--- Log opened Sat May 26 00:00:33 2018
00:07 < mawk> when trying to hack my LTE usb stick I discovered it's running linux
00:07 < mawk> digging further it's actually running some modified version of Android
00:07 < mawk> dirty dirty
00:08 < torncolours> Lots of weird stuff runs android
00:09 < torncolours> my dvd player does but thats pretty common
00:09 < mawk> yeah but it's a usb stick
00:09 < mawk> I'd have expected some light µC doing the usb work, a LTE module, that's all
00:09 < mawk> not a full linux OS
00:10 < ||JD||> dvd player? are we back in 2003?
00:13 < mawk> I don't want to brick the stick, that thing costs $50
00:17 < Some_Person> I wonder how hard it would be to set up something similar to rabb.it for personal use
00:17 < Some_Person> and not limited to just running Firefox
00:18 < Some_Person> For those unaware, rabb.it is a service intended for watching videos together remotely. It uses a virtualized Linux environment that is controlled using a browser which runs Firefox
00:20 < twainwek> but why
00:20 < mawk> why does it do that ?
00:20 < mawk> that seems useless
00:20 < Some_Person> mawk: how is it useless?
00:20 < mawk> for sharing a browser ok, for playing videos I'm sure you have simpler ways to share a screen
00:20 < acresearch> people is there a way to install ubuntu-GUI on a server and access it through SSH?
00:20 < mawk> than virtualizing linux to launch a browser in it
00:21 < infinisil> Some_Person: Damn that seems horribly overcomplicated for what it does
00:21 < mawk> I thought I was alone
00:21 < Some_Person> mawk: How would you do it?
00:22 < twainwek> acresearch: ssh with x forwarding
00:22 < infinisil> I wouldn't even use x forwarding, that's terribly slow
00:22 < mawk> x2go is pretty fast
00:22 < mawk> you could start from that for the screen sharing bit
00:23 < mawk> then for shared control of the computer, you need to explain a bit more how that would work
00:23 < acresearch> twainwek: hmm ok, but what about the actuall installation, i am using vultr, and i can only setup a machine by choosing an OS
00:23 < acresearch> twainwek: sorry i am doing this for the first time
00:23 < infinisil> Ah, shared control is a bit harder
00:23 < Some_Person> Rabbit does have shared control, it has an option to pass control to someone else
00:24 < mawk> it could be a simple fifo for people requesting control of the computer
00:24 < mawk> yeah
00:24 < infinisil> Oh how about this
00:24 < infinisil> mpv supports a messaging protocol
00:24 < Some_Person> Also keep in mind situations like shitty upload pipes
00:25 < infinisil> shitty upload pipes?
00:25 < infinisil> UDP for stream, TCP for control
00:25 < Some_Person> My upload speed at home is 5 Mbps... how well is that gonna work for sharing a screen directly?
00:26 < mawk> just try x2go Some_Person
00:26 < Some_Person> Honestly, I find Rabbit actually works pretty darn well
00:26 < mawk> it has pretty smart algorithms for fast remote screen view
00:27 < infinisil> Some_Person: Wait so the goal is to have like a shared X session pretty much, but only with a video player?
00:27 < ziggylazer> Some_Person, well
00:27 < Some_Person> infinisil: That's essentially what that service does, although I'm looking at possibly setting something else up for myself that's not so limited
00:27 < infinisil> In what ways is it limited?
00:28 < Some_Person> infinisil: Rabbit runs Firefox and only Firefox
00:28 < infinisil> And you want to run it in terminal?
00:28 < infinisil> Yeah that's reasonable to want
00:29 < Some_Person> Well, I'm just thinking of just setting up something similar on my dedicated server, since it has a good upload pipe
00:30 < acresearch> i am running ubuntu 18.04 and pip3 is not working, ImportError: cannot import name 'sysconfig'   help?
00:30 < Some_Person> Y'all mentioned x2go? I can definitely look into that
00:32 < triceratux> https://phys.org/news/2018-05-news-websites-blocked-eu-law.html
00:38 < Acheron> and ...?
00:40 < Acheron> something needed to be done and lot of companies just put off getting ready for compliance until the last minute as most do and stress their workers during crunch time or thought maybe it would be over turned
00:42 < Acheron> Even though the rules were officially adopted two years ago, with a grace period until now to adapt to them, companies have been slow to act, resulting in a last-minute scramble this week.
00:42 < Acheron> yup
00:42 < Acheron> been there, done that
00:42 < rascul> i'd rather block eu than deal with that ridiculous crap
00:42 < Acheron> at HP everything was done at the last minute
00:43 < rascul> except it doesn't matter because there's nothing the eu can do to me in the us if i don't comply
00:43 < Acheron> oh, please, congress is paid off so these companies can do as they please
00:46 < Acheron> Facebook lied and said they would apply the EU standard world wide but then went back on their word for US users
00:46 < Acheron> just blatant lying
00:47 < todkon> yo dudes, I wanna use imagemagick to convert both png and jpg files into one pdf, is there away to run something like "convert ./*/*png&./*/*jpg blah.pdf?"
00:47 < Acheron> so say what you want  i'm not libtard but i believe in Harvey Firestone's Motto:I believe fundamental honesty is the keystone of business.
00:48 < rascul> google and mozilla also lied
00:48 < rascul> facebook isn't alone
00:48 < Acheron> Capital isn't that important in business. Experience isn't that important. You can get both of these things. What is important is ideas.
00:49 < todkon> lol nevermind, just did &&
00:49 < Acheron> is doesn't matter what you do, lying is wrong whether you are in the US or the EU
00:49 < todkon> seems to be working
00:49 < Acheron> i'm done
00:50 < rascul> i agree completely about lying and honesty and stuff
01:00 < T-Rog> Could not start kdeinit5, check your installation
01:02 < kemisten> söker svenskar! pma om du fatttar vad jag skriver har en gåva åt just dig!!
01:03 < infinisil> T-Rog: ??
01:04 < T-Rog> infinisil: I just upgraded from an old ATI graphics card to a newer AMD graphics card. I was having some problems with an application and decided to install Mesa-gallium. I rebooted and now instead of a login screen I get "Could not start kdeinit5, check your installation."
01:04 < Acheron> what a shame about KDE
01:04 < koala_man> kemisten: isn't there a #linux.se or something?
01:05 < T-Rog> infinisil: I googled and it said uninstall freetype2. I did and rebooted and I get the same error message
01:05 < Acheron> there is a #KDE but just 700 people idling in there
01:05 < infinisil> T-Rog: You think uninstalling freetype after installing mesa-gallium would fix the problem you got by installing mesa-gallium??
01:05 < T-Rog> maybe?
01:06 < T-Rog> infinisil: I tried zypper remove Mesa-gallium and it said it was going to remove every package on my system
01:06 < T-Rog> so I said no
01:06 < T-Rog> now I don't know what to do
01:06 < infinisil> Okay I guess that isn't too far off the charts, but I'd start with reverting your system first
01:06 < T-Rog> how do I do that
01:06 < infinisil> No idea
01:06 < T-Rog> fuck
01:07 < infinisil> Do you at least know how to get a terminal?
01:07 < T-Rog> yes
01:07 < infinisil> Well that's something
01:07 < T-Rog> I'm sitting on one right now
01:07 < infinisil> Try running `undo`
01:07 < T-Rog> not a valid command
01:07 < infinisil> Well it was worth a try, I'm out of ideas
01:09  * infinisil thinks about advertising how easy it is to revert your broken system with NixOS, but refrains from it.
01:10 < rascul> iirc zypper can do fancy rollback type stuff if you use btrfs
01:12 < infinisil> NixOS creates a new bootable grub entry for every change to your system, so you can just boot into a previous configuration (but the state in the filesystems won't be touched)
01:12 < infinisil> Saved my machine a couple times already
01:15 < swift110> goodness that sounds pretty cool
01:16 < rascul> https://doc.opensuse.org/documentation/leap/reference/html/book.opensuse.reference/cha.snapper.html
01:17 < Acheron> hello swift110
01:17  * Acheron is waiting for the Mint Beta *
01:18 < swift110> how are you acetakwas
01:18 < swift110> how are you Acheron
01:18 < Acheron> doing quite well
01:18 < swift110> Acheron: I intend on putting a machine or two on Linux Mint 19
01:19 < Acheron> i have partitions open on two computers to dual boot 19 on
01:22 < swift110> oh really
01:22 < T-Rog> I fixed it by using snapper to roll back the changes
01:22 < swift110> thatds cool
01:22 < T-Rog> that was scary
01:23 < swift110> I have several laptops that will have several distros on them
01:32 < TRS-80> I'm having heck of a time trying to compile something using GNU autotools, anyone know anything about that? I been at it all day. :/
01:33 < rascul> TRS-80 lots of people know about autotools
01:33 < rascul> nobody knows if they can help you though until you explain your issue
01:33 < lnnb> i hate it when theres no configure file, or it's broken and you have to use autoreconf but it won't work unless you passs it specific commands.... uugggghhhhhh
01:34 < lone-wolf> Is possible run a software in monitor secundary after run a command in terminal?
01:34 < lone-wolf> For example, I run leafpad in terminal, i want that the leafpad open in monitor secundary.
01:35 < rascul> lone-wolf depends on how you have your monitor setup, you may be able to do 'DISPLAY=:0.1 leafpad'
01:39 < lone-wolf> rascul: In here show leafpad: Cannot open display:
01:40 < rascul> so you don't have your monitors setup in such a way where that would work
01:40 < TRS-80> https://bpaste.net/show/aac41c25700e
01:40 < TRS-80> :)
01:41 < rascul> TRS-80 could you put the output of configure in a pastebin?
01:41 < TRS-80> sure
01:42 < TRS-80> that's the full log actually, what I just posted ^
01:43 < rascul> yes, i'm aware of that
01:43 < TRS-80> more useful that the messages output from configure, I think, but if you want me to I will OK
01:43 < rascul> it's not immediately clear what configure choked on
01:44 < TRS-80> https://bpaste.net/show/e82e39f726da
01:44 < rascul> you need to install the fuse dev package(s)
01:44 < rascul> the last lines of configure make it quite obvious
01:45 < TRS-80> I found fuse.h on my system and pointed it to that but then it gives other error about compiler
01:45 < TRS-80> sorry I been round and round on this
01:45 < rascul> so then i need to see configure output when you told it where to find fuse
01:47 < TRS-80> OK I installed libfuse-dev, lemme try again
01:48 < TRS-80> OK progress, lol: configure: error: Encfs requires OpenSSL
01:48 < TRS-80> lemme guess openssl-dev? :)
01:48 < rascul> possibly
01:49 < TRS-80> now I do think that the version of openssl may be important but lemme have a look
01:49 < rascul> you are building encfs?
01:49 < TRS-80> yes
01:50 < rascul> is there not a package for it available?
01:50 < TRS-80> yes but package on sid is like 1.9 and I need 1.7.4 for compatibility with Android
01:51 < xamithan> better grab it manually
01:51 < TRS-80> did some testing earlier syncing files back and forth and it wouldn't work, wouldn't mount the filesystem
01:51 < TRS-80> xamithan: I couldn't seem to find it anywhere?
01:53 < lone-wolf> rascul: How do I configure my monitor?
01:54 < rascul> lone-wolf not sure, haven't done multiple monitors in awhile and don't really remember how to do it except with xrandr, and i don't think that gives you what you want
01:54 < TRS-80> lone-wolf: what DE are you using?
01:55 < lone-wolf> rascul: Whats DE? lxde?
01:55 < TRS-80> yeah Desktop Environment
01:55 < rascul> i think you need to use xinerama to use DISPLAY in such a manner but i don't recall for sure
01:56 < xamithan> There is a jessie package available so I'm not sure where you looked
01:56 < TRS-80> I setup 20 30 20 PLP on KDE Plasma fairly recently (1-1.5 yr ago?) and at that time I found a lot of info on xinerama and xrandr but it's all deprecated
01:56 < TRS-80> xamithan: but how to install when I am on sid?
01:56 < rascul> xrandr isn't deprecated
01:57 < xamithan> The same way you install anything else,  download it and apt install
01:57 < TRS-80> rascul: what I mean is, settting things manually with config files, etc
01:57 < rascul> that's not xrandr if you're using config files
01:57 < TRS-80> you should let it autodetect. basically, although we may be talking about 2 different things
01:58 < rascul> the problem is he wants to launch a program on a specific screen and afaik that's only doable with xinerama
01:59 < rascul> which probably requires manually configuring xorg.conf
01:59 < TRS-80> I know there are ways you can set it in KDE Plasma, not suggesting to change, just dunno how to do it in another DE
01:59 < rascul> that's not xinerama
02:00 < TRS-80> I know that
02:01 < TRS-80> I dunno, maybe you are right
02:01 < lone-wolf> E: Unable to locate package xinerama
02:01 < lone-wolf> sudo apt-get install xinerama
02:03 < rascul> xinerama kinda sucks compared to randr, it's potentially difficult, requires manual configuration of an archaic configuration file, and is probably not worth it just to save the effort of dragging a window to another monitor
02:03 < lone-wolf> rascul: Ok, thank you.
02:04 < rascul> i could of course be wrong about some of these things though, it's been awhile since this knowledge was fresh in my mind
02:04 < TRS-80> xamithan: dang man I never noticed you could get a deb file from here, only ever found the links for sources: https://packages.debian.org/jessie/encfs wow I feel like a real idiot now
02:05 < TRS-80> no rascul that's kind of the sense I got as well, xinerama almost deprecated; IIRC they just about say so on their page
02:05 < xamithan> Yeah just got to worry about the dependecies
02:06 < TRS-80> xamithan: I'm so far into this now, I almost feel like I have to finish now DAMOT X|
02:07 < ldiamond_> Anyone here ever tried ptunnel to get internet access on a cruise ship?
02:07 < rascul> why would you need ptunnel?
02:08 < ldiamond_> if there's wifi but they block everything except they forget to block icmp
02:08 < infinisil> They block *everything*?
02:08 < infinisil> Why would they provide wifi then?
02:08 < ldiamond_> until you pay
02:09 < rascul> then pay
02:09 < rascul> we're not going to knowingly help you illegally obtain access to anything
02:09 < ldiamond_> Pretty sure that's not illegal.
02:10 < infinisil> It depends on their TOS :)
02:10 < ldiamond_> no, cause you don't go through their TOS
02:10 < rascul> maybe not when you're in the middle of the ocean outside of any country's jurisdiction...
02:10 < infinisil> ldiamond_: Huh?
02:10 < ldiamond_> even if you're in a country. They let icmp through, you use icmp.
02:11 < ldiamond_> their TOS is when you go on the wifi, accept the charge, accept TOS
02:11 < infinisil> You most likely agreed to a TOS by buying this trip
02:11 < ldiamond_> if you just have a wifi connection and use an icmp tunnel you never get their tos or accept it
02:11 < ldiamond_> not for the wifi
02:11 < infinisil> Well whatever, but if you wanna use ptunnel, just try it
02:12 < ldiamond_> well I'm wondering if anyone have tried it before I setup a ptunnel server
02:15 < jpleau> hi. https://gist.github.com/jpleau/5b16a39a0fa8f9c45c82de0deb7967f1 why is "w / who" telling me there are 6 users, while only showing me one in the list?
02:21 < Whoroo> for i in Contents/; do convert $i -resize 1080 -quality 90 $i.res.jpg; done
02:21 < Whoroo> does this look correct for a batch resize on a directory?
02:22 < infinisil> Whoroo: file "Contents/" not found
02:22 < mattfly> does anyone here have nvidia cards on a debian based distro?
02:22 < mattfly> and is able to hibernate on the kernel 4.15
02:22 < mattfly> ?
02:23 < Whoroo> infinisil, yeah it's been failing
02:23 < Whoroo> if I'm in the parent directory at the time it should be ok?
02:24 < mattfly> I am having an issue with ubuntu bionic that isnt nothing common, after installing the nvidia driver 390, on all of my hibernation attempts i get stuck ad s2disk message "snapshooting system", then the fans run faster for a while and im forced to force shutdown (alt+print REISUB works)
02:24 < mattfly> ubuntu uses uswsusp and s2disk
02:24 < mattfly> have anyone ever seen something similar?
02:25 < infinisil> Whoroo: You probably want `for i in Contents/*; ...`
02:25 < mnrmnaugh> Whoroo for i in Contents/*; do convert "$i" -resize 1080 -quality 90 "$i".res.jpg; done
02:25 < mattfly> and is there any alternative way i can hibernate?
02:25 < Whoroo> ok thanks
02:26 < mattfly> I suspect is a problem with the meltdown patches but I have more issues when i try to use the newest kernel, specially that the nvidia drivers dont work then
02:27 < mnrmnaugh> Whoroo https://mywiki.wooledge.org/glob
02:27 < mattfly> here is my bug report to ubuntu team... https://bugs.launchpad.net/ubuntu/+source/uswsusp/+bug/1770491
02:28 < Whoroo> convert-im6.q16: no images defined `Contents/IMG_2342.CR2.res.jpg' @ error/convert.c/ConvertImageCommand/3258.
02:28 < Whoroo> still fails for me atm
02:29 < mnrmnaugh> maybe the convert command is wrong. i dont know convert at all
02:30 < Whoroo> it's worked under debian 8, this is debian 9, but been a while too
02:35 < afidegnum> hello, i m having issue, i trying to mount a raid disk to /home but how do i find out if they are mounted as expected?
02:35 < tds> afidegnum: you can run "mount" to see what fses are mounted where
02:36 < afidegnum> ok
02:37 < afidegnum> tds: it seems it hasn't been mounted, https://ghostbin.com/paste/v78q8
02:40 < tds> did you say you were trying to mount it to /home? there's a mdadm raid device there, but it's mounted to /
02:40 < jsgrant> Cleaned out this old 'Samsung Chronos 7' Laptop today and dual-vents were dustier than Father Time ...
02:40 < afidegnum> ok, I have RAID SSD if i can remember and RAID HDD
02:40 < afidegnum> so i wanted to mount RAID HDD to /home
02:41 < afidegnum> so / will be left for the SSD to have system files,
02:42 < afidegnum> tds: how do i see all the raid disks ?
02:43 < afidegnum> and where they are mounted
02:44 < tds> do you mean which arrays mdadm is exposing as block devices, rather than mounted?
02:44 < tds> since that last command's output shows you the mounted ones
02:45 < tds> I can never remember the mdadm commands, but I think looking in /etc/mdadm.conf and /proc/mdstat may get you somewhere
02:46 < afidegnum> yes, listing raid arrays
02:54 < syb0rg> afidegnum, use lsblk. I'm pretty sure the raid devices don't have to actually be mounted to show up there, so long as they are mapped.
02:56 < afidegnum> syb0rg: ok, thanks,
03:00 < afidegnum> syb0rg: here is my raid structure, how do i /4
03:01 < syb0rg> I don't know, I am not experienced in /4'ing
03:01 < syb0rg> =P
03:02 < Windoa> What editor do you use?
03:02 < Windoa> I use vim
03:02 < syb0rg> you aren't fooling anybody, we know what you're doing Windoa
03:02 < syb0rg> >:|
03:03 < syb0rg> did candlejack getcha afidegnum?
03:06 < afidegnum> ???
03:06 < afidegnum> Windoa: i use nano
03:06 < syb0rg> you kinda cut off halfway through a sentence there and stopped talking afidegnum
03:06 < syb0rg> there is a meme on the internet where you stop talking halfway through a sentence, and it means candlejack got you.
03:07 < syb0rg> In any case I can't see your raid structure, I don't know what it says on your side but I see this: "<afidegnum> syb0rg: here is my raid structure, how do i /4"
03:07 < Windoa> I heard that the people who use vim and who use Emacs usually argue.
03:08 < syb0rg> oh diiiid you now, Windoa >>:|
03:08 < Windoa> HWAT
03:08 < syb0rg> I'll tell you hwat
03:09 < Windoa> pLEAsE
03:11 < syb0rg> so afidegnum, you gonna actually share your raid structure and ask your question or what?
03:12 < Windoa> syb0rg:Please tell me what and why
03:12 < syb0rg> lol. Okay. I use vim, because I messed with it before I messed with emacs and see no reason to use both
03:12 < syb0rg> I only use it to edit config files anyway
03:13 < syb0rg> I use kate or mousepad to write code, or eclipse if I am writing java
03:13 < Windoa> I am writing C++
03:14 < syb0rg> nice, what are you making? Or are you learning it?
03:14 < [R]> TOTAL ECLIPSE OF THE HEART
03:15 < syb0rg> sounds depressing bruh
03:15 < Windoa> I am making a game with ue4
03:15 < syb0rg> lunar eclipse or solar eclipse
03:15 < [R]> syb0rg: both at the same time
03:15 < syb0rg> what kind of game?
03:16 < Windoa> Cars
03:16 < jim> ok, so what do you want to know?
03:17 < syb0rg> a racing game? Does it support peripherals, like a wheel or pedals?
03:17 < syb0rg> aw come on jim, there's nothing else going on right now
03:18 < Windoa> I must leave there for a miniute,sorry
03:18 < jim> come on what? what are you asking for? :P
03:18 < syb0rg> lol
03:19 < syb0rg> oh I mean:
03:19  * syb0rg laughs out loud
03:19 < syb0rg> =P
03:19 < syb0rg> just giving you a hard time jim ;)
03:20 < jim> I'm happy to let it go for awhile (and yes I saw him go),,, and, it would be nice to know what he's after
03:21 < syb0rg> yeah, he was probably just trying to talk about coding on linux
03:21 < syb0rg> at first I thought he was trying to start a vim vs emacs flame war
03:22 < syb0rg> but it seemed like he actually had some kinda project going
03:22 < jim> I got: emacs and vim people argue, I'm using c++, I'm writing a game... Windoa, what would you like to know?
03:22 < Windoa> I am programming
03:22 < jim> ok
03:22 < tomty89> sometimes people just like to chit chat
03:24 < jim> Windoa, are you having any problems with it right now?
03:25 < Windoa> jim: You can send emails to me at postmaster@mocoder.cc
03:25 < tomty89> huh
03:25 < Windoa> It's the website of my studio
03:25 < afidegnum> syb0rg: sorry i was disc  https://ghostbin.com/paste/vuzeh   https://ghostbin.com/paste/g6pca
03:25 < jim> Windoa, you're right here :P it seems a bit inconvenient...
03:26 < Windoa> But it hasn't been ready
03:26 < syb0rg> ok afidegnum, and what are you trying to accomplish? (I was afk when you started asking for help)
03:27 < syb0rg> err away from keyboard
03:27 < syb0rg> god, that rule always gets me
03:27 < jim> syb0rg, don't worry, you're doing a good job thinking about it
03:28 < nekOwOseam> How much memory does systemd use over SysV/
03:28 < afidegnum> i have 2 type of raids, 1 for sdd and another for hdd 2TB
03:29 < afidegnum> but i noticed it's only the sdd raid which is mounted,
03:29 < afidegnum> when i check on fstab, the hdd is not mounted
03:29 < afidegnum> i wanted to mount the hdd on hoe
03:29 < syb0rg> okay, sounds like you need to edit fstab afidegnum
03:30 < syb0rg> do you need help writing an fstab entry?
03:30 < afidegnum> yes, i think i am somehow getting on track but i have an issue
03:30 < [R]> nekOwOseam: 10 gb
03:30 < syb0rg> can you show your current fstab?
03:30 < nekOwOseam> [R]: that low?
03:30 < afidegnum> /dev/md/3 /home ext4 defaults 0 0
03:30 < afidegnum> is this the correct one or /dev/md/3 /home ext4 defaults 0 0
03:31 < afidegnum> /dev/md3 ?
03:31 < syb0rg> you don't want the slash between md and 3
03:31 < afidegnum> well there is an available one with slash
03:31 < afidegnum> wait let me show you
03:31 < syb0rg> I can see in your lsblk output that the device is /dev/md3
03:31 < jim> 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"
03:32 < afidegnum> here is an existing entry https://ghostbin.com/paste/vuzeh
03:32 < afidegnum> jim: thx
03:33 < jim> and, please expand thx :)
03:34 < syb0rg> afidegnum, can you manually mount /dev/md3 to /mnt?
03:34 < afidegnum> i m on a remote server,
03:35 < afidegnum> so i will need to use the terminal
03:35 < syb0rg> okay afidegnum
03:35 < syb0rg> the command is: sudo mount /dev/md3 /mnt
03:35 < syb0rg> assuming you use sudo
03:35 < syb0rg> otherwise become root and do the same thing
03:36 < syb0rg> I just want to verify the device can mount
03:36 < afidegnum> syb0rg: i wanted it to mount on /home
03:36 < afidegnum> should i stil type the command ?
03:37 < syb0rg> I know afidegnum, I just want to see if the raid device is functiona;
03:37 < syb0rg> *functional
03:37 < afidegnum> ok
03:37 < syb0rg> yes, still do it
03:37 < afidegnum> i think it's mounted,
03:37 < afidegnum> there wasn't an output
03:37 < afidegnum> should i unmount it ?
03:37 < syb0rg> yeah go ahead
03:37 < syb0rg> if it didn't complain it is fine
03:37 < afidegnum> unmount, right ?
03:38 < syb0rg> umount
03:38 < jim> before you do that...
03:38 < afidegnum> ...
03:38 < syb0rg> lol trust jim
03:38 < jim> take a look at the output of ls /mnt
03:39 < jim> I'm just wondering if it looks good
03:39 < afidegnum> :)
03:39 < jim> if you unmounted it, thats ok, we can do that after filling in fstab
03:40 < afidegnum> ls /mnt
03:40 < afidegnum> lost+found
03:40 < jim> ok, so it's just empty
03:40 < syb0rg> is that after unmounting?
03:40 < afidegnum> no
03:40 < jim> ok hold one sec...
03:40 < syb0rg> hmm, you have no home directory ther
03:40 < syb0rg> *there
03:41 < jim> do you have /home right now? is that the one you want on the raid /home?
03:41 < afidegnum> yes
03:41 < afidegnum> and #mount gave me this https://ghostbin.com/paste/vuzeh
03:42 < syb0rg> you are gonna need to copy your home directory onto that raid device
03:42 < jim> ok... keep the raid mounted on /mnt, and we can copy it... do you want to do that?
03:42 < syb0rg> then edit fstab and reboot to see if it worked
03:42 < jim> syb0rg, got a new task, one sec :)
03:43 < syb0rg> eh fine, keep trusting jim =P
03:43 < afidegnum> wait i dont understand
03:43 < jim> afidegnum, ok
03:44 < jim> so, you have a /home right now, right?
03:44 < jim> (when you ls /home, you see stuff (looks like usernames) right?
03:44 < afidegnum> yes, i have home
03:45 < jim> do you want -that- /home on your raid?
03:45 < afidegnum> but /home is empty
03:45 < afidegnum> yes,
03:45 < jim> what?
03:45 < jim> ok
03:45 < jim> we can do that, and I'll make sure before we do it...
03:46 < afidegnum> but ~ contains the root's home,
03:46 < afidegnum> this is a bit confusing here,
03:46 < jim> yes, which is /root
03:46 < afidegnum> but /home is empty
03:46 < afidegnum> i mean ~ suppose to represent the current use's home right ?
03:47 < jim> I dunno, they decided to put root's home dir in /root a long time ago
03:47 < afidegnum> ah ok,
03:47 < afidegnum> yes, i saw it
03:47 < afidegnum> ok, let's place /home on /md3
03:48 < jim> could you do this: mount | nc termbin.com 9999
03:48 < jim> it will give a url
03:48 < afidegnum> ok,
03:49 < phogg> root's home is in /root because root may need to log in when an nfs-mounted /home is not available (single user mode)
03:49 < jim> that url will be a pastebin of the output of mount
03:49 < phogg> a more traditional homedir for root is /, but don't do that
03:49 < afidegnum> jim: i like dark background :) http://termbin.com/muop
03:50 < afidegnum> so what happens when we have a new use?
03:50 < afidegnum> will it go to home or he will have /$user ?
03:50 < phogg> afidegnum: it will be /home/$user for most distributions
03:51 < phogg> you can change it to anything you want, but that's the default for UID >= 1000 on typical distributions
03:51 < [R]> phogg: you're nfs
03:51 < phogg> [R]: no, only my face is nfs
03:51 < [R]> lol
03:51 < jim> ok, great... found your /, /boot and the temporary /mnt mounts
03:52 < afidegnum> jim: so should i unmount it ?
03:52 < jim> now, could you do: ls -CFa /home | nc termbin.com 9999
03:52 < jim> no, don't unmount it yet
03:53 < jim> in case we end up copying stuff
03:54 < afidegnum> jim: http://termbin.com/epx6c
03:54 < jim> right now I'm making sure whether there's anything in your current /home
03:55 < tomty89> oops
03:55 < jim> it's totally empty... we don't need to copy anything
03:55 < jim> so: umount /mnt
03:55 < phogg> tomty89: oops?
03:55 < afidegnum> ok
03:56 < tomty89> phogg: nothing in /home
03:56 < afidegnum> done
03:57 < afidegnum> next
03:57 < jim> syb0rg, that's it... you could continue if you want, with making sure fstab is right
03:57 < phogg> tomty89: not necessarily oops
03:58 < jim> looks like maybe there was nothing ever there?
03:58 < TRS-80> evening peeps!
03:58 < jim> hi
03:58 < afidegnum> so which one is correct, ? /dev/md3 /home ext4 defaults 0 0  or /dev/md/3 /home ext4 defaults 0 0
03:58 < jim> the first one
03:59 < tomty89> or both
03:59 < jim> and, the next thing, is you need to make sure the raid is assembled before it gets mounted
04:00 < TRS-80> I have manually installed some dependencies (older version of some boost libraries) and trying to install manually using dpkg -i my.deb but dpkg is not seeing the boost package I manually installed to /usr/local/lib/ - do I need to update path or something?
04:00 < afidegnum> here is my existing fstab http://termbin.com/ernv
04:01 < tomty89> TRS-80: dpkg has no idea about libraries
04:01 < tomty89> TRS-80: but packages
04:01 < jim> afidegnum, ok, you need to get rid of the slash between md and 0
04:02 < tomty89> TRS-80: maybe you should just extract this deb manually as well
04:02 < jim> so instead of /dev/md/0 it should be /dev/md0
04:02 < afidegnum> for all of them or only md0 ?
04:02 < jim> for all of them
04:03 < afidegnum> ok,
04:03 < jim> show the new one :)
04:05 < jim> let me check mine
04:05  * Some_Person has been wondering how it might be possible to set up a remotely accessible Linux desktop with little lag and audio support, preferably in a web browser
04:06 < phogg> Some_Person: "little lag" might be a problem
04:07 < jim> I'll be back in about a half hour
04:07 < phogg> no, don't abandon me!
04:07 < afidegnum> jim: here is the new settings http://termbin.com/2x6m
04:07 < Some_Person> phogg: Well, I've seen it done before. https://www.rabb.it somehow manages it
04:08 < phogg> Some_Person: I'll look but it sounds suspiciously like magic. It matters a lot *how* they do it.
04:08 < Some_Person> phogg: Unfortunately, I have no idea how they accomplish it
04:08 < jim> afidegnum, ok, have you ever tried to reboot with the raid?
04:08 < phogg> Some_Person: that site is light on technical details to the point of being useless
04:09 < toothe> err...how do I delete a file that starts with a - ?
04:09 < toothe> its a directory called -D
04:09 < toothe> probably some mistake I made a bit ago.
04:09 < [R]> rm -- -D
04:09 < [R]> rmdir*
04:09 < toothe> thanks!
04:09 < jim> rm -- -thefile
04:09 < phogg> Some_Person: I know how to share X and X-like display things, some of those protocols also support audio. None of them are especially fast.
04:09 < afidegnum> now rebooting
04:09 < toothe> why doesn't \- work ?
04:09 < Some_Person> phogg: All I can tell as a user is that they give you a virtualized Linux desktop locked down to only run Firefox, and it works well enough to play video smoothly
04:09 < toothe> oh wait...cuz - isn't a special character/
04:09 < toothe> i think I read about this "problem" in phrack a while ago.
04:10 < toothe> like, a phrack from 20 years ago.
04:10 < phogg> Some_Person: A browser-based client for any of those normal remoting protocols is possible and (I assume) must exist, and such clients could use modern APIs to even include audio. Performance might not be great.
04:10 < jim> toothe, because the rm command still sees the - and thinks it's an option
04:11 < dannylee> jo jo jo
04:11 < phogg> toothe: you can also just use ./-
04:11 < jim> the rm command doesn't see the \, the shell does
04:11 < Some_Person> phogg: Yeah, I get what you're saying. It's too bad that service doesn't explain how the hell it works
04:11 < toothe> phogg: that's interesting.
04:12 < phogg> Some_Person: It may be that some open source equivalent does exist, it's just not my area so I do not know for sure. I know that enough of the technologies you would need *individually* exist, so someone could do it.
04:12 < Some_Person> phogg: I've used VNC and RDP for various things over the years, and it's never been particularly smooth
04:13 < afidegnum> jim: how do i verify the entries again?
04:13 < phogg> Some_Person: I imagine they use something like what is used for "steam link"
04:14 < jim> afidegnum, well what happened when you rebooted?
04:14 < afidegnum> nothing, it rebooted, and i checked on fstab, the new entries are there, that's why i want to verify if /home is mounted
04:15 < phogg> Some_Person: some kind of super low latency command protocol coupled with highly compressed video and some cleverness
04:15 < Some_Person> phogg: I see... so basically, a custom job?
04:15 < phogg> Some_Person: That's my guess. Although perhaps not entirely... webrtc is a thing, it might help some.
04:17 < afidegnum> jim: we are good,
04:17 < jim> afidegnum, ohh, just look through the output of mount (with no params)
04:17 < afidegnum> lsblk shows md3 mounted on /hom
04:17 < jim> you can also do df
04:17 < Some_Person> phogg: I believe they use webrtc
04:18 < jim> cool, looks like you got it... you might want to check if it comes up all the way from power off
04:18 < jim> anyway, looks good so far... I'll be back in awhile
04:18 < afidegnum> jim: but wait, there is a litle issue
04:19 < jim> ok, can we deal with it when I get back?
04:20 < afidegnum> https://ghostbin.com/paste/vuzeh
04:20 < afidegnum> /dev/md3         16G   45M   15G   1% /home
04:20 < afidegnum> i thought it supposed to show 1.8T
04:21 < afidegnum> i mean 1.6T instead of 16G
04:21 < TRS-80> OK so how do I manually extract / install just the data.tar.xz portion of the .deb file such that everything gets installed in the right place? extract to /  ???
04:22 < afidegnum> jim: any idea?
04:22 < jim> ok, afidegnum, we'll deal with it when I get back, shouldn't be too long
04:23 < [R]> TRS-80: you don't...
04:23 < afidegnum> ok, thanks in advance,
04:23 < [R]> TRS-80: you use dpkg or a dpkg frontend
04:24 < TRS-80> [R]: can't because it's not recognizing the older version of libboost I had to manually install as a dependency
04:24 < [R]> sounds like a great way to kill your system
04:24 < [R]> installing random things outside your pacakge manager
04:25 < [R]> i.e. you're not installing a package made for your dist
04:25 < TRS-80> yes it is
04:25 < TRS-80> just manually
04:26 < [R]> theres a reason dpkg is complaining
04:26 < TRS-80> because when I do dpkg -i my.deb it tells me " encfs depends on libboost-filesystem1.55.0; however:   Package libboost-filesystem1.55.0 is not installed." even though it is
04:26 < [R]> sounds like it isn't
04:27 < xamithan> Maybe it is in the wrong place ?
04:27 < TRS-80> er rather the [i]package[/i] is not installed, but the libs are
04:27 < [R]> dpkg doesnt care what random crap you're spewed all over your filesystem willy nilly
04:27 < [R]> you've*
04:27 < [R]> it cares about properly installing things that will work
04:27 < TRS-80> yes I know, hence needing to install manually
04:28 < TRS-80> I can't use dpkg, so where to install safely so as not to break anything? under ~ somewhere?
04:28 < [R]> its not made for your dist
04:28 < [R]> you don't
04:28 < jim> afidegnum, also I don't know much about the raid part, I never had an mdraid before
04:30 < jim> TRS-80, well you -can- do it, but you have to make sure everything's right, as far as libs and other programs it needs
04:30 < TRS-80> jim I did already with dpkg -I my.deb
04:30 < jim> TRS-80, it's a LOT of work
04:31 < [R]> jim: the packages work hard for the money... so hard for the money
04:31 < phogg> TRS-80: you would have to make your own .deb package and call it libboost-filesystem with a version number >= that required and install that.
04:31 < jim> I know that's a song, I don't remember whjich... anyway, ok, I'm gone
04:32 < afidegnum> jim: what do you suggest ?
04:33 < TRS-80> I downloaded a .deb package straight from Debian, it's just for Stretch and I'm on sid so couple things had wrong version (actually, only libboost)
04:33 < phogg> TRS-80: on Debian you can always install things into /usr/local/lib that would not interfere with dpkg
04:33 < [R]> lol
04:33 < TRS-80> phogg thank you, that's what I wanted to know. I thought there was some way.
04:33 < phogg> TRS-80: If it's for a different release it's *not compatible*. You should do a simple backport.
04:34 < phogg> TRS-80: you can *install* into /usr/local but that will *not* make dpkg agree to install a package depending on the libs
04:35 < phogg> TRS-80: do you know the dpkg bot in #debian? have you tried /msg dpkg simple sid backport ?
04:35 < phogg> backporting an existing .deb to a different release is usually not terribly hard.
04:36 < TRS-80> backports are not listed in here?: https://packages.debian.org/search?keywords=encfs
04:36 < phogg> even thought it says "simple sid backport" and was meant for porting newer packages from sid to a stable release you can use the same general form for porting a package from any release to any other, with only sometimes having complications
04:36 < phogg> TRS-80: no
04:36 < TRS-80> oh ok, I'll try that bot then
04:36 < phogg> TRS-80: backports.debian.org
04:36 < TRS-80> aah ok
04:41 < TRS-80> phogg: yeah not coming up with anything there
04:41 < TRS-80> bot has same info, I assume?
04:41 < afidegnum> hi
04:41 < TRS-80> afidegnum: hello
04:42 < afidegnum> jim: u there?
04:43 < Psi-Jack> afidegnum: u isn't in this channel at this time.
04:44 < afidegnum> Psi-Jack: :D :D :D :D
04:45 < Psi-Jack> What're you smiling about? It's not really something I'd smile about.
04:45 < afidegnum> you said U is not in the channel
04:45 < afidegnum> i m talknig about jim not U
04:46 < Psi-Jack> Well, u isn't in the channel. Basically, 'u' by itself is not a word. Don't use 'u', use 'you' to communicate properly.
04:46 < afidegnum> ok, comprende,
04:46 < SaEeDIRHA> hello guys , would you know how can i redirect some of the web request to other webserver on a local network?  for example i have server X that is accessible from internet , however , for example when user types x.com/otherserver/  i want to pass the requests to my other http server on my local network which is not accessible directly from internet (in NAT network)
04:46 < Psi-Jack> And, don't mix English and Spanish together. :p
04:47 < SaEeDIRHA> do you know how this can be done ?
04:47 < Psi-Jack> SaEeDIRHA: There's many ways to do that.
04:47 < phogg> SaEeDIRHA: you can use e.g. apache mod_proxy and the lke for that sort of thing
04:47 < phogg> s/lke/like
04:47 < Psi-Jack> And not NAT.
04:48 < phogg> well you could use NAT, but only for whole ports and not for portions of HTTP URLs
04:48 < nudoge> SaEeDIRHA: give the secondary a sub.domain
04:48 < Psi-Jack> Hence, could not, for his actual asked question.
04:48 < Psi-Jack> nudoge: Ignoring the question itself. :p
04:48 < phogg> wouldn't be any point anyway
04:48 < phogg> Psi-Jack: care to expand on some of the other "many ways" to do that? It's not really my area.
04:49 < Psi-Jack> SaEeDIRHA: Apache mod_proxy, nginx location rules, fabiolb, traefik, haproxy, etc...
04:49 < SaEeDIRHA> nudoge, how can i redirect the secondary sub domain to the different server in my network ? i can only forward ports to one ip address in my network which has my apache server running
04:49 < SaEeDIRHA> Psi-Jack, googling the mod_proxy now, thanks :)
04:50 < Psi-Jack> mod_proxy is a bit of a bear to use, and gets slower the more you use.
04:50 < nudoge> SaEeDIRHA: httpd and named entries
04:50 < nudoge> httpd.conf and add a virtual host
04:51 < SaEeDIRHA> nudoge, thats not what i want to do , as the other apache server is on a different machine on a local network
04:51 < SaEeDIRHA> not on the same server
04:51 < Psi-Jack> Yes, you'd still use a named vhost with mod_proxy.
04:51 < TRS-80> ok phogg so by doing backport can I get older version (1.7.4) of software I need instead of what is current version in sid?
04:51 < TRS-80> (whick is like 1.9 or something)
04:52 < TRS-80> looks like that dpgk bot will give you instructions how to do so
04:52 < SaEeDIRHA> Psi-Jack, so certain requestes would be passed to different server on the network ? like proxying the request ?
04:52 < phogg> Psi-Jack: traefik looks quite nice!
04:52 < Psi-Jack> Not like, actually proxying.
04:52 < Psi-Jack> phogg: It's a royal PITA, actually.
04:52 < Psi-Jack> But, for some, it works.
04:52 < nudoge> ^^
04:53 < phogg> Psi-Jack: compared to mod_proxy?
04:53 < Psi-Jack> phogg: You can't even change the Host header from frontend-to-backend. Only passthrough.
04:53 < phogg> TRS-80: should work just fine
04:53 < nudoge> /etc/nginx/conf.d/ becomes your best friend.
04:53 < SaEeDIRHA> Psi-Jack, would the client that send a http request able to detect if its proxied ?
04:53 < TRS-80> phogg: I'll have you know I spent all day trying to build this from source already, lol
04:54 < Psi-Jack> SaEeDIRHA: No.
04:54 < TRS-80> ALL DAY fighting with automake tools...
04:54 < SaEeDIRHA> Psi-Jack, perfect, :) thanks guys , i am googling it to see how it can be done :)
04:54 < TRS-80> until someone said "it's available right theree in Jessie (or whatever) as a .deb!
04:54 < phogg> TRS-80: Welcome to the club. I believe you are entitled to a companionable nod of understanding.
04:54 < TRS-80> :)
04:55 < afidegnum> hello is there a way to modify this ? # mdadm --create /dev/md3 --raid-devices=2 --metadata=1.2 --level=1 /dev/sdc1 /dev/sdd1
04:55 < afidegnum> so i can change to sdc3 and sdd4 ?
04:55 < [R]> there is no sdc3...
04:55 < afidegnum> https://ghostbin.com/paste/vuzeh
04:55 < afidegnum> sd4 rather
04:56 < [R]> what?
04:56 < afidegnum> sdc4
04:56 < [R]> well you already have an md3... so first you have to delete it, then you can make it whatever you want
04:56 < afidegnum> ??
04:56 < afidegnum> what's the command
04:58 < afidegnum> is it possible to use sdc instead of sdc1?
04:58 < nudoge> why aren't you using uuid?
04:58 < [R]> wtf mate @ cloudflare dns
04:58 < [R]> they don't resolve archive.is
04:58 < phogg> nudoge: let's not confuse him further
04:59 < nudoge> k
04:59 < afidegnum> ?
04:59 < afidegnum> nudoge: you mean to use mdadm to add the device uuid ?
05:00 < afidegnum> [R]: cloudflare?
05:00 < [R]> 1.1.1.1
05:00 < Psi-Jack> [R]: Found out that cloudflare filters DNS.
05:00 < [R]> lol
05:01 < [R]> https://community.cloudflare.com/t/archive-is-error-1001/18227/3
05:01 < [R]> they claim it's the fault of the dns server hosting it
05:01 < [R]> but i doubt that...
05:01 < Psi-Jack> It's not. I resolve it just fine.
05:01 < nudoge> That is the best organization method I know of.  using s{a,b,c,d}{1..3} gets too confusing.  Come up with a sane layout using unique lables for each disk.  Then your mdadm command will be more robust.  Deleting the current raid should be the last thing you do.
05:02 < [R]> yeah, when i use google its fine
05:02 < nudoge> 9.9.9.9  IBM quad9 DNS
05:03 < nudoge> I've been using quad1 and quad9 for all the servers I've setup lately.
05:03 < tds> looks like their name servers aren't behaving properly wrt edns: http://dnsviz.net/d/archive.is/dnssec/
05:03 < afidegnum> nudoge: i m just following an available directive, can you please provide an explanatory link ?
05:03 < tds> I'm surprised cloudflare return nxdomain because of that though
05:04 < nudoge> afidegnum: just a moment.
05:04 < tds> unless cloudflare have cached old invalid ns records or something
05:04 < TRS-80> phogg: so how do I specify an older version when trying to build a backport?
05:05 < TRS-80> I just tried putting version number but I got Unable to find a source package for encfs_1.7.4-5
05:05 < [R]> back to google for me
05:05 < phogg> TRS-80: you can specify a release with -t or a version with /
05:05 < TRS-80> oh wow ok
05:05 < TRS-80> lemme try that
05:05 < phogg> TRS-80: e.g. apt-get source foobar/1.2.3
05:06 < nudoge> https://www.tecmint.com/create-raid0-in-linux/
05:06 < nudoge> https://www.tecmint.com/grow-raid-array-in-linux/
05:06 < phogg> TRS-80: or I guess in your case aptitude build-dep packagename/1.2.3.whatever
05:06 < TRS-80> just had the syntax wrong, I was pretty close actually, not a bad guess if I do say so myself :)
05:06 < irwinz> i can't ssh into my linux server
05:07 < irwinz> i'm looking at the authorized keys file from a monitor
05:07 < irwinz> looks the same as the ssh key i'm trying to use with TeraTerm
05:07 < nudoge> those are nice pages for learning the basics.  You can instead of using the /dev/s* layout just use uuid's https://unix.stackexchange.com/questions/52321/using-uuids-with-mdadm
05:08 < afidegnum> nudoge: thanks,
05:08 < nudoge> no worries, good luck.
05:10 < nudoge> afidegnum: blkid and lsblk will help too.  There is a way to customize uuids too.
05:11 < afidegnum> ok
05:14 < afidegnum> nudoge: so is there a way to use  sdd or sdc instead of sdc1 or sdd1 or 2,3... i.e https://ghostbin.com/paste/vuzeh ?
05:17 < nudoge> afidegnum: https://ghostbin.com/paste/g468c what matters is the fact that the md0 is already configured to use the /dev/sd* set up and will require work to rebuild the raid.  Are you trying to just swap a drive or rebuild from scratch?
05:17 < irwinz> nvm boys i got it
05:17 < ayecee> glad we could help
05:17 < irwinz> heh
05:18 < irwinz> sometimes asking puts you in the right mindstate
05:18 < irwinz> hopefully i didn't bother anybody
05:18 < irwinz> cheers
05:18 < ayecee> :)
05:18 < irwinz> now on to figuring out how to set up plex
05:18 < irwinz> is there any benefit to running my nzb client on my linux server instead of this win pc?
05:18 < irwinz> aside from resources--those are aplenty
05:18 < afidegnum> nudoge: i think i will have to rebuild from scratch
05:19 < afidegnum> i just want to map /home to sdd and sdc
05:19 < ayecee> linux server probably doesn't get rebooted as often
05:19 < ayecee> those windows updates amirite
05:19 < irwinz> true
05:19 < irwinz> so i'd just be accessing sabnzbd the same
05:19 < irwinz> but with it running on that box
05:19 < irwinz> and then plex to stream to this pc and/or smart tv
05:20 < phogg> another satisfied customer!
05:20 < irwinz> damn it i just realized i bought a 92mm fan for my hdd cage
05:20 < phogg> irwinz: the only hard part about plex is that it's finicky about permissions
05:20 < irwinz> its supposed to be 120mm
05:20 < irwinz> i'm a dunce
05:21 < irwinz> phogg: oh god, i'm terrible with those
05:21 < irwinz> my weakest asset in regards to linux
05:21 < irwinz> i have some fundamental block when it comes to understanding the numbering and how they correspond to read/write, etc
05:21 < irwinz> need to work on that.
05:22 < phogg> irwinz: the numbering is rather irrelevant
05:22 < nudoge> afidegnum: you are trying to mount two drives to the same directory?  If so there will be issues with one getting written to an not the other.  My recommendation is to get the data backed up and start formating drives.  Once they are all formatted you can go discover their new UUID's and start setting up /dev/md0 with those instead of device/array.  mdadm respects uuid's and these uuid's can be added to the /etc/fstab in place of a /dev/s* destination.  Wh
05:22 < phogg> irwinz: if you want to learn it, though, I recommend you read the 'info chmod' document entirely once.
05:23 < nudoge> chmod is a sticky subject
05:23 < phogg> nudoge: only if you use +s
05:24 < irwinz> yeah i'm not even gonna start
05:24 < phogg> irwinz: it's really, really simple and not at all bad.
05:24 < irwinz> i know that's the worst part
05:24 < afidegnum> nudoge: the 2 drives will be mounted as raid on md3
05:24 < irwinz> i know it's simple i just can't grasp it
05:24 < afidegnum> on which /home will be mapped on
05:25 < irwinz> not good for the self esteem
05:25 < irwinz> so i end up kinda stabbing in the dark and testing, adjusting if needed
05:25 < TRS-80> I don't understand backports. Say I need encfs_1.7.4 which was available in jessie, but I am currently on sid. What do I put into my sources file? jessie-backports?
05:26 < nudoge> afidegnum : https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/4/html/Introduction_To_System_Administration/s3-storage-raid-day2day-add.html something like this?
05:26 < phogg> irwinz: it's easy. If you take zero as no permissions and you want read permission you add 4. If you want write permission you add 2. If you want execution permission you add 1. That's it.
05:27 < irwinz> and then there are different 'locations' or spots in the ordering that discern the user/root/whatever
05:27 < phogg> irwinz: because there is no way to add any two or three of these numbers and come to the same value twice a single number can tell which of those values you summed and thus which of the 3 permissions you have.
05:27 < irwinz> hm, neat
05:27 < phogg> irwinz: That, too, is straightfoward: It's always user, then group, then other ("world").
05:28 < phogg> irwinz: so 660 is "4 + 2" aka "read + write" for user, the same for group, then no permissions for people who are not the file's owner or people in the file's group.
05:28 < phogg> irwinz: there are a ton of corner cases and extra bits which you can ignore 99% of the time
05:30 < irwinz> cool, thanks for the info
05:30 < irwinz> always appreciated
05:31 < irwinz> while i'm here--perhaps a superficial and useless concern--but i have my ssh client (TeraTerm) looking super nice and easy on the eyes, can i change the system font on a more local level so that it looks nice via a display plugged in to it?
05:31 < irwinz> currently it's that ugly courier new/fixedsys looking font on the monitor
05:34 < phogg> irwinz: I don't know one way or the another about that.
05:35 < imofftopic> hi
05:35 < afidegnum> nudoge: you there?
05:36 < afidegnum> see this https://ghostbin.com/paste/qcd7y
05:36 < nudoge> yes
05:36 < afidegnum> this seems to be confusing,
05:36 < afidegnum> where should the partition mapped to?
05:36 < nudoge> looks right
05:36 < afidegnum> sdc or sdc1-4 ?
05:37 < nudoge> what does /etc/fstab look like? and did you run lsblk and blkid?
05:37 < afidegnum> ok, let me give you their output
05:38 < kurahaupo> irwinz: perhaps you mean your terminal program TeraTerm? (which happens to include a built-in ssh client, though that's irrelevant to any questions about fonts and displays)
05:38 < irwinz> kurahaupo: i probably worded that the worst way possible
05:39 < irwinz> let me simplify: when i plug in a monitor locally to my linux server, the font is ugly and i want to improve it
05:39 < afidegnum> nudoge: here are they https://ghostbin.com/paste/qcd7y
05:39 < irwinz> (the font i use in TeraTerm i custom specified, obviously it won't show on the monitor
05:40 < surrealpie> im trying to write a simple wrapper for the pinentry program
05:40 < kurahaupo> Is it running X or Wayland or something using direct framebuffer?
05:41 < kurahaupo> Eg SVGA emulation?
05:41 < irwinz> is it a stupid/dunce/shortsighted idea to try and retro fit a 92mm fan into a loc in my case that should have a 120mm fan?
05:41 < nudoge> afidegnum : remember when Psi said you would need to get rid of the current raid?  Well its not like you can just swap stuff unless you are swapping to replace something.  If you only want to use specific drives and raid those two its going to take a lot of work.
05:41 < ayecee> it _sounds_ stupid, doesn't it
05:41 < day> is it possible to clone a linux system while being on it to create a 1:1 copy of it that i can run in a VM?
05:42 < irwinz> my current setup is a large square fan facing the front point blank, and a small 10"-12" fan on the top in place of the also decomissioned 200mm fan
05:42 < ayecee> it's not like 120mm fans are hard to come by
05:42 < day> something like cp / > usbstick/ -r
05:42 < irwinz> ayecee: this is true, fry's is far though and i grabbed the wrong one
05:42 < irwinz> i convinced myself it had to be right because it was the same brand as my case
05:42 < irwinz> wtf Antec...
05:42 < irwinz> 92mm? who uses this?
05:42 < nudoge> afidegnum : thus, why you should verify backup integrety and start deleting raids
05:42 < ayecee> they can ship them to you these days. they have the technology.
05:43 < afidegnum> ok, i m going to get rid of the current raid
05:43 < irwinz> i digressed, and this is not a hardware channel--sorry
05:43 < afidegnum> i don't have anything saved on the raid HDD yet
05:43 < nudoge> ya, we can't bend that
05:43 < nudoge> well the raid looks good. Is it not working?
05:43 < afidegnum> they are all healthy
05:43 < nudoge> Its a lot of hard work to get to this point
05:44 < nudoge> see how they have not only uuid's but sub-uuid's
05:45 < afidegnum> yes
05:46 < afidegnum> nudoge: what is still not clear to me, is the raid partitions,
05:46 < afidegnum> why the sdc1-4 instead of sdc only?
05:47 < nudoge> afidegnum : its an array
05:48 < afidegnum> ok,
05:48 < nudoge> This device list is built on boot.  /dev/sd[a-z]{0..9}
05:48 < afidegnum> so what device/array do we suppose to use?
05:48 < nudoge> if all of the disks are there an good and fstab passes say 0 1 then it will boot.  but with a 0 1 and one of the drives goes out it will not boot.
05:50 < nudoge> so if the machine is expecting a sdc3 to be present and it is broken this it will need to be replaced, or you can set to 0 0 and it will still boot with the failed disk.  so.  the array was used to designte devices to the operating system.  now uuid's are used in advanced systems.  While learning you should not need to get that deep.  But I recommend then, to get 5 usb storage devices and do a standard mdadm setup with email alerts.
05:52 < afidegnum> ok, thanks for the insight
05:52 < nekOwOseam> Hewwo frens UwU
05:53 < TRS-80> Unable to find an archive "1.7.4" for the package "encfs"; Unable to find the source package for "encfs"
05:53 < nudoge> Ya, mdadm is super awesome.  But think about this.  Now, lets say you have uuid set up and a drive has to be swapped while the file system is offline.
05:54 < TRS-80> >not writing numbers on the ends of your HDDs with a paint marker
05:57 < afidegnum> nudoge: i m getting the point, but still need more light
06:00 < afidegnum> nudoge  how do i rewrite this ?  mdadm --create /dev/md3 --raid-devices=2 --metadata=1.2 --level=1 /dev/sdc1 /dev/sdd1
06:08 < kerframil> rewrite to what end?
06:18 < storge> to victory
06:23 < ayecee> to the pain
06:44 < Chirrups> How do I change the temporary working directory for a cat * | sort | uniq > file command?
06:44 < Chirrups> it's storing files in /temp and I'd like to offload them to another drive
06:46 < ayecee> what shows that it's storing in /temp ?
06:47 < Chirrups> I'm not even sure which command is doing it; I'd assume it's cat because it starts right off the rip, but it may be passing the text to sort and uniq live
06:47 < Chirrups> Is there an easy way for me to search a handle?
06:47 < ayecee> right, but what shows that?
06:48 < ayecee> like, what is the problem that led you to this question?
06:49 < Chirrups> I'm running the OS on a 32gb SD card and I have to combine\sort and prune duplicates on ~250gb of log files
06:49 < ayecee> okay, so what happens when you try?
06:49 < Chirrups> It fills up the /temp/ directory and stalls
06:49 < Chirrups> sorry /tmp/
06:50 < ayecee> probably sort, then. see the -T option for sort.
06:51 < ayecee> the thing about sorting is that you pretty much have to load the whole list before you can sort it
06:51 < ayecee> also consider using sort -u instead of sort | uniq
06:51 < Chirrups> OH GOD
06:51 < Chirrups> Wow that filled up fast
06:52 < Chirrups> I should not be putting that at the root of the drive
06:52 < Chirrups> So I could do sort -u and my thought about piping it was because I have like. 9 billion 10mb files
06:53 < ayecee> the piping is reasonable
06:54 < ayecee> having sort do the uniq itself with -u means it doesn't have to store duplicate lines in its temp files.
06:54 < Chirrups> Is there some kind of alternitive to sort that I could use? It seems a bit dumb to make individual files
06:54 < Chirrups> I would think it would load it into a temporary indexed DB
06:55 < Chirrups> Or would that more be a job of something like nosql\postgres
06:55 < ayecee> interesting idea. i bet if you passed the files to sort directly to sort, it could do something clever like that.
06:55 < ayecee> if sort doesn't know which files the data came from, it can't really index it
06:56 < Chirrups> Ahh
06:56 < ayecee> so, sort -u *
06:56 < ayecee> easy peasy
06:57 < Chirrups> And then pass that command to output to a file with a carrot?
06:57 < ayecee> with a pipe, sure
06:58 < ayecee> or probably with a redirect,  >
06:58 < Chirrups> That's what I meant by a carrot
06:58 < Chirrups> Sorry
06:58 < ayecee> ah, caret
06:58 < sauvin> That's not a carrot, nor is it a carrot.
06:58 < sauvin> Or a caret, even.
06:58 < ayecee> yeah, caret is ^
06:58 < Chirrups> Greater then?
06:58 < sauvin> Exactly.
06:59 < sauvin> Right angle bracket.
06:59 < ayecee> maybe just call it a redirect
07:00 < kerframil> you can call it a redirection operator
07:00 < sauvin> I'll certainly sit still for that.
07:00 < kerframil> but most people would understand redirect, I think
07:00 < kerframil> carrot ... not so much
07:00 < Chirrups>     -S, --buffer-size=SIZE               use SIZE for main memory buffer
07:00 < Chirrups> What is it looking for do you think here, bytes?
07:01 < ayecee> probably yes
07:01 < Chirrups> ((finally reading the man page))
07:01 < kerframil> the man pages for GNU utilities are generally lacking
07:01 < ayecee> also keep in mind that a smaller buffer means larger temp files
07:01 < ayecee> kerframil: how so?
07:02 < kerframil> Chirrups: info sort will explain the default - and accepted - units
07:02 < kerframil> ayecee: ^ one example
07:02 < ayecee> so does the manpage
07:02 < Chirrups> Right I've got 128gb of ram
07:03 < Chirrups> Figure it probably could use some permission to use a bit more than whatever is default
07:03 < imofftopic> 128GB ram... you need an upgrade man
07:03 < ayecee> Chirrups: would be useful to compare with and without that option. it could be that sort is already smart enough to choose a reasonable size.
07:03 < Chirrups> It's DDR4
07:03 < kerframil> ayecee: sure, right at the bottom and it doesn't explicitly state the default unit. anyway, I'm not getting into an argument over this. the GNU's stance on man pages, and their focus on info documentation is well known.
07:03 < kerframil> FSF, I mean
07:04 < ayecee> kerframil: in my experience, the manpage is almost always a verbatim copy of the info page.
07:05 < kerframil> a verbatim copy? ok, whatever you say.
07:06 < ayecee> comes from years of "for more details, see the info page", and finding no new info in the info page.
07:06 < Chirrups> kerframil what do you have to prove here o.o like legit, why the patronizing tone.
07:06 < Chirrups> Like, everyone does crap differently
07:06 < ayecee> patronizing is what we do here
07:06 < ayecee> up your game man
07:07 < kerframil> Chirrups: nothing. but I reserve the right to dispute matters of fact.
07:07 < Chirrups> If i'ts that much better people will gravitate there themselves
07:07 < Chirrups> That's interesting
07:07 < kerframil> Chirrups: I don't care one iota whether the next user prefers to use man or info. it's an indisputable fact that the info documentation for, say, coreutils is more comprehensive.
07:07 < ayecee> yeah, info hasn't made big strides in that area.
07:07 < Chirrups> sort -S 17179869184 -T /media/root/STOR/cat | uniq > /media/root/STOR/Stage/sort.txt is just sitting there
07:07 < ayecee> and yet here i am disputing it
07:08 < ayecee> Chirrups: what should it do instead?
07:08 < Chirrups> ... use resources
07:08 < ayecee> how are you checking?
07:08 < Chirrups> htop and top
07:08 < ayecee> oh, wait.
07:08 < Chirrups> No disk, no ram, no cpu
07:08 < ayecee> you haven't provided any files to sort.
07:09 < Chirrups> oh oh, it won't just use the files in the directory?
07:09 < Chirrups> yeah adding a * at the end did it
07:09 < kerframil> ayecee: diff -u <(man sort) <(info sort)
07:10 < kerframil> way, way more information
07:10 < ayecee> and yet no information i've ever needed.
07:10 < Chirrups> Ha
07:10 < kerframil> what happened to "a verbatim copy"?
07:11 < ayecee> i suppose i didn't actually diff the output. good observation.
07:11 < kerframil> tar is a fantastic example. it's actually not possible to understand fully how GNU tar works without reading the info documentaiton.
07:11 < Chirrups> What do I have to do again to redirect cat to stdout and make it follow a file?
07:11 < kerframil> but, really, this pattern is repeated pretty much everywhere for GNU stuff
07:11 < kerframil> I'm not making it up for the fun of it
07:11 < ayecee> Chirrups: how do you mean?
07:11 < Chirrups> oh wait
07:11 < Chirrups> This is linux
07:11 < Chirrups> I use tail
07:11 < Chirrups> sorry
07:12 < Chirrups> I don't work with linux much
07:12 < Chirrups> I'm one of the scorned and hated primarily windows sysadmin and voip people
07:12  * storge shakes a hate fist
07:13  * ayecee gets the pitchforks
07:13 < Chirrups> So this is interesting
07:14 < Chirrups> What it seems to be doing is running 3 concurant threads on 2 different files and then using a third to merge
07:14 < Chirrups> That seems .. counterproductive as the complexity of the merge must at somepoint get much larger
07:14 < ayecee> doesn't sound like the sort i know
07:15 < ayecee> i thought it was single threaded
07:15 < srgj931>  I get some artefacts on my screen when using VLC in fullscreen. They also persists when exit VLC fullscreen mode. I am running arch linux on the newest kernel. 4.16.11-1. Intel Graphics HD 520. Driver in use: i915.
07:15 < srgj931> Looks like this: https://imgur.com/a/IpJjwqP
07:16 < Chirrups> Yeah dude it's totally multithreaded
07:16 < kerframil> srgj931: do the artifacts go away if you tell vlc not to use hardware acceleration? if so, you may have an issue with VA-API, assuming that the supporting components are installed (not that I know how to fix VA-API problems).
07:18 < kerframil> also, fwiw, I have found hardware acceleration to be somewhat unreliable in VLC. I'm running it on a much older GPU than you have, though.
07:20 < srgj931> kerframil: it doesnt matter if I set hardware acceleration to disabled/automatic. Same outcome.
07:20 < kerframil> srgj931: oh dear
07:21 < Chirrups> ayecee: https://imgur.com/a/SlNHfuu
07:22 < storge> what am i looking for in that picture
07:22 < ayecee> the multiple instances of sort
07:22 < ayecee> i guess it's probably multiprocess rather than multithread
07:23 < Chirrups> pegging all 4 cores for sure
07:23 < storge> is that unusual?
07:23 < ayecee> i don't know, i had always thought it was single threaded
07:23 < kerframil> srgj931: I would recommend checking the media for errors, just to be sure. if it's a file, something like ffmpeg -v error -i <file> -f null
07:23 < Chirrups> I just need to make sure that it's not opening every file in a different thread\process
07:23 < storge> but it's not pegging all for cores
07:23 < kerframil> srgj931: if it seems healthy, try another player. if the problem can be only reproduced in vlc, and it happens evern with pure software decoding, file a bug I suppose.
07:24 < kerframil> even*
07:24 < Chirrups> It's for sure multi process, it's different PID
07:24 < storge> multiprocess = multithread?
07:24 < Chirrups> Nope
07:24 < ayecee> nope, different things
07:24 < storge> i didn't think so
07:24 < Chirrups> It's going to do a final merge
07:24 < Chirrups> I think
07:25 < storge> i thought that's what you were saying, mentioning first multithread then just now saying multiprocess
07:25 < Chirrups> Yeah I thought it was multithread till I saw the PID
07:25 < Chirrups> it's cool that it's smart enough to move the processes across cores
07:25 < kerframil> srgj931: one other thing; you can activate a console window in vlc itself and set the threshold to debug. you may need that if you have to raise it with the vlc folks.
07:25 < Chirrups> Going to have to move this to the server now though :/
07:26 < Chirrups> vmware can be such a PITA with usb 3
07:27 < srgj931> kerframil: the artifacts goes away whenever i move the mouse.
07:27 < srgj931> kerframil: no media issue, as it happens completely random over mutliple files. Just when entering fullscreen.
07:28 < srgj931> kerframil: I reinstalled arch recently. Definitly didnt had that problem before. So maybe I am missing a piece of software or configuration ...
07:28 < kerframil> srgj931: arch is a fast moving system so it could just as well be a regression
07:29 < kerframil> srgj931: at any rate, see if the problem only manifests itself in vlc
07:30 < srgj931> kerframil: alright, i ll test with other video players. Thanks.
07:32 < kerframil> srgj931: don't suppose you remember which kernel you were using in the prior installation?
07:33 < srgj931> kerframil: No. But I can downgrade to 4.15.x and see if its better.
07:33 < kerframil> srgj931: I was going to say, you can get older packages from archive.archlinux.org if you're trying to check for the possibility of a regression in a particular package. the kernel is obviously the lowest entity in the stack here that matters.
07:34 < srgj931> kerframil: Dont have a problems with mplayer.
07:34 < kerframil> but it might just be vlc
07:34 < kerframil> ah
07:34 < srgj931> Ok. Let me downgrade VLC.
07:35 < blackgatonegro> VLC on linux tends to be... not as good as it can be. I honesty just keep it for nostaligia and mostly use mplayer
07:35 < kerframil> keep in mind other possibilities too, like whether you are using xorg or wayland, as compared to before
07:35 < blackgatonegro> mostly with a gui
07:35 < kerframil> I agree; VLC is disappointing on Linux
07:36 < blackgatonegro> I mostly use it to test how badly coded up a video is.
07:36 < kerframil> and macos, for that matter. windows seems to be the most stable platform for it, oddly enough.
07:37 < blackgatonegro> Also vlc tends to work worse if you use propietary video drivers on linux for some reason. While literraly no other video player I use on linux does.
07:38 < blackgatonegro> on linux use mplayer, or mplayer2, want gui? We got gui. SMPlayer works fine.
07:39 < luna> going to OpenSUSE Hangout and LUG meetup in Uppsala in a couple of minutes :)
07:40 < kerframil> mpv also has a good rep. totem is acceptable if one doesn't require bells and whistles - just make sure that the -good and -libav gstreamer plugins are installed.
07:40 < storge> does anyone know if sprunge.us purged old files recently?
07:42 < blackgatonegro> I honesty wish my memory was better, I keep forgetting Terminal commands. Yet I can still remenber most of the ones for dos, I blame videogames. When a ton of games need you to know stuff to play them, you learn that stuff.
07:43 < srgj931> kerframil: alright. I ll have a look. Thanks for helping.
07:43 < srgj931> kerframil: Oh. I just have the artifacts now also on mplayer.
07:44 < kerframil> that's bad
07:51 < srgj931> Ok this is bad; Installed xf86-video-intel package now. And it looks like its gone.
07:52 < kerframil> that's the correct driver. makes me wonder what xorg was using prior. possibly something very inefficient like vesa.
07:53 < kerframil> not that that excuses the artifacts, of course but still
07:53 < kerframil> by the way, you also need to install libva-intel-driver if you want to offload decoding to the GPU. there's a whole article on it at the arch wiki.
07:56 < srgj931> Ok. Let me have a read on that.
08:12 < muffindrake> Is there a comprehensive/widely-used format that can be used for binary patches of files?
08:20 < kerframil> muffindrake: xdelta, maybe
08:20 < TaZeR> anyone know of a script or program that does simple to input/ouput conversions for data like bytes to kilobytes or megabytes etc..
08:20 < kerframil> TaZeR: man units
08:21 < kerframil> wait, that's not it
08:21 < kerframil> there should be a package named units, though
08:22 < TaZeR> thanks i found it, will check it out
08:22 < kerframil> once you install that package, man units should give you the correct information. otherwise it will probably bring up units(7) instead.
08:22 < TaZeR> looks like what i want, even does conversions for all sorts of other units, nice
08:23 < TaZeR> oh i see
08:23 < kerframil> units(1) is the right page
08:23 < TaZeR> i was searching all the wrong phrases like "bit converter" lol
08:23 < syb0rg> .... man units. I'll add that to the list with man mount and finger
08:23 < TaZeR> heheh
08:24 < kerframil> that said, man 7 units does provide the information necessary to do the calculation yourself
08:24 < kerframil> nice to have a utility though
08:25 < TaZeR> where did the other man page go though if it replaced it
08:25 < TaZeR> or does it include the same information
08:25 < TaZeR> i didnt really read too far into the original one
08:25 < kerframil> nowhere. man just searches the 'sections' in a specific order.
08:25 < kerframil> 7 is the lowest priority section
08:26 < TaZeR> oh ok i can bring it with the other number then
08:26 < TaZeR> bring it up*
08:26 < kerframil> you can always run man <n> <name>, where <n> is the desired section
08:26 < TaZeR> cool it works
08:26 < kerframil> should it ever be necessary
08:50 < RayTracer> there is also numfmt
09:16 < supernovah> Anyone know a good place to discuss engineering signals (ECE stuff)?
09:24 < storge> supernovah: maybe ##electronics
09:29 < supernovah> also random question, how do I search channels in irssi
09:30 < geirha> supernovah: /msg alis help list
09:35 < storge> note: that's how to search freenode from an irc client like irssi; alis is a freenode thing
09:36 < Juesto> how do i offset a partition inside a partition from 1049kb to 0?
09:39 < hexnewbie> Juesto: You want to create a partition table inside a partition (nested partition tables), with the nested partition starting at offset 0 of the parent partition (i.e. them being one and the same)?
09:41 < sauvin> Under the old DOS partition table scheme, I'd have accused such a thing as being a "logical partition", or "extended", rather than a "physical" or "primary" partition.
09:42 < Juesto> hexnewbie: I'm trying to put a image file converted from virtual box to raw onto a partition on my drive
09:43 < storge> i've been running with logicals in an extended with no primaries for ages and no problems
09:43 < Juesto> my drive is gpt
09:43 < hexnewbie> Juesto: OK. Do you want to do just that? How is it related with editing partition tables?
09:43 < hexnewbie> Juesto: Or your image has a partition at offset 1049kb?
09:44 < Juesto> hexnewbie: i just dd'd it to a new partition and it's offsetted to 1049kb
09:44 < Juesto> parted says
09:44 < Juesto> it's a disk image
09:44 < Juesto> I'm not sure how many partitions it has
09:44 < hexnewbie> Juesto: Which object is offsettted from which object? What's the actual output parted uses to describe this?
09:45 < Juesto> the partition inside the primary partition is reporting this in parted
09:45 < Juesto> just like that, starting offset: 1049kb
09:46 < hexnewbie> Juesto: Perhaps you want to run losetup on the image to create a loop device, then run kpartx -a (apt-get install kpartx) on the loop device, and copy individual partitions from /dev/mapper/loop-...
09:46 < hexnewbie> But I can't tell for sure with the way you're asking it
09:46 < Juesto> I'm running gparted live image
09:46 < Juesto> the source file doesn't mount via loop
09:46 < Juesto> it's technically a hard drive
09:47 < hexnewbie> In what way it doesn't mount via loop?
09:47 < Juesto> gives error
09:47 < hexnewbie> Because it has partitions and you need offsets?
09:47 < hexnewbie> Juesto: “...then run kpartx -a (apt-get install kpartx) on the loop device...”
09:48 < Juesto> it's not a loop device
09:49 < Juesto> I'm trying to fix an offsetted partition
09:49 < Juesto> inside a partition
09:49 < hexnewbie> It's no longer a disk image, it now resides in a partition?
09:49 < Juesto> i used dd from that disk image
09:50 < hexnewbie> All partitions are offsetted, and usually don't need to be fixed
09:50 < hexnewbie> Partitions *are* offsets. That being said, you can run kpartx on a partition if you mistakenly created nested partitions.
09:51 < hexnewbie> But if you did that, it would be probably easier to start over, and work with your loop device again.
09:51 < Juesto> it's not a loop device
09:51 < hexnewbie> <Juesto> the source file doesn't mount via loop
09:51 < Juesto> well by default it tried that
09:52 < Juesto> nvm
09:52 < hexnewbie> I still believe that you need to run losetup, kpartx and then copy the partitions. The rest of it makes no sense to me.
09:53 < rcf> Juesto: if I understand what's going on, you can create a loop device using losetup -f -o youroffset /dev/whateverhtepartitionis which will give you a device with the filesystem you're trying to mount.
09:53 < Juesto> it says invalid argument for the loop
09:53 < hexnewbie> If you did do what I think you did, it can be fixed by doing a lot of math and manually creating partitions at specific sectors, but if someone told me I had to do that, I'd bail out and start all over by copying again from the source the right way
09:53 < Juesto> when i try to use mount
09:54 < hexnewbie> Juesto: Yes, because the loop device contains an entire disk with partitions, you need to run kpartx on it to create /dev/mapper devices for the partitions and mount those
09:55 < rcf> hexnewbie: Actually if you have a recent-enough losetup you can use -P now to automatically scan for partitions.
09:55 < hexnewbie> Juesto: Or what rcf says. I just haven't used partitions with loop devices in the last 10 years :)
09:56 < Juesto> how do i fix the offset in this image?
09:57 < Juesto> i can access to the partition now
09:57 < hexnewbie> Juesto: losetup -P -f /path/to/image
09:57 < Juesto> but i dd'd the image to a physical partition and it's not recognized
09:57 < Juesto> Should i dd the newly mapped loop?
09:58 < Juesto> ....
09:58 < rcf> Juesto: what you want is to image the partition that you can actually mount, so yes.
09:58 < rcf> If you're just trying to get the filesystem.
09:58 < hexnewbie> Juesto: Yes, but not the main loop device, the partition of the loop device (it would be named loop0pX or loop0-partX or something).
09:58 < csierra_> I'm looking for a a script that will monitor the network uptime on a system which isn't publicly facing. I was thinking of a two part script. One on the internal server to poll the second script, which would be running on a webhost I control. The second script would log the datetime to a database and create the page with uptime information. Does such a thing already exist?
09:59 < Juesto> yes loop1p1
09:59 < rcf> Do that and you'll have a mountable filesystem image.
09:59 < Juesto> okay
10:00 < Juesto> it's 21gb so, waiting
10:01 < Juesto> loop1p1 is a link to /dev/dm-0
10:01 < Juesto> huh
10:02 < hexnewbie> If you used kpartx, it uses dm devices. I suspect rcf's suggestion for losetup -P wouldn't, but I may be wrong
10:03 < Juesto> i would like to know why that offset is present
10:03 < Juesto> again, it's a raw image converted from virtualbox
10:04 < Juesto> is the first sections reserved for the partition table data?
10:04 < rcf> Juesto: Yes. The VM presents the image as if it were a physical hard drive, so it's partitioned as such by the guest OS.
10:04 < Juesto> yeah right
10:05 < Juesto> so I'm supposed to dd it to a disk instead of a partition?
10:05 < Stryyker> maybe vbox has some extra stuff at the start? Checked in a hex editor? I know VMWare Workstation normally does
10:05 < rcf> Stryyker: I think he converted it to raw, so that stuff would be stripped out.
10:05 < Juesto> Stryyker: i used their vboxmanage to make it a raw image straight away
10:07 < rcf> hexnewbie: I would think losetup would be using the same kernel features as kpartx, so dm devices wouldn't be too much of a surprise.
10:07 < Juesto> alright then
10:13 < hexnewbie> Juesto: You either dd to a disk, or dd the individual partitions (more complicated options are also available)
10:14 < Juesto> How do i do that without needing to mount the image as partitions
10:14 < hexnewbie> The word mount would need a definition, because if you're using dd it's essential that *NOTHING* is mounted.
10:14 < Juesto> hm
10:15 < rcf> Juesto: if you're already using dd, skip=N where N is your offset divided by the input blocksize.
10:15 < Juesto> hexnewbie: more complicated stuff such as?
10:15 < Juesto> rcf I'm not using that :/
10:16 < Juesto> but okay
10:16 < hexnewbie> If you mean how to do it without a loop device and partition detection, the answer is a lot of offset math, and using the correct bs= and skip= on dd
10:16 < Juesto> til for the next time
10:16 < Juesto> and how i would be figuring out where the partition
10:16 < hexnewbie> Entering the correct skip= and bs= is more complicated (as in, you can get it wrong), than trusting the kernel and/or kpartx to get those right and give you the partition devices.the
10:16 < Juesto> begins?
10:17 < Juesto> is there a known numbers for those offsets?
10:17 < hexnewbie> Juesto: Partition table tools (gdisk, fdisk) will tell you at which sector the partition starts, and the partition size. You'd compute the byte size and convert to block number.
10:18 < hexnewbie> Oh, and the sector size.
10:18 < Juesto> but can they analyze an image file?
10:19 < hexnewbie> So you'd multiple the sector size by the starting sector number to get the starting byte, then find a big power of two the starting byte is divisible by to use as bs= for dd, then divide the starting byte by the bs= to compute the skip=
10:19 < hexnewbie> Juesto: Everything is a file
10:19 < bR4bbit> test
10:19 < ayecee> i am not a file
10:20 < Juesto> yeah well so i would specify the raw image i guess?
10:20 < ayecee> probably
10:20 < hexnewbie> Juesto: Yes
10:22 < hexnewbie> But lazy people like me just use the kernel and/or losetup and/or kpartx to do that work for them :)
10:23 < kerframil> justanotheruser: fdisk, at least, does not work with image files (probably due to missing ioctls). file -s usually works.
10:24 < kerframil> sorry, that was for Juesto
10:24 < rcf> hexnewbie, Juesto: partx --show apparently always uses 512-byte sectors, so with bs=512 skip=$START count=$SECTORS dd wouldn't actually be that difficult.
10:24 < hexnewbie> Em, fdisk has always worked on files for me. Maybe that's a new thing?
10:24 < Juesto> hmm
10:25 < Juesto> yeah i guess it's simpler to do it off the loop device arrangement
10:26 < rcf> It also gives you a guarantee in that you can always be sure that the partition will mount before you image it.
10:26 < kerframil> hexnewbie: could be. it's not working for me with a recent version (just tested with a simple MBR layout).
10:26 < Juesto> yeah there we go thanks
10:27 < kerframil> hexnewbie: oh wait. I tested on a box with an outdated version of fdisk. that might be it.
10:27 < rcf> Whereas offsets are, as you've discovered, really annoying.
10:27 < rcf> kerframil: it works for me on util-linux 2.32
10:28 < kerframil> yeah, it does
10:28 < hexnewbie> It will certainly fail to ‘reread’ the partition table, but fdisk -l and gdisk -l ought to work. And I've managed to create partition tables on files as well, that was some time ago actually. So maybe it stopped working for some versions, and then got fixed again
10:28 < kerframil> the box I chose has an old userspace; sorry for the confusion. in any case, file -s should also work.
10:30 < kerframil> but fdisk is a much nicer option
10:30 < rcf> kerframil: file -s is not quite as pretty, though it might be really useful if you hack dd to work with CHS addressing....
10:30 < kerframil> rcf: absolutely. it's not consistent with its output either.
10:31 < kerframil> seems to report the sector number in certain situations, but not others
10:33 < FManTropyx> bash gave me a syntax error near unexpected token
10:33 < kerframil> for which command?
10:33 < hexnewbie> Most syntax errors occur in the vicinity of unexpected tokens
10:33 < rcf> kerframil: it's just doing a very literal reading of the MBR, which is... not quite up to modern standards, as anyone with a disk >2TB knows very well.
10:34 < FManTropyx> fixed it, thanks!
10:34 < kerframil> rcf: so it would onl read the protective MBR for GPT. that explains it.
10:34 < kerframil> only*
10:35 < rcf> I didn't try it on a GPT disk (because all the full raw HD images I have at hand are from old computers) but that would definitely explain it.
10:37 < muffindrake> kerframil: Thanks, xdelta looks like a step in the right direction
10:39 < quint> I'm trying to use "salsa20-cbc-essiv:sha256" as a cipher using cryptsetup, but it returns "Check that kernel supports salsa20-cbc-essiv:sha256 cipher" .. How do I add support for this cipher to my kernel?
10:39 < ebar_pomasti> hi
10:42 < Juesto> how do i convert a linux install to uefi
10:42 < Juesto> or how i can boot on a per partition basis?
10:42 < Juesto> should i tell my distro to generate and detect systems?
10:43 < rcf> Juesto: what do you mean by boot on a per-partition basis?
10:43 < Juesto> this was possible in mbr, you just had to set a partition active
10:44 < rcf> In UEFI you need an EFI system partition which contains the bootloaders
10:44 < Juesto> yeah that's done i already have grub
10:45 < Juesto> now how do i load another grub configuration from partition?
10:45 < SuperSeriousCat> I _think_ all you need to do is change a flag on a partition, have an EFI partition and then grub-install with EFI + do grub-mkconfig -o
10:46 < Juesto> uh just add another distro is done that way?
10:46 < Juesto> or i boot once to the distro then reinstall grub as uefi correct?
10:47 < kerframil> quint: you might want to begin by looking at: zgrep CONFIG_CRYPTO /proc/config.gz | sort
10:47 < Juesto> and I'll not have grub conflicts when updates happen??
10:47 < kerframil> quint: if you don't have /proc/config.gz, then you may have a config under /boot
10:47 < rcf> Juesto: there might well be conflicts between distros, but the easy way out is just to install grub only on one of them.
10:47 < rcf> Or delete it from one, as the case may be.
10:48 < Juesto> yeah
10:48 < kerframil> quint: also, the options that begin with "CONFIG_CRYPTO_USER_API" are potentially important, as they provide a userspace API with which cryptsetup may need to interact with
10:48 < Juesto> Should any distro detect the different distro?
10:49 < Juesto> rcf, I'll just leave the self contained grub alone, but wouldn't it affect the disk?
10:49 < morf> there is that os-probe thingy
10:49 < Juesto> whenever it attempts to reinstalls the bootloader
10:49 < Juesto> what would happen
10:50 < quint> kerframil: what are the "m" values in that file exactly?
10:51 < kerframil> quint: it means that the option was enabled, and would be provided as a loadable kernel module
10:51 < kerframil> quint: if you don't see a similarly named module by running lsmod, you could use modprobe to load the module, for instance
10:52 < kerframil> quint: in most cases, modules are loaded as and when they are required. there are some exceptions.
10:52 < rcf> Juesto: if you have grub from one distro, and don't have it in the other, whenever it updates you'll be fine once you configure grub to detect the other distro.
10:52 < quint> kerframil: neat. thanks
10:53 < kerframil> quint: you need _SALSA20 and any other relevant options to be either 'y' or 'm'. if that's not the case, an alternate kernel - or a custom build - would be required.
10:55 < rcf> Juesto: no matter what happens though, you can boot always salvage the system by booting from the grub command line. It's not like Windows Update wiping EFI settings whenever the Windows loader changes.
10:57 < notmike> Beware. The spammer is coming.
11:00 < Juesto> hey
11:00 < Juesto> i have a question
11:00 < notmike> Ask it
11:00 < Juesto> how do i know what to pass to the kernel?
11:00 < Juesto> im on grub and i booted a system
11:00 < Juesto> but it complained about init= not passed
11:01 < notmike> That's a weird error. What kernel u running
11:03 < rcf> Juesto: the most important thing to pass is root=
11:03 < SuperSeriousCat> Doubt it is a kernel problem. This is after you changed from BIOS to EFI grub, Juesto?
11:04 < Juesto> SuperSeriousCat: no, nothing like that, I'm just trying to boot up the system by just sort of guessing
11:05 < hexnewbie> I don't believe I have ever passed init= to a kernel, expect when I was restoring root password on a non-initrd install.
11:05 < Juesto> or i could boot to my installed distro and update grub there i guess
11:05 < Juesto> I'm trying to boot manually into a install
11:06 < Juesto> it works via the root= ?
11:06 < Juesto> how do i pass something more easier to type than a uuid
11:07 < rcf> root=/dev/your/partition/goes/here
11:07 < Juesto> that works?
11:07 < rcf> Juesto: yes, as long as the partition exists, obviously.
11:07 < Juesto> in grub root is in the form of hd0,gpt1 etc
11:08 < Juesto> alright
11:08 < Juesto> and as for the initrd command do i pass something there?
11:09 < Juesto> nvm seems just nothing
11:09 < Juesto> nope
11:09 < Juesto> no init found it complains
11:10 < Juesto> lol it tried to mount the entire thing in /root
11:10 < Juesto> so yeah
11:11 < Juesto> how do i properly boot?
11:14 < rcf> For a proper introduction to manual grub booting I'd suggest https://www.linux.com/learn/how-rescue-non-booting-grub-2-linux%20
11:14 < Juesto> why is the system trying to mount into /root
11:14 < Juesto> I'm doing this for testing purposes then I'll regen grub efi under fedora
11:15 < rcf> Juesto: that means it can't find /etc/fstab
11:15 < Juesto> actually it's pointing somewhere else
11:15 < Juesto> should i edit it beforehand?
11:17 < rcf> Juesto: no, because if it could find your fstab in the first place it would be mounting things where you expect.
11:18 < Juesto> hm
11:18 < rcf> You do probably need an initrd.
11:20 < Juesto> yeah well
11:20 < rcf> On the BIOS system you'd find it in /boot alongside the kernel.
11:21 < rcf> Since you are trying this for testing and haven't reconfigured grub properly yet.
11:21 < Juesto> i would still need to reconfigure fstab regardless
11:22 < rcf> Only if you repartitioned the disk.
11:22 < qswz> genuine question, my bind -P shows `capitalize-word can be found on "\ec".` how would I trigger this? in my terminal?
11:22 < qswz> I've read http://www.softpanorama.org/Scripting/Shellorama/Bash_as_command_interpreter/inputrc.shtml, but cant find out
11:22 < Juesto> rcf technically yes because it's a completely different configuration
11:23 < Juesto> vm vs real
11:23 < Juesto> plus it's on sda9
11:24 < rcf> If your fstab was using device names and they've changed on a real computer, yeah, you'll want to edit that.
11:24 < Juesto> it's complaining because of invalid superblock checksum
11:24 < Juesto> and no fsck
11:25 < rcf> That's filesystem corruption.
11:25 < Juesto> I've just dd'd a partition
11:26 < qswz> you did dd
11:27 < Juesto> ...
11:33 < Juesto> now it booted
11:33 < Juesto> but i don't get it
11:33 < Juesto> why there isn't a fsck on initrd
11:34 < rcf> Juesto: That's a question for the distro maintainers.
11:36 < hexnewbie> Juesto: Was the filesystem mounted while you're using dd on it?
11:37 < Juesto> hexnewbie: i dd'd it over a existing one and i did unmount it while dding
11:37 < Juesto> it's fixed anyway
11:38 < hexnewbie> Unmounting *while* dding can cause filesystem corruption, in particular invalid superblock checksum is a possible outcome.
11:38 < qswz> What's the readline command for CapsLock, like \C- is for Control,  but CapsLock?
11:38 < hexnewbie> (Having it mounted while dding as well, but the unmount itself is a form of a write command, so like any write - more so)
11:39 < oleo> capslock is shift
11:39 < oleo> + some modifiers maybe
11:39 < qswz> ok thanks
11:39 < hexnewbie> qswz: CapsLock does not exist there. In terminals, you don't send key strokes but character sequences. CapsLock doesn't get sent to your terminal
11:40 < qswz> hmm, yea ok, makes sense
11:40 < hexnewbie> Control key also doesn't get sent, but you do get control characters in the end (such as Ctrl+A)
11:41 < qswz> I actully rebound it to TAB at xkb level
11:41 < Juesto> capslock is like holding shift, right?
11:41 < oleo> yes
11:41 < oleo> it's sticky
11:41 < oleo> until you release it
11:42 < oleo> or you can make it that way i suppose
11:43 < hexnewbie> Yeah. Shift is a shift which switches to capital letters. With Caps Lock holding it, partially - it's slightly different from Shift Lock, though. Control, on the other hand, is a shift that switches down to control characters, instead of up to capitals. Those used to happen on a low level in the teletype's keyboard, so the computer only got the small a, capital A, or Ctrl A.
11:48 < qswz> it's annoying that ~/inputrc overwites /etc/inputrc and not extends it
11:48 < qswz> ~/.inputrc*
11:50 < pankaj_> What is the difference between the usage of /mnt and /media. They both look like the same.
11:50 < hexnewbie> qswz: #include  /etc/inputrc
11:50 < hexnewbie> qswz: Errr, $include  /etc/inputrc
11:50 < qswz> ah thanks
11:50 < qswz> also how can I know if I have mode=emacs?
11:51 < qswz> yes $include works
11:56 < pankaj_> What is the difference between the usage of /mnt and /media. They both work the same.
11:57 < hassoon> anyone who got xbacklight working in debian stretch ?
11:57 < hassoon> i always get this error message:  No outputs have backlight property
12:03 < qswz> google that error prefixed by your OS distrib
12:03 < qswz> I fixed that a while back on my lubuntu
12:05 < afidegnum> hi, good morning, i m back again, i am having issue with my RAID arrais,
12:05 < afidegnum> arrays
12:05 < afidegnum> can anyone assist pls?
12:13 < afidegnum> my raid disk is 2Tb in size but due to initial configuration, i m having only 16GB to use, how do i make use of all the 2TB ?  https://ghostbin.com/paste/vuzeh
12:31 < noodlepie> Hiya guys! What's up today? Nice weather here, bit cloudy but sunny. @:P-~
12:31 < afidegnum> its sunny her
12:31 < afidegnum> here
12:33 < lovetruth> hello Ș)
12:33 < lovetruth> do you know any way in linux to darken & contrast scanned pdf files?... :)
12:38 < djph> lovetruth: change your scanner's settings
12:38 < lovetruth> can't, djph
12:39 < lovetruth> the files are already scanned and lost... :)
12:39 < lovetruth> unfortunately...
12:39 < blueglass> lovetruth: use adobe acrobat with wine
12:39 < blueglass> or in a VM
12:39 < blueglass> I recommend not attemptinng to use third-party tools to edit PDFs
12:39 < lovetruth> what can acrobat reader do?...
12:40 < djph> acrobat reader can't
12:40 < blueglass> oh, scanned
12:40 < blueglass> I'd just convert it to an image and edit in gimp
12:40 < blueglass> and then convert back to PDF
12:40 < lovetruth> I can manually do that in Gimp, but it's page after page and it's nasty, takes time, memory, etc...
12:40 < lovetruth> what I actually do it's this: I improve contrast, gama, brightness by using the  and then I duplicate the layer three times, I make the foreground layers color (white) to alpha and I slightly move the foreground layer a pixel to the right. Then the other foreground layer I move it slightly to down. And, eventually, I move the most foreground layer to left or up (and maybe continue with more layers, if needed, depending on the reso
12:40 < blueglass> maybe try image magick
12:40 < djph> you're somewhat outta luck, it's all jumping through hoops.
12:40 < blueglass> you could use the commandline utils
12:40 < blueglass> https://stackoverflow.com/questions/6605006/convert-pdf-to-image-with-high-resolution
12:40 < djph> I mean, maybe imagemagick; but even there, it's the whole 'convert it to image' bit first
12:41 < blueglass> I think you might be able to do it with just convert
12:41 < blueglass> andn script it pretty easily
12:41 < djph> oh, it's more the 'jump through all the hoops' bit.  If they're *his* scans; just fix the problem at the source
12:42 < lovetruth> the thing is that I don't think it's going to be really readable without overlapping layers, as I said above... :)
12:42 < blueglass> hmm
12:43 < lovetruth> and they are not my scans... they are from a copybook, as I' m going to have an exam in few days... :)
12:43 < lovetruth> advanced chemistry, quantum physics... :)
13:02 < pankaj_> I am practicing building applications from source. But downloading many of the dependency takes much time. Can somebody please give advice on downloading application and dependency and building them (any set of commands that can make this process easy) or any advice will be helpful.
13:02 < djph> wget
13:02 < djph> realistically, you can't fix that with anything other than "purchase better internet"
13:03 < djph> unless the problem is compile times.  In which case, "purchase a better PC"
13:04 < lovetruth> pankaj_: perl?...
13:04 < lovetruth> python?...
13:05 < luke-jr> pankaj_: Gentoo with emerge -f
13:10 < sodhi> I'm afraid I've borked my LVM-setup (server running proxmox) by accidentlally removing the logical volume 'pve/data' using vgreduce --removemissing. Is anyone able to assist me possibly recreating the volume and/or data?
13:11 < djph> restore from the backups you take.
13:11 < sodhi> That's cute. If I had backups, I wouldn't be here, would I ? :)
13:11 < pankaj_> luke-jr: I am using archlinux right now. But have seen lfs and tried it. Soon going again and may be this time be successfull. But I was talking about if their are tricks or tools to fasten the compilation and build process of package with multiple dependency (if this is possible via some commands or tricks) rather then finding and downloading all packages one by one and installing with concern on
13:11 < djph> well, you're fucked.
13:12 < pankaj_> dependency.
13:12 < sodhi> i.e. the data is lost?
13:13 < sodhi> Maybe I'm missing something here, but how would losing a logical volume lose the data, if the physical volume is still there?
13:13 < sodhi> Heck, there are still 2 working logical volumes on the physical volumes.
13:14 < luke-jr> pankaj_: that's why I suggested Gentoo; it automates stuff for you
13:15 < luke-jr> emerge -f automatically downloads the source code for a given program and all its dependencies
13:15 < pankaj_> luke-jr: But I also want to understand what happens when I build a package (properly) or how to build a large package easily rather then using easy way like using package manager.
13:15 < TheWild> hello
13:16 < luke-jr> pankaj_: emerge -f *only* downloads the sources
13:16 < pankaj_> luke-jr: But I have not tried gentoo. What is difference between arch and gentoo? Can you please tell.
13:16 < pankaj_> luke-jr: OK. It sounds cool.
13:16 < luke-jr> I'm not sure there's really anything in common between Arch and Gentoo. Arch is a binary distro; Gentoo builds everything from source.
13:17 < pankaj_> luke-jr: How much near is it to lfs?
13:17 < Sapphirus> Arch can be configured to be a source/binary hybrid.
13:17 < luke-jr> simply using Gentoo gives you all the options that building from source gives you; but you could in theory just build what you want manually too
13:17 < Sapphirus> That's what I love about it.
13:17 < luke-jr> Sapphirus: not really
13:18 < RustyJ> anyone decent with zend? i have an app i need hosted.
13:18 < Sapphirus> luke-jr, I can easily build replacement packages and slip them in.
13:18 < Sapphirus> I'd rather do that than build everything from source.
13:18 < luke-jr> Sapphirus: you can with Debian too
13:18 < BluesKaj> Hiyas all
13:18 < luke-jr> Sapphirus: but you don't get any of the build-time options easily available for example
13:19 < Sapphirus> PKGBUILDs suffice.
13:19 < Sapphirus> I'd rather not build my whole system from scratch.
13:20 < luke-jr> Sapphirus: that's your choice of course, but to pretend it's comparable is overlooking a lot
13:21 < TheWild> do you remember that somewhat bad HDD I was talking about some time ago. One morning it started producing IO errors. I took a look at SMART and it got "Current pending sectors: 3". I shut it down immediately, got some storage then attempted to recover the partition I haven't backed up. On 50 GB of data it couldn't read about 13 MB. Damaged sectors numbers ranged from 67095824 to 67122415 (= 0x3FFCD10 to 0x40034EF - very suspicious
13:22 < compdoc> no
13:22 < TheWild> after recovery attempt, the "Current pending sectors" count hit to about 600
13:22 < luke-jr> I'm not sure sector numbers have any relation to the physical disk
13:22 < TheWild> I scanned it once more and the range of bad sectors decreased somewhat.
13:22 < compdoc> pending sectors are bad. did you get your data?
13:23 < TheWild> I checked SMART today, and to my surprise "Current pending sectors" is now 16 and the previously bad sectors are readable again. "Reallocated Event Count" is still 0.
13:23 < TheWild> wtf? Was this HDD packed with some dust inside?
13:24 < Sapphirus> luke-jr, I'm making a oversimplified comparison.
13:24 < Sapphirus> There's no need to get into semantics.
13:25 < TheWild> I lost 8 MB of data (+ 5 MB unallocated space ;)), but I tracked it down to be a bunch of some flash games.
13:25 < TheWild> I backed remaining stuff up of course.
14:36 < hexnewbie> TheWild: Happens often. It's why one needs to keep backups.
14:38 < hexnewbie> It makes it difficult to decide what to do with warranty returns - a drive with hundreds of badblocks according to badblocks and multiple pending sectors, now has 0 reallocated or pending sectors. So, on what grounds do we return this, or do we simply put the ‘Entertainment collection’ on it?
14:44 < SkunkyFone> hexnewbie: run a wipe on it over and over until it shows the failure in smart?  don't put anything on it you want to keep.
14:49 < TheWild> hexnewbie: warranty of mine (that bad one) expired about 4 years ago
14:49 < TheWild> once it went bad, it's now used as a "scratchpad CPU register" ;D
14:50 < TheWild> could serve also as a quite large swap space
14:50 < hexnewbie> That name is intended to be humorous I presume, because that's like naming a snail ‘space shuttle’.
14:51 < hexnewbie> Although, on a second thought, that's a good name for your pet snail
14:52 < TheWild> - It did not move for two days. Is it dead?
14:52 < TheWild> - Oh, come on. It's downloading updates.
15:02 < cousin_luigi> Greetings.
15:02 < cousin_luigi> Anyone with experience in fixing DSDT tables?
15:03 < mawk> probably
15:03 < mawk> someone here made a website about that
15:03 < mawk> but I can't remember who
15:05 < forgon> Often when I listen to something, I decrease the sound volume via alsamixer (or amixer). This is only necessary for some files, but quite tedious.
15:06 < forgon> Is there a technology that will automatically adjust audio output so that it never exceeds a particular loudness?
15:09 < MrElendig> forgon: yes
15:10 < MrElendig> can sort of do it in software, most reliable is to do it in hardware
15:10 < MrElendig> read up on replaygain etc
15:10 < rascul> there are tools to normalize the volume of music collections
15:11 < MrElendig> the problem is defining "loudness"
15:11 < MrElendig> clipping at a certain amplitude is easy, but probably won't do whta you want
15:12 < mawk> dividing all amplitudes by the right coefficient ?
15:12 < mawk> but amplitudes aren't meaningful to humans
15:13 < rascul> it's also an issue with tv, turn the volume up for one channel and the next channel is blasting you, also throw some commercials in the mix and you might constantly be adjusting volume
15:13 < mawk> better talk about volume
15:14 < rascul> MrElendig i wasn't aware there were hardware solutions, but it doesn't surprise me, i guess i just never looked into it
15:14 < forgon> mawk: I didn't want to be too specific, since I don't know what I'm talking about :|
15:16 < MrElendig> https://en.wikipedia.org/wiki/ReplayGain
15:47 < afidegnum> i m having this error /usr/sbin/grub-probe: error: cannot find a device for / (is /dev mounted?).
15:51 < MrElendig> more context needed
15:51 < MrElendig> are you in a chroot?
15:51 < afidegnum> yes
15:51 < MrElendig> so you forgot to bind sys, dev etc then
15:51 < MrElendig> exit the chroot and fix that
15:51 < revel> (don't bindmount /etc)
15:52 < MrElendig> https://wiki.archlinux.org/index.php/Change_root#Using_chroot
15:52 < MrElendig> mostly distro agnostic
15:52 < afidegnum> MrElendig: can you please give a hint?
15:52 < MrElendig> I just did
15:53 < afidegnum> what's the command to fix it ?
15:53 < MrElendig> how about reading that link?
15:53 < Alexander-47u> hi all
15:54 < Alexander-47u> is there any tool that works like a C&C but without all the malicious acts built in it? i want to be able to access a bunch of 3g connected raspberry pi's
15:54 < revel> MrElendig: Is that "mount --rbind" warning systemd-specific? Since it doesn't have that effect on my system.
15:54 < revel> And lsof reported systemd as being the thing blocking it.
15:55 < MrElendig> ymmv
15:55 < revel> (on a systemd machine, it did, anyway)
15:55 < Alexander-47u> anyone?
15:59 < spare> ircd servers work well for multiplexing remote shells
15:59 < revel> Oh, --make-rslave works for systemd, maybe.
16:00 < Alexander-47u> nobody :P?
16:00 < MrElendig> Alexander-47u: there are lots of those, depending on just what you are trying to do
16:00 < klotz> pacman -Qi aspell
16:00 < klotz> sudo pacman -S aspell-en
16:00 < klotz> worldeater
16:00 < klotz> y
16:00 < MrElendig> quite a lot of pub/sub based cluster software
16:01 < Alexander-47u> well, im trying to gain command line behind a firewalled environment for a bunch of 3g connected raspberry pi's
16:01 < MrElendig> everyone and their grandmother at universities makes one of those for their master
16:01 < klotz> lolz sry, testing new config
16:01 < Guest57044> afidegnum: so you're asking here too now
16:01 < Guest57044> after spamming 3 other random channels
16:02 < Guest57044> afidegnum: just join proxmax and ask in there, it's the right place
16:02 < Guest57044> afidegnum: but again, why are you touching grub if you're just changing your home directory disk
16:02 < afidegnum> Guest57044: stop that, i m stuck in the mess and you are stalking me,
16:02 < Guest57044> afidegnum: I'm not stalking you, I'm already in the channel
16:02 < spare> Alexander-47u: https://github.com/tcejorperaps/gentoo/raw/master/mount/misc/ash/freeshell < - shell script that throws shells to freenode locked to the nick supplied just run your own ircd takes two minutes to irc based remote shells
16:02 < Guest57044> afidegnum: and you stop cross-posting
16:02 < Guest57044> use the correct channel which is proxmax
16:02 < Guest57044> proxmox sorry
16:03 < afidegnum> this is a VM created
16:03 < Guest57044> so ?
16:03 < afidegnum> i m working in a Linux VM,
16:03 < Guest57044> a.) your home directory means you should not be touching grub b.) proxmox is the right channel
16:03 < Guest57044> address those two problems
16:03 < afidegnum> i explained earlier i m mapping /home to the HDD raid
16:03 < Guest57044> right, so you don't need to touch grub
16:04 < Guest57044> as grub has nothing to do with your home directory
16:04 < afidegnum> i have been asked to add entries to the fstab which i did
16:04 < afidegnum> i m following the given steps
16:04 < Guest57044> ok, so ?
16:04 < Guest57044> your steps are wrong
16:04 < Guest57044> you don't need to touch grub
16:04 < afidegnum> that's where i m having this erro,
16:04 < MrElendig> seems like reading isn't his best skill
16:04 < Alexander-47u> spare, i was looking more for nc kinda shell
16:04 < Guest57044> no, you're having this error for other reasons
16:05 < Alexander-47u> im not doing anything weird/hackish
16:05 < Alexander-47u> i really just need simple command line access to the pi's, but since the connection needs to be initiated from the raspberry pi, all the C&C seem to fit the use case, but i dont want the weird features such as keylogging added
16:05 < MrElendig> Alexander-47u: if you are just going to send commands then look at things like fabric/celery
16:06 < spare> you can swap out openssl s_client for nc and use it :P
16:06 < MrElendig> or one of the pub/sub ones
16:06 < MrElendig> pub/sub is nice for things that are not 24/7 avaialable
16:06 < MrElendig> available*
16:07 < spare> if you do reverse shells without encryption your literally throwing a shell at anyone along the path to mitm on : /
16:08 < Alexander-47u> yes some encryption would be nice of course, also different authentication per client would be great
16:09 < MrElendig> there are 0 reasons to not atleast do tls >= 1.2
16:09 < MrElendig> pinned selfsigned certs are fine for the task
16:09 < Alexander-47u> celery seems old, fabric uses ssh so that wont work
16:09 < spare> you can switch out netcat with openssl s_client and it will do full client / server x509 and then literally works like netcat read writing to a fifo
16:10 < afidegnum> this is one of the reason why / is not mounted, https://ghostbin.com/paste/s35dt
16:10 < Alexander-47u> i cannot open any ports on 3G (dahr)
16:10 < Alexander-47u> so it has to be a reverse shell
16:10 < MrElendig> "despite" being old, celery is really useful
16:10 < spare> srop stunnel infront of a irc server and you can do full ecc client / server authentication
16:10 < Alexander-47u> MrElendig, does it work reverse shell'ish?
16:10 < MrElendig> and it is still maintained
16:11 < MrElendig> Alexander-47u: it is a task queu
16:11 < MrElendig> you can use anything you like as the transport
16:12 < Alexander-47u> can i use it with 3g clients ?
16:12 < MrElendig> sure
16:12 < Alexander-47u> like creating a service to execute the executable that then initiates the connection
16:12 < Alexander-47u> to my vps
16:12 < MrElendig> could even send the commands over sms if you like
16:14 < Alexander-47u> celery seems to be used for preconfigured tasks
16:14 < Alexander-47u> i just want to drop a shell
16:14 < Alexander-47u> just like i would get with ssh
16:14 < MrElendig> you would put that "shell" on top of celery
16:15 < MrElendig> also look at mqtt etc
16:17 < Guest57044> afidegnum: why are you not listening
16:18 < Guest57044> afidegnum: you don't need to run grub if you're just changing your home directory disk
16:23 < afidegnum> i did without running the grub, and was locked out,
16:24 < afidegnum> i needed to wipe everything and restart afresh
16:24 < afidegnum> this is a remote server, i only have access to it via te terminal
16:25 < Guest57044> that has nothing to do with it
16:25 < Guest57044> changing /home is nothing to do with grub
16:25 < Alexander-47u> MrElendig, isnt there anymore ready-to-use? :P
16:26 < mawk> my LTE USB stick is running linux and I want to hack it a bit but it looks like that networking chip handling is really locked down by qualcomm/Android, I can't find any docs whatsoever
16:26 < mawk> isn't there like an open driver for that ?
16:27 < Guest57044> you've got a usb stick that actually runs linux /
16:28 < Guest57044> do you mean you're using your usb stick with linux ?
16:28 < Guest57044> eg: plugging it into a linux machine
16:28 < mawk> no no, the usb stick runs linux
16:28 < mawk> I was surprised too
16:28 < mawk> I thought that a LTE stick just needed some specialized µC for the usb logic and that's all
16:28 < mawk> but there is a full linux in it, apparently android-based
16:29 < Guest57044> how is a usb stick running linux ?
16:29 < mawk> with a microprocessor ?
16:29 < mawk> https://www.ebay.fr/itm/271819147163
16:29 < mawk> it's that thing
16:29 < Guest57044> ahhh so is't a modem
16:29 < Guest57044> you mean the firmware on it that drives the hardware
16:30 < mawk> yeah
16:30 < Guest57044> you don't want to touch that though
16:31 < RebelCoderRU> Guys, how do I add Alias to my local server IP? I want to type ssh myserver instead of ssh user@IP
16:31 < Alexander-47u> i was looking for something more like this
16:31 < Alexander-47u> https://cockpit-project.org/
16:31 < mawk> why not ? I have a few modifications to make
16:31 < mawk> like, make that annoying LED stop blinking
16:31 < Alexander-47u> but for clients that cant open ports
16:31 < mawk> or make it remember my PIN code so that I don't have to type it everytime
16:32 < mawk> I opened it and there are UART pins so that I can have an emergency shell if I break it
16:32 < Guest57044> mawk: is it not going to be a ROM that is locked
16:33 < Guest57044> only put into write mode with the software update tool
16:33 < mawk> what ?
16:33 < mawk> I'm already into the stick
16:33 < Guest57044> (I'm askig)
16:33 < mawk> I have a root shell
16:33 < Guest57044> ahhh
16:33 < mawk> ah
16:33 < Guest57044> mawk: so is the actual file system a read/write file system
16:33 < Alexander-47u> remote administration without opening ports? anyone know good tools?
16:33 < Guest57044> eg: if you made a change and rebooted would it stick
16:34 < Guest57044> Alexander-47u: how can you connect to a remote device without opening ports ?
16:34 < mawk> / is a temporary file system unpacked from a rom, but there are persisting directories yes
16:34 < Guest57044> mawk: how is the persistant file system stored/saved
16:34 < Guest57044> (I'm curious)
16:34 < mawk> it's a flash partition
16:34 < Alexander-47u> reverse connection, the connection is initiated from the client, just like when you use teamviewer
16:35 < mawk> it's just mounted from the embedded flash memory
16:35 < mawk> it's a "yaffs2" filesystem
16:35 < tds> Alexander-47u: if it's a linux box, reverse ssh tunnel + autossh as a service would work
16:35 < Guest57044> mawk: very interesting
16:35 < Alexander-47u> tds, that is exactly what i have right now
16:35 < mawk> yeah
16:35 < spare> Alexander-47u: you can take 4 lines from the irc rfc and make bash through a shell at freenode dont think it gets more simple than that
16:35 < mawk> the flash partitions look like /dev/mtdblock18
16:35 < Alexander-47u> tds, but that is not feasible when talking about hundreds of hosts
16:36 < Alexander-47u> maybe thousands :p
16:36 < mawk> also some flash partition look like it's mounted on / so it may be persisting after all, but it has a strange name; let me try if anything persists
16:36 < spare> create a channel set it moderated admin only and you have one to many command execution or invite to smaller channels and group them its really easy /shrug anything complex script a client side pull and execute
16:36 < Alexander-47u> Guest57044, these is called 'reverse' connections
16:37 < Alexander-47u> are*
16:37 < Alexander-47u> lol
16:37 < tds> Alexander-47u: are all of these devices in a single location? if so I guess openvpn/whatever running on the router would let you easily jump to any of them
16:37 < Guest57044> Alexander-47u: thats still opening ports
16:37 < Guest57044> just outgoing ones
16:37 < Guest57044> and it means you need incoming ports on your local machine
16:37 < Guest57044> so again, opening ports
16:37 < tds> I assume the use case is just escaping a NATed network, effectively?
16:37 < mawk> yes everything is persisted actually Guest57044
16:37 < Alexander-47u> yes but if all your ports are closed, then you dont have any network connections :D
16:37 < mawk> that dongle is surprising
16:37 < Guest57044> mawk: very interesting (and surprising)
16:38 < Guest57044> I'd have at least expected / to be on a ROM base setup
16:38 < mawk> yeah
16:38 < Alexander-47u> im talking from a completely legit perspective, no IDS and stuff is at play here
16:38 < Alexander-47u> its about managing a shitload of raspberry pi's
16:38 < tds> well you don't have to have traffic with ports at all, there are protocols other than udp/tcp ;)
16:39 < Alexander-47u> yes
16:39 < mawk> they made a mess out of the initrd tho, instead of moving /newroot to / and chrooting into it, they just mounted it to / directly, so the initrd files are still under the mountpoint
16:39 < mawk> also the scripts are full of typos and broken english
16:39 < tds> Alexander-47u: I saw a talk from someone doing something similar a little while ago, I think they were using a tool called teleport?
16:39 < mawk> but the web pages look like well coded
16:40 < mawk> Alexander-47u: I'd use a vpn with a persistent keepalive
16:40 < mawk> sending probes every 25 seconds to keep the NAT mapping open on the firewall
16:40 < mawk> then any reverse connection can occur
16:40 < mawk> wireguard is perfect for that usage imo
16:41 < Alexander-47u> im trying to keep the client side setup as simple as possible, for example just executing a binary/script to initiate the reverse connection
16:42 < mawk> it could sound invasive but wireguard has very good performances, 1Gbps on a 1Gbps line; but if you don't want that you can code/look for something based on detecting publics ports using STUN, à la SIP
16:42 < mawk> or just making the clients initiate the tcp session, and use that
16:42 < mawk> you have the choice
16:42 < Alexander-47u> tds, seems to fit the use case :)
16:43 < Alexander-47u> but teleport isnt free :D
16:43 < tds> isn't there some free community/open source/whatever version?
16:43 < tds> btw, this was the talk, in case you care: https://www.youtube.com/watch?v=l-fYg0B7c00
16:43 < Alexander-47u> oh ye shit i see lol
16:44 < Alexander-47u> thanks
16:44 < Alexander-47u> they also have a ARM binary, perfect
16:44 < tds> personally I think I'd still go with openvpn/whatever as mawk suggested, but it's your call
16:44 < Alexander-47u> i would also do that
16:45 < Alexander-47u> but im trying to keep this as simple as possible
16:45 < mawk> wireguard is very light
16:45 < mawk> compared to openvpn or ipsec
16:45 < Alexander-47u> because the company that is going to use this prototype setup, isn't an IT company, so ye
16:45 < mawk> configuration is very simple too, you just copy-paste a public key
16:46 < Alexander-47u> i want no' configuration, just binary execution and connection starts, that would be great :P
16:50 < spare> if your really doing that at scale you still need round robin backend all the clients need to generate entropy so you dont ddos you c&c aslong as you can push/pull updates without killing your infra its all pretty easy
16:51 < tds> I'd be tempted to write something custom to do the config for openvpn/wireguard/whatever and then package it up to easily install it, then just identify the pis eg by serial number
16:54 < spare> dont think vpns make a good command channel : / but even things like spreading out your keep alive pings and having clients generate a random sleep timeout before responding to one to many commands
17:05 < screwsss> any linux users in here
17:06 < Volleo> ...yes
17:06 < Alexander-47u> lol
17:06 < Alexander-47u> no, we all use mac
17:07 < Volleo> </sarcasm>
17:09 < sauvin> Screw your Mac, I'm a WINDOWS user!
17:09 < Volleo> so there's more joins/leaves than actual messages. k
17:09  * sauvin hides
17:10 < sauvin> Volleo,  check out the channel's *size*.
17:10 < Alexander-47u> well, irc is really old.
17:12 < Volleo> 777 isn't jackpot for you, but it is jackpot for whoever is trying to find a security hole in your sistem
17:12 < Volleo> system*
17:12 < Alexander-47u> wut?
17:13 < Volleo> chmod 777 file.name # ls -l will output -rwxrwxrwx
17:13 < Volleo> 777 means everyone can read/write/execute
17:16 < Alexander-47u> ye i know that, but why do you start talking about that all of a sudden lol
17:16 < Volleo> because the chat is very inactive except for all these people joining/leaving
17:17 < misternumberone> Volleo I am waiting for people to talk so I can ask about my problem lol
17:18 < Alexander-47u> just ask man
17:18 < Volleo> oh.
17:18 < misternumberone>  hi, on multiple systems running multiple versions of debian, one of which is the latest debian 9.4 stretch DebianLive CD image,installing the package grub2-common and then running the command grub-mkrescue in any configuration that includes the option "--compress" prints "Segmentation fault" and exits
17:18 < misternumberone> for example, sudo grub-mkrescue --modules="lvm cryptodisk scsi ahci disk luks" --compress xz -o floppy.img ./img, or alternatively simply grub-mkrescue --compress
17:18 < BluesKaj> Volleo, not true , try disabling joins/parts ion your client
17:18 < misternumberone>  all calls I have tested that do not include "--compress" appear to work as expected
17:18 < Alexander-47u> in irc, with answers, you always have leechers and seeders
17:18 < Alexander-47u> maybe one day i'll be a seeder as well lol.
17:20 < Volleo> I don't know... segfaults give you 139 on bash for some reason (128 + 11, where 11 is the numerical value of SIGSEGV), but I would have no idea why it's doing that
17:20 < Volleo> it's accessing memory that it shouldn't ... is it not allocating enough?
17:23 < misternumberone> Volleo: minor detail: when sudo is used "Segmentation fault" does not print rather the command just does nothing. When sudo is not used nothing still happens but "Segmentation fault" prints.
17:24 < Volleo> maybe that's just sudo... nevermind about that
17:24 < sudo_halt> hmmmm.....
17:26 < qrvpzvb> is anyone using OpenIndiana ?
17:26 < Volleo> anyway there's like ten messages in five minutes, so SIGHUP
17:27 < ziggylazer> jim, online?
17:27 < ziggylazer> Or anyone with some vsftpd knowledge?
17:27 < ziggylazer> On a RHEL server
17:33 < Alexander-47u> btw coming back on the stuff
17:33 < Alexander-47u> remote admin for thousands of pi's
17:33 < Alexander-47u> remot3.it seems to be perfect.
17:44 < rascul> i want thousands of pies
17:45 < rascul> not all at once though, i couldn't eat them fast enough
17:46 < tds> Alexander-47u: hmm, I thought you weren't keen on the other project due to the cost?
17:46 < tds> if it works for you then fair enough, though :)
17:48 < Alexander-47u> tds, indeed but this fits the case to good lol.
17:48 < Alexander-47u> too*
17:49 < Alexander-47u> i will present to them a free alternative and a paid one, with remot3.it / dataplicity
17:49 < Alexander-47u> which are dedicated to IoT devices, which the raspberry pi's will be used as.
17:51 < Alexander-47u> thanks for all the input though :)
17:52 < Alexander-47u> tds, the reason that this catches my attention is the ease of use
17:52 < tds> if you're deploying a pi-based iot solution, make sure you have a good way to push out updates as well :)
17:52 < Alexander-47u> yes, this remot3.it has it all,.
17:53 < Alexander-47u> the ease of use is necessary because the company receiving this prototype is a water dispenser company, so i'm guessing they don't have a very tech savvy R&D department
18:00 < pankaj> In the grub menu I cannot use my arrow keys to toggle around as I have to dual boot to another destro that I have set up. I googled but it talks about old computer keyboards and their interface problem but I have laptop.
18:02 < screwsss> any linux users in here
18:02 < screwsss> lol i mean, kali linux...
18:04 < pankaj> In the grub menu I cannot use my arrow keys to toggle around as I have to dual boot to another destro that I have set up. I googled but it talks about old computer keyboards and their interface problem but I have laptop.
18:05 < Pentode> what kind of laptop is it?
18:05 < djph> is this your (still probably very broken) LFS install?
18:07 < pankaj> Pentode: I am using toshiba satellite c850
18:07 < pankaj> djph: Me?
18:10 < Pentode> i dont see any reason it shouldnt work, unless it's one of those wacky laptops that has an i2c keyboard
18:13 < pankaj> Pentode: So, is their any solution to it?
18:15 < screwsss> Im back
18:15 < Pentode> short of plugging in a usb keyboard, probably not.
18:23 < pankaj> djph: Hello.
18:30 < pankaj> who #.
18:30 < mattfly> hi
18:30 < pankaj> mattfly: Hello
18:31 < sirwilliam> mattfly: hillo
18:32 < mattfly> I am trying to like set up an audio studio with kxstudio and im having  multiple jack windows and spectrometers and waveform view in a bunch of windows, is there any thing that can colapse this to a single window, like in a docked view?
18:33 < mattfly> like if there was a tilling window application that works as a i3wm inside another DE
18:33 < mattfly> lol i might be wanting something impossible
18:34 < pankaj> mattfly: I use awesome window manager which is absolutely suitable for these types of work. It is also a tiling window manager.
18:34 < mattfly> so i could like replace kwin with that?
18:35 < phogg> mattfly: you could
18:35 < mattfly> interesting
18:35 < mattfly> it works kinda like i3 ?
18:35 < phogg> mattfly: pretty sure kwin has a tiling mode of some kind, though
18:36 < phogg> like maybe this https://github.com/faho/kwin-tiling
18:36 < pankaj> mattfly: Yes
18:36 < rascul> yeah it's had it for awhile... i seem to remember it at least as far back as kde3
18:37 < rascul> i don't care for tiling, so i have no clue how good it is
18:37 < phogg> indeed
18:37 < phogg> kwin is actually quite extensively powerful; the usual KDE setup doesn't scratch the surface
18:37 < phogg> they only expose so many knobs in the UI
18:37 < Henry151> howdy-hey ##linux :)
18:38 < mattfly> hm thanks but it doesnt handle activities :-(
18:38 < brutser> to disable screensaver locking, i added this line to kickstarter file, but it gives a problem: gconftool-2 --direct --config-source=xml:readwrite:/etc/gconf/gconf.xml.defaults -s -t bool /apps/gnome-screensaver/lock_enabled false >/dev/null
18:38 < brutser> Warning: Schema ?org.gnome.crypto.cache? has path ?/desktop/gnome/crypto/cache/?.  Paths starting with ?/apps/?, ?/desktop/? or ?/system/? are deprecated.
18:38 < Henry151> I really like my colors for my bash prompt and all, but I want to use tmux, and when I do, it makes things black-and-white again. Any guidance?
18:38 < brutser> how can i fix this?
18:39 < rascul> iirc even windows 3.1 could do some tiling ;)
18:40 < phogg> mattfly: for your scenario what I do (in e16) is set all of the windows involved to borderless, then I position then just as I want, mark their position as remembered by the WM so they reopen there every time
18:40 < phogg> it takes a bit of tweaking to get the sizes and positions just right (not automatic like in a tiling WM) but it works for me
18:41 < phogg> Henry151: which colors?
18:41 < phogg> Henry151: why are colors important when you run tmux?
18:42 < Henry151> phogg: henry@thinkpad:~$ echo $PS1
18:42 < Henry151> \[\e]0;\u@\h: \w\a\]${debian_chroot:+($debian_chroot)}\[\033[01;32m\]\u@\h\[\033[00m\]:\[\033[01;34m\]\w$(__git_ps1 " (%s)")\[\033[00m\]\$
18:42 < Henry151> colors are not important, but I like them
18:42 < Henry151> green if user, red if root
18:42 < DLange> Henry151: and inside tmux? What is PS1 there?
18:43 < brutser> it seems i cannot use gconftool-2 anymore in kickstarter?
18:43 < Henry151> but i have started using tmux a lot and ${debian_chroot:+($debian_chroot)}\u@\h:\w\$
18:43 < Henry151> i'm not sure where to set it.
18:43 < phogg> Henry151: why does your PS1 change inside tmux? There's no reason it should.
18:43 < Henry151> well that's what I'm here to ask about i suppose..
18:44 < phogg> Henry151: PS1 is set from your shell's startup script e.g. ~/.bashrc if you have one.
18:44 < Henry151> phogg: yeah, that's where i set all my preferred colors
18:44 < phogg> Henry151: unless you tell it to do otherwise tmux will invoke your default shell with default options which will read from default locations and do default things, which will pick up any ~/.bashrc you have created with any fancy PS1. So what are you doing that's not usual?
18:45 < Henry151> but in tmux, when starting tmux with just the command "tmux," it gives me black and white
18:45 < phogg> Henry151: then you have configured tmux to do something unusual
18:45 < Henry151> it's freshly installed with "sudo apt install tmux -y"
18:45 < Henry151> and I haven't touched any tmux configs
18:45 < phogg> Henry151: If your shell is usually bash run bash manually inside tmux. Does PS1 change?
18:46 < phogg> Henry151: Are you setting PS1 yourself inside ~/.bashrc or relying on your distributions /etc/bash.bashrc or the like?
18:46 < Henry151> phogg: nope it stays the same
18:46 < Henry151> setting it in ~/.bashrc
18:46 < DLange> Henry151: type bash    inside a tmux window
18:46 < phogg> Henry151: start bash inside tmux with: bash --rcfile ~/.bashrc
18:46 < DLange> looks better?
18:46 < phogg> DLange: he just did
18:47 < Henry151> phogg: still doesn't give me color
18:47 < Henry151> here's my .bashrc
18:47 < Henry151> http://termbin.com/vcb01
18:47 < phogg> Henry151: well there's your problem
18:47 < phogg> Henry151: You are setting PS1 only when TERM looks a specific way. Why?
18:48 < Henry151> I have some stuff set up for git in there
18:48 < DLange> echo $TERM
18:48 < phogg> Henry151: case "$TERM" in xterm*|rxvt*) PS1= ... ;; esac
18:48 < phogg> TERM won
18:48 < Henry151> phogg: I made that .bashrc a long time ago and have been using it without changing it/looking at it ever since
18:48 < phogg> er, TERM will not be xterm or rxvt inside tmux
18:49 < phogg> Henry151: That's stupid. Just don't check TERM, virtually every terminal will support those escape codes anyway
18:49 < ziggylazer> jim,!
18:49 < Henry151> phogg: how and where can I fix it?
18:49 < phogg> Henry151: you do it in two places; remove both
18:50 < Henry151> remove lines 43, 45, 73, 79?
18:50 < DLange> or set force_color_prompt=yes
18:51 < rascul> what is the PS1 that you are trying to set?
18:51 < rascul> oh ignore that, was scrolled up and didn't realize it
18:56 < Henry151> phogg: thanks for the guidance, got it fixed
18:56 < lotus> Hiya~  I have the "dreaded grub prompt" and I followed http://fedoraproject.org/wiki/GRUB_2#Encountering_the_dreaded_GRUB_2_boot_prompt  -- problem is I have an encrypted luks partition.  After running the commands in the guide, I was able to enter my LUKS password; this led me to a dracut CLI prompt (after some timeouts) -- I did a blkid at the dracut prompt after the timeout and got the UUID of the mapper partition containing my root 
18:56 < Juesto> hey
18:57 < lotus> :waves to Juesto:
18:57 < Juesto> about grub
18:57 < Juesto> i got something working and kind of safe guards happening
18:57 < Juesto> grub knows it's a efi system and tries to add entry to firmware
18:57 < Juesto> but esp isn't mounted
19:10 < Roey> hi.  Why do I get this error?  https://pastebin.com/Prh50khz  <-- I am attaching an external backup hard drive and this is what I see from /var/log/syslog when I turn it on.  "lsblk" doesn't list the drive at all, either.
19:10 < cberg> am I seeing this right that when I have two LUKS volumes sharing the same passphrase, I cannot use an unlocked one to unlock the other one, without actually knowing the passphrase, right?
19:11 < ayecee> drive is dead. press F to pay respects.
19:12 < Roey> ayecee: to whom are you responding?
19:12 < ayecee> to you. i didn't think it was that ambiguous.
19:14 < Roey> ayecee: there was another person here talking about LUKS volumes, who knows.
19:14 < Roey> anyway,
19:14 < ayecee> anyway
19:14 < cberg> mine was a hypothetical question though
19:14 < Roey> ayecee: how positive are you that it's the drive?
19:14 < ayecee> about a 4
19:14 < Roey> ayecee: and not say the encosure?
19:15 < Roey> also, dmesg shows me too many messages like "hpet1: lost 6 rtc interrupts".  Couldn't find a real fix for it online.  Have you seen this error before?
19:15 < ayecee> all i can tell for this is that the drive stopped responding shortly after connected.
19:15 < Roey> ayecee: ahh, ok
19:15 < ayecee> i am not good at diagnosing "something like" errors.
19:17 < Roey> ayecee: ok, it says that shit
19:17 < Roey> again and again and again.
19:17 < Roey> ayecee: not "like this", but "this exactly".
19:17 < Roey> repeatedly.
19:17 < ayecee> that error alone says little.
19:17 < Roey> you can gogole for it and enough folks ahve apparently seen this problem
19:18 < ayecee> the context that it appears in and the events leading up to it are what's important.
19:18 < DLange> does the enclosure have external power?
19:18 < DLange> 'cause often the USB ports just don't supply enough current
19:19 < lotus> GRUB2 prompt, encrypted drive, halp -- this tut was great but only for non-encrypted drives :(  My encryption screen popped up after this tut, but entering password sent me into a loop that timed out eventually
19:19 < lotus> http://fedoraproject.org/wiki/GRUB_2#Encountering_the_dreaded_GRUB_2_boot_prompt  <  tut
19:20 < syborg> lotus, what kind of loop? Did it say anything?
19:20 < lotus> yeah it is a dracut-initqueue loop
19:20 < Roey> DLange: it does, yes
19:20 < Roey> DLange: and it spins up
19:20 < Juesto> hey
19:20 < lotus> Warning: dracut-initqueue timeout - starting timeout scripts
19:21 < Juesto> so I'm having a little of trouble with grub
19:21 < lotus> @syborg ^
19:21 < syborg> yeah gotit
19:21 < lotus> ty
19:22 < syborg> lotus, I don't suppose you have an emergency/recovery boot option?
19:22 < lotus> I was able to boot into windows by selecting that partition as the primary boot option
19:23 < syborg> ew not windows =P
19:23 < syborg> for fedora
19:23 < lotus> Otherwise, my bootloader does not even open -- I was able to use `ls` in the grub menu and identify my /boot as (hd0,5) i think
19:23 < lotus> I have a Ubuntu live and maybe even a fedora live device
19:24 < lotus> brb 5ish minutes I want a smoke.  Thank you syborg for jumping in and asking questions.
19:25 < syborg> lotus, I have not experience with dracut, but from what I can see with a bit of googling the issue may be a bad initramfs, which you can try to rebuild my chrooting into your system from a live media and running dracut -f
19:25 < syborg> np
19:26 < syb0rg> specifically run dracut [target_kernel.img] -f
19:28 < lotus> syb0rg: thanks I will try that right when I get back.  Amazing to have a new step -- I was quite blocked
19:28 < lotus> [poi[
19:28 < lotus> [
19:28 < syb0rg> yeah no guarantee it is the right step though =P
19:28 < lotus> better than nothing
19:28 < lotus> brb
19:29 < syb0rg> look here for some intermediary steps (mounting certain system directories in the chroot) https://www.centos.org/forums/viewtopic.php?t=63988
19:40 < lotus> So I think you are right syb0rg -- I just went to restart to try and use the live disk (looking for rescue option) and it suggested I should regenerate my initramfs
19:40 < syb0rg> nice! Sounds like this might work, then
19:40 < lotus> It is in the process of shutting down, I guess -- I would not be surprised if it hangs here
19:41 < lotus> I will try the fedora live disk and if it has a rescue option I will use that, otherwise I will boot to live and try and mount the parititions and regenerate the initramfs
19:41 < syb0rg> sounds like a plan
19:41 < syb0rg> have you done this before?
19:41 < syb0rg> (the live system chroot and rebuild initramfs)
19:41 < lotus> hah... a start job is running for dev-sda5.device ... 1h27min16s
19:41 < syb0rg> lol
19:42 < lotus> I guess the job was hanging the whole time I was at the prompt...
19:42 < syb0rg> that is a crazy timeout
19:42 < syb0rg> I mean why not make it infinite, or like five minutes?
19:42 < lotus> spamming ctrl-alt-delete reboots it
19:43 < lotus> no recovery option on fedora media
19:43 < lotus> booting to live mode
19:43 < Juesto> use single or something like that
19:43 < syb0rg> yeah, when I asked that question I meant a recovery kernel in grub
19:43 < lotus> ooooh~
19:44 < syb0rg> never seen a live media with a rescue mode, but honestly a live boot is the best possible rescue tool anyway
19:44 < lotus> the link you pasted says: Hit Escape to get to the "Boot:" command, and executed: "linux rescue" command
19:44 < Juesto> there used to be a rescue kernel, yes
19:49 < syb0rg> so lotus you didn't answer my question earlier, have you generated an initramfs from a chroot before?
19:49 < syb0rg> Just wanna know if you generally know the steps or not
19:50 < Dagmar> Why would that require a chroot?
19:51 < syb0rg> because he is doing it from a live boot Dagmar
19:51 < syb0rg> to make his system bootable
19:51 < Dagmar> That doesn't require a chroot
19:51 < syb0rg> hmm, interesting. How do you do it without?
19:52 < Dagmar> I'd just run the commands to create the filesystem
19:52 < Dagmar> No chroot is required for that
19:52 < syb0rg> the guide I followed when I had to do this had me mount --bind various system directories, mount all my partitions, chroot into my system and do it from there
19:54 < syb0rg> Dagmar, does that rely on the live system being the same distro as the target system, or is that irrelevant? Could you use another tool entirely - say, ubuntu's update-initramfs on a fedora system?
19:54 < Dagmar> Probably
19:55 < sadasaulna> Do I cut the red wire?  "Probably."  Or the black wire?  "Probably."  ;)
19:55 < syb0rg> huh, maybe you should tell lotus the steps for this once he gets booted, I'll watch and learn =P (if you have the time and inclination)
19:55 < syb0rg> ^ Dagmar
19:55 < syb0rg> sadasaulna, lol. At least in this case the worst consequence is that the unbootable system remains unbootable
19:55 < sadasaulna> true
19:56 < syb0rg> You're making me think of Archer now: "M as in Mancy"...
19:57 < sadasaulna> ones blue with a white stripe...
19:58 < gunix> can somebody please explain why ip route is needed and how everything you do with ip route couldn't be done with iptables nat and ip route ?
19:59 < syb0rg> gunix, did you just ask why you can't do with ip route what you can do with ip route?
19:59 < gunix> syb0rg: ahaha
19:59 < gunix> yes i did
20:00 < sadasaulna> hah
20:00 < syb0rg> =P
20:00 < gunix> i will rewrite the question
20:00 < gunix> sorry
20:00 < gunix> can somebody please explain why "ip rule" is needed and how everything you do with ip route couldn't be done with iptables nat and ip route ?
20:00 < gunix> :D
20:01 < syb0rg> well to that question: no idea. iptables is dark magic voodoo, and I only really use ip to find ip adresses, set up interfaces, etc.
20:02 < sadasaulna> nah iptables is a cinch
20:02 < syb0rg> you meant *ufw is a cinch ;)
20:02 < sadasaulna> i thought ip rule was for policy routing, different thing than for ip tables and nat
20:02 < sadasaulna> ok, there is a learning curve with iptables but once you've figured it, its genuinely easy to work with
20:03 < sadasaulna> compared with something like Cisco CBAC
20:03 < syb0rg> yeah, but like everything I don't know until I learn it it is dark magic voodoo
20:03 < gunix> sadasaulna: yea, but can't you just to that with iptables and nat? :)) i mean, except checking the TOS field with ip rule (which i never used in my life), why would you use it ?
20:04 < sadasaulna> gunix, i'm not sure what it is you're trying to achieve with either ip rules or iptables tbh!
20:05 < sadasaulna> syb0rg: iptables worth learning, its good.
20:06 < syb0rg> yeah, I know it is a powerful tool... I even have a script that relies on iptables that I'd love to be able to customize a bit
20:06 < syb0rg> so I really should
20:06 < phinxy> Is there some key close to the right hand that could be mapped to tab for tab-completion?
20:06 < syb0rg> maybe insert if you don't use that phinxy?
20:08  * sadasaulna wonders how lotus live recovery is going
20:10 < syb0rg> I think he must have gone for another smoke break lol sadasaulna
20:11 < sadasaulna> thats why i like to vape.  Can sit at pooter doing it :)
20:13 < phinxy> Why whouldnt the menu-key register any event in showkey?
20:13 < phinxy> insert is 110 for example
20:14 < _j> Hi guys
20:14 < no_gravity> I wish when toggling through past history via CTRL+R there would be a shortkey to delete entries from history.
20:14 < _j> I had a quick question - how hard is it to port a device that is on a 2.6 kernel to mainline?
20:15 < _j> Device in question is a kobo, I'm porting it to debian
20:16 < rascul> seems like that would be specific to each case, and there would be no way to answer that question in general
20:17 < djph> ^
20:17 < rascul> i'm not much of a kernel hacker though so i don't truly know
20:17 < _j> hmm
20:18 < _j> I'm just suprised that it's got wifi and runs 2.6 still - what the hell are rakuten thinking?
20:18 < rascul> what's wrong with having wifi on 2.6?
20:19 < _j> rascul: isn't it insecure? I imagine it doesn't have updates for stuff like wpa_supplicant too
20:20 < gunix> sadasaulna: i am trying to learn the ip rule command
20:20 < rascul> _j do you have information to point to wireless insecurities in the 2.6 kernel?
20:21 < _j> rascul: I don't, it was just anecdotal
20:21 < gunix> sadasaulna: i am studying cumulus linux at the moment, and they use VRFs for routing ... just like other network operating systems. in the background, VRFs use ip rule. i am trying to understand ip rule, before i jump into VRFs, because it's better to understand the technology in the background
20:21 < rascul> _j simply because it's older doesn't mean it's insecure (though it might be)
20:22 < rascul> and also, simply because it's newer doesn't mean it's more secure (though it might be)
20:22 < sadasaulna> gunix, google "a quick introduction to linux policy routing" might be what you're after
20:22 < phinxy> Insert was quite nice as a secondary <tab>
20:22 < jim> ziggylazer, hmm?
20:22 < rascul> _j can you get the specific kernel and wpa_supplicant versions?
20:22 < _j> rascul: true, but newer kernels get security updates?
20:23 < jatt> 2.6 is 15 years old
20:23 < rascul> so do older kernels
20:23 < gunix> sadasaulna: yea, i read that, i still don't get it why this was invented in the first place. it says it helps you route based on interface/source/destination ... well thanks, ip route and iptables can do that already :)))
20:23 < rascul> rhel6 is on 2.6 and won't eol until 2020 iirc
20:23 < _j> rascul: fair enough, but I imagine they backport fixes
20:24 < djph> yrah, they're backporting fixes, but 2.6 proper isn't maintained by kernel devs anymore
20:24 < rascul> they probably do, and those patches are almost certainly available somewhere for others to use
20:24 < _j>  
20:25 < _j> rascul: I'll keep that in mind, thanks ;-)
20:25 < rascul> well, i don't know about rhel specifically, but i've pulled red hat patches before to use on non red hat kernels
20:26 < rascul> _j what issues have you run into so far porting to debian?
20:26 < rascul> or have you not started yet?
20:27 < _j> rascul: there's already a project that has done so for the a few kobo's - I'm porting it to the aura h2o
20:28 < _j> So, It should be pretty simple in cherry-picking certain patches, maybe messing around with uboot
20:29 < sadasaulna> gunix, not sure you make decisions on source addresses with just ip route you'd need NAT right, and what if i'm doing end to end routing where I don't want NAT?
20:30 < gunix> sadasaulna: can you give me an example ?
20:32 < sadasaulna> would help if you gave me an example.  If you're just using it for your home network, i'm sure NAT has you covered.  But say I was in a network where network A's default route needs to go one place, and network Bs default needs to go another.  How you doing that with just iptables and ip route?
20:33 < ziggylazer> jim, I need help with vsftpd
20:33 < ziggylazer> If you have time
20:33 < fendur> why not ask the whole channel?
20:33 < rascul> eww ftp :(
20:33 < rascul> ziggylazer what issues are you having? someone can possibly help
20:33 < tds> sadasaulna: you shouldn't even need iptables with marks for that, you can just do ip rule and then have multiple routing tables
20:33 < tds> (with the rules based on source interface or subnet)
20:34 < gunix> sadasaulna: you can prerouting based on interface, wihtout masquarade... so if you have a package coming on port45, you force it to go down another path
20:34 < ziggylazer> Ok so this is the issues: I setup vsftpd on a RHEL server with SElinux disabeld
20:34 < ziggylazer> I cant get the service to start. This is the conf
20:34 < gunix> tds: hey, man! you are just what i need!
20:34  * tds waves
20:34 < tds> what are you trying to do?
20:35 < gunix> tds: i am trying to understand in which cases "ip rule" is better than "iptables" and "ip route"
20:35 < ziggylazer> https://pastebin.com/LtBzNZaX
20:35 < faLUCE> hello. do you know how to download videos from dailymotion website using linux? I tried the online services, but they can't download a file bigger than 4.66 GB
20:35 < gunix> tds: i need to understand it, because it's used in the background by cumulus, the new cool kid in the network world ...
20:35 < ziggylazer> The error code I get is 2. And the logs dosent tell me anything at all. Besides error code=2
20:35 < jatt> faLUCE: wget
20:36 < tds> gunix: doing marks with iptables is a lot more useful if you need to match on certain things that aren't available in ip rule, or if you need to do it based on states with connmarks
20:36 < faLUCE> jatt: wget which url?
20:36 < nixfreak> does anyone have a ddns and a domain ?
20:36 < ziggylazer> When goofling on error code=2 I get nothing usefull at all
20:36 < faLUCE> jatt: how can I fecth the url?
20:36 < sadasaulna> gunix, so how would you specify the routing decisions made for that pre-routed packet?
20:36 < tds> for simple things (eg routing based on source address), a single ip rule is much nicer than iptables marking and then ip rule to switch routing table, though
20:37 < jim> ziggylazer, is it a sustemd unit?
20:37 < ziggylazer> RHEL
20:37 < jim> understood
20:37 < gunix> tds: are rules set by "ip rule" processed faster than the ones set with "ip tables -t nat"?
20:38 < jim> does dmesg say anything about the failure?
20:39 < tds> if you're on fancy equipment where they're handled in hardware I'm sure they are, not sure about on a plain x86 box though
20:39 < gunix> tds: well, most switches have a trident II or tomahawk ASIC, which handle most network operations ... not sure if iptables too
20:39 < ziggylazer> jim, no
20:40 < jim> also, this will probably be better to use as a pastebin: 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"
20:40 < tds> assuming there's some kind of interaction between iptables having to mark the packet and then ip rule having to route based on the mark, I'd expect plain ip rules to be faster, but you'd need to do performance testing to see
20:40 < ziggylazer> Yeah sure.
20:40 < jim> (pastebin does nasty stuff)
20:40 < gunix> sadasaulna: tds: while reading about this, i actually found for the first time that you can do "ip route add nat". i've been using iptables masquarade for years now. i think this is one example where RHCE turned me into an idiot
20:41 < gunix> at least from now on it will be easier to do nat :D
20:41 < tds> gunix: I think that's rather more dumb NAT of the IP packet itself, rather than the state tracking and port translation that iptables does
20:41 < sadasaulna> tds, but you're still using ip rule, gunix seems to be talking about something that doesn't use ip rule at all
20:41 < jim> what are you running (exact command line) to try starting vsftpd?
20:41 < ziggylazer> sudo systemctl start vstfpd.service
20:41 < aaa__> d
20:41 < ziggylazer> or sudo service start vsftpd
20:42 < aaa__> apparently freenode is compromised
20:42 < sadasaulna> still not sure what he's trying to do.  but for source based policy routing i'd be using ip rule i'm sure
20:42 < aaa__> need change password and all
20:42 < jim> ok, so it is systemd
20:42 < ziggylazer> Yeah In RHEL
20:42 < ziggylazer> aaa__, freenode is compromised?
20:42 < jatt> lolwut?, again?
20:42 < gunix> sadasaulna: it uses ip rule. look: http://www.routereflector.com/2016/11/working-with-vrf-on-linux/
20:43 < gunix> sadasaulna: or even better: https://docs.cumulusnetworks.com/display/DOCS/Virtual+Routing+and+Forwarding+-+VRF
20:44 < jim> do you have a file /etc/init.d/vsftpd ?
20:44 < ziggylazer> jim, no
20:45 < ziggylazer> Doesnt start att boot
20:46 < sadasaulna> yeah well with vrf you have overlapping ip spaces so that there is something you need ip rule for I presume, cos you need multiple routing tables.  I've only ever done vrf on Cisco though.
20:47 < gunix> sadasaulna: i never used cisco. i actually never cared about network, till linux became a thing for network. so now i am trying to learn a lot in a small timeframe and i get confused
20:47 < jim> ziggylazer, well there's a little something to go on, but we have to find out exactly what's throwing this error 2, once we know, we can check the man page for that thing
20:49 < ziggylazer> jim, I cant find the one little setting that is causing the service from staring . I do not understand it
20:49 < jim> I wonder if we can start the daemon in the foreground
20:49 < sadasaulna> gunix, well imagine a scenario where you have customer A, and B, and both are using 10.0.0.0/8 - and you want different routing tables for each customer.  Thats the kind of thing I presume ip rule is handy for.
20:49 < jim> what does which vsftpd say?
20:50 < ziggylazer> as it shoud sbin
20:50 < jim> show me exactly
20:50 < gunix> sadasaulna: hmm ... cause you can create the routing table based on interface, so that means you can have CIDR based on interface ... so if another tenant has the same CIDR but uses another interface, you can mix them together
20:51 < sadasaulna> gunix, yep
20:51 < ziggylazer> "/usr/sbin/vsftpd"
20:51 < unixfreak> anyone have a monitor with "nvidia lightboost" features?
20:51 < tds> ip rule + multiple routing tables is also useful in the case where you have multiple routes out to the internet, and want to have traffic from a certain source go via one and another go via another
20:51 < gunix> sadasaulna: tds: i feel like i just discovered fire. you can do ip -br show to get brief so that your terminal/brain doesn't explode from too much info
20:51 < jim> ok... one sec
20:52 < tds> gunix: very neat, TIL :)
20:52 < sadasaulna> gunix, how come you into networking on Linux btw?  Just curious, or using it for a project?
20:52 < gunix> tds: works also for link ... ip -br link. this is briliant
20:53 < gunix> sadasaulna: i am mostly a cloud admit... and i got invited to a "network project" within my company. I was mostly "wtf" at first but than i got it ... the project is using only linux for networking, so they need stuff like ansible, kubernetes, docker, prometheus, ELF stack and so on ...
20:54  * sadasaulna cowers remembering docker and cloud project networking from last job
20:54 < gunix> sadasaulna: this is a new trend in networking, because 10 switches with transcievers and support from cisco cost 200k euros... and if you get THE SAME hardware from the manufacturer, you pay only 30k for it :) and you can run linux on top of that.
20:55 < gunix> sadasaulna: so ... cost reduction almost 90% ... and you have all the linux tools to manage it. seems a lot of companies are doing this... including microsoft, which is amazing.
20:55 < gunix> sadasaulna: sounds like a thing that is worth investigating, no? :D
20:56 < ziggylazer> gunix, what?
20:56 < sadasaulna> gunix, people been doing networking with linux since forever but I think that cloud stuff is accelerating it
20:56 < sadasaulna> stuff like vxlan and the likes probably
20:56 < ziggylazer> Cisco charges for support hand HW
20:56 < gunix> sadasaulna: yea, check project sonic from microsoft. or cumulus linux ... or openswitch.. or FRR project
20:56 < ziggylazer> I hate cisco but thats what they do
20:57 < ziggylazer> So what should be investigated  ?
20:57 < ziggylazer> The number of backdoors they have?
20:57 < gunix> ziggylazer: network on linux should be investigated
20:57 < gunix> since... money.
20:57 < ziggylazer> Well fuck, MS have that in every OS ever released
20:57 < ziggylazer> And so does apple
20:57 < ziggylazer> Linux, however does not
20:58 < gunix> sadasaulna: think about this awesome idea: do you know kubernetes? and calico on kubernetes?
20:58 < gunix> ziggylazer: yes. that's why networking on linux is awesome
20:58 < ziggylazer> So go peddle that shit in another channel
20:58 < ziggylazer> Please
20:58 < sadasaulna> i was just starting to play with kubernetes before I went insane and left my job, so its all a distant memory
20:58 < rascul> what doesn't linux have?
20:59 < ziggylazer> The same "mandatory" backdoor that cisco has
20:59 < SuperSeriousCat> Hookers and blackjack ^
20:59 < rascul> oh
20:59 < gunix> sadasaulna: well, calico is used for network overlays. and it uses BGP to share routes ...
20:59 < ziggylazer> BGP is a freaking routing protocall
20:59 < gunix> sadasaulna: you can run calico on the network operating systems and spread the routes from the containers to the fabric. :D because of linux :D
21:00 < ziggylazer> gunix, please share the specs of another routing protocall that is used
21:00 < ziggylazer> And not OSPF and ERGP
21:00 < ziggylazer> Since they are internal
21:00 < gunix> ziggylazer: EVPN
21:01 < ziggylazer> And how widley spread is that protocal today?
21:01 < gunix> ziggylazer: it's just emerging lol. the ASICs doing VXLAN got out only a few years ago ...
21:02 < ziggylazer> Yeah..
21:03 < gunix> ziggylazer: i share your hate for cisco. i also hate microsoft. and vmware. and tbh every company doing propriatary :D
21:03 < gunix> so ... having open source stuff in the network is awesome.
21:03 < NewbProgrammer10> That's already been accomplished for many, *many* years.
21:04 < ziggylazer> You pay cisco for what you get
21:04 < ziggylazer> And what you get is some awesome shit, with a few backdoors insode
21:05 < ayecee> nothing's perfect, amirite
21:06  * SuperSeriousCat is perfect
21:06 < iodev> right, ayecee
21:15 < jim> ziggylazer, hmm, I don't see anything in either man vsftpd or man vsftpd.conf that tells how to put the thing into the foreground
21:16 < jim> also, it's going to read that config file no matter what
21:17 < jim> unless we ove it
21:17 < jim> move
21:21 < ziggylazer> Yeah but I do want it to read that conf file
21:22 < ziggylazer> I'll figure it out man.
21:22 < ziggylazer> Dont spend time on it
21:25 < sadasaulna> jim, isn't there a 'background=NO'
21:26 < jim> ziggylazer, seems like the best thing to do, is watch its log while you start it
21:27 < ziggylazer> yeah I will. Thanks man
21:27 < jim> maybe stuff is already in the log
21:27 < sadasaulna> you mean you hadn't check the log?
21:27 < jim> I dunno if he did or not
21:28 < sadasaulna> basics first :)
21:29 < Juesto> what was the utility used to discover and mount partitions off a raw image?
21:30 < jim> how would it know where to mount them?
21:30 < Juesto> loop/dm stuff
21:31 < jim> hmm, not much to go on
21:31 < Juesto> jim it mounts them on loop via dm
21:31 < Juesto> stufd on kernel
21:32 < prussian> losetup should be able to discover partition tables
21:32 < prussian> and setup a loopback for you
21:32 < Juesto> there's an alternative to losetup
21:32 < prussian> it might be -P or some opt I don't remember
21:32 < Juesto> the other day it got suggested
21:32 < Juesto> mentioned instead of losetup
21:33 < Juesto> don't remember, but k something
21:33 < day> is there a pasteservice like pastebin for binary files?
21:33 < Juesto> day https://ptpb.pw
21:33 < prussian> kpartx? what?
21:33 < Juesto> hmm
21:34 < Juesto> close i think
21:34 < Juesto> i forgot how to convert a grub install to efi
21:35 < day> Juesto: that excepts binary files?
21:35 < Juesto> day see the site
21:35 < Juesto> it's browsable and it goes via curl
21:35 < day> im reading it
21:37 < faLUCE> I'm trying to get a dailymotion video with youtube-dl. Unfortunately, it allows to get maximum size of 4.66 GB, but the video is longer. Why? How can I solve that?
21:37 < faLUCE> I'm trying to get a dailymotion video with youtube-dl. Unfortunately, it allows to get maximum size of 4.66 GB, but the video is longer. Why? How can I solve that?
21:41 < mouses> using cloudflare dns myself and resolving archive.is with no issue
21:41 < mouses> oh lol
21:41 < mouses> I just responded to something days old, ignore me
21:42 < mouses> faLUCE: I think you can pass --max-filesize to the command and override that
21:46 < faLUCE> mouses: It calculates 4.66 GB as the filesize. This is the video url:  https://www.dailymotion.com/video/x17wq88
21:47 < mouses> okay, so when you run youtube-dl - add the --max-filesize 6gb flag to the command
21:47 < mouses> I think that should work
21:49 < faLUCE> mouses: same result...
21:50 < mouses> faLUCE: Ack.  Give me a second, let me futz with it
21:50 < faLUCE> thanks very much mouses
21:51 < mouses> faLUCE: oh!  the --max-filesize flag means don't download anything 'larger' than what you set
21:51 < mouses> that's weird
21:52 < mouses> faLUCE: now i'm just not sure lol
21:52 < faLUCE> mouses: in fact it's weird
21:53 < mouses> faLUCE: that's weird, on my server I just did a: youtube-dl https://www.dailymotion.com/video/x17wq88
21:53 < mouses> it seems to be working fine
21:53 < mouses> faLUCE: does it abort later in the download process for you?
21:53 < hexnewbie> faLUCE: Maybe update your youtube-dl code?
21:53 < faLUCE> mouses: and you are sure that the duration is the same of dailymotion? more than 2 hours
21:54 < mouses> faLUCE: it's showing 4.66GB total
21:54 < mouses> so I assmue that's > 2 hours
21:54 < faLUCE> mouses: I know, but it doesn't correspond to > 2 hours. It's about 1h 49m
21:54 < mouses> weird
21:54 < kekePower> latest version is 2018-05-18 afaic
21:55 < mouses> maybe it's something to do with the server side (dailymotion?)
21:55 < kekePower> latest version is actually 2018-05-26
21:55 < faLUCE> mouses: maybe the file is fragmented on the server side
21:55 < mouses> yeah
21:55 < faLUCE> mouses: I wonder if is there a way to specify a start time
21:55 < Dagmar> Perhaps you could just ask the original publisher to send you a copy
21:56 < mouses> faLUCE: don't think so, not seeing that in the man page
21:57 < faLUCE> I see :-(
21:57 < faLUCE> I wonder if is it possible to see the video with vlc or mplayer
21:57 < faLUCE> (so I can record it while viewing)
21:57 < Psi-Jack> mpv
21:58 < faLUCE> Psi-Jack: how?
21:58 < hexnewbie> Well, doesn't mpv use youtube-dl? If youtube-dl's code can't get the whole thing, I doubt mpv will be able to record it
22:00 < Psi-Jack> mpv does use youtube-dl heh
22:00 < hexnewbie> faLUCE: Did you check the video actually stops before the end?
22:01 < hexnewbie> faLUCE: Meaning, that it's not a corruption making the player report the wrong time?
22:03 < Psi-Jack> Hmmm... Maybe you need to specify the specific format you actually want.
22:03 < Psi-Jack> youtube-dl --list-formats URL
22:04 < Psi-Jack> And with mpv, you could pass that -f format with --ytdl-format=<code>
22:06 < Psi-Jack> I for example, could not do the http-1080, but the http-720 is working fine.
22:08 < gunix> you can also do -F to get a list of all formats
22:08 < faLUCE> Psi-Jack: I don't think that could solve the issue
22:08 < faLUCE> of 4.66 GB
22:09 < gunix> faLUCE: format can get you a smaller video
22:09 < Psi-Jack> ^
22:09 < gunix> faLUCE: you can pick the resolution... or only audio if you want only audio
22:09 < gunix> i usually do youtube-dl -f 251 cause that gets me only audio at the best quality
22:09 < faLUCE> gunix: I can't get audio or video separately
22:09 < gunix> faLUCE: you actually can with youtube-dl
22:10 < faLUCE> gunix: I tried to list all the formats, but it doesn't show audio and video separately for that link
22:10 < gunix> let me try
22:10 < Psi-Jack> Yeah, it doesn't show audio/video only format-codes.
22:11 < Psi-Jack> Only combined.
22:11 < eletious> Am I missing some kind of weird documentation that will just tell me what tools exist for software dev?  There are so many things that I wouldn't think I need in C dev that I need, and not a great way to sort it out...
22:11 < gunix> yea, seems it is limitted for dailymotion in that regard ...
22:12 < sadasaulna> eletious: bit of an open question, no?
22:12 < eletious> yeah, it is...
22:12 < gunix> eletious: there are tons of tools for software devs ...
22:12 < faLUCE> in addition, I see:   hls-1080-0,1,2,3   mp4        1920x1080
22:12 < faLUCE> why 1,2,3,4 ?
22:12 < Psi-Jack> hehe
22:12 < Psi-Jack> faLUCE: I dunno, but none of the hls-* format codes worked for me.
22:12 < Psi-Jack> Would get 403's with those.
22:13 < faLUCE> :-(
22:13 < eletious> It just seems that other software devs just *know* what's going on and what tools they like and should use to get work done
22:13 < Psi-Jack> http-* codes worked though, hls-* codes do not.
22:13 < gunix> Psi-Jack: you just downloaded all formats, didn't you ?
22:14 < Psi-Jack> Nope.
22:14 < gunix> :D
22:14 < Psi-Jack> I tested http-1080 and http-720 with mpv, that's all.
22:14 < Psi-Jack> With seeking. :)
22:14 < sadasaulna> eletious: C is open ended though isn't it, you can do more or less anything with it.
22:15 < sadasaulna> eletious: what is your use case?
22:15 < eletious> I'm just learning C - I've been putting together some smaller programs, the one I'm working on now makes a little REST endpoint that spits out Hello World
22:16 < eletious> There are features I'm trying to implement though - this isn't just a C exercise, I'm playing with CI frameworks so I'm trying to add tests and eventually deployment steps to my CI pipeline
22:17 < gunix> electricbear: i've been running linux for 11 years and worked a lot as dev and sysadmin. i still have no idea what tools i like most, and I have a tendancy to work only with vim, since it's really portable. i am no sure you will get an answer to your question soon ... just enjoy trying stuff out, and use what you like most.
22:18 < faLUCE> so, there's no solution for that
22:18 < gunix> there is no "best tool". most people use what they like most ... and with time they also start contributing to it :)
22:20 < sadasaulna> sounds like question is more about CI frameworks than C itself.  Wouldn't have a clue about CI, i'm not a developer, I only tinker with C for shits and giggles.
22:21 < sadasaulna> eletious: probably better dev focused channels on irc though
22:22 < eletious> most likely.  I'll just have to keep playing around and figure out what I like to use best.  Thanks guys :)
22:22 < sadasaulna> no worries, keep playing with C and have fun
22:23 < Li> how to send a notification email from terminal without having to setup any email account? something like those spams we daily receive inBOXes. I want to use this to confirm the operational status of rPi every hour.
22:23 < rpifan> hello
22:26 < elichai2> hey
22:26 < elichai2> I'm writing a script that can rotate 180* the screen+touchpad+touchscreen using xrandr+xinput, how can I check the current rotation status?
22:29 < rascul> you can parse the output of xrandr, but that isn't always reliable
22:29 < elichai2> but I don't find anywhere inside the output of xrandr an indication of current rotation status
22:30 < rascul> eDP1 connected 1366x768+0+0 inverted (normal left inverted right x axis y axis) 260mm x 140mm
22:30 < rascul> mine says inverted
22:30 < BSODjunkie> Quick question from new user, I know you can boot some linux distros from a USB relatively easily, so theoretically would I be able to take that usb and then use it on my other PC and it would work on the new hardware with all my files/preferences stored on USB?
22:30 < elichai2> rascul: yeah I see now that it says that when you add verbose
22:31 < rascul> i don't need to add verbose here
22:31 < gunix> sadasaulna: regarding our discussion earlier, found this: https://www.kernel.org/doc/Documentation/networking/vrf.txt
22:31 < rascul> see how it isn't always reliable? ;)
22:31 < gunix> it seems to have a lot of cool info
22:31 < rascul> probably the best way is to interface with libxrandr but that might be more than you're interested in doing
22:34 < sadasaulna> gunix: ah brings back memories, I realise I have done some vrf stuff with linux, and it was using that guide I think
22:35 < sadasaulna> i was playing around with it in GNS3 ISTR
22:37 < sadasaulna> GNS3 is pretty cool for playing with this stuff
22:57 < solidfox> hello if someone is spamming me strange links and saying they're 10 minutes away from me, who should I report that too?
22:57 < solidfox> to*
22:58 < blackpawn> the FBI?
22:58 < solidfox> nvm. I'll tell freenode I think
22:58 < hexnewbie> FreeNode staff, channel ops (if common), police, abuse emails on the IP whois
22:58 < solidfox> blackpawn, its not that serious I think
22:58 < ayjay_t> okay so i went through a lot of pain to install postfix, and realized that i really don't need a mail server, just a command line, programatic mail client, preferably one that supports oauth2 (gmail, ugh)... any recommendations?
22:58 < day> solidfox: getting the same shit
22:58 < solidfox> day, yeah. from lawgangfo92 ?
22:59 < jim> solidfox, start with freenode staff
22:59 < blackpawn> what distro everyone use?
22:59 < day> im pretty sure they already know about it though. dunno he was changing his nick constantly. i saw the message a few hours late and the message tab had like 50+ name changes in it
22:59 < blackpawn> i just setup linux mint with cinnamon
22:59 < jim> first time installing linux?
23:00 < blackpawn> nah just curious what other ppl using now
23:00 < blackpawn> i setup a nwe machine from mining burst coin
23:00 < jim> I run debian
23:00 < blackpawn> solid
23:01 < jim> yep, has been for a long time
23:01 < triceratux> blackpawn: congrats. lubuntu lxqt, mx-17, extix, siduction, swagarch, xubuntu-core. that kind of thing
23:01 < bitslicer> slackware/suse depending on my mood
23:01 < Anticom> Hi. I've got a sieve script that's forwarding mails by `redirect :copy "<email>". I'd like to move the mails instead of copying. How is this done?
23:01 < blackpawn> i saw manjaro or something is the tops now
23:01 < solidfox> day, ah
23:01 < solidfox> blackpawn, I use freebsd
23:01 < solidfox> oh wait this is ##linux
23:01 < solidfox> I use ubuntu 7.04
23:02 < blackpawn> i coded my first stuff for linux recently, was really fun! i'm hooked. used GTK+
23:02 < FreakingOut1987> why is this printing 'no-fail'?
23:02 < FreakingOut1987> if [[ $(ls /root/) == 0 ]]; then echo "fail!" && exit 1; else echo "no-fail"; fi
23:02 < FreakingOut1987> I checked ls /root/ the exit code is 2
23:04 < jim> well 0 exit status, isn't that success?
23:04 < day> does $(ls /root/) actually return the error code?
23:04 < day> isnt the error code stored in $?
23:05 < blackpawn> is cool Cinnamon can show progress in the task bar. what other ones do that? i think Ubuntu can
23:06 < FreakingOut1987> gotcha thanks guys!
23:07 < jim> FreakingOut1987, plus, you probably want to see whether $(ls) gives you the exit status... maybe just ls does?
23:08 < FreakingOut1987> jim, nope my mistake was using ==. I should have been using -eq
23:08 < ziggylazer> hackthebox.eu invite done
23:08 < ziggylazer> jim,  can you pass that?
23:08 < day> FreakingOut1987: odd. echo $(ls /root/) doesnt return an error code. why does it work in your scenario at all
23:09 < FreakingOut1987> day, ~$ $(ls /root/); echo $?;
23:09 < FreakingOut1987> ls: cannot open directory '/root/': Permission denied
23:09 < FreakingOut1987> 2
23:09 < FreakingOut1987> i'm not root
23:09 < jim> FreakingOut1987, ok
23:10 < day> so bash translates the '2' into a clear text error? interesting
23:10 < day> to get the '2' you can also do 'ls /root/; echo $?
23:11 < rascul> no, that's ls printing an error, 'echo $?' will print the exit code of the last command
23:11 < jim> day, well ls actually prints the error message
23:11 < day> jim: but wouldnt it do so in the if [[ $(ls /root/) == 0 ]]; scenario as well?
23:12 < day> and 'some string error' will never be equal to 0
23:12 < FreakingOut1987> day, thats wrong i changed syntax
23:12 < rascul> don't need the $() there
23:12 < jim> not sure... I'll try it
23:12 < day> FreakingOut1987: yeah i know. but from my understanding the -eq shouldt work either
23:12 < FreakingOut1987> I may be messing up here somewhere. Let me turn on trace
23:12 < rascul> probably don't need the if, either
23:12 < rascul> ls /root && echo "success" || echo "failure"
23:13 < phantomcircuit> im trying to install debian on an x1 carbon with i915 graphics
23:13 < FreakingOut1987> rascul, you're correct from my experiment. But why wouldn't failure echo?
23:14 < rascul> failure wouldn't echo if it didn't fail
23:14 < FreakingOut1987> ah... I get it now
23:14 < jim> phantomcircuit, ok...
23:14 < rascul> it's either going to echo success if it succeeded (exit code is 0) or failure if it failed (exit code not 0)
23:14 < FreakingOut1987> wow I don't know what I was thinking lol
23:17 < jim> phantomcircuit, is it that you're stuck somewhere? or something's not working right?
23:17 < phantomcircuit> sorry got distracted
23:17 < phantomcircuit> grub comes up but when i select an option and the actual kernel boots the screen is just black
23:18 < phantomcircuit> i tried hitting the hardware screen brightness adjustments but it's not working
23:18 < phantomcircuit> the thing is it boots into arch and ubuntu fine
23:19 < jim> do you see where it starts loading the kernel (or anything else)?
23:21 < phantomcircuit> jim, no the screen is just black
23:21 < phantomcircuit> the backlight is on
23:21 < jim> ok
23:21 < jim> this is a laptop?
23:22 < phantomcircuit> jim, yeah
23:23 < jim> one thing I could suggest, is see if #debian has any extra info they can give you when you ask the question
23:23 < phantomcircuit> asked a while ago
23:24 < phantomcircuit> channels seems to be pretty dead
23:24 < Psi-Jack> It's definitely not dead.
23:25 < Psi-Jack> I don't even get why you could say that, because since you asked that, there's been constant communications, just none towards you.
23:28 < phantomcircuit> Psi-Jack, cause all the discussion is about trivial stuff...
23:28 < phantomcircuit> anyways
23:30 < day> irc doesnt work for complex questions. takes too long. you need to be able to break it down into simple questions that one can expect someone to know the answer to
23:31 < Li> fucking nice channel
23:31 < day> no you wouldnt D:
23:37 < badsekter> gg
--- Log closed Sun May 27 00:00:31 2018