--- Log opened Fri Apr 06 00:00:20 2018
--- Day changed Fri Apr 06 2018
00:00 < jim> volty, if both mounts succeed, one will "mask" the other
00:01 < volty> thx jim.
00:01 < jim> volty, welcome... and please expand things like thx, y, u... please spell out u as you, it helps people (particularly new english speakers) to understand, at least, most of what's going on
00:04 < volty> another question I asked yesterday on grub chan but didn' t get an answer.  I did clone my partition on a partition of a new ssd. later I had to fix UUID (was the same). So I have two partition with the same OS. But when grub-update runs, the menu for the update-grub ran, it found both, but in the second (ssd), puts root=UUID_of_old_one. So, to boot into it I edited manually whiel booting (e-option). How can I fix this
00:04 < volty> automatically without hunting and editing manually the right grub.cfg ?
00:07 < volty> my case:  ubuntu14.04 on /dev/sda6, then 14.04 on /dev/sdc1/, but the grub menu for the second one has, as param (kernel) ROOT=UUID of /dev/sda6, while the other lines, of the same menu entry, have the correct UUID. How that? Why that?
00:07 < msiism> i just discovered something arguably weird about processes on my system: when i run a command in a x terminal window, the process started there will create another x terminal process (which i cannot see) as a child process to itself, then launch bash as a child process to itself as well and then run the script as a child process of that bash instance. see: http://paste.debian.net/plainh/2bc36359. why the additonal x term emulator process?
00:08 < msiism> i'm running devuan 1.0, my x term is urxvt.
00:09 < mutante> so does anyone use strict SPF checking with Google as mail servers?  as in "-all" at the end and not "~all"?  It seems like Google really wants to keep the ~ and even if you use ""v=spf1 mx include:_spf.google.com -all"" that would expand to something with ~ at the end
00:10 < spammcoin> yourxvt is doing weird stuff
00:11 < msiism> spammcoin: is that directed at me?
00:13 < msiism> ok, i'll check again, using xterm this time.
00:13 < spammcoin> myxterm just creates xterm->bash->command
00:14 < msiism> spammcoin: ok, same here. so, it seems to be an urxvt issue.
00:14 < bls> does your terminal support mutiple tabs?
00:15 < msiism> bls: no.
00:16 < bls> or could be urxvt spawning off another process for its smoothing/double buffering stuff
00:16 < msiism> bls: well, i'm not starting urxvt directly but via invoking x-terminal-emulator. may that has sth to do with it. i'll check that.
00:16 < bls> or x-terminal-emulator could just be a script that doesn't call exec at the end
00:17 < phogg> msiism: sth? Statement Handle?
00:17 < phogg> x-terminal-emulator is a symlink to your chosen terminal emulator binary. On debian urxvt is not a script.
00:18 < msiism> bls: running urxvt directly results in urxvt -> urxvt -> bash -> command
00:19 < msiism> phogg: i know about the symlink. but i don't know what a statement handle is.
00:20 < msiism> maybe it's the way i'm starting the whole thing from my window manager's menu
00:22 < msiism> ok, checked that. it doesn't make any difference (running urxvt from wm menu or executing it from the run dialog).
00:22 < msiism> it's creating the additional instance either way.
00:25 < msiism> and i was wrong about running urxvt directly before. it does not make any difference compared to running x-terminal-emulator.
00:25 < darkhanb> command >>log_file 2>&1
00:25 < darkhanb> does the stderr append to the log file? or overwrite it?
00:25 < msiism> darkhanb: let me try that
00:26 < bls> darkhanb: >> is append
00:26 < darkhanb> bls msiism ok thanks
00:27 < msiism> bls: but it will create the file if it isn't there (on my system)
00:27 < msiism> bls: but ok, we are also looking for overwriting. got it.
00:28 < phogg> append is still O_CREAT
00:28 < jim> here's another way: command | tee logfile
00:28 < jim> or if you want to include any error messages generated by command,
00:29 < jim> command 2>&1 | tee logfile
00:29 < msiism> darkhanb: which command could i use. "top" (which i was using) gives me a lot of character salat in the log file.
00:30 < msiism> maybe "who" is better.
00:30 < phogg> msiism: use for what?
00:30 < jml2> msiism, hoo?
00:30 < bls> who and top do completely different things
00:31 < phogg> "whom" is better (-;
00:31  * phogg waits for someone to complain that he hasn't got any whom(1)
00:31 < phogg> that person: IT WAS A JOKE
00:31 < jim> you just did :)
00:31 < treefrob> I haven't got the whom command. Is my system disk corrupt?
00:32 < phogg> treefrob: No, just ungrammatical
00:32 < jml2> treefrob, the joke was corrupt
00:32 < geheimnisse> alias whomst to who
00:32 < phogg> corrupt joke error, please retransmit
00:32 < msiism> bls: ok, i think there was a misunderstanding here at some point. i'm only trying to find out why i get an additional instance of urxvt.
00:32 < supernovah> What does inotify do with spaces in paths/filenames?
00:32 < msiism> that does not depend on an actual command being executed in the terminal window
00:32 < supernovah> Does it escape them with a literal backslash?
00:33 < treefrob> djph
00:33 < phogg> supernovah: nothing
00:33 < treefrob> oops, sorry
00:33 < bls> msiism: then you probably want to just read the source code or run strace on it
00:34 < bn_work> hi, I have written a SysV-LSB compliant init script.  I wish the service to start under a system service account but when I try to do `su -c 'nohup myservice param1 param2 > mylog.nohup' `, I get `su: must be run from a terminal`.  Why?  and how do I get the service to run under the service account?
00:34 < phogg> supernovah: shells need to worry about escaping file names, not programs using C APIs
00:34 < mattfly> Hello i need a software;method/script/whatever suggestion to use a laptop with linux as a virtual secondary monitor to my computer(linux too)
00:34 < mattfly> like i have synergy, i could use synergy and a vnc on the monitor laptop but im looking for something that really behaves like a secondary monitor
00:34 < phogg> mattfly: mirroring your existing monitor or displaying some other content?
00:34 < mattfly> is there a program for this some non Xorg messy way
00:35 < mattfly> no not mirrowing, using for other content
00:35 < Psilocyber> hey fam, my system is hard locking and something is puking a huge string of "^@^@^@^@^@^@^@^@" characters to the syslog when it locks up. Does the ^@ indicate anything in particular?
00:35 < mattfly> like when you plug another monitor and can send other windows to there and move and such
00:35 < phogg> mattfly: Start a 2nd X server and connect to it via either xdmcp or VNC depending on your preference
00:35 < mattfly> a secondary monitor for editing videos with kdenlive
00:35 < Psilocyber> I'm running Ubuntu Server 16.04
00:35 < msiism> bls: i'm not qualified to understand urxvt's source code. it's probably written in C, which i have just begun to learn a few days ago.
00:36 < mattfly> wait phogg, start a 2nd X server on my current machine and connect to it from the oother using vnc and making my mouse work there with synergy
00:36 < mattfly> is there any tutorial for this?
00:36 < koala_man> Psilocyber: those are \0 bytes. Not sure what it's supposed to mean
00:37 < mattfly> and will my system recognize this 2nd Xorg it as a second monitor? Will i be able to move windows there?
00:37 < supernovah> phogg: What does that comment about shell vs C API's mean? Why did you mention that?
00:37 < Psilocyber> koala_man: thanks, was hoping I would get lucky and it would be the golden clue to the hard lock XD
00:38 < Psilocyber> koala_man: when you say \0 bytes, what do you mean exactly?
00:38 < koala_man> Psilocyber: ascii NUL bytes
00:38 < phogg> msiism: it looks like rxvt uses as two process model
00:38 < Psilocyber> ahhhh thanks
00:40 < msiism> phogg: ok. so, it's probably not a bug then.
00:41 < kurahaupo> mattfly: no you can't drag windows between separate X servers.
00:41 < mattfly> I would appreciate a tutorial for this
00:41 < msiism> phogg: i'm gonna ask about this in #rxvt-unicode.
00:41 < mattfly> well thats what i wanted to work
00:41 < mattfly> is it really hard to make a program like that?
00:41 < phogg> msiism: that's a good idea, I can't be too certain after 3 minutes with the source code
00:42 < kurahaupo> mattfly: so you want a single X server spanning all the physical display devices
00:42 < bls> xorg has no mechanism for doing this the way say a mac does
00:42 < phogg> mattfly: you don't need synergy for this
00:42 < ironpillow> hi all, when i am trying to run a binary (os: ubuntu) I am getting this error: Invalid uid or gid. Both must be greater than 0. -1, -1. Any advice? thanks!
00:42 < msiism> phogg: but thanks for having a look.
00:42 < kurahaupo> bls: I routinely have two monitors in one X session
00:42 < ironpillow> I am doing sudo ./thebinary start
00:43 < kurahaupo> mattfly: man xrandr
00:43 < bls> kurahaupo: this isn't two monitors though, he wants to use a laptop display as a second monitor for a different machine
00:43 < phogg> mattfly: there's no way to drive the GPU on a remote system from the X server on your local host. That's what you'd need to drag a window from the local X to a monitor plugged in to a GPU attached to another box. That or some serious, fanciful tunneling.
00:44 < mattfly> damn...
00:44 < bls> i.e. in the way two macs can share a GPU over TB
00:44 < kurahaupo> Ah. Well you don't *need* the GPU, it's just faster with it
00:44 < phogg> mattfly: there are however a variety of not-quite-that things you can do instead
00:44 < phogg> kurahaupo: I don't know of any way to add a monitor device to X that is not locally attached
00:45 < kurahaupo> Xvnc?
00:45 < kurahaupo> phogg: proxy
00:45 < phogg> kurahaupo: that would still be a second screen, no?
00:45 < mattfly> why cant a  program make a virtual secondary monitor and virtualize that and stream that to another computr?
00:45 < mattfly> a vritual streaming monitor
00:46 < phogg> kurahaupo: yeah, some kind tunneling like I was talking about. But is there any such thing? Technically you only need a socket and some hand waving but I don't recall anything like that.
00:46 < mattfly> and i cant set it up as if it was a real monitor
00:46 < bls> it could, just no one's seen the point to undertake the effort to do it
00:46 < phogg> mattfly: It's not technically infeasible but there is no such program you can just get and run. If you want to try writing it, though...
00:46 < kurahaupo> mattfly: it can have an entire X server; federating multiple displays is tricky
00:46 < rypervenche> ironpillow: The binary doesn't want you to run it as the root user. Don't use sudo.
00:47 < bls> it'd be way less headache to just buy a second monitor
00:47 < mattfly> I would try writting it if it turns out it really doesnt exist
00:47 < mattfly> i already have 2 laptops
00:47 < mattfly> an extra computer is more than a monitor
00:47  * phogg contemplates how tricky it would be to write a kernel module to present a remote display as a local device.
00:47 < bn_work> anyone?
00:47 < cheater> hi
00:47 < mattfly> should do its work without any effort
00:47 < kurahaupo> mattfly: you can build nearly that out of parts (at least 3)
00:47 < phogg> that's about ten leagues above my head
00:48 < cheater> if i have an amazon image id like ami-0123456 how can i find out what it is exactly?
00:48 < mattfly> hm but maybe someone already has almost done this
00:48 < phogg> cheater: file. Also, I believe it's a proprietary format.
00:48 < ironpillow> rypervenche: just tried to cat it, it's a bash script. when I try to run without sudo: The script needs to be run using sudo
00:48 < ironpillow> rypervenche: sorry for saying it's a binary
00:48 < jml2> mattfly, https://superuser.com/questions/115076/how-to-vertically-split-widescreen-into-two-virtual-workspaces-on-ubuntu-gnome
00:49 < rypervenche> ironpillow: If it's telling you that it has to be run with a uid/gid greater than 0 then it is saying that you need to not be the root user.
00:49 < bn_work> no one has experience writting init scripts?
00:49 < jml2> mattfly, and then maybe using vnc to specify 1 of the displays
00:49 < cheater> phogg: i'm talking about an id of an image that exists on the amazon infrastructure... i just have the id, i don't have the file on disk.
00:49 < mattfly> oh cool
00:49 < rypervenche> ironpillow: Maybe a poorly-written script. I don't know.
00:49 < kurahaupo> bn_work: sure, 10 years ago
00:49 < jml2> mattfly, maybe  FakeXRandR
00:50 < ironpillow> rypervenche: i see. i didn't write this script. its from amazon :(
00:50 < mattfly> thanks i think i can do somethingk with these
00:50 < rypervenche> ironpillow: I'd have to see what it is and the output that you're getting in order to be able to help further.
00:51 < ironpillow> rypervenche: sure, let me create a gist link
00:51 < phogg> jml2: it sounds nice but could you get the mouse from the host box to appear on the other system's screen?
00:51 < phogg> This FakeXRandR thing looks really fun, I will say.
00:51 < mattfly> i could use synergy with for that?
00:51 < jml2> no clue -- i assume it should be possible to use vnc against the virtual monitor, and have synergy working on both hosts
00:52 < phogg> bn_work: some, why?
00:52 < bls> that's still not going to allow you to drag windows across
00:52 < jml2> synergy of course needs to be started manually prior running vnc from station 2
00:52 < bn_work> phogg/kurahaupo: how would I run a service under a specific service account I created?  I tried `su -c 'mycmd ...'` in the start section of the init script but it keeps saying `su: must be run from a terminal`
00:52 < phogg> bn_work: typically the daemon should change user itself
00:53 < bn_work> phogg: in the init script itself?
00:53 < bls> and if su is giving that message, it likely means you're trying to run it as a normal user
00:53 < jml2> or maybe synergy is not even needed
00:53 < jml2> mattfly, ^^
00:53 < phogg> bn_work: no, in the binary it runs
00:53 < mattfly> okay
00:54 < bn_work> bls: It's in the init script
00:54 < bn_work> but to test it I'm obviously invoking it from the terminal
00:54 < mattfly> thanks for the tips
00:54 < ironpillow> rypervenche: here is the file: https://gist.github.com/ironpillow/6b455190c7821c240b5124636c8dfa82
00:55 < phogg> bn_work: if you don't want to write the daemony stuff yourself you can wrap an existing program with http://www.libslack.org/daemon/
00:56 < bn_work> phogg: hmm, so the app I'm trying to daemonize to run on boot is a 3rd party app I've been given
00:57 < bn_work> phogg:  with a non-functional init script
00:57 < phogg> bn_work: complain to its author if it does not provide a way to run as another user, meanwhile use the daemon command linked above to this for you
00:58 < phogg> I've replaced the stock init scripts from a lot of programs because their authors were crazy or heavily drugged when the scripts were written.
00:58 < bn_work> phogg:  so their use of `su` in the init script is wrong?
00:58 < bn_work> lmao
00:58 < phogg> bn_work: Not exactly wrong, but not normal.
01:01 < bn_work> the weird thing is, when I tested a similar script on CentOS7, I believe it didn't complain about `su` being used there?
01:03 < bn_work> admittedly CentOS7 though, so it's really using systemd under the hood, so not sure if that's altering the perceived success?
01:03 < bls> it's odd that su is giving that message. it usually implies that it wants to prompt for a password
01:03 < bn_work> on Ubuntu it seems it doesn't like that
01:04 < bls> which would only be required if su wasn't really su, or a non-root user was executing the init script
01:04 < bn_work> bls:  should I just run the script with sudo to simulate boot conditions running as root?
01:04 < ayecee> sounds like a plan
01:05 < bls> does ubuntu do something nutty like put a wrapper around su to actually call sudo?
01:05 < ayecee> afaik no
01:05 < ayecee> i missed the intro. what was the error?
01:06 < phogg> bn_work: su always requires a tty
01:06 < bn_work> ayecee: `su: must be run from a terminal`
01:06 < ayecee> how were you running the script?
01:07 < phogg> I suppose it may have been less strict in past versions.
01:07 < ironpillow> rypervenche: sorry to bother. did you find why the script is throwing an error. thanks :)
01:07 < bn_work> I'm using `service start appname`
01:07 < ayecee> in an interactive session?
01:07 < bn_work> yes
01:07 < ayecee> huh.
01:07 < rypervenche> ironpillow: Can you show us your exact output please?
01:07 < bn_work> of course the init script will be run on boot too though
01:07 < phogg> Hmm, I'm wrong. When run as root su will not complain about not having a terminal.
01:09 < jim> ironpillow, this could be handy for sharing results and other command output... you can pastebin the output of an arbitrary command by running "anArbitraryCommand | nc termbin.com 9999", and to include error messages, "anArbitraryCommand 2>&1 | nc termbin.com 9999"
01:09 < bls> yeah, used to depend on that all the time on RHEL4/5/6. would be curious where this su binary is from
01:09 < phogg> su as found in Debian checks isatty(0) and in the else case does: if(!caller_is_root){ /* shows error message and exits*/ }
01:09 < bn_work> bls:  I'm testing on Ubuntu 14.04
01:10 < bls> so you're getting this error when running the `service start` command as a normal user?
01:10 < bn_work> yeah
01:10 < phogg> bls: it appears that Debian may have an entirely different su from red hat http://pkg-shadow.alioth.debian.org/
01:11 < ironpillow> I updated the gist with error message: https://gist.github.com/ironpillow/6b455190c7821c240b5124636c8dfa82
01:11 < jml2> matt_, here's another cuz im bored lol https://wiki.ubuntu.com/multihead-over-network-gui
01:11 < ironpillow> jim: cool didn't know that. I added the error message to the gist itself so that I have documentation
01:11 < bls> bn_work: ah, then yes, you'll need sudo to invoke service to su to work correctly
01:12 < bls> bn_work: it's pretty rare to allow non-root users to start and stop services
01:12 < jml2> phogg, in case you're checking that fakexrandr things, there's that xdmx alternative ... i'm quite sure this is possible but I'm not quite going much into, sort of poking and reading about it
01:12 < bls> phogg: weird. I vaguely remember something about them doing that so they could get rid of wheel for RMS
01:13 < bn_work> also, was there at some point a `/etc/init.d/functions` on Ubuntu?  I read some post that claimed that there is a `daemonize()` in there but that file doesn't seem to exist on my box... except at `/usr/share/initramfs-tools/scripts/functions` which I'm wondering is something different?
01:13 < phogg> bls: why does RMS hate wheel?
01:14 < bn_work> bls:  ok, yeah, that works, I completely forgot about doing that... although the launched process tree is kinda weird
01:14 < phogg> bn_work: probably thinking of /lib/lsb/init-functions
01:14 < ironpillow> termbin link: http://termbin.com/r0r3 but it's doesn't show the actual output after the script exits
01:14 < bn_work> phogg:  https://unix.stackexchange.com/questions/155700/init-script-to-run-a-script-as-specific-user
01:15 < phogg> bn_work: on Debian-based systems you will also habe the start-stop-daemon command installed which can do part of what daemon(1) does.
01:15 < bn_work> oops, wrong link
01:16 < bn_work> phogg:  ah, so just wrap it with that?
01:16 < jml2> ah mattfly left lol.. oh well
01:17 < bls> phogg: https://ftp.gnu.org/old-gnu/Manuals/coreutils-4.5.4/html_node/coreutils_149.html
01:17 < phogg> bn_work: read the man page carefully, but yes. See specfically the --chuid option
01:18 < phogg> bls: I bet he really hates sudo, then.
01:18 < bls> or maybe it was the other way around, they replaced su from coreutils so they could continute to use wheel
01:19 < phogg> bls: possible; the upstream for the shadow package is ftp://ftp.pld-linux.org/software/shadow/ apparently. I don't have a RH box to check for theirs but I'll bet it's different.
01:20 < bls> $ rpm -qf /bin/su -> coreutils-8.4-31.el6.x86_64 on RHEL6.5
01:21 < phogg> hmm, I don't see a su.c in coreutils
01:21 < bls> I believe it may have been dropped at some point
01:21 < bn_work> bls:  hmm, how is not checking for wheel achieving what RMS is saying there?
01:21  * phogg goes to check 8.4
01:21 < bls> that info entry I referred to no longer exists
01:22 < phogg> bingo, 8.4 has a su.c but 8.28 does not
01:22 < bls> bn_work: it allows anyone that knows the password to become root, not just the people in the wheel group
01:23 < bn_work> ah, so if someone is told it, they can still admin-like stuff
01:23 < bn_work> s/still/still do/
01:24 < bn_work> VS being restricted to the original group
01:24 < phogg> bn_work: It was a different era. He's arguing in favor of *sharing the root password* and doesn't want wheel because it might allow you to not need to.
01:24 < bls> bn_work: yes. which was a bit deal back when a large portion of a company/university were all expected to work on the same machine
01:25 < bls> you could get the root password, but unless you could convince someone else in the wheel group to add you, it'd be worthless
01:25 < bn_work> so is that still the case?
01:26 < bls> multi user systems are so rare, and sudo so prevalent, it doesn't really matter
01:29 < roygbiv> hey all, what’s the “proper” way to save my default route in alpine 3.7 so that comes back after a reboot
01:29 < phogg> bn_work: Choose a Linux box you maintain. How many people need admin access to that box?
01:29 < triceratux> bn_work: as a general rule, in linux the wheel group is never the answer
01:31 < bn_work> so apparently the infamous LP thinks "sudo" is a "broken concept", is it?
01:33 < spammcoin> probably
01:34 < phogg> bn_work: no
01:37 < phogg> Having the ability to define rules that allow users to gain each other's privileges is not a broken concept. Sometimes a user needs to elevate his privileges for some purpose and there must be a way to do that. If allowing the policy deciding who can do it and when up to the system administrator is not the answer then I don't know what is.
01:38 < phogg> that said nobody understands sudoers and everyone writes broken sudoers files
01:38 < bn_work> yeah, it seems no really leverages the full power of it
01:38 < bn_work> s/no really/no one really/
01:39 < mateothegreat> true story
01:39 < mateothegreat> it's the same as disabling selinux first thing after install :D
01:40 < bn_work> selinux is a pain because it doesn't use sane timestamp formats
01:40 < bn_work> if they changed that one thing, it'd be a slightly easier pill to swallow
01:41 < bn_work> why no one has done that yet baffles me
01:42 < dh128> let's say I want to release a compiled binary for multiple platforms, including various Ubuntu and Fedora versions. what's the best way to go about doing this? set up 4 OS's? Let Launchpad PPA take care of the ubuntu side and make chroots for fedora?
01:42 < bn_work> selinux feels a lot like AD GPO settings, ie:  a huge table of security lockdown rules that one has to just know/lookup to properly manage it
01:44 < bn_work> oh well, thanks for your help phogg , bls!
01:44 < spammcoin> dh128: depends what the dependencies are
01:44 < bn_work> now to get something to eat before I fall over from starvation :/
01:45 < dh128> spammcoin, i'm pretty new to linux development but i'm talking about the dolphin-emu package
01:45 < dh128> so its dependencies are numerous, with stuff like gtk and qt
01:45 < spammcoin> RIP
01:45 < spammcoin> find something else that uses a sane toolkit
01:46 < truexfan81> you know what i find amusing about linux? when using secure boot, you have to manually sign the nvidia drivers or the kernel will reject them
01:47 < bazhang> dh128, is that the wii and gamecube emulator?
01:47 < dh128> bazhang, yessir
01:47 < bazhang> dh128, ubuntu has that in repos already
01:48 < dh128> oh sure
01:48 < dh128> but this is half a thought experiment half an actual test
01:48 < CompanionCube> well, if you don't care about the downsides you can just static link everything
01:49 < dh128> gtk can't be statically linked?
01:50 < CompanionCube> there's also appimage and friends
01:51 < navy_seal9614> hi guys. So I am having trouble with setting up a firewall rule for one of my docker containers. So to test, I did ufw deny 80 which should block everything, but it doesn't. After some googling I found out that docker adds some rules straight to iptables. Anybody knows what is the simplest way to make container ufw rules, without breaking existing rules in iptables
01:54 < dh128> CompanionCube, appimage sounds good but i'd still need to get dependencies and compile for multiple distros
02:01 < spammcoin> most of the time (MOST of the time) you don't need distro specific dependencies if the libs are of the same major version number
02:02 < spammcoin> or the distro does unexpected things with dynamic linkage
02:04 < dh128> i mean i tried compiling on 17.10 and running on 16.04 and that did not work
02:04 < spammcoin> why not
02:04 < dannylee> its a great day for linux...but windows must die a painful death????
02:09 < spammcoin> theres usually a way to make it work, unless it's bad, buggy code that requires some specific major + minor version number
02:16 < storge> it's a great day for linux?!
02:16 < jaziz> a great day indeed
02:16 < jaziz> even though I'm on windows
02:17 < jaziz> b/c my graphics card is garbage
02:17 < spammcoin> what card?
02:17 < jaziz> 750 ti
02:18 < jaziz> I think it's dying on me
02:18 < jaziz> screen just crashes and I can't do anything so have to manually reboot
02:18 < jaziz> happens when I play games on windows as well
02:18 < spammcoin> lol
02:18 < Psi-Jack> jaziz: "because", not "b/c", for future corrections on your part.
02:18 < jaziz> definitely the card b/c the crashes aren't so frequent when I lower fps
02:19 < jaziz> Psi-Jack, got it haha
02:19 < Psi-Jack> Thank you. :)
02:19 < jaziz> I mean if it's understood as shorthand, is it really so bad
02:20 < Psi-Jack> jaziz: It's against the rules. :)
02:20 < spammcoin> you're going to be fined 3 credits
02:20 < jaziz> oh no
02:20 < Sveta> i'm not sure how a smiling face is pertinent in such a sentence
02:20 < Psi-Jack> Nah. 3 bitcoin fines would be so much more better. :)
02:20 < jaziz> does he lose credits, too?
02:21 < spammcoin> no, credits are payable to spammcoin, LLC
02:36 < geheimnisse> Your current credit: Beryllium Level
02:52 < autopsy> Hello Nevolution2.
02:57 < ironpillow> rypervenche: pinging to see if the error output helped? thanks!
03:02 < rypervenche> ironpillow: What user are you running it as? This is AWS software, so you might have to ask them. Do you see anything useful in the crash log?
03:03 < rypervenche> ironpillow: In what directory are you running this?
03:03 < ironpillow> rypervenche: I am running as ironpillow and not as root, hence I am using sudo. The script is under root. Crash log show that same error. Directory: /greengrass/ggc/core/
03:04 < ironpillow> do you think hardware might not be supported?
03:04 < rypervenche> ironpillow: namei -om /greengrass/ggc/core/
03:06 < ironpillow> rypervenche: https://gist.github.com/ironpillow/6b455190c7821c240b5124636c8dfa82#file-01_namei
03:08 < rypervenche> ironpillow: I see online people running the software as a dedicated system user.
03:08 < rypervenche> https://docs.aws.amazon.com/greengrass/latest/developerguide/module1.html
03:08 < rypervenche> ironpillow: you also need to make sure you're doing your cgroup mounting, as they mentioned.
03:10 < ironpillow> rypervenche: ah...i see. thanks for you help :) I will follow that guide and get back if I encounter any problem.
03:35 < dannylee> hi
03:37 < mynameisdebian> Does anyone here use Selenium, and can recommend ANY way of getting it working?  This sounds like a dumb question, but if you actually use Selenium you will know wtf I am talking about
03:37 < dannylee> some video i can`t watch..i just need html5....but i still love fedora27...youtube is ok?
03:38 < mateothegreat> mynameisdebian: what _isn't_ working
03:40 < jim> I think youtube will do html5 videos
03:40 < dannylee> i wont do update on my machine...update mess my computer up
03:40 < jim> i.e., you don't need flash
03:41 < dannylee> ok man can you fix my machine
03:41 < mynameisdebian> mateothegreat: I used to run scripts every single day, but since the introduction of Geckodriver and Chromedriver I am unable to get it running in any capacity.  I have spent around 50 hours on various occasions trying to get the standalone version working.  I get tons of errors, which seem to be linked to known issues everyone is having, yet none of the posted temporary fixes ever work.  So last night I tried the official
03:41 < mynameisdebian> Selenium Docker container, which ran into an issue that was said to be related to the FF version, so I manually logged into the container and downgraded FF, which only produced a different issue
03:41 < jim> as once said before I said it again, that's 2 names!
03:42 < mateothegreat> mynameisdebian: yea, selenium is a beast... cypress is def. worth the switch
03:43 < Truxx> why is there no alpine channel? or did I miss its name?
03:43 < mateothegreat> #alpine-linux
03:43 < Truxx> ok, thx
03:43 < dannylee> i love the linux terminal..ill never quit using linux....windows spend alott of money to kill linux and there update
03:44 < mynameisdebian> matetothegreat: thank you, Selenium used to be great for me but I don't know what to do as a developer if I can't find a way to get back to old versions of the software.  I will try Cypress, which looks pretty nice
03:44 < mateothegreat> yep, I never looked back
03:45 < dannylee> openSuse is still a great choice for linux user...
03:45 < dannylee> openSuse is made in germany
03:46 < dannylee> i just like the fedora 27 terminal
03:46 < mynameisdebian> matetothegreat, do you know if tests can be written in Ruby?
03:47 < mateothegreat> you'll write tests in js
03:47 < dannylee> i love you guys an gals
03:47 < mateothegreat> I'm sure you could transpile but no need, it's as simple as it gets
03:47 < mateothegreat> https://docs.cypress.io/guides/getting-started/writing-your-first-test.html#Write-a-Simple-Test
03:55 < littlepython> is there a way to open new terminal from vagrant
03:55 < littlepython> i dont hav a gui yet
03:56 < mateothegreat> what do you mean "from" vagrant?
03:57 < mateothegreat> ...you're logged in to your vagrant machine over ssh... and want to ..?
03:57 < jim> littlepython, you just have virtual consoles?
03:57 < mateothegreat> or you're trying to figure out _how_ to connect to your vagrant machine?
03:57 < littlepython> jim: right now I am running through a git bash console
03:57 < mynameisdebian> littlepython: type "vagrant ssh"
03:58 < littlepython> mynameisdebian: i am already inside the vagrant
03:58 < mynameisdebian> oh sry
03:58 < littlepython> i want to open multiple terminal windows
03:58 < mateothegreat> use tmux
03:58 < mateothegreat> or screen
03:58 < littlepython> mateothegreat: sure
03:58 < jim> yeah, tmux or screen (probably tmux is better?)
03:58 < littlepython> yup
03:59 < jim> https://tmuxcheatsheet.com/ (config https://github.com/tony/tmux-config/blob/master/.tmux.conf)
03:59 < mateothegreat> after a spiffy ~/.tmux conf ;)
04:02 < jim> and don'/t worry, you can use tmux without a config
04:03 < mawk> but you need at least to write a few helpful lines
04:03 < mawk> to enable the mouse, at the very least
04:03 < mawk> and xterm keys for vim or emacs or nano or whatever
04:06 < littlepython> how can i continue the process to run in background
04:06 < littlepython> right now the process is running in foreground
04:06 < littlepython> I did not launch it with &
04:06 < jim> try ctrl z
04:06 < jim> then bg
04:06 < littlepython> jim: thats going to stop the process
04:07 < jim> pause it :)
04:07 < jim> it will pause when you ctrl z, then continue when you bg
04:08 < littlepython> sure
04:08 < jim> I dunno how you can do that "in one shot"
04:09 < littlepython> jim: syncing panes options is there in tmux
04:10 < littlepython> i am trying to use this command to display some content curl -X GET http://localhost:8090
04:10 < littlepython> but nothing is displaying for me
04:10 < jim> what's running on port 8090?
04:11 < littlepython> druid is running http://druid.io/docs/latest/tutorials/quickstart.html
04:15 < littlepython> jim: http://druid.io/docs/latest/tutorials/quickstart.html is there a way to run all the java commands listed in middle of the page to run as a bg job from a shell script
04:37 < daishun> Can someone explain to me why I would need to have a different block size for input and output when using dd?
04:41 < ananke> daishun: because the default one results in a lot of overhead
04:43 < daishun> I mean, having different values for input and output: ie input 512, output 1024. How does dd write more bytes than it reads?
04:49 < Sitri> daishun: dd can do some conversions to the data
04:49 < Sitri> Some of which will transform the amount of bytes
05:17 < Drakonan> ok so what should i install in my wsl
05:18 < Sitri> Whatever you're more comfortable with
05:18 < Drakonan> can you install everything in kali in ubuntu or vice versa?
05:18 < Drakonan> well im really wanting to do learning on it... idk maybe even try something different just to learn
05:18 < Drakonan> i am doing a lot of pentesting stuff... but need it as a general purpose linux sometimes
05:19 < ccqwtxt> run some VMs
05:19 < slidinghorn> Kali isn't meant to be used as a desktop, it's more meant to be a non-persistent live distro
05:19 < Drakonan> i was using msfvenom earlier today but that installed just fine on windows...
05:19 < Drakonan> wish you could install centos in wsl
05:20 < Drakonan> because im not super familiar with it anymore or the differences between it and the debian side
05:23 < Drakonan> was surprised it was even an offer really...
05:23 < alexey-nemovff> Drakonan: what is wsl?
05:24 < Drakonan> windows subsystem linux
05:24 < alexey-nemovff> got it
05:24 < alexey-nemovff> thx
05:24 < Drakonan> np its pretty nice... beats cygwin
05:24 < mateothegreat> Drakonan, docker -> centos :D
05:24 < Drakonan> inside wsl?
05:25 < Drakonan> i need to play around with that... they are talking about it at work and hadn't really messed with it yet
05:25 < mateothegreat> well, inside of wsl, you can have docker client
05:25  * mateothegreat wsl's all day long ... _into_ my centos containers
05:27 < mateothegreat> one nifty nugget most do not know about wsl is if you want to increase your speed 10-fold you'll want to add the system dir for wsl to the exclude paths for windows defender ^_^
05:29 < Psi-Jack> xem: Pick a nick and let it stick?
05:30 < xem> hi
05:30 < mateothegreat> mehhh, roll the dice.. see how many friends you make :D
05:30 < xem> the nicks are want are taken damnit
05:30 < xem> i want kex
05:30 < alexey-nemovff> wsl is a vile deed for the Linux world..
05:31 < mateothegreat> wsl + ConEmu
05:31 < justsomeguy> It's better with minty. Ncurses actually works with that terminal emulator.
05:31 < justsomeguy> *mintty
05:36 < nohop> No idea where to go with this. I'm adding a debug interface on an embedded board.
05:36 < nohop> What ansi code should i send (from the device to the serial terminal) whenever the backspace is pressed, to remove the last characted on the line ?
05:36 < nohop> backspace, space, backspace seems... dumb :)
05:37 < royal_screwup21> anyone know how to turn stuff happening on a given video into a gif? Speifically, I'm running a desktop app on Ubuntu 16.04 through Android Studio and when I run the app, it shows up on a window which I want to convert to a gif
05:37 < royal_screwup21> s/video/window
05:40 < mateothegreat> royal_screwup21, you're looking for a "screen recorder" .. peek, gifine are some apps
05:40 < royal_screwup21> ah thanks
05:51 < freeman42x]NixOS> question about Evince bookmarks: https://unix.stackexchange.com/q/435890/30584
05:54 < nejni-marji> was there ever a reason for /usr/bin/yes?
05:54 < Kumool> yes nejni-marji
05:55 < nejni-marji> I suppose it does make it easy to create a file with $n lines of a single string
05:55 < nejni-marji> yes $string | head -n $n
05:56 < nejni-marji> oh, it's for piping input into things to run non-interactively
05:56 < nejni-marji> that's clever
06:26 < rajrajraj> Can a websocket server initiate connection with websocket client?
06:27 < Sitri> No
06:27 < Sitri> It could initiate a normal TCP connection though...
06:31 < avis> hello
06:36 < paddy|> <beep>
06:38 < rajrajraj> Sitri:  how could it initiate a normal tcp connection
06:39 < Sitri> Whatever script is handling the websocket usually has a library that can also make TCP requests itself.
06:45 < mateothegreat> rajrajraj, if your client is "listening" for a connection, and your server knows how to connect to the client, you got it
06:45 < mateothegreat> effectively making it a client & server heh
06:46 < mateothegreat> more likely the case.. the client would connect to the server and your application layer would handle the bi-directional communication capabilities. This would be a "passive" approach, vs the previous "active" approach.
06:46 < mateothegreat> so if your node.js app and client subscribe to the same socket/channel, you would have your bi-directional communication
06:47 < rajrajraj> mateothegreat:  yeah got it. btw are request and response merely messages between two machines
06:48 < mateothegreat> yes
06:48 < rajrajraj> Sitri:  i could not understand that
06:48 < mateothegreat> it's a "communication pattern"
06:48 < mateothegreat> are you cheating on your homework? :)
06:49 < rajrajraj> mateothegreat:  no man, i am just confused with this websocket
06:49 < mateothegreat> gotcha.. pastebin your code if you get stuck
06:49 < ccqwtxt> mateothegreat "No I'm just curious. Also, what's the answer to this red black tree problem?"
06:49 < ccqwtxt> just kidding
06:49 < rajrajraj> ccqwtxt: lol
06:50 < Sitri> rajrajraj: What are you even asking then?
06:50 < rajrajraj> mateothegreat:  i need to decide between http and websocket. if websocket has active bidirectional capabilities than its a plus
06:50 < Sitri> Websocket is still HTTP
06:51 < rajrajraj> Sitri:  the communication handshake happens via http i understand that
06:51 < mateothegreat> rajrajraj, you're going to want to use the socket.io library
06:52 < mateothegreat> you can have a pub/sub channel which effectively gives you "bi-directional" communication
06:52 < mateothegreat> ..at the simplest level of abstraction
06:52 < rajrajraj> Sitri:  but lets say once the handshake happened. i turn off both machines and then server comes up and sends request to client. is this potentially possible, via any kind of hack.
06:52 < ccqwtxt> hm... a webhook?
06:52 < Sitri> WhY?
06:53 < Sitri> What are you trying to do?
06:53 < rajrajraj> ccqwtxt:  i dont know what that is.
06:53 < rajrajraj> Sitri:  i am trying to establish an active bidirectional communication
06:54 < Sitri> Then websocket covers your needs alone
06:54 < mateothegreat> rajrajraj, https://socket.io .. few lines of code & you're up and running
06:55 < rajrajraj> Sitri:  my need is, if client comes up, then server should be able to send request to client. the client WILL NOT initiate a new connection, if possible use the old one.
06:55 < rajrajraj> mateothegreat:  this conversation is on protocol, not really on a third party library
06:55 < rajrajraj> i need to strictly know if that is possible or not
06:57 < Sitri> rajrajraj: Then the client is no longer a client, but a server.
07:01 < ccqwtxt> rajrajraj at work what we have set up is that in case of losing connection for any reason, the client simply tries to establish connection with the server on a loop
07:01 < ccqwtxt> the server gets rebooted, whatever; it's the client's responsibility, and we just write that into the client side code
07:03 < rajrajraj> ccqwtxt:  got it
07:03 < toothe> anyone here use Digital Ocean? if so, can it have Kali easily installed?
07:04 < ccqwtxt> rajrajraj also-- since you said you didn't know, this is a decent first article on webhooks https://hackernoon.com/webhook-vs-api-whats-the-difference-8d41e6661652
07:05 < JimBuntu> toothe, I also use DO. Unsure about Kali being "easily installed", but it's doable.
07:05 < JimBuntu> toothe, I should warn, that using it could violate the TOS... it would be suspicious to say the least.
07:06 < rajrajraj> thank you mateothegreat  and ccqwtxt
07:06 < centrex> Hello.  I have a question.  I can't seem to find a decent explanation of the difference between systemctl list-units, list-units --all, and list-unit-files, could someone help me out?
07:06 < ccqwtxt> rajrajraj you're very welcome
07:08 < JimBuntu> toothe, DO say they do not support installing it, use the docker instead
07:14 < Hey> how do you create a ddtgz from a disk image?
07:14 < CoJaBo> a wat
07:15 < Hey> its a filetype
07:15 < centrex> a tgz with dd?
07:15 < jim> a compressed tarball?
07:15 < mateothegreat> thats some way of the future stuff there
07:15 < Hey> I have a vhdx file.. I converted it to raw.  do I then dd then tar gzip?
07:15 < jim> if you can mount the image, you can use tar then
07:16 < CoJaBo> Hey: you're going to have to specify what the goal is
07:16 < Hey> goal is MAAS
07:16 < jim> chrismaas?
07:17 < CoJaBo> IIRC, mount it, then `tar cz` the contents
07:17 < jim> (and to think, it's only april)
07:17 < mateothegreat> maaso menos
07:17 < Hey> I prepared the image vhdx.  Now I need to convert to RAW disk image and compress tar.gz
07:18 < CoJaBo> Hey: you'll need to mount the raw image somewhere first
07:18 < Hey> So. after converting.. I then need to mount it?
07:18 < jim> so convert, then mount
07:19 < Hey> ok.. I'll try that
07:19 < jim> yeah, that's what I'[m guessing
07:19 < CoJaBo> Some of the VM disk formats can be mounted natively if there's a package for it
07:19 < jim> how long will the image take to convert?
07:20 < Hey> CoJaBo: I can mount the vhdx file
07:20 < CoJaBo> If you can mount the file as is, you don't need to convert it to raw
07:20 < jim> oh, then you might not need to convert
07:20 < Hey> hmm.
07:21 < jim> once you mount it, can you see the files in it?
07:21 < CoJaBo> something like: cd /mnt/diskimage;  tar czf /home/stuff/root.tar.gz .
07:21 < Hey> what is the relevance of ddtgz  "dd"
07:21 < CoJaBo> I have no idea what ddtgz would refer to
07:22 < jim> I guess that might be some other way of doing it
07:22 < CoJaBo> I'm assuming MAAS is asking for a root.tar.gz
07:22 < Hey> This is the clue I got:
07:22 < Hey> Please note, that sysprep may take a long time depending on whether or not you selected to also run updates. The final image is a RAW disk image compressed into a tar.gz. If you still see a vhdx, wait for the commandlet to exit. You may follow the progress of the sysprep by starting up Hyper-V manager and double clicking on the MaaS-SysprepXXX machine.
07:23 < CoJaBo> Yeh, that *sounds* like it wants the root.tar.gz
07:23 < Hey> So... it essentically fires up the vhdx file, syspreps it.. shuts it down then does the following
07:23 < CoJaBo> Kind of like LXD, which has HORRIBLE documentation, and you have to read the source to figure out what the input needs to be <_<
07:23 < Hey> CoJaBo: and how to achieve root.tar.gz?
07:23 < CoJaBo> That command I posted
07:24 < Hey> So.. mount the vhdx file
07:24 < Hey> then tar.gz the contents
07:24 < jim> where will you mount it?
07:25 < Hey> windows allows you to mount directly.. it creates the mounts in disk manager
07:25 < Hey> it actually has 2 partitions
07:25 < Hey> so.. not sure how to do that one.
07:26 < Hey> maybe that's where dd comes in
07:26 < CoJaBo> You'll have to check the documentation for whatever tool you're using to mount the disk image
07:28 < jim> I would do it this way: cd /where/its/mounted ; tar cfz root.tar.gz .
07:29 < CoJaBo> jim: The root.tar.gz MUST be outside of the current directory
07:29 < CoJaBo> Don't leave that bit out; I discovered this the hard way. Twice.
07:30 < CoJaBo> My personal record is 382TB >_>
07:31 < Hey> When I mount the disk image.. there are 2 partitions.. a reserved D: and F: partition
07:32 < CoJaBo> Hey: is the image a linux filesystem? if so, it may be better to mount it on linux; windows might not be able to read it properly
07:33 < Hey> CoJaBo: its a windows10 image
07:33 < CoJaBo> Hey: if it's a windows image on a windows host.. why are you asking in ##linux lol
07:34 < mateothegreat> busted!
07:36 < Hey> CoJaBo: good point.  MAAS is Ubuntu.  and it uses Linux conventions
07:37 < Hey> they use the qemu tool to convert images
07:37 < Hey> qemu-img
07:38 < CoJaBo> Hey: It sounds like you're finding the linux instructions; you'll need the ones for Windows. Assuming it's documented at all
07:38 < Hey> https://github.com/cloudbase/windows-openstack-imaging-tools/blob/experimental/WinImageBuilder.psm1
07:39 < Hey> I did everything other than create the final image.
07:54 < Hey> ok.. I was correct. it converts vhd to raw
07:54 < mathu> anyone use EternalTerminal and have tips for how to get it to play nicely with the 2fac setup I have here? i have a similar flow on this linux server and on a mac, and et can only connect into the linux server
07:59 < RahulAN> Hi all
07:59 < RahulAN> i have a 64 gb drive
08:00 < RahulAN> i am unable to mount it
08:00 < RahulAN> https://bpaste.net/show/b53e14429756
08:00 < ayecee> cool story
08:00 < RahulAN> ayecee: :(
08:01 < RahulAN> How can i mount it .. Any suggestions ?
08:01 < ayecee> i would suggest to describe what happens when you try
08:01 < RahulAN> when i try to mount
08:01 < RahulAN> mount: wrong fs type, bad option, bad superblock on /dev/sdc2,
08:01 < RahulAN> it says this
08:02 < ayecee> and also how you are trying to mount it
08:02 < RahulAN> I pasted link
08:02 < RahulAN> mount /dev/sdc2 /mnt
08:02 < ayecee> what should be on the disk?
08:02 < RahulAN> ayecee: in fdisk -l it says partition is unknown
08:03 < RahulAN> ayecee: it have normal documents .. PS : it is not bootable ,
08:03 < RahulAN> it was sd card in my mobile
08:03 < ayecee> what filesystem should it have?
08:03 < RahulAN> ayecee:  Not sure  .. but it should be fat FS
08:03 < ayecee> does it work in a different computer?
08:04 < ayecee> aside from the mobile
08:04 < RahulAN> ayecee: No it didn't work
08:04 < ayecee> what is the output of "file -s /dev/sdc2" ?
08:04 < RahulAN> ayecee: I tried to check in gparted , it says same
08:04 < RahulAN> ayecee: /dev/sdc2: data
08:05 < ayecee> this is not a normal filesystem. it may be encrypted or otherwise specially formatted for the mobile.
08:05 < heeen> so my chroot time was totally off, etc/localtime was not a symlink. I updated it to match my host one which is Europe/Berlin
08:05 < RahulAN> ayecee: Is it ? How to open it then ?
08:05 < heeen> now date still shows 6:00 when it is 8:00
08:06 < supernovah> IS it normal for inotifywatch to gives me a create event when I watch a newly created folder, that says the newly created folder itself was just created?
08:06 < ayecee> RahulAN: put it back in the mobile :P
08:06 < heeen> any idea?
08:06 < jim> did you ever use it?
08:07 < RahulAN> ayecee: It was not detecting with my mobile
08:07 < supernovah> I have a script that watches for a folder to be created, then kills the inotifywait that got that event, and watches that folder - but upon doing that it gets another event that THAT folder was just created also...
08:07 < RahulAN> ayecee: I will try with other mobiles as well
08:08 < jim> RahulAN, did you ever use it?
08:08 < RahulAN> jim: Yes 1 day back it was all fine
08:09 < jim> ok, what wereyou using it for?
08:09 < RahulAN> jim: It was an external storage in my mobile
08:10 < jim> what os is your mobile running?
08:10 < supernovah> oh I get it, multiple sources try to create the same folder, I assume those creates fail - is there a way to tell with inotifywait if an event was succesful?
08:10 < supernovah> example a create failed due to it already existing
08:11 < RahulAN> jim: it is Android
08:11 < Dagmar> Not with inotify.  That's a thing you do by checking the exit status of the call that created the directory
08:11 < supernovah> which would have to be done with strace?
08:12 < jim> RahulAN, ok, do you remember formatting it? how did you do it?
08:12 < supernovah> i g2g for a birt
08:12 < RahulAN> jim: I didn't formatted it :|
08:12 < jim> get me w birts when you get back!
08:13 < RahulAN> jim: It was all working till yesterday
08:13 < jim> RahulAN, ok, what exactly is it?
08:13 < RahulAN> jim: It have photos and files ..
08:13 < jim> ok, is it a usb stick?
08:14 < RahulAN> jim: No it is a SD card , now i have connected to my linux machine using USB Card Reader
08:15 < jim> ok
08:15 < jim> is that working?
08:15 < jim> also
08:15 < RahulAN> jim: It is getting detected, but unable to mount it
08:15 < JimBuntu> RahulAN, You said you were using it in your mobile, to be clear... you mean cellular telephone, right?
08:15 < jim> when you got it, it was already formatted?
08:15 < RahulAN> jim: mount: wrong fs type, bad option, bad superblock on /dev/sdc2,
08:16 < RahulAN> JimBuntu: Yes correct, a Smart phone
08:16 < RahulAN> jim: Yes  .. it was
08:16 < ayecee> also file -s /dev/sdc2 says "data", so no filesystem is detected.
08:16 < jim> sdc2... what about sdc1?
08:16 < ayecee> sdc1 is 16 megabytes
08:16 < jim> RahulAN, ok, how did you obtain it? (also brb)
08:17 < RahulAN> jim: sdc1 also showing as "data"
08:18 < USERNAME00> How many people have ever stayed at the Radisson Hotel while listening to that Tool album made in 1993?
08:18 < USERNAME00> I feel like a time traveling spy
08:18 < JimBuntu> 438
08:18 < USERNAME00> Oh Saskatoon.
08:18 < RahulAN> jim: https://bpaste.net/show/b53e14429756
08:19 < VjdfMQ> Hey all
08:19 < VjdfMQ> Is there any possibility to check if device sends any signal ?
08:19 < USERNAME00> Oh I was wrong it was release is in 1996, maybe?
08:19 < ayecee> vague question is vague
08:19 < RahulAN> jim: blkid /dev/sdc2 says : /dev/sdc2: PARTLABEL="android_expand" PARTUUID="e4cb36e7-c685-f5bb-e666-01ce85737b06"
08:20 < RahulAN> ayecee: jim  :  /dev/sdc1: PARTLABEL="android_meta" PARTUUID="0e5214ae-7b97-4439-804f-1b9c59f82443"
08:21 < VjdfMQ> I mean, is there any possible way to check if device, for example, touchpad is sending any signal when it's buttons are clicked ?
08:21 < Dagmar> RahulAN: Were you using Adoptive Storage on that sdcard previousluy?
08:21 < ayecee> sure. oscilloscope or logic analyzer.
08:22 < JimBuntu> xinput?
08:22 < VjdfMQ> xinput shows devices connected
08:22 < RahulAN> Dagmar: what is Adoptive storage ?
08:22 < Dagmar> RahulAN: That
08:22 < jim> hmm. ok, how did you get this thing?
08:23 < Dagmar> RahulAN: That's basically when you tell the phone to sort of spread out it's internal storage to include hte microSD card
08:23 < Dagmar> RahulAN: It'll repartition the microSD card for that
08:23 < Dagmar> RahulAN: This also makes the microSD card generally unusuable for anything else (or any other device) until you go and reformat the entire thing with only one partition, as exFAT
08:23 < Dagmar> RahulAN: Provided it's not currently bound to _that_ device you should be able to just tell the phone to format the thing
08:24 < VjdfMQ> So, xinput dones't show anything when buttons on "red area" ( https://i.imgur.com/cB6EzCc.png ) are being clicked
08:24 < supernovah> can someone tell me what it means when they say android is based off of linux, does it use a inux kernel?
08:24 < Dagmar> It does
08:25 < supernovah> and we're done, ty Dagmar
08:25 < VjdfMQ> supernovah: #android
08:25 < jim> RahulAN, hmm. ok, how did you get this thing?
08:25 < RahulAN> jim: Which thing
08:25 < Dagmar> The kernel is a _bit_ modified, and it's not the sort of thing you're going to ever want to try to compile yourself without *lots* of technical knowledge
08:25 < jim> the sdcard
08:26 < RahulAN> jim: I bought it  ..
08:26 < Dagmar> jim: I'm pretty sure it was previously being used in a phone that someone told it to use as adoptive storage
08:26 < Dagmar> It can be reformatted and erased without a big deal
08:26 < jim> ok, so you bought it formatted?
08:27 < jim> do you want to recover the data on the sdcard, or what do you want to do?
08:27 < Dagmar> jim: Recovering the data is not going to be possible
08:27 < Dagmar> Not without the original device
08:28 < supernovah> is there a default package that will tell me the absolute path of a globby one such as "../1/2/3/./../../"?
08:28 < jim> do you not have the original device, RahulAN
08:28 < RahulAN> jim: I want to recover data ..
08:29 < Dagmar> RahulAN: Without the device that was using it as adoptive storage, that data is _unrecoverable_
08:29 < Dagmar> It's entirely encrypted
08:29 < RahulAN> jim: I have that device ... But it saying unknown device
08:29 < RahulAN> Dagmar: is it not possible to recover :(
08:30 < Dagmar> Nope
08:30 < VjdfMQ> https://i.imgur.com/cB6EzCc.png At this image you can see that the green area is being shown in events ( xinput ), but the red one doesn't. What could this mean ? The thing which is important here is that after a kernel's update the whole thing, but touchpad weren't working. And after rebooting with old kernel it worked again, but not these buttons in the red area.
08:31 < CoJaBo> RahulAN: what's wrong with the device?
08:31 < jim> ok... it's a very good chance Dagmar is right and the data is not recoverable... but I do have a few more questions, starting with, did you buy the mobile with the sdcard in it?
08:32 < Dagmar> There's a 100% chance Dagmar is right because Dagmar actually spent a lot of time messing with that particular feature
08:32 < shan> hi'
08:32 < ayecee> that's a pretty good chance
08:32 < jim> hi
08:32 < shan> hi
08:32 < RahulAN> Dagmar: I have that device with me
08:32 < shan> what's the dnf equivalent of apt update?
08:33 < RahulAN> Dagmar: If i put the card in there .. it says unknwon device.
08:33 < jim> Dagmar, is it that you need a password to get to the data?
08:33 < ayecee> was the mobile wiped?
08:33 < ayecee> jim: it's that the encryption key is gone
08:33 < Dagmar> If the phone was reset, it'll no longer have the key required for decrypting what was on the sdcard
08:34 < Dagmar> Adoptive Storage basically _requires_ that the microSD card be encrypted
08:34 < RahulAN> Dagmar: Is the mobile defaultly do the encryption of SDcard FS ?
08:34 < ayecee> that is what he said, yes
08:35 < RahulAN> ayecee: Yes . .I seen, my message went late ;)
08:35 < RahulAN> Dagmar: So if it is encrypted .. does the key is in my mobile FS ?
08:36 < CoJaBo> RahulAN: Was the phone reset/wiped?
08:36 < jim> RahulAN, this is probably the last question I have... when you rember back to the time it was working, what happened near the time it stopped working?
08:36 < RahulAN> CoJaBo: No the phone is not reset/wiped
08:37 < jim> RahulAN, this is probably the last question I have... when you rember back to the time it was working, what happened near the time it stopped working?
08:37 < RahulAN> jim:  I am not sure about this .. I think it was rebooted and Bang ;|
08:37 < CoJaBo> RahulAN: If the phone was not wiped, and the data is important enough, there's a fair chance to have it professionally recovered; but it'd likely cost a few hundred dollars.
08:38 < Dagmar> The only way to get at the data on the thing now is _through the device it's bound to_
08:38 < Dagmar> It will, for all intents and purposes, be part of the internal storage of the phone
08:39 < RahulAN> CoJaBo: but as the Dagmar said it is encrypted one
08:39 < RahulAN> CoJaBo: How will it get recovered ?
08:39 < CoJaBo> RahulAN: They'll need the phone and card both to have any chance
08:39 < Dagmar> You can no longer read anything useful from it by attempting to directly access the thing
08:40 < RahulAN> Dagmar: I will try to put this card in mobile (other mobiles) to check if it works
08:40 < CoJaBo> RahulAN: That could only destroy the card
08:40 < Dagmar> They won't be able to read it
08:40 < CoJaBo> RahulAN: (Unless there's a chance that you got it mixed up with some other device..)
08:40 < jim> Dagmar, so rebooting the phone could have rendered it useless?
08:40 < Dagmar> They'll be able to see that it has two partitions, and it'll be able to see the partition labels (which you pasted here) and that's about it
08:40 < Dagmar> jim: It should not have done that
08:40 < Dagmar> jim: When it's working fine, adoptive storage works pretty nicely
08:41 < jim> why would someone choose adoptive storage?
08:41 < RahulAN> CoJaBo: Dagmar: If i keep it in the phone (original one where it was before)  can there be any possibility ?
08:41 < Dagmar> It's not the same as spanning volumes with LVM, and there are some caveats to how some apps will behave since sdcards are generally slower than the internal flash storage on a phone, but it works more or less exactly as advertised
08:41 < CoJaBo> jim: To get more storage space
08:42 < Dagmar> jim: Because basically having a phone with only 16Gb internal storage for apps can be a PITA if you want to install more apps than that
08:42 < RahulAN> jim: I think there is nothing a user choose,  it is all default
08:42 < CoJaBo> RahulAN: If it's not recognizing it, there's probably not much you can do without proprietary tools that only recovery companies have access to
08:42 < RahulAN> CoJaBo: Ohkk ..
08:42 < CoJaBo> AFAIK, most reputable places don't charge if unsuccessful, so there's that.. but if they can get it, it'll be expensive.
08:42 < Dagmar> It never played well on my S5 for reasons I couldn't figure out (well, I could tell it was hideously slow access but couldn't nail down why) and it was completely out of hte question on my S3
08:43 < Dagmar> I suspect Samsung had some special sauce in there that LineageOS lacked
08:44 < Dagmar> f2fs shouldn't have been as abominatively slow as it turned out to be
08:44 < jim> Dagmar, how does adoptive storage work?
08:44 < CoJaBo> Could've been a bad card
08:44 < Dagmar> jim: Basically, you stick a microsd card into the phone (and have a new enough version of Android), go into Settings->Storage and tell it to start using the card as adoptive storage
08:44 < Dagmar> There are some preliminary tests the phone does before it'll even _show you
08:45 < Dagmar> _ that option, most of which are speed related
08:45 < Dagmar> Like, if you didn't stick in a class-10 card or better, it won't show the option
08:45 < jim> Dagmar, so it's a form of compression?
08:45 < Dagmar> On a slow microsd card, any app attempting to use storage _may_ wind up touching storage on the microsd card, and it'll suck
08:45 < Dagmar> No, it's not compression
08:46 < Dagmar> It just spans the internal storage to include the microSd card stop an encryption layer, using FUSE
08:46 < RahulAN> Dagmar: when we have a encrypted File system, every time decrypting it for any operation , doesnt it cost time ?
08:46 < ayecee> kind of a union mount
08:46 < Dagmar> but keep in mind how horribly an app will behave if you're used to it having 40Mb/s throughput, and suddenly it's sometimes 10Mb/s or lower
08:47 < ayecee> RahulAN: technically yes, but the time spent on encryption is small compared to the time spend retrieving the data from the device.
08:47 < Dagmar> RahulAN: No, it's done transparently on the fly.  CPUs are plenty fast enough to do block chaining ciphers this way now without you even noticing
08:48 < Dagmar> One great example of an app that does _not_ work well anymore if you enable adoptive storage and the sdcard isn't up to snuff: Pokemon GO
08:48 < jim> is it possible to use the card "normally" (i.e., not "adoptive")?
08:48 < CoJaBo> Most modern CPUs have hardware accel for crypto operations too, so battery impact is also small.
08:48 < Dagmar> jim: Nope it's an all or nothing thing
08:48 < RahulAN> Dagmar: Or is it be like, it decrypt the card content on phone start up and then do what ever needed
08:48 < Dagmar> In theory you could manually partition the thing yourself and keep using a third, non-encrypted partition, but that would require a level of technical knowledge beyond which this question would not need to be asked
08:49 < jim> I don't understand... can you not format the card with (say) ext4 or vfat?
08:49 < Dagmar> RahulAN: It decrypts the data on-the-fly, just like it does to _all_ your data once you've enabled full filesytsem encryption
08:49 < ayecee> jim: yes, but you can't store apps on it that way
08:49 < Dagmar> jim>
08:49 < Dagmar> jim: By default it just allocates the _entire_ sdcard to those two partitions you saw
08:49 < Dagmar> You're not normally allowed to stick apps on an sdcard without (at the very least) rooting the phone, which is going to cause some other problems
08:49 < RahulAN> Dagmar: is it done by all phone vendors ?
08:50 < jim> and the first (and smaller) partition holds what?
08:50 < Dagmar> RahulAN: Is what done by all phone vendors?
08:50 < Dagmar> metadata
08:50 < CoJaBo> RahulAN: Any phone that allows apps on the SD, has to do that
08:50 < RahulAN> Dagmar: I mean this encrypting Sdcard is done in all mobile phones ?
08:50 < ayecee> RahulAN: it's a feature of android, so yes
08:50 < Dagmar> jim: Note that exFAT has about the same permissions/ownership information as vfat, i.e., next to none
08:50 < jim> where is this encryption key held? can it be backed up?
08:50 < CoJaBo> RahulAN: If you just want to put photos on it, don't chose the adoptive/encrypted option.
08:50 < Dagmar> This is _woefully_ insufficient for what's required, so there's a FUSE metadata layer that handles all the SElinux contexts and unix permissions
08:50 < ayecee> jim: internal to the phone, and no.
08:51 < RahulAN> jim: It is in internal storage .
08:51  * TyrfingMjolnir listens to ABRACADABRA - Listen to the Picture
08:51 < RahulAN> ayecee: Why can't it be backed up
08:51 < Dagmar> jim: The encryption key for the adoptive storage volume is held in the internal storage of the phone
08:51 < RahulAN> ayecee:  as we have phone working .. it mean intenal storage is working
08:51 < jim> right
08:51 < Dagmar> So if someone FRs it, it's basically gone
08:51 < ayecee> RahulAN: because you can't retrieve it to back up.
08:51 < jim> I got that part :)
08:51 < Dagmar> They did a good job making this pretty massively difficult to bypas
08:52 < CoJaBo> The key is usually in some kind of secure storage that you can't read without special proprietary hardware, or sometimes even with special proprietary hardware.
08:52 < jim> what's FR?
08:52 < ayecee> factory reset
08:52 < Dagmar> CoJaBo: On devices that shipped with 7.x or newer, yeah
08:52 < jim> oh
08:52 < ayecee> (i'm guessing)
08:52 < RahulAN> ayecee: can't we access it from adb ?
08:52 < Dagmar> The dm_verity stuff is pretty hardcore
08:53 < ayecee> RahulAN: i don't know, but i doubt it.
08:54 < CoJaBo> If you could, that'd be a pretty huge exploit lol
08:54 < jim> this all sounds pretty unstable to me
08:54 < ayecee> it's intended that the sdcard remain permanently in the device.
08:54 < ayecee> if you're using adoptive storage, you can't remove it.
08:54 < CoJaBo> It's almost funny the level of protection modern devices have, when it all gets backed up insecurely to the clown anyway..
08:55 < ayecee> heh, clown
08:55 < jim> and the clown would say "will there be fries with that"
08:55 < RahulAN> ayecee: It means if removed this is the problem we can face ?
08:55 < ayecee> yes
08:56 < jim> looks that way
08:56 < RahulAN> jim: That's insane :(
08:56 < CoJaBo> I, for one, welcome our New World Overlords https://i.imgur.com/Y7DH5ad.png
09:00 < rendar> [root@rasp0 /]# mount /dev/sdb /backup
09:00 < rendar> mount: /dev/sdb: can't read superblock
09:00 < rendar> can i solve this?
09:00 < ayecee> doubtful
09:00 < rendar> fuck this! i have just formatted my hard drive with ext4 and copied some data
09:01 < rendar> is ext4 so fragile?!
09:01 < rendar> it's not the first time that this shit happens
09:01 < ayecee> okay mister ranty
09:01 < ayecee> don't lose your shit
09:01 < ayecee> do you have a partition table on the device?
09:01 < rendar> nope
09:01 < rendar> /dev/sdb is directly ext4, no partitions
09:02 < ayecee> what does file -s /dev/sdb say?
09:02 < rendar> /dev/sdb: ERROR: cannot read `/dev/sdb' (Input/output error)
09:02 < ayecee> think we've found the problem
09:02 < ayecee> and it's not an ext4 problem
09:03 < rendar> is the hdd damaged?
09:03 < ayecee> don't know yet, but that would be the kind of error you'd see if it was
09:03 < ayecee> maybe check dmesg output
09:03 < mirazi_heket1> hey, i want to set up vnc server so i can connect to my existing session (same screen on both pc's) - on the server station im using deepinos (debian fork) with tightvnc. So far i can connect but ive got empty gray screen. What should i correct?
09:04 < rendar> [339891.874084] blk_update_request: I/O error, dev sdb, sector 240
09:04 < rendar> [339891.874857] blk_update_request: I/O error, dev sdb, sector 480
09:04 < ayecee> sounds like hdd damage. you can maybe disconnect and reconnect it, or reboot
09:06 < ayecee> err. not reboot, but power off/wait/power on
09:07 < rendar> ok
09:10 < supernovah> !isarray
09:10 < supernovah> oops
09:32 < mirazi_heket1> hey, i want to set up vnc server so i can connect to my existing session (same screen on both pc's) - on the server station im using deepinos (debian fork) with xtightvnc. So far i have connected, but ive got new desktop session. Config: http://dpaste.com/0ZT81XZ http://dpaste.com/2SZVWFM
09:33 < mirazi_heket> and on a sidenote, lightdm here
09:33 < CrazyTux> hello, how a non technical end user assess the level of security a distro has?
09:34 < Deusdeorum> I have site which is protected with .htpasswd -> /etc/nginx/.htpasswd , now I need to add a another site with a different user:passwd, can anyone redirect me to a useful link on how to do this? It has been years since I did it last time
09:34 < CrazyTux> suppose I want to test Mint's security vis a vis Ubuntu's.
09:34 < greenit> hi, does anyone have a hp envy x360 15z (Ryzen edition) ? I'd like to know how well linux works on this notebook and had hoped that someone already tried it
09:35 < CoolerX> i am getting an error trying to build the gmp library
09:35 < peetaur2> CrazyTux: security is complex... and people disagree on fundamental things. Like our great leader Linus Torvalds himself has a strange view that if you can't enable the security feature for everyone by default, it's a garbage feature (even though the rest of the world accepts that security features compromise other goals, so you should pick carefully per system)
09:35 < mirazi_heket> Deusdeorum: man htpasswd
09:36 < CrazyTux> peetaur2, what's your opinion on Linux Mint. I want to install it for some of my friends.
09:36 < Deusdeorum> mirazi_heket yes, but I don't see how to use different user:pass for different sites
09:36 < peetaur2> CrazyTux: so I think you have to it highly subjectively, and based on patterns resulting in estimations and general impressions, not some hard metrics.
09:37 < peetaur2> CrazyTux: so for example read the mailing list...do they fix vulnerabilities quickly? do they have more vulnerabilities than others (stable distros like old debian and centos will have fewer than bleeding edge like arch/gentoo)
09:38 < peetaur2> CrazyTux: and look at how they handle things...like when systemd gets a bug report that setting umask=0000 (like chmod 777) is a horrid idea and has plenty of side effects (linking lots of bugs caused by it), they just ignore it and WONTFIX
09:38 < CrazyTux> I was advised here to stick to some main distros only and I feel Mint is both popular and newbie friendly.
09:38 < CoolerX> can someone help?
09:38 < CoolerX> https://paste.linux.community/view/raw/52a87d33
09:38 < greenit> does gentoo have many vulnerabilities? didn't see many mailing-list posts about it
09:38 < mirazi_heket> Deusdeorum: https://docs.nginx.com/nginx/admin-guide/security-controls/configuring-http-basic-authentication/
09:38 < CrazyTux> I myself am a non technical end user.
09:39 < peetaur2> or when they get a report that when they handle a User=3whoever and it's a valid linux name even though systemd rejects it and just assumes User=0  (as if that somehow makes sense), it's a feature, not a bug https://news.ycombinator.com/item?id=14682210
09:40 < Deusdeorum> mirazi_heket wouldn't all the user:passwd work for every site then?
09:40 < peetaur2> CrazyTux: mint is popular and well liked by its users, but it is considered one with slopper design with more potential for security issues, but still convenient etc. (like I said...security is like a tradeoff...convenience vs security; can't have default 100% both)
09:40 < greenit> CrazyTux: mint is good and pretty user-friendly. a non-technical buddy managed to set it up on his notebook on his own. another good distro (imo, though not checked for security) is manjaro. my parents use it after i've installed it and they are able to do most of the things on their own (except if the printer doesn't want to print, but that's another story)
09:40 < mirazi_heket> Deusdeorum: if you want the same user name with different password, you can create and use different file
09:41 < Deusdeorum> mirazi_heket ah
09:41 < mirazi_heket> if that's what you meant
09:41 < CrazyTux> peetaur2, suppose I apply updates of all levels in Mint's update manager can I expect the OS to be as secure as Ubuntu?
09:41 < peetaur2> CrazyTux: one time Mint's site got hacked and a malware infested iso was posted with wrong checksum, and to fix it all they did was replace the iso...then it got replaced with the hacked version again. So there are fails like that..they should have figured out how the thing got replaced in the first place, and something automated to take it down the instant it's bad again. Or maybe simply apparmor would prevent it (plus log it).
09:41 < Deusdeorum> mirazi_heket most likely ye, and just to be clear, i can put .htpasswd anywhere right?
09:41 < Deusdeorum> and just configure my sites nginx conf
09:42 < CrazyTux> by default Mint doesn't apply all the updates/upgrades. That is the issue that people talk a lot about, I think.
09:42 < mirazi_heket> Deusdeorum: as long as it's not /tmp /dev /proc and so on
09:42 < Deusdeorum> thanks
09:43 < greenit> doesn't mint have the "update-stages" like 1, 2, 3 where 1 and 2 get installed automatically and 3 only if the user says so?
09:43 < CrazyTux> peetaur2, what is the closest substitute of Mint? in beginner friendliness.
09:43 < peetaur2> CrazyTux: I can't say surely, and I don't use mint, but I would think it would be slightly less secure...just some sloppiness will make them miss a fix or delay one; but updating often is a good plan on any system.
09:43 < peetaur2> updating within the same stable release = more secure, updating to new bleeding edge = randomly more, less, same security
09:43 < CrazyTux> peetaur2, ok
09:44 < greenit> CrazyTux: take a look at Manjaro too ( https://manjaro.org/ ), maybe it's something to consider for you too :)
09:44 < peetaur2> CrazyTux: I think Ubuntu is probably the closest for such a goal...and without Unity it wasn't so buggy, and now Unity is gone, so maybe it's fine always
09:45 < peetaur2> I don't know how anyone tolerated Unity....any time I had to help someone fix a machine with Unity it was so bad... worst DE ever stability-wise
09:45 < CrazyTux> greenit, I am using Manjaro also.
09:45 < greenit> but a nice side-bar^^
09:45 < nindustries> Hi, has expansion come to ZFS-on-linux yet?
09:45 < CrazyTux> I needed to install a Linux OS for some of my friends.
09:45 < peetaur2> nindustries: expansion?
09:45 < Schrapnel> oi, can anyone tell me why I got banned from ##linux?
09:45 < nindustries> Adding an extra disk to a vdev peetaur2
09:46 < Schrapnel> 09:44:30 -!- Schrapnel ##linux Cannot change nickname while banned on channel
09:46 < CrazyTux> peetaur2, ok. I will try Ubuntu.
09:46 < greenit> Schrapnel: i may be wrong, but if you are _in_ the channel, are you really banned?
09:47 < Schrapnel> You can be banned, but not kicked I guess...
09:47 < peetaur2> CrazyTux: is it for you, or someone else? If you're already using Manjaro, it's probably pretty good.... it's like arch, more bleeding edge, so surely more stability and security bugs, but more convenience for new users
09:47 < Schrapnel> Anyway, I believe it has something with nick services.
09:47 < greenit> oh, ok
09:47 < Schrapnel> *todo
09:47 < peetaur2> I'm basically still using manjaro... at some point moving all machines to artix since manjaro dropped openrc support and the devs for it created artix
09:48 < greenit> afaik manjaro gets security updates faster than mint
09:48 < greenit> peetaur2: so you don't like systemd either?^^
09:48 < CrazyTux> peetaur2, I will try Ubuntu for others. I am using Manjaro now.
09:51 < VjdfMQ> Anyone ?
09:52 < VjdfMQ> https://i.imgur.com/cB6EzCc.png At this image you can see that the green area is being shown in events ( xinput ), but the red one doesn't. What could this mean ? The thing which is important here is that after a kernel's update the whole thing, but touchpad weren't working. And after rebooting with old kernel it worked again, but not these buttons in the red area.
09:53 < CrazyTux> peetaur2, btw, is Manjaro as secure as Ubuntu?
09:53 < Disconsented> Its as secure as you make it
09:53 < nindustries> ^
09:54 < junka> i just farted
09:54 < greenit> VjdfMQ: which distro?
09:54 < junka> sorry
09:55 < VjdfMQ> greenit: KDE Neon. Based on Ubuntu 16.04
09:56 < greenit> VjdfMQ: do you have a libinput.conf in your /etc/X11/ directory?
09:56 < greenit> including subdirectories
09:57 < VjdfMQ> It's could be because of apt full-u[grade. I've though that this could cause kernel's update to 4.13.0-38 from 4.13.0-37, but after rebooting into the previous one, it stayed the same.
09:57 < VjdfMQ> greenit: Nope. There's no such file.
09:57 < greenit> VjdfMQ: ok, what do you get when you run "xinput list" ?
09:58 < peetaur2> CrazyTux: if you use an old ubuntu lts (like released 6 months ago), you could feel pretty secure compared to a bleeding edge thing like Manjaro
09:58 < VjdfMQ> http://termbin.com/g429
09:58 < peetaur2> but manjaro seems ok...they aren't too sloppy; they've been sloppy with some things on the forum (like solution to ssl cert problem is change your clock :D) but otherwise they seem sane enough
09:59 < greenit> VjdfMQ: what's the output of "xinput list-props 11" ?
09:59 < peetaur2> greenit: no I think systemd is some ok goals done completely wrong, and the marketing is worse... full adoption before it's even half ready, before any sane polices are in place (like seriously, it's out there on production servers with umask=000?)
09:59 < VjdfMQ> greenit: http://termbin.com/p97j
10:00 < CrazyTux> peetaur2, ok
10:00 < peetaur2> openrc solves many of the goals, while still maintaining flexibility and funciontality of sysv (full bash syntax in your files if you want, but still can be done like declarative style, and no having to source the functions like start-stop-daemon and logging stuff)
10:01 < peetaur2> and it also solves the problem of badly written programs that are NOT daemons that call themselves daemons ... like systemd, you run it in non-forking foreground mode (non-daemon mode), and the rc system handles the stopping/restarting and such
10:03 < peetaur2> if you want to research the umask thing...look up the principle of least privilege; it's one of the core points in security, and permissive by default is the exact opposite
10:04 < peetaur2> like look at selinux and apparmor... you forbid everything by default, and then you whitelist things that you verified the process actually needs; you don't just assume it's fine to allow everything else until someone notices a side effect
10:06 < Dagmar> What's this about "foreground mode"?
10:06 < rendar> if i remotely ssh to a remote server i own, bash starts but it doesn't read .bashrc, if i run bash again, by typing 'bash' it will -- why?!
10:07 < peetaur2> Dagmar: a proper daemon must keep track of its processes and be able to terminate them; a forking one is hard to track because the pids change...so new rc systems come along and fixes problems in that with cgroups (which keep the processes together no matter how many subprocesses spawn), or making it so it doesn't fork at all (foreground mode)
10:07 < heftig> rendar: if it's a login shell, bash will run .bash_profile but not .bashrc
10:07 < Dagmar> rendar: Because of the long-standing tradition that shell rc files should be bewildering in their execution
10:07 < greenit> VjdfMQ: what does "xinput get-button-map 11" say?
10:07 < heftig> rendar: so source .bashrc from .bash_profile
10:08 < VjdfMQ> greenit: 1 2 3 4 5 6 7 8 9 10 11 12
10:08 < Dagmar> peetaur2: When exactly did this part ways from a "normal process"?
10:08 < peetaur2> I think all we have talked about are "normal" processes
10:08 < rendar> heftig: that's awesome thanks
10:08 < greenit> VjdfMQ: xinput list --long 11
10:08 < peetaur2> but look up the definition of a daemon...you'll see it's more than just "runs when you boot" or "runs in the background"
10:08 < Dagmar> rendar: If you check the bash man page it explains when each of the rc files is run and why, but it's going to make your head ache
10:09 < VjdfMQ> greenit: http://termbin.com/cpp1
10:09 < Dagmar> peetaur2: I'm perfectly aware of the formal requirements for something to be a daemon
10:09 < Dagmar> I still want to know exactly what's so special about a "foreground process"
10:09 < IntoxicatedHippo> I have an executable (firefox) which won't start and bash shows the error "No such file or directory", file says that it's an executable and ldd says that I have all the required dynamic libraries, what else could cause that error? https://paste.linux.community/view/cecc536b
10:10 < peetaur2> Dagmar: just that the pid the rc system sees is the same pid all the time it runs. this is like systemd with  Type=simple
10:10 < DLange> Dagmar: it has a controlling terminal connected
10:10 < rendar> Dagmar: lol ok
10:10 < Zad> hello, does this smartctl output mean that I should replace the disk asap?
10:10 < Zad> http://paste.debian.net/1018963/
10:10 < VjdfMQ> greenit: brb
10:11 < Dagmar> peetaur: So how is that different from a *normal* process.  I think you've wandered a bit too deeply into September with some of this stuff
10:11 < heftig> Dagmar: what are you arguing about?
10:11 < peetaur2> it's still a "normal" process
10:12 < Dagmar> Then why are you using "foreground" instead of "normal"?
10:12 < Zad> this is also from the same test: http://paste.debian.net/1018964/
10:12 < heftig> Dagmar: normal is very vague
10:12 < Dagmar> heftig: ...and yet "foreground" is both vague *and* misleading
10:12 < heftig> "foreground" at least draws attention to the fact that it's not daemonizing
10:13 < Dagmar> OKay, let me be frank
10:13 < Dagmar> Calling these things "foreground" and "background" is insane.
10:13 < peetaur2> because foreground has a meaning ...many daemons have a foreground or non-forking mode. like in sshd,   -D      When this option is specified, sshd will not detach and does not become a daemon.  This allows easy monitoring of sshd.
10:13 < Dagmar> They're just bloody processes.  They all have the same nice level in the vast majority of cases, and that's pretty much all the kernel gives a damn about
10:14 < Dagmar> That _daemons_ have a "foreground" mode is something specific to how they were build so that an admin can manually run the things, generally to get their log messages and or verbose operational details right there in that terminal
10:14 < Dagmar> s/build/built/
10:15 < peetaur2> and I was talking about non-daemons that call themselves daemons or are assumed to act like daemons, but really run in foreground/non-forking mode, or simply don't track their pids (like tomcat for example, which uses tcp sockets to talk to the main process)
10:15 < Dagmar> If they did not have that capability built into them specifically, they'd still just be normal bloody processes which have the ability to detatch from the tty, fork & exec, and get their own session
10:15 < DLange> Dagmar: man bash, / current terminal process group ID
10:16 < Dagmar> There's nothing magical about this and once you've read the specs it's not even exciting anymore
10:16 < Dagmar> ...and this buisness of carping about umask _for init_ is also quite mad
10:17 < greenit> VjdfMQ: what does "xinput list-props 2" say?
10:17 < Dagmar> That's a problem for the systemd people to worry about, but a sane and reasonable init doesn't ever create bloody files, and anyone writing their own daemon should know well enough to manage their umask correctly
10:18 < peetaur2> except it does create files, and there have been vulnerabilities found as a side effect, and instead of fixing the cause, they just fix the code that handles that one particular file
10:18 < Dagmar> DLange: The problem there is that's bash job control, which has eff all to do with daemons
10:19 < peetaur2> but at least we agree an init system should not create files, and systemd is not just an init system...it's a monster that includes more than init  (and the whole time I have been talking about rc, not init... look above and you'll see I said rc a few times and init 0 times)
10:19 < DLange> it is the kernels, just wrapped in bash or any other shell
10:19 < Dagmar> Unless, of course, someone's in the business of attempting to write a freakin' daemon in _bash script_ which I can tell you is just about as hideous a way to accomplish that as I have ever seen (despite that I've done a closest-possible one or two)
10:20  * DLange has written daemons in bash. Best practice. Unless perl :)
10:20 < peetaur2> and if the software is on your system, it's also a problem for you to worry about.
10:20 < peetaur2> there's a dhcp client written in bash I think.... but other than that I don't think I've seen a bash daemon :D  (who would do such a thing...)
10:21 < Dagmar> I'm no so sure that can really be called something written in bash
10:21 < Dagmar> They had to have called _some_ external tool to craft the request packets, not to mention get the responses
10:22 < Dagmar> peetaur2: Well, for things like Minecraft it's that or screen/tmux to keep the thing going
10:22 < Dagmar> I'm not saying what I wrote was debuggable or anything but an eyesore, but I did eventually get rid of the dependency on screen
10:22 < peetaur2> hehe yeah, but foreground/non-forking mode with the rc system makes it pretty easy too
10:24 < shortCircuit__> hello. I need a little help with a debugging thing
10:25 < Dagmar> I'm going to have to apologize to my wife tomorrow.
10:25 < Dagmar> I just adopted her Galaxy S7 and I've been running a custom firmware for so long I had no idea the amount of vendor bullsh*t she was dealing with
10:26 < shortCircuit__> basically I have a java app run using puma and it logs to syslog. but the problem is after I say, restart rsyslg, the date and time  for the logs appear fine. but after like 30 40 mintes they seem to start lagging by an hour or so
10:26 < shortCircuit__> in syslog
10:26 < shortCircuit__> how do I figure out why this lag occurs
10:26 < Dagmar> o.O
10:26 < Triffid_Hunter> Dagmar: heh that's why I'm waiting for lineageOS release for xiaomi mix 2 before I buy one
10:26 < Triffid_Hunter> shortCircuit__: DST?
10:27 < Dagmar> Triffid_Hunter: I have killed off this Visual Voicemail monstrosity before and I'll do it again tonight by god
10:27 < shortCircuit__> DST as in day light saving?
10:27 < Triffid_Hunter> Dagmar: I'm yet to use a smartphone without a 3rd party rom
10:27 < Triffid_Hunter> shortCircuit__: yes
10:28 < Dagmar> shortCircuit__: An hour "or so"?  If it's not exactly an hour you should consider piping the output of the date command to `logger` on some boxes to see whose timestamp you're really looking at
10:28 < Dagmar> If it's _precisely_ an hour, you might wanna check that everything is configured to be in the same time zone
10:29 < Dagmar> Like, _system wide_ not just via setting TZ in an rc script somewhere
10:29 < shortCircuit__> Fri Apr  6 15:28:37 WIB 2018 this is what date gives me. and Apr  6 14:17:46 is what is getting logged in the syslog
10:29 < hexnewbie> Isn't syslogd setting the timestamps? I'm not even sure if you *can* change TZ on the fly for the process.
10:29 < shortCircuit__> and this happens in all boxes
10:29 < hexnewbie> shortCircuit__: Is that the full date in syslog, or does it have year and TZ as well?
10:29 < Dagmar> Well, you can wind up with /etc/localtime holding some other spec, and someone setting TZ using login.defs or something
10:30 < shortCircuit__> its the full date. no TZ is there
10:30 < Dagmar> No no no...
10:30 < shortCircuit__> as soon as i restart rsyslog . this two will sync up for sometime
10:30 < Dagmar> TZ is an environment variable that when set affects the standard glibc functions for _formatting datestrings_
10:30 < hexnewbie> And a shorthand for timezone ;)
10:30 < anddam> hello, semi-OT but is there a graphical PDF viewer that will handle password on a file?
10:31 < hexnewbie> anddam: Okular has an option to ignore DRM.
10:31 < Dagmar> The /etc/localtime file is a spec file that is often improperly symlinked to a glibc timespec file that is the system's default time zone setting, basically
10:31 < shortCircuit__> the year is there. 2018 .. `Fri Apr  6 15:28:37 WIB 2018`
10:31 < VjdfMQ> ;So, greenit, any ideas :) ?
10:32 < Dagmar> But if you've got a difference of an hour and a few minutes, use the logger command to write _the current time_ being reported from `date` right into the syslog
10:32 < shortCircuit__> ok.
10:32 < greenit> VjdfMQ: not really, sorry.... however, what does "xinput list-props 2" output?
10:32 < Dagmar> ...because for sure your clocks _aren't_ synched up correctly and it's long past time you got NTPd working on these machines
10:32 < hexnewbie> shortCircuit__: Is it currently broken, i.e. off by an hour or so?
10:32 < shortCircuit__> yes
10:33 < VjdfMQ> greenit: http://termbin.com/4twy
10:33 < VjdfMQ> greenit: Thank yo uvery much for trying
10:33 < VjdfMQ> Really.
10:33 < Dagmar> It's entirely possible the time is just bloody _wrong_ on one or more machines
10:33 < shortCircuit__> and if something happens in production there is no way of knowing what on earth happened :D
10:33 < VjdfMQ> Just ... don't know what could this be. Just ... don't work.
10:33 < hexnewbie> shortCircuit__: Then do what Dagmar says, run ‘logger "TEST: $(date)"’ and paste here what test message appeared in syslog
10:34 < anddam> hexnewbie: thanks, more than ignore I want to change the password
10:34 < shortCircuit__> ok
10:34 < greenit> VjdfMQ: did you do "xinput list --long 2" or "xinput list-props 2" ? looks more like the first command
10:35 < shortCircuit__> Apr 06 15:34:57 p-some-box-name alexday[25730]: TEST: Fri Apr  6 15:34:57 WIB 2018
10:36 < Dagmar> OKay, so far so good
10:36 < VjdfMQ> greenit: Oh. Sorry. http://termbin.com/bl4wu
10:36 < shortCircuit__> hmh this is strange
10:36 < hexnewbie> anddam: qpdf perhaps, but I don't guarantee it
10:36 < shortCircuit__> journalctl -f and tail -f /var/log/syslog is different
10:37 < Dagmar> Well, yes
10:37 < shortCircuit__> in journalctl the time for the logs coming from puma.service shows correct time
10:37 < Dagmar> /var/log/syslog being a simple text file, and journalctl being a complex sin against god
10:37 < greenit> VjdfMQ: I'm curious, can you paste the output of "xinput list-props 13" please? :)
10:38 < VjdfMQ> greenit: http://termbin.com/39p6
10:39 < shortCircuit__> :D .. I have to see how journalctl and /var/log/syslog works. how come one is correct and other is wrong. is it like some queue buildup over socket , :/ possibly not
10:41 < hexnewbie> shortCircuit__: It's only incorrect for your service, but not for the logger test, right?
10:41 < shortCircuit__> yes
10:42 < shortCircuit__> the logger test didnot show up in the tail -f /v/l/syslog
10:43 < Dagmar> Seriously.  syslog is just a text file.  It'll never change.  What journalctl outputs is binary data that's been parsed back into text
10:44 < Dagmar> What you're seeing is the result of one of the reasons the older heads *do not* like journalctl
10:45 < Dagmar> That there are *any* occasions where it can output _incorrect data_ for _any reason_ are things we find deeply and fundamentally offensive and anathema to proper accounting
10:45 < anddam> hexnewbie: that's what I'm reading right now
10:46 < anddam> its help I mean, I find it odd that the only way to provide password is by using an option with the password in clear
10:46 < Zajt> Hi! Very stupid question but if I want to follow this tutorial: https://www.digitalocean.com/community/tutorials/how-to-set-up-apache-virtual-hosts-on-ubuntu-16-04 , how can I get the IP I can use instead of domain name?
10:46 < VjdfMQ> Zajt: #httpd
10:47 < shortCircuit__> ok I see. I tried the logger date thing in a vbox locally, that shows up in syslog. so I guess because a lot of logs are waiting to be written to syslog
10:47 < raindev> Hi there. Is it possible to do a bash expansion like [ab] but where a and b are char sequences, not individual characters?
10:47 < well_laid_lawn> there is a command called sync
10:47 < raindev> ^My googling skills failing me :(
10:47 < Triffid_Hunter> raindev: {ab,cd} ?
10:48 < VjdfMQ> raindev: #bash
10:48 < raindev> Triffid_Hunter: but that would try to match both sequences, not one of.
10:48 < raindev> VjdfMQ: alright, thanks
10:48 < shortCircuit__> Dagmar: can it be the applications fault?
10:49 < Zad> hello, is SMART test not applicable on GPT-partitioned disks?
10:49 < DarsVaeda> hi, how can I get the whole string passed to my bash script? I need to call something like "foo bar something" now I want to make a script that adds some parameters like "foo -p1 -p1 bar something", so I would need to pass "bar something" to my script
10:50 < Klaus_Dieter> smart is independent form partitions
10:50 < peetaur2> Zad: smart is on a low level independent of whatever data happens to be on the disk
10:50 < Zad> peetaur2, you're right. I just checked an mbr partition, and still gives the error 'unknown USB bridge'
10:51 < greenit> VjdfMQ: can you try if the following command helps? "xinput --set-int-prop 11 282 32 0"
10:51 < Zad> SMART rejects drives with this error
10:51 < heftig> DarsVaeda: if you have a script that's just `exec foo -p1 -p1 "$@"` (without the backticks) that will do what you want
10:52 < TJ-> Zad: you can pass a --device=TYPE argument to smartctl - check "man smartctl" for the options (to work with a few USB bridges)
10:53 < VjdfMQ> greenit: http://termbin.com/xhcu
10:53 < peetaur2> DarsVaeda: https://bpaste.net/show/31ce180d1926
10:53 < heftig> DarsVaeda: bash will replace "$@" with the current arguments (and not join or split them as "$*" or a bare $@ would do)
10:54 < DarsVaeda> but it is not arguments
10:54 < DarsVaeda> its like: "php myscript something" -> "php -p1 -p2 myscript something"
10:55 < DarsVaeda> the bashscript should add -p1 -p2 statically
10:55 < greenit> VjdfMQ: ok, maybe you'll have to write the option you want to edit:  xinput --set-int-prop 11 "Synaptics Off" 32 0
10:55 < heftig> DarsVaeda: you want to use "php" in your script and each time automatically add arguments?
10:56 < Zad> what do I need to do just to make an ordinary usb to be checked by smart? it used to work all the time
10:56 < VjdfMQ> greenit: http://termbin.com/2e85
10:56 < greenit> VjdfMQ: oh wait, that's deprecated, I read the documentation wrong. maybe this works: xinput --set-prop --type=int  --format=32 11 282 0
10:58 < peetaur2> DarsVaeda: your questions are vague....just make a vaporware pastebin... show the input data, the command, and desired output
10:58 < VjdfMQ> greenit: http://termbin.com/h5q1
10:58 < peetaur2> I'm not sure if you want $0 in there... like do you want  ./whatever 1 2 3  to run a command    ./whatever -p1 -p2 1 2 3? or  php -p1 -p1 1 2 3?
10:58 < greenit> VjdfMQ: what if you remove "--type=int" and "--format=32" from the command?
11:00 < greenit> so "xinput --set-prop 11 282 0"
11:00 < VjdfMQ> xinput --set-prop --type=int 11 282 0 and xinput --set-prop 11 282 0 worked
11:01 < greenit> do the buttons work now?
11:02 < VjdfMQ> greenit: Sorry, .. no
11:02 < peetaur2> (and btw, you don't need a while like my example...it just gives you more control.  more_args+=("$@")  would do the same  (other than not shifting)
11:03 < heeen> can you measure memory bus usage/contention
11:03 < heeen> on armv7, that is
11:04 < s10gopal> can anyone please build a kernel for me , it takes more than 3.5hours on my machine . 408c9861c6979db974455b9e7a9bcadd60e0934c is good and cbcd4f08aa637b74f575268770da86a00fabde6d is bad
11:05 < peetaur2> s10gopal: with -j4 it takes 3.5 hours???
11:05 < s10gopal> i can guide too
11:06 < V7> greenit: So ... really interesting, why it could be something like that...
11:06 < V7> I mean ... what could it be ... really.
11:06 < greenit> V7: what if you execute this command too? "xinput --set-prop 11 274 0"
11:07 < V7> greenit: No changes at all
11:07 < DarsVaeda> heftig: peetaur2: https://bpaste.net/show/f7bde5b5a4ed
11:07 < greenit> hm... that's a really strange problem^^
11:07 < V7> :)
11:07 < V7> mb
11:08 < V7> I'll to unmount it right now and mount again
11:08 < V7> I mean, the touchpad
11:08 < V7> I will try *
11:08 < peetaur2> DarsVaeda: this is not really the format I wanted.... so like this? https://bpaste.net/show/52e95e58129b
11:09 < s10gopal> can anyone please build a kernel for me , it takes more than 3.5hours on my machine . 408c9861c6979db974455b9e7a9bcadd60e0934c is good and cbcd4f08aa637b74f575268770da86a00fabde6d is bad
11:10 < DarsVaeda> peetaur2: yeah but that script is actually correct, it just did not echo it, which I tried first, ups... ^^
11:10 < peetaur2> DarsVaeda: https://bpaste.net/show/e3a8ced10527
11:11 < peetaur2> and I think an alias can do that too....  alias phpthing="php -some -argument"
11:11 < peetaur2> nd then   phpthing myscript data
11:12 < s10gopal> peetaur2: it takes more than 3.5hours to complete , you know any remote computer where i can build it  ?
11:12 < heftig> peetaur2: that's total overkill, just use "$@"
11:12 < Skunky> s10gopal: WTF is your computer that it takes that long?
11:12 < nothos> Hey all, I've got dovecot setup with a mysql backend, and I just tried adding a wildcard forwarder in the db as *@example.com, but when I try emailing I get a bounceback due to unfound user in the virtual mailbox table
11:12 < nothos> Is there an option that needs enabling for wildcards or does it use a different syntax than *?
11:12 < s10gopal> Skunky: it is just core i5 laptop
11:13 < V7> greenit: Really. Thank you for give a try
11:13 < Skunky> s10gopal: then you're obviously doing something wrong with your compile or your config.
11:13 < heftig> peetaur2: e.g. more_args+=("$@") does the same thing as your loop, without shifting
11:13 < peetaur2> s10gopal: I know plenty of computers where I could build faster than that, but I don't know where you can get one for free...maybe you should get a multi-core VPS ...maybe one that is prorated per cpu time you use
11:13 < s10gopal> Skunky: i am using default config
11:14 < Skunky> s10gopal: well, that explains a lot.  turn off all the stuff you don't need!
11:14 < peetaur2> like this machine would probably build it in 2 or 3 minutes  http://brockmann-consult.de/peter2/EPYC7401htop.png
11:14 < Skunky> s10gopal: my rpi only takes an hour
11:14 < s10gopal> peetaur2: i have collage email id , they provide free for students ? i tried google cloud but they are not accepting my debit card , i can pay too
11:15 < s10gopal> Skunky: i dont know how to make custom config
11:15 < anddam> hexnewbie: qpdf did the trick, thanks
11:16 < Skunky> s10gopal: in the 3.5 hours it takes for a kernel compile, you could have easily read enough to know what to do.
11:16 < peetaur2> s10gopal: and can you verify if that 3.5h number is when you used -j4?
11:17 < s10gopal> peetaur2: yes
11:17 < peetaur2> did you also let it finish without -j4?
11:18 < s10gopal> yes
11:18 < peetaur2> don't you have a desktop somewhere? most notebooks are pretty underpowered by comparison
11:18 < peetaur2> how long was it without -j4?
11:18 < s10gopal> and it is just core 2 duo 1.8ghz
11:18 < s10gopal> no ssd
11:18 < peetaur2> an outdated i7 920 I had before this one...like 7 years old... would build the kernel in something like 19-25 minutes
11:18 < s10gopal> amd 1gb ram
11:19 < peetaur2> ok that sounds slower than the 4 core 2.3 GHz or whatever you said before ...but you could try it
11:19 < peetaur2> 3.5 hours is overly high...remember I estimated it would be like 30-40 min?
11:20 < s10gopal> yes
11:20 < SuperSeriousCat> This 7 year old laptop also build kernel in 30ish min
11:20 < peetaur2> when you ran it with -j4, did you verify (like in top/htop) that it was using multiple cores?
11:20 < s10gopal> my desktop get crash easily
11:20 < s10gopal> yes
11:20 < s10gopal> but packing kernel was only using 1 core
11:21 < s10gopal> packing took around 1.5 hours
11:21 < peetaur2> heh oh that's pretty bad ....
11:21 < peetaur2> I foudn on the i7 920 it was about equal time 4corees+HT to build kernel vs 1 core packing
11:21 < Triffid_Hunter> s10gopal: hours? wth? I've never had a kernel take longer than 40 minutes even on a duron 800..
11:22 < SuperSeriousCat> Dont enable ALL the options :p
11:22 < s10gopal> Triffid_Hunter: i was using default confi
11:22 < peetaur2> if you want to be lazy and make a mess, you can cut the time in half... you can "make -j4" instead of deb-pkg, and then "sudo make install" (which ought to just put a file in /boot/vmlinuz-... and /lib/modules/...)
11:23 < peetaur2> but maybe it makes some other files you lose track of, and they leave a mess behind
11:23 < peetaur2> default config? I told you to take your distro's config and copy it there
11:23 < Triffid_Hunter> peetaur2: last time I checked, make install can call a system-provided out-of-tree script which can be customised.. very useful if you have a specific way you like your kernels installed
11:23 < s10gopal> i used time make -j4 deb-pkg
11:23 < s10gopal> peetaur2: yes i am calling it default confi
11:24 < peetaur2> well "default config" is a term used by the linux make file which is something entirely different (something that likely won't even boot)
11:25 < s10gopal> can i get aws ec2 or google cloud  without credit card ?
11:25 < V7> Could anyone suggest why ssh doesn't refresh a screen affer screen changes ?
11:26 < V7> Only after Ctrl+L
11:26 < djph> refresh what screen in wha... ?
11:26 < Triffid_Hunter> s10gopal: can buy pre-paid visas in many places specifically for buying things on the internet
11:26 < jim> what's a screen affer?
11:26 < V7> So, 70% screen just messes up, but after Ctrl+L it ok
11:26 < V7> after *
11:26 < Triffid_Hunter> V7: what makes the screen "mess up
11:27 < Triffid_Hunter> "?
11:27 < jim> oh :)
11:27 < V7> For example, text appendings
11:27 < s10gopal> can i buy them online ?
11:27 < V7> s10gopal: https://hostadvice.com/lp2/hosting-services/vps/?creative=208826910699&targetid=kwd-297024763871&matchtype=e&device=c&campaignid=837723732&adgroupid=45696376074&feeditemid=&loc_physical_ms=1011914&loc_interest_ms=&network=g&devicemodel=&placement=&keyword=cheap%20vps&target=&aceid=&adposition=1t1&gclid=EAIaIQobChMIzq6j06ql2gIVjk8YCh1TAQbqEAAYASAAEgL4BvD_BwE
11:27 < V7> oh dear
11:27 < V7> Sorry for such a big link
11:28 < s10gopal> b7thx
11:28 < s10gopal> V7: thx
11:28 < V7> http://www.hostingadvice.com/how-to/best-cheap-vps-hosting/
11:28 < V7> Take a look at google. There're a lot of good VPSs there
11:29 < V7> So. The problem. Default openssh-server is installed and default openssh-client is used.
11:29 < V7> So, sometimes the screen on connected ssh client is just messed up.
11:29 < V7> So, to refresh it the command Ctrl+L is used.
11:30 < V7> How to make it refresh the screen automatically ? The real mess starts when scrolling a file in nano or some others text editors
11:31 < jim> please expand thx
11:31 < djph> fix your $TERM perhaps?
11:32 < Triffid_Hunter> V7: can't reproduce, works fine for me
11:32 < V7> djph: xterm-256color is set
11:32 < TJ-> V7: that often happens if you've got a local 256 colour TERM setting which is exported. Usually it helps to reset to TERM=screen on the ssh session
11:33 < V7> s10gopal: Check this out http://www.uncensoredhosting.com/cheapest-uk-vps/
11:33 < V7> There're 1$-30$
11:35 < V7> TJ-: Might be that this command worked
11:38 < V7> TJ-: I have "export TERM=xterm-256color" in bashrc to make it colored
11:38 < V7> Is it possible to make it colored with TERM=screen ?
11:39 < TJ-> V7: see "man ssh" and the "ENVIRONMENT" section, about controlling what gets exported to the remote
11:46 < V7> TJ-: ENVIRONMENT doesn't have anything about TERM
11:51 < TJ-> V7: So? TERM is an environment variable. That section explains how exporting those variables is controlled, read the last bit "Additionally, ssh reads ~/.ssh/environment, ..."
11:52 < heftig> you shouldn't change TERM
11:52 < heftig> it's set by your terminal emulator and forwarded by the ssh client
11:52 < TJ-> V7: You'd add "TERM=screen" to $HOME/.ssh/environment
11:53 < heftig> don't do that
11:54 < heftig> if you really have to, use something like "case $TERM in xterm) TERM=xterm-256color ;; screen) TERM=screen-256color;; esac" in your bashrc or similar
11:55 < shortCircuit__> ping
11:56 < heftig> V7: that "export TERM=xterm-256color" is wrong
11:59 < TJ-> V7: if you want to set the remote terminfo permanently without needing to play with TERM= then do " infocmp | ssh user@remote-host 'tic -'  "
12:00 < greenit> the only time i encountered a buffer-problem with the terminal was when i used bash4windows...
12:06 < TJ-> V7: if the remote has an old version of 'tic' it won't take stdin (-) so you would need to do: infocmp | ssh user@remote-host 'cat > "$TERM.info" &&  tic -x "$TERM.info" '
12:31 < BluesKaj> Hi folks
12:32 < nobody> hi BluesKaj
12:32 < BluesKaj> hi nobody ...whoever you are :-)
12:50 < TaZeR> configuring as a loghost with syslog-ng, that sounds cool, i can review all the logs from my pcs on the laptop for example instead of running around to each one error checking
12:50 < TaZeR> is this as easy to do with systemd-journald?
12:51 < TaZeR> i find the way syslog did it by writing everything to a txt file kind of easier, for me at least
12:51 < TaZeR> i just found out about syslog-ng replacement
12:51 < TaZeR> if anyone uses it, id like to hear your experience :)
12:52 < s10gopal> how to setup google cloud to build kernel ?
12:52 < peetaur2> s10gopal: do you have a vps there with your distro?
12:53 < Guest82136> microsoft why https://screenshots.firefox.com/myvpZEgND8B0LFNV/www.howtogeek.com
12:53 < s10gopal> peetaur2, yes but it is command line
12:53 < ananke> TaZeR: rsyslog is a modern standard
12:53 < peetaur2> s10gopal: so just do it like you did before...install git, gcc, etc. and then build it
12:54 < s10gopal> peetaur2, how to copy .configure file from my laptop to that cloud ?
12:54 < TaZeR> ananke: oh wow i didnt know there was another one to chose from
12:54 < TaZeR> would you say its better than syslog-ng?
12:55 < peetaur2> s10gopal: if it has an ssh daemon and ip address, just use scp
12:55 < TaZeR> as in syslog-ng is not up to the modern standard perhaps?
12:55 < s10gopal> i dont know , but it was saying ssh when it was loading
12:55 < s10gopal> i dont know scp
12:55 < peetaur2> s10gopal: so configure ssh then.... or test if it's already there, and try using a client
12:55 < hexnewbie> Guest82136: Microsoft is embracing GNU/Linux and the command line, pining for their XENIX days?
12:56 < peetaur2> scp is just a hacky terrible script (fails at space handling, except when you give a dir without spaces and -r) that uses ssh to transfer files
12:56 < s10gopal> peetaur2, plz tell me cmd
12:56 < peetaur2> you need to learn how to research and study yourself or you won't progress very well
12:57 < Guest82136> hexnewbie, now i think they are just trying to annoy people into switching to Linux
12:57 < peetaur2> I gave you the terms you need to look it up... like ssh, so just look up how to configure that, and then when it's done, we can add to it (like I'd suggest restrict to pubkey auth only)
12:57 < peetaur2> (and disable dsa if they idiotically have it enabled like they shouldn't have since like 5 years ago)
12:57 < hexnewbie> peetaur2: Which is why I use tar, manually escaping what is needed on both ends.
12:57 < peetaur2> (I think I've seen dsa enabled even on builds where upstream removed dsa support ..like they patched it back in or something)
12:58 < ananke> TaZeR: yeah, rsyslog is better
12:58 < peetaur2> hexnewbie: I just scp -r the dir, use rsync instead, or put * in place of all the spaces :D
12:59 < peetaur2> if it is worth something, I am also a fan of rsyslog and have no idea why anyone uses syslog-ng ... I tried it and found nothing better, only issues with finding docs and examples which are easy with rsyslog
13:00 < hexnewbie> I only used syslog-ng in my Gentoo days because that was in the manual IIRC. I got somewhat used to the config syntax, and rsyslog has some confusion (legacy syntax vs. the modern one, plus bad documentation on the latter), but I no longer remember how you configured syslog-ng in the first place. I just remember it being fancy
13:00 < hexnewbie> Does syslog-ng even support the legacy syntax that rsyslog does, though?
13:00 < konimex> psst... socklog
13:00 < peetaur2> yeah the choice of multiple incompatible unequivalent syntaxes in rsyslog can be confusing, but I don't find it that bad since you can just use either
13:03 < s10gopal> peetaur2, it is possible to upload that file on google drive and then move it to vm?
13:04 < peetaur2> you should be able to just upload it right to the vm... just learn how to do that as part of your linux education
13:05 < peetaur2> and using ssh to connect should be your preferred way to build the kernel anyway, which is all you need to use scp too
13:05 < s10gopal> ok , i need to setup ssh on vm or on my laptop?
13:06 < peetaur2> ssh client on laptop (is it windows? try putty), and server on VM
13:06 < peetaur2> and also learn how to use either screen or tmux, so you can build the kernel inside that....that way if your client disconnects, you won't have to restart the build
13:07 < peetaur2> basically just install screen, then "screen -dR" and use it as normal (but the scrollbar won't work normally). And if you disconnect, just  screen -dR again to get it back
13:08 < JimBuntu> s10gopal, if using tmux... in it's simplest form... `tmux` on remote host to start a session... CTRL-B then D to disconnect from the session... `tmux a` to re-connect to session.
13:08 < s10gopal> peetaur2, done on my laptop and key is 2b:f1:23:61:5f:fe:ea:af:4e:c3:bb:80:6f:40:5b:45 gopal@gopal-HP-Notebook
13:09 < JimBuntu> CTRL-B then D are in caps for example purposes, you shouldn't also be using SHIFT
13:11 < s10gopal> how to fix it ? s10gopal@instance-1:~/linux$ ssh gopal@ 171.79.238.142
13:11 < s10gopal> ssh: Could not resolve hostname : Name or service not known
13:11 < edd_lc> Anyone know why I can't toggle bewteen keyboard layouts? I used `localectl --no-convert set-x11-keymap us,cz pc105 ,dvorak grp:alt_shift_toggle` as per https://wiki.archlinux.org/index.php/Keyboard_configuration_in_Xorg#Using_X_configuration_files. The /etc/X11/xorg.conf.d/00-keyboard.conf generated file looks ok, I've uncommented the language in /etc/locale.gen and ran locale-gen, can manually switch between layouts using
13:11 < edd_lc> `setxkbmap blabla`, grp:alt_shift_toggle is in /usr/share/X11/xkb/rules/base.lst and I've restarded X server. STILL, alt+shift will not toggle between languages
13:11 < peetaur2> and for choosing which to learn, I dunno... I never learned tmux; I have just heard people say things like screen has more, but the bindings suck, and screen is GPL and tmux is BSD if you care. (I don't until I want to contribute a patch...I don't like contributing with a license where some corp can steal it and monopolize your own patch)
13:11 < revel> JimBuntu: ctrl-b+d and ctrl-b+D act differently.
13:11 < JimBuntu> revel, that's why I added the statement about not using SHIFT
13:11 < s10gopal> peetaur2, done
13:12 < s10gopal> i am inside ssh  , now ?
13:13 < s10gopal> put i cant see prompt sigh
13:13 < s10gopal> sign
13:13 < peetaur2> s10gopal: install screen, gcc, make, etc. like you did before on your laptop
13:13 < konimex> screen is GNU so it must have some arcane emacs keybinding stuff
13:14 < s10gopal> peetaur2, already did sudo apt-get install git build-essential kernel-package fakeroot libncurses5-dev libssl-dev ccache
13:15 < zenix_2k2> guys is /etc/rc.local a file that available on most of the popular linux distros cause mine doesn't have it ( Ubuntu 17. )
13:15 < s10gopal> peetaur2, i am getting this error on google shell ssh: connect to host 171.79.238.142 port 22: Connection timed out
13:15 < V7> Thank you TJ- heftig
13:15 < lopid> not necessarily, zenix_2k2
13:15 < lopid> and apparently not
13:15 < zenix_2k2> so Ubuntu doesn't have that by default ?
13:16 < lopid> i don't have any ubuntus
13:16 < lopid> ubunti
13:16 < hexnewbie> zenix_2k2: Its use has been discouraged, and the implementation of services has become easier with systemd, which is why many distros have removed it. But you can create and re-enable the service for it (or make one if it is gone now)
13:16 < peetaur2> s10gopal: with missing context I'm not sure what you are doing that failed.... so did you ever connect from your client to the vm ssh server?
13:16 < peetaur2> the missing prompt sign might mean no
13:17 < s10gopal> ssh gopal@171.79.238.142
13:17 < hexnewbie> Which is annoying, since I don't want to create a service for all the temporary things I want at boot, when I can just have one.
13:17 < V7> So, the problem stayed
13:17 < s10gopal> it is wrong ?
13:17 < V7> Touchpad buttons don't work
13:19 < peetaur2> s10gopal: that command looks right except I can't validate ip or username...but the error basically says the server isn't running or firewall DROPs the connection
13:19 < peetaur2> s10gopal: so on the VM, check that it's running.... service ssh status
13:19 < peetaur2> and if not, start it... and if you want it always running, enable it on startup
13:20 < s10gopal> peetaur2, ssh start/running, process 1453
13:20 < peetaur2> so check the IP (ip a) and firewall (iptables -nvL)
13:21 < s10gopal> peetaur2, Perhaps iptables or your kernel needs to be upgraded.
13:22 < zenix_2k2> so another question, let's say i have created a startup.service in /etc/systemd/system and its contents look like this --> https://pastebin.com/NbfNwgAJ, so will /root/Desktop/script.sh be started when i restart ?
13:22 < peetaur2> uh what's that? iptables output? maybe it means the vps has no kernel access (like vserver, lxc, openvz maybe)
13:22 < zenix_2k2> and plus i have already chmod script.sh
13:22 < peetaur2> (unlike kvm, vmware, xen)
13:22 < s10gopal> i need to run thos command on vm ?  i was running them on my lapto[
13:22 < s10gopal> i need to run thos command on vm ?  i was running them on my laptop
13:22 < peetaur2> yeah on the server
13:22 < peetaur2> and run the iptables one as root
13:23 < s10gopal> server is my laptop right ?
13:23 < peetaur2> no, server is the VM
13:23 < s10gopal> i installed it on my laptop
13:23 < zenix_2k2> so will it work guys ???
13:24 < zenix_2k2> or i should replace [Unit] with something else
13:25 < dw> hey. super dumb question. i'm a monolingual english speaker and never used a linux box in a foreign language intentionally in my life. how would i enable e.g. german translations of tool output on debian or ubuntu?  i'm writing some code sensitive to the text of error messages, need to test it
13:25 < dw> i thought it was triggered by LANG or LANGUAGE or LC_ALL, but everything i've tried doesn't seem to have any effect
13:26 < s10gopal> peetaur2, done on vm , but how to find ip and username on vm?
13:27 < s10gopal> peetaur2, how to connect to ssh server now ?
13:29 < peetaur2> s10gopal: follow instructions from your vps provider to see how to access it... did they give you a console in the browser? some ssh instructions? maybe you didn't start the vm yet?
13:29 < s10gopal> peetaur2, got username by echo $USER and it is s10gopal
13:30 < s10gopal> peetaur2, yes i have console
13:30 < peetaur2> if you're not on the vm yet, it doesn't matter what $USER is
13:30 < peetaur2> ok so on the console  find out if ssh is running, and if there is a firewall block
13:31 < s10gopal> s10gopal@instance-1:~/linux$ service ssh status
13:31 < s10gopal> ssh start/running, process 10179
13:31 < s10gopal> .
13:31 < peetaur2> ok so now find otu the ip.... ip a
13:31 < peetaur2> is it the same as you had before? 171.79.238.142 ?
13:31 < xse> dw: i guess it depends on the distribution, in archlinux you set for example LANG=fr_FR.UTF-8 in /etc/locale.conf, but for it to be effective you need to uncomment the same locale in /etc/locale.gen, run "locale-gen" and reboot
13:32 < peetaur2> dw: you also need to install the locale package and run localegen.
13:32 < peetaur2> dw: locale -a   ought to list what you have installed
13:32 < peetaur2> * locale-gen
13:32 < dw> i have de_DE.utf8 installed, and doing e.g.  LANG=de_DE.utf8 ls --help, i get english. isn't there some way for that text to become translated too?
13:33 < s10gopal> peetaur2, http://paste.ubuntu.com/p/B3TQCrY87k/
13:33 < peetaur2> configuring /etc/locale.conf is system wide default... setting the environment just in your shell is also enough
13:33 < s10gopal> 171.79.238.142 is got by visiting whatismyip.com
13:33 < peetaur2> s10gopal: ok so this machine has a private IP address, not a public WAN one. So is there some forward rule or something to reach it that they gave you?
13:33 < s10gopal> no
13:34 < peetaur2> (see here it's in the class A network range https://en.wikipedia.org/wiki/Private_network#Private_IPv4_address_spaces)
13:34 < JimBuntu> s10gopal, is this a Google Cloud compute instance?
13:34 < s10gopal> yes
13:34 < peetaur2> ok well you could connect backwards to something that has routing/port forwarding...like your laptop, and then your laptop can ssh in to localhost
13:34 < peetaur2> a reverse tunnel
13:34 < s10gopal> i was doing that before
13:35 < peetaur2> if JimBuntu has Google Cloud compute instance knowlege, listen to his suggestions first
13:35 < JimBuntu> peetaur2, I don't use it much, but I'm checking where the public IP is on one of mine/etc. I rarely have to shell into them.
13:36 < s10gopal> peetaur2, this is external ip 35.185.79.26
13:36 < s10gopal> and this is internal ip 10.142.0.2
13:36 < peetaur2> ok, so try to ssh to that address then
13:37 < s10gopal> external or internal ?
13:37 < peetaur2> external
13:37 < JimBuntu> s10gopal, from the Cloud Platform console, on the left, hover on 'Compute Engine' so it opens the context menu, then clock on 'VM instances'
13:37 < JimBuntu> ok, you got it I see
13:38 < s10gopal> The authenticity of host '35.185.79.26 (35.185.79.26)' can't be established.
13:38 < s10gopal> ECDSA key fingerprint is 5e:57:55:47:3c:d8:77:ba:1f:24:58:b1:15:71:b6:0e.
13:38 < s10gopal> Are you sure you want to continue connecting (yes/no)? yes
13:38 < s10gopal> Warning: Permanently added '35.185.79.26' (ECDSA) to the list of known hosts.
13:38 < s10gopal> Permission denied (publickey).
13:40 < s10gopal> peetaur2, what does this mean ?
13:40 < peetaur2> looks good
13:40 < peetaur2> so now just generate a key on your client, and copy the .pub file appending it to the ~/.ssh/authorized_keys file on the server
13:41 < peetaur2> it says it's already configured like I said, to allow pubkey only, not password authentication
13:41 < peetaur2> which is not normally a default...so good job google on that one
13:41 < peetaur2> I guess they don't like hosting botnets
13:41 < s10gopal> ssh-keygen -t rsa ?
13:42 < peetaur2> yes... and I'd add -b 4096 for extra paranoia
13:42 < s10gopal> i ran that command on google cloud server
13:42 < funksh0n> Hello all.
13:42 < peetaur2> no run that on the client
13:42 < peetaur2> the cat ~/.ssh/id_rsa.pub  and copy and paste that into the server ~/.ssh/authorized_keys
13:42 < peetaur2> (or depending on the server config, it could be ~/.ssh/authorized_keys2)
13:43 < s10gopal> it is 72:e3:02:7a:5a:21:e2:fb:04:c4:24:f1:39:0c:e8:ff gopal@gopal-HP-Notebook
13:43 < peetaur2> that's the fingerprint, not the key
13:43 < peetaur2> put the pubkey in the authorized_keys file
13:44 < funksh0n> Is it safe to `dd if=/dev/zero of=/dev/sdx` on an SSD?  I want to wipe it but am not concerned about it being a secure wipe.
13:44 < JimBuntu> funksh0n, sure
13:44 < dw> zomg, so i figured it out. frickin' localepurge, maaan.
13:44 < funksh0n> JimBuntu: any tricks for speeding it up?  bs=4M or something?
13:44 < JimBuntu> funksh0n, obviously, be really careful that you get the drive/partition correct
13:45 < dw> "LANGUAGE=de ls foo" is sufficient to get german error messages, /if/ the box still has the relevant /usr/share/locale/de/LC_MESSAGES/coreutils.mo file installed. but habitual localepurge use going back decades means they're not :)
13:45 < funksh0n> Of course :)  All other drives removed anyway, booted off live usb just to do some tidying
13:45 < dw> lessons learned :)  thanks all
13:45 < JimBuntu> funksh0n, yes, bs=4M or maybe even bs=8M. This is going to be system specific though
13:45 < s10gopal> peetaur2, how ?
13:45 < peetaur2> s10gopal:  [13:42] <peetaur2> the cat ~/.ssh/id_rsa.pub  and copy and paste that into the server ~/.ssh/authorized_keys
13:45 < peetaur2> s/the/then/
13:46 < s10gopal> cat: /home/gopal/.ssh/id_rsa.pub: No such file or directory
13:47 < peetaur2> is that your notebook?
13:47 < peetaur2> run both the ssh-keygen command, and the cat on your notebook
13:47 < peetaur2> put it in authorized_keys on the vps
13:47 < s10gopal> yes
13:49 < s10gopal> how to paste on vm ?
13:50 < nobrain> you should use ssh-copy-id
13:50 < revel> ^
13:50 < peetaur2> ssh-copy-id won't work if he still doesn't have any way to connect via ssh
13:50 < revel> He doesn't? Oh.
13:50 < revel> s10gopal: VirtualBox guest additions?
13:51 < s10gopal> revel, ?
13:51 < peetaur2> one way is like this.... on server,  nc -l 9999 > id_rsa.pub   and then on client:     cat id_rsa.pub | nc 35.185.79.26 9999
13:51 < peetaur2> and then cat on server just to see if it's right...man in the middle attack is unlikely but easy to check.... and then append that into the authorized_keys file
13:51 < revel> You're using VirtualBox or VMware, right? They have "guest additions" that can let you copy text in the host and then paste it in the guest.
13:51 < peetaur2> revel: no he's using some google cloud thing
13:52 < revel> nvm then.
13:53 < JimBuntu> I think copy/paste will work in the VM window from Google.
13:53 < JimBuntu> Actually, I know it will.
13:54 < s10gopal>  nc -l 9999 > id_rsa.pub after this i am unable to see prompt on google cloud
13:55 < revel> That's fine. Just run the other nc command on your notebook.
13:55 < revel> It'll give you your prompt back after the file's transferred.
13:56 < s10gopal> cat: id_rsa.pub: No such file or directory
13:56 < revel> Generate it with ssh-keygen
13:56 < s10gopal> on laptop or on google cloud ?
13:57 < revel> Laptop.
13:58 < funksh0n> Does fsck only work on partitions e.g. `fsck.ext4 /dev/sdxy` or can I use it on a whole drive i.e. `fsck /dev/sda`?
13:58 < SkunkyFone> funksh0n: fsck works on filesystems
13:58 < SkunkyFone> funksh0n: so if you have a filesystem on a whole drive device, sure.  but that's almost never the case
13:58 < vlt> funksh0n: A file system can live on any block device.
13:58 < hexnewbie> funksh0n: fsck works on devices that contain filesystems, be they whole disks, partitions, device mapper devices, floppies, network block devices. What are you trying to do?
13:59 < funksh0n> Is there some equivalent disk checker SkunkyFone ?  I've deleted all the partitions and run `dd if=/dev/zero of=/dev/sda`.  I just want to ensure the drive is healthy before I sell it.
13:59 < s10gopal> on both device now i am not getting prompt sign
13:59 < revel> badblocks
13:59 < SkunkyFone> funksh0n: badblocks -w
13:59 < hexnewbie> funksh0n: A combination of badblocks and smartctl -t long, perhaps
13:59 < qman__> that's not what fsck is for - you want to use smartmontools to check the drive's intenral health reporting, and possibly badblocks to run tests
14:00 < funksh0n> I see, thank you all.
14:00 < s10gopal> peetaur2, what should i do ?
14:00 < revel> s10gopal: Nothing whatsoever on your laptop?
14:00 < hexnewbie> funksh0n: ‘badblocks -w’ is destructive, so be careful - it will overwrite disk with patterns, so *only* run it on that drive you wrote zeroes over.
14:00 < s10gopal> getting blinking cursor
14:00 < revel> s10gopal: It should be prompting you for a file path, though you can just hit enter like 5 times.
14:00 < funksh0n> Thanks for the heads up hexnewbie
14:01 < s10gopal> i can hit enter
14:01 < s10gopal> but still blinking
14:03 < s10gopal> got promot again now ?
14:03 < s10gopal> prompt
14:03 < revel> s10gopal: Just curious, but are you blind, not reading the output on the terminal or do you have a weird setup...?
14:04 < s10gopal> weired setup
14:04 < peetaur2> s10gopal:  cat ~/.ssh/id_rsa.pub | nc 35.185.79.26 9999
14:04 < s10gopal> on laptop or google ?
14:04 < revel> peetaur2: He doesn't have a key yet.
14:05 < nobrain> s10gopal: I think you should first try to understand what are you doing
14:05 < s10gopal> can i use pastebin it to get key from google cloud and use cat command to paste it on my laptop ?
14:05 < s10gopal> bastebinit
14:05 < s10gopal> pastebinit
14:06 < JimBuntu> s10gopal, You are creating the key on your laptop and then placing it on Google Cloud Instance.
14:06 < s10gopal> how to place on google instance ?
14:06 < JimBuntu> s10gopal, do you have the VM instance open in a browser window?
14:06 < s10gopal> yes
14:07 < JimBuntu> You can the netcat route that peetaur2 has been so kindly working on you with... if you follow their instructions closely...
14:08 < s10gopal> ok i will try again , please explain again
14:08 < JimBuntu> s10gopal, you can also create the key, then cat the key on your laptop, highlight the key, copy the key and then open the ~/authorized_keys file in the cloud instance and simply paste it right into it. The browser VM window accepts copy/paste
14:09 < JimBuntu> sorry peetaur2 I'm not trying to step on your toes, simply offering what may be a simpler route (since the VM window allows for copy.paste)
14:09 < s10gopal> copied key from my laptop terimnal
14:09 < s10gopal> now how to paste on vm ?
14:09 < JimBuntu> `sudo nano ~/.ssh/authorized_keys`
14:10 < JimBuntu> Once that opens, do you see it has at least one line of text already?
14:10 < revel> JimBuntu: sudo?
14:10 < JimBuntu> I was just thinking that revel, lol
14:11 < s10gopal> done
14:11 < JimBuntu> s10gopal, let's double check without the sudo line, and make sure that the file you opened has the newly pasted line in it, and that it's on it's own line
14:12 < JimBuntu> 'nano ~/.ssh/authorized_keys`
14:12 < s10gopal> still getting this on my laptop gopal@gopal-HP-Notebook:~$ ssh s10gopal@35.185.79.26
14:12 < s10gopal> Permission denied (publickey).
14:12 < JimBuntu> right s10gopal you are skipping ahead
14:13 < JimBuntu> now let's double-check the owner/permissinos are right for the keys file, on the vm... `ls -alf ~/.ssh/authorized_keys`
14:13 < peetaur2> JimBuntu: you're the one with experience with that...I already said he should prefer your suggestions
14:13 < s10gopal> getting game key on vm
14:13 < JimBuntu> You should bee it's owned by s10gopal and the permissinos should be "-rw-------"
14:13 < JimBuntu> s/bee/see
14:14 < JimBuntu> peetaur2, if I remember correctly, these google instances don't even have an 'll' alias
14:14 < s10gopal> it prints /home/s10gopal/.ssh/authorized_keys
14:15 < JimBuntu> s10gopal, try again with `ls -alF ~/.ssh/authorized_keys` please
14:15 < s10gopal> -rw-r--r-- 1 root root 1616 Apr  6 12:11 /home/s10gopal/.ssh/authorized_keys
14:16 < JimBuntu> `chmod 600 ~/.ssh/authorized_keys`
14:16 < s10gopal> i need to use chmod 600 ?
14:17 < ecraven> is there any reason why I should not be able to set my ELF entry point and virtual adress to 0x0? everything below 0x10000 seems to segfault :-/
14:17 < s10gopal> -rw------- 1 root root 1616 Apr  6 12:11 /home/s10gopal/.ssh/authorized_keys
14:17 < qazey> u+x.
14:17 < SkunkyFone> ecraven: where's the kernel?
14:17 < s10gopal> now ?
14:17 < ecraven> (hand-crafted elf, no sections, only one program header)
14:17 < JimBuntu> Well, 1616 seems a tad big... but we will come back to that
14:17 < qazey>  /boot/vmlinuz.
14:18 < ecraven> SkunkyFone: I'm assuming in the higher half of memory?
14:18 < JimBuntu> Next up, almost there... we need to change your SSH command line to include the identity file (on the laptop)... while I prefer to use a ~/.ssh/config ... we will test first at the command line. One moment
14:18 < SkunkyFone> ecraven: never assume...
14:19 < ecraven> SkunkyFone: well, it wouldn't fit below #x10000, would it :-/
14:19 < stevendale> Hi there
14:19 < stevendale> I'd like to sell all my personal info
14:19 < stevendale> Any buyers?
14:19 < JimBuntu> s10gopal,   `ssh s10gopal@35.185.79.26 -i ~/.ssh/id_rsa`
14:20 < revel> Shouldn't -i have the pubkey?
14:20 < JimBuntu> I'm guessing that's where we stored the private key, I didn't see that explicitly mentioned in the convo above
14:20 < revel> Guess not.
14:20 < s10gopal> Warning: Identity file /home/gopal/.ssh/id_rs not accessible: No such file or directory.
14:20 < s10gopal> Permission denied (publickey).
14:20 < revel> You missed an a.
14:21 < s10gopal> Warning: Identity file /home/gopal/.ssh/id_rsa not accessible: No such file or directory.
14:21 < s10gopal> Permission denied (publickey).
14:21 < JimBuntu> `find ~ -iname "id_rsa"`
14:22 < qazey> I prefer using Bash scripts for automatication.
14:22 < s10gopal> find: `/home/gopal/.cache/dconf': Permission denied
14:22 < s10gopal> find: `/home/gopal/.config/enchant': Permission denied
14:22 < s10gopal> find: `/home/gopal/.gvfs': Permission denied
14:22 < s10gopal> and when i use sudo , no output
14:23 < JimBuntu> s10gopal, We shouuldn't need sudo. Where was the pub file that you copied the key out of, to paste into the authorized_keys file?
14:23 < s10gopal> JimBuntu, sorry i am unable to understand it
14:24 < qazey> yep.  haiku programming for hunger.
14:24 < JimBuntu> s10gopal, You cat'd some text on your laptop a bit ago, and pasted it into the authorized_keys file on the VM. What is the path to the file you cat'd?
14:25 < s10gopal> JimBuntu, gopal@gopal-HP-Notebook:~$ cat id_rsa.pub
14:26 < JimBuntu> Ok, a couple ways to solve this... I'm going with this one... `cp ~/id_rsa ~/.ssh/id_rsa_gcp
14:26 < JimBuntu> then your new command line for SSH would be `ssh s10gopal@35.185.79.26 -i ~/.ssh/id_rsa_gcp`
14:26 < s10gopal> cp: cannot stat ‘/home/gopal/id_rsa’: No such file or directory
14:26 < qazey> missed the .ssh dir.
14:26 < kiwan> hi. can i dd a whole drive then restore it to different hardware and expect it to boot or is it bad practice?
14:27 < s10gopal> filex by using .pub
14:27 < JimBuntu> sorry s10gopal I'm confused. It's like it make a pub, but no private key? Also confused because we didn't find the id_rsa when running the find command earlier
14:27 < qazey> darn it tootin'.
14:27 < JimBuntu> ok, not confused about not finding the pub key, as I didn't say to look for "id_rsa*"
14:28 < s10gopal> gopal@gopal-HP-Notebook:~$ ssh s10gopal@35.185.79.26 -i ~/.ssh/id_rsa_gcp
14:28 < s10gopal> Permission denied (publickey).
14:28 < qazey> then use private.
14:28 < qazey> whoa.
14:29 < s10gopal> done
14:29 < s10gopal> i can find them
14:31 < s10gopal> gopal@gopal-HP-Notebook:~$ ssh s10gopal@35.185.79.26 -i /home/gopal/.ssh/id_rsa_gcp
14:31 < s10gopal> Permission denied (publickey).
14:31 < s10gopal> gopal@gopal-HP-Notebook:~$
14:31 < s10gopal> gopal@gopal-HP-Notebook:~$ ssh s10gopal@35.185.79.26 -i /home/gopal/id_rsa.pub
14:31 < s10gopal> Permission denied (publickey).
14:31 < qazey> tuts.  der ez.
14:32 < JimBuntu> s10gopal, `ssh-keygen -t rsa -b 4096 -f ~/.ssh/id_rsa_gcp && cat ~/.ssh/id_rsa_gcp.pub && echo "copy and paste the displayed pub key from this laptop into your GCP VM at ~/.ssh/authorized_keys"`
14:33 < JimBuntu> When asked for a passphrase, hit enter... when asked again, hit enter...
14:33 < s10gopal> i am doing the same
14:33 < JimBuntu> You should get something like 'Your identification has been saved in /home/gopal/.ssh/id_rsa_gcp.' and another line with .pub at the end. Please confirm
14:34 < qazey> firm.
14:35 < s10gopal> done
14:35 < qazey> At what coast...?
14:35 < qazey> Ahem, cost?
14:35 < JimBuntu> s10gopal, If so... and if `ssh s10gopal@35.185.79.26 -i ~/.ssh/id_rsa_gcp` fails again... I think you need to make a backup of the authorized_keys file and create a new one, with only the single line of output from ~/.ssh/id_rsa_gcp.pub
14:35 < s10gopal> i dont got this s10gopal@instance-1:~/linux$ sudo vi ~/.ssh/authorized_keys
14:35 < s10gopal> s10gopal@instance-1:~/linux$
14:36 < qazey> right.  we need more FHS stanzas to highlight the dynamic links and libs within shared environments.
14:36 < s10gopal> still geting denied on laptop
14:37 < JimBuntu> s10gopal, ok, backup the authorized_keys file, make a new one. The only line that should be in it, it the copy/pasted line from the .pub file on your laptop
14:37 < s10gopal> it is possible to remove publickey use it with no security ?
14:37 < s10gopal> how ?
14:37 < qazey> No.  Security always.  No password, no security.
14:37 < JimBuntu> s10gopal, yes, you could change it to allow password entry... not a good idea.
14:38 < s10gopal> i dont have any private info stored
14:38 < s10gopal> and i dont mind if an hacker steal it
14:38 < s10gopal> how to change it to password entry ?
14:39 < JimBuntu> yeah, kinda of a good pint since you are working on the kernel... also kind of a bad idea as someone could take over your machine, shut you out, do what they want... the right person could even try to get bad code into the kernel you want to fix and submit
14:39 < peetaur2> s10gopal: when you drop security, you add more problems in another way...mysterious ones that might even make you lose your payments and other things
14:40 < s10gopal> i dont have any payment
14:40 < peetaur2> like if it's hacked and they think you did it on purpose, they think you violated the terms, and you don't get a refund; sometimes they know it's not you and blame you anyway just to rip you off (google probably won't do that...smaller ones would)
14:40 < s10gopal> account connected with google is debit card which is only having .1$
14:40 < JimBuntu> s10gopal, if we can get you squared away with using a key, then using a key becomes easier as you don't even have to enter a password, we will set up your SSH client to auto-grab the proper key and send it for you.
14:41 < JimBuntu> within the /etc/ssh/sshd_config file is a "PasswordAuthentication no" line... you would replace the no with a yes... then sudo service ssh reload
14:41 < s10gopal> on vm right ?
14:41 < JimBuntu> I can't say for certain Google didn't make other changes, that's normally the first thing I change to "no" in normal distros
14:41 < JimBuntu> s10gopal, yes, on the VM
14:42 < JimBuntu> This method will still require a password, but you wont have to deal with a key... which is basically a really long password
14:43 < JimBuntu> Oh, s10gopal , you will now forever carry the sin of having changed that value to "yes"... but that's on you ;-)
14:43 < s10gopal> done
14:44 < peetaur2> just change it back to no once you deploy your pubkey
14:44 < JimBuntu> can you now SSH in, and not get slammed with "Permission denied (publickey)" ?
14:44 < s10gopal> i ony want to make 8 kernel on it
14:44 < s10gopal> and then i am not going to use it
14:44 < s10gopal> now what to do?
14:45 < JimBuntu> If you already did the sudo service ssh reload... then `ssh s10gopal@35.185.79.26`
14:45 < JimBuntu> the reload command was meant to be run on the VM, the ssh command on your laptop
14:46 < peetaur2> pubkey also has the advantage that you just ssh-add one time, and then after that you don't need to enter the password again for each new connections
14:46 < peetaur2> -s
14:46 < s10gopal> ssh: Could not resolve hostname reload: Name or service not known
14:46 < s10gopal> ssh restart ?
14:46 < JimBuntu> restart
14:46 < qazey> reload.
14:46 < qazey> service reload.
14:47 < s10gopal> sudo restart ssh ?
14:47 < JimBuntu> sudo service ssh restart
14:47 < qazey> exit; restart terminal.
14:48 < thebigj> https://superuser.com/questions/421959/wait-for-kernel-after-creating-partition-with-sftdisk-on-linux
14:49 < thebigj> Are there other better options than the correct answer given here?
14:49 < s10gopal> what is password?
14:49 < JimBuntu> s10gopal, I wouldn't know that. You can change the password from the VM terminal...
14:49 < thebigj> I am running sfdisk and then mkfs
14:49 < qazey> Password relates to login.
14:49 < s10gopal> how ?
14:49 < JimBuntu> `passwd`
14:49 < peetaur2> s10gopal: on the VPS, type "passwd" and hit enter, and it'll get you to set a password. If you don't know the old one, but have root access, run it as root like  "passwd usernamehere" and it won't ask for old.
14:50 < JimBuntu> s10gopal, have you had to enter a password once in a while when using sudo?
14:50 < thebigj> Most times it failed, because there request was unable to complete.
14:50 < peetaur2> s10gopal: and DO NOT set your username, hostname, ip, etc. as the password or it'll be hacked by a botnet in like 30 seconds, and don't set simple words, or yourusername1 or simple things added or it'll take like 5 min instead
14:50 < s10gopal> done :)
14:50 < peetaur2> and seriously, once you're in, just deploy the pubkey and only use that ,and disable password auth to forget about the whole problem of script kiddie attacks
14:50 < qazey> 10 second login.
14:50 < JimBuntu> What peetaur2 says is especially true since bots are watching IRC
14:52 < userus> lol script kiddies...best internet nickname ever
14:52 < qazey> lol script kid.
14:53 < s10gopal> now how to start building kernel on vm ?
14:53 < qazey> vi.
14:53 < peetaur2> s10gopal: screen, and then just like before...git clone, cd linux, etc.
14:53 < s10gopal> first i need to copy that .congif
14:55 < qazey> s10gopal, Have you ever heard of "Rollin' Tundah" source?
14:55 < s10gopal> no
14:55 < qazey> It's great.
14:55 < qazey> should check it on google.
14:55 < ananke> s10gopal: why do you keep wanting to build a kernel, when you know so little about the process?
14:56 < s10gopal> ananke, i want to get it fixed
14:56 < peetaur2> with enough effort, he'll be the leetest linuxer ever
14:56 < ananke> s10gopal: fix what exactly?
14:56 < s10gopal> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1745646
14:57 < JimBuntu> TL;DR -> He is going to do some bisects to find the change that caused his issue.
14:57 < s10gopal> qazey, tv series ?
14:58 < ananke> s10gopal: uhmm, why build a kernel from source, when the developers provided test packages?
14:58 < ananke> s10gopal: look at the most recent post
14:58 < s10gopal> ananke, i have to wait him to come online
14:59 < s10gopal> if i can build it myself i can test 4 5 kernel in one day
14:59 < s10gopal> how to copy .config ?
14:59 < peetaur2> s10gopal: doesn't this bug report say that Joseph Salisbury will build the kernels for you to test?
14:59 < ananke> s10gopal: did you even try the latest kernel that was provided?
14:59 < s10gopal> peetaur2, yes , but he only one a kernel in a day
14:59 < s10gopal> ananke, yes
14:59 < peetaur2> and how many kernels per day have you averaged? ;)
15:00 < s10gopal> 1.5
15:00 < s10gopal> i ihink with 8 cores i can build a kernel in 5 minutes
15:01 < ananke> s10gopal: frankly, I'm not convinced you have the necessary skillset to work on this on your own
15:01 < s10gopal> ananke, so i will learn them :)
15:01 < peetaur2> I agree, but I don't agree that discouraging him is the best way for him to gain those skills.
15:01 < ananke> peetaur2: I'm not discouraging him
15:02 < s10gopal> i need to run this cmd in ssh ? sudo cp /boot/config-4.12.0-041200-generic .config
15:03 < ananke> s10gopal: start with the basics: https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel
15:03 < s10gopal> ananke, peetaur2 told me basic
15:03 < peetaur2> the name Salisbury brings back highschool memories ... they wanted me to write about the salzbury incident, and I had no idea what that was, so I just titled my paper that and rambled nonsensically, and got marks on it anyway and realized they don't read your paper, so from that point forward I just filled in garbage and got marks... a good moment of progress for me
15:04 < ananke> s10gopal: except you keep asking about the basics. start by reading that wiki page
15:04 < s10gopal> sure i will
15:08 < s10gopal> peetaur2, how to copy file ?
15:08 < peetaur2> irc is not the place for the most basic questions...please look it up online first
15:09 < peetaur2> I suggested things before such as: rsync, scp, nc, and JimBuntu suggested copying and pasting in the browser
15:09 < s10gopal> peetaur2, scp /boot/config-4.12.0-041200-generic .config gopal:s10gopal right ?
15:10 < s10gopal>  he told about text files
15:10 < peetaur2> close ...    scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:path/to/linux/.config
15:11 < peetaur2> or grab it from the VPS's /boot if it has one
15:11 < peetaur2> (and config is a text file...it's just really huge)
15:11 < V7> What're the refresh bit of screen in "screen" and "tmux" ?
15:11 < peetaur2> V7: refresh bit?
15:11 < V7> Which is sent by Ctrl+L, you know
15:12 < s10gopal>  it is source then dest
15:12 < s10gopal> ?
15:12 < V7> redraw *
15:12 < peetaur2> s10gopal: yes basically everything is
15:12 < s10gopal> gopal@35.185.79.26 is dest ?
15:12 < peetaur2> V7: man says   C-l ('el') will redraw the screen.
15:13 < peetaur2> s10gopal: yes thats username + @ + host, and then also add : and then the path if you want (default would be same filename in your homedir if you don't set a path)
15:14 < peetaur2> s10gopal: do you know man pages? to look up commands, just type eg.:    man scp     and q to quit
15:14 < s10gopal> gopal@35.185.79.26 is my vm
15:14 < s10gopal> got it
15:14 < peetaur2> yes (so far it's yours....but soon a botnet's if you don't disable password auth)
15:15 < V7> So yes. I've heared that it's done automatically in ssh ( environment variable )
15:15 < peetaur2> while password auth is enabled, you can now easily use ssh-copy-id to deploy your pubkey
15:15 < peetaur2> and before you ask how to do that,  man ssh-copy-id
15:15 < V7> ssh refreshes screen when some refresh bit is received
15:15 < emberquill> ux
15:15 < V7> Isn't it ?
15:15 < fenrig_> Hi I have a tap device, which my program writes to but then the kernel responds with EIO
15:16 < fenrig_> I dont know much of tap devices, so I was hoping to ask for some pointers.
15:19 < Psi-Jack> tap is basically a virtual ethernet device. It's layer 2 of the OSI model.
15:20 < s10gopal> i am unable to enter password
15:20 < s10gopal> scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:path/to/linux/.config
15:20 < peetaur2> s10gopal: did you do that command literally, or did you replace path/to/linux/ with the path to the git repo dir?
15:21 < s10gopal> scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:/linux/.config
15:21 < peetaur2> what does it output? if you have password auth only, it should ask for a password...and if it uses a pubkey and you used ssh-agent and ssh-add, then it should not have to ask, but it would finish successfully
15:21 < s10gopal> scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:/linux/
15:21 < peetaur2> you put a / in there after the : ... that means absolute path in the root dir... take out the / to make it relative to your homedir
15:22 < peetaur2> it ought to say "Permisson denied"
15:22 < Psi-Jack> Hmmm
15:23 < Psi-Jack> What is the problem scenario there?
15:23 < peetaur2> Psi-Jack: he's linux beginner, and he's trying to git bisect the kernel
15:24 < peetaur2> his laptop takes 3.5h to compile, so he bought a google cloud VPS
15:24 < Dominian> Psi-Jack: 1.1.1.1
15:24 < Psi-Jack> Seems like he's way over his head.
15:24 < zenix_2k2> guys when i want to install my very own program, which directory is the most valid one for it ?
15:24 < Psi-Jack> Dominian: Heh, there he is!
15:24 < Dominian> :)
15:24 < zenix_2k2> some say usr/local, some say usr/local/bin and some say /opt
15:25 < peetaur2> zenix_2k2: if you want it system-wide and it respects the usual linuxy file layout...like libs in /lib, config in /etc/, etc. then put the config in /etc, and if no package, the rest in /usr/local, and if package then the usual places
15:25 < ananke> zenix_2k2: that depends
15:25 < peetaur2> zenix_2k2: if the paths are not linuxy at all, put it in /opt/
15:25 < Psi-Jack> Dominian: I changed up my site from Drupal, finally, to a CMS engine that's totally flat-file based, yet incredibly easy and fast as heck.
15:25 < peetaur2> zenix_2k2: and if it's not system-wide, stick it in ~/.local/ or wherever custom place in your homedir
15:25 < Psi-Jack> And in glorious Markdown. :)
15:25 < s10gopal> peetaur2, scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:/home/s10gopal/linux/.config whats wrong with it
15:25 < s10gopal> on vm , i was in my git directory
15:26 < s10gopal> and did pwd and got /home/s10gopal/linux/
15:26 < peetaur2> s10gopal: on vm? you run that command on your laptop to send the file remotely to the VM
15:26 < a-yogi> Hi, does anyone know what's the difference between "cat /proc/sys/fs/file-nr" and "ls /proc/$pid/fd"
15:26 < s10gopal> ion vm i did pwd
15:26 < Psi-Jack> a-yogi: One prints a file, one lists a directory.
15:26 < s10gopal> and i am using the avove command on ssh on my laptop
15:26 < zenix_2k2> peetaur2: sorry but i don't get it when you say "linuxy"
15:27 < a-yogi> what I am trying to find is the list of open files for the process
15:27 < ananke> a-yogi: they're unrelated to each other. try each, see what you get
15:27 < Dominian> Psi-Jack: hmm which one did you use?
15:27 < Dominian> Psi-Jack: ther ewas one I was looking at a long time ago but can't remember the damn name of it
15:27 < s10gopal> after using the command cursor starts blinking
15:27 < ananke> a-yogi: for that you'd use the second command
15:27 < peetaur2> a-yogi: lsof -Pn -p $pid
15:27 < s10gopal> cursor blinking mean it is copying files right ?
15:28 < jim> maybe "path not linuxy" means "path not FHS conformant"
15:28 < peetaur2> s10gopal: pastebin what you see and we can say what it means....the cursor likely tells nothing
15:28 < peetaur2> jim: :)
15:28 < s10gopal> s10gopal@instance-1:~$ scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:home/s10gopal/linux/.config
15:28 < s10gopal> then i see only cursor blinking no output
15:29 < a-yogi> ananke, then what's the siginificance of "cat /proc/sys/fs/file-nr"? becasue "cat /proc/sys/fs/file-nr" says 2784 and "ls /proc/$pid/fd | wc -l" says 678
15:29 < peetaur2> zenix_2k2: yes what jim said.... see man hier  for a full system explanation of the FHS (filesystem hierarchy)
15:29 < cnrhkiyf> s10gopal: you have to exit the ssh connection first
15:29 < a-yogi> which one is the exact number of open files, by the process
15:29 < zenix_2k2> peetaur2:no i mean "linuxy", like this --> "if the paths are not linuxy at all..."
15:29 < zenix_2k2> don't get that part
15:29 < s10gopal> ssh: connect to host 35.185.79.26 port 22: Connection timed out
15:29 < s10gopal> lost connection
15:29 < JimBuntu> that means super cool and utilitarian
15:30 < peetaur2> zenix_2k2: I mean like if you have a tomcat dir which has its own conf, temp, workdir, etc. then it's a mess and you'll waste loads of time reorganizing it, so just stick it in a dir so you hav epaths like /opt/apache-tomcat-versionnumber/temp
15:30 < ananke> a-yogi: /proc/sys/fs/file-nr is not PID specific. not sure why you keep comparing those two
15:31 < Psi-Jack> Dominian: Well, I'm using Grav.
15:31 < peetaur2> zenix_2k2: but if you have something that has its own /etc, /usr/lib, /usr/share, etc. all nicely organized, then you don't need to give up and stick it in /opt, but put it in the normal places. But if you didn't use system packages, you prepend /usr/local/ to all the paths except /etc/ (except on BSD where they still prepend /usr/local)
15:31 < Psi-Jack> Dominian: It's funny. Apparently the creator of Grav was one of the co-founders of Joomla.
15:31 < Dominian> Psi-Jack: ahh
15:32 < Dominian> Psi-Jack: there's another I used to use.. well test.. that had a logo like an octopus
15:32 < peetaur2> in the worst case with a non FHS compliant thing, you just symlink some parts so you don't need to modify $PATH... like  ln -s /opt/whatever/bin/somebinary /usr/local/bin/
15:32 < Dominian> trying to remember what it was.
15:32 < a-yogi> ananke, so what does "cat /proc/sys/fs/file-nr" really signifies? what does it mean?
15:32 < Psi-Jack> Dominian: PHP based?
15:32 < Dominian> flat-file I think
15:32 < Dominian> ]I don't remember
15:32 < Dominian> it looked slick though
15:32 < jim> and, in "not getting it", zenix_2k2 I think maybe is indicating he doesn't have enough experience to be able to tell when a path is not "linuxy" (aka doesn't fit in the standard)
15:33 < peetaur2> a-yogi: man procfs  will tell you about things in /proc/
15:33 < Psi-Jack> Dominian: Well, this is pretty slick, too. :)
15:33 < jim> and, the man page and your assistance now will help with that
15:33 < peetaur2> and if it doesn't know, then look at the text files in the kernel sources, or the source code itself
15:33 < Psi-Jack> Dominian: Grav's pretty much on the top shelf of the top 10 flat-file cms engines.
15:33 < zenix_2k2> man <arg> ???
15:33 < zenix_2k2> that can give me full help
15:34 < jim> zenix_2k2, they're saying the Filesystem Hierarchy Standard is described in man hier
15:34 < triceratux> Psi-Jack: the new grav site looks great. you might consider updating your recommendations on the distro page. it leaves the uninitiated with the impression that ubuntu is somehow acceptable ;)
15:34 < Psi-Jack> triceratux: LOL
15:35 < ananke> a-yogi: system wide
15:35 < jim> zenix_2k2, and, that may or may not give "full help", but at least will describe where in the linux filesystem some things go
15:35 < ananke> a-yogi: not sure why you're conflating the concept of 'here's what the entire system is doing' with 'here's what a single process is doing'
15:35 < Dominian> Psi-Jack: I'll find out and let you know.. I can see the sit ein my head.. I just cannot remember the name...
15:36 < Psi-Jack> hehe
15:36 < Psi-Jack> Dominian: Was it git-backed?
15:36 < Dominian> Psi-Jack: ha.. as I said it, I remembered: http://octopress.org/
15:36 < Psi-Jack> Ahhhh
15:36 < Dominian> but it appears it hasn't bene touched in forever
15:36 < Psi-Jack> Yeaaaah.
15:36 < peetaur2> a-yogi: as a rule, if it's in /proc/$somepid/ it's probably about a single process... if it's not, and not a symlink to such a place (such as /proc/self and I forget what else), then it is probably system-wide
15:36 < Dominian> 3.0 has been in the works for 3 years lol
15:36 < Psi-Jack> Dominian: Sheash.
15:37 < Psi-Jack> Dominian: You remember my site at all? ;)
15:37 < a-yogi> ananke, because recently a process was complaining about "Too many open files", but now when I see the "ls /proc/$pid/fd | wc -l" it's only 678 and the limit is 65k
15:37 < jim> (helpers: please note how I did that, bringing both sides of the conversation a little closer to understanding the other side)
15:37 < a-yogi> ananke, sorry the limit is "4096"
15:39 < ananke> a-yogi: finally, we're getting to the real issue. limits on individual processes can be set in various ways, your best bet is to look up documentation for your specific distro & release
15:40 < jim> zenix_2k2, did that help at least a little, and at least toward understanding about the existance of the FHS and how it might figure into your present situation?
15:40 < ananke> a-yogi: also, what the process has open now is likely very different from what it had when you got that error
15:41 < a-yogi> ananke, that is true
15:41 < zenix_2k2> jim: well i am still reading, cause it is truly a very long manual
15:42 < jim> yeah,I suppose that's true
15:44 < jim> anyway gotta go for about an hour and a half
15:45 < zenix_2k2> anyway i have another question, when i type "ls" in terminal, my shell is gonna look for the "ls" file in the $PATH right ?
15:45 < zenix_2k2> just to be sure
15:46 < peetaur2> zenix_2k2: yeah....similar to which you run:   which ls   it searches
15:47 < zenix_2k2> so what if... i create another "ls" file somewhere else in $PATH ?
15:47 < zenix_2k2> will the shell run mine or the system's one
15:47 < ananke> zenix_2k2: depends where that location falls in $PATH
15:47 < otirc> zenix_2k2: it all depends on the order of the $PATH
15:47 < ananke> zenix_2k2: first found wins
15:48 < zenix_2k2> HHHHmmmm... interesting
15:48 < peetaur2> zenix_2k2: it should run the first found in $PATH
15:48 < peetaur2> but in cases it won't ...not sure in modern days; but long ago, there was a command "rehash" you sometimes had to run to refresh it
15:49 < zenix_2k2> so here is my path --> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/usr/lib/jvm/java-8-oracle/bin:/usr/lib/jvm/java-8-oracle/db/bin:/usr/lib/jvm/java-8-oracle/jre/bin so what if i add another "ls" file in the "usr/local/sbin" ?
15:49 < zenix_2k2> $PATH*
15:49 < peetaur2> oh I think it's  `hash -r` in modern days, but I think you likely never need to run it
15:50 < zenix_2k2> and i think i shouldn't have used "another" cause usr/local/sbin is currently empty
15:50 < BCMM> is there a simple http server that can report on the bandwidth of clients?
15:50 < zenix_2k2> let's say another "ls" file in the $PATH which its directory is usr/local/sbin
15:50 < peetaur2> /snap/bin is such an ugly path
15:50 < s10gopal> peetaur2, i am getting Connection closed by 35.185.79.26
15:50 < s10gopal> lost connection
15:50 < BCMM> i'm trying to benchmark bandwidth over my wifi, between my linux computer and my android phone. would also be welcome to other suggestions for testing this.
15:51 < peetaur2> s10gopal: let's see the full command and output in a pastebin
15:51 < zenix_2k2> but will my "ls" run first if i add it in usr/local/sbin in my case ?
15:51 < peetaur2> maybe you were haxored by some script kiddies and they repurposed your VM for their own needs and accidentally broke ssh (they normally don't change things you notice...they want to keep using your system long before you notice)
15:51 < zenix_2k2> and i am not sure what is the original ls's path anyway
15:52 < section1> BCMM, i think ntop
15:52 < peetaur2> and I'm half kidding...don't jump to any conclusions until you look. Use the VM console and check again whether ssh is running; check the system logs, etc.
15:52 < s10gopal> peetaur2, sudo scp /boot/config-4.12.0-041200-generic gopal@35.185.79.26:/home/s10gopal/linux/.config
15:52 < section1> BCMM, iftop from console(real time monitor.)
15:52 < s10gopal> and i am unable to get output
15:53 < BCMM> section1: i tried nethogs, which is the same concept i think? but it sorta jumps around the whole time. something that measures overall bandwidth like an internet speed test would be ideal
15:53 < peetaur2> zenix_2k2: yeah if it's in a dir earlier in $PATH it will use that one instead of the usual /bin/ls (or likely /usr/bin/ls on systemdified systems)
15:54 < peetaur2> s10gopal: don't use sudo for scp
15:54 < BCMM> oh, there's iperf apps for android. derp.
15:54 < iflema> zenix_2k2: why not use the "usual" ls?
15:55 < zenix_2k2> iflema: well of cause i will, i am just wondering what would happen if i messed around like that
15:55 < iflema> zenix_2k2:
15:55 < iflema> k
15:55 < iflema> thought maybe you had a alias
15:55 < zenix_2k2> but i can't really save back alias
15:55 < s10gopal> peetaur2, still getting same problem
15:56 < zenix_2k2> it will be deleted once i reboot
15:56 < iflema> huh
15:57 < s10gopal> peetaur2, ssh: connect to host 35.185.79.26 port 22: Connection timed out
15:57 < s10gopal> lost connection
15:57 < s10gopal> You are trying to send an empty document, exiting.
15:59 < s10gopal> peetaur2, there is option in google cloud to upload and download file and i selected that file and uploaded it
15:59 < s10gopal> peetaur2, can you plz help me verify it is at correct position ?
16:00 < peetaur2> s10gopal: put it in the git repo dir, named .config
16:01 < peetaur2> keep another copy somewhere else though....because make menuconfig, etc. will modify it, and each time it might be different if you do things in a different order
16:01 < s10gopal> peetaur2, my vm password keeps getting changing
16:01 < peetaur2> so put one in your homedir with the original name, and then the other named .config
16:01 < s10gopal> peetaur2, i got connection lost and the old password dont work
16:05 < s10gopal> peetaur2, how to verify i did it correct or not ?
16:06 < peetaur2> don't know...gather information to find out what the system is doing.
16:06 < s10gopal> peetaur2, i did :~/linux$ mv config-4.12.0-041200-generic ./config
16:07 < peetaur2> if you think it's hacked (no idea how likely... my estimation is 100% chance if your password sucked like "p@ssword1") then nuke the VM and start over
16:07 < s10gopal> i moved the file correctly ?
16:07 < peetaur2> your password should not be changing unless you are doing it
16:07 < s10gopal> this time i will choose very strong password
16:09 < peetaur2> s10gopal: here's your new password:    dd if=/dev/urandom count=1 2>/dev/null | base64 | tr -d '\n' | grep -Eo "^.{8}"
16:10 < s10gopal> thx
16:10 < s10gopal> peetaur2, i i moved file correctly?
16:10 < peetaur2> or if you prefer, here's your new passphrase:   grep -E "^[a-z]{5,8}$" /usr/share/dict/cracklib-small | sort -R | head -n4 | tr '\n' ' '
16:11 < peetaur2> (location of that words/dict/cracklib file varies per distro)
16:11 < peetaur2> on an Ubuntu 16.04 I see the path is  /usr/share/dict/american-english
16:11 < peetaur2> or I guess  /usr/share/dict/words
16:12 < peetaur2> s10gopal: I have no idea if you moved the file...yo uhave to show some output, like ls -l or something I can use to answer that
16:14 < s10gopal> http://paste.ubuntu.com/p/zdqvNSJJ6Z/
16:14 < peetaur2> this looks like ls -l in the git repo... in that case the file is named .config and hidden so you do ls -la
16:14 < peetaur2> but also let's make sure we can see a config-4.12.0-041200-generic too, in another dir like ~/config-4.12.0-041200-generic
16:15 < s10gopal> http://paste.ubuntu.com/p/zdqvNSJJ6Z/
16:15 < peetaur2> that is the same link again
16:15 < s10gopal> http://paste.ubuntu.com/p/TmpPp3tk3j/
16:17 < peetaur2> so now you can see a .config, but not sure how to verify it's the same one... you could use sha256sum .config    and then compare to   sha256 /boot/config-4.12.0-041200-generic
16:19 < iflema> diff -sq
16:19 < iflema> :D
16:19 < peetaur2> yeah but he might not have both copies there...I don't see the config-... one anywhere
16:23 < AbleBacon> what do video devices look like in "dev"? i'm plugging/unplugging a display panel controller board, and "dmesg" is just saying that a USB device was plugged/unplugged
16:24 < AbleBacon> not sure how to get the dern thing to show up in /dev/
16:24 < s10gopal> what does this mean ? http://paste.ubuntu.com/p/rCzHGSjXph/
16:24 < peetaur2> AbleBacon: find /dev/ -type f | sort > before.txt   ....   find /dev/ -type f | sort > after.txt   ... diff -u before.txt after.txt
16:24 < s10gopal> Cannot use CONFIG_CC_STACKPROTECTOR_STRONG: -fstack-protector-strong not supported by compiler
16:24 < peetaur2> and the other answer I forget but it's probably simply like /dev/video1
16:24 < AbleBacon> ooooooooh good idea
16:25 < section1> AbleBacon, try ls -rtl /dev/   .. to see the last created file/node/etc
16:26 < peetaur2> oh and...er. type f is wrong... I guess  -not -type f -a -not -type d
16:26 < s10gopal> peetaur2, https://paste.ubuntu.com/p/mqmJB2Z72j/ please see it
16:27 < AbleBacon> wait, what's the full command then, peetaur?
16:28 < AbleBacon> it's not liking the "-not"s
16:29 < peetaur2> AbleBacon: find /dev -not -type f -a -not -type d
16:29 < peetaur2> and you can also try section1's suggestion...based on timestamp rather than diff
16:30 < AbleBacon> my primitive version of "find" doesn't like "-not". i'll just do "find" without options and see what all changes
16:30 < section1> yeah but thinking now if for example it put under /dev/video  or other sub-dir..fails
16:31 < AbleBacon> ugh. nothing about video in the diff. for the life of me i cannot get the system to detect this display device
16:31 < AbleBacon> i'm going to lose my mind
16:33 < Psi-Jack> AbleBacon: Too late. It's already mine.
16:33 < AbleBacon> O_O
16:34 < peetaur2> display device? I was assuming something like a camera
16:34 < peetaur2> I kinda don't think you'll see monitors in /dev/
16:34 < peetaur2> and if you add pci devices (like thunderbolt), I would expect them in /sys but not /dev but don't have experience with that
16:35 < AbleBacon> Bus 002 Device 005: ID 0eef:c000 D-WAV Scientific Co., Ltd
16:35 < AbleBacon> this is what it is
16:35 < AbleBacon> it's a screen and backlight with a controller board
16:36 < AbleBacon> i mean, presumably i don't have the drivers for it i'd guess
16:36 < ayecee> damn, thought that was d-wave for a bit
16:36 < ayecee> which could potentially be more interesting
16:36 < AbleBacon> no quantum monitor here :(
16:36 < ayecee> (they're in quantum computing field)
16:39 < poptix> AbleBacon: those are capacitive touchscreens
16:40 < section1> AbleBacon, http://karuppuswamy.com/wordpress/2013/01/17/how-to-get-evtouch-touch-screen-kit-working-in-linux/  its a similar touchscreen
16:41 < poptix> AbleBacon: look at the EETI touchscreen support, eGalax, etc
16:41 < poptix> ah, yes, section1's link should do it.
16:42 < AbleBacon> yeah, just figured that out--it's not the video it's the touchscreen
16:42 < AbleBacon> oh god no please don't make me compile anything
16:42 < AbleBacon> i've compiled and compiled and compiled
16:44 < AbleBacon> well, for all i know the touchscreen could be working. but apparently those wires that i thought went to the display went to the touchscreen
16:44 < qazey> gauging i's.
16:45 < qazey> take it easy for the lenses are britannica.
16:48 < AbleBacon> i've compiled until the cows turned blue
16:48 < AbleBacon> it was a big rush when i finally got a command line working though
16:51 < AbleBacon> if i could get the backlight working that'd be great... but /sys/class/backlight is empty
16:54 < triceratux> https://stratechery.com/2018/the-end-of-windows/
16:58 < fendur> poptix is a familiar nick
16:58 < DrDonkey> Hello people.
16:58 < paddy|> triceratux: it feels like the author wants to say 3 things at the same time
16:58 < paddy|> very hard to read for me
17:00 < DrDonkey> What terminal/Font you guys use?
17:01 < Psi-Jack> The one I like.
17:02 < DrDonkey> And that is?
17:02 < DrDonkey> I am using termite with Fantasque Sans Mono.
17:02 < kekePower> hi fellow penguins
17:02 < DrDonkey> Hello.
17:02 < kekePower> first time here :-)
17:04 < paddy|> :)
17:05 < kekePower> so, what's a hot topic these days?
17:05 < paddy|> kekePower: ##chat, ##, ##social etc.
17:05 < kekePower> a hot linux topic
17:06 < dgurney> dunno, is there supposed to be one that's above every other linux-related thing?
17:06 < kekePower> don't know. Could be a new program, new development, something exciting
17:07 < kekePower> paddy|: "To join us at ##Linux, please see How to Connect. You will need an IRC client such as irssi or XChat and a sense of humor 😉 ."
17:07 < paddy|> this chat is not about gathering excitings. its for support
17:07 < kekePower> paddy|: OK
17:07 < kekePower> so anything off-topic is off limits?
17:08 < paddy|> thats about right, yeah
17:08 < kekePower> OK
17:08 < paddy|> and mind me saying that bluntly: be a bit less pushing. real newbies are too afraid for this attitude
17:09 < kekePower> paddy|: Thanks. I've been in the game for so long I've forgotten how it is to be new
17:09 < paddy|> i misunderstood your intro comment then
17:09 < triceratux> paddy|: yep the author has been running windows far too long. needs to boot an xubuntu liveiso
17:10 < kekePower> paddy|: I think you may have. I just wanted to understand the type of discussions that goes on here
17:10 < paddy|> kekePower: you usually get them by observing not by asking and pushing
17:11 < kekePower> paddy|: Try teaching an old horse to sit
17:11 < kekePower> :)
17:12 < paddy|> i just pass along what was told to me before
17:12 < kekePower> paddy|: and that's how we learn
17:12 < paddy|> now you got your offtopic chat. arent we all nice
17:13 < peetaur2> triceratux: but when windows dies, what will be the next thing? mac is even worse... and you know that freedom like in Linux doesn't take over as fast as lock in aggressively marketed stuff, so it won't be simply Linux
17:14 < kekePower> paddy|: I think it may be called an introductory chat
17:14 < peetaur2> and if windows is gone, will we still get fun videos to watch like this?  https://www.youtube.com/watch?v=UOqBBggxLTo
17:15 < paddy|> kekePower: the stage is yours *clap, clap, clap*
17:16 < bipul> Is that true that mysql root password is reside inside configuration file?
17:17 < kekePower> bipul: Don't think so. I think you can add it to a config file, but it's not recommended or safe
17:17 < kekePower> bipul: When you set the root password, it gets set in the sql database
17:17 < kekePower> if my memory serves me correct
17:17 < triceratux> peetaur2: its simpler than all that. 30yrs ago an OS was a planet altering achievement. these days its just a detail you have to include to get your browsers & the rest of the apps up, theres going to be a bunch of them & theyll come from anywhere. just look at android
17:17 < moniker-> lol https://www.bleachbit.org/cloth-or-something
17:21 < peetaur2> bipul: mysql root password is hashed and is in the database
17:21 < peetaur2> SELECT * FROM mysql.user;
17:22 < kekePower> thanks peetaur2. It's been a while since I fiddled with mysql.
17:22 < peetaur2> lucky you
17:22 < bipul> peetaur2, Yes, but i don't have access to mysql.user
17:23 < peetaur2> bipul: if you have root on the box, just stop mysql, and run it with auth disabled (and network too if you don't like haxors connecting), and then you can log in as root without a password   /usr/bin/mysqld_safe --skip-grant-tables --skip-networking
17:23 < bipul> kekePower, Thank you.
17:23 < bipul> peetaur2, Thank you too.
17:26 < nobrain> I have seen systems with MySQL root password stored in my.cnf file
17:26 < nobrain> even in plaintext
17:26 < nobrain> don't even ask me why people do that
17:27 < peetaur2> they're testing what they can get away with before they are fired
17:27 < LordRyan> Is there a way to get touch events on a touchscreen if I'm not running X?
17:29 < nobrain> I mean google a bit about that and you will see it is/was a common practice
17:30 < nobrain> as stupid as it sounds
17:31 < kekePower> LordRyan: I don't see how that can be done.
17:31 < Psi-Jack> LordRyan: better question is... Why? Lol
17:31 < Psi-Jack> Why?
17:32 < kekePower> Psi-Jack: I guess he has his reasons and I think it's a good question
17:32 < kekePower> I have a touch screen, but have never thought about using it without X
17:32 < LordRyan> Psi-Jack: the product writes on a layer lower than X, to get more processing power out of it
17:33  * Psi-Jack tilts his head.
17:33 < kekePower> as far as see, there are only two input methods on the console, Keyboard and Mouse
17:33 < kekePower> and for the mouse there is gpm
17:33 < Psi-Jack> X... Is not really a "layer". So  that satement makes no sense.
17:34 < kekePower> nitpicking
17:35 < Psi-Jack> Being accurate in a situation is not nitpicking. it's being thorough, and accurate.
17:35 < kekePower> but as far as we are in the know, we understand what's being said
17:35 < paddy|> it seems like bipul lost interest in that subject
17:35 < Psi-Jack> Actually no. No "we" do not. because what was said makes literally no sense. You can't possibly know what he actually said.
17:36 < xse> LordRyan: gpm supports touchscreens
17:36 < kekePower> everything can be considered a layer. Hardware to Kernel to User space to UI to Application
17:36 < paddy|> bipul: sorry, wrong tab
17:36 < Psi-Jack> xse: Yeah, I was wondering about that. Does it really?
17:37 < bipul> paddy|, ? i am sorry?
17:37 < LordRyan> xse: cool :D
17:37 < bipul> I'm interested into database.
17:37 < paddy|> bipul: false alert, wrong tab, my fault - sorry
17:38 < xse> Psi-Jack: i havn't tested it but "man 7 gpm-type" says so
17:38 < bipul> It's ok.
17:40 < zumba_addict> Morning folks. I'm used to seeing a populated /etc/resolv.conf. I'm analyzing a blackbox right now. It's redhat. It's running bind/named as shown by ps and netstat. However, /etc/resolv.conf is empty. Does the network stack directly forward the dns query to the locally running dns daemon(BIND) if no resolv.conf is found?
17:42 < junka> Are Arch users really the vegans of the Linux world ?
17:42 < zumba_addict> hmm, I do see a file /etc/named.caching-nameserver.conf interesting
17:44 < LordRyan> xse: thx but unfortunately it only works on the Linux terminal
17:44 < LordRyan> oh wait no i'm a liar
17:46 < xse> LordRyan: i mean where did you wanted to make it work ? wayland/xorg should support touchscreens using libinput otherwise
17:47 < LordRyan> xse: it works if I don't have X open in the background
17:47 < a-yogi> ananke, peetaur2, what's the difference between "lsof -Pn -p $pid" and "ls /proc/$pid/fd/"
17:47 < LordRyan> but for dumb reasons, sometimes we do
17:49 < xse> LordRyan: yeah gpm does work without X, it might even be able to relay events to X with tweaking
17:50 < xse> but X you shouldn't need gpm for it to be working under X
17:50 < LordRyan> well yea it works without x
17:50 < LordRyan> but sometimes I have X open, sometimes I don't
17:50 < LordRyan> i know, it's super weird
17:50 < LordRyan> will do some more testing
17:51 < tyzoid> Any possible reason I'm seeing 950mb available memory on a system with 7979mb ram with no tmpfs usage, and total process rss size is 3252mb?
17:52 < tyzoid> The system is running three lxc containers, so I suspect that has something to do with it
17:52 < kpel> is anybody here using fou tunnels through nat?
17:53 < tyzoid> but I'm not sure how to measure the memory overhead of that
17:53 < tyzoid> I'm trying to figure out what's causing the difference
17:55 < NDx33xsy> Hello. Which best terminal emulator?
17:55 < LordRyan> okay so X doesn't support the touchscreen, gpm does but only as long as X is running
17:55 < LordRyan> NDx33xsy: I've had the least amount of problems with Termite
17:55 < LordRyan> However, on some systems, Termite's a bit hard to install
17:55 < ngc3982> hi. i installed a new server, and im adding my older hard drives. due to me being a total cluts i installed grub on one of the drives. when i add the drive it tries to start from grub on that drive. how can i hinder that?
17:56 < LordRyan> wipe the partition, change the order of drives in BIOS
17:56 < ngc3982> wipe, as in format?
17:56 < ngc3982> it contains important data :/
17:57 < ngc3982> it is not a system drive in the first place.
17:57 < xse> LordRyan: strange, you should be able to start the gpm service and have it working without X
17:59 < LordRyan> xse: right
17:59  * ngc3982 looks into bios.
17:59 < LordRyan> but I still need X running
17:59 < LordRyan> I just don't want the events to go through X
17:59 < xse> yeah :/ that's weird
17:59 < LordRyan> GPM works as long as X isn't running but if X *is* running it breaks.
17:59 < NDx33xsy> LordRyan: Thanks. That's what I was looking for)
18:00 < LordRyan> And xev won't print out touch events - which is weird because I *swear* Cookie Clicker worked.
18:00 < qazey> Cookie clicker?
18:00 < LordRyan> qazey: yea it's one of the oldest idle clickers
18:01 < LordRyan> gonna see if I can get it working again
18:02 < absurdistani> has anyone had trouble with NVMe in Linux kernel 4.14 and later? for me, the same intel nvme ssd is visible without issue in 4.9 but isn't visible at all in 4.14 according to lsblk and /dev
18:04 < solidfox> absurdistani, isnt 4.9 > 4.14
18:04 < solidfox> unless you meant 4.09?
18:04 < xz> or 4.014?
18:04 < absurdistani> solidfox: 4.09 vs 4.14 sure
18:05 < solidfox> sorry I know that's not the way version numbers work lol
18:05 < absurdistani> i've also tried 4.15 and 4.16 (4.16 being latest) and it isn't working
18:05 < qazey> solidfox, Got a sister?
18:05 < solidfox> qazey, yeah I've got 3
18:05 < kekePower> absurdistani: have you mentioned this on the kernel dev mailing list?
18:06 < absurdistani> kekePower: no. im not a kernel dev.
18:06 < qazey> Toto says hi.
18:06 < solidfox> ?
18:06 < kekePower> absurdistani: It's a good place to start since I don't think any kernel devs are here
18:07 < kekePower> absurdistani: Did you compile your own kernel or use your distro kernel?
18:07 < solidfox> qazey, who are you, and how do you know my middle name?
18:07 < qazey> Toto told me.
18:07 < solidfox> I have no idea who that is!
18:07 < qazey> He said it was a blast with #2.
18:07 < absurdistani> kekePower: compiled my own. the thing is, I am hoping that someone else has had this issue and could point me to a config option or other issue that could be causing this
18:07 < solidfox> O_O
18:08 < kekePower> absurdistani: It's tricky business, but also really fun and rewarding
18:08 < absurdistani> even weirder is that the samsung 960 pro nvme is just fine, only the intel ssd is not showing up
18:08 < kekePower> absurdistani: just look through the options in the menues and I think you'll find the missing module
18:10 < solidfox> qazey, i dont git it
18:10 < absurdistani> kekePower: yeah. done that. im still missing something. the only reason this is really an issue is because i have a threadripper machine with vega graphics, so i need a more recent kernel to take advantage of the hardware... but in doing so lost the ability to use one of my disks...
18:10 < solidfox> qazey, what do you WANT FROM ME
18:11 < absurdistani> like wtf. this reminds me of early wireless issue with linux where the kernel just dropped support for half of the wireless hardware in existence.
18:13 < LordRyan> xse: haha. i'm dumb. :)
18:13 < LordRyan> xev needed -root
18:13 < Psi-Jack> solidfox: Your doom.
18:13 < xse> nice :)
18:14 < LordRyan> so I can use X's event handling system
18:15 < LordRyan> and I know X will be up because that's how we display the streams :D
18:18 < qazey> Doom Nukem.
18:19 < qazey> Ey doo nukem.
18:20 < epder> i've setup 2 nameservers in my interfaces-file. but when i restart networking it only adds one to resolv.conf. Ive put 127.0.0.1 first, and this is the only one thats being added to resolv.conf. This makes it hard to resolve any other hosts than the local ones. Anyone knows how to fix this?
18:21 < spammcoin> i dunno, my resolv.conf is a simple text file that is saved across reboots
18:22 < absurdistani> so, some more information about this issue, the nvme ssd isn't present on kernels ive compiled but its also not present in arch or suse. it is present in slackware 14.2 default kernel and updated kernels through slackpkg. so it's something that changed with a kernel later than 4.10 (ive a custom 4.10 kernel that is also working without error with this disk)
18:23 < absurdistani> i did try the boot parameter nvme_core.default_ps_max_latency_us=5500, and it solved nothing
18:23 < xz> absurdistani what is is that you are trying to achieve?
18:23 < absurdistani> xz: i just want the nvme disk to be usable with a newer kernel
18:23 < xz> absurdistani whatever distro?
18:23 < absurdistani> xz: the disk works without issue in 4.9 and 4.10, but it isn't usable with 4.14 and higher
18:24 < absurdistani> xz: specifically, I need a newer kernel because I am using ryzen and vega which get support in newer kernels
18:24 < absurdistani> but with a newer kernel the disk doesn't work
18:25 < nobrain> AMD fan uh, thought they all died
18:25 < xse> epder: dunno how did you set them up, here i'm using netctl for my network profile and it's working like a charm : https://i.imgur.com/uTm7wIj.png
18:25 < xz> isn't debian 9 currently on 4.11 or something like that?
18:26 < absurdistani> xz: sure, but 4.11 won't have the other hardware support I am looking for :/
18:26 < xz> so you need at least 4.14 then?
18:27 < epder> this is my setup of the interfaces-file: https://pastebin.com/KZWaUyPV
18:28 < absurdistani> xz: yeah
18:28 < rendar> i have like 500Mb of logs in /var/log, how i can get the process writing those?
18:29 < xz> absurdistani would you know PCI device id (lspci) of that NVMe card?
18:29 < spammcoin> what do the logs say, rendar
18:29 < xse> epder: you can try to modify your resolv.conf manually, and protect it against write with "chattr +i"
18:31 < absurdistani> xz: none of the nvme disks show up in lspci
18:31 < absurdistani> just the controllers
18:32 < absurdistani> but they do show up in lsblk
18:32 < xz> absurdistani and you see the controler on both, 4.9 and 4.14?
18:32 < absurdistani> xz: that is correct
18:32 < rendar> spammcoin: wait, i will paste it, basically it's like it logs all ssh packets, and i don't know why
18:32 < xz> what's the PCI ID of the controler then?
18:32 < absurdistani> xz: and the other nvme disk does work without issue
18:33 < absurdistani> 41:00.0
18:33 < spammcoin> it's one log file then?  what is the name of the log file?
18:33 < xz> that's Bus/Device/Function I think
18:34 < absurdistani> that's the intel that doesn't work with newer kernels and samsung is on 42:00.0
18:34 < xz> absurdistani try that: https://wiki.debian.org/HowToIdentifyADevice/PCI
18:34 < absurdistani> ah
18:34 < absurdistani> okay
18:34 < absurdistani> other number
18:34 < xz> the Intel card should have PCI id something like 8086:....
18:36 < absurdistani> I am not seeing any number anything like that anywhere... I remember seeing stuff like that setting up vfio once upon a time, but...
18:36 < absurdistani> lscpi doesn't show anything of the sort
18:36 < absurdistani> lspci*
18:37 < xz> absurdistani you might have that in dmesg as well when device was enumerated
18:37 < xz> you could try something like dmesg | grep NVMe
18:37 < rendar> spammcoin: here we go https://pastebin.com/kp7ij9dH
18:37 < rendar> i have dozens of 100Mb of those logs: https://pastebin.com/kp7ij9dH
18:37 < rendar> what are they?
18:37 < spammcoin> i'm not digging through your logs
18:37 < spammcoin> whats the filename
18:37 < rendar> digging?! it's a line of text!
18:37 < xz> absurdistani in order to be able to track what happened with that device between the 2 kernels, you need to identify PCI id somehow
18:37 < spammcoin> i'm not clicking on your pastebin link
18:37 < spammcoin> paste the line here
18:38 < rendar> spammcoin: ok, it's a long line: Apr  6 18:35:23 enyo kernel: [930984.880626] IN=he-ipv6 OUT=eth0 MAC=b8:27:eb:32:3a:c7:f0:b4:29:f5:f2:92:08:00:45:00:01:9f:db:d7:40:00:f8:29:95:b6:d8:42:57:0e:c0:a8:1f:ae:68:c0:00:00:01:63:06:34:2a:03:28:80 TUNNEL=216.66.87.14->192.168.31.174 SRC=2a03:2880:f008:0001:face:b00c:0000:0001 DST=2001:0470:22ae:babe:4519:e6ec:5f40:5afe LEN=395 TC=140 HOPLIMIT=52
18:38 < rendar> FLOWLBL=0 PROTO=TCP SPT=443 DPT=56059 WINDOW=1576 RES=0x00 ACK PSH URGP=0
18:38 < absurdistani> xz: ah okay, and yeah syslog has it
18:39 < absurdistani> xz: okay. well, i will tackle it later. thx. gives me an okay place to start
18:39 < spammcoin> what's the filename
18:39 < absurdistani> in the end, I will probably just stop using this disk.
18:39 < absurdistani> nothing but headache
18:39 < rendar> spammcoin: both kern.log and messages
18:40 < spammcoin> is yoru iptables set up to log packets?
18:40 < rendar> spammcoin: how i can see that?
18:43 < spammcoin> iptables -L  && ip6tables -L
18:43 < rendar> it's empty
18:43 < rendar> because i tried to delete everything from iptables
18:44 < spammcoin> this mesage is from before you did that?
18:44 < spammcoin> does lsns only show 1 net ?
18:44 < rendar> before and after
18:45 < rendar> spammcoin: https://pastebin.com/3XRqEd77
19:04 < oatie> My server is accessible via http, but when I try to ping websites from the shell, it just hangs. When I try to install things with pip, I get this error:
19:04 < oatie>   Retrying (Retry(total=0, connect=None, read=None, redirect=None)) after connection broken by 'NewConnectionError('<pip._vendor.requests.packages.urllib3.connection.VerifiedHTTPSConnection object at 0x7fd88c7e6390>: Failed to establish a new connection: [Errno 101] Network is unreachable',)': /simple/django/
19:05 < spammcoin> is that the one you get when you didn't set a default route?
19:05 < oatie> huh, what do you mean a default route?
19:05 < spammcoin> type ip route
19:06 < meyou> is there an IP that can be guaranteed to never be connectable over the internet
19:07 < meyou> err, locally either
19:07 < mallu> Which file in cups has printer driver info? I checked /etc/cups/printers.conf and there is no info as which driver a printer should use
19:07 < bindi> meyou: 0.0.0.0
19:07 < hexnewbie> meyou: Are you after a private network, or after an example IP?
19:07 < oatie> spammcoin: okay I typed ip route, now what
19:07 < spammcoin> is there a default route?
19:07 < oatie> " default via 172.31.80.1 dev eth0"
19:08 < spammcoin> damn!
19:08 < oatie> what happened
19:08 < spammcoin> your network is allegedly unreachable
19:09 < oatie> but what does that default mean?
19:09 < oatie> what is that AP address?
19:09 < oatie> IP*
19:09 < meyou> i want to make an A record that i know will never ever ever connect to anything
19:09 < spammcoin> i assume your router or something that is acting as a router
19:09 < meyou> even if they're running a webserver on localhost or use that ip scheme locally
19:09 < meyou> etc
19:10 < hexnewbie> mallu: The ‘driver’ is a PPD in the PPD directory. For PostScript/PDF printers, that's just a description of the printer's parameters (PPD = PostScript Description). For non-PS printers, it contains a cupsFilter attribute containing the name of the CUPS driver (filter program which converts a PostScript/PDF file to the printer data)
19:10 < hexnewbie> Er, PostScript Printer Description
19:11 < hexnewbie> mallu: And "ppd" directory, lower case. Sorry. :)
19:11 < mallu> hexnewbie: we want to copy printers from one server to another. What all do we need to copy from source server to dest server?
19:11 < hexnewbie> mallu: Copy what?
19:11 < gronke> Anyone here use Galaxy? (i.e. galaxyproject.org) I'm having trouble with a dependency resolver that it seems should be working. (I recall someone saying here yesterday that they used it but I forgot who)
19:12 < hexnewbie> mallu: THe .ppd file needs to be copied, but also the filter (if used) needs to be installed, so you need to verify it exists in /usr/lib/cups/filter
19:12 < mallu> hexnewbie: we are retiring a server. We want to copy all the servers configured in the old server to new server
19:13 < hexnewbie> mallu: And the entry in printers.conf, too
19:14 < mallu> hexnewbie: ok so /etc/cups/printers.conf, /usr/lib/cups/filter and /usr/share/cups/ppd? Using Redhat
19:15 < gronke> It is supposed to use Conda as a dependency resolution by default in the settings, but when I try to install a repo that needs something from Conda, it says "Please activate the Conda dependency resolution." And the thing I need IS on Conda.
19:15 < hexnewbie> mallu: No, /etc/cups/printers.conf and /etc/cups/ppd/ need to be copied. THe /usr/lib/cups/filter need to be installed from the distro (unless you installed non-standard drivers). RedHat may use different directory for CUPS filters, too
19:16 < mallu> ok thank you hexnewbie
19:20 < leshaste> is it possible to save the state of vncserver so I can kill it and resume it later?
19:24 < lopid> what's the best independent tool that will detect a mouse button click and let me configure what happens next, eg turn it into a double click, send keystrokes instead, etc?
19:25 < lopid> i can use xdotool for the latter, but i need something else to detect the trigger
19:25 < spammcoin> lopid: i wonder if theres a way to emulate a /dev/input device
19:27 < spammcoin> something device node level would be ideal because you could route other inputs to it and change events before any input libraries can do what they will
19:27 < lopid> maybe xinput?
19:32 < nobrain> I hate cups with passion, most ugly unfriendly shit I ever found in *nix world
19:39 < Dagmar> That feeling is normal
19:41 < nszceta> /usr/bin/ld: cannot find -lzlib
19:41 < nszceta> zlib1g-dev is already the newest version (1:1.2.8.dfsg-2ubuntu4.1).
19:41 < nszceta> what do
19:42 < lopid> something is not looking in the right place. is your /etc/ld.so.conf correct?
19:42 < fendur> whatever is looking for it may need help identifying zlibs location.
19:42 < prussian> maybe you need zlib-static or whatever
19:44 < nszceta> thanks guys
19:44 < nszceta> figured it out
19:50 < xse> Anyone familiar with seccomp here ? i don't get why there is a feature "When adding a new filter, synchronize all other threads of the calling process to the same seccomp filter tree.", does it mean that i can have different seccomp filters for multiple threads launched by the same program ?
19:55 < LordRyan> xev gives off a "PropertyEvent" when I touch my touchscreen or push buttons on a mouse. I'd expect some other kind of event, is this normal?
19:55 < LordRyan> sorry, PropertyNotify event*
19:55 < DracoDan> does anyone know of a channel for linux-iscsi/LIO/target-devel or whatever you want to call it?
19:56 < LordRyan> I'd test on an actual Linux system but my laptop's at home and work's putting me on this awful Mac. >:I
19:57 < spammcoin> xse: process == thread, every process has it's own filter
19:59 < xse> spammcoin: thanks ! i'm gonna try it out to get it better :)
19:59 < gronke> Does anyone here use Galaxy (galaxyproject.org) I have a simple question if you do, thanks
20:02 < Psi-Jack> gronke: It
20:02 < gronke> Psi-Jack, Que?
20:02 < Psi-Jack> gronke: It's very unlikely if anyone here would.
20:03 < DrunkRhino> Is there any easy way using dnsmasq to associate an ipv6 address with it's corresponding ipv4 host? I'm running a pi-hole with the DHCP server enabled for v4 and the SLAAC+RA option turned on for v6, but I can only resolve the hosts v4 addresses
20:03 < gronke> Psi-Jack, Someone yesterday had used it, I'm just having this inexplicable issue
20:03 < Psi-Jack> Perhaps try their channel, forums, or more direct approach.
20:03 < Psi-Jack> gronke: Someone here tried it, and ran away from it. :)
20:03 < gronke> That's what they had said
20:04 < Psi-Jack> It's a very ... Specialized piece of software.
20:07 < hexnewbie> If that's the Galaxy I think it is, we just ended up using Makefiles and bash scripts after some folks recommended us against it and told that all the tech guys hated it.
20:07 < rajrajraj> suggest me a phone under 150usd please :D
20:07 < revel> rajrajraj: Check ##android
20:07 < rajrajraj> ok
20:08 < gronke> hexnewbie, bioinformatics software?
20:08 < hexnewbie> gronke: Yeah. But then, I wasn't the one with the more complex pipelines, so I don't know much about the integration part.
20:09 < gronke> hexnewbie, it's just odd. I'm installing this software from their tool shed that requires dependencies. Galaxy is supposed to have Conda resolve dependencies by default, but it's not doing it.
20:09 < gronke> hexnewbie, and I'm supposed to have a checkbox that says: "When available, install externally managed dependencies (e.g. conda)? Beta" but that is nowhere to be found.
20:09 < hexnewbie> The colleagues who didn't more sophisticated processing did use makefiles to define the processing steps and reproduce, though (I wasn't too in on it, and used bash scripts for mine)
20:10 < gronke> yeah we have grad students and we dont want to give them all linux access, students as well in courses
20:10 < hexnewbie> s/didn't/did the/
20:12 < hexnewbie> gronke: Ah, then disregard what I said. It's  good for teaching (probably), and trying to tell the students with biology background to run the tools from the command line proved too difficult here.
20:12 < gronke> yeah that's why we need a tool like it, we were using Mobyle but that has been abandoned for like 4 years and I finally convinced the boss man to move on to a more relevant peice of software
20:12 < gronke> and often i'm the one tasked with writing the new tools which is why i'll be hanging out in #python asking a lot of q's as well
20:12 < hexnewbie> gronke: Without being familiar with Galaxy in particular, why don't you just install all the tools (most do come with Ubuntu, etc) and skip the option to let it install things?
20:13 < gronke> well for example I'm trying to install a Trimmomatic tool, which requires the software Trimmomatic, which is not installed, but it supposed to be resolved by Conda, and it's ON Conda, yet it's not recognizing that it can use conda to resolve dependencies
20:14 < gronke> and it says "Please activate Conda dependency resolution", but according to the docs *That is enabled by default*
20:15 < gronke> so it's a bit maddening
20:15 < gronke> in the docs: "The short answer is that as of 17.01, Galaxy should install Conda the first time it starts up and be configured to use it by default."
20:16 < hexnewbie> trimmomatic does come with Debian, so I assume it also comes with Ubuntu?
20:16 < hexnewbie> Er, I didn't ask for distro first. Sorry for assuming Ubuntu.
20:17 < gronke> CentOS 7
20:22 < s10gopal> most of the linux users report that there battery draines when laptop is off , but still linux dev are not doing anything yet , it cant be fixed ?
20:23 < Thedarkb> I want to pipe the output of a program to a file as well as the screen.
20:23 < hexnewbie> s10gopal: Most Linux users can't agree if they like their init system, I doubt they would be reporting anything in unison. How does Linux affect what your battery does when your laptop is *off*?
20:23 < s10gopal> Thedarkb: use tee command
20:23 < Psi-Jack> I like my init system.
20:23 < Thedarkb> s10gopal, go on?
20:24 < Sitri> s10gopal: The battery naturally drains when the laptop is off.  Windows has the same issue as a result
20:24 < Psi-Jack> I specifically do not like upstart, and the broken designs that lsb-init had when people didn't follow the spec as it was designed.
20:24 < Sitri> Linux can't magically fix hardware issues
20:24 < Psi-Jack> heh
20:24 < Thedarkb> How do I use tee for that/
20:24 < slidinghorn> Thedarkb - what's the command you're using?
20:25 < Psi-Jack> s10gopal: man tee
20:25 < Psi-Jack> Err, Thedarkb: man tee
20:25 < Thedarkb> I'm running a dedicated server and I want to log what's going on.
20:26 < Psi-Jack> Dedicated server and log what's going on, don't mix in the context.
20:26 < Thedarkb> and still be able to use the console.
20:26 < Dagmar> The man page explains it pretty well.  It's not like tee is complex
20:26 < s10gopal> Thedarkb: https://www.computerhope.com/unix/utee.htm
20:26 < Dagmar> You wont be able to continue using the "console" of whatever it was
20:26 < Thedarkb> eh, nevermind.
20:27 < s10gopal> and even most of the users think it is hardware problem , but it is kernel problem see https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1745646
20:27 < Psi-Jack> Heh
20:27 < Dagmar> Make whoever wrote the dedicated whatever actually write some logging functionality
20:27 < hexnewbie> Thedarkb: That sounds like a job for ‘tail -f’ ran on a log, or ‘journalctl -f’ with a systemd service
20:27 < Psi-Jack> hexnewbie: Combined with tmux
20:27 < vezveth> hello everyone
20:28 < Psi-Jack> vezveth: Hello mere Earthling.
20:30 < s10gopal> Sitri: but what if battery draines on linux only not on windows?
20:30 < Psi-Jack> s10gopal: Batteries always drain, :
20:30 < Psi-Jack> OS not even a factor at all.
20:30 < Sitri> ^
20:30 < hexnewbie> Why would you possibly want the trusted platform running when your laptop is off?
20:31 < s10gopal> Psi-Jack: https://bugzilla.kernel.org/show_bug.cgi?id=198665
20:31 < s10gopal> even i have turned off tpm
20:32 < hexnewbie> Someone used to debate the benefits of hard OFF on AT power supplies here, but not even ATX or laptops of any era thought it wise to do anything battery-intensive while reporting to be off.
20:33 < Psi-Jack> s10gopal: There's a reason why it's flagged "Needs more information"
20:33 < Psi-Jack> That whole bug report is vague as hell.
20:34 < Psi-Jack> I'm betting if it is a bug, it's a hardware level bug.
20:34 < hexnewbie> Or a really horrible design decision
20:34 < mallu> hexnewbie: is it possible to share printers.conf and ppd file between multiple servers? One printers.conf file in a shared directory and multiple servers access it?
20:34 < Psi-Jack> Keeping the battery activate while powered down to enable charging for USB ports for portable devices and such, even while "off".
20:34 < Dagmar> Wake on keyboard/serial/lan
20:35 < hexnewbie> mallu: I seem to recall CUPS tends to write to printers.conf (for storing the printer status, etc.), so possibly not.
20:35 < s10gopal> WOL + TPM are disabled
20:35 < s10gopal> and no external device connected
20:36 < Dagmar> Well, we don't have your hardware so there's nothing we can do about the problem you have
20:36 < Psi-Jack> s10gopal: Whether or not a device is plugged in is immaterial.
20:36 < Psi-Jack> s10gopal: As in, this is an HP problem.
20:36 < s10gopal>  google it , it happen with dell and msi too
20:37 < s10gopal> asus lenovo
20:37 < Dagmar> What precisely are you expecting we're going to be able to do about this?
20:37 < Psi-Jack> I still say it's a hardware problem.
20:37 < mawk> my buddy's iphone just exploded on his face
20:37 < mawk> his nose is burnt now
20:37 < revel> mawk: Pics?
20:37 < Dagmar> Maybe he shouldn't bite the phone
20:38 < mawk> revel: https://pix.watch/YCluJF/87dgQh.jpeg
20:38 < Psi-Jack> Just because it looks tasty doesn't mean it's like a brownie. :)
20:38 < mawk> lol
20:38 < Psi-Jack> mawk: Heh, why did it blow up?
20:39 < revel> Naet.
20:39 < mawk> I suppose that's because he changed the battery himself a while ago
20:39 < revel> lol
20:39 < mawk> and the battery got damaged somewhere in the process
20:39 < Psi-Jack> Ahhhhhh
20:39 < hexnewbie> Probably another Linux kernel fault!
20:39 < Psi-Jack> Quite possible, yes.
20:39 < s10gopal> Psi-Jack: it dont happen on windows 10 , but happen when i install linux , even it dont happen on 4.12
20:40 < Psi-Jack> hexnewbie: Yeah. I think Linux somehow got onto that iPhone and made the battery overcharge itself while not even plugged in, just tesla'd energy over the air while in his hand.
20:40 < Psi-Jack> s10gopal: It's still absolutely and technically a hardware problem.
20:40 < Psi-Jack> No ifs ands or butts about it.
20:41 < meyou> lol
20:42 < meyou> "bug report is too vague, it's definitely hardware"
20:42 < hexnewbie> When the next Destroyers Update for Redmond OS triggers the same hardware bug in the laptop, I'm sure it will still somehow be a Linux problem :)
20:43 < Psi-Jack> I mean, sure, whatever hacky bad work-around that may be going on in ACPI or such that "makes it work", doesn't mean it's a Linux problem, when hardware should simply be better designed to do what it's designed to do, and when told to power off, if nothing needs power, power down fully.
20:44 < Psi-Jack> That in effect is indeed a hardware problem.
20:44 < Psi-Jack> At least its not vPro. :)
20:44 < s10gopal> Psi-Jack: and these hardware bug cant be fixed?
20:45 < hexnewbie> Linux tells the laptop to power off. It does not. That part of the report is not unclear.
20:45 < Psi-Jack> s10gopal: Often times fixed via BIOS/Firmware updates from the manufacturer.
20:45 < moqq> hello... i'm having some troubles configuring a masquerade nat setup. the NAT router machine has a single eth interface which has ip forwarding enabled and a masquerade iptables rule setup. on the "internal" machine that wants to get to the internet, when i do a single static route, it works fine: 'route add 1.1.1.1 gw 10.10.0.79 eth0' then ping 1.1.1.1 works fine
20:45 < spammcoin> speculative execution side channel attacks are hardware bugs, didn't stop the patches from flowing
20:46 < hexnewbie> Granted, future people may laugh at that notion of they get fully software-defined meaning of what ‘off’ means, but we're not there yet, so from our POV that's a hardware bug.
20:46 < phogg> spammcoin: because patching hardware is too hard
20:46 < mawk> yes moqq
20:46 < moqq> however, when i issue the 'route add default gw 10.10.0.79 eth0' i'm disconnected from the machine and can't get back in. anyone have any idea what this might point to?
20:46 < mawk> why are you using route tho ? isn't that thing deprecated ?
20:46 < mawk> use ip
20:46 < moqq> orly
20:46 < mawk> how do you access that machine moqq ? please show the current rules before your modification
20:46 < mawk> paste output of ip route show
20:47 < moqq> i'm SSH'd to it via the same interface. ok
20:47 < Psi-Jack> moqq: "oh really" for future corrections. It does matter.
20:47 < mawk> and your current computer has a 10.10.0.0/16 address moqq ?
20:48 < hexnewbie> What eagles are we talking about? https://en.wiktionary.org/wiki/orly#Czech
20:48 < moqq> mawk: i'm SSH'd to it from another computer that has a 10.10.0.0/16 address, yes
20:50 < moqq> ubuntu@ip-10-10-0-141:~$ ip route show \\\ default via 10.10.0.1 dev eth0  \\\  10.10.0.0/24 dev eth0  proto kernel  scope link  src 10.10.0.141
20:50 < hexnewbie> moqq: What IP does "who" list for your SSH connection?
20:50 < moqq> the rules before i change anything
20:50 < moqq> 10.10.2.154
20:50 < moqq> oh
20:50 < hexnewbie> moqq: Yeah, you're in a different network.
20:51 < moqq> yes, goooood call. oops. thank you
20:51 < moqq> sorry, Psi-Jack, i mean, "Yes, good call."
20:53 < Dagmar> This is why people should take the time to actually learn what little there is to ipv4 routing
20:53 < s10gopal> Psi-Jack: which laptop brand is good if we need to run linux on it ?
20:53 < Dagmar> It's only hard when you're using voodoo to configure your network
20:53 < fendur> s10gopal: google has good answers for that (I just looked recently)
20:54 < phogg> Dagmar: but cargo cult networking seems to make the packets arrive OK, we mustn't risk angering the network gods by changing anything.
20:55 < Dagmar> It's too late for that now.  Two default gateways in one machine means they must FIGHT TO THE DEATH
20:56 < phogg> two gateways enter, one dell leaves
20:57 < hexnewbie> Quick way to check which branch you need to cut so you fall to the ground: ip route get "${SSH_CLIENT%% *}"
21:03 < Dagmar> =D
21:04 < triceratux> rofl http://downtoearthlinux.com/posts/11-reasons-to-avoid-linux/
21:05 < Psi-Jack> s10gopal: *
21:07 < s10gopal> Psi-Jack: you are kidding right ?how linux is slow than window , and it has best community
21:07 < Psi-Jack> What?
21:07 < s10gopal> http://downtoearthlinux.com/posts/11-reasons-to-avoid-linux/
21:07 < Psi-Jack> Notice who posted that.
21:09 < fendur> "pointless competition"? wtf
21:09 < triceratux> s10gopal: "*" meant "all laptops", not "that was an exceptionally fine blog post"
21:10 < CoJaBo> s10gopal: His problem is that he's using Ewwbuntu.
21:10 < Psi-Jack> CoJaBo: Elitism isn't becomming of you here.
21:10 < Psi-Jack> :p
21:11 < CoJaBo> Many of those are valid reasons why Ubuntu specifically sucks, but not linux in general :P
21:11 < Psi-Jack> However, this particular page does focus primarilly around Ubuntu == Linux, as if it were the same thing, which is extreme bias.
21:11 < Psi-Jack> heh
21:11 < CoJaBo> Kubuntu and probably the Firefox 4k fix would likely resolve almost all of his issues.
21:12 < Psi-Jack> If only they had KDE 3.x
21:12 < Psi-Jack> Or at least finally fixed KDE 5
21:12 < CoJaBo> I can't believe Firefox still hasn't fixed that tho :/
21:12 < hexnewbie> Reason #12: Satya Nadella didn't show off flying to the ISS on a Russian rocket.
21:13 < gnulligan> Are there any decent desktop Android x86 forks?
21:13 < gnulligan> remix os died
21:13 < Psi-Jack> gnulligan: Got a Linux question?
21:13 < gnulligan> Android isnt linux?
21:13 < CoJaBo> With a stock install, my P4 desktop with 1024×768 running Xewwbuntu runs laps around my gaming laptop with any distro on a 4K screen.
21:13 < gnulligan> :P
21:13 < Psi-Jack> gnulligan: This channel primarily focuses on GNU/Linux, not Android/Linux. Perhaps you're looking for #android
21:13 < CoJaBo> Stupid thing defaults to NO hardware accel, which is.. uhh, stupid.
21:15 < s10gopal> Psi-Jack: changing linux destro can fix my bug ?
21:16 < Psi-Jack> s10gopal: TIAS
21:16 < hexnewbie> s10gopal: Didn't you say earlier that changing your kernel fixed the issue for you?
21:16 < Psi-Jack> Doubt it, since it's a hardware bug, but hey!
21:17 < hexnewbie> <s10gopal> Psi-Jack: it dont happen on windows 10 , but happen when i install linux , even it dont happen on 4.12
21:17 < s10gopal> hexnewbie: yes it dont happen upto  kernel 4.12
21:17 < hexnewbie> s10gopal: 4.12 is out of support, but 4.9 is LTS, so you may stick with that one (or distro that has it)
21:18 < s10gopal> hexnewbie: ubuntu 14.04lts still uses 4.4
21:19 < stranglerfish> hey, i'm having a bit of trouble extracting text using grep / awk. i have very little experience with either / regex, could someone recommend a resource for getting a handle on it fairly quickly? I'm just trying to get a working understanding of it so i can fix something that's not working for methat i got off someone's repo
21:19 < jeffree> is anyone using cloudflare's DNS over https?
21:19 < stranglerfish> using grep to extract weather info from an api call
21:19 < Psi-Jack> jeffree: Got a Linux question?
21:20 < Psi-Jack> DNS != http protocol, let alone https.
21:20 < jeffree> is it possible to setup DoH on a linux-based router?
21:20 < Psi-Jack> DoH?
21:20 < jeffree> see above
21:20 < stranglerfish> to be specific, i'm trying to fix this: https://github.com/Anachron/i3blocks/blob/master/blocks/weather so that the weather_main actually grabs the right info as it's returning nothing for me
21:20 < Psi-Jack> Department of Humans?
21:21 < hexnewbie> Using a DNS over HTTPS? Is that an extension for RFC3252?
21:21 < Psi-Jack> That.... Sounds... Horrible.
21:21 < Psi-Jack> Stop doing stupidly bad things! ;)\
21:22 < jeffree> rationale would be more valuable than just callling it stupid
21:23 < hexnewbie> Rationale: HTML is parsed using a HTML parsing library, not ad-hoc Bash commands?
21:23 < Dagmar> SOme of the reporting on the new Cloudflare thing has been well... I'm sure they meant well.
21:23 < Dagmar> ...and that their mother thinks they're bright.
21:23 < jeffree> lol
21:23 < jeffree> https is not html
21:24 < Dagmar> I'm probably only a few more days of bullshit news articles away from sending some howler mails at reporters
21:24 < Dagmar> Guys who _clearly_ are just copying stuff they don't understand from other articles written by people who also didn't understand how DNS works
21:25 < Psi-Jack> I mean, DNS over HTTPS is a backasswaters hackery without thought. DNSCurve would be literally lightyears better.
21:26 < Psi-Jack> Even better would be TCP-Curve, encrypt the entire TCP traffic itself.
21:27 < hexnewbie> stranglerfish: Fix your variables, first one needs to be ICON_SUNNY="🌞" as that's the proper Unicode and not PU, besides it works with Unifont, which is 8x16 and size-compatible with console VGA and Terminus.
21:27 < hexnewbie> s/PU/PUA/
21:29 < T-Rog> how do I build Arc and install it to ~/.local/share/themes instead of /usr/share/themes
21:29 < Sitri> <Psi-Jack> Even better would be TCP-Curve, encrypt the entire TCP traffic itself. <-- IPSec?
21:30 < Psi-Jack> Sitri: No. TCP-Curve
21:30 < Psi-Jack> Dan Bernstein may've failed making a worthwhile email server (qmail), but his crypto is solid and sound.
21:32 < hexnewbie> I've been adopting ed25519 where the SSH is new enough to support it.
21:33 < stranglerfish> it's not my code, the main part i'm concerned with hexnewbie is that it doesn't seem to be actually grabbing the WEATHER_MAIN
21:35 < hexnewbie> stranglerfish: The owner doesn't appear to have a valid API key
21:35 < stranglerfish> i'm testing it, i stored the output into a file 'weather' and am doing 'cat weather | grep -o -e '\"main\":\"C' and that works
21:35 < stranglerfish> ignore all that stuff, hexnewbie. assume i'm working with my own local file that has a valid api key and instance for my city
21:35 < hexnewbie> stranglerfish: I mean, the person who wrote the script (or whose API key was misused in it) doesn't appear to have a working API key anymore
21:36 < stranglerfish> yeah no i'm not using that exact script, mine is using my working api key, and my own city
21:36 < stranglerfish> it's the WEATHER_MAIN part that's not working for me, everything else appears to work fine
21:36 < Dagmar> Is the output XML?
21:36 < stranglerfish> as far as i know, just plain text
21:36 < stranglerfish> ok let's change the example
21:36 < Dagmar> so pastebin one of the outputs
21:37 < Dagmar> Also, on general principle, chaining grep to awk to extract an argument is a waste of execution time
21:37 < stranglerfish> https://pastebin.com/aNZXS1TP
21:38 < Dagmar> This is looking like something sed can do in one round
21:38 < hexnewbie> That's JSON
21:38 < Dagmar> That's filthy json data
21:38 < gronke> hexnewbie, if you care, I figured out what the problem was with my conda thing
21:38 < Dagmar> Use an actual programming language to parse it
21:39 < stranglerfish> i'm just trying to make it work with the current script. there must be some way to implement it with what is there, it's just something wrong with the regex i think. but the thing is i don't know regex all that well
21:39 < hexnewbie> gronke: What was it?
21:39 < scyld> Hello. Heads up. I'm just wondering if such a thing exists in linux. I think about storage with two disks (classes) slow and fast. I would like to create a filesystem with the functionality of migrating data based on some policies, between this two disks (classes). For instance policy specifies that if file is not accessed for 3 weeks it is being migrated to slower disk, etc. Can be done under linux?
21:40 < gronke> hexnewbie, when you run the run.sh script for the server, it does all this install/dependency check stuff, it did that for conda, but the problem was that I didn't have bzip2 installed so it couldn't unzip the bz2 files to install conda, so it would just say "error" and then continue and launch the server with me none the wiser
21:40 < hexnewbie> stranglerfish: Cutting JSON with Regexes won't work. The Python code for this is really: data = json.load(urllib2.urlopen(url)); print data['main']['temp']
21:40 < Dagmar> stranglerfish: This is why you should be using an actual language that has a JSON parsing library
21:40 < meyou_> scyld, so BYO Tiered SAN
21:40 < stranglerfish> ok, i'll learn python then
21:41 < Dagmar> Honestly it shoudln't tkae more than a day to figure that much of it out
21:41 < hexnewbie> gronke: Yeah, errors shouldn't be silently ignored (and I hate distros that don't install bzip2 by default)
21:41 < gronke> I mean it did say it, but it was in a big spam of install script stuff that fills up the screen so I figured that the server successfully starting meant everything went ok
21:41 < meyou_> scyld, google's telling me to look into the flashcache kernel module, sounds like it may do what you're after
21:41 < stranglerfish> fair, i just figured it worked at one point for whoever wrote this script so whether or not it's the most optimal / easy method, i should be able to fix it but i could just use it as an opportunity to make something similar in python
21:42 < hexnewbie> gronke: I didn't mean by you, I meant by the installer :)
21:42 < gronke> ah yes :P
21:42 < scyld> meyou_, hmm… ok I will check it out…
21:42 < edd_lc> stranglerfish: python -c "import json; json.loads(\"\"\"${data}\"\"\")['weather'].pop()['main']"
21:42 < gronke> hexnewbie, any idea what this means? "!!! no internal routing support, rebuild with pcre support !!!"
21:43 < stranglerfish> ok cool
21:43 < stranglerfish> i gotta jet, thanks for the help
21:43 < hexnewbie> gronke: Is it installing things with pip?
21:43 < azizLIGHT> i am backing up hdd to a usb. the usb has 100gb space left. the hdd is 230 gb big, but only 20gb data on it.... i used "sudo dd if=/dev/sda conv=sparse status=progress | gzip -c | split -b 4000m - img.gz."
21:43 < azizLIGHT> will it be able to backup to the usb? or will usb fill up
21:44 < gronke> hexnewbie, I don't think so, this is happening when I launch the shell script and it's doing all the install/dependency checks, here it is: https://paste.pound-python.org/show/mJZFo5puByiXlT6BSrws/
21:44 < hexnewbie> gronke: Hm, this uwsgi was installed by Galaxy?
21:44 < hexnewbie> In a virtualenv as far as I can tell
21:45 < gronke> hexnewbie, I believe so, it does a big check to see that everything it needs is installed before launching the server, now that I had that conda issue I'm giving a closer look to all the error messages
21:45 < gronke> It may be a non issue but who knows
21:46 < hexnewbie> gronke: Install libpcre and pcre-devel (don't know how CentOS packages are called), and try to rebuild/reinstall wsgi in Galaxy's the virtualenv (*not* on your system)
21:46 < hexnewbie> gronke: Or, of course, if it works, don't mess with it :)
21:50 < hexnewbie> azizLIGHT: Hm, that may not work if the free space is not actually zeroes. Why not instead rsync?
21:50 < azizLIGHT> well i want to recover data from free space if possible
21:52 < hexnewbie> azizLIGHT: I see. But it still isn't guaranteed to fit. You may go shopping for a larger drive while it's copying? Or free some space on the target filesystem?
21:52 < azizLIGHT> hmmm. i think ill have to save to a network share
21:52 < azizLIGHT> but ive been having trouble with samba permissions so much
21:53 < azizLIGHT> well.... its gone through 96GB already and the resulting backup gzips are 11gb total
21:53 < azizLIGHT> so thats looking good
21:53 < azizLIGHT> i just saw that
21:57 < nobody> hi :)
22:02 < azizLIGHT> 157GB processed, and file size of backup gzips on usb is only 12gb
22:02 < azizLIGHT> i am lookin good
22:04 < Psi-Jack> azizLIGHT: Are you sure the mirror is properly tuned to be able to confirm such a bold statement? ;)
22:05 < azizLIGHT> i am looking sleepy :D
22:06 < Psi-Jack> Perhaps you need more light! :D
22:06 < azizLIGHT> heheheheeh
22:07 < triceratux> woo woo https://www.linux.com/learn/intro-to-linux/2018/4/mx-linux-mid-weight-distro-focused-simplicity
22:11 < s10gopal>  why it is too much costly https://training.linuxfoundation.org/linux-courses/development-training/introduction-to-linux-open-source-development-and-git ?
22:14 < Psi-Jack> s10gopal: Give me your money instead.
22:14 < s10gopal> can i get these course video for fee ?
22:15 < s10gopal> that money is equivalent for my 3 year college fee
22:15 < Psi-Jack> Obviously not.
22:16 < SuperSeriousCat> Hah. Feel like s10gopal have asked questions non-stop this whole day :p Whenever I glanced in here since 12 hours ago, he was asking a question
22:16 < Psi-Jack> heh
22:24 < zenix_2k2> guys do most of distros nowadays use gnome-terminal ?
22:24 < zenix_2k2> or have ?
22:25 < doubtful> where can I find more of them?
22:25 < doubtful> what are these kinds of wallpapers called https://imgur.com/r/unixporn/ElgzA?
22:25 < doubtful> where can I find more of them?
22:26 < phogg> zenix_2k2: Distributions don't use terminals, people do. Distributions which ship GNOME tend to set gnome-terminal as the default. Not all distributions ship GNOME.
22:26 < phogg> zenix_2k2: personally I find gnome-terminal to be rather horrible in every way and would never choose to use it.
22:27 < zenix_2k2> well it looks nice though
22:27 < zenix_2k2> but do most distros nowadays have it builtin ?
22:27 < Psi-Jack> I actually like gnome-terminal.
22:28 < Psi-Jack> I used to like xfce-terminal, but it lacks fullcolor proper support.
22:28 < triceratux> zenix_2k2: gnome-terminal is trivial to install on any sane distro. most distros ship with the terminal specific to their de for starters tho
22:29 < zenix_2k2> triceratux: yea but i ain't talking about shipping them for starters, i mean do they available on most of the distros nowadays ???
22:29 < Psi-Jack> zenix_2k2: Have what? X-terminal emulators? Yes.
22:29 < zenix_2k2> No
22:29 < zenix_2k2> gnome
22:30 < Psi-Jack> All major distros have gnome.
22:30 < Pentode> Psi-Jack, it does?
22:30 < zenix_2k2> and if you mean "X" by a bunch of emulators then ok
22:30 < Psi-Jack> zenix_2k2: No, by X I mean X.
22:30 < doubtful> wallpapers :)?
22:30 < Psi-Jack> Aka X11, X Window System, X.org, XFree86, etc.
22:31 < zenix_2k2> Oh, nice to hear
22:31 < Psi-Jack> So far, though you've been dancing around seemingly randomly.
22:32 < zenix_2k2> yea, not only here
22:32 < zenix_2k2> but seriously, gnome-terminal -e <command> doesn't work anymore
22:32 < zenix_2k2> any solution ?
22:32 < zenix_2k2> it said something like "Option “-e” is deprecated and might be removed in a later version of gnome-terminal."
22:33 < phogg> zenix_2k2: use a terminal emulator that still supports -e
22:33 < zenix_2k2> oh com'on...
22:34 < phogg> zenix_2k2: Get the gnome-terminal source code, re-add support and rebuild it.
22:34 < zenix_2k2> but if there is no other option rather than that then
22:34 < zenix_2k2> ok i will switch to xterm instead
22:34 < Psi-Jack> What's the -e do?
22:34 < phogg> zenix_2k2: use an older version of gnome-terminal
22:34 < phogg> Psi-Jack: executes the command specified after running the terminal. Think su -c, but for a terminal.
22:34 < zenix_2k2> Psi-Jack: it will run the inputted command
22:34 < zenix_2k2> like an argument that we pass in
22:35 < Psi-Jack> You can do that without -e
22:35 < zenix_2k2> No i am trying to do that via a script
22:35 < zenix_2k2> .sh
22:35 < zenix_2k2> not manually typing
22:36 < zenix_2k2> and if there are modules that support that in python then it would be nice
22:36 < Psi-Jack> And?
22:36 < zenix_2k2> And wut ?
22:36 < phogg> Psi-Jack: Blame xterm; it has supported -e for this since forever, so most terminals include it for compatibility.
22:36 < zenix_2k2> well but gnome still looks better than xterm
22:36 < Psi-Jack> None is what you said, zenix_2k2 changes what I said at all.
22:37 < zenix_2k2> ok, i am kinda lost you there
22:38 < Psi-Jack> What are you lost about? The fact that you don't need -e?
22:38 < zenix_2k2> no i do need -e
22:38 < Psi-Jack> No, you do not.
22:38 < zenix_2k2> so what do i need instead ?
22:38 < Psi-Jack> gnome-termina [COMMAND]
22:39 < Psi-Jack> +l
22:39 < phogg> zenix_2k2: I think Psi-Jack means to say "gnome-terminal does not require that you specify -e in order to execute a command you give it"
22:39 < zenix_2k2> it doesn't ?
22:39 < Psi-Jack> gnome-terminal --help
22:39 < zenix_2k2> Hm, let's me try
22:39 < hexnewbie> zenix_2k2: gnome-terminal -- command arg1 arg2 arg3
22:39 < Psi-Jack> -- works too. :)
22:39 < Psi-Jack> Ahh -- is needed actually. :)
22:40 < Psi-Jack> hexnewbie: Thank you. Corrected noted. :)
22:40 < phogg> I would certainly hope so.
22:40 < zenix_2k2> and do "gnome-terminal -- command arg1 arg2 arg3" also work on older versions of gnome ?
22:40 < zenix_2k2> cause i don't know whether my friends are up-to-dated or not
22:41 < phogg> zenix_2k2: it probably has worked since at least the time they started warning about use of -e
22:41 < zenix_2k2> oh yay
22:42 < Psi-Jack> zenix_2k2: It's worked since the beginning of time.
22:42 < hexnewbie> The ‘standard’ -e is no good, since konsole, gnome-terminal and xterm have a slightly different interpretation of it. Or rather, xterm interprets it in two different ways depending on context, and gnome-terminal support(ed) one of them, konsole another, and I assume it's a bigger mess across the rest of the terminal emulators
22:42 < zenix_2k2> thank god it does
22:42 < Psi-Jack> I believe even xterm, rxvt, konsole, etc understand -- command
22:43 < Psi-Jack> But, I don't even have anything other than gnome-terminal even installed, so. ;)
22:43 < zenix_2k2> well gnome looks nice and cool though
22:43 < zenix_2k2> even tho it eats too much RAM
22:43 < phogg> hexnewbie: oh? What are the two contexts for xterm's -e?
22:44 < zenix_2k2> not too much but still kinda much
22:44 < Psi-Jack> Heh. I don't use Gnome.
22:44 < Psi-Jack> Just parts of it.
22:44 < hexnewbie> phogg: As a single shell command, or as multiple command arguments.
22:44 < phogg> hexnewbie: ah yes, of course
22:44 < zenix_2k2> yea right, only its terminal ;)
22:44 < phogg> hexnewbie: this is why sane people don't ever try args with -e!
22:45  * phogg learned to wrap such things in shell scripts so long ago he'd actually forgotten why
22:45 < Psi-Jack> heh
22:46 < ejr> i am running only a tty, no X, and I want to deactivate the laptop display so that the output is only on my external display. "sudo vbetool dpms off" does not work ("Real mode call failed."). Are there other ways to switch off a screen when only using the framebuffer?
22:47 < lukey_> ejr: You could try with the video= kernel parameter
22:47 < hexnewbie> ejr: There might be a way, but the virtual console's support for graphics is poor, so you may have better luck by installing X
22:47 < phogg> lukey_: without rebooting?
22:48 < Psi-Jack> ejr: Sure. Install X, choose whatever window-manager or desktop-environment you want, profit. :)
22:48 < Psi-Jack> Or you could choose not to run any desktop/windowmanager and just run a fullscreen x-terminal session, like a console. :)
22:50 < ejr> Psi-Jack / hexnewbie: thx but no X
22:50 < ejr> lukey_ ok, I will look into that, thx
22:50 < hexnewbie> ejr: If you do install it, you'll have it.
22:50 < Psi-Jack> ejr: "thanks", not "thx" for future self corrections on your part. Yes, it really does matter.
22:50 < zenix_2k2> and one more thing, do i need the full path when i run this --> gnome-terminal -- "python" "calculate.py" ( the path of calculate.py )
22:51 < zenix_2k2> cause i am currently on Desktop and calculate.py is also stored there
22:51 < zenix_2k2> but it doesn't work
22:51 < hexnewbie> zenix_2k2: Did it work when you tried full path?
22:51 < ejr> Psi-Jack: y? :P
22:51 < zenix_2k2> Nope, doesn't either
22:52 < Psi-Jack> ejr: "why", not "y" for future self corrections. It is literally in the rules not to use sms-speak (aka shtspk).
22:52 < hexnewbie> zenix_2k2: What error did you get, or what happened?
22:52 < hexnewbie> zenix_2k2: Does calculate.py exit immediately?
22:52 < ejr> Psi-Jack: "short message-speak", not "sms-speak" for future self corrections then.
22:53 < Psi-Jack> ejr: Actually SMS is known world-wide. :p
22:53 < cnrhkiyf> says who?
22:53 < Psi-Jack> Says the telecom industry. Since it's world wife.
22:53 < Psi-Jack> wide*
22:53 < zenix_2k2> hexnewbie: this creepy form showed up --> https://ibb.co/iuOdKx
22:53 < ejr> i would argue that "thx"is know all over the internet, but whatever :P if it's in the rules, I shall abide by them.
22:53 < zenix_2k2> and actually i should have called that terminal
22:53 < lukey_> ejr: You could also play with the stuff in /sys/class/drm/
22:54 < zenix_2k2> and i remember correctly that /home/zenix/calculate.py does exist
22:54 < Psi-Jack> ejr: Thank you. We adhere to the understanding not everyone is native English speaking/literate. Not only that, shtspk just looks lazy. Do you want to help a lazy person? There's a difference. :)
22:54 < Psi-Jack> zenix_2k2: Don't remember. Verify
22:55 < ejr> lukey_ thanks! looks interesting, though complicated. I will read some more about what those individual files in drm/ mean
22:55 < irrgit_1> Psi-Jack, hello you useless git, can you help me with my problem?  https://paste.pound-python.org/show/30M75amsxoF6RtlT7enn/
22:55 < zenix_2k2> i tried "python /home/zenix/Desktop/calculate.py" and it worked
22:55 < zenix_2k2> same on ./home/zenix/Desktop/calculate.py
22:56 < irrgit_1> Im missing that file and im not sure how to install it, i installed jdk 1.7 from an rpm provided by oracle
22:56 < hexnewbie> zenix_2k2: ./home/zenix/Desktop/calculate.py is not a full path
22:56 < zenix_2k2> wait, so which is ?
22:56 < Psi-Jack> irrgit_1: Sure. I'll start by pounding your face in 20 times, breaking each limb of your body, one by one, over the course of the next few hours, and we'll see what's next after that.
22:56 < hexnewbie> zenix_2k2: Any absolute path that starts with the root (/)
22:57 < zenix_2k2> well /home starts with "/"
22:57 < fendur> whoa.
22:57 < irrgit_1> Well too bad, look at the time, its almost 5, i guess ill be here monday again :D
22:57 < hexnewbie> zenix_2k2: ./home does not; . is the current directory
22:58 < zenix_2k2> i don't really get it, so what is the correct path ?
22:58 < zenix_2k2> and i don't think /home is the current directory
22:58 < fendur> I think there are better ways to diffuse a problem than clearly trying to escalate it. heh.
22:58 < zenix_2k2> i am currently in Desktop
22:58 < hexnewbie> zenix_2k2: I'm not on your computer, but I suspect the correct path is /home/zenix/Desktop/calculate.py
22:58 < Psi-Jack> fendur: What're you on about?
22:59 < fendur> Psi-Jack: just adult stuff.
22:59 < Psi-Jack> Well, you have the whole internet, afterall, internet is for what again? ;)
22:59 < zenix_2k2> no but i wanna execute it, i know that  /home/zenix/Desktop/calculate.py is my absolute path
22:59 < hexnewbie> fendur: You need mpv for that
22:59 < zenix_2k2> like something like this
23:00 < zenix_2k2> ./file.py ( with chmod +x file.py and there is a shebang line in it )
23:00 < zenix_2k2> so i tried with . /home/zenix/Desktop/calculate.py
23:00 < zenix_2k2> ./home/zenix/Desktop/calculate.py*
23:00 < Psi-Jack> zenix_2k2: . again?
23:00 < SuperSeriousCat> "python /home/zenx/Desktop/calculate.py
23:00 < zenix_2k2> why not . ?
23:00 < Psi-Jack> zenix_2k2: Do you not understand what . is?
23:00 < zenix_2k2> well i tried but it doesn't work when i open it in another terminal
23:00 < SuperSeriousCat> Or just "python file.py"
23:01 < zenix_2k2> gnome-terminal -- "python /home/zenix/Desktop/calculate.py"
23:01 < hexnewbie> zenix_2k2: . means the current directory, just like .. means the parent directory (run ‘ls -la --color=always | head’ and see the two directories on top). Placing it at the start of a path makes the path *relative* to the current directory.
23:02 < hexnewbie> zenix_2k2: No. Remove those quotes.
23:02 < zenix_2k2> also showed this creepy form --> https://ibb.co/iuOdKx
23:02 < zenix_2k2> Oh...
23:02 < zenix_2k2> ok then
23:02 < zenix_2k2> so what should i put in instead ?
23:02 < hexnewbie> zenix_2k2: python is the program,  /home/zenix/Desktop/calculate.py is the argument. By placing it in quotes, you're looking for a program file named "python /home/zenix/Desktop/calculate.py" (which wouldn't exist)
23:03 < Psi-Jack> zenix_2k2: . literally means "current working directory", as in what directory is currently active now, same as what cwd would tell you.
23:03 < zenix_2k2> gnome-terminal -- "python" "/home/zenix/Desktop/calculate.py", also tried
23:03 < zenix_2k2> but didn't work
23:03 < Psi-Jack> So, ./ means current working directory too
23:03 < zenix_2k2> well ok
23:03 < zenix_2k2> but it still doesn't seem to work
23:03 < hexnewbie> zenix_2k2: Does this command work: "python" "/home/zenix/Desktop/calculate.py"
23:04 < Psi-Jack> head -n1 /home/zenix/Desktop/calculate.py
23:04 < Psi-Jack> What does that show?
23:04 < zenix_2k2> my shebang line
23:04 < Psi-Jack> And that is?
23:04 < zenix_2k2> #!/usr/bin/python2
23:04 < Psi-Jack> And is that correct?
23:04 < zenix_2k2> yea
23:05 < Psi-Jack> Is that file chmdo +x?
23:05 < Psi-Jack> chmod*
23:05 < hexnewbie> If you have a shebang, and it has execute permissions, you don't need the "python" part. Still, it shouldn't hurt.
23:05 < zenix_2k2> hexnewbie: but it doesn't work when i combine it with gnome-terminal --
23:05 < hexnewbie> zenix_2k2: We know that. Does it work when you don't, and what happens?
23:05 < zenix_2k2> it does work when i don't
23:05 < zenix_2k2> but i am trying to combine it with gnome-terminal
23:05 < hexnewbie> zenix_2k2: Does it exit immediately?
23:06 < zenix_2k2> with or without the terminal part ?
23:06 < Psi-Jack> Either.
23:06 < hexnewbie> zenix_2k2: Without the terminal part (or with) - does your script exit immediately, or not?
23:06 < Psi-Jack> When you run it, does it do what it does, and immediately return you to a shell prompt again?
23:06 < zenix_2k2> with terminal, it does... but i can't see the output, without it, i can see the output
23:06 < zenix_2k2> and it also exits, i think most programs do
23:07 < Psi-Jack> zenix_2k2: You need to put a sleep, or something in the code to allow it to run without immediately exiting, then.
23:07 < zenix_2k2> a loop ?
23:07 < Psi-Jack> For it to work in the manner of being called into a brand new gnome-terminal.
23:07 < hexnewbie> zenix_2k2: This means that the  ‘gnome-terminal -- "python" "/home/zenix/Desktop/calculate.py"’ command works as expected, your script runs, exits and gnome-terminal closes with it.
23:07 < zenix_2k2> but why when i tried "gnome-terminal -- "echo" "hello world""
23:08 < zenix_2k2> it ran but then it pending a bit ?
23:08 < zenix_2k2> like it doesn't close immediately
23:08 < Psi-Jack> It doesn't.
23:08 < hexnewbie> zenix_2k2: gnome-terminal -- sh -c '"python" "/home/zenix/Desktop/calculate.py"; sleep 20'
23:08 < Psi-Jack> Then again, though, echo itself is a internal bash function too.
23:09 < zenix_2k2> well now it works, when i put a little loop[ in it
23:09 < redrapscallion> I need to be able to run any arbitrary Java/Haskell/Python code and record the outputs of each program. What's the easiest software/technique for safely running this arbitrary code?
23:09 < Psi-Jack> zenix_2k2: It was working before.
23:09 < Psi-Jack> :p
23:09 < zenix_2k2> yea, but computer's magics are too fast to observe
23:09 < redrapscallion> my first thought is to just make a new user with basically no privileges, but is that even sufficient?
23:09 < hexnewbie> zenix_2k2: (My last example is just for testing; it's NOT a recommended way to run things)
23:09 < Psi-Jack> zenix_2k2: That doesn't mean it wasn't working. :)
23:10 < borsin> redrapscallion: use tee or a redirect to a file?
23:10 < zenix_2k2> well ok then
23:10 < zenix_2k2> thank
23:10 < lukey_> redrapscallion: Have a look at firejail
23:10 < Psi-Jack> zenix_2k2: Gotta remember. When you run a terminal, or shell, and you tell that terminal/shell to run a command instead of it's normal routine, it will start  said command, and upon finishing that command, exit and return to the original session.
23:11 < Psi-Jack> So, gnome-terminal -- some command, if some-command executes instructions and finishes and returns/exits, so too will the calling gnome-terminal itself exit.
23:11 < redrapscallion> lukey_: oh, that's perfect. thanks :)
23:12 < zenix_2k2> but echo doesn't exit automatically
23:12 < zenix_2k2> i have tried
23:12 < Psi-Jack> it does.
23:12 < zenix_2k2> No it doesn't, it only does when i press enter
23:12 < Psi-Jack> echo is an internal bash command, so gnome-terminal ran bash to run the internal command "echo", and output.
23:13 < zenix_2k2> oh wait, it does exist
23:13 < zenix_2k2> what is going on around here ?
23:13 < borsin> heh. gnome-terminal running bash.
23:13 < zenix_2k2> my computer is so weird
23:13 < Psi-Jack> zenix_2k2: Now, there is a way to configure gnome-terminal to wait always.
23:14 < zenix_2k2> well i am gonna do that tomorrow
23:14 < zenix_2k2> anyway thank
23:14 < Some_Person> Is it possible to have a cron schedule that triggers on the Friday of the 3rd week of the month?
23:14 < Psi-Jack> There's always --wait too
23:14 < Psi-Jack> Oh, but he didn't stick around and learn. :0
23:16 < Psi-Jack> There we go. MemoServ. :D
23:16 < mawk> I need to invent a programming challenge
23:16 < mawk> but every thing I think of already exists
23:17 < Psi-Jack> mawk: Code to detail how to do cold fusion.
23:17 < mawk> lol
23:17 < SuperSeriousCat> Do a challenge where you get points for fewer lines of code
23:18 < redrapscallion> ie; code golf
23:18 < borsin> mawk: not true. come up with a new idea for completing a task or job. no one here is going to give you those ideas. otherwise contribute to an existing project if you can't come up with something
23:18 < hexnewbie> Tide pod challenge but with Perl golf?
23:18 < TJ-> Obfuscated BASH ?
23:18 < revel> SuperSeriousCat: A single 5k character line of C?
23:19 < Psi-Jack> Wait, I knwo!
23:19 < borsin> create a new face-book. :p
23:19 < TJ-> mawk: how long is the challenge supposed to last, 2 hours was it?
23:19 < mawk> yeah something like that
23:19 < revel> Or just a shell script with ; instead of \n
23:19 < mawk> between 0.5 and 2 hours
23:19 < Psi-Jack> Do a "Hello World" project that actually contacts EVERYONE in the world to say "Hello" to them./
23:19 < Psi-Jack> It's never been done before. :)
23:19 < hexnewbie> Psi-Jack: From orbit.
23:19 < mawk> I'd like to do something with ttys because it's fun and complex to understand fully
23:19 < borsin> Psi-Jack: pretty sure that's already been done by accident
23:19 < TJ-> mawk: is there any particular set of tools or topic it's supposed to use?
23:19 < diogenese> blindfolded
23:19 < borsin> decades ago
23:20 < Psi-Jack> borsin: Nope.
23:20 < mawk> not really TJ- , you usually do it the way you want in the least time
23:20 < borsin> Psi-Jack: I don't have the link, but you could fine it. I bet you're good with MSN search. :p
23:21 < TJ-> mawk: I once set the challenge to create a basic HTTP (GET/POST) server using only bash shell
23:21 < borsin> s/fine/find
23:21 < Psi-Jack> borsin: Instead, I'll fine you. That'll be $5,000,000,000,000,000 payable only in US Currency cash.
23:21 < borsin> Psi-Jack: If I had that I wouldn't bore myself with IRC. I would find me a nice peaceful place to meditate.
23:22 < Psi-Jack> You mean party, right?
23:22 < hexnewbie> Before starting the implementation remember that in extraterrestrial tongue ‘Hello’ is spelled ‘Missile from Mars inbound’
23:22 < borsin> Psi-Jack: Or that.
23:22 < lukey_> Some_Person: No
23:23 < mawk> that's nice TJ-
23:23 < mawk> it's about a real challenge borsin
23:23 < mawk> something entertaining, not especially useful
23:23 < Psi-Jack> mawk: You could implement the full functionality of a teapot brewing system, and error code proper for if a coffee maker tries to use said teapot...
23:23 < borsin> mawk: as I said a new face-book would be a real challenge and would be very useful to some.
23:24 < TJ-> mawk: when you said using tty, you mean using obscure terminal codes?
23:24 < mawk> not especially the control codes but the whole tty system TJ- , with foreground process groups, controlling terminals, session leaders, etc
23:24 < borsin> and make it with a plastic arm, preferrably recycable materials.
23:24 < Psi-Jack> Implement "MyFace" ;)
23:24 < zapotah> does anyone remember a link to the backstory of that return code?
23:25 < TJ-> mawk: that might be a bit extreme for a 2-hour challenge, that's a lot of learning curve for a novice
23:25 < Psi-Jack> A link? No. I remember the actual story. :p
23:25 < Some_Person> lukey_: Well crap... that means I can't use cron to schedule this thing
23:25 < mawk> they won't be novices I think TJ- , it's a CTF event
23:25 < Psi-Jack> Google remembers links. :)
23:25 < mawk> I fear of doing something too easy after all
23:26 < lukey_> Some_Person: You could schedule an Script Daily that checks if your conditions are met
23:26 < borsin> mawk: do a loop of hello world's.
23:27 < borsin> for every human on the planet.
23:27 < borsin> maybe use a real paper bound phone book as a reference
23:28 < Some_Person> lukey_: It ends up being the 3rd Friday of the month in most cases, the only exception being when the month starts on a Saturday
23:28 < borsin> .. the boredom that creeps in while you're waiting for a partner to review code. :p
23:28 < TJ-> mawk: have you see the new kernel 'lock down' patch-set? Maybe you could set a challenge around (trying to) compromise it
23:29 < TJ-> mawk: that'd be very topical, and no-one will have done it before
23:29 < mawk> the one Linus yelled after ?
23:29 < Psi-Jack> mawk: Maybe. Bring a painball gun to the event, and a flag. PLay REAL CTF. :)
23:29 < mawk> lol
23:29 < TJ-> mawk: right, it's looking like it'll land soon
23:29 < hexnewbie> Make Linus mad, and submit a kernel patch supporting floating point IPv6 addresses (RFC 8135).
23:29 < TJ-> mawk: basically the challenge would be to compromise the boot loader
23:31 < TBotNik> All: Need help with issuing a linux cmd from PHP!  Did what the manual said, but not running the cmd!  Write up at:  https://www.linuxquestions.org/questions/showthread.php?p=5840172#post5840172
23:32 < lukey_> Some_Person: I mean do something like "if [ $(date '+%d') -gt 14 ] && [ $(date '+%d') -lt 21 ] && ... "
23:33 < lukey_> Some_Person: And execute that daily using cron
23:33 < mawk> it seems tricky TJ-
23:34 < hexnewbie> TBotNik: I believe someone mentioned to you here that you shouldn't need --display=':n' argument to gedit if you have a DISPLAY variable set
23:34 < mawk> that patchset is useful with secure boot, and with that I don't see how you could flip the right bit to disable the lockdown
23:36 < TJ-> mawk: it ticks all the boxes for a CTF though :)
23:36 < hexnewbie> TBotNik: To run a X program from a web page, you'd need a script on the X side to cooperate (e.g. by providing the display number through some shared memory). You could discover the display by crawling /tmp, or using whatever "who" uses to get it for the login session
23:37 < mawk> yeah but it will take ∞ hours to flag it
23:37 < DrunkRhino> Anyone have experience in getting dnsmasq to resolve hostnames to IPv6 addresses as well as IPv4? I'm trying to get a Pi 2 set up with pi-hole adblocking/dns using its DHCP server
23:37 < hexnewbie> TBotNik: It's unlikely that the web server user will have permissions to open the display, though
23:37 < Psi-Jack> mawk: I already got your flag BTW. :)
23:37  * Psi-Jack runs.
23:37 < mawk> :(
23:38 < poptix> if you're executing X programs from PHP you've already failed
23:38 < poptix> may as well 'xhost ++'
23:38 < hexnewbie> poptix: It's more common than you think. But gedit is a bit much :)
23:40 < poptix> he'd probably be better of adding a url handler for gedit://
23:40 < poptix> and letting the browser execute gedit.
23:40 < poptix> i didn't read his writeup though, maybe it's a different box.
23:41  * hexnewbie is trying to figure which of the two is a bigger security risk
23:42 < hexnewbie> I think I'll prefer my web server messing with my display directly, over creating an URL handler that allows all of Internet to run commands on my computer (or even just open files)
23:42 < hexnewbie> Admittedly, the web server thing will probably not work due to other permission issues
23:43 < poptix> hexnewbie: you're assuming that any args would be passed, there's nothing inherently wrong with executing an application like gedit
--- Log closed Sat Apr 07 00:00:36 2018