingfloyd> that's nvidia and ati's problem
13:01 < seven-eleven> is /tmp == tmpfs?
13:02 < epicmetal> oiaohm: never used qingy
13:02 < oiaohm> pingfloyd: things have improved since then.
13:02 < epicmetal> seven-eleven: /tmp is a directory, tmpfs is a virtual filesystem
13:02 < oiaohm> epicmetal: that lock screen stuff was only proto type stuff becuase it had major isuses.
13:03 < epicmetal> Isn't AMD good now
13:03 < pingfloyd> ATI has gotten better about it, but nvidia should quit making their users have to endure their proprietary drivers and support foss already.
13:03 < epicmetal> s/ATI/AMD/g
13:03 < seven-eleven> epicmetal, so stuff inside /tmp is stored on HDD not in RAM?
13:03 < pingfloyd> there really is no excuse for them other than being greedy and controlling assholes
13:03 < seven-eleven> nor in SWAP
13:03 < oiaohm> epicmetal: AMD has improved a lot.   Still some areas that are not finshed yet.
13:03 < epicmetal> seven-eleven: if tmpfs is mounted on /tmp, then stuff inside /tmp is stored in RAM
13:04 < seven-eleven> epicmetal, ahhh i see, its not mounted to /tmp its mounted to /dev/shm
13:04 < widp> I am trying to add a pythons script using crontab
13:04 < epicmetal> seven-eleven: if /tmp is not a mountpoint, then stuff inside /tmp is stored wherever stuff in / is stored
13:04 < seven-eleven> epicmetal, thanks!
13:04 < widp> I run crontab -e but I don't get any editor at all
13:04 < pingfloyd> AMD pretty much absorbed ATI
13:04 < epicmetal> seven-eleven: no worries
13:04 < oiaohm> epicmetal: wayland dependance on raw kms with drm really does provide a possible path out.   So that something can claim the screen to lock it and not leave a path for something else to bipass it.
13:04  * epicmetal makes slurp noise
13:04 < widp> ohh, nvm, it was open in an editor window which I didn't notice
13:04 < epicmetal> I miss the old days
13:04 < veridiam> nvidia proprietary is still the more performant, unfortunately
13:05 < epicmetal> No job, 486es, and good games
13:05 < pingfloyd> veridiam: yeah, because nvidia imposes that
13:05  * epicmetal shakes fist at life
13:05 < pingfloyd> if you bought their hardware, you shouldn't be expect to be put through that.
13:05 < pingfloyd> expected
13:06 < epicmetal> i hear VFIO is the way to do it
13:06 < epicmetal> (game on linux)
13:08 < gr8> re
13:08 < gr8> did you find any other interesting X11 security solutions / distros? Is someone here using Wayland?
13:10 < oiaohm> gr8: security wise X11 is basically not fixable.
13:10 < oiaohm> gr8: its the reason why wayland started.
13:10 < PlasmaHH> Hi, I have a host that does show empty conntrack tables despite nf_conntrack_ipv4 et al being loaded, any idea what might be missing?
13:11 < epicmetal> gr8: not sure how much of this is still relevant https://lwn.net/Articles/589147/
13:11 < pingfloyd> fedora has pretty progressive support of wayland btw
13:12 < oiaohm> pingfloyd: all the security faults in that 2014 article about X11 are just as true today as back then.
13:12 < pingfloyd> I know
13:13 < pingfloyd> I'm implying that gr8 may want to consider that in his dist policy equation
13:13 < oiaohm> gr8: problem is with X11 is that is broken and when you attempt to fix lot of existing X11 applications bust.   So the only real valid way forwards is the xwayland for legacy applications and direct wayland ones for those you need secure.
13:14 < gr8> thanks for the hint, I should take a closer look at Wayland then
13:14 < oiaohm> gr8: problem is getting all the frameworks done so Linux can move off X11 default to wayland default.
13:14 < oiaohm> gr8: that includes getting Nvidia to be cooperative that is kind of a uphill battle.
13:15 < pingfloyd> I think every nvidia on linux user should write a really nasty and profane letter to nvidia.
13:17 < PlasmaHH> as long as there are no alternatives to things like dmx X11 will stay alive
13:17 < phil42> do that if you are ready to switch to nouveau
13:17 < phil42> otherwise play nice with nvidia
14:37 < adrian_1908> When typing `groups alice`, then next line would always start with `alice : `, right? Or is there a case where another name appears there?
14:37 < adrian_1908> (by next line I mean the output of the command)
14:38 < revel> adrian_1908: I think that's just because most users generally have a group with the same name.
14:38 < adrian_1908> revel: ah good point. I'll test that.
14:39 < revel> Yeah, just tested it.
14:39 < revel> An user not in a group by their own name won't have their name as the start of the output. Or something. You get the idea.
14:40 < adrian_1908> Ok, thanks!
14:46 < adrian_1908> Hmm, I tried it myself, and to me it looks like the pattern remains "username : group1 group2 groupN", even if the user doesn't have a group is its own name anymore.
14:49 < revel> There's no colon in the output of my `groups`
14:49 < mnemon> ü
14:49 < stevendale> Hey
14:50 < stevendale> How do I find the size of a folder using cli
14:50 < revel> Anyways, I tested it with guixbuilder1, which is only in the group guixbuild, and it only shows that, for what it's worth.
14:50 < oleo> du -sh /
14:50 < revel> stevendale: du
14:50 < djph> revel: try 'groups revel' (or whatever your username is)
14:50 < oleo> s is summmary h is human readable
14:50 < stevendale> Thanks oleo,  revel :D
14:51 < revel> djph: The point is that the nick isn't always the first thing shown, period.
14:51 < adrian_1908> revel: maybe envrionment/defaults differences. I'm on Ubuntu which comes with its own idea of defaults. In your case it makes sense that the first item would change.
14:51 < oleo> np
14:51 < revel> It may be a different implemenatation of groups.
14:51 < revel> Though mine appears to have Debian as the upstream...
14:52 < adrian_1908> Or that. For me it's "user : usergroup othergroups ...", for your it's "usergroup othergroups" and so my second argument changes, your first.
14:52 < revel> But Ubuntu has NIH syndrome, so.
14:52 < djph> revel: quite.
14:52 < revel> Yeah, there's no "${user}: " in the beginning for me.
14:52 < pingfloyd> a cruder wheel usually
14:53 < pingfloyd> a wheel with training wheels duck taped to it
14:53 < revel> i.e `groups guixbuilder1` => "guixbuild\n"
14:53 < djph> If I run 'groups' I just get the group listing of the current user (i.e. me).  If I run 'groups user', I get "user : $list"
14:53 < stevendale> What does Zenity do?
14:53 < pingfloyd> remember greatest hits like upstart?
14:54 < revel> I don't know what it's even sorted by, but $username isn't the first thing I get with `groups`, it's wheel.
14:54 < pingfloyd> somehow something came along that managed to be worse but get get massive support by dists.
14:54 < adrian_1908> anyway, looks like `groups alice` shouldn't print `bob : alice others` which I think is a typo I stumbled upon on a website.
14:55 < djph> revel: are you running 'groups wheel' or something?
14:55 < revel> No, `groups revel`
14:55 < djph> or is your MAIN GROUP wheel?
14:55 < revel> Nope.
14:55 < revel> wheel uucp audio cdrom video games kvm usb users plugdev portage docker android wireshark vboxusers revel
14:56 < revel> Wait... Should probably get rid of portage. Not that it should make much of a difference with what I've done though.
15:11 < imofftopic> hi
15:12 < Alexander-47u> whats this
15:12 < BluesKaj> Hi folks
15:12 < Alexander-47u> tracker-miner-f and tracker-store eating my cpu
15:12 < Alexander-47u> is it necessary?
15:12 < jim> hi blues
15:13 < Alexander-47u> where is my manners, hello good people :)
15:13 < jim> hi
15:15 < BluesKaj> Hi jim
15:15 < graff> Alexander-47u: it doesn't strike me as part of your core system
15:15 < imofftopic> Hi
15:15 < graff> but it's probably necessary for something you are running
15:16 < imofftopic> What do people think about Automation and jobs?
15:16 < imofftopic> Looks hard to tell what the future has in store...
15:16 < graff> well there may be a job automating things
15:16 < graff> essentially it's the programmers job to make themselves obsolete
15:17 < graff> programmers'
15:17 < imofftopic> graff I guess they get the advantage of being the last ones to become obsolete?
15:17 < graff> imofftopic: heh yeah
15:17 < graff> haha
15:17 < adrian_1908> Will be interesting to see whether tech jobs disappear before social jobs. They might. Afaik office jobs are under particular threat.
15:17 < graff> well people will always want service jobs
15:18 < imofftopic> Yeah but how do the service jobs work if the office people lost there jobs?
15:18 < graff> obviously mcdonalds could be automated, but there are millions of people who depend on having temporary servants at restaurants
15:18 < imofftopic> Whos going to have the money?
15:18 < graff> even fancy restaurants. all about paying to have our time as king
15:18 < adrian_1908> right, and the last job remaining will be the oldest profession, handjobs, blowjobs, etc.
15:18 < imofftopic> adrian_1908: .... you're actually probably correct there :O
15:18 < graff> so all jobs may someday simply be temporary servant service so as to provude the illusion that an advanced society values wealth
15:18 < imofftopic> Can't automate that one
15:19 < imofftopic> And this is going to happen soon
15:19 < imofftopic> These changes are... like within 10 years
15:19 < graff> it may have already happened
15:19 < imofftopic> Yeah true
15:19 < imofftopic> its happening
15:19 < graff> if you think about it companies like monstanto and bayer are older than their interests
15:19 < graff> they are in essence paper work based AI
15:20 < graff> they live and grow with no real need for humans to enjoy or benefit from what they do
15:20 < Alexander-47u> graff, no i read it was some kind of indexing thing of gnome
15:20 < Alexander-47u> but it only uses high cpu at boot time ive noticed now
15:21 < adrian_1908> Right, the argument has been made that many governments are full of low utility jobs, that are basically a "silent" welfare system, which gives people a sense of meaning vs being unemployed.
15:21 < graff> ay program that uses 1 percent CPU or above is suspect
15:21 < graff> in fact most programs should be hardly registering on the system monitor
15:22 < graff> especially event driven daemons, monitors and whatnot
15:22 < Alexander-47u> ye its 0-1% cpu idle now
15:22 < graff> that seems ok yeah
15:22 < imofftopic> Full of silent welfare jobs?
15:22 < imofftopic> So basically just keep things afloat?
15:23 < imofftopic> Spending on jobs just to keep the cycle going?
15:23 < graff> imofftopic: honestly, I am pretty smart. and make very little money
15:23 < graff> and I see all of pretty much everyone as having "welfare jobs"
15:23 < graff> from trump to many programmers
15:23 < graff> to the guy sweeping the street with a hand broom
15:23 < imofftopic> You mean jobs that aren't really needed?
15:23 < adrian_1908> imofftopic: well, the person who passes the paper to the person who reads the paper to the person who stamps or shreds the paper. That kind of thing. Since society isn't ready to admit this, I've heard the arguments are governments are consciously bloated for the greater good.
15:24 < graff> imofftopic: they are not only not needed, they are being done badly
15:24 < messiah_> hey
15:24 < graff> which means actual loss of energy in the form of displacemnt into pollution and corruption
15:24 < imofftopic> So where does this go?
15:24 < messiah_> got a problem in make
15:24 < messiah_> anybody help?
15:24 < messiah_> error: conflicting types for ‘int64_t’  typedef long long int int64_t;
15:25 < oleo> depopulation
15:25 < graff> no one should be defining int64_t other than the underlying libc impl messiah_
15:25 < imofftopic> oleo wait what?
15:25 < adrian_1908> Alexander-47u: those process names would have made me suspicious too btw, just like "rtkit-daemon" :)
15:26 < messiah_> graff, tell me more popular please
15:26 < graff> try getting rid of that definition and instead use #define <stdint.h>
15:26 < graff> sorry
15:26 < messiah_> i can share link with more full error
15:26 < graff> #include <stdint.h>
15:26 < graff> messiah_: it's ok, i already know what is causing it
15:26 < Alexander-47u> adrian_1908, lol no if it was a malware miner, then it wouldnt be named tracker-miner lol
15:26 < graff> someone is trying to redefine int64_t
15:26 < Alexander-47u> its jsut a indexing stuff of gnome, just like what windows does at boot time
15:27 < adrian_1908> Alexander-47u: yeh, good point. Luckily I seem to have been spared so far.
15:27 < graff> and they are going about it fairly stupidly. a long could be 64 bits on some platforms
15:27 < Pentode> oleo, what graff suggested is your best bet.
15:27 < Pentode> im not sure why they did that?
15:27 < Pentode> maybe compatibility for some other platform / compiler i dunno
15:28 < Alexander-47u> adrian_1908, if you only install stuff from repository and have noscript in your browser, with some common sense, you are fairly safe
15:28 < messiah_> graff, tell your opinion about it if you know
15:28 < messiah_> oh
15:28 < messiah_> you already
15:29 < messiah_> i will try with stdint.h
15:29 < imofftopic> I really don't know what the future holds
15:29 < Alexander-47u> hey btw, what would be a good book to learn C with exercises
15:29 < imofftopic> but... change is coming big change and fast
15:29 < graff> Alexander-47u: i used K&R2. but every year you wait it gets harder and harder to interpret what it is saying
15:29 < jim> [[tcpl]]
15:30 < graff> and nearlyeverything is now "worng" and you have to understand how to update the examples
15:30 < messiah_> how to mention somebody?
15:30 < graff> you also need to rename the examples because they will conflict with the namespace of your system's libc impl
15:30 < neachdainn> Hey all. I'm looking into setting up dm-crypt on my machine. Is it possible to encrypt a partition without erasing everything on it?
15:30 < Alexander-47u> does it also have handson exercises that make sense in between paragraphs?:P
15:30 < graff> (trying to reuse names)
15:31 < graff> messiah_: type  gr then hit tab
15:31 < pingfloyd> it becomes an extra exercise
15:31 < graff> gr [and hit tab]
15:31 < jim> put their nick in the message, typilly at the staart with a comma or colon
15:31 < graff> it will autocomplete my nick
15:31 < imofftopic> ... just noticed theres a small "Coca Cola" label in one Music video i've seen so many times. It was hidden. Hit pause then noticed it
15:31 < imofftopic> Interesting
15:32 < shrdlu68> Casually off-topic, aye?
15:32 < messiah_> graff: where i need to change it? in stdint.h?
15:32 < imofftopic> shrdlu68: really love to know where that stuff is actually on topic
15:32 < graff> messiah_: find the place that the typedef is made
15:32 < imofftopic> outside of social media
15:32 < graff> and comment out the line and replace it with #include <stdint.h>
15:33 < graff> then try recompiling
15:33 < graff> depending on how many bad typedefs they made you might have to do this for a number of lines
15:34 < graff> it seems though that you probably just are not activating the proper compiling mode
15:34 < messiah_> graff: tell me?
15:34 < graff> and then the program thinks it is on a system without stdint.h and tries to use its own definitions
15:34 < graff> messiah_: what is the program, linux distro and other details
15:35 < Dan39> Alexander-47u: gnome tracker/miner will also do stuff later after boot, but it should be like nice and ionice +20 so it doesn't interfere with anything else, and actually only attempt to do stuff while system is idle iirc
15:35 < messiah_> kali (same thing in debian vm). its osmocom
15:35 < Dan39> but its famous for being buggy and bogging down the whole system :P
15:36 < jim> who asked about c programming excercise book?
15:36 < adrian_1908> messiah_: do you have a URL to the source, or did you get it via the package manager?
15:37 < Alexander-47u> jim, me
15:37 < Dan39> the gnome people will deny it to their death, though in the changelog for next release have something like "fix tracker bug that caused slow system..."
15:37 < messiah_> graff: here full error log - https://pastebin.com/uSuJGRGM
15:37 < TyrfingMjolnir> Anyone using ZeroTier One?
15:37 < messiah_> adrian_1908: second
15:37 < jim> here;s one: http://www.dipmat.univpm.it/~demeio/public/the_c_programming_language_2.pdf
15:38 < jim> Alexander-47u, ^^
15:38 < TyrfingMjolnir> All according to this video: https://www.youtube.com/watch?v=H0PPgSZUIjE However can not ping
15:38 < Alexander-47u> thanks jim
15:38 < Alexander-47u> :)
15:39 < adrian_1908> messiah_: Maybe this helps. Looks like there are certain requirements for cross compilation to ARM: https://github.com/osmocom/osmocom-bb/tree/master/src
15:39 < messiah_> adrian_1908: git://git.osmocom.org/osmocom-bb
15:39 < jim> Alexander-47u, also find a book about writing state machines, you gonna need to know
15:41 < Alexander-47u> il start with C lol
15:41 < adrian_1908> gotta go, bye all!
15:42 < Alexander-47u> bye
15:42 < messiah_> by the way - its like same distr that adrian gave me
15:42 < messiah_> but i getting another error - fatal error: sys/_types.h: No such file or directory  #include <sys/_types.h>
15:43 < graff> messiah_: your toolchain looks messed up
15:43 < messiah_> graff: there build finished
15:43 < messiah_> :)
15:44 < graff> cool
15:44 < messiah_> graff: there wasnt any mistakes and it was compiled by same instruction to osmocom
15:44 < messiah_> graff: so you think i need rebuild it?
15:45 < graff> messiah_: the problem is that firmware/ has its own include/ directory
15:45 < messiah_> graff: problem is that found osmocom vm image in addition to my kali and there same error
15:45 < graff> and it's also trying to use the /usr/include from your system
15:45 < messiah_> then why its not using its own include?
15:45 < graff> it should not have it tbh
15:46 < graff> the only people that need their include/ of system headers is a libc
15:46 < graff> their own include/  *
15:46 < graff> but nonetheless, a lot can be accomplished simply by having access to headers
15:47 < graff> so by providing the prototype we can compile a program --- and then LATER dynamically link in the needed library at run time
15:47 < WebHome> In a server environment, does something like ksplice make sense, or is it a nightmare I should avoid?  I’ve heard the ladder a lot.
15:47 < graff> or simple attach a static library to it using `ar'
15:47 < messiah_> well
15:47 < graff> former, latter yeah
15:47 < messiah_> im bit newbie in linux
15:47 < messiah_> i say...
15:47 < graff> i don't know much about ksplice. i can look it up though if you like
15:48 < graff> messiah_: everyone starts somewhere
15:49 < graff> anyway the key words in your error log are: previous declaration of ‘uint64_t’ was here
15:49 < graff> and you can see then that uint64_t was included from two seperate files
15:49 < neachdainn> Is there any way to encrypt a partition with dm-crypt without erasing the partition first?
15:49 < graff> this would be fine is the definitions were identical, but if they are different we get this error
15:55 < Deele> Hello! I am creating a a-bit-more-complex cron-job per-second script, that is launched and with given time interval, runs another command. This is my current working solution https://gist.github.com/Deele/36a0bc37f20de9cf91cb20ff051744cf
15:56 < Deele> But because that another command could take longer than single time interval, I want that command be called as a new process without waiting for previous command to finish - how to do that?
15:56 < m00dy> What type of process scheduling algorithms exist in the kernel ?
15:58 < graff> Deele: so 1> use /bin/sh 2> use printf -- instead of echo 3> this looks like a  (( i++ )) bashism
15:58 < graff> the POSIX specification for echo is hopelessly underspecified
15:58 < graff> and this -e stuff will cause you nothing but grief
15:59 < Deele> echo is commented out - was used for debug only
15:59 < graff> printf handles escape sequences automatically anyway
15:59 < graff> so yo uneed to decide if yo uare using bash or POSIX shell
15:59 < graff> this looks closer to POSIX shell, which is good
15:59 < Deele> I run this with "sh script.sh"
15:59 < graff> just change (i++) to i=$((i + 1))
16:00 < graff> also don't mix [[]] and [] style tests
16:00 < graff> [[]] is a bashism, [] is a POSIX style test
16:00 < graff> Deele: may i rewrite this for you real quick in proper POSIX shell?
16:01 < graff> this way it will work on freebsd, netbsd, busybox based distros like apline and in general with ash
16:01 < Deele> graff, sure, go ahead, it is already glued from leftover stuff anyway
16:01 < graff> kk
16:02 < Deele> I was reading that I should add "&" to the end of call, to make it run as new process, but it seems that it doesn't work
16:02 < ayecee> what happens when you try
16:02 < Deele> desn't change behavior
16:03 < Deele> is it "nohup <command> &"?
16:03 < graff> oh nvm yo uuse a bash array
16:03 < ayecee> where did you add it?
16:03 < graff> i can't easily do the switch
16:03 < graff> so you should use bash for the whole thing then
16:03 < graff> and make all the [] into [[]]
16:03 < Deele> ayecee, look at my link, there is whole script
16:03 < graff> and call it specifically #!/bin/bash
16:04 < WebHome> && does sequential processing, & just makes it a background process
16:04 < Deele> graff, sorry, I don't know the difference yet, but I usually use sh
16:05 < uberwag> need some help on encryption topic, any good channel here to talk about it? luks vs deluks , etc
16:05 < graff> Deele: then your system has symlinked sh to bash
16:05 < stevendale> Hey
16:05 < TaZeR> would anyone care to play some bashfu on this im trying to make a human readable form of the data i get from journalctl --no-hostname |grep "Lid opened" and journalctl --no-hostname |grep "Lid closed" so i can determine exactly when i have been at the computer and my sleeping habits
16:05 < graff> printf -- "\n%s\n" "Please call $0 <workerId> to run this command!"
16:05 < graff> use that instead of echo, for better programming practice
16:05 < Deele> graff, if you can propose anything better for setting working dir to current and ad storing current one in variable for later use, it would remove chunk 12-21 that seems to be with invalid style
16:06 < ayecee> Deele: pushd/popd
16:06 < TaZeR> i need some program that can somehow intertwine the two numbers showing the difference and making that to hours logged from and to
16:06 < TaZeR> or something like that i guess heh
16:06 < graff> pushd and popd are also bashisms
16:06 < ayecee> so use bash
16:07 < ayecee> not everything has to be a sh script
16:07 < graff> i think if he is using POSIX shell then he should use POSIX shell, if not he should specify /bin/bash
16:07 < TaZeR> my bash skills are still in their infancy im not adept enough to handle it
16:07 < rascul> writing for posix is good until it isn't
16:07 < graff> right now though it is an amalgamation of the two
16:07 < ayecee> yeah, the shebang should be changed
16:07 < rascul> posix is good at being ambiguous which leads to different posix compliant shells implementing things differently
16:07 < rascul> you're often much better off writing for a specific shell so you know it works
16:08 < graff> rascul: that's not true.
16:08 < graff> most of this is just from broken echo
16:08 < ayecee> well i'm convinced
16:08 < rascul> https://stackoverflow.com/a/16376043
16:08 < graff> echo itself is underspecified, if use printf most of the problems you desribe will go away
16:08 < graff> anyway, in no way is it ever better to write for a specific shell than to write for POSIX shell
16:09 < graff> unless you need a specific feature of that shell
16:09 < rascul> or if you want it just work instead of potentially debugging incompatible shells
16:09 < graff> bbl
16:09 < ayecee> or you've become tired of reinventing wheels
16:09 < Deele> is bash "better"? I can try anything at this point. I mean, this script is working, I just need it to spawn new process in line 44, rather than wait for previous one to finish.
16:09 < graff> no you guys are making stuff up
16:09 < rascul> bash is "better" if you want to use bashisms
16:10 < graff> and i left this channel years ago, because of this particular anti-POSIX meta dialogue
16:10 < ayecee> Deele: you're already using bashisms. you should make that explicit in the shebang on line 1.
16:10 < graff> it's so stupid, that it is just mind numbing
16:10 < Deele> I agree, I prefer not to make shell specific script
16:10 < ayecee> GrapeApe: get off your damned soapbox.
16:10 < Deele> I am using both bash and zsh
16:10 < ayecee> err.
16:10 < rascul> you missed
16:10 < ayecee> dang.
16:12 < djph> wait, we're anti-posix in this channel?
16:12 < ayecee> ikr
16:12 < ayecee> that escalated quickly
16:12 < WebHome> And deescalated just as fast
16:13 < rascul> like an escalator
16:13 < WebHome> Going up? xD
16:13 < ayecee> love on an escalator
16:13 < robak> I'm getting a lot of defunct sd-pam processes that are owned by systemd on ubuntu 16.04 and I have no idea why. Any way to debug the issue? It happens on more than one machine and elevates system load through the roof
16:13 < rascul> i was expecting a discussion, not a rage quit
16:14 < WebHome> I’m still looking for advice on ksplice.  Yes/no/alternates?  RHEL servers
16:14 < ayecee> WebHome: no one remembers the question
16:15 < ayecee> or was that the whole question?
16:16 < WebHome> In a server environment, does it make sense to use ksplice. I’ve heard nothing but nightmare stories, but keep getting told I should use it. I’m looking for advice on yes or no, or alternatives, or legit reasons to or not to use it.
16:16 < mnemon> WebHome: kpatch is from redhat so might work better
16:17 < mnemon> alternatively build the system in a way that you can reboot without impact on the service :P
16:17 < WebHome> That’s what I’ve been doing the entire time, which is why I’m considering a rebootless option.
16:17 < robak> anyone would have any clue what to look at?
16:18 < rascul> robak look at for what?
16:18 < rascul> oh, i see your question now
16:19 < robak> rascul: the problem I've described few lines above
16:20 < WebHome> mnemon: thanks for the kpatch nod, that looks more promising
16:23 < mnemon> WebHome: I think the rebootless is only reasonable if you for some reason can't reliably reboot the servers in a reasonable time
16:24 < mnemon> the areas you can live patch are anyways limited
16:24 < ayecee> also if your application is structured such that it has no redundancy :P
16:25 < WebHome> mnemon: for now, I can, but the server is running VMs and it’s becoming a time issue as more VMs are placed on the server
16:26 < ayecee> there isn't a second vm server, i guess
16:27 < WebHome> ayecee: I agree, but I can’t afford the redundancy of a second server. I have the best safeguards in place (RAID, DR backups, etc) that I can.
16:27 < ayecee> fair enough
16:28 < WebHome> But suspending each vm takes time, then rebooting, then unsuspending VMs   5 VMs isn’t bad, 30 is.
16:28 < ayecee> yup
16:29 < messiah_> im getting - make: arm-elf-gcc: Command not found
16:29 < messiah_> its means that my gnuarm no visible from that make file
16:30 < messiah_> how to explain it where my gnuarm?
16:30 < robak> I'll repeat my question, I'm getting a lot of defunct sd-pam processes that are 'parented' by systemd on ubuntu 16.04 (kept fairly up to date) and I have no idea why. Any way to debug the issue? It happens on more than one machine and elevates system load through the roof
16:33 < lukey> robak: kernel log, output of journalctl, /var/log/messages
16:40 < messiah_> how to check my default gnuarm compiler?
16:40 < messiah_> or what compiler is will be used
16:42 < robak> lukey: well, nothing there on this subject
16:45 < slondr> hello
16:49 < Pentode> hi
16:51 < messiah_> im still getting make[1]: arm-elf-gcc: Command not found
16:51 < messiah_> how to give understand this compiler where my gnuarm
16:53 < Pentode> messiah_, what happens when you invoke arm-elf-gcc in bash? is it installed?
16:54 < messiah_> sry me
16:54 < messiah_> im newb
16:54 < messiah_> it is
16:54 < messiah_> im compiled it
16:54 < messiah_> it was lasting for 30 minutes
16:54 < messiah_> in my gnuarm
16:55 < Pentode> so it's it in your path? ie, arm-elf-gcc in bash executes the compiler?
16:55 < messiah_> nope
16:56 < messiah_> root@linux:~/trx/src# (arm-elf-gcc) bash: arm-elf-gcc: command not found
16:56 < messiah_> root@linux:~/trx/src# arm-elf-gcc bash: arm-elf-gcc: command not found
16:56 < Pentode> where is it located?
16:57 < revel> I have aarch64-unknown-linux-gnu-gcc and armv7a-hardfloat-linux-gnueabi-gcc :D
16:58 < messiah_> Pentode: /root/gnuarm/install
16:58 < messiah_> sry
16:58 < messiah_> Pentode: /root/gnuarm/
16:59 < Pentode> well no wonder it wont work. have you actually done a make install?
16:59 < revel> Or maybe it's just installed to somewhere that's not in his PATH
16:59 < Pentode> if so and it / and or you _really_ want to keep it there, you have to add that to your path.
17:00 < messiah_> i need to make install after it was compiling right?
17:01 < Pentode> typically, yeah. otherwise it is compiled in it's own source tree but it hasnt been installed yet
17:01 < messiah_> i download it then just execute -
17:01 < messiah_> # ./download.sh # ./build.sh
17:01 < ses1984> i'm trying to write a udev rule to run a script when a bluetooth device is connected/disconnectd, but it's not behaving quite like i would expect,
17:02 < ses1984> when i run `hcitool con` as an interactive user i see connected bluetooth devices but when i run this as part of the udev RUN script i never see any devices
17:02 < messiah_> Pentode: === INSTALL  Make sure that you already have installed following packages: build-essential libgmp3-dev libmpfr-dev libx11-6 libx11-dev flex bison libncurses5 libncurses5-dbg libncurses5-dev libncursesw5 libncursesw5-dbg libncursesw5-dev zlibc zlib1g-dev libmpfr4 libmpc-dev.  1. Download the latest installer: git clone https://github.com/axilirator/gnu-arm-installer.git cd gnu-arm-installer  2. Run this scripts: ./download.sh
17:03 < Pentode> a separate installer? what is the world coming to?
17:04 < Pentode> revel, will it work properly if he just adds that to his path?
17:04 < messiah_> in install dir there just dirs no files
17:04 < revel> Well, my cross-compiler is in aarch64-unknown-linux-gnu-gcc-7.3.0 and I have no idea why it's in my PATH.
17:05 < revel> Something in a profile file somewhere, probably.
17:07 < revel> `printf "$PATH" | wc -c` => 1136 :D
17:07 < Pentode> lol
17:08 < messiah_> here what i have in tutorial which i used to install that
17:08 < messiah_> # ./download.sh # ./build.sh  # export PATH=$PATH:/root/gnuarm/install/bin
17:09 < messiah_> here link to tutorial - https://bastienbaranoff.wordpress.com/2016/02/08/gsm-base-station-with-two-osmocom-bb-compatible-phones-on-kali-rolling/
17:10 < TRS-80> I love you guys
17:10 < Pentode> well if you exported it as above, and it is indeed in /gnuarm/install/bin it should be working ;)
17:10 < revel> TRS-80: "Don't come to Github tomorrow"?
17:11 < messiah_> hooray
17:12 < messiah_> its done
17:12 < messiah_> its compiled now
17:12 < messiah_> thanks
17:12 < TRS-80> revel wat?
17:12 < messiah_> with your superior admistration over my my compile machine
17:12 < messiah_> its done now
17:12 < Pentode> oh good
17:13 < messiah_> i was entering #export PATH=$PATH:/root/gnuarm/install/bin - yesterday
17:13 < ses1984> why would running `hcitool con` as an interactive user display my connected devices but running the same from a udev script display nothing?
17:13 < messiah_> after restart its possible reseted
17:13 < Pentode> doh, lol
17:13 < messiah_> or resets* how to better spell it
17:14 < Pentode> you can add it to bashrc or your bash_profile, however your distro does it.
17:14 < Pentode> doesn't hurt to just throw it in ~/.bashrc though for simplicity if you wanted to
17:14 < messiah_> Pentode: can you tell me in steps
17:14 < Pentode> just add that export command to /root/.bashrc/ thats all.
17:15 < messiah_> so #export PATH=$PATH:/root/gnuarm/install/bin - is always for compiler? there nothing sad like complilet or gnuarm in that line
17:15 < Conder> hello, sorry for partial OT, but hashcat gave me this wifi password: $HEX[3132333435363738393000] when i try to decode it, i get 1234567890 and some weird character at the end. then how should i connect?
17:16 < revel> Conder: Try 1234567890 :D
17:16 < djph> 1234567890[weirdcharacter]
17:16 < revel> Maybe it's a NUL at the end or something.
17:16 < revel> lol, or what djph said.
17:16 < Pentode> messiah_, with that command make was able to determine where the compiler is located so it can execute it.
17:17 < backnforth> Hi, what's the best todo system to use when using Linux?
17:17 < messiah_> Pentode: that what i asked - so #export PATH - is meant compiler
17:18 < Pentode> adding that to your "path" means that all of the binaries in that directory are accessible root user wide by simply invoking it at the prompt, without the need to enter the containing directory or specify the full path to the binary
17:19 < Pentode> messiah_, the compiler, and anything or anyone else.
17:19 < Pentode> -,
17:19 < TRS-80> backnforth: todo.txt
17:19 < backnforth> TRS-80,  OK thank you
17:20 < TRS-80> baccknf
17:20 < dunpeal> Hi. I got a fancy used Xeon workstation with tons of RAM and cores, but it occasionally just freezes on me. The mouse cursor keeps moving, but there's no response or change otherwise. How do I debug this?
17:21 < dunpeal> *just got
17:21 < revel> dunpeal: I think I had that with Manjaro...
17:21 < TRS-80> backnforth: what other devices you use? Android? That's my use case, but an important philosophical approach (from GTD) is that whatever system you choose should be available everywhere and easy to use, or you won't have faith in it, and won't use it.
17:21 < backnforth> TRS-80,  I use Android yeah
17:22 < revel> No idea, maybe it's some specific graphics driver not playing ball.
17:22 < TRS-80> backnforth: do you follow any certain methodology (like GTD) or you just looking for something simple?
17:22 < sl4ck> .
17:22 < backnforth> TRS-80,  I'm willing to learn something new
17:22 < backnforth> I never heard of GTD before
17:23 < dunpeal> revel: this is Xubuntu 18.04
17:24 < jim> how can I tell (maybe using ip) whether my host is using ipv6 or ipv4?
17:25 < revel> `ip a` should tell you if you have any ipv6 addresses, I think.
17:25 < TRS-80> backnforth: let's just say I been thinking about these sort of issues and refining my various systems for a number of years. Even if you don't adopt all of the GTD ideas, there are some very good ones in there. Like contexts. I'll leave the search up to you, but it's such a popular book it's likely available at your local library system, or through loan. That will get you thinking in the right way about the whole problem to start
17:25 < TRS-80> out. It's short for "Getting Things Done" and it's actually a technology agnostic system (you could do it on index cards).
17:26 < dunpeal> Is there a way to read any last error messages from right before the system froze and I had to reboot it?
17:27 < jim> dmesg (if you have access)?
17:27 < dunpeal> jim: I have root, sure.
17:27 < dunpeal> dmesg only seems to show messages since the last boot.
17:28 < TRS-80> backnforth: Having said that, then there is a tech component. And I have evolved to thinking that you want lowest common denominator, which to me means text files. They are universal across platforms. Hence my choice of todo.txt. Just because the format is universal, doesn't mean it needs to look ugly or you need to use Emacs, etc. Check out SimpleTask for Android which is available on FDroid.
17:28 < TRS-80> bacnkf
17:30 < djph> TRS-80: so, pencil and paper?
17:31 < TRS-80> backnforth: So I also have my bookmarks in text files, and other notes and thoughts (and increasingly, links) in a TiddlyWiki (self editable HTML file). All these files are synced across all my devices on a self hosted cloud instance. I have NextCloud (fork of OwnCloud) running on a Cubietruck here in my bedroom. :) It only uses like $20 per year in electricity. Lovely little home GNU/Linux server. But you could sync with anything
17:31 < TRS-80> if you weren't into setting that infrastructure up. Again, I have evolved this system in bits and pieces over years.
17:32 < Pentode> im a fan of doing it that way also..
17:32 < TRS-80> djph yep, absolutely. Although I prefer the ability to edit things without copying that devices give you, there are absolutely people who do GTD on index cards and notebooks.
17:33 < TRS-80> it is a methodology, a philosophy
17:34 < TRS-80> if you are on computers there are other cool things you can also do with text based files obviously and automation, esp in Linux...
17:34 < cxc99> how would i test a port forward besides actually logging in from another server?
17:34 < djph> TRS-80: ah.  I actually recently found a nroff example for writing RFC-style things ... that may actually come in handy (and also give the other team an anneurism when they get 40+ pages of plaintext)
17:34 < djph> cxc99: nmap from the outside
17:35 < cxc99> there's no other way?
17:35 < djph> not really, no
17:35 < dunpeal> Does anyone know how I can check the error log _before_ my last boot (to figure out why my system crashed)?
17:35 < djph> dunpeal: /var/log/dmesg# might have a previous boot
17:36 < djph> *previous log
17:36 < dunpeal> djph: thanks.
17:37 < djph> dunpeal: looks like I've got 3 or 4 boots worth (plus the current)
17:38 < TRS-80> but twhat I came to share with the channel today actually is  this cool project I came across: https://github.com/Eloston/ungoogled-chromium
17:39 < backnforth> TRS-80,  I had my own owncloud instance before on a Digital Ocean droplet. I've never used SimpleTask before though. Nor have I used TiddlyWiki before.
17:40 < j0seph> I would set up NextCloud on my droplet but eh, I'm not sure lol
17:41 < backnforth> TRS-80,  Would I have to change my default router settings if I set up a NextCloud instance in my home?
17:42 < TRS-80> backnforth: how so? I mean, you shouldn't other than allowing the port(s) through. It's essentially hosting a web server.
17:42 < backnforth> TRS-80,  I'm worried about doing that for security reasons
17:44 < djph> backnforth: if you're running it from home *AND* want to access it from the internet
17:44 < TRS-80> backnforth: I haven't had any problems, been up couple years now almost and it works flawless. Only downtime is power outages and then it boots right back up (yes I know, I should have some power backup to gracefully shutdown, which I will in time as resources allow)
17:45 < TRS-80> j0seph: what are you not sure about? It's a pretty mature product at this point with a lot of development and support behind it
17:45 < j0seph> TRS-80: the problem isn't so much the product itself rather than the resources I have to host it on
17:46 < TRS-80> j0seph: I run it on a Cubietruck, which is no speed demon. Works a treat for file, contact, and calendar sync for our family. Web interface is a little slow, but I only really use that "in a pinch" like when logging in from a friend's computer or whatever.
17:47 < marko_> Is their a way to resolve kernel panic issue in linux.I accidently ran a command as root.
17:48 < j0seph> TRS-80: 1GB RAM, 25GBSSD, and 1TB transfer should be fine then?
17:48 < Sitri> marko_: reboot
17:48 < TRS-80> Cubietruck also doubles as our home media server. Has a multi TB SATA drive attached. SATA being one of the main reasons to choose the Cubietruck over other SBSs (along with good community/Linux support)
17:48 < TRS-80> j0seph: lemme look up CT specs real quick, standby
17:48 < j0seph> TRS-80: gotcha
17:50 < turkeyhand> if I repair the windows install, /j #lineageos
17:50 < turkeyhand> oops
17:50 < dunpeal> These are the errors right before the GUI completely froze last night: http://dpaste.com/10VE1PX
17:52 < TRS-80> j0seph: Cubietruck SoC is an A20 which is only a dual core ARM Cortex-A7 1GHz processor, although it does have 2GiB DDR3 @ 480MHz
17:52 < dunpeal> Is this a reasonable explanation?  "Chrome tried to Hardware Acclerate and due to some bug caused a critical part of the nouveau driver to crash".
17:53 < j0seph> TRS-80: i see. i'm paying for my VPS with a rate of $5 a month for the specs I mentioned earlier (plus some i5 core that I forgot the specs of)
17:53 < prussian> dunpeal: sure
17:53 < djph> dunpeal: nouveau going south is an explanation, yes.
17:53 < TRS-80> j0seph: and obviously a 25GB would only be good for contact, calendar, bookmark, todo, and other text/personal/work files, and not tons and tons of large media (video) files
17:54 < djph> TRS-80: that's what sftp is for :)
17:54 < prussian> dunpeal: update your kernel. lots of nouveau fixes are in the mainline kernel
17:54 < prussian> that one looks familiar
17:54 < TRS-80> j0seph: if you prefer to host offsite that's up to you but in my mind what's the point of hosting your "own cloud" if it's not physically in your possession / control?
17:55 < j0seph> TRS-80: well, that's part of the reason i'm not so sure, eh? :)
17:55 < TRS-80> j0seph: I mean, sounds like your CPU is likely way faster than mine, and cost is reasonable, so depens what considerations are important to you, I suppose. To me privacy is most important.
17:55 < djph> if "security" is a concern, get something better than your typical consumer-grade "naaah, they'll never need more than one LAN" router.
17:56 < TRS-80> djph: all in due time my friend ;)
17:56 < djph> TRS-80: oh, I thought that was his "use a VPS" reason
17:57 < TRS-80> but I gotta tell you djph (and j0seph, and anyone elsee following along) I been real real happy with the Cubietruck for the $80 it cost me!
17:57 < djph> I have an old dell rackmount server in the basement
17:57 < j0seph> i have a VPS because I had a coupon
17:57 < j0seph> i am a simple guy. :P
17:58 < TRS-80> djph: I probably made my $80 back in electricity over yours in a couple months. XD
17:58 < dunpeal> prussian: this is with the latest 4.15.0-20-generic kernel available on Ubuntu 18.04
17:58 < djph> TRS-80: probably. But the fact I can run half a dozen VMs on it all doing different things was fun
17:58 < prussian> then make an issue with ubuntu's kernel maintainer
17:59 < TRS-80> main difference of Cubietruck over other SBCs is SATA connection. Then you can just connect nowadays inexpensive SATA drive for tons of storage / server. Also it doesn't share the network with the USB ports on the same bus like the RPi, lol
18:00 < j0seph> TRS-80: in any case, it's a case of weighing up what i value the most. i do like privacy and i'd like to get away from hosting my files on dropbox and google chrome where i don't have 'control' but like you said, using nextcloud in the name of privacy and then hosting it on an offsite VPS is a bit contrarian situation aha
18:01 < dunpeal> prussian: I think I'll just install the prop Nvidia driver.
18:01 < prussian> i see
18:02 < TRS-80> djph: I'm busting your stones. Truth is I have a more powerful server build in the works based on a ASUS KGPE-D16 server board which is supported in LibreBoot (https://libreboot.org/docs/hardware/kgpe-d16.html)
18:02 < TRS-80> It's an addiction. Amirite? :)
18:02 < dunpeal> Otherwise it's a hassle to either pressure Ubuntu to update the kernel, or maintain my own forked kernal.
18:02 < fendur> TRS-80: does the sata port get good speeds?
18:03 < TRS-80> fendur: I'm not one of these benchmark people, but I'm sure you could find data on that. I do know it's faster than other SBCs because it doesn't share the bus, and performance wise I see no visible lag. I have the drives mounted via NFS at boot on this machine and it's just like they are local.
18:04 < TRS-80> fendur: er, drive, rather. There is also 8GB NAND which comes on the Cubietruck, besides the SATA drive you can attach.
18:04 < prussian> dunpeal: well ya. I'm running custom kernel on fedora because of this nouveau adventures
18:05 < prussian> started at 4.14 for me and only got worse. now I have a patch free kernel with 4.16.x
18:06 < fendur> TRS-80: yeah. asking because the specs are just specs. and because I want to make sure I'm not going to be limited to 10MB/sec if I use it as a NAS or something. wife gets cranky if it take sa while to copy a file :)
18:07 < TRS-80> j0seph: it's not impossible to do. You would get the CT, install Linux like normal (there are good solid distros like Armbian readily available) and then basically set up a LAMP stack (or other web server of your choice. NextCloud sits on top of that. It's not hard to do if you allow yourself some time. And you will learn a lot about Linux (I know I did).
18:08 < revel> "Armbian"? Doesn't regular Debian have armv7 and aarch64 support anyway?
18:08 < KernelBloomer> what codec should I get to watch wma videos
18:08 < KernelBloomer> on linux
18:08 < KernelBloomer> hmm?
18:08 < revel> (and a better name :D)
18:08 < j0seph> TRS-80: sounds cool, but how different is a CT to an RPi?
18:09 < cutesona> ls
18:09 < cutesona> hs  file	file.py  xy.txt  wav_file
18:09 < cutesona> rm xy.txt
18:09 < cutesona> hs/            file     file.py  xy.txt      wav_file/    <— when i execute command of ‘rm proxy.txt’ then it don’t delete xy.txt file, instead it show me list. what’s wrong?
18:09 < TRS-80> fendur: well we stream 1080p to the living room TV through KODI on an RPi3 with no issues. It does choke on "Big Buck Bunny" but that's like the gold standard in streaming tests and the RPi is only connected via wifi (CT is wired to router).
18:10 < Sonolin> yea never had any issues with my SBC streaming 1080p video (from the net even)
18:10 < fendur> TRS-80: sure. I use a rpi2 for that.
18:10 < fendur> haven't put my 3 to work yet
18:10 < fendur> But I don't think NAS is the right job for it
18:11 < TRS-80> j0seph: they are in concept similar (both SBCs) however hardware is better, and actually, more open than RPi
18:11 < cutesona> No one know it?
18:11 < Sonolin> I've been interested in the OrangePi
18:11 < TRS-80> fendur: yeah NAS no good for RPi due to shares bus issue. Great front end media player though because large community support and HDMI out. No HDMO out on Cubietruck.
18:11 < Sonolin> bootloader less crazy than the rpi, which is the main seller for me
18:12 < j0seph> TRS-80: sounds good. more expensive though, I imagine?
18:12 < fendur> Sonolin: ah, for the cubie, you mean?
18:12 < jim> cutesona, maybe there's a hackspace or makerspace near you that has people who do
18:12 < Sonolin> fendur: not sure what you mean, just logged in, but this is the website:
18:12 < Sonolin> http://www.orangepi.org/
18:13 < Sonolin> much more powerful than the cubieboards
18:13 < TRS-80> j0seph: $80 when I bought mine year and a half ago, but remember that includees power supply and 8GB NAND onboard, with RPi you have to add those in so it gets much closer
18:13 < cutesona> jim: then it is not error?
18:13 < j0seph> TRS-80: suppose so, yeah. i've more experience with an RPi in any case. I actually have it with me, I really should use it more. It's one of the older Version 1 models.
18:14 < jim> dunno :) I have no idea about the device you speak of
18:14 < cutesona> thanks :)
18:14 < jim> welcome
18:15 < FrozenFire[alt]> I am trying to add a directive to /etc/pam.d/sshd which invokes a script via pam_exec to do something with the user's environment on SSH login. However, I cannot seem to access the user's full environment - specifically the environment variables forwarded by SSH. Is there some flag I need to specify?
18:15 < TRS-80> Sonolin: there are tons of SBCs that "on paper" have impressive hardware specs, but when you dig into it a little deeper, they are on some ancient kernel because the mfr has no concept of what open source means, and they don't support the community, release documentation to make drivers, etc... more info here: https://linux-sunxi.org/Buying_guide
18:15 < Sonolin> TRS-80 no, check linux-sunxi.org (I think that's the site)
18:15 < Sonolin> the orangepi runs the latest kernel with no modifications
18:15 < Sonolin> and its $24
18:16 < TRS-80> j0seph: CT will be very similar if you are running Linux on RPi, and CT is actually more open
18:16 < TRS-80> Sonolin: yes I have heard that things are progressing nicely, I made my decision year and a half ago and haven't looked back so you could be right
18:17 < Sonolin> yea, I'm very aware most boards have issues - which is why I'm looking into that one, since my current is a pos
18:17 < j0seph> TRS-80: i actually do not know what's running on there. probably raspbian.
18:17 < Sonolin> well, I say that, but at least it plays 1080p without stuttering I guess...
18:18 < TRS-80> j0seph: well it's some sort of Linux though. Problem with RPi is the proprietary GPU/bootloader blob which starts first before everything else, including OS. RPi therefore, while being most famous and bringing a lot of eyeballs and feet to the space, is actually one of the least "open" pieces of hardware...
18:19 < camer0n_> Can anyone hear me?
18:20 < TRS-80> Sonolin: yeah that's awesome. I actually used to play some video games with a guy who codes some of the drivers and stuff for that and other newer devices and he was telling me same. Which is great news, things are coming along.
18:20 < j0seph> TRS-80: i'll prioritise performance and usability over whether something is more open / familiar or not, but if cubietruck does have better hardware like you say then i'll look into it more
18:20 < j0seph> camer0n_: of course we can't, it's IRC, there's no voice chat
18:20 < prussian> you forgot the RPi funky bluetooth and custom bluez stuff too
18:20 < camer0n_> lol thx
18:20 < camer0n_> Just setting up irssi
18:20 < Sonolin> j0seph why does performance matter when 99% of these boards have similar specs?
18:20 < Sonolin> *new boards
18:20 < prussian> depends what you're doing I guess
18:20 < camer0n_> Still, it doesn't have my username right...
18:21 < Sonolin> I guess if you're mining for BC or something it makes sense, but people usually go dedicated GPU for that
18:21 < j0seph> Sonolin: That was a broad statement, not just in regards to those boards
18:21 < prussian> if you want performance. I'd just ask NXP what they can give you and go to some company to make design you your own board with whatever
18:21 < Sonolin> j0seph eh, same could be said for most modern computing
18:21 < Sonolin> intel vs. amd is mainly politics & splitting hairs at this point
18:22 < j0seph> Sonolin: as much as i'd like that to be the case, it's not quite reached that point fully yet.
18:22 < Sonolin> unless you're load average is >3 most of the time, I won't take your point seriously at all :P
18:22 < TRS-80> Sonolin: if that's the case with OrangePi, how come nothing on Buying Guide page of linux-sunxi.org as I linked above? Is it just out of date?
18:22 < Sonolin> TRS-80 probably, I just use the mainline status page to check compatibility
18:23 < Sonolin> its a wiki, so some things will be out of date
18:23 < j0seph> Sonolin: that's okay, i don't expect to be taken seriously on IRC regardless. :P
18:24 < Sonolin> well at least you're sane ;)
18:24 < j0seph> Sonolin: haha, borderline sane
18:25 < TRS-80> I guess I look at things different. The only reason we can have any nice things at all is because some hardware mfrs. don't have their head up their arse. I feel very strongly that those sort of hardware mfrs should be rewarded by all of us voting with our feet/dollars. Certainly dev support seems to flow in those directions, because the devs understand the issues. And a device with good dev community around it will work well and be
18:25 < TRS-80> supported for a long time. That's the one I want to buy.
18:27 < j0seph> TRS-80: is RPi not that? genuine question, i've not ventured into this topic for quite some time now
18:28 < fendur> Sonolin: ah thanks (sorry for delay). So you have no problem getting linux going? Do you trust the images that are avialable? Or do you install yourself from a standard dist?
18:28 < TRS-80> j0seph: RPi has the biggest community for sure, but not of knowledgeable people. It's the blob I discussed earlier that's the problem. There is one guy trying to reverse engineer the MALI (or whatever) graphics drivers though, bless his heart.
18:29 < prussian> it's pretty bad. videocore and bt radio are all kinds of what and it basically only runs raspbian reliably.
18:29 < prussian> and who would want to run raspbian. yikes
18:29 < j0seph> prussian: what would you run
18:30 < prussian> some custom yocto or timesys linuxlink
18:30 < prussian> though timesys tells me they "support" RPi boards now.
18:31 < prussian> but even something like fedora server on rpi 3 was pretty lackluster
18:33 < prussian> maybe it's better now. I thought videocore stuff was foss now if you don't want h.264
18:34 < Sonolin> fendur: I always try mainline kernel first, but something like rpi will likely need custom firmware/drivers
18:35 < Hobadee> Hey everyone, if you are a GSuite admin, there is a new channel over at #gsuite for us!  Head on over!
18:35 < dunpeal> prussian: why not just use the prop driver instead of going through all this trouble?
18:38 < prussian> because I'm not going to run nonfree software unless my life depends on it.
18:38 < Sonolin> ^^
18:39 < prussian> i can easily buy some functioning alt discrete profile card to replace this stupid fermi card that is in my work desktop
18:39 < prussian> but for the most part nouveau works now since 4.16.
18:39 < TRS-80> dunpeal: do you even know where you are? XD
18:40 < jim> Hobadee, probably not a good thing to do channel advertising or mass invites
18:41 < dunpeal> TRS-80: hey, I like Linux, and prefer FOSS, but I won't kill myself over not installing one piece of non-free software on it :)
18:42 < koala_man> well, maybe you should
18:43 < koala_man> there is no GNU but GNU and Stallman is its prophet
18:43 < j0seph> i feel like that two-line interaction between dunpeal and koala_man really does summarise the FOSS philosophy. :P
18:43 < emberquill> Oh boy, here we go...
18:43 < dunpeal> Amen brother.
18:44 < Sonolin> dunpeal I think he said "unless my life depends on it", not the other way around...
18:44 < koala_man> I just logged on and saw that one line, so I knew I had to fight for freedom
18:44 < oehansen> Haleluja!
18:44 < Sonolin> *he/she/xey/they/it/dogperson/whatever
18:44 < oehansen> May the GNU be with you brother
18:44 < Sonolin> koala_man preach, brotha!
18:45 < j0seph> Sonolin: goddamn it, stop excluding my demographic: "empty lemonade bottle"
18:45 < koala_man> there's 72 root accounts waiting for us in the cloud
18:45 < Sonolin> sorry, I'll just put ____ next time
18:45  * oehansen google fuss philosophy
18:45 < Sonolin> :P
18:45 < TRS-80> emberquill: kek
18:45 < prussian> just put other (please specify)
18:46 < oehansen> man!
18:46 < j0seph> woman!
18:46 < oehansen> where!
18:46  * oehansen looks around
18:46 < Sonolin> I'll put whateverthefuckingpronounIwant next time ;)
18:46 < TRS-80> oehansen: not in here
18:46 < oehansen> why you ... you got me all aroused there for a sec bro! :-)
18:46 < Sonolin> oh, right, this isn't #bsd
18:46  * bls puts his BSD necklace in his shirt to escape the scrutiny of the morality police
18:47 < anonnumberanon> Hi. If I put the IP address of a computer in my browser, what port will be attempt connected by default, 80?
18:47 < Sonolin> haha :)
18:47 < oehansen> haha!
18:47 < anonnumberanon> Trying to figure out what port I should give my web server.
18:47  * oehansen noticed BSD ... that must me a Mac OSX dude!
18:47 < prussian> http:// implies 80 and https:// implies 443...
18:47 < bls> anonnumberanon: yes, that's the default http port
18:47 < prussian> those schemas anyhow
18:47  * oehansen scurily hides his ryzen MACOSX pirate copy
18:47 < TRS-80> prussian +1
18:48 < oehansen> jawhol!
18:48 < j0seph> jaw hole?
18:48 < anonnumberanon> bls, i see, awesome
18:49 < anonnumberanon> prussian, wow thanks, makes sense
18:49 < bls> anonnumberanon: and have a look in /etc/services if you're interested
18:51 < bls> anonnumberanon: that's where most *nixes establish service to port number relations
18:55 < wad> I've got several hundred linux servers that I'm working with. Using some tools to run commands on them in bulk, and gather results. I need to see the status of a service (logstash). Problem: Some of them require "service logstash status", and some require "service status logstash". Is there a way, in one line, I can have it choose the right command to execute? For stopping and starting, I just do both. But for getting the status, I have to get the result.
18:55 < anonnumberanon> bls, i do not have that path, i am running:   Linux 4.13.0-41-generic #46~16.04.1-Ubuntu SMP Thu May 3 10:06:43 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
18:55 < bls> wad: ansible
18:56 < TRS-80> wad make a script which, upon error, does the other one?
18:56 < THE_GFR|WORK> hey everyone have a SAS HBA in one of my systems running Linux Mint.  If I plug in drive to the external SAS to SATA cable too many times they stop showing in Linux Mint Disk Utility.  The only way to get them to show back up in disk utility is to reboot the system.  The drives show in the MegaRaid Storage Manager though, but they do not have a device ID aka /dev/sdc or something similar ideas?
18:56 < bls> not a surprise. it was a simple, long used, well established scheme. I assume someone has replaced it with something way more complicated
18:57 < bls> wad: you give it a list of hosts, a service name, and it'll figure out the correct way to interact with the service based on the distro
18:57 < wad> So yes, I could make scripts, and go that route, but that invovles other teams doing extra work for me, and increases the complexity of an already-complex system. I was hoping to find a one-line solution somehow.
18:57 < THE_GFR|WORK> so I am not really sure why the drives stop showing up in Disk Utility... do I need to restart the udisks service?
18:58 < THE_GFR|WORK> disks on USB show up in Disk Utility without issues
18:58 < THE_GFR|WORK> that's gnome disk utility aka Linux Mint but yea
18:58 < tomreyn> wad: you can probably do the if then else yourself in one line?
18:58 < wad> Yeah, I'd love to use ansible. Trouble is, the system I'm working with also hits a bunch of crappy Windows boxen. (And the world sucks more because of it.)
18:59 < THE_GFR|WORK> udisks2 or something
18:59 < wad> tomreyn, what would I check on, in the "if"?
18:59 < bls> ansible has windows support now, assuming you've got a sane way to remote in
18:59  * THE_GFR|WORK has an Ansible script to deploy his software
19:00 < wad> bls: Really! I tried to help someone to get ansible set up, and they were using Windows as their OS, and they were dead in the water.
19:00 < tomreyn> wad: i don't know. i'm only used to "service <servicename> <command>". which kind of system does it the other way around?
19:00 < wad> There are amazon-linux boxen, and centos 7.
19:01 < THE_GFR|WORK> would I just restart a specific service like the disk utility service or something?
19:01 < tomreyn> wad: amazon-linux is something rpm'ish, too, isn't it? maybe check lsb_release -ds
19:01 < bls> http://docs.ansible.com/ansible/latest/user_guide/intro_windows.html
19:02 < wad> Sounds good.
19:03 < wad> bls, yeah, I'd totally use ansible, but in this case, it's not an option. Too much investment in the other tool.
19:03 < bls> fair enough
19:04 < wad> Appreciate the info and support though. :) If I get to greenfield, ansible will definitely be on the table!
19:05 < djph> TRS-80: yeah, getting new kit is :) (grrr phone calls)
19:10 < anonnumberanon> prussian, bls , it worked for my port 80 thing, awesome, thanks.
19:17 < rx7a> Hey all!
19:17 < rx7a> I am blocked on setting Odoo upon Apache!
19:17 < quul> hey rx7a!
19:17 < rx7a> Hi quul o/
19:17 < jim> hi
19:17 < rx7a> Yep, I need some advice about the conf of Apache
19:18 < rx7a> Here is my "/etc/apache2/sites-available/my_domain.conf"
19:18 < rx7a> http://paste.debian.net/1025284/
19:18 < rx7a> I ran those next commands:
19:18 < rx7a> a2ensite my_domain
19:18 < rx7a> systemctl reload apache2
19:19 < rx7a> Then I went to "my_domain"... And it doesn't work :'(
19:19  * mouses dances with mnrmnaugh
19:20 < mouses> lol wrong channel
19:21 < turkeyhand> mouses
19:21 < rx7a> Oooops, Didn't say hi to jim... Scuse
19:21 < turkeyhand> who told you boot-repair fixes arch installs?
19:21 < turkeyhand> why'd you insist that it would for me D:
19:25 < zerowaitstate> rx7a: when you said you specified the IP in ProxyPass, is that the interface address or localhost?
19:26 < jim> rx7a, no big deal :)
19:27 < sla3k> Hi there, any good video streaming open source which are not python based? I searched but looks like mediadrop is the only option for video streaming
19:27 < sla3k> Basically this is for internal training videos that we have for new hires.
19:28 < zack6849> sla3k: why such a strange requirement?
19:28 < zack6849> did python kill your family?
19:28 < zack6849> either way, yes, there definitley are
19:28 < zack6849> https://github.com/obsproject/obs-studio
19:29 < sla3k> nothing against python but I am not a python person, too much learning curve for me, and believe me, if python would've killed any of my family, python would've been dead already!
19:29 < rx7a> zerowaitstate: it is the interface_address
19:29 < zack6849> obs is pretty good i assume it can work on linux last i recall
19:29 < j0seph> learning curve... for python?
19:29 < j0seph> hm..
19:29 < ses1984> i'm trying to automate something to happen after a bluetooth device is connected using udev. whenever i run a script as part of a udev rule, it can't seem to see any connected devices. but the same script run interactively works fine
19:30 < zerowaitstate> rx7a: is your app server bound to 0.0.0.0 or the interface_address, or is it bound to the loopback interface (127.0.0.1)?
19:30 < ses1984> for example `hcitool con` -- lists all devices when i run interactively, lists nothing when run in a udev rule
19:31 < indigoblue> ses1984: maybe has to do with user permissions?
19:31 < rx7a> zerowaitstate: Excuse me, you mean the server_ip with interface_addresse ?... I have set IP to the server_addr
19:31 < sla3k> zack6849: thanks, I'll look into obs.
19:31 < rx7a> I have set IP to the server_IP
19:32 < ses1984> indigoblue: i tried using sudo in the udev script with the same result. i'm teeing STDOUT and STDERR to a file so i can monitor what it's doing and i'm not getting errors
19:32 < indigoblue> ses1984: udev rule might be executed on a different user? add a line in your script to record the user
19:32 < indigoblue> see if its different than when you run the script interactively
19:33 < zerowaitstate> rx7a: is 0doo or whatever listening on 127.0.0.1?
19:33 < ses1984> indigoblue: i added `whoami >> /tmp/log` and it says `root`
19:33 < ses1984> which is weird, because if i `sudo su` and run the script it does give the results i want/expect
19:34 < indigoblue> hmm, probably something else then
19:34 < bls> ses1984: some udev setups run their scripts in a restricted environment
19:34 < zerowaitstate> rx7a: the command "ss -tlpn" will tell you what IP addresses and TCP ports your applications are listening on
19:35 < ses1984> bls this sounds reasonable, i tried searching for it and didn't find anything
19:35 < bls> where you may not have access to the entire file system, all devices, etc
19:49 < giaco> hello
19:50 < giaco> could someone explain me why tar sometimes archives files as "filename" and sometimes as "./filename" ? It is a very very stupid nonsense behaviour
19:51 < noodlepie> "tar zcvf filename.tgz /foldersToInvlude
19:52 < noodlepie> or do you mean the files inside it? Here I assume you are trying to unzip it a noncurrent folder?
19:53 < giaco> noodlepie: no zip, no compression. Plain tar. I am talking about the files already inside the tar, the ones displayed with tar tf file.tar
19:54 < noodlepie> maybe if from making the tar file with ./file rather than just "file" as the source data? @:P-~
19:55 < bls> ./file and file are functionally equivalent
19:56 < giaco> bls: ask the software that is parsing the tarfile
19:56 < bls> you mean tar?
19:57 < Dagmar> ...if you're sitting in /usr/bin or something maybe'
19:57 < meyou> someone should make a compression app called feathered imo
19:57 < koala_man> giaco: files are stored with Unix path, and like most tools it mimics the path you give it
19:57 < giaco> bls: no, any program, such a python script, that expects "file.txt" in the archive but founds "./file.txt"
19:57 < bls> that's a problem in the python script then if it doesn't understand that the two are the same thing
19:59 < giaco> bls: sure, because tar needs two way to express a file in the root of the archive. Much information there, so cool
19:59 < meyou> giaco, tar just doesn't bother to strip out ./, much like whatever script you ran into this problem with
19:59 < bls> it doesn't need two ways. it does what it's told to do
19:59 < meyou> "two spidermans pointing at eachother"
20:00 < bls> and it sounds like the script is trying to manually parse filenames/paths instead of using something like os.path
20:00 < meyou> i wonder how the script would handle if /etc/passwd was tarred with an absolute path :o
20:02 < bls> heh, saw a rant on that the other day, how the -P flag not being on by default was the greatest injustice to humanity ever
20:02 < kremator> guys, which is closer to RHEL? CentOS or Fedora?
20:03 < bls> kremator: do you understand the project's lifecycle?
20:03 < kremator> bls, link please?
20:03 < giaco> bls: the script is just looking for "file.txt" in the root of the archive. Python tarfile.open("foo.tar", 'r').getmember("bar.txt")
20:03 < kremator> bls, only about Fedora not really on RHEL
20:04 < bls> kremator: fedora is essentially the pre-release/proving ground for what eventually becomes RHEL, CentOS then takes RHEL and strips off the copyrighted stuffs
20:05 < bls> https://docs.fedoraproject.org/quick-docs/en-US/fedora-and-red-hat-enterprise-linux.html
20:06 < kremator> bls, usually, what kind of copyrighted stuff i would lose between RHEL and CentOS?
20:06 < koala_man> giaco: well, it's already making non-canonical use of tar archives
20:06 < kremator> bls, i do want to download the iso to try out and lern it on a VM, but i would like to get the closest experience to RHEL
20:06 < kremator> so far i think i will go on CentOS
20:06 < bls> kremator: then use CentOS
20:07 < kremator> bls, yeah, thanks
20:07 < bls> kremator: mainly themes and documentation
20:07 < giaco> koala_man: what do you mean?
20:07 < kremator> bls, but all the service maanagment and systemd is there right? so if i learn how to setup X service in CentOs i would be 90% th same on RHEL right?
20:07 < koala_man> giaco: tar files conventionally have a single top level directory, unlike zip files
20:08 < koala_man> so that when you e.g. untar the linux sources, all the files are in a dir  linux-2.6.20/*  instead of in the base dir
20:08 < bls> kremator: yes, the majority of it is a search-relace Red Hat -> CentOS
20:08 < giaco> koala_man: same problem, "prefix/bar.txt" or "./prefix/bar.txt"
20:09 < giaco> koala_man: which of the two is the conventional way?
20:09 < Rihannsu> opensuse > fedora, haters gonna hate
20:09 < koala_man> giaco: prefix/bar.txt
20:09 < bls> kremator: or rm RedHatLogo.png; cp CentOSLogo.png
20:10 < giaco> koala_man: any idea how to compress a folder in the conventional way while using the C option? I am in . , files I have to compress are in foo/bar/*, I want the tar to not contain any "./"
20:10 < bls> kremator: so as long as you're not trying to use any of Red Hat's java or cloud stuff, it's going to be a 1:1 experience
20:11 < koala_man> giaco: tar cf file.tar foo/bar
20:11 < ses1984> still trying to troubleshoot the udev/hcitool thing, i think i learned a little more
20:11 < bls> kremator: at my last place, we used RHEL on production servers and CentOS on integration and dev
20:11 < ses1984> i tried to specify the bluetooth device using `hcitool -i hci0` and see  "Address family not supported by the protocol"
20:12 < giaco> koala_man: no, it would contain "foo/bar/file.txt", not "file.txt"
20:12 < koala_man> giaco: yes. do you still want "file.txt"?
20:12 < kremator> bls, oh pretty great!, why did you use centOS on integration and dev area? to save some bucks?
20:12 < koala_man> giaco: if so, tar cf file.tar -C foo/bar file.txt
20:13 < bls> kremator: mainly just so we didn't have to go through all the RHN stuff
20:13 < ses1984> i think that means the udev script doesn't have access to the bluetooth driver?
20:13 < kremator> RHN?
20:13 < koala_man> giaco: GNU tar also has filters to strip any prefix
20:14 < bls> kremator: Red Hat Network, their subscription management and update distribution service
20:14 < kremator> bls, are RH services that good? i mean, for a medium to big data company, is it worth?
20:17 < bls> kremator: depends on the field. we were doing lots of advanced NAS and SAN setups and RH would either fix of drive someone else to make bug fixes for us
20:19 < bls> like most stuff like that though, if you're not paying them enough, you're not going to get the same level of attention or traction
20:21 < quul> who bannt me from ##linux-overflow!?
20:22 < bls> heh, ask ##linux-overflow-ops
20:22 < bls> or is it ##linux-ops-overflow?
20:30 < bls> probably also helped that we were right down the road from them and could easily send someone over for face to face meetings if the need came up
20:33 < kremator> bls, wait, they literally send someone to the building?
20:33 < kremator> i though they only did remote support
20:36 < bls> it was remote/online, but it's one of those things where you interact differently with people next door than across the country/on a different continent
20:51 < mawk> quul: I guess it has a channel ban for ##linux
20:51 < mawk> can't be on both
20:52 < quul> who comes up with this stuff
20:52 < jim> hmm?
20:52 < candidat>  hmmmMMMMmm
20:54 < jim> oh, so the whole question is about ##linux-overflow?
20:55 < ayecee> the plot thickens
20:56 < quul> and then it just redirects here
20:56 < quul> laaame
20:56 < jim> the basic deal as I understand it, is overflow gets more people when the channel has a +l (which I don't remember it having)... we ask that if you get in there, you don't just idle there
20:57 < quul> why not this channel is too boring
20:58 < ayecee> also i keep hearing this whining sound
20:58 < quul> me too, every time someone mis-spells a word, or uses incorrect punctuation
20:58 < jim> you don't have to be here if you don't wanna... and you don't have to listen here actively if you donm'/t want
20:58 < quul> seems to originate from a select few users
20:58 < kremator> guys, in case the end user does have a decent internet connection  (let's say 100 Mbps) is there any advantage of using a geographically local mirror of repositories instead of using the main one in USA/Canada??
20:59 < ayecee> kremator: distributes the load among the mirrors
21:00 < kremator> ayecee, so it's more like a technique to benefit the project's daily expenses right?
21:01 < jim> ##linux isn';t meant as entertainment, so if you're bored and you don't need anything or want to discuss linux stuff, you don't have to listen to the channel, you can pretty much come and go as you please
21:01 < ayecee> sure, something like that
21:02 < kremator> jim, what if one gets entertained by DE fights and systemD protest/riots?
21:02 < jim> kremator, a geographically closer mirror matters less than if the mirror is well connected
21:02 < bls> a 100 Mbps connection in timbuktu is still going to be faster downloading from a local mirror than from the US
21:03 < jim> then cook some popcorn
21:03 < ayecee> better than running traffic over transocean links if you don't need to
21:03 < dgurney> kremator, was the wrong capitalization of systemd intentional? :p
21:04 < kremator> dgurney, not really, i have seen plenty of it with the D capitalize
21:04 < kremator> capitalized*
21:04 < dgurney> well that's wrong
21:04 < dgurney> it's systemd
21:05 < jim> kremator, and having said "cook popcorn", I'll request that you don't add to them when they occur, and I'll be trying to calm folks down so it gets quieter
21:05 < kremator> jim, dont worry, you cannot join to a DE fight if you use no DE
21:06 < dgurney> WM fights can also get vicious
21:06 < bls> especially with the anti-tiling crowd
21:07 < kremator> dgurney, let me rephrase: you cannot join to a DE fight if you dont have x11 installed to begin with
21:07 < dgurney> oh, but then you only have "core" stuff to fight about
21:07 < kremator> dgurney, exactly, i'd join in a fight for minimalism purist
21:07 < ayecee> kremator: pfft. of course you can.
21:08 < kremator> ayecee, by using wayland lol?
21:08 < bls> then you can get into the "why can't I display this unicode character/use TTF/copy-paste/browse the web? linux sucks!" argument
21:08 < ayecee> you don't have to _use_ something to have an opinion about it
21:09 < kremator> bls, well, fi you are usign a virtual terminal, you know better that you could browse the web with links or w3m
21:10 < bls> kremator: right, the usual whinge is why isn't there a firefox/chrome/modern/full JS stack browser equivalent
21:10 < quul> wayland is minimalism? heh
21:10 < lukey> You can even watch videos as mpv etc. support direct output to framebuffer
21:11 < phogg> or with libcaca
21:11 < bls> yeah, the framebuffer is cheating, *true* purists are ASCII only
21:12 < dgurney> true purists shun every modern comfort!
21:12  * bls puts on his goat hair shirt
21:12 < kremator> bls, i agree, UTF-8 are only for those dudes that doesnt care about all the memory wasted everytime they reserve a character var!
21:13 < lukey> True purists connect via Teletype to the Serial-port https://www.youtube.com/watch?v=X9ctLFYSDfQ
21:13 < quul> nah that's utf-32
21:13 < phogg> UTF8 does not waste any memory as long as you stick with a strict subset.
21:13 < kremator> lukey, no serial port here sadly :/
21:13 < vook> true purists only use a 300 baud TRS-80 thermal transfer data-terminal
21:14 < phogg> The important thing isn't to use multibyte characters, the important thing is to be sure that if you do you're not using something other than UTF8
21:14 < bls> what are you talking about? I *love* using the swedish subset of latin-1 for everything
21:14 < ayecee> bork bork bork
21:15 < phogg> bls: please, maybe if you'd said Shift JS I would have believed you
21:15 < phogg> s/JS/JIS/
21:15 < TRS-80> vook you rang?!
21:15 < TRS-80> :)
21:15 < bls> nah, I reserve my CJK issues for the weekends
21:19 < sy> Hello
21:20 < TRS-80> hello sy
21:20 < sy> Is anyone aware of a system that lets you have the same media on multiple devices, and track watched videos and sync playlists among them?
21:20 < giaco> koala_man: tar cf file.tar -C foo/bar file.txt still keep "./file.txt"
21:20 < sy> read: netflix minus drm and self hosted
21:20 < NetTerminalGene> does kernel mitigation of spectre and meltdown is only effective if you installed new microcode?
21:20 < sy> i'm currently using Kodi but it's pretty slow
21:21 < TRS-80> sy Kodi with centralized database
21:21 < sy> That's what I have
21:21 < sy> with a central mysql database(s)
21:21 < giaco> koala_man: also cd "$path" && find . -type d -print0 | xargs -0 tar cvf "$outputdir/"$outputname".tar generates "./filename.txt"
21:21 < TRS-80> if it's slow, it's due to network and/or hardware issues, not Kodi; but wait: are you transcoding?
21:22 < giaco> I am ranting. Tar is plain stupid
21:22 < ananke> sy: plex
21:23 < TRS-80> ananke plex usecase is only for transcoding, hence my question
21:23 < Elec_A> I get many data drops, how can I know the reason ? this is the output of ifconfig : https://pastebin.com/Dziffhv0
21:23 < ananke> TRS-80: it fits presented requirement,s not just transcoding
21:23 < sy> TRS-80: transcoding?
21:23 < TRS-80> it's muvh less open than kodi, much much less
21:23 < ananke> open wasn't a requirement
21:24 < TRS-80> sy yes downscaling to stream to for example mobile
21:24 < sy> i'm not sure
21:24 < TRS-80> ananke it's a requirement for my support, and many others in this channel
21:25 < ananke> TRS-80: isn't that fancy
21:25 < TRS-80> sy what are the receiving clients? PCs? RPis? tablets?
21:26 < sy> TRS-80: tablets, phones and desktops
21:26 < twainwek> Elec_A: is this recent?
21:26 < xamithan> trakt has plugins for most kodi plugins and also plex
21:26 < xamithan> It'll sync watched and lists
21:26 < Elec_A> twainwek: yes, I'm generating the packets using a FPGA board, the data transmission rate is 100Gbits/s .
21:27 < bls> Elec_A: tried getting diagnostics with ethtool?
21:27 < Dagmar> You're generating them incorrectly then.  Look at them carefully
21:28 < TRS-80> sy I'd look into tuning whatever your performance issues are first, as kodi does what you want. Check your hardware and netwok setup and consult Kodi performance docs. We run Kodi here and it's great.
21:28 < bls> oh, so it could be something other than a hardware problem
21:28 < Elec_A> bls: I have executed ethtool -S eth2, and it only shows some statistics. are there other type of output ?
21:28 < Dagmar> Elec_A: Doesn't really matter.  That "dropped" category should generally be _zero_
21:28 < sy> TRS-80: to be fair picking a server a couple thousand miles away would probably do it
21:28 < Elec_A> Dagmar: which is not.
21:29 < sy> TRS-80: is the mysql database big at all? if not I can simply stick that on a closer server
21:29 < sy> i just went with the maximum storage i could get for an ok price
21:29 < TRS-80> sy your server not at home? We have ours running on a Cubietruck and it works great
21:29 < Dagmar> Elec_A: I suspect perhaps you're not filling in the IP field properly when you're generating the packets, or basically just firing malformed packets at the MAC address
21:29 < bls> Elec_A: ethtool can do a ton of things, but I've had to research them the few times I've needed it
21:30 < strykerawb> does anyone here know how to mix microphone with computer sounds when doing something like skype, google hangouts, an online radio show, etc etc?   My mic seems to pick up stuff from the speaker, but not so well.  I'd like to be able to play random sound files while on these calls
21:30 < sy> TRS-80: nope, i have a server a few countries over
21:30 < bls> wireshark somewhere else should show if you're generating invalid packets
21:30 < strykerawb> im not sure if i need a software or hardware solution for this
21:30 < sy> i don't have any home servers
21:30 < TRS-80> sy tthe media streaming is what takes the bandwidth, not the small database
21:30 < sy> althrough i guess i could repurpose this rpi
21:30 < Elec_A> Dagmar: I recieve the packets with "tcpdump" and "wireshark" and it shows the correct format. like source and destincation IP and mac addres.
21:30 < Dagmar> strykerawb: There's ways to do it with ALSA.  I'm not saying they're _pleasant_ ways, but there are ways.
21:30 < sy> TRS-80: but isnt the database what you access when you flip through pages
21:30 < sy> that's the slow bit
21:30 < TRS-80> sy RPi will not work well as a server, only as a front end
21:30 < sy> streaming is great
21:31 < TRS-80> sy I gotcha, well could be then yes
21:31 < xamithan> a few countries over?...
21:31 < sy> TRS-80: by server do you just mean video storage? or is there a kodi server?
21:31 < xamithan> Isn't that laggy
21:31 < Elec_A> bls: yes, wireshark shows the data correctly, the size of the packet correctly. the ethernet, Ipv4 and UDP headers seem to be fine as well .
21:31 < sy> xamithan: it's high latency but the link speed is actually quite good
21:31 < TRS-80> yes xamithan I think that's the problem
21:31 < sy> once you get going, that is
21:31 < Dagmar> Does it show the correct number of packets recieved?
21:31 < Elec_A> Dagmar: yes.
21:32 < xamithan> Yeah but you have to push data through that server back to you so it makes it worse
21:32 < strykerawb> hrmm ok guess i'll see what i can do there.
21:32 < sy> xamithan: huh?
21:32 < TRS-80> sy you willing to spring few bucks for some hardware for better performance?
21:33 < sy> TRS-80: i can't run things overnight here
21:33 < sy> only the one room
21:33 < Elec_A> Dagmar: bls isn't it because of NIC ring buffer filling up fast?
21:33 < TRS-80> SBCs are silent
21:33 < sy> TRS-80: why do I need something beefier? transcoding or network throughput?
21:34 < sy> because i doubt i'll ever be in a place with more than 100 mbits which is what the rpi does
21:34 < Dagmar> Elec_A: It could be a number of things.  See also https://unix.stackexchange.com/questions/205141/what-exactly-is-an-ifconfig-dropped-rx-packet
21:35 < Elec_A> this is the output of dropwatch : https://pastebin.com/E54Xc3P9
21:35 < TRS-80> sy neither, sound like network latency for the UI in your case. but RPi does not make a good server because it shares bus between USB where drive attaches and internet
21:35 < sy> ah
21:35 < Dagmar> Elec_A: Have you changed a sysctl related to udp buffer sizes?
21:36 < TRS-80> sy Cubietruck on other hand has SATA connector and does not share bus
21:36 < M6HZ> ayecee, I've tried with another computer (windows), and it worked.
21:36 < Elec_A> Dagmar: I have not, but I totally appreciate if you could explain more. I might have to change it to make it work.
21:36 < bls> is this FPGA interrupt driven? what's it's buffer size? could you have issues there?
21:36 < TRS-80> sy CT also has no HDMI out, making it useless as a front end. different hardware for different purposes
21:37 < Dagmar> Elec_A: They're the problem.  The kernel reserves a buffer to store packets in until a process has a chance to retrieve them
21:37 < Elec_A> bls: The packet size is 512bits and I believe I have packet pause frames correctly.
21:37 < Dagmar> Elec_A: The default was set when Fast Ethernet was new
21:37 < Dagmar> Elec_A: It is *not* suitable for gigabit ethernet
21:37 < TRS-80> Dagmar: lol
21:37 < FrozenFire[alt]> Is there a way, with socat, using UNIX-CONNECT to kill the process when the target socket disappears?
21:38 < sy> TRS-80: the isea of my project was to find a system to replace netflix and save me a few bucks each month
21:38 < Dagmar> Elec_A: USually this comes up when people are running a large monitoring solution and a crapton of SNMP tests mysteriously fail all at once
21:38 < Elec_A> Dagmar: Interesting.
21:38 < Dagmar> Elec_A: Two jumbo frames can basically overload it.  See also https://access.redhat.com/documentation/en-US/JBoss_Enterprise_Web_Platform/5/html/Administration_And_Configuration_Guide/jgroups-perf-udpbuffer.html
21:38 < sy> so i'm leasing a server which has pretty good links to everywhere i want to watch from
21:38 < sy> my home connection is like 5 up so i don't really want to set it up there
21:38 < FrozenFire[alt]> Currently I'm finding that it is persisting when the target socket is gone, and connections simply fail when connecting on the UNIX-LISTEN it communicates with
21:38 < Dagmar> Elec_A: Basically, 131k isn't going to cut it for the buffer size.  Jack it up by 10 for starters
21:39 < Elec_A> Dagmar: Should I install "real-time" kernels? how can I fix it ?
21:39 < Dagmar> Elec_A: You just need to make the buffer larger is all
21:39 < TRS-80> sy this is exactly what we do. plus now we have a Linux server I also set up a NextCloud instance to self host/sync our family contacts, calendar, files, etc.
21:39 < Elec_A> Dagmar: Okay, this is a new concept for me. How can I increase the buffer size?
21:39 < Elec_A> What's the name of the buffer
21:39 < TRS-80> sy are clients usually outside your home network when streaming?
21:40 < sy> TRS-80: I use EteSync for that (calendar and contacts only)
21:40 < Dagmar> Elec_A: No matter how fast your code is it's only likely to get CPU time when there's a "tick" available for it, so if the buffer fills before it gets another tick to drain the buffer, packets get overwritten
21:40 < Dagmar> Elec_A: I jsut linked you the RHEL documentation ont he matter
21:40 < sy> TRS-80: yeah, I hardly ever stream at home because I have local media mostly
21:40 < Elec_A> Dagmar: So you mean this is CPU's issue, do you think DMA of GPU card would solve the issue?
21:40 < Dagmar> Elec_A: Just read it and do what it says.  It should at least radically reduce the dropped packet count
21:41 < Dagmar> Elec_A: Are you even reading the words I'm typing?  I just told you what it was.
21:41 < Elec_A> Dagmar: Absolutely. Thank you so much.
21:41 < sy> TRS-80: i'm still a bit fuzzy on how I want this to work
21:41 < sy> ideally i would run it from home
21:41 < Dagmar> It's not a CPU issue or a DMA issue.  It's simply a matter of the buffer packets go into from the NIC being set to a value that was appropriate in 1995 and it's now 2018.
21:41 < Elec_A> Dagmar: I'm just throwing some ideas, I will defintiely read that RH page.
21:41 < sy> but 5 up will just make it a nightmare to do anything
21:41 < sy> and if you have multiple clients then rip
21:42 < Elec_A> Dagmar: Sure, Thank you so much.
21:42 < TRS-80> sy yeah because to me "streaming" means streaming local downloaded content, but your last reply makes me  think it means something different to you
21:42 < Dagmar> Elec_A: I ain't some random guy guesses.  This is definitely a buffer that needs to be made larger once gig-E is involved
21:42 < bls> and a lot FPGA solutions I've seen for this are more for dev work/testing/toying around and can't handle any serious workloads
21:42 < Dagmar> s/gusses/guessing/
21:43 < sy> TRS-80: so with a server at home streaming from home is over lan so my shitty network speed isn't a bottleneck
21:43 < TRS-80> sy but then again whatever server you have must cost the same as Netflix?
21:43 < TRS-80> sy yes precisely
21:43 < sy> but from eg work where i usually do most of my streaming dont judge me, my home upload would be a huge bottleneck
21:43 < sy> TRS-80: https://www.time4vps.eu/storage-servers/
21:44 < TRS-80> sy  not judging, just trying to get the use case nailed down in order to make a proper recommendation
21:44 < sy> don't judge me for streaming at work* :)
21:44 < TRS-80> if you can get away with streaming at work, more power to you, lol
21:44 < TRS-80> :)
21:44 < sy> we have 100megs there so
21:44 < stevendale> Hi o/
21:45 < TRS-80> hi stevendale
21:45 < stevendale> I have 730 K/s down and 35 K/s up
21:45 < stevendale> sy
21:45 < sy> oh wow i used to have something like that
21:45 < stevendale> How are you TRS-80?
21:45 < sy> but they put new cables in or something
21:45 < TRS-80> sy can you hide a Cubietruck in a closet there and connect it to network? LOL
21:45 < stevendale> sy I haven't gotten the new cables yet unfortunately
21:45 < sy> TRS-80: yes
21:46 < sy> you're a genius
21:46 < sy> we actually have nuge NASes at work
21:46 < sy> huge compared to a sbc anyway
21:46 < TRS-80> sy I mean it only uses like $20 per year electricity, you could probably get permission maybe
21:47 < TRS-80> or were you thinking using their hardware too? lol
21:47 < sy> yeah lol
21:47 < sy> colocation @ work
21:47 < sy> wwell i can think of one problem and it has 3 letters
21:48 < sy> and rhymes with bat
21:48 < kremator> guys, in debian 9, when i try to run a VM in virtualbox, it trows the error about "not having virtualbox dkms" but i do have it installed, and i already did the "modprobe vboxdrv" but it trows the error http://dpaste.com/3ZP9KQ0
21:48 < TRS-80> sy well that's up to you and whatever sort of relationship you have at your work
21:48 < sy> TRS-80: you have a setup for music streaming also?
21:49 < sy> TRS-80: I think I will just mirror my server onto a local hard drive
21:49 < sy> However doesn't local media also take up space in the databases along with network content (kodi)
21:50 < sy> or could I exclude it?
21:50 < sy> so, local media i would only have at home
21:50 < Alina-malina> https://pastebin.com/raw/Za6iyAWM   someone ban this idiot from here please....
21:50 < sy> Hm. I could just set /etc/hosts to make my home machine connect to itself for that and still have all the buffs of resuming playback...
21:51 < twainwek> Alina-malina: there's an ignore feature
21:51 < sy> there's been an ignore feature for several thousand years, people just need to use it
21:51 < Alina-malina> or at least dont provide any help.
21:52 < Alina-malina> this guy is under names of Neo4 and disturbing people for years on these channels, so i want to warn people so they not help this idiot, hes pming bullshit to peole..
21:53 < TRS-80> sy for music I have my music files in a folder in NextCloud, which syncs to whichever devices I want (phone, desktop, laptop, but not tablet). In addition to that I use Pandora with an Exposed module on Android to block ads and allow unlimited skips. And sometimes YouTube streamed through a client like WebTube which is available in F-Droid and allows "watching" (listening) to a video with screen off and also over TOR.
21:53 < bls> Alina-malina: we can't do anything other than say, yeah, we know. #linux-ops should be able to help you though
21:53 < jml2> kremator, apt-get --reinstall install <vbox package>  can fix this
21:54 < Alina-malina> bls, just want to warn people to not provide any help to this guy, hes harasing people in PMs under different nicknames for many years, hes also know under Neo4 nickname
21:54 < Alina-malina> jus sayin
21:54 < sy> TRS-80: Neat. I was considering using MPD somehow, along with MALP (fdroid) on my phone along with some kind of magical streaming
21:54 < jml2> Alina-malina, #freenode
21:55 < stevendale> jml2 Alina tried
21:55 < TRS-80> sy yeah there is tons of software, depending on your use case(s). that's what's great about linux. there is also Syncthing if you want to sync all your devices without needing to run a centralized server like I do.
21:55 < jml2> stevendale, #freenode
21:55 < stevendale> They said /ignore too, jml2
21:55 < jml2> use /ignore
21:55 < jml2> so ues it
21:55 < stevendale> This is a network issue
21:55 < jml2> use it
21:55 < stevendale> Are you an idiot
21:55 < jml2> stevendale, #freenode
21:55 < kremator> jml2, it didnt, but i found the problem
21:55 < stevendale> jml2, ##linux
21:55 < kremator> jml2, i actually didnt had the headers installed
21:55 < stevendale> jml2, ##linux
21:55 < stevendale> I can do that too
21:56 < ayecee> stevendale: please stop
21:56 < jml2> kremator, ok
21:56 < jml2> kremator, tends to happen when you upgrade kernels
21:56 < Elec_A> Dagmar: increasing the buffer size didn't help reducing the packets drop.
21:56 < jml2> kremator, there's another issue often with vbox that you need to add yourself to the vbox user group
21:57 < jml2> kremator, (group "vboxusers" in debian)
21:57 < TRS-80> jml2: I just ate a bowl of upgraded kernels, I upgraded them through the microwave into popcorn XD
21:57 < TRS-80> er 2400MHz device
21:57 < TRS-80> XD
21:57 < kremator> jml2, really? even if i execute vbox as root?
21:58 < jml2> kremator, i dont recall what -- maybe it was due to me using vbox "extensions" (this is not the guest additions)
21:58 < kremator> TRS-80, dude if you kernel works at 2400Mhz to make popcorn, its slow af
21:58 < jml2> kremator, but that had to be used in order to use vbox properly
21:58 < kremator> in this age microwave ovens works at 5 Ghz at least, go big or go home
21:58 < phogg> kremator: that's really not true
21:58 < kremator> jml2, oh yeah, i know which extension you are refering but i do not use them
21:59 < jml2> kremator, it's an extension that allows to use physical usb3 with vbox guests
21:59 < kremator> jml2, yeah
21:59 < jml2> kremator, likely you will never use it
22:00 < arooni> is it ok to have directories in the path variable that dont actually exist?
22:00 < Elec_A> Actually, I found something insteresting. I opened the port and read the data through low-lever raw socket C programming and droped packets reduced from 200000/s to 140000/s
22:00 < bls> arooni: yes
22:00 < jml2> kremator, btw you shouldn't be use root in order to use vbox properly .. add yourself to the vboxusers group
22:00 < arooni> ok cool
22:00 < bls> arooni: it's inefficient in an OCD kind of way, but won't hurt anything
22:00 < kremator> jml2, i will do it, it was jut an hypotethical solution, because in my home ubuntu, it works flawlessly
22:00 < jml2> arooni, no it wouldn't create damage?
22:01 < arooni> bls: i just am copying config across ubuntu + mac ; i need to add a few paths for gnu utils on mac that dont exist on linux
22:01 < phogg> Elec_A: how were you reading them before?
22:01 < Elec_A> phogg: dropwatch, tcpdump and wireshark
22:01 < jml2> arooni, "help test" -- shows you can test a string to see if it matches an actual file or directory
22:02 < phogg> jml2: I don't think he cares
22:02 < jml2> arooni, macs use ancient bash shells ... better practice to using things under vbox
22:02 < arooni> i use fish shell because i like to be weird
22:02 < FreeFull> Because Apple didn't like the GPLv3
22:02 < FreeFull> So they use really old stuff
22:02 < FreeFull> arooni: I like fish
22:02 < jml2> arooni, #macos :)
22:02 < jml2> lol
22:02 < jml2> phogg, right.
22:03 < FreeFull> Much nicer for scripting than posix shells
22:03 < arooni> im chatting on an ubuntu laptop; i think i can be here :P lol jml2
22:03 < phogg> arooni: what do you like about fish?
22:03 < jml2> an non-existing path in $VAR is "good enough" to ask about.
22:03 < THE_GFR|WORK> hey everyone have a SAS HBA in one of my systems running Linux Mint.  If I plug in drive to the external SAS to SATA cable too many times they stop showing in Linux Mint Disk Utility.  The only way to get them to show back up in disk utility is to reboot the system.  The drives show in the MegaRaid Storage Manager though, but they do not have a device ID aka /dev/sdc or something similar ideas?
22:03 < arooni> phogg: its like the hipster shell; just makes me feel happy.  also like the command completion
22:03 < jml2> arooni, but you use a mac
22:03 < arooni> jml2: i have a mac and and ubuntu laptop
22:04 < jml2> arooni, nope
22:04 < phogg> THE_GFR|WORK: do you stop seeing them appear in kern.log, too?
22:04 < jml2> arooni, you use mac things.
22:04 < arooni> hahahah
22:04 < arooni> i must be evil then
22:04 < jml2> arooni, you're an Apple boy
22:04 < arooni> lol no;  i bought this mini used on craigslist and upgraded the memory and ssd because paying the apple tax for hardware is dumb
22:04 < phogg> jml2: lots of people buy Apple hardware to run Linux. It's nice hardware.
22:05 < jml2> arooni, did you fix your bib? keep biting that apple
22:05 < stevendale> OpenSSL got an update on Debian Unstable
22:05 < phogg> THE_GFR|WORK: I ask because it could be that only udev is the issue.
22:05 < stevendale> Make sure to upgrade
22:06 < phogg> stevendale: and how many people pull updates from sid?
22:06 < phogg> you should wait for the version packaged against the version of Debian you actually run, not sid
22:06 < jml2> upgrade on debian unstable. great advice for all of us to break our machines
22:06 < jml2> phogg, at least if he said "testing" he might make more sense
22:06 < stevendale> phogg I run Sid
22:07 < phogg> stevendale: that's cute
22:07 < jml2> stevendale, good keep sid to yourself
22:07 < bls> if it's a security issue, it'll be backported. if it's not, there's no reason to bother
22:07 < jml2> dont tell others to upgrade with debian unstable
22:07 < jml2> that's 100% wrong
22:07 < phogg> for a chance jml2 is entirely correct
22:07 < jml2> and idiotic.
22:07 < bls> phogg: shhhhh
22:08  * jml2 ignores stevendale 
22:08 < jml2> better for me.
22:08 < jml2> he was calling me an idiot for mentioning #freenode sooner.  they're just trolling.
22:09 < phogg> jml2: you could be less antagonistic yourself
22:09 < bls> it's like they were meant for each other
22:09 < jml2> lol .. not good to break all our machines.. that's bad enough XD
22:09 < stevendale> Almost as if it's love at first sight
22:10 < FrozenFire[alt]> Is there a way, with socat, using UNIX-CONNECT to kill the process when the target socket disappears? Currently I'm finding that it is persisting when the target socket is gone, and connections simply fail when connecting on the UNIX-LISTEN it communicates with
22:13 < jml2> FrozenFire[alt], there's a timeout with netcat when it received an eof ...
22:13 < THE_GFR|WORK> phogg: no, because they show up on the MegaRaid Storage Manager fine without issues
22:14 < THE_GFR|WORK> phogg: they are in the log file, but they show up fine if connected USB also
22:14 < THE_GFR|WORK> but just the SAS to SATA cable they stop showing up if I plug and unplug them too many times
22:14 < THE_GFR|WORK> or even once or something
22:14 < THE_GFR|WORK> so yea
22:15 < THE_GFR|WORK> phogg: would I just restart the udev service?
22:18 < phogg> THE_GFR|WORK: I'm not sure it would matter, but you could try stop/start on that the next time before resorting to a full reboot.
22:18 < phogg> theoretically udev won't notice anything different after it comes back up
22:18 < THE_GFR|WORK> phogg: how do I restart udev?
22:19 < THE_GFR|WORK> servicestop udev
22:19 < phogg> THE_GFR|WORK: /etc/init.d/udev restart # classical sysvinit
22:19 < THE_GFR|WORK> ah
22:19 < THE_GFR|WORK> ok
22:19 < phogg> or however you restart daemons on your distro
22:19 < THE_GFR|WORK> thx
22:19 < phogg> it's just another one
22:19 < phogg> you can even leave it stopped if you want... but some things do not expect this and become unhappy
22:20 < THE_GFR|WORK> phogg: restarting udev didn't fix the drive issue with it not showing in Disk Utility or having an id
22:20 < jnewt> how do i scan long receipts?   my scanner is Fujitsu Scansnap iX500, OS is ubuntu 16.04, it cuts off at 1 page on linux, works fine on windows, but just realized simple scan doesn't adjust to the size, it's always expecting 8.5x11
22:21 < phogg> THE_GFR|WORK: what do you mean by "not having an id"? And what is "Disk Utility"?
22:21 < phogg> THE_GFR|WORK: is there a device file in /dev? Is there a device dir in /sys?
22:22 < THE_GFR|WORK> phogg: the device isn't showing in /dev
22:22 < phogg> THE_GFR|WORK: then that is a udev problem of some kind. The primary job of udev is to notice device events and adjust /dev/ accordingly
22:23 < phogg> THE_GFR|WORK: you can do its work for it by hand in a pinch, same as we used to do before udev
22:23 < phogg> THE_GFR|WORK: or you can try figuring out what it is about device events that occur after the system has been online for a while that make udev's rules stop working
22:23 < THE_GFR|WORK> phogg: oh?
22:24 < THE_GFR|WORK> ok
22:24 < THE_GFR|WORK> phogg: that's what I am wondering but restarting udev should have caused it to work again but alas it did not
22:24 < phogg> THE_GFR|WORK: that's why I asked about the notice in kern.log. When the device appears the kernel knows, but something in the udev script that handles it breaks. What? Why?
22:25 < THE_GFR|WORK> phogg: not really sure because it's only specific to this HBA I have
22:25 < phogg> THE_GFR|WORK: right, udev only cares about *events*. There is also some kind of coldplug mechanism for changes that happen when the system is offline, but I don't know how that works since udev came about
22:25 < phogg> THE_GFR|WORK: yeah that's pretty weird
22:25 < THE_GFR|WORK> phogg: because USB stuff works fine specific to drives being connected or disconnected
22:26 < THE_GFR|WORK> phogg: but also specific to disk utility and /dev because the HBA shows the drive as attached in the MegaRaid Storage Manager
22:26 < phogg> THE_GFR|WORK: what I would try, if it were me, is making the correct device files by hand after it goes wrong. If I can and the device still works then I know that udev should be handling it. If it's not possible then the fault is on the HBA side
22:26 < bls> some of the q-logic stuff I worked with essentially required removing and readding the kernel module any time a LUN changed state
22:26 < THE_GFR|WORK> phogg: I would guess it's on the HBAs side
22:26 < phogg> THE_GFR|WORK: I am much more inclined to think the hba is the problem, but you never know
22:27 < THE_GFR|WORK> phogg: if I wait long enough it starts working again most of the time...
22:27 < THE_GFR|WORK> but that could take hours or days depending
22:27 < phogg> bls: I suppose a module reload would fix this, too... but who wants to umount the storage for that?
22:28 < bls> agreed it's a pain, but that was the most reliabe way we found to keep things sane
22:28 < phogg> The most reliable way is "Don't buy qlogic"
22:28 < THE_GFR|WORK> lol
22:28 < phogg> if you have the option
22:29 < THE_GFR|WORK> phogg: my HBA is an LSI aka Avago aka Broadcom
22:29 < bls> heh, yeah, this was nursing a set of legacy system through the later stages of their deployment
22:31 < bls> the sloppy cable pulls to the behemoth brocade silkworm didn't help much either
22:38 < THE_GFR|WORK> phogg: thanks for the help I'll have to just reboot since I am not sure at the moment what's going on and no time now... ran out.
22:39 < phogg> THE_GFR|WORK: that'll happen
22:39 < kubast2> Hey how can I calculate a CVT reduced for refresh rates other than 60hz like I can with nvidia settings on windows ?
22:40 < bls> kubast2: looked at the cvt tool?
22:40 < kubast2> ERROR: Multiple of 60Hz refresh rate required for  reduced blanking.
22:40 < kubast2> but I think I understand how it works looking at it
22:41 < jml2> kubast2, omg windows?  cvt are fallback timings .. like vesa-compatible.... better you let the driver automatically detect edid information
22:45 < kubast2> jml2: I use cvt timings to squese more bandwidth over single-link DVI-D cables
22:45 < jml2> kubast2, in theory you should be able to force a timing but you need to be sure that timing is supported (you can check in windows of course -- but you'll need to find the tool to show you the timing info)
22:46 < jml2> kubast2, otherwise you could possibly break hardware with wrong timing -- so be careful..
22:50 < FrozenFire[alt]> If I have a sessionid, how can I find the processes in that session? I tried pgrep -s <sessionid>, and it returns nothing, despite /proc/<processid>/sessionid containing that value
22:52 < phogg> jml2: driver .inf files in windows embed that in plain text, I believe
23:02 < widp> how do clipboards work in linux?
23:02 < csan> Hi guys can anyone help? I get `No protocol specified\nUnable to init server: Could not connect: Connection refused\nCannot parse arguments: Cannot open display:` at every single GUI app I try and launch (terminals included). It happens since I mounted my old home folder, which is "old" because I just reinstalled my linux distro right after this exact same issue occurred... Anyone has a clue?
23:02 < widp> I have i3
23:02 < widp> i3wm, and the whole clipboard thing is a complete mess.
23:03 < widp> I use xsel it seems to work well only inside urxvt
23:03 < widp> but can't copy anything outside.
23:03 < Deele> Hello guys, I'm back with my shell script and being unable to launch separate process. Remember https://gist.github.com/Deele/36a0bc37f20de9cf91cb20ff051744cf ?
23:04 < Deele> at line 46 I added "&" at the end, so that program is deattached from this script, but I still get output in that same console and it seems that it still is acting the same as without &
23:04 < Elec_A> Hoe to disable optimization step (ip_rcv_finish) ?
23:07 < bls> Deele: not sure who was here, but do you just want to not see the output from the backgrounded command?
23:07 < Elec_A> found it : "$ sudo sysctl -w net.ipv4.ip_early_demux=0"
23:10 < Deele> bls, no, but The gist is - this is supposed to be able to split CRON job that goes every minute into a job that goes every second (if necessary), but every process is launched in a loop followed by sleep to delay next launch and it seems that it currently goes to next loop only after previous
23:13 < bls> I'm not understanding. Isn't that how loops are supposed to work?
23:14 < Sveta> Deele: your message may have cut off
23:14 < Sveta> Deele: the last words in it are 'only after previous'
23:14 < Deele> ah
23:14 < bls> also, have you looked at GNU parallel?
23:17 < Deele> so, the best case scenario is to launch with a script 30 commands with interval of 2 seconds so that they live their own life
23:20 < bls> parallel can launch a process with given instance count and interprocess delay
23:21 < bls> you should also read up on https://mywiki.wooledge.org/BashFAQ/028
23:21 < TheWild> hello
23:22 < TheWild> "tar xf whatever.tar" will reject files that have something funky in their path, like leading / or .. somewhere in the middle, right?
23:22 < koala_man> by default, yes
23:23 < koala_man> it'll strip such prefixes
23:25 < TheWild> oh, very good. I'm extracting something on /media/thewild/some-long-guid, and I was afraid I would mess up the stuff that's outside if something strange was in the tar. Thanks koala_man
23:25 < bls> it's actually pretty sane about things like (contrary to some strong opinions otherwise)
23:25 < prussian> was gonna say, you could try it and see. even if I intentionally add the / to the file it strips it. at least gnu tar does
23:26 < koala_man> strips when adding, strips when extracting
23:27 < TheWild> man tar - 'The GNU version of the tar archiving utility'. Yeah, that's the GNU one.
23:27 < prussian> tar -cf - --transform='s|^|/|' * | tar -tvf -
23:27 < za1b1tsu> Hello, anyone using vivaldi or brave?
23:27 < prussian> just so you can see
23:27 < prussian> the tvf one will emit tar: Removing leading `/' from member names
23:29 < compdoc> Ive been told Im very brave
23:29 < cfoch> hello
23:29 < jml2> TheWild, you can use "tf" to list the tarfile , -- if there's no top directory, you can use -C   to specify an empty target folder
23:29 < cfoch> what book do you recommend to start to contribute to the Linux kernel?
23:30 < jml2> cfoch, there's a kernel newbies site to help for that
23:30 < jml2> cfoch, https://kernelnewbies.org/  ---
23:30 < TheWild> oh f. I extracted TrueCrypt volume from the backup. Now what was the password? :(
23:30 < jml2> TheWild, the password is your birthday
23:31 < jml2> TheWild, joke XD
23:32 < TheWild> the file was named a.tc
23:32 < TheWild> but there's also a.txt
23:32 < TheWild> hmm...
23:33 < bls> cfoch: not a book, but there's also http://eudyptula-challenge.org seems to be on hold now, but something to come back to
23:34 < bls> cfoch: not sure how deep it goes into the kernel, but there's also the LPI book
23:37 < jim> echo Cyrum
23:38 < jim> Cyrum, if you wish to speak in this channel, you need to register your nick and authenticate using the resulting nickserv account
23:40 < phogg> yikes when did RAM get so expensive
23:42 < oiaohm> phogg: really ram is cheaper than it was 20 years ago.
23:45 < FrozenFire[alt]> If I have a sessionid, how can I find the processes in that session? I tried pgrep -s <sessionid>, and it returns nothing, despite /proc/<processid>/sessionid containing that value
23:49 < rcf> FrozenFire[alt]: in my case I have to run as root to see all the processes in the session
23:51 < rcf> ls
23:53 < kurahaupo> rcf: . ..
23:53 < Cyrum> Hola
23:53 < Cyrum> yay i can finally talk
23:53 < Cyrum> silly irc
23:54 < Cyrum> as for ram its all relative haha
--- Log closed Sat May 19 00:00:25 2018