--- Log opened Sun Jul 08 00:00:26 2018
--- Day changed Sun Jul 08 2018
00:00 < ThePortWhisperer> hm, i want to do a quick curl through a site, is there a simple 1 liner to do curl site.com/?page=x for x in 1-20
00:01 < pnbeast> ThePortWhisperer, you might like the {n..m} construct in bash.
00:01 < dfcnvt> Alarm worked at 6pm.  :)
00:01 < dfcnvt> Time to move my car.
00:01 < dfcnvt> :)
00:02 < alexey-nemovff> lol
00:05 < msiism> the LSB spec requires all scripts in /etc/profile.d/ to be read automatically whenever /etc/profile is read. why does that make sense? i mean, if you have a view global settings special to bash, you'd probably want to put them in a file "bash_profile.sh" inside /etc/profile.d/ and then tell /etc/profile to run that whenever bash is detected. but that's not LSB-compliant then.
00:05 < msiism> s/view/few
00:06 < bls> you're not supposed to have shell specific stuff there, and if you do, it's your job to make sure the right one gets loaded
00:06 < jim> for page in $(seq 1 20); do curl site.com/?page=$page ; done
00:07 < msiism> bls: ok, i see. where am i supposed to have shell-specific (global) configs then?
00:08 < bls> they can go there, you just have to ensure if you have shell specific code in them, it only gets executed when the right shell is loading the file
00:09 < Sitri> Check against $SHELL
00:09 < jim> ThePortWhisperer, for page in $(seq 1 20); do curl site.com/?page=$page ; done
00:09 < bls> case $SHELL in; /bin/bash) ;; /bin/ksh) ;; /bin/zsh) ;; /bin/tcsh) ;;
00:09 < bls> etc etc
00:11 < msiism> Sitri: $SHELL is not the right thing to check, afaik. because it refers to the user's shell as set in /etc/passwd.
00:11 < Sitri> It's good enough
00:12 < msiism> Sitri: how so?
00:12 < msiism> Sitri: if i run "ksh -l" from within bash, $SHELL will still be /bin/bash
00:12 < green-> suggestions on how to spot a TTFB bottleneck? i’m experiencing intermittent, but frequent long TTFB (>10s) … but I can’t explain it. Have htop, iotop and mtop running simultaneously, and when the TTFB spikes occur — I’m seeing nothing that would suggest it should be … load isn’t spiking, io isn’t spiking, htop isn’t showing any processes in iowait … I’m clueless.
00:13 < utack> is it sane procedure to "dd" a system ssd to a new ssd, or can that somehow end up with misalinged blocks that do not correspond to underlying 4k blocks of the flash?
00:13 < Dan39> utack: i wouldn't recommend it
00:13 < utack> ok. make fresh partitions and rsync?
00:14 < Dan39> that should do
00:14 < utack> i guess that is the sane way to do it
00:14 < Sitri> Ah, so it isn't.  I thought the shell set it, my bad
00:14 < Dan39> unless your partition scheme is a PITA, id just copy the files
00:14 < Dan39> there are backup programs that will handle partitions and stuff though
00:14 < utack> efi and ext4, that's it
00:14 < Dan39> i usually just do it manually though
00:14 < Dan39> utack: super easy then, don't even have to worry about grub-install!
00:15 < bls> and dd will be reading/writing to what would normally be empty space
00:15 < utack> kinda, efi has no option to add things, i need to live disc chroot into the installation, but it is easy
00:15 < Dan39> utack: you can transfer the efi files and all at same time too, just make sure you mkdir/mount efi partition first
00:15 < utack> i plan to do so
00:16 < Dan39> i do it all the time with tarballs
00:16 < msiism> bls, Sitri: ok, i could put a shellcheck at the top of every shell-specific "module" in /etc/profile.d/. but it seems kind of illogical to me, though. having a bash_profile.sh in there that is run no matter what and then having to check for the shell actually being bash inside of it. might be the wrong approach in the first place.
00:17 < bls> forcing it all to be in one file makes package management more difficult though
00:18 < bls> which is the whole reason for a directory full of files to exist
00:18 < msiism> bls: ok, right.
00:18 < bls> so it's a tradeoff
00:19 < Dan39> utack: don't forget with rsync to add -HAX
00:20 < utack> dirty hax for it to work? jk, i will look at the paramters properly before starting :D
00:20 < ThePortWhisperer> jim someone else showed me.. curl site.com/?page=[1-20]
00:20 < ThePortWhisperer> curl auto loops
00:20 < msiism> bls: this whole issue with package managers trying to overwrite confis has been bugging me today. imo, it's a bad habit to do that. ok, slackpkg, for example will ask you about this, iirc.
00:20 < ThePortWhisperer> or interprets that as a loop, subs 1-20
00:21 < Dan39> utack: probably be fine without them, but if you are on like fedora, the bringing the selinux labels along saves you the pain of having to do a relabel at first boot
00:21 < msiism> bls: i'll just go with te shellcheck in every shell-specific file inside profile.d/ then. it's ok.
00:21 < Dan39> and ACLs seem to rarely be used... i dunno why ¯\_(ツ)_/¯
00:21 < utack> a sounds useful, preservse everything
00:22 < bls> which is what this is <foo>.d directory thing is meant to solve. the package manager owns /etc/<foo>.conf, no package should ever touch it, and you add your extensions to /etc/<foo>.conf.d/
00:22 < Dan39> utack: -a doesnt include -HAX, as it says :p
00:22 < utack> so -aHAX?
00:22 < Dan39> yea
00:22 < Dan39> -a is the most common, that was assumed :p
00:23 < msiism> bls: ok, i see. thanks.
00:23 < bls> except for shells it was decided that other /etc/bash_profile /etc/bash_profile.d/ /etc/zsh_profile /etc/zsh_profile.d/ where you might be duplicating most posix compliant code, they'd all just share /etc/profile.d and guard their non-posix code
00:23 < jim> ThePortWhisperer, oh cool :) well anyway, that's one way to do it using a shell, maybe you need to sequence something else...
00:24 < utack> no idea, i never use rsync really
00:24 < msiism> bls, Sitri: btw, i check for bash by checking for the existance of "$BASH" ;)
00:25 < jim> ThePortWhisperer, plus, bash (and probably others) have ways where you can list things to loop over, like: for color in red green blue; do echo $color; echo or put anything else here; done
00:27 < bls> $BASH would be better, the whole $SHELL thing was from the days where it was a safe assumption that when you'd check it in profile, profile was only being loaded once, at login, when invoked based on that /etc/passwd setting
00:28 < pnbeast> Dan39, I use ACLs a lot in work environments, but never need them at home.
00:29 < Dan39> i see, makes sense.
00:29 < msiism> bls: ok, i'm aiming to have a setup that is compatible between bash, ksh, mksh at least. i'm slowly getting there...
00:29 < Dan39> i've used them, but only in rare specific use
00:30 < jinie> i use ACLs on my NAS (Odroid-HC * 4 with GlusterFS), but that's mostly since it relies on POSIX permissions, and everybody and their mothers can fake a unix user id these days
00:40 < ryouma> is there a command line i can run that will take a partition and a mountpoint and will determine if mounted in more than one place?  (what i actually want is to mount if unmounted, NOT error if already mounted, and error if mounted in more than one place or mountpoint not exist.  but this is exceeding my scripting ability atm.)
00:40 < bls> like findmnt?
00:40 < [R]> ryouma: cat /proc/mounts
00:42 < phogg> ryouma: not a single command, no, but you could make one
00:43 < ryouma> bls: nothing in findmnt man page seems to detect if something is mounted more than once
00:43 < bls> oh, you're wanting a single command to do all of it
00:43 < ryouma> or shell function yes
00:44 < ryouma> basically i want to ensure that a is mounted on b.  if not, then it should mount.  but if it is already mounted /anywhere/ then it should throw a fit.
00:45 < jim> well throwing fits isn't one of your shell's abilities :)
00:45 < ryouma> exit 2 is good enough :)
00:45 < phogg> ryouma: you can write a small amount of shell code to do that
00:45 < phogg> or your language of choice, of course
00:46 < ryouma> that is what i am having trouble with
00:46 < ryouma> hang on
00:46 < jim> you can do... echo there was a problem dave 1>&2
00:47 < jim> the 1>&2 will take anything that would have gone to stdout, and send it instead to stderr
00:48 < dfcnvt> I wonder if there's something beyond than the ordinary Operating System...Say, an AI operating system....I wonder how that work
00:48 < ryouma> on reflection i want a shell function that does this: is a mounted on b?  return silently.  else is b missing?  error.  else did i already mount a?  say that it would be mounted in more than one place and error.  else mount a on b.
00:49 < dfcnvt> Maybe, Kernel AI
00:49 < ryouma> it is to be called ensuremount
00:49 < jim> ryouma, you should have enough to do that... if not, where are you stuck?
00:50 < ryouma> finding whether it is mounted in the wrong place
00:50 < ryouma> mount(8) errors if it is already mounted, AND if the mountpoint does not exist.  this is not useful.
00:51 < ryouma> also idk what command asks whether a is mounted on b specifically
00:52 < jim> look at the output of mount
00:52 < [R]> [03:40:53] <[R]> ryouma: cat /proc/mounts
00:54 < phogg> ryouma: one moment please
01:01 < eletious> Hey guys, I'm trying to use the tty in the hopes that it helps me focus a little bit - but my nitpicky self can't help but notice that the cursor isn't blinking evenly!  It blinks, but not for the same amount of time per blink.  Does anyone know what might cause that?
01:02 < phogg> ryouma: try http://termbin.com/s5qs
01:02 < pnbeast> eletious, maybe a lack of focus?
01:02 < [R]> lol
01:03 < Khaotic> so im teaching a class on reverse engineering later this week, and just discovered the -S switch for objdump
01:04 < phogg> ryouma: cleaned up the error messages a bit: http://termbin.com/o9vw
01:04 < phogg> eletious: cursors don't blink on the console. What "tty" are you using?
01:06 < phogg> ryouma: FYI no command does what you want, not even mountpoint(1)
01:06 < ryouma> phogg: thank you.  interesting.  i was hoping i could do mountpoint (something) || mount (something) or something
01:07 < phogg> ryouma: as you can see it is more complicated than that. mountpoint(1) can tell you if a directory is a mount point, but not whether it is the mountpoint *for a specific device*.
01:08 < phogg> instead I just checked all places the given device was mounted and report an error if any of them is not what you asked for
01:08 < eletious> phogg: I'm using tty2 in ubuntu - accessed via ctrl+alt+f2
01:08 < ryouma> i will try it.  thanks again.
01:09 < phogg> technically it would be better to not require positional args but to entirely wrap mount
01:10 < [R]> don't be a fool, wrap your mount
01:10 < esselfe> [R] you're a mount
01:11 < [R]> i have the largest mount
01:11 < ryouma> you did wrap mount?
01:11 < msiism> so, to make sure your root prompt is colored red, no matter if you're running a login shell or not, you'll have to put one and the same PS1 setting into $gloabl_conf_file and ~/.bashrc, where it's not even supposed to be in the first, since it would normally be set in ~/.bash_profile and export-ed. lol.
01:12 < ryouma> i thought bashrc was run for every interactive shell.  probably tghat is incorrect.
01:12 < ryouma> but do you need prompts for non-interactive?
01:12 < ryouma> hmm i suppose you do
01:13 < kerframil> msiism: that's not correct - .bashrc is
01:13 < msiism> ryouma: well, if i use su to gain root access an suddently my prompt is only # or "bash...#", well that's kind of not what i want.
01:13 < esselfe> .bash_profile is non-interactive
01:13 < [R]> so tell su to give you a logi shell
01:13 < ryouma> most people do not use su like that
01:14 < esselfe> it's for variables like LANG and TZ
01:14 < kerframil> msiism: sorry, keyboard accident - .bashrc is the correct place to define shell options, including the likes of PS1
01:14 < msiism> esselfe: how's that? afaik, it's the pe-user config for bash login shells.
01:14 < esselfe> .bashrc has PS1 and EDITOR and such
01:14 < ryouma> can you set ps1 for a chroot?
01:14 < [R]> you set PS1 for a shell
01:14 < ryouma> or must you do it inside it?
01:14 < ryouma> btw, for a chroot, is there any reason to do "mount -t proc  none              "${mountedroot:?}"/proc || exit" instead of "mount --rbind --make-rslave "$dir" "$mountedroot/proc" || exit"?  same question for /sys.
01:15 < ryouma> i.e. create new fs instead of bind?
01:15 < [R]> ryouma: bind is probably better but it oent matter
01:15 < [R]> you gotta bind dev
01:15 < [R]> o why not just bind them all?
01:15 < phogg> I don't know. I guess that mounting twice is probably better.
01:15 < phogg> but they're about the same
01:16 < eletious> i'm still confused... why would my cursor be blinking if cursors in console don't blink
01:16 < [R]> eletious: maybe you're blinking?
01:16 < bls> they can
01:16 < eletious> oh jeez, I *am* blinking
01:16 < eletious>  /s
01:16 < aaro> eletious: mine does blink
01:17 < eletious> well at least i'm not imagining it... that's good.  Do yours blink regularly, or are some of the blinks shorter than others?
01:17 < ryouma> eletious: cursors in console do blink.  this is an accessibility nightmare.
01:17 < ayecee> depends how many drinks i've had
01:18 < ryouma> eletious: yes they blink irregularly.  you are not insane.  according to the evidence we have at least.
01:18 < ayecee> not insane because of this, anyhow
01:18 < [R]> how do you know he's not insane?
01:18 < msiism> kerframil: mywiki.wooledge.org/DotFiles, which i consider a sort of an authorative source, says that environment vars that can be handed to subprocesses should be put in .bash_profile. but then, it also says: always source .basrc at the end of .bash_profile. so if i put PS1 into .bashrc ist would still be there.
01:19 < eletious> yeah there's not a whole lot of evidence supporting the idea that I'm sane... i just don't think that should affect my cursor
01:19 < ryouma> would that lead to bashrc being loaded twice sometimes?
01:19 < ryouma> lol
01:19 < msiism> ryouma: no, not that i know.
01:19 < msiism> ryouma: it's done to pull setting from .bashrc in for login shell, if i understand it right.
01:20 < ryouma> but there are all sorts of corner cases, like non-interactive ssh or something which make me wonder
01:20 < ryouma> (might not want it at least)
01:20 < [R]> ryouma: read the bash man page... it exaplins the rules exactly for what loads when
01:20 < ryouma> reading the bash man page does have a risk of changing your cursor
01:21 < msiism> ryouma: for bash, you'd always have to "if it's not interactive, don't do anything" code block into your files.
01:21 < phogg> msiism: you should probably re-read INVOCATION in bash(1). Interactive shells read .bash_profile, period. Non-interactive read only .bashrc, not .bash_profile. Which things you put in which places depends on where you want them. Most of the time you want all non-interactive things also interactively, but not the other way around, so you define *most* env vars in .bashrc and you source it from .bash_profile.
01:21 < eletious> dang, i knew those man pages are evil
01:22 < ryouma>  info page might be marginally better
01:22 < msiism> phogg: the man page of bash on my system says: .bash_profile is read for login shells.
01:22 < phogg> msiism: you could define PS1 only in .bash_profile, but since env vars are more typically defined in .bashrc it is common to put it there. *Logically* it makes more sense to put any setting *configuring the shell* in the shell's rc file.
01:22 < kerframil> msiism: yes. further, PS1 is a shell variable, and defined as such - not exported as an environment variable. hence, it belongs in .bashrc.
01:22 < bls> and things have gotten blurred these days because of DMs/DEs that don't ensure the parent shell of your windowing session was invoked as a login shell
01:23 < phogg> bls: wait, they don't?!
01:23 < bls> so things that should only need to be set once, like environment variables, now get set every time a new pty/shell is opened
01:23 < kerframil> msiism: EDITOR is a good example of something that you would want in .profile/.bash_profile. PS1, by contrast, is typically re-defined as non-exported shell variable and only affects interactive shells. hence, .bashrc is a sensible place for it.
01:24 < msiism> kerframil: ok, makes sense.
01:25 < phogg> it gets more fun when you run multiple shells and want to share configuration that should also be set for /bin/sh
01:25 < msiism> phogg: yes, i'm trying to share system-wide aliases between bourne-compatible shell, for example.
01:25 < bls> hence the earlier discussion about $SHELL vs $0 vs $BASH
01:25 < phogg> bls: I must have missed that one
01:29 < kerframil> phogg: that's a bit muddled up. interactive login shells read .bash_profile, interactive non-login shells don't, and non-interactive shells don't actually read .bashrc at all - unless you define BASH_ENV="$HOME/.bashrc", for instance.
01:30 < kerframil> or source .bashrc from the code being executed by the non-interactive shell
01:32 < msiism> kerframil: but interactive non-login shells read .bashrc, afaik
01:32 < kerframil> msiism: that's right
01:33 < kerframil> it is pretty confusing at first
01:33 < msiism> well, i'ts still confusing to me after several days...
01:34 < ryouma> i don't even know what use cases specifically qualify for interactive non-login shell
01:34 < msiism> ryouma: using su to gain root access. i do it all the time.
01:34 < kerframil> ryouma: every time you open an terminal emulator under X, typically
01:34 < msiism> ryouma: also, x terminals
01:34 < bls> think your terminal with a bunch of tabs/screen/tmux windows
01:34 < kerframil> that's a pretty important use case
01:35 < bls> each of those is interactive, but you didn't get into it via a login
01:35 < ryouma> for my purposes i'd want an xterm and console to be the same setup in most cases
01:35 < jim> ryouma, sometimes I'll start a shell just because later I want to exit and come back to exactly where I was before I ran the the shell... that would be an interactive nonlogin shell
01:36 < kerframil> it's easier to rationalise if you think about it in terms of "what do I want to export to my environment?" on the one hand, and "how do I want to customise my (interactive) shells?" on the other
01:36 < bls> they're supposed to both be set up the same way, but some DMs/DEs don't, which is why most terminals offer the option to start the shell in them as a login shell
01:38 < bls> and why I've always been in the habit of not auto-starting X
01:39 < msiism> bls: it should be ok as long as you use startx to do it, i guess.
01:39 < bls> yeah, that's what I do
01:39 < ryouma> startx here too
01:41 < msiism> but recetly, the X server seems to have problems when you witch between tty's running X sessions a lot. i've experiecned crashes and freezes on three (very) different distros with different hardware. debian introduced a nice crash festure into their bash package in version 9.
01:41 < msiism> s/witch/switch
01:41 < msiism> s/festure/feature
01:41 < kerframil> I can't speak for other graphical login managers, but gdm will honour .profile. therefore, I define my env vars there, and also source .profile from .bash_profile (when logging in via the virtual console, bash will prefer .bash_profile over .profile)
01:42 < bls> pretty sure it was lightdm that didn't do it
01:45 < JodaZ> ext4 or xfs?
01:48 < kerframil> seeing as you asked, xfs in its modern v5 incarnation, but ext4 is still good. just don't pick xfs if you like to shrink filesystems, as it can't be done.
01:49 < msiism> kerframil: about shell vars vs. environment vars: now, man bash lists PATH as a shell variable as well. so would you define that in .bashrc as well instead of having it in /etc/profile or .bash_profile and export it?
01:49 < eletious> unrelated: I wish social media feeds had pages, and tech manuals/guides had infinite scroll
01:50 < ryouma> info can be scolled infinitely with just spc or it can be concatenated intoa  single long thing
01:50 < bls> PAGER=cat man cat
01:51 < bls> and use your terminal's scolling
01:51 < ryouma> or even man cat | cat # this will annoy some people who think the name of cat should determine its uses
01:51  * msiism recently learned that one can actually scroll back in 'more', using 'b'
01:52 < kerframil> msiism: I define it in the profile, because it's something that I want to be exported and its effect extends beyond just bash
01:52 < msiism> kerframil: ok, thanks.
01:52 < bls> same here
01:55 < kerframil> my rule of thumb is that if one is exporting it, it probably doesn't belong in .bashrc
01:57 < [R]> export kerframil
01:57 < bls> right, that's how you get a PATH like: PATH=/bin:/usr/bin:/bin:/usr/bin:/bin:/usr/bin
01:58 < msiism> i have PATH set in /etc/profile and export it there. but that might actually not be such a good idea. it's because i wante dit to be available for ksh as well. but then, if i want to use ksh, i shouls configure that right.
01:59 < bls> I generally stay away from any personal settings in /etc, easier to ensure everything is encapsulated in ~ so a setup can reproduced across systems
01:59 < kerframil> and if you do use /etc, use /etc/profile.d if that's how your distro rolls
02:00 < msiism> bls: ok, eveyone tells me to stay away fomr /etc/profile. i will do that now.
02:01 < kerframil> if you consider that /etc/profile is usually an owned file in terms of package management, it does make sense
02:02 < kerframil> unset -v '[R]'
02:02 < kerframil> just kidding
02:02 < msiism> kerframil: yes, it does.
02:04 < [R]> lol
02:09 < ThePendulum> 'lo
02:21 < pnbeast> [R], why did you drive all those mademexico bots away?
02:26 < [R]> ?
02:27 < iflema> the world needs a hero...
02:27 < iflema> how many
02:27 < iflema> lost count
02:29 < pnbeast> There were a dozen joins or more, then a few seconds later a dozen parts, all the nicks were the same.
02:30 < amrx> Those were my soliders!
02:31 < amrx> just kidding!
02:31 < amrx> missed the action!
02:37 < ThePendulum> when I hardlink a file with ln, no -s, ls -i should show the inode to be the same, no?
02:37 < ThePendulum> for some reason they're different, yet I can still delete the 'original' and the hardlinked file will work fine. not quite sure what's going on there
02:38 < [R]> ThePendulum: sure you're looking at the inode? sounds like you made a copy
02:38 < ThePendulum> well if that's what ls -i does
02:39 < ThePendulum> and that would be a quick copy
02:53 < Sitri> ThePendulum: yes, they should have the same inode
03:00 < nitrix> Hi, why does fdisk -l /dev/sdb lists my /dev/sdb1 as 1.8T and when I mount the drive, df -h displays it as 487M?
03:00 < nitrix> I tried formatting the partition as ext4 again with no luck. What am I doing wrong?
03:01 < [R]> pastebin the output of fdisk, mkfs.ext4, and df
03:01 < nitrix> https://lpaste.net/3846257799000489984
03:02 < ryouma> nitrix: maybe hardware error
03:02 < [R]> and the mkfs, and the df after the mkfs
03:04 < nitrix> https://lpaste.net/8165146263247912960
03:04 < nitrix> ^ [R]: ryouma
03:04 < nitrix> Well that fixed things.
03:04 < nitrix> I don't know what I did wrong the first time. Maybe it was still mounted and I didn't notice.
03:04 < [R]> lol
03:05 < ThePendulum> Sitri: really weird then, they behave like hard copies but the inode just doesn't match
03:05 < ThePendulum> bit worried proceeding with this
03:05 < nitrix> 9 more drives to go :)
03:05 < Sitri> They're on the same partition?
03:05 < bls> a paste of what commands you're running and the outputs you're actually seeing would be useful to make sure you're not misreading something
03:13 < lostfile> any one here still use gopher im working on a ubuntu bucktooth server
03:14 < nitrix> I should just be able to paste the most recent lines of /proc/mounts into /etc/fstab, right?
03:14 < ryouma> i doubt it
03:14 < ryouma> mine says fusectl /sys/fs/fuse/connections fusectl rw,relatime 0 0
03:14 < nitrix> Oh. Well. Oh. Yeah, it's not responding to ping anymore. Rescue mode time x]
03:14 < ryouma> which, i have no clue
03:14 < lostfile> im thinking about running it inside of docker but i dont think thats possible
03:15 < nitrix> /dev/sdb1 /mnt/drive_b ext4 rw,relatime,data=ordered 0 0
03:15 < nitrix> /dev/sdc1 /mnt/drive_c ext4 rw,relatime,data=ordered 0 0
03:15 < nitrix> /dev/sdd1 /mnt/drive_d ext4 rw,relatime,data=ordered 0 0
03:15 < nitrix> ryouma: ^
03:16 < ryouma> nitrix: sdx notation is unreliable.  try uuid.  also, /mnt is for temporary use.
03:16 < [R]> ryouma: you're unreliable
03:16 < ryouma> also perhaps relatime is default
03:16 < nitrix> It's definitely not booting now ;-;
03:17 < nitrix> Nevermind! It's on!
03:18 < jim> so it booted?
03:18 < [R]> like donkey kong?
03:18 < nitrix> Yep. So what was the recommendation anyway? I'd rather do things cleany.
03:18 < nitrix> Using the drives UUID?
03:20 < [R]> using the filesystem's uuid...
03:22 < jim> nitrix, you might have noticed this before, but your sda3 is 1.7t
03:23 < nitrix> Mhm :)
03:23 < jim> or at least it was around the time you first joined
03:23 < jim> question... had you ever heard of LVM?
03:23 < nitrix> sda1 through sdj1 will be 1.7T :)
03:24 < nitrix> jim: Tell me more
03:24 < jim> nitrix, with LVM, instead of putting filesystems in partitions directly, you put 'LVM physical volumes' in partitions, then you make a 'volume group' (which is just a list of physical volumes), and put physical volumes in the volume groups, then you can make 'logical volumes', and these are allocated out of the volume groups
03:25 < jim> and it's the logical volumes that you would format and mount
03:26 < jim> it's essentially another way of organizing your storage
03:26 < ThePendulum> Sitri: yep, same partition
03:27 < nitrix> jim: What happens if one of the physical drives fail?
03:27 < jim> it depends on what you put the lvm physical volumes on...
03:28 < nitrix> I will look into it in the future. This is for a server and I'd rather keep it to what I'm familiar in case something goes wrong.
03:29 < jim> I guess you could put a large pv on a raid array, then failing of a drive would have whatever consequences on the raid array
03:29 < nitrix> I'm guessing LVM is strongly encouraged?
03:29 < jim> not necessasarilly
03:29 < jim> you can do it if you want to
03:29 < bls> nitrix: a lot of things that are difficult with just partitions and FSs become easier with LVM
03:29 < ryouma> if you are new i would say do not do it
03:30 < ryouma> tehre are corner cases
03:30 < nitrix> ryouma: Thank you.
03:30 < jim> but I like it, because of how it's arranged, and because it's been very solid for a lot of years
03:30 < nitrix> I'll go without LVM for this server, but try it on my laptop :)
03:30 < bls> if your needs are simple, it's just a layer of redirection, but it's also the default on most distros these days, so by avoiding it you're making yourself less common and more difficult to support
03:31 < ryouma> 3 sets of commands to learn, corner cases to learn, and configuration files to set to take account of it
03:31 < ryouma> but good poitn if it makes you more unusual
03:32 < ryouma> less*
03:33 < ryouma> i get confused enough without adding layers
03:34 < jim> ryouma, generally with layering, you can work on the layer you want to, and it is likely less complex than having to look at the whole thing at once
03:36 < ryouma> jim: i still recommend to new people to avoid them.  the point is not merely to solve a problem at a layer, but to understand what goes where.  you might not know that this layer belongs in that file.  if there are fewer layers, then you have fewer things to choose from to go into the files.
03:36 < ryouma> for example
03:37 < jim> specifically for lvm, I knew the basics, then all I had to do if I wanted to know all the volume group commands, I would type vg, leave the cursor at the end, and hit tab a couple times :)
03:38 < ryouma> dunno, i found in instructions for dealing with various issues that ther are special cases for lvm
03:39 < ryouma> if you use it allt he time it should be no problem persumably
03:39 < ryouma> re
03:40 < pnbeast> I used it, once.  I lost 15 pounds, my hair grew thicker and my smile whitened overnight.  I recommend it.
03:40 < ryouma> that's because you rolled around in the grease and you drank the bleach.  i do not recommend it.
03:41 < jim> I'll have to admit I used to get nervous about what happens if you have a vg that spans several drives, and the way I put that to rest, is only have a vg cover one drive... worked like a champ :)
03:42 < jim> so at that time I used lvm as a way of getting around the 4-partition limit
03:42 < pnbeast> jim, did your hair grow thicker?  Tell the truth.
03:42 < ryouma> what is wrong with extended partitions?
03:42 < jim> oh yeah :)
03:42 < lostfile> why do you need more then 4
03:43 < jim> I don't think it was all about lvm though :)
03:43 < lostfile> oh
03:44 < jim> lostfile, when I would build a system, I typically had separate /var, /usr, /usr/src, /tmp, /home and /
03:45 < jim> because I was trying to protect / from filling up
03:46 < lostfile> i typically just go with the auto partition thing or i would just make one partition with a swap
03:46 < pnbeast> I just tell all my users that I'm going to email their mothers the contents of their "private" directory if the disk fills up.  I've never had a problem.
03:47 < lostfile> well i make 3
03:47 < lostfile> my pc uses efi
03:47 < lostfile> or whats that boot thingy called again
03:49 < jim> I had this one lab assistant where I was lab manager, one time he filled the drive which was used by maybe 2000 students with mp3s... so I moved all the lab assistant's home dirs to a smaller and separate partition
03:51 < lostfile> mp3s how much music where these kids stealing
03:51 < jim> there was a program called pighunt going around at the time, so before moving their home dirs, I would periodically run pighunt and email the results to all the lab assistants
03:51 < ryouma> when i compare the contents of initrd files (that started out identical and then i ran grub-install, update-grub, and update-initramfs), i find that ./lib/x86_64-linux-gnu/libgcrypt.so.20 have different contents.  what could account for this?
03:52 < ziggylazer> anyone have a clue what cat is in windows?
03:52 < kerframil> ziggylazer: type
03:52 < ziggylazer> Thank you
03:52 < jim> ziggylazer, type
03:52 < bls> has libgcypt been updated since the last time your initrd was built?
03:53 < jim> but I dunno if type is a direct substitution, unless you want to type out one file
03:53 < ryouma> bls: nope, i rsync over /boot, then do the commands in a chroot, and voila, that file is different (as are the few expected files)
03:53 < ryouma> it is the only file that does not make sense to be different
03:56 < ryouma> basically i start with identical initrd, then run those 3 commands, and that file differs as follows (forgive 4 line spam)
03:56 < ryouma> --rw-r--r--,root,root,49        ./conf/conf.d/resume
03:56 < ryouma> +-rw-r--r--,root,root,30        ./conf/conf.d/resume
03:56 < ryouma> --rw-r--r--,root,root,920000    ./lib/x86_64-linux-gnu/libgcrypt.so.20
03:56 < ryouma> +-rw-r--r--,root,root,924096    ./lib/x86_64-linux-gnu/libgcrypt.so.20
03:57 < ryouma> resume is expected to differ.  but libgcrypt is not.
03:57 < ryouma> i disassembled both sides of it and the diff goes on forever
03:58 < ryouma> (i.e. quite different files)
03:58 < bls> checked your initrd generation scripts? they could be doing something with randomization for security purposes
03:58 < ryouma> ah, like a hardcoded seed or something?  i have no clue on debian jessie where to look for that.
03:59 < bls> yeah, that or function address randomization
03:59 < ryouma> no, not a seed, that's silly.  i guess you suggest reordering functions or something.
04:02 < ElinKattunge> Where can I get a minimalistic browser alternative to Suckless Surf?
04:02 < ElinKattunge> It's just Surf seems to leak.
04:02 < ryouma> bls: that is at least plausible, though idk why it would be useful.  the only other plausible thing would be some attack already peformed on update-initramfs.
04:04 < ryouma> if anybody wants to confirm this anomaly, all you have to do is "cpio -i -d -H newc --no-absolute-filenames" on your initrd file after running update-initramfs, update-grub, and grub-install on an identical copy of it.
04:05 < ryouma> then do a sha
04:05 < ryouma> (debian)
04:06 < ryouma> (do this in a chroot on the place to which you copied it)
04:06 < ElinKattunge> Oh, I love chroot's, they are really fun!
04:07 < bls> maybe as a variant of https://en.wikipedia.org/wiki/Address_space_layout_randomization but just for the crypto routines
04:08 < ryouma> bls: yeah i was trying to look that up.  i just got a bunch of pdfs by cs depts.
04:08 < ryouma> side channel attack whatever that means
04:08 < storge> ElinKattunge: i like luakit but i think it's unsupported for years now
04:08 < ElinKattunge> How to prevent escaping from a chroot jail?
04:08 < Sitri> ElinKattunge: uzbl, dillo (HTTP and HTML only, and CSS I think)
04:08 < ElinKattunge> storge: That is the issue, even xombrero is dead
04:08 < ElinKattunge> dillo isn't minimal enough
04:09 < ElinKattunge> Surf is as minimal as I want to go
04:09 < storge> uzbl migt be the only other option
04:09 < ElinKattunge> I'll take a look
04:09 < justsomeguy> Are we talking about vim like web browsers?
04:09 < Sitri> <ElinKattunge> Where can I get a minimalistic browser alternative to Suckless Surf?
04:09 < justsomeguy> ... because Qutebrowser is amazing, and everyone should try it.
04:10 < ElinKattunge> Interesting.
04:10 < ryouma> bls: but a search for +libgcrypt Address space layout randomization does not yield anything apparent
04:10 < ryouma> and you'd think it would be mentioned
04:10 < ElinKattunge> I tried it, didn't work, I'll try it again, give you feedback on logs and stuff.
04:10 < justsomeguy> Thanks for the context, Sitri
04:10 < ryouma> justsomeguy: does it do js?
04:11 < ElinKattunge> justsomeguy: All I get is a download button
04:11 < justsomeguy> ryouma: Sure does. It's actually actively maintained. ...and rapidly developed, even.
04:11 < ryouma> emacs has 3 web browsers.  none do js, but all should work.
04:11 < ryouma> justsomeguy: great
04:11 < bls> it's just a wrapper around webkit
04:12 < justsomeguy> ElinKattunge: What distro are you on, and how are you trying to install it?
04:12 < ryouma> ok yeah people are working on that for emacs i think
04:12 < pnbeast> Geez, I remember the good, ol' days when we were *happy* if our text editors had only two web browsers.
04:12 < ElinKattunge> Ubuntu
04:12 < ryouma> only two are in common use, but w3 was useful for people who do emacspeak several years ago
04:13 < ElinKattunge> I just did apt-get install qutebrowser
04:13 < ryouma> the commonly used ones are emacs-w3m and eww
04:13 < ElinKattunge> I'll run in terminal, see what the output is
04:13 < ryouma> emacspeak is for accessibility
04:14 < ElinKattunge> Weird, it doesn't work well when loaded from the dmenu.
04:14 < justsomeguy> What do you mean, exactly? What happens?
04:15 < ElinKattunge> dmenu is my app launcher
04:15 < jim> lynx and netscape?
04:15 < ElinKattunge> From the cli, qutebrowser works fine, loading it from demu, it breaks it.
04:15 < justsomeguy> Hmm...
04:15 < justsomeguy> But what is broken about it?
04:16 < justsomeguy> Are you having graphical issues? Something else? Error output?
04:17 < ElinKattunge> It just displays a blue download button in the browser session when I run qutebrowser from the dmenu.
04:18 < ElinKattunge> Weird
04:20 < justsomeguy> Can you navigate to another url?
04:21 < ElinKattunge> Nope
04:21 < ElinKattunge> If I run from cli, all works fine.
04:27 < eblip> hey i run gentoo, but i have another machine that i want another real reliable stable minimal linux distro that is really fast to install and update and has modern packages...
04:27 < eblip> i would use arch linux but i cant afford to have this thing break on me
04:28 < eblip> gentoo is great but i need a machine i dont have to rebuild for dev purposes doing django and python and postgresql stuff
04:28 < eblip> i was thinking opensuse leap...minimal x windows install
04:28 < esselfe> try openSUSE
04:29 < eblip> anyone else using thais
04:29 < eblip> nice esselfe
04:30 < eblip> maybe ill run with it...its a departure from my normal gentoo..but in this case i need reliability and prett recent packages and no messing about rapid environment set up
04:31 < [R]> what does "pretty recent packages" mean
04:31 < storge> newer than 2010
04:32 < notmike> Might as well use Debian if you're considering OpenSUSE
04:33 < [R]> lol
04:42 < dannylee> <>;<>
04:42 < lessthan0> what is the command to print domain or print hostname to the terminal?
04:43 < lessthan0> I see a few different ones on google
04:43 < lessthan0> uname -n
04:43 < lessthan0> $hostname
04:43 < fryguy> why are you asking a question you know the answer to? what is wrong with the answers provided by google?
04:43 < lessthan0> they are not for debian
04:44 < lessthan0> and I was wondering why use echo and why single quotes
04:44 < fryguy> why not?
04:44 < lessthan0> and why use one over the other
04:44 < lessthan0> I see $(hostname) but I don't think debian uses ()
04:44 < lessthan0> thats why I ask here
04:45 < lessthan0> and it does not say if uname -n is hostname or domain name
04:45 < [R]> $() is standard shell syntax...
04:46 < lessthan0> but if I echo $PATH no ()
04:46 < [R]> thats a variable
04:46 < lessthan0> ok
04:46 < [R]> hostname is a command
04:46 < lessthan0> so $() is always a command?
04:46 < [R]> man bash
04:47 < lessthan0> yeah I have an actual book on bash
04:47 < lessthan0> it didn't cover this
04:47 < lessthan0> old book maybe thats why
04:48 < triceratux> lessthan0: hostname is a command. $(hostname) is the result of that command. debian uses ()
04:49 < lessthan0> I am pretty sure I setup my hostname with .home suffix but echo $(hostname) returns without .home suffix
04:49 < lessthan0> what is the easiest way to change it?
04:50 < lessthan0> I see I can edit the config but maybe there is an easy command?
04:51 < lessthan0> shit I think I put my hostname in my domain name
04:52 < lessthan0> my domain is (none)
04:54 < pnbeast> Don't do that!  If someone looks up your hostname, their DNS resolver could recurse infinitely and it'll take down the entire DNS system, and then the internet!  An angry world will blame you, lessthan0.
04:54 < dannylee> fedora has a really great terminal..just like redhat and centOS...but i will switch too ubuntu gnome if my computer gets sick...i just really hate doing updates...fdora has a great terminal 8-)
04:54 < fryguy> 1. why does this matter.  2. why are you doing echo $(hostname) instead of hostname
04:54 < ElinKattunge> I really like qutebrowser!
04:56 < dannylee> ill just wait for fedora 29 or 30 comes out
04:57 < lessthan0> pnbeast yes I think that happened already
04:58 < lessthan0> the device name in gnome had fios-router
04:58 < lessthan0> echo $(domainname) returns (none)
04:58 < lessthan0> I fixed it in gnome prefs gui
04:59 < lessthan0> well I fixed hostname by fixing device name in the gui
04:59 < lessthan0> domain is still (none)
04:59 < ElinKattunge> pnbeast: That sounds over dramatic to me.
05:00 < lessthan0> I have man domainname
05:00 < pnbeast> ElinKattunge, really?  I guess you read it perfectly, then.
05:01 < lessthan0> echo $(dnsdomainname) returns \n
05:08 < Tech_8> hi
05:08 < [R]> so anyone ever use intel optane... the concept seems interesting
05:08 < Disconsented> Which one?
05:09 < esselfe> [R] you're a concept
05:09 < [R]> i'm abstract
05:12 < lessthan0> it is mostly not the best option
05:13 < lessthan0> they are cheap but shitty
05:13 < [R]> how so?
05:13 < lessthan0> if your machine was maxed out on ram but you have an available NVME slot that you will never use then you can put optane instead of swap partition
05:14 < lessthan0> like a machine that supports 64GB maximum DDR3
05:14 < lessthan0> or maybe a laptop with one DDR3L slot max 16GB
05:14 < Disconsented> Eh, you tend to loose a PCIe slot then
05:14 < lessthan0> yes
05:14 < Disconsented> Which you can throw an HBA in
05:14 < lessthan0> so basically optane is a horrible idea
05:14 < Disconsented> Optane memory is
05:15 < [R]> well im in the market for a new laptop, i figure if it comes with a regular drive and an optane, i'll leave it alone... instead of putting in an ssd
05:15 < Disconsented> but the 900p/905p are amazing
05:15 < lessthan0> the optane can be either sorta hybrid spinning HD enhancement
05:15 < luke-jr> my machine has way more RAM slots than PCI-e..\
05:15 < lessthan0> or L5 ram if you consider real ram L4
05:15 < luke-jr> I wish I could use the RAM slots for persistent storage
05:16 < Disconsented> Intel is trying to do that with optane
05:16 < [R]> i was planniing on doing bcache with it
05:16 < luke-jr> Disconsented: but does it work in standard RAM slots?
05:16 < lessthan0> you can also use optane for the sleep states that need non-volatile storage
05:16 < lessthan0> intel rapid start I think its called
05:16 < Disconsented> And NVRAM is a thing even if it is really expensive
05:16 < Disconsented> luke-jr> https://www.extremetech.com/extreme/270270-intel-announces-new-optane-dc-persistent-memory
05:16 < luke-jr> Disconsented: I thought they require non-standard RAM slots
05:16 < lessthan0> so it copies RAM to the optane and goes into deep sleep
05:17 < [R]> so suspend to disk...
05:17 < lessthan0> but better with optane
05:17 < lessthan0> it works fine on an ssd too
05:17 < lessthan0> but without the ssd you would be kinda eh
05:17 < Disconsented> Standard slots but the motherboard needs to support it luke-jr
05:17 < Disconsented> Which means server boards only
05:17 < lessthan0> you usually pickup a laptop when you close it
05:18 < luke-jr> Disconsented: my motherboard is open source..
05:18 < Disconsented> Thats a thing?
05:18 < luke-jr> at least the firmware is
05:19 < Disconsented> Figured
05:19 < luke-jr> except for the Ethernet and SAS stuff
05:19 < lessthan0> I would be skeptical about mixing optane with linux. I could be wrong.
05:19 < luke-jr> but RAM shouldn't care about that
05:19 < [R]> richard stallman would be dissapointed
05:19 < [R]> lessthan0: how so?
05:19 < luke-jr> [R]: Stallman doesn't care about free hardware
05:20 < lessthan0> intel will directly work with microsoft and motherboard vendors
05:20 < [R]> and?
05:20 < lessthan0> my guess is that linux foundation will lag behind if intel does not put extra effort into them
05:20 < luke-jr> lessthan0: will they? My motherboard isn't Intel-based.
05:21 < lessthan0> microsoft is on the board of the linux foundation now
05:21 < [R]> lessthan0: no clue what you are rambling about
05:21 < [R]> but okay
05:21 < lessthan0> conflict of interest
05:21 < lessthan0> NVME is less supported on AMD then it is on intel chipsets in windows at the moment
05:22 < lessthan0> because intel is involved in setting the standards for NVME and optane
05:22 < naan_> um I somehow have an ssh session still open even though no terminal is open
05:22 < lessthan0> try using optane on an AMD cpu/chipset
05:22 < naan_> I must have not quit gracefully.. how do I close it now ?
05:23 < lessthan0> kill it
05:23 < luke-jr> lessthan0: yeah, whcih means it's even less likely to work on IBM :/
05:23 < naan_> ok
05:23 < lessthan0> sigterm sigkill
05:23 < bls> pgrep -fl ssh, and then kill it
05:23 < Disconsented> lessthan0> Less supported in what way?
05:23 < bls> and don't use sigkill
05:23 < lessthan0> some people simply could not get NVME running on some AMD cpu/mobo combos
05:23 < triceratux> https://www.maketecheasier.com/kill-command-in-linux/
05:24 < lessthan0> it could be fixed now
05:24 < Disconsented> It is
05:24 < Disconsented> The contary is true
05:24 < naan_> ok that worked thank you lessthan0 and bls
05:24 < Disconsented> Espically on HEDT
05:24  * lessthan0 googles HEDT
05:24 < Disconsented> High End Desktop
05:24 < Disconsented> TR4/LGA 2066 (X399/X299)
05:24 < luke-jr> Disconsented: even the RAM form factor ones?
05:25 < Disconsented> luke-jr> Those arnt out
05:25 < lessthan0> 2066 is gonna die soon
05:25 < luke-jr> o
05:25 < lessthan0> 3647 replaced 2011-3
05:25 < Disconsented> Nah its got another gen in it
05:25 < lessthan0> 2066 is like 1150
05:26 < luke-jr> I have like 32 DDR4 slots, but only 5 PCI-e and 4 SATA :/
05:26 < lessthan0> just a few years and then *poof* gone
05:26 < Disconsented> TR4 gives you enough PCIe lanes to run 4x NVMe drives in RAID and still have plenty of lanes left over as well as not giving two shits what drives you do use
05:26 < lessthan0> 32 DDR4 orly?
05:26 < Disconsented> luke-jr> Sounds like its intended to be used with a SAN
05:27 < luke-jr> sorry, 16
05:27 < lessthan0> I like 7800X
05:27 < lessthan0> it has really good AVX512 performance
05:27 < lessthan0> 2 per core x 6 cores
05:27 < lessthan0> fast clock
05:27 < luke-jr> https://www.raptorcs.com/content/TL2MB1/intro.html
05:27 < luke-jr> lessthan0: lol
05:28 < Disconsented> 7800x is only useful for AVX
05:28 < luke-jr> lessthan0: I have dual 8-core SMT4 CPUs (so 64 threads)
05:31  * luke-jr ponders if PCI-e 4.0 slots can be split out to multiple PCI-e 3.0 slots.
05:31 < Disconsented> Yes
05:31 < luke-jr> since no cards seem to actully support 4.0
05:31 < Disconsented> Thats already done on AM4 through the chipset
05:31 < Disconsented> We wont have PCIe4 cards until we have the 7nm GPU's at minimum
05:31 < Disconsented> likley a generation after
05:31 < lessthan0> luke-jr this is very interesting
05:32 < lessthan0> the way you can work with microcode and verify it
05:32 < lessthan0> this excites me
05:32 < lessthan0> but wikipedia says max is 24 cores per smt4 cpu
05:32 < luke-jr> it doesn't really have microcode in the Intel sense
05:32 < luke-jr> I think most cores sold atm is 22-core
05:32 < lessthan0> I like that you can have an easy time doing your own bios
05:33 < Disconsented> 32 on the EPYC 7601
05:33 < Disconsented> Intel has a 28c part called... ?
05:33 < luke-jr> Disconsented: that's SMT1 though, isn't it?
05:33 < lessthan0> yes and 64 or 72 on intel coprocessor
05:33 < lessthan0> 54 cores for $300 on ebay
05:33 < Disconsented> 64t/56t respectivley
05:33 < Disconsented> Thats in a single socket
05:34 < lessthan0> intel coprocessor has no HT
05:34 < luke-jr> 16-core POWER9 is 64 threads
05:34 < lessthan0> it has 64C
05:34 < Disconsented> im not talk about a coprocessor
05:34 < luke-jr> that 22-core is 88 threads
05:34 < luke-jr> and they're not just HT either
05:34 < Disconsented> https://ark.intel.com/products/120508/Intel-Xeon-Platinum-8176-Processor-38_5M-Cache-2_10-GHz that thing
05:34 < lessthan0> I wonder if power9 is vulnerable to the HT attack that runs in VM?
05:34 < luke-jr> lessthan0: Spectre you mean? no
05:34 < lessthan0> the cache peaking attach?
05:34 < lessthan0> not spectre
05:35 < lessthan0> that is speculative execution side channel attack
05:35 < lessthan0> this is a VM attack
05:35 < [R]> Disconsented: $8700!?
05:35 < Disconsented> Power 9 _is_ vulnerable to specter
05:35 < luke-jr> Disconsented: no, it isn't.
05:35 < lessthan0> interesting
05:35 < Disconsented> IBM seems to think it is
05:35 < luke-jr> Disconsented: pre-production CPUs were
05:35 < Disconsented> All CPU's are
05:35 < luke-jr> they fixed itfor the final
05:35 < Disconsented> _All_
05:36 < lessthan0> you can also change things around so much that it won't be vulnerable
05:36 < Disconsented> Unless its a new model it is vulnerable
05:36 < luke-jr> Disconsented: they disabled the problematic cache
05:36 < lessthan0> if you can change the way command run and disable some commands then you are good
05:36 < luke-jr> it wasn't as vulnerable as Intel, so the cost isn't as bad
05:36 < Disconsented> So it is still vulnerable
05:36 < luke-jr> nope, it ships fixed
05:36 < Disconsented> Disabling isnt fixing
05:36 < lessthan0> intel just disables prefetching and speculative execution
05:37 < Disconsented> Nobody has a silicon fix yet
05:37 < Disconsented> Thats what i mean by a fix
05:37 < lessthan0> I think if something is impossible to fix in software then it is "vulnerable"
05:37 < luke-jr> as shipped, POWER9 is not vulnerable to Spectre
05:38 < lessthan0> is IBM cell-be dead?
05:38 < luke-jr> ?
05:38 < lessthan0> can I buy some old cell be on ebay?
05:38 < lessthan0> to play with?
05:39 < lessthan0> it was supposed to be the new cluster architechure in 2003
05:39 < lessthan0> there were supercomputers on it
05:39 < lessthan0> ps3 was on it
05:40 < lessthan0> this is why they had gigabit lan on the ps3
05:40 < lessthan0> it was a 8 core mini cluster you could scale through the lan
05:41 < lessthan0> can't trust ps3 on ebay cause they bricked linux after the software update
05:41 < jbit> leachim6: cell is dead, you can buy "spursengine" PCI cards that have the SPUs of the cell on them
05:41 < jbit> err
05:41 < jbit> lessthan0
05:41 < lessthan0> no desktops or rack servers?
05:42 < naan_> I'm not sure how ngrok will behave on my remote machine if the internet connection drops
05:42 < naan_> I want it to be able to reconnect
05:42 < jbit> lessthan0: there were some ibm bladeservers with cell processors in, if you find one for sale let me know :D
05:43 < lessthan0> I would only buy it if I can buy two of them to cluster
05:43 < lessthan0> that is the whole point
05:44 < luke-jr> didn't they hack PS3 to run Linux native after Sony screwed the official port over?
05:44 < lessthan0> yes they did
05:44 < Disconsented> It was enabled on the first generation of the console, it was locked down via firmware later
05:44 < lessthan0> but you can't jailbreak them after the latest update unless you already have the signed keys from before
05:44 < lessthan0> when you stored the keys
05:45 < lessthan0> or you need to mod chip it just to run linux
05:45 < luke-jr> :/
05:45 < lessthan0> if you got in when it was 3 of 8 cores running you can escalate privileges
05:45 < lessthan0> then use all 8 cores
05:45 < lessthan0> and control firmware from there
05:46 < lessthan0> it was puppy linux officially supported in collaboration with sony
05:46 < jbit> lessthan0: https://www.ebay.com/itm/370886660266 oh wow, there are actually some available :D
05:46 < lessthan0> then users were dumping blu ray keys for the drive and each disk / movie under linux
05:47 < lessthan0> it is in the official bluray agreement and the HDCP agreement to blacklist the PS3
05:47 < lessthan0> and sony lost a lawsuit
05:47 < lessthan0> so they were going to lose linux or blueray + the store so guess what was more important? haha
05:48 < lessthan0> jbit thanks
05:48 < lessthan0> thats kinda cheap
05:48 < lessthan0> I would need a blade chassis
05:49 < jbit> yeah, god knows how you'd actually use that thing...
05:49 < lessthan0> why is that?
05:49 < lessthan0> you use it for HPC
05:49 < dysfigured> is there a way to get workspace/desktop info like names etc without trying to parse xprop
05:49 < lessthan0> parallel compute on clusters
05:50 < dysfigured> like where does xprop get its info from?
05:50 < [R]> dysfigured: xlib
05:50 < jbit> lessthan0: oh no... i just meant... i have no idea what blade chassis it'd use, what bootloader, if mainline linux supports it, etc :)
05:50 < lessthan0> it is ancient I know but it is interesting as a museam piece
05:51 < lessthan0> you can track this stuff down as far as the distro
05:51 < jbit> lessthan0: if i were still in europe i'd be tempted to buy one of those myself :)
05:51 < lessthan0> not sure about the chassis and the bios
05:51 < jbit> lessthan0: i used to develop on cell processors for a living, so kind of nostalgic
05:51 < lessthan0> oh cool!
05:52 < lessthan0> puppy linux got me reading about HPC for the first time
05:52 < lessthan0> it blew my mind
05:52 < lessthan0> the IBM machine the size of a basketball court
05:53 < dysfigured> [R]: so, can't like parse from /proc or something
05:53 < bls> dysfigured: proc has nothing to do with X primitives
05:53 < [R]> bls: you're primitive
05:54 < jbit> lessthan0: i'm curious if the powerxcell-8i has an upgraded POWER cpu core, the POWER core on the cell-be used in the ps3 was kind of lousy
05:54 < dysfigured> i just wanted to make a small script that displays workspace info
05:55 < dysfigured> only reliable way i can seem to figure out is parsing from xprop
05:56 < bls> that's likely poing to be the best way, that stuff wasn't really intended to be used for parsing/scripting
05:56 < luke-jr> xprop is open source. but if you want a script, parsing from xprop is probably the way to go
05:59 < lessthan0> jbit I would do it for performance but I would do it for proof of concept to learn and look
05:59 < lessthan0> *not for performance I mean
06:00 < jbit> lessthan0: well the SPUs were super crazy fast :
06:01 < jbit> lessthan0: (each sell is made up of eight high-speed-but-limited SPU units, and one low-speed-but-compatible PPU unit that usually runs the operating system)
06:04 < jbit> lessthan0: linux would run on the PPU (Which is POWER/PowerPC), and distribute tasks to the SPU units using a framework like SPURS or something. the idea is that the SPUs should be maxed out at 100% all the time, with the PPU dealing with network io/disk io/etc. mainframes have a similar layout with "io processors" and channels
06:06 < lessthan0> so you would not need to set cpu usage to %99 then
06:06 < lessthan0> just a different way of doing things
06:07 < lessthan0> the intel coprocessor runs linux inside
06:08 < lessthan0> kinda ironic
06:08 < lessthan0> you can telnet in
06:08 < [R]> don't ya think
06:08 < lessthan0> your main cpu usage is %0
06:09 < lessthan0> it has different ways to pass messages and access IO
06:09 < lessthan0> you can even distribute loads with the system CPU and the pcie cpu shared resources
06:10 < lessthan0> if you wanted to use all the cpu for example
06:10 < lessthan0> but not all uses cases would benefit
06:10 < oerheks> why running linux as it seems already inside?
06:12 < vimal2012> I want to change a user's login shell from /sbin/nologin to /bin/bash.  How to do that?  Can I directly edit /etc/passwd?  Or should I run some commands?
06:12 < bls> vimal2012: chsh
06:12 < vimal2012> What will happen if I directly edit /etc/passwd?
06:13 < bls> maybe nothing, maybe you screw up and lock yourself out
06:13 < [R]> vimal2012: your computer will exploe
06:13 < esselfe> vimal2012: everyuthing will lock up
06:13 < luke-jr> vimal2012: what bls said is correct
06:14 < lessthan0> oerheks you can have the coprocessor run %100 independant from the main system
06:14 < lessthan0> it has an independant kernel
06:16 < lessthan0> the advantage is compiling something to run on xeon phi as is the main cpu IS xeon phi
06:17 < lessthan0> *as if
06:17 < lessthan0> so you could just compile without all the extra API stuff
06:17 < lessthan0> there was a desktop xeon phi that would run everything native for testing
06:18 < jbit> lessthan0: there were some compiler hacks for Cell later in its life that let you compile "normal code" and run it on the SPUs
06:18 < lessthan0> then you deploy to the cloud that has the pcie coprocessors
06:18 < lessthan0> and never need to compile again when you deploy
06:18 < lessthan0> you run the same binary
06:18 < jbit> lessthan0: it seemed pretty silly to do so though, since usually you'd get better performance just running the code on a xeon or something...
06:19 < lessthan0> if it works on the phi desktop it just works on the cloud
06:19 < lessthan0> using the api means testing with a xeon and a phi compatible motherboard 64 bit BAR PCIe
06:20 < lessthan0> so you need a xeon cpu and a phi pcie card
06:20 < lessthan0> the code has to offload all avx512 to the card or it would not run
06:21 < lessthan0> at that time the phi was avx512 and the xeon were avx2
06:21 < lessthan0> and you needed to learn the API + the phi library stuff
06:21 < lessthan0> easier to just learn the phi lib stuff and be done with it
06:22 < lessthan0> there is openMP lib if you want free free
06:22 < lord|> it would be nice if I was rich enough to be able to throw 72 cores and 512 bit SIMD at problems
06:23 < lessthan0> its $300 for 52 core phi
06:23 < lessthan0> just get the A model
06:23 < lessthan0> not the P. P is passive
06:23 < lessthan0> you need a rack server that pushes air through pci slots
06:23 < lord|> 52 cores clearly is not enough
06:24 < lessthan0> not sure if serious
06:24 < doug16k> really cores? or logical processors? PHI is 4 way SMT
06:24 < lessthan0> not HT
06:24 < matsaman> I wonder what kind of useful information we've actually ever acquired from such processing power? Full-res weather models, maybe?
06:24 < lessthan0> only slow cores with big fat AVX512
06:24 < lord|> matsaman: mostly just AI
06:24 < lessthan0> 1GHZ 52 cores
06:25 < doug16k> so it has 208 logical processors?
06:25 < matsaman> AI? We don't even have AI anymore. The term is so popular it just means heuristic anymore
06:25 < lord|> not all problems can be easily parallelized
06:25 < lessthan0> AI = software
06:25 < matsaman> "people" "think" facebook is an "AI" these days
06:25 < matsaman> it's ridiculous
06:25 < lessthan0> it is a misnomer
06:25 < matsaman> yeah
06:25 < lord|> lessthan0: actually, AI is humans
06:25 < lessthan0> haha
06:25 < lord|> lessthan0: https://www.theguardian.com/technology/2018/jul/06/artificial-intelligence-ai-humans-bots-tech-companies
06:26 < lessthan0> mturk = himans
06:26 < lord|> AI startups are literally secretly using humans
06:26 < lessthan0> humans
06:26 < lord|> before a real solution is developed
06:26 < misternumberone> hi, I have a realtek RTL81111/8168/8411 ethernet network adapter and I need to use it. however when i run a ubuntu 17.10 live image, the network adapter is not working. when i found an identical problem documented online the suggested solution is to use a realtek provided driver. however I do not want to use a closed source driver.
06:26 < lessthan0> yes exactly
06:26 < matsaman> and if we really did have real AI, we'd just get annihilated
06:26 < matsaman> =P
06:26 < lessthan0> you hire amazon for AI and they hire people with mturk pay them less than a human
06:27 < lessthan0> so AI is just code for human slave
06:27 < lessthan0> or software
06:27 < misternumberone> is there any way to make it work with the open source driver even if functionality is lost? i just need it to work even if it is slow
06:27 < lessthan0> either way
06:27 < lord|> matsaman: we won't be annihalated, we'll just be put aside for the more important goal of stamp collection
06:27 < lord|> which may or may not mean annihalation
06:27 < lord|> or enslavement
06:27 < doug16k> I guess intel is continuing it's ridiculous tradition of completely changing an architecture and continuing to use the same name
06:28 < matsaman> misternumberone: what's the device?
06:28 < doug16k> its*
06:28 < matsaman> lord|: hah, not even a real AI would have the patience to wage war with the USPS
06:28 < lessthan0> if someone makes money tax it
06:29 < lessthan0> if they make a lot of money tax it more
06:29 < matsaman> I can't tell what argument you're making
06:29 < lessthan0> that sums up USPS vs amazon
06:29 < misternumberone> matsaman: the system is a ASUS K53SV main board from a laptop running in a custom enclosure. the network adapter is integrated into the board.
06:29 < lessthan0> its greed
06:29 < matsaman> misternumberone: what does lspci say the network adapter is?
06:30 < lessthan0> the more you do good in business there will be someone bigger and more powerful to rob you of everything you worked hard to build
06:30 < matsaman> doesn't matter how much the government is charged with protecting the USPS, nobody is going to use it if they don't deliver things, and they don't
06:30 < lord|> USPS reliably delivers drugs from other countries
06:30 < matsaman> that I believe
06:31 < matsaman> they just won't deliver my tax refund or presents from my mother
06:31 < lessthan0> USPS makes a shit load more money through ecommerce than before ecommerce
06:31 < matsaman> they make their money by selling your address to other people
06:31 < lessthan0> they should thank ebay and amazon
06:31 < matsaman> they should thank me for not eliminating them from existence
06:33 < Loshki> lord|: semi-reliably
06:34 < lord|> more reliably than any private company
06:35 < matsaman> besides the CIA, you mean? =P
06:37 < misternumberone> matsaman: https://pastebin.com/raw/qG5nH4kB sorry for delay the system cannot upload the paste because the network adapter is not working
06:38 < matsaman> sure
06:40 < e^1> guys still after removing auto-away plugin still my auto-away is activated, how can i remove it ?
06:41 < e^1> opps ignore it :)
06:42 < matsaman> misternumberone: if it's a temporary solution I would just install the firmware, or just spend $20-30 on a better piece of hardware
06:42 < CrystalMath> hi all
06:43 < matsaman> hi cryst
06:43 < MrElendig> the firmware for that chip is in linux-firmware
06:44 < CrystalMath> does anyone know of a terminal emulator that is lightweight, has no scrollbar / thick window borders (or that there's an option to disable them), no menu (or an option to remove it), supports semitransparency (the real kind, a compositor is running) and that doesn't get broken whenever a double-width character shows up?
06:44 < MrElendig> however the in kernel driver is pretty crappy, so you might want to use the r8168 driver instead
06:44 < MrElendig> misternumberone: ^
06:44 < CrystalMath> i'm currently using urxvt which satisfies all those conditions except for the last one
06:45 < sacules> CrystalMath: st
06:45 < CrystalMath> sacules: st supports real semitransparency?
06:45 < lopid> xterm and your window manager
06:45 < sacules> yeah, with a little patch
06:46 < bls> if you want eye candy, you're generally going to get all the eye candy
06:46 < matsaman> CrystalMath: that's a lot of them
06:46 < CrystalMath> btw do other people have the same problem with urxvt?
06:46 < matsaman> CrystalMath: urxvt can handle that with the right compile options
06:46 < matsaman> CrystalMath: what's a char code that breaks your urxvt?
06:47 < CrystalMath> any emoji
06:47 < misternumberone> MrElendig: the r8169 open source driver is currently in use. however the network adapter is not working at all. no link is detected and the interface is down. i understand that the open source driver might have less functionality, but is it possible to make it work at all?
06:47 < matsaman> CrystalMath: pick one
06:47 < CrystalMath> matsaman: the regular smiley one perhaps
06:47 < CrystalMath> i have no idea how to type it
06:47 < CrystalMath> maybe if i find it somewhere and copy it
06:47 < matsaman> ☺?
06:48 < MrElendig> what does miitools say about the interface?
06:48 < sacules> have you got the correct font?
06:48 < CrystalMath> matsaman: hmm... no that works, it's not double-width
06:48 < MrElendig> also, sure the cable actually is good?
06:48 < matsaman> CrystalMath: uhuh
06:48 < CrystalMath> matsaman: i need the one that's double-sized
06:48 < MrElendig> and do you have a link light
06:49 < aaro> CrystalMath: what do you mean by break it?
06:49 < matsaman> CrystalMath: pick one
06:49 < CrystalMath> aaro: the width of the line seems to be miscalculated in all curses applications after it appears
06:49 < MrElendig> CrystalMath: which version of glibc?
06:50 < matsaman> CrystalMath: 'ろ'?
06:50 < CrystalMath> MrElendig: libc6 2.24-11+deb9u3
06:50 < bls> and does your ncurses have wide character support
06:50 < MrElendig> (urxvt fails at unicode btw, it doesn't do width correctly)
06:50 < MrElendig> CrystalMath: well, way too old glibc too
06:50 < matsaman> it does fine, if you build it right
06:50 < CrystalMath> 😀
06:50 < CrystalMath> that one i think
06:50 < CrystalMath> ok yes
06:50 < matsaman> CrystalMath: that one breaks your urxvt?
06:50 < CrystalMath> it's breaking
06:50 < CrystalMath> yes
06:51 < sacules> CrystalMath: do you have the correct font?
06:51 < matsaman> CrystalMath: define breaking
06:51 < CrystalMath> yes, it shows uup corectly
06:51 < MrElendig> CrystalMath: that one doesn't support unicode >= 8, so it doesn't do 2 char wide unicode correctly
06:51 < CrystalMath> matsaman: all the lines are now shifted by one
06:51 < MrElendig> CrystalMath: so you have two broken things, and sadly those don't negate eachother
06:51 < MrElendig> CrystalMath: I suggest upgrading the system
06:52 < CrystalMath> i don't want to use something other than debian stable
06:52 < CrystalMath> is there a way to block these characters?
06:52 < MrElendig> oh wait, 2.24 do have uc8
06:52 < MrElendig> would still upgrade
06:53 < matsaman> well, there's your real problem
06:53 < matsaman> debian stable is for servers
06:53 < matsaman> and servers don't need interactive terminals with toothy emoji
06:53 < MrElendig> test in a vte based terminal
06:53 < MrElendig> debian probably have a too old vte though
06:54 < CrystalMath> libvte works
06:54 < bls> it works fine in xterm with utf-8 turned on
06:54 < CrystalMath> using sakura right now
06:54 < bls> oh, forgot about the transparency thing
06:54 < misternumberone> MrElendig: the hardware does not have ethernet port link or activity leds. if you mean ethtool output then here is that within ubuntu 17.10 live image https://pastebin.com/raw/9ADTNUqw
06:55 < CrystalMath> oh, no
06:55 < CrystalMath> it's not working
06:55 < CrystalMath> it does break, but only when it scrolls with an emoji
06:55 < MrElendig> use a sane pastebin
06:55 < CrystalMath> with an emoji on the screen
06:56 < aaro> CrystalMath: that char in my urxvt shows as a square but there's no line shifting, anyway you could try termite
06:56 < MrElendig> and I suggest getting the info from the installed system, not some random live cd
06:56 < matsaman> what DE are you using, anyway?
06:56 < CrystalMath> matsaman: are you asking me?
06:56 < CrystalMath> matsaman: i don't use any DE
06:57 < CrystalMath> just a standalone window manager
06:57 < CrystalMath> and i run panels and conky and stuff
06:57 < sacules> are you using a patched font? one of these always works http://nerdfonts.com/
06:57 < matsaman> CrystalMath: oh which wm?
06:57 < CrystalMath> matsaman: compiz
06:57 < CrystalMath> sometimes openbox
06:58 < sacules> i doubt it's a wm issue tbh
06:58 < sacules> it's either a font or terminal problem
06:58 < MrElendig> use glibc >= 2.26 and a less broken TE than urxvt
06:58 < misternumberone> MrElendig: i have not yet installed a system because first I need to determine that the hardware will function under my chosen operating system. here is the paste in the topic's pastebin https://paste.linux.community/view/raw/a2a15d58
06:58 < MrElendig> misternumberone: don't test with 17.10 then
06:58 < CrystalMath> MrElendig: so glibc is to blame?
06:59 < MrElendig> CrystalMath: both glibc and urxvt, as I said, urxvt doesn't do widths correctly even if you have a up to date glibc
06:59 < sacules> i really doubt so
06:59 < CrystalMath> perhaps i can use st to block out such characters
06:59 < CrystalMath> i guess they're the ones that are outside the first 64K
07:00 < sacules> sometimes i had issues with emojis and glyphs, solved it by adding ':width=1' to the font in the sonfig file
07:00 < MrElendig> older glibc versions has a wcwidth which doesn't understand double width characters
07:00 < sacules> maybe you can use a fallback font
07:01 < MrElendig> urxvt just doesn't understand them period
07:01 < CrystalMath> MrElendig: ok but the japanese hiragana "ro" you wrote displayed correctly
07:01 < sacules> like your regular one, and another like Font Awesome for special characters
07:01 < CrystalMath> japanese characters are double-width and they work
07:01 < MrElendig> there is a patch for urxvt, that sort of works, but that upstream have refused to include
07:01 < CrystalMath> i think only characters bigger than 0xFFFF cause the problem
07:02 < matsaman> comes with Gentoo's version =P
07:02 < matsaman> https://github.com/exg/rxvt-unicode/blob/master/doc/wcwidth.patch
07:02 < CrystalMath> sadly width=1 doesn't fix it, but rather it makes it behave the same as sakura
07:03 < CrystalMath> in urxvt
07:03 < sacules> try with dfferent widths tbh
07:03 < MrElendig> (you sort of want unicode >= 10 anyway though..=
07:03 < MrElendig> )
07:03 < MrElendig> which means getting a more up to date glibc
07:04 < matsaman> well no, he's on Debian stable
07:04 < matsaman> so he doesn't want new things =P
07:04 < CrystalMath> yeah, i'd remove unicode 8 too if i could
07:04 < CrystalMath> 2new4me
07:05 < CrystalMath> sacules: widht width=2 the breakage is localized to just the line with the emoji, and not everything after it
07:05 < CrystalMath> *with
07:05 < CrystalMath> ok the breakage is kinda preventing me from seeing what i'm typing properly
07:05 < CrystalMath> i'll have to clear the buffer
07:05 < aaro> CrystalMath: then use rxvt an not urxvt
07:05 < CrystalMath> aaro: rxvt also breaks
07:06 < CrystalMath> oh, hmm...
07:06 < CrystalMath> i think in debian, rxvt is actually urxvt
07:06 < CrystalMath> there's no rxvt packaged
07:06 < MrElendig> rxvt doesn't do unicode at all
07:06 < matsaman> well, it has a few names it goes by
07:06 < aaro> CrystalMath: and termite?
07:06 < MrElendig> rxvt and rxvt-unicode is not the same software
07:06 < CrystalMath> aaro: no package at all
07:08 < CrystalMath> i installed rxvt, it's a symlink to /etc/alternatives/rxvt, which is a symlink to... /usr/bin/urxvt
07:08 < misternumberone> MrElendig: my connection dropped and freenode would not let me reconnect with registered username - to avoid asking for recommendation, i will ask, would it be acceptable to test using a ubuntu 18.04 lts live image?
07:08 < AndrzejL> Hey guys
07:09 < matsaman> hi
07:09 < AndrzejL> I know i am in a wrong place but I don't even know how to search for it...
07:09 < bls> misternumberone: that's what it's for
07:10 < AndrzejL> So here is my problem - I have nginx running on a server inside my network - you can access it at https://andrzejl.eu:30303 - its accessible to the machines that are outside my network as in not connected to my router but not to the machines that are connected to my router.
07:11 < AndrzejL> I changed router recently and I know that the router setting is a culprit
07:11 < misternumberone> bls: previously, i said I was using an ubuntu 17.10 live image and he said "don't use a random live cd... don't use 17.10 then"
07:11 < AndrzejL> bbut I have no idea how to phrase the question / google it
07:11 < MrElendig> misternumberone: you want 18.04 not 17.10
07:11 < AndrzejL> if I disconnect from wifi on my phone and use 3G - site works in the phone browser
07:11 < AndrzejL> if I connect to the wifi on the phone - site times out
07:12 < matsaman> AndrzejL: maybe you have some router-based port forwarding
07:12 < bls> AndrzejL: possibly something relating to routing table, subnet, bound address, there's also ##networking
07:12 < matsaman> most routers can reset to a default config, if that's all you've changed
07:12 < AndrzejL> bls: thanks! I will try them
07:12 < AndrzejL> matsaman: port forward works otherwise the website wouldnt work from outside network too
07:13 < matsaman> well, the one that works works
07:13 < misternumberone> MrElendig: ok thank you i will update the usb disk with live image to 18.04 lts and retry in that case
07:14 < MrElendig> 17.10 is end of life
07:15 < cncr04s> when is 14.04
07:16 < matsaman> doesn't matter, nobody who uses software that goes EOL who doesn't already know when it goes EOL can be helped =P
07:16 < matsaman> lost causes
07:16 < cncr04s> I run all my servers on 14.04
07:16 < MrElendig> cncr04s: already bugfix only, eol in april next year
07:16 < cncr04s> still waiting for them to get images for 18.04
07:16 < MrElendig> er... security fix only
07:17 < MrElendig> no more bugfixes for 14.04
07:17 < cncr04s> I only mostly use mysql (from their repo) and apache
07:17 < cncr04s> too much downtime to upgrade
07:17 < cncr04s> lose millions
07:18 < MrElendig> only if you have no idea what you are doing
07:18 < MrElendig> if you will lose millions from downtime, why don't you have failover and test boxes?
07:18 < MrElendig> (and someone competent managing them=
07:18 < cncr04s> because, I pocket all the $
07:18 < matsaman> gonna lose millions when everything breaks and you have no upgrade plan
07:19 < cncr04s> nah
07:19 < matsaman> unless you dump it all on some sucker just before that happens
07:19 < matsaman> which is a legitimate business plan, even if it makes you an a-hole
07:19 < cncr04s> i'm selling the business potentially
07:19 < matsaman> yes obviously
07:19 < matsaman> good ole republicans
07:19 < cncr04s> i otherwise write all the other software being used
07:19 < matsaman> nevermind those who will suffer after your stupid butt is deceased
07:20 < cncr04s> my build box is still 14.04 too =/
07:20 < matsaman> be quiet
07:20 < cncr04s> i assume a clean install vs an inplace upgrade should still resolve deps the same
07:21 < cncr04s> don't need issues when i do upgrade them all, only to ahve issues running the built software from an upgraded box, on a clean install of a new vps
07:21 < matsaman> shhh
07:21 < matsaman> good lad
07:21 < MrElendig> trying to distupgrade 14.04 to 18.04 will probably be a mess
07:21 < matsaman> upgrading ubuntu usually is
07:21 < cncr04s> already upgraded from 12
07:22 < MrElendig> I would rather do a clean install, set up and test everything and then hotswap in
07:22 < cncr04s> don't want to spend the time to setup an identical 40k$ box setup for a swap
07:23 < cncr04s> employees cost too much
07:23 < matsaman> #aynrand called
07:23 < matsaman> or was it ##alisarosenbum
07:23 < MrElendig> outsource your hosting to china or india
07:23 < MrElendig> problem solved
07:23 < cncr04s> nah
07:24 < cncr04s> service relies on low ping
07:24 < matsaman> I just don't get it
07:24 < matsaman> if you want to shaft people, go all out
07:24 < matsaman> start a sweatshop or something
07:24 < MrElendig> they have servers all over the world
07:24 < cncr04s> I have regional vpses as end nodes
07:24 < matsaman> this ridiculous selling of a business that will immediately fail is the slow way to be an a-hole
07:24 < cncr04s> but a main server rack here in my city
07:25 < cncr04s> they want to keep me on
07:25 < system16> is it true that google is blocked in china ?
07:25 < cncr04s> so, will still be my problem
07:25 < cncr04s> I will care less if they pay me what I want though
07:26 < cncr04s> already got a nice 10 room house in Aspen picked out
07:26 < matsaman> yes I'm sure
07:26 < matsaman> diaf, thanks
07:32 < misternumberone> MrElendig: ok i am now using ubuntu mate 18.04 lts live image and the situation appears to be the same, with no changes in the output of the commands
07:33 < kiki`> "matsman: this ridiculous selling of a business that will immediately fail is the slow way to be an a-hole" .. i thought when buying
07:34 < MrElendig> test with a different cable
07:34 < MrElendig> and check what dmesg says
07:34 < MrElendig> and lsmod
07:34 < MrElendig> and try the 8168 driver
07:36 < jim> kiki`, hmmm? there was no one on this channel named matsman...
07:36 < cncr04s> he left
07:36 < cncr04s> I blew his mind
07:37 < ayecee> it was matsaman, and yeah, he gone
07:37 < kiki`> oh .. jim: it was matsaman .. yeah .. .https://www.antitrustinstitute.org/sites/default/files/Cartels.pdf
07:37 < jim> you mean kiki`?
07:37 < ayecee> who means kiki`
07:37 < kiki`> lol
07:37 < kiki`> shut up and read the paper. it's urgent.
07:37 < jim> oh, typoed
07:38 < ayecee> nah
07:38 < jim> the only paper I have is the paper I wrote a song on
07:38 < ayecee> i have some paper to wipe my butt with
07:38 < cmj> single ply?
07:39 < ayecee> what is this, a mcdonalds
07:39 < cmj> at times
07:39 < kiki`> penetrate the ocean on our .. moonlight drive
07:40 < jim> imprinted with our favorite billionaire...
07:40 < ayecee> i don't know what that means, but i like it
07:41 < kiki`> the grand cornholio - ayecee plays the imbecile hand
07:41 < kiki`> (1/4 in the deck)
07:42 < ayecee> hey, i resemble that remark
07:42 < jim> you do? I mean, yaknow, you're right...
07:42 < kiki`> alright. have a good one. make linux better. peace out.
07:43 < jim> one line at a time
07:48 < {DV8}> I'm guessing this is a stupid question, but does anyone know if there's any harm setting my 13" 1080p display to 1600x900?  I haven't been able to find another solution in Cinnamon that works for all windows without making text too small to read without straining my eyes
07:49 < MrElendig> that is not a good solution
07:49 < MrElendig> fix the broken programs, or use less broken ones instead
07:49 < {DV8}> yea, didn't think so :)
07:50 < MrElendig> many laptops have displays that don't scale to non-native, so ymmv
07:50 < {DV8}> Thunderbird, for instance, displays all the toolbars and icons much too small.
07:51 < MrElendig> also using a non-native res generally quite noticable reduces the quality
07:52 < MrElendig>  Edit → Preferences → Advanced → Config editor.
07:52 < nyAgEO> what would happen is you try modifying dpi setting?
07:52 < MrElendig> layout.css.devPixelsPerPx 2
07:52 < MrElendig> (or whatever value you want)
07:52 < MrElendig> also GDK_DPI_SCALE
07:54 < {DV8}> hah, I searched the settings trying to find something like that.  Thanks so much :)
07:56 < MrElendig> https://wiki.archlinux.org/index.php/HiDPI
07:58 < {DV8}> That's been the challenge is trying to set the HiDPI at a setting between 1 and 2.  2 makes everything much too large, however 1 is too small.  It seems most environments don't allow a non integer value.
08:01 < nyAgEO> the one time I had to change the dpi setting I added a line in the monitor section of xorg.conf to adjust it to the monitor
08:02 < nyAgEO> i think the default is 96 dpi and on small screens its no good
08:02 < MrElendig> 1080p on a 13" isn't that high of an dpi
08:02 < system16> what distro is the most light weight OS ?
08:02 < ayecee> ms dos 3.3
08:03 < MrElendig> 2.22 you mean
08:03 < LissajousPattern> system16, more like desktop environment
08:03 < ThePortWhisperer> hi
08:03 < MrElendig> 2.2*
08:03 < ayecee> is that what i mean
08:03 < ThePortWhisperer> a box has a static ip assigned, i looked in /etc/network/interfaces and it didn't reference any static IP there
08:03 < system16> LissajousPattern, xfce ?
08:03 < LissajousPattern> yeah
08:04 < {DV8}> It's Cinnamon on an XPS 13.  I did change some of the font-sizes in the settings, however it didn't change everywhere
08:04 < LissajousPattern> I use it on an old netbook
08:04 < ThePortWhisperer> i know it's being set inside the VM - any other place i can look to get rid of that static IP
08:04 < LissajousPattern> it runs very well
08:04 < system16> LissajousPattern, i have lubuntu installed on my usb drive.
08:04 < LissajousPattern> atom n450 with 1 GB and it runs well
08:04 < jim> ThePortWhisperer, there are a few different ways a static IP could be assigned
08:05 < LissajousPattern> system16, yeah I used to use lubuntu
08:05 < ThePortWhisperer> i know its being done by the VM because my dhcp server isn't able to override whatever setting the VM has
08:05 < LissajousPattern> it really depends on your use case and the hardware for sure
08:06 < jim> ThePortWhisperer, does that machine have net right now?
08:06 < LissajousPattern> as many say your mileage may vary slightly.
08:06 < ThePortWhisperer> ist not internet facing but other boxes on my lan can see it
08:06 < ThePortWhisperer> its just putting itself on the wrong subnet
08:07 < ThePortWhisperer> my firewall rules prevent any meaningful interaction with that subnet
08:07 < jim> can you ping random stuff, like yahoo.com?
08:07 < ThePortWhisperer> its not supposed to be able to call out to the net
08:07 < ThePortWhisperer> i dont have commandline access to it either, its part of an exercise
08:08 < ThePortWhisperer> i can just mount the disk and mess with configs
08:09 < jim> exercise?
08:09 < LissajousPattern> red team?
08:09 < ThePortWhisperer> LissajousPattern, basically
08:09 < LissajousPattern> yeah sounds liek it
08:09 < LissajousPattern> like**
08:10 < LissajousPattern> so if you can mess with configs than you need to first get your self a some permissions
08:11 < LissajousPattern> and then grant yourself a shell somehow
08:11 < ThePortWhisperer> no, i just need to figure out how to unset the static ip
08:11 < LissajousPattern> oh
08:11 < ThePortWhisperer> i want my dhcp server to put it in the proper subnet. its putting itself in the wrong subnet and my boxes cant4 talk to it
08:11 < ThePortWhisperer> im not sure where else in linux this could be configured outside /etc/network/interfaces
08:13 < LissajousPattern> try that
08:13 < PowerPCM_> smoking bones by the staircase?
08:14 < ThePortWhisperer> try what?
08:14 < LissajousPattern> does it have to dhcp or would resolving it to the proper subnet work solely?
08:14 < LissajousPattern> because you can resolve it to public/private in that conf file
08:15 < LissajousPattern> but the ip may still be static...?
08:15 < MrElendig> fix your dhcp server config?
08:15 < ThePortWhisperer> it just has to be in the proper subnet
08:15 < ThePortWhisperer> dhcp server can be overridden by host settings?
08:15 < ThePortWhisperer> ip allocation*
08:16 < MrElendig> you can not use dhcp
08:16 < ThePortWhisperer> my goal is to determine how this linux box is assigning itself a specific IP
08:16 < MrElendig> but really, just fix it so that your dhcp server gives you the correct subnet?
08:17 < LissajousPattern> all yo have to do is change the /24 to a /17 or visa virsa
08:17 < MrElendig> uhm, why would that involve changing the subnet
08:17 < LissajousPattern> in the network interface conf file you specified
08:17 < ThePortWhisperer> i dont want to change my network
08:17 < ThePortWhisperer> i want to change the proper file on this host which is causing it to get assigned this IP
08:18  * MrElendig smell a xyproblem
08:18 < PowerPCM_> smoking bones by the staircase?
08:18 < ThePortWhisperer> the VM is specified in the setup guide  has having this IP
08:18 < ThePortWhisperer> as*
08:18 < ThePortWhisperer> they dont say where or how they set it
08:18 < ThePortWhisperer> so i was hoping someone in here might know
08:19 < ThePortWhisperer> im on the verge of doing a find . -exec cat {} \; |grep ip
08:19 < LissajousPattern> port scan that hoe
08:20 < ThePortWhisperer> sigh,
08:20 < LissajousPattern> IKR
08:20 < LissajousPattern> i was joking
08:21 < ayecee> should tell funnier jokes
08:22 < LissajousPattern> ayecee, let me think of some..., it may take a minute.
08:23 < andrei-n> Hello.  Is there a way to make stty temporarily disable ^C and ^Z whenever the less program runs?  I tried to make it in a function, but it doesn't work because it needs the original terminal....
08:24 < ayecee> make less an alias?
08:25 < andrei-n> ayecee, Yeah, I tried this to make an alias to a function, but it didn't work...  Also it should support pipe.
08:25 < ayecee> what happened when you tried
08:26 < ThePortWhisperer> dudes
08:26 < ThePortWhisperer> so much rage
08:26 < andrei-n> ayecee, an error: stty: 'standard input': Inappropriate ioctl for device
08:26 < ThePortWhisperer> @jim, LissajousPattern MrElendig
08:26 < ayecee> andrei-n: what was your function
08:27 < MrElendig> rage?
08:27 < LissajousPattern> ThePortWhisperer, sorry breo
08:27 < LissajousPattern> bro-ski
08:27 < ThePortWhisperer> i mounted the disk but migrated to the wrong directory
08:27 < LissajousPattern> ouch
08:27 < ThePortWhisperer> was looking at my main hosts etc/network/interfaces
08:27 < ThePortWhisperer> fat finger confirmed
08:28 < LissajousPattern> oh f***
08:28 < andrei-n> ayecee, stty intr undef; stty susp undef; less "$@"; stty intr '^C'; stty susp '^Z'
08:29 < ayecee> andrei-n: could use stty's -f option to tell it what device to frob
08:30 < andrei-n> ayecee, I don't think so.  Since it's in a pipe, how can I get the device?
08:30 < ayecee> you'd get it beforehand and put it in an env variable, if it's not already there.
08:31 < ayecee> like, something you'd set from a login script
08:31 < andrei-n> ayecee, thanks I'll try it.
08:37 < andrei-n> ayecee, Thanks!  It worked.  Even with the man command.
08:38 < ThePortWhisperer> sup
08:38 < ayecee> nice
08:38 < ThePortWhisperer> what does the line `iface eth0 inet dhcp` mean
08:38 < ThePortWhisperer> particularly the inet part
08:39 < ayecee> inet is the protocol family
08:39 < ayecee> inet6 would be another example
08:39 < ThePortWhisperer> so inet is ipv4
08:39 < ayecee> looks that way
08:40 < esselfe> dhcp is the opposite of static
08:40 < ThePortWhisperer> thanks im good now
08:40 < ThePortWhisperer> frick
08:40 < ayecee> man interfaces goes into more detail
08:41 < ThePortWhisperer> after about 3 hours of troubleshooting i finally have this virtual machine up
08:41 < LissajousPattern> sweet
08:41 < LissajousPattern> well not the 3 hour part but the fact that its up
08:42 < ThePortWhisperer> yeah im trying to use these to learn about linux
08:42 < LissajousPattern> yeah its all good stuff to learn
08:42 < ThePortWhisperer> tshooting..good learning experience as well but gets old after a couple hours
08:42 < LissajousPattern> yeah for sure
08:43 < jim> well I haven'/t seen anything but inet, so it could just mean " here comes an interface
08:43 < jim> "
08:43 < LissajousPattern> I have been learning linux for the past 7 years and am still learning new stuff daily about it
08:43 < jim> ThePortWhisperer, and, debian invented that
08:44 < LissajousPattern> the power of linux  is beast mode
08:44 < jim> best noodle?
08:44 < storge> beast mode
08:44 < LissajousPattern> storge, yes indeed
08:44 < ThePortWhisperer> yeah i just did something that would have required python in windows
08:45 < ThePortWhisperer> curl site.com/?page[1-20] >>foo.txt
08:45 < LissajousPattern> man I cant wait to play a game in 4k for the first time
08:45 < ThePortWhisperer> looped through all 20 pages
08:45 < ThePortWhisperer> with such a short one liner
08:46 < LissajousPattern> that is truly what I call beast mode
08:46 < LissajousPattern> one and done mic drop
08:50 < pingfloyd> LissajousPattern: that's unix-likes in general
08:50 < my_mind> ther is a program called oprint for non-linux-OS that creates a virtual printer for iphones to use network printewrs
08:50 < my_mind> is there one for linux?
08:50 < ThePortWhisperer> LissajousPattern, whiich windows manager do you use
08:50 < LissajousPattern> pingfloyd, no doubt
08:50 < ThePortWhisperer> i just switched to xfce, it blows away gnome3
08:50 < pingfloyd> LissajousPattern: that is they have a good userland that was designed with unix tenet in mind.  The result is all this power and flexibility at the shell.
08:51 < ThePortWhisperer> its powerful and whatnot, sometimes python is just as effective. sometimes those standard nix one liners are better.
08:52 < pingfloyd> unixes and unix-likes don't fight you like Windows and MacOS does.  They do as you tell them as long as you tell them right.
08:52 < LissajousPattern> ThePortWhisperer, yeah I have been using xfce for a few years
08:52 < LissajousPattern> I like basic
08:52 < pingfloyd> I like xfce also
08:52 < LissajousPattern> but yet highly functional
08:53 < pingfloyd> same here
08:53 < jim> do you like line numbers?
08:53 < pingfloyd> I like the UI to be out of my way
08:53 < LissajousPattern> honestly I appreciate elegance in simplicity
08:53 < ThePortWhisperer> line numbers..indifferent
08:54 < pingfloyd> xfce has the important stuff covered
08:54 < LissajousPattern> man I am loving my new setup
08:55 < LissajousPattern> 55 4k TV as my monitor for a i7 6700 gtx1070 rig man talk about real estate
08:55 < LissajousPattern> and I still want to add another monitor
08:56 < notmike> Money
08:56 < LissajousPattern> bout to get down on a few games of fortnite just for shits
08:57 < LissajousPattern> I am just waiting on a few updates so I can reboot
08:59 < ThePortWhisperer> pretty good
08:59 < ThePortWhisperer> similar but i run 4 portrait 23 inchers
08:59 < ThePortWhisperer> 1x4
09:01 < justsomeguy> Yeah? Well I run a bitchin' raspberry pi hooked up to a resistive touch screen. 720p. 3" diagonal. Loaded.
09:01 < justsomeguy> :~p
09:02 < justsomeguy> Nice setups, tho, for real.
09:02 < LissajousPattern> ^ same
09:05 < ThePortWhisperer> yeah mine is good for what i do
09:05 < ThePortWhisperer> 4 screens allow for snapping windows  to a display quickly
09:05 < LissajousPattern> mine is way overkill for me at the moment
09:05 < ThePortWhisperer> portrait, good for code/log viewingr
09:06 < ThePortWhisperer> landscape....more for entertainment/media consumption
09:06 < LissajousPattern> most def
09:06 < LissajousPattern> I have a 32 1080 panel that I may orient as portrait
09:17 < LissajousPattern> I had no idea there was a "Gaming Music" genre
09:17 < LissajousPattern> cool I guess
09:25 < notmike> No trolling
09:31 < felix_vs> Hi i'm trying to map Menu key to Super key, is this possible? (using XFCE4 Linux Mint 18.3)
09:31 < bls> yes
09:31 < bls> https://wiki.archlinux.org/index.php/xmodmap
09:41 < MrElendig> might even be a xbkmap option for it
09:53 < felix_vs> bls: thanks for suggestion! guide is very good. solved the key mapping issue with it.
10:03 < bipul> FreeBSD uses a linux kernel?
10:08 < Stryyker> no
10:11 < abramovich> how to setup persistence storage in a live USB?
10:12 < abramovich> like i dont wanna use the WHOLE pendrive for flashing the iso, only 2GB tops..rest of 60 gb i want to store some movies and stuff. can it possible?
10:12 < abramovich> help will be appreciated
10:14 < RayTracer> abramovich: maybe there is already documentation for the distro, like https://docs.fedoraproject.org/quick-docs/en-US/creating-and-using-a-live-installation-image.html
10:15 < abramovich> i use arch btw (unironically)
10:18 < bipul> What does this kernel stat mean? "sysctl -n kern.ipc.shmall" ?
10:19 < storge> abramovich: grml has a whole howto on their website of how to do persistence
10:19 < storge> they even have a package (at least in debian) grml2usb
10:19 < abramovich> storge: thank you brother. will look into it now
10:19 < RayTracer> abramovich: the web search I did doesn't work out that well if I just replace fedora with arch.. seems they head for a direct install to usb
10:20 < storge> it's a pretty well written logical approach. i have a live linux plus persistence and a third FAT partition so i can so a lot with that one drive
10:21 < abramovich> storge: true
10:23 < RayTracer> bipul: it's a wrong name, it's kernel.shmall, documented at https://www.kernel.org/doc/Documentation/sysctl/kernel.txt
10:26 < jim> send the partition to the gym
10:32 < abramovich> kek
10:36 < abramovich> 1145044992 bytes (1.1 GB, 1.1 GiB) copied, 9 s, 121.2 MB/s
10:37 < X2> SSD -- abramovich ?
10:37 < abramovich> X2: how u know :D
10:37 < X2> Lucky guess
10:37 < abramovich> :D
10:37 < abramovich> inb4 doxxed! im out folks. thanks storge. i will give you 10 XMR when i become crypto millionaire
10:50 < rozick> Hey, I don't know if this is the right channel for this. But I have a zip file that includes a DMG file on Linux with 0777 permissions
10:50 < rozick> When I download it to a mac, I run ls -l and see that it indeed still has 0777 permissions. However when I try to run the file it says "can't open the file".
10:51 < rozick> I then, on mac this time, set permissions to the folder again to chmod -R 777 and run it
10:51 < rozick> and this time it works
10:51 < rozick> But it already showed all the permissions as 0777
10:51 < rozick> so why do I Have to run the command again on the mac?
10:54 < lopid> it changed from file to folder
10:56 < TheDoudou> hi
10:56 < rozick> So how do I solve it Lopid? So that it works?
10:57 < lopid> it changed in your description. nothing in it suggests why it wouldn't work
10:57 < TheDoudou> it's possible in bash to send logs to a syslog server ?
10:57 < lopid> it has nothing to do with bash, TheDoudou
10:57 < TheDoudou> logger ?
10:57 < lopid> yes
10:58 < lopid> how did you run it, rozick?
10:58 < TheDoudou> for bash script :p
10:58 < lopid> exactly
10:58 < rozick> I ran it through a mac user
10:58 < rozick> I Created the dmg file on linux, set permissionst o 0777
10:58 < rozick> ran it on mac, and it says can't run it
10:58 < rozick> i then, on mac, run chmod -R 777 on the folder and i'm able to run the DMG file
10:58 < lopid> ran it how?
10:59 < rozick> but before I Ran 777 in any case all the files were 777...
10:59 < rozick> double clicking on the dmg file as a user
10:59 < lopid> and what folder?
10:59 < jim> TheDoudou, it's possible to send logs to a syslog server... I don't understand what you're saying with "in bash"? what exactly do you want to do?
10:59 < rozick> on the dmg folder
10:59 < rozick> on the dmg file
10:59 < lopid> you said it was a file
11:00 < lopid> anyway, sounds like a question for #mac
11:00 < rozick> its a folder
11:00 < rozick> dmg folder
11:01 < TheDoudou> https://ideone.com/mA6lp5
11:01 < hexnewbie> rozick: ‘chmod -R 777’ (aka ‘a command that you should never ever use’) does have very noticeable effects when called on *directories*, because it changes them recursively and thus changes more than just the 777 permissions of the directory you already see
11:02 < rozick> What does it change besides permissions?
11:02 < rozick> I understand it shouldn't be used but I'm trying to test.
11:02 < hexnewbie> rozick: Permissions. But those not of the directory, those of the files inside it.
11:03 < rozick> it's recursive.
11:03 < rozick> I know, I'm doing that on purpose
11:03 < rozick> My question is why isn't the mac automatically setting the correct permissions to the download
11:03 < hexnewbie> rozick: Yeah, but if the directory shows up as 777 in whatever, that speaks nothing of the permissions inside
11:03 < LissajousPattern> hi
11:04 < rozick> the permissions inside are also 777
11:04 < LissajousPattern> test
11:04 < hexnewbie> rozick: Now that's an exclusively macOS question. But I suspect that it is because macOS is sane.
11:04 < LissajousPattern> well I guess its working
11:05 < lopid> did something just say "test"? it didn't appear with a name
11:05 < TheDoudou> lopid: with netcat work fine (echo -n "test message" | nc -4u -w1 192.168.0.10 514)
11:05 < LissajousPattern> lopid: what? thats wierd its me
11:05 < LissajousPattern> i have a nick
11:05 < hexnewbie> lopid: I don't think it did, I don't see it.
11:05 < lopid> it happened again
11:06 < hexnewbie> Maybe it's a ghost?
11:06 < LissajousPattern> what?
11:06 < LissajousPattern> oh my
11:06 < LissajousPattern> please help me if you can't see my nick?
11:06 < lopid> we can
11:06 < LissajousPattern> oh
11:06 < lopid> at least, i can…
11:06 < LissajousPattern> cool
11:06 < LissajousPattern> sorry
11:07 < lopid> so how's it going, SaraJ81?
11:07 < LissajousPattern> you were freaking me out I happen to be using a completely different client than usual
11:08 < lopid> from that output, it looks like you have more problems than syslog, TheDoudou
11:09 < TheDoudou> lopid: ?
11:09 < lopid> "ping: command not found"
11:10 < TheDoudou> at ideone ?
11:10 < lopid> yes
11:10 < TheDoudou> it's normal no network command, file, lib, etc.
11:11 < lopid> there is also no network
11:12 < lopid> …support for that destination
11:12 < LissajousPattern> that client will take some getting used to but i bet once i configure it then it will be amazing.
11:12 < hexnewbie> What are the chances, a server reached a load average of 3452, 4 minutes after I accidentally saw it was about to run out of memory as I was reading the monitoring reports (instead of the news) with my morning coffee. That timing.
11:14 < lopid> it doesn't look like syslog is even listening on that port, especially since you were able to listen on it with netcat
11:16 < TheDoudou> send log with netcat
11:16 < TheDoudou> not listen
11:16 < lopid> oh yes
11:17 < TheDoudou> syslog have port 514 open udp (work fine with my routeur to syslog serv)
11:17 < hexnewbie> It's running some badly written fork-happy daemon that never exits its children, so it's constantly like a slow-running fork bomb (like time travel explosions from movies), but you almost never see the high load average before someone has caught it, let alone at the same time.
11:19 < lopid> is it a cut down os? maybe that logger doesn't support sending messages remotely. google that servname message
11:20 < lopid> sounds like you need to monitor and alert on that automatically, hexnewbie
11:20 < TheDoudou> lopid: "-n, --server server               Write to the specified remote syslog server instead of to the               system log socket.  Unless --udp or --tcp is specified, logger               will first try to use UDP, but if this fails a TCP connection               is attempted."
11:21 < TheDoudou> man logger
11:22 < FightingFalcon> will a hard reset cause data loss?
11:22 < jim> it can
11:23 < hexnewbie> lopid: That was the automatic alert on the alert page I was reading (less depressing than the news), which nobody saw for 1 day and 18 hours. It doesn't send emails or SMS alerts because well, there are constantly like 20 pressing problems on it.
11:23 < FightingFalcon> you lose data on hard disk?
11:23 < jim> it's always best to shut down a linux properly, so that all filesystems are cleanly unmounted
11:23 < sh1ro> what data is lost will is dependant on what you have open and your fs
11:24 < FightingFalcon>  jim how do i know if i lost data or not?
11:24 < lopid> doesn't sound like anybody considers them so pressing :)
11:24 < jim> FightingFalcon, what you might lose is what's in ram about to be written to the disk (as a result of opening and then writing a file)
11:25 < FightingFalcon> jim, how about files in hard disk (ssd) ?
11:25 < jim> well it's the same thing, whether the disk is spinnning or an ssd
11:26 < FightingFalcon> i mean older files wont be lost right?
11:26 < jim> closed files will -probably- be safer
11:27 < FightingFalcon> is there a way to see if data is lost or not?
11:27 < jim> I don't know myself, I've never had to deal with that
11:28 < hexnewbie> lopid: We've been thinking of implementing alert levels that separate problems not only on severity, but whether how immediately they need to resolved and do they have lasting impact. No free developers.
11:28 < jim> FightingFalcon, did you have to reset?
11:29 < FightingFalcon> yes
11:29 < jim> were you editing a file at the time?
11:30 < FightingFalcon> everything works fine but im not sure for example if 1 picture out of 1000+ in the web site is lost...
11:30 < FightingFalcon> no
11:31 < jim> so, there were probably a few files open for writing, and those include log files
11:31 < jim> were you downloading or uploading any files?
11:32 < FightingFalcon> Nope
11:32 < jim> ok, do you use any databases?
11:32 < FightingFalcon> yes i do
11:33 < FightingFalcon> i have wordpress sites
11:33 < jim> which machine did you have to reset?
11:33 < Disconsented> >wordpress
11:33 < Disconsented> Im sorry for your loss
11:33 < FightingFalcon> what do you mean?
11:33 < FightingFalcon> which machine?
11:34 < jim> your home machine? a work machine? a wordpress site?"
11:35 < jim> you say you had to reset a machine, which one?
11:35 < FightingFalcon> My vps in hetzner
11:35 < jim> ok.
11:35 < MrElendig> just compare to your backup
11:35 < jim> does that machine have wordpress sites on it?
11:36 < FightingFalcon> yes
11:36 < lukey> FightingFalcon: You may look in /lost+found
11:37 < jim> do you have a way of comparing the backed up version of the database to the running version?
11:37 < hernan604> hi!
11:37 < jim> hi
11:37 < hernan604> anyone know if its possible for a GUI software to open another(s) GUI software inside it ?
11:38 < FightingFalcon> jim, well no, all the websites are working fine right now. And i havent updated them for 2 weeks. So i didnt change anything.
11:38 < lukey> hernan604: What do you mean exactly?
11:38 < jim> FightingFalcon, do they collect data you want to save?
11:39 < FightingFalcon> nope
11:39 < FightingFalcon> its a read-only web site
11:39 < MrElendig> (why use wordpress in the first place then)
11:39 < jim> oh ok... if everything is ok on that site, you're good... if not, you might need to restore your backup
11:40 < FightingFalcon> jim, my only concern is i have 1000+ pictures hosted on the site. none of them might have been lost right?
11:41 < jim> dunno :) you'll have to check
11:41 < jim> I don't think so, if they were only read, necer written
11:41 < MrElendig> FightingFalcon: as said, check against your backup
11:42 < MrElendig> only way to actually know
11:42 < lukey> FightingFalcon: Look at the log of you system booting / fsck output of the boot after the system crash
11:43 < FightingFalcon> lukey where do i find that?
11:43 < lukey> FightingFalcon: If you didn't reboot since the incident "journalctl" then search with "/fsck"
11:43 < jim> FightingFalcon, do you keep backups of the site?
11:43 < FightingFalcon> yes
11:44 < jim> maybe the safest thing to do is restore the entire site
11:45 < FightingFalcon> lukey no such thing in the log
11:45 < hernan604> lukey: i want to run software as embeded apps, inside my GUI software
11:45 < hernan604> lukey: is there a terminology for this ?
11:45 < jim> wiktorpp, hi
11:47 < MrElendig> a window manager? :p
11:54 < hexnewbie> platform?
12:13 < ExtremeFMan> morn
12:23 < ElinKattunge> I was just wondering, I've symlinked /bin/bash to /bin/zsh and I've had people tell me that this is a very bad idea, however I didn't believe them because zsh is basically just ksh with improvements and ksh is POSIX-Compliant. I was just wondering what the rest of you guys think of this?
12:24 < ElinKattunge> I mean, I can understand if I symlinked /bin/bash to a completely different shell, there are gonna be syntax issues when executing system scripts.
12:25 < MrElendig> ElinKattunge: don't do this period
12:25 < revel> ElinKattunge: Well, scripts with a #!/bin/bash shebang might expect actual bash-like behaviour that zsh may not have.
12:25 < MrElendig> it is stupid and serves no purpose except breaking your system
12:25 < revel> Plus ^
12:25 < MrElendig> if you want to change your shell then just use chsh
12:25 < ElinKattunge> MrElendig: Why not? There is never a "period", considering when I do things that are considered mad, it's usually in a VM I've created for the intents of doing things like this.
12:26 < ElinKattunge> "except breaking your system"
12:26 < ElinKattunge> Not entirely accurate.
12:26 < revel> Pros: None. Cons: Likely to break tons of system scripts due to missing or unexpected features.
12:26 < ElinKattunge> Not when it comes to substituting bash with zsh.
12:27 < MrElendig> there are more productive ways to break your system, like trying to set up ldap/radius/heimdal
12:28 < ElinKattunge> Thing is, in a production environment, I do things the proper way, I don't symlink shells, etc, I take advantage of /etc/shell and the likes. I am always experimenting with Linux and the FNU Userland in a VM
12:28 < ElinKattunge> *GNU
12:28 < ElinKattunge> I actually tried this on a setup for 5 years, I and with zsh, it worked fine, no problems.
12:31 < ElinKattunge> I even removed bash, turning it into a bash-free system, lol
12:35 < revel> Well, as my package manager makes heavy usage of bash, I'd rather not risk it based on you saying "it didn't break something for me"
12:35 < ElinKattunge> revel: I never suggested others should try.
12:36 < revel> Why'd you do it though? Just for fun?
12:36 < ElinKattunge> I did this as an experiment myself, and if others are to do it, I expect them to treat what they are doing also as an experiment in a production environment.
12:37 < one_roOt> does this message get through? :S
12:38 < revel> Yes.
12:38 < ElinKattunge> revel: Thing is, I know the standards on how a GNU/Linux-based system is setup, I don't need to be told I'm doing something wrong, and if I do, I usually have a sense I'm doing something wrong beforehand and I will correct my mistakes. But, the thing is, that gets boring, there are more things that can be tried and played with on Linux such as symlinking shells which a sysadmin would screw their face at, and
12:38 < ElinKattunge> I understand why.
12:39 < ElinKattunge> I don't like to be told "No, that's stupid" because it doesn't mean I don't know it's stupid, when I do. I just like to know what happens and the outcomes.
12:39 < revel> I don't see where the fun is to be had with potentially breaking scripts that assume some bash functionality is there while it actually isn't in some other shell's "compatibility mode".
12:39 < ElinKattunge> I will do something just to see what happens, so why shouldn't I link /bin/bash to /bin/zsh ?
12:40 < revel> Well, I literally just told you why I don't think you should/why I don't, didn't I?
12:40 < ExtremeFMan> because that's stupid ;)
12:40 < ElinKattunge> Thing is, it isn't about symlinking shells, it's about what shell you symlink bash to. Don't forget zsh and ksh are almost the same thing, with their differences, and the differences risk breakage, however what I found is most scripts wont break.
12:40 < revel> Because zsh's "compatibility mode" may be incomplete and break some scripts.
12:41 < ElinKattunge> brb
12:41 < revel> And I'd rather not try to unbreak large parts of my package manager.
12:42 < ice9> whats the best easy to use video editor
12:43 < HaMsTeRs> I still couldn't install Office 2013 nor 2016 on my Ubuntu 18.06 box.  I tried many many many many many many methods, none works!
12:43 < MrElendig> ice9: cinerella
12:43 < MrElendig> ice9: it is the one that crashes the least
12:43 < MrElendig> (and the only one that isn't a toy)
12:44 < ice9> MrElendig, what do you think about openshot?
12:44 < MrElendig> unuseable most of the time
12:45 < HaMsTeRs> help anyone?
12:46 < revel> HaMsTeRs: Install LibreOffice :^)
12:46 < MrElendig> HaMsTeRs: use libreoffice or the web version of office
12:46 < HaMsTeRs> revel, can't,
12:46 < revel> Why not?
12:46 < MrElendig> HaMsTeRs: other than that: contact microsoft customer support
12:47 < ElinKattunge> Back
12:47 < HaMsTeRs> because it cannot replace MS office.  it has some spacing problem
12:47 < MrElendig> HaMsTeRs: it is closed source, pay to play, we cant fix it for you
12:47 < ElinKattunge> Is Wayland actually confirmed to replace Xorg, and is there an ETA?
12:47 < HaMsTeRs> obviously they doesn't provide support to linux
12:48 < ElinKattunge> And I don't understand Canonical's direction with Mir, they decide to develop a display server which is a huge task, only to abandon it? I wouldn't be happy to be on their dev team...
12:48 < ElinKattunge> I've heard abandonments time and time again from Canonical, it's annoying.
12:49 < hexnewbie> Office 2016? Is that like SoftMaker Office 2016? Did they have a 2016 version?
12:49 < hexnewbie> HaMsTeRs: Was any of the methods: Using MS Office 2013 *32-bit* installer, using Wine 3.11, running all winetricks pertaining to Office (not all listed on the appdb entry for the installer of this version)?
12:50 < ElinKattunge> hexnewbie: You'd be best using a more Commercial solution like Crosswire, if you want to properly support Microsoft applications on your Linux platform.
12:50 < hexnewbie> MrElendig: LibreOffice is completely unusable for many documents out there (e.g. journal articles)
12:50 < ElinKattunge> and that's gonna cost you money, so good luck with using bare wine.
12:50 < HaMsTeRs> win 3.0 not 11
12:51 < HaMsTeRs> and tired many different playonlinux scripts
12:51 < ElinKattunge> HaMsTeRs: LibreOffice is annoying
12:51 < ElinKattunge> It pushes me to Google Docs...
12:51 < hexnewbie> s/LibreOffice/any non-Microsoft office/, I did run a Softmaker Office trial, and that borked all the documents as well
12:52 < hexnewbie> HaMsTeRs: Office 2013 had garbage rating until recently, I don't remember if I ever saw reasonable ratings on 3.0. It's possible it wasn't supported in 3.0
12:53 < hexnewbie> HaMsTeRs: Office 2016 has garbage rating right now, so chances are it won't work at all (I'm speaking of the installer, you need to check appdb seperately for individual components)
12:53 < HaMsTeRs> hexnewbie, i don't mind to install 2013
12:54 < ElinKattunge> LibreOffice is annoying cause, it's buggy as hell, and that's since it's gone from first being StarOffice => OpenOffice => LibreOffice, that is 3 different names with no real changes! LibreOffice is stagnant. Apache forks OpenOffice but does nothing with it. I don't care about the UI, I don't expect Microsoft Ribbon, but there are bugs which are come across and I think "Hmm, those will be easy to clean up"
12:54 < HaMsTeRs> ElinKattunge, indeed
12:54 < MrElendig> ElinKattunge: it is not stagnant
12:55 < MrElendig> you have clearly never looked at the changelogs/codebase/news
12:55 < ElinKattunge> MrElendig: So then why does each and every release work almost the same with bugs then?
12:55 < MrElendig> as for compability, ms office isn't even compatible with itself
12:56 < ElinKattunge> I actually can't use the Suite in a production environment because it's too buggy.
12:56 < sh1ro> schools are starting to use libreoffice now and it's kinda amazing
12:56 < hexnewbie> ElinKattunge: Well, LibreOffice has improved substantially (and it's easy to notice even if I'm not a serious user of it), unlike the Apache one. But you're right that it's riddled with compatibility problems and bugs.
12:57 < sh1ro> granted that is partially because it's cheaper
12:57 < ElinKattunge> My business has been considering getting Microsoft Office licnses to replace LibreOffice because it's not mature enough.
12:57 < hexnewbie> It fails badly at editing a simple table, and stuff like that.
12:57 < sh1ro> there's also openoffice but i've not tried it and it's less popular
12:58 < sh1ro> or to be more accurate i've not tried it since openoffice was a thing
12:58 < ElinKattunge> Just cause it comes in handy to quickly open a document you stumbled upon, doesn't mean it should be used in a daily-to-daily scenario, where you got a Secretary at the desk.
12:58 < HaMsTeRs> well, thinking crossover now
12:58 < MrElendig> openoffice is dead
12:58 < sh1ro> *open opensolaris
12:58 < hexnewbie> Or I create a circle around an item on the page, and I cannot move the circle. Like, all I want is to print a number to show a clerk, and circle the number so that the clerk knows which is the relevant part
12:58 < MrElendig> oracle decided to suicide it
12:59 < sh1ro> MrElendig: again? i knew people using it less than a year ago
12:59 < MrElendig> sh1ro: people are also still using windows xp
12:59 < MrElendig> sh1ro: and homeopathy and religion
12:59 < ElinKattunge> Eww
12:59 < sh1ro> i glanced at it and it looked like it was back
12:59 < MrElendig> and imperial units
12:59 < MrElendig> it isn't
12:59 < sh1ro> oracle, just cursing everything they touch
13:00 < ElinKattunge> Windows XP drove my Irlen Disease crazy
13:00 < MrElendig> all the devs fled to librepffice
13:00 < hexnewbie> MrElendig: After Oracle killed it, it became Apache OpenOffice, which gotten a serious interface lift from the Lotus Symphony fork. So OpenOffice right now refers to something way after Oracle. LibreOffice imported the interface lift, so it is dead, but died years after Oracle.
13:00 < ElinKattunge> and I used to get sick and tired of people telling me off for changing the UI on their systems to a dark black
13:01 < hexnewbie> ElinKattunge: I'd tell you off for changing 1px of padding on *my* system, sure.
13:01 < searedvandal> wps office is pretty good. their support for ms office documents is a bit better than libreoffice in my opinion
13:01 < ElinKattunge> hexnewbie: Like in college
13:01 < ElinKattunge> Still "someones" system
13:02 < ElinKattunge> One staffer would permit you, another would tell you off...
13:02 < ElinKattunge> There is like no win with people when it comes to doing the right thing.
13:02 < hexnewbie> College? Well, then that's a different place. For some people, using a dark theme may even constitute ‘hacking’.
13:03 < ElinKattunge> I was always called a "hacker"
13:03 < ElinKattunge> It's annoying, but if a dark theme is better for your vision then it's a bit OTT to call people hackers based on their UI choices.
13:04 < hassoon> 'morning
13:04 < sh1ro> on a related note if you style your terminal to black on white with margins and a font like arial, sans or helvetica people will think you're doing normal word stuff
13:05 < ElinKattunge> Seriously?
13:05 < hexnewbie> No, I seriously have an academic colleague who is 70% likely to tell everybody ‘You know, X is a hacker now, everything on his laptop is black, hahaha’ (he actually laughs, but is mostly serious about it)
13:05 < ElinKattunge> I have my terminal style 8,Monospace, green on black, and I get called a hacker for it...
13:06 < ElinKattunge> I don't want margins, I like my terminal space... not interested in detering people, that just restricts myself.
13:06 < hexnewbie> He's berated people for using mc (calling it Norton Commander).
13:07 < searedvandal> norton commander <3
13:07 < ElinKattunge> People :/
13:07 < ElinKattunge> What is wrong with the world we live in?
13:08 < searedvandal> mc was/is more or less a nc clone though
13:08 < ElinKattunge> And it like goes from calling you a hacker cause they've seen your netbook screen, to pressing the power button for on me while I was working for shits and giggles
13:09 < ElinKattunge> Luckily they were dumb enough that all I had to do was disable the SIG trigger in my Linux environment so they had to hold the button.
13:09 < sh1ro> try using a green terminal anywhere on a college campus, public library area or coffee shop with black on white then green
13:09 < ElinKattunge> But then I wanted to actually turn off that button, I didn't like having an easily accessible power switch on my netbook in the end.
13:09 < sh1ro> i really need to fix that X'D
13:09 < ElinKattunge> I have
13:10 < hexnewbie> ElinKattunge: I suspect you could treatened to sue them for hacking for pressing the power button in a serious voice.
13:10 < ElinKattunge> Been in COSTA with my terminal
13:10 < searedvandal> never had anyone comment on my UI choices
13:10 < searedvandal> except for other linux users on the interwebz, of course.
13:10 < sh1ro> thing with spice is messing up what i'm typing
13:11 < ElinKattunge> Like one time I was just walkin' down the street, as you do, minding my own friggin business, when some reason guy who was walking past me said "Hacker", and I said "What do you mean?", and he replied "I've seen you on the bus with your lime green text, that's sick!"
13:11 < sh1ro> searedvandal: really? i've had random people walk up to me for using green on black and cyan on black atleast half a dosen t imes
13:12 < ElinKattunge> and I don't know whether that is a good thing, or a bad thing.
13:12 < sh1ro> yeah it's kinda weird to respond, this is just what updates look like on my computer
13:13 < searedvandal> sh1ro, nope. normal people don't really care, at least not in places where I go. only people who likes to comment on my UI choices are other linux users as I mentioned. those are some harsh critics.
13:13 < ElinKattunge> Fuck critics
13:14 < ElinKattunge> A UI is a personal choice, it is not this way or that way. It is whatever you make it to be, and whatever your friends make their UIs to be. Don't take critics as they are bringing you down on your personal environment, the environment which works for you.
13:15 < jim> ElinKattunge, all cool... except... please watch your language while you're here
13:15  * sh1ro will critique the security of ones UI
13:15  * sh1ro plugs wayland and zephyr
13:15 < searedvandal> oh, I don't care what other people say. I use whatever works for me. Like nano. I love me some nano action.
13:15 < ElinKattunge> jim: My apologies...
13:16 < ElinKattunge> Where is Wayland up to? When are distros gonna start using it as their default manager?
13:16 < jim> ElinKattunge, no problem; thanks for considering.
13:16 < revel> Doesn't Fedora already use it by default?
13:16 < jim> well it's gotta be stable firsrt
13:17 < searedvandal> I believe Fedora ships with Wayland as default.
13:17 < sh1ro> and ubuntu
13:17 < ElinKattunge> searedvandal: Just take the critics as ideas to think about and decide for yourself whether or not they will serve a good purpose for you.
13:17 < searedvandal> Ubuntu tried on 17.10, but went for xorg on the 18.04 release
13:17 < sh1ro> and you can use it on gentoo or arch if you want
13:17 < ElinKattunge> Ubuntu 18.04 Server sux
13:17 < sh1ro> it's more the DEs that need support
13:17 < revel> I'm sure you can use it on most distros.
13:17 < ElinKattunge> The installer doesn't allow you to skip the Network config to be put in later.
13:18 < revel> You're just not given a lot of DEs/WMs that support it.
13:18 < sh1ro> so throw money at XFCE
13:18 < ElinKattunge> So it is mature?
13:18 < revel> I don't use it, so I only know of two, Sway (i3-ish) and Gnome.
13:18 < sh1ro> kde
13:18 < revel> Oh, and KDE.
13:18 < ElinKattunge> I have used it a little bit in the past, but is it actually mature?
13:18 < sh1ro> they're slowly getting things more wayland
13:18  * bittin pets ElinKattunge 
13:19 < ElinKattunge> Also, I use i3wm, and I heard there are forks with Wayland support.
13:19 < hassoon> 'morning
13:19 < searedvandal> yeah, sway is the i3 fork for wayland
13:20 < sh1ro> those are really the main ones. flux doesn't need to be ported, mate maybe... those are pretty much all the common ones
13:21 < ElinKattunge> I heard Wayland has an in-built X server for the purpose of providing backward compatibility to Xorg apps, however last time I used Wayland, I could never get Xorg apps to work.
13:21 < revel> Might have to install Xwayland seperately.
13:21 < revel> It's provided by the Xorg package over here.
13:21 < ElinKattunge> Aaaah
13:23 < jim> ElinKattunge, yeah, there's supposed to be a wayland i3wm
13:23 < searedvandal> swaywm.org
13:23 < revel> Does devtmpfs have a nodev mount option? :^)
13:23 < hpotter> :-)
13:25 < ElinKattunge> I'm gonna install Sway and Wayland into a VM
13:37 < sensibel> I want to extend root partition as it is running out of space, how this can be done ?
13:42 < phogg> sensibel: is the partition an LVM volume? If not is there empty space on the disk after the end of the partition?
13:42 < jim> yep that's where I was just about to go
13:42 < phogg> sensibel: if not, rather than extend the partition you could simply mount some extra space at a place in the filesystem where you need it.
13:43 < sensibel> phogg: /dev/mapper/centos-root xfs        50G   50G  504M 100% /
13:43 < iflema> bingo
13:44 < phogg> sensibel: e.g. if /var is 25% of your disk space you could make a new partition, put a filesystem on it and then mount it on /mnt/tmp, copy everything from /var in to it, mount --bind /var to /mnt/tmp2, umount /mnt/tmp and re-mount it on /var, then rm -r /mnt/tmp2/* to clean up the data on the original disk
13:44 < phogg> sensibel: ah good, LVM it is
13:44 < phogg> that makes everything easier
13:44 < jim> sensibel, so you're very low on disk space... what are your other mounts?
13:44 < phogg> sensibel: does your volume group have any free space to allocate?
13:44 < iflema> wheres /home
13:45 < iflema> oh centos
13:45 < iflema> ugh
13:45 < phogg> CentOS isn't fun but LVM is the same everywhere.
13:46 < jim> maybe we add a pv somewhere if his vg isn't enough right now
13:47 < iflema> i avoided for ever, now i just go LVM on luks everywhere
13:47 < jim> sensibel, are you pretty familiar with lvm?
13:48 < ksk> hola
13:48 < jim> hi
13:48 < iflema> he might be outy
13:48 < fiter> phogg: how to whether root is lvm or not ?
13:48 < fiter> *check
13:49 < iflema> *He
13:49 < pingfloyd> fiter: try typing 'lvs'
13:49 < jim> yep he sems top be gone
13:49 < iflema> later...
13:51 < ksk> I have some veth devices of containers, some have routable ipv4-addresses. Right now I setup routes to these devices via network/interfaces - but this kind of seems 1980 style to me. Do you know of a (lightweight if possible at all) solution to this?
13:51 < fiter>   LV   VG     Attr       LSize  Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert   home centos -wi-ao----  3.58t                                                       root centos -wi-ao---- 50.00g                                                       swap centos -wi-ao----  4.00g
13:51 < pingfloyd> fiter: yep, it's an lv
13:51 < iflema> 3t
13:52 < fiter> pingfloyd: how can I add more space to it?
13:52 < jim> fiter, are you pretty good with lvm?
13:52 < pingfloyd> fiter: so you could do something like lvexpand -r -l +20G /dev/centos/root
13:52 < fiter> jim: no
13:52 < pingfloyd> s/lvexpand/lvextend/
13:53 < jim> ok, you've seen that normally a filesystem occupies a partition?
13:53 < fiter> jim: yes
13:53 < fiter> [dellemc@localhost reads]$ s/lvexpand/lvextend/ bash: s/lvexpand/lvextend/: No such file or directory
13:54 < pingfloyd> fiter: lvextend -r -l +20G /dev/centos/root
13:54 < hassoon> god damn it 4G of RAM is not enough anymore nowadays
13:54  * hassoon eyes his debian
13:54 < jim> fiter, ok... with LVM, instead of putting filesystems in partitions directly, you put 'LVM physical volumes' in partitions, then you make a 'volume group' (which is just a list of physical volumes), and put physical volumes in the volume groups, then you can make 'logical volumes', and these are allocated out of the volume groups, and then you put filesystems in lvs by formatting them, then you can also mount them
13:54 < phogg> hassoon: depends on what you're running
13:54 < pingfloyd> hassoon: what are you running that is so memory hungry?
13:54 < iflema> hassoon: ^^
13:55 < hassoon> ending up always with ~0.9G of free RAM
13:55 < phogg> hassoon: how are you calculating free ram?
13:55 < pingfloyd> it's not free that matters, it's available
13:55 < ExtremeFMan> how do I concatenate files? since they are text files, I suppose I can just "cat >>" - however, there is probably a way similar to the join command on Windows?
13:55 < hassoon> pingfloyd: meh few firefox tabs, few terminal windows, and the visualcode code editor
13:55 < pingfloyd> hassoon: paste free -m output
13:55 < hassoon> don't blame me for choosing that editor, i can vim whenever i want anyway, but oh well
13:55 < hassoon> this is still silly
13:55 < fiter> jim: ok
13:56 < hassoon> pingfloyd: meh here http://paste.debian.net/1032844/
13:56 < BluesKaj> Hey folks
13:56 < phogg> hassoon: Firefox is somewhat memory hungry. I'm betting that's most of the problem.
13:57 < hassoon> phogg: oh yeah? what would you say about chromium then ?
13:57 < hassoon> phogg: starving for memory ?
13:57 < jim> fiter, those are the very basics, and doesn't even name the commands... as you've seen, lvs lists your logical volumes... likewise, vgs lists your volume groups, and pvs lists your lvm physical volumes
13:57 < pingfloyd> phogg: firefox isn't usually that bad unless you have a million tabs open
13:57 < hassoon> pingfloyd: agreed
13:57 < phogg> hassoon: All browsers are either memory hogs OR non-modern. Take your pick.
13:57 < pingfloyd> it's not light, but not that insane on usage either
13:57 < fiter> jim: ok
13:58 < hassoon> phogg: meh what about...textual browsers :D
13:58  * hassoon highfives lynx
13:58 < pingfloyd> hassoon: you should see what's the worse memory hog with top or htop
13:58 < hassoon> pingfloyd: i knw sure
13:58 < hassoon> <3 htop
13:58 < jim> fiter, you're pretty lucky to have your system on lvm, it makes things really easy to add storage, and do other things as well
13:59 < pingfloyd> hassoon: I found it surprising how much memory your env requires.
13:59 < fiter> jim: ok
13:59 < phogg> hassoon: non-modern, like I said
13:59 < fiter> jim: does lvm allows raid1?
13:59 < pingfloyd> phogg: yeah, that's the dilemma
14:00 < jim> you can make a raid1, and put pvs on it
14:00 < hassoon> pingfloyd: i3wm here fyi
14:00 < jim> or even -a- pv
14:00 < fiter> jim: how can I check whether sytem is RAID1 currently or not ?
14:00 < pingfloyd> hassoon: I didn't mean DE, I'm mean the software environment
14:00  * hassoon donates some brain cells for his inspiron laptop :D
14:00 < hassoon> now it does have a higher RAM
14:01 < jim> I never done raid before... so it'd be better to ask another person... but do you have other questions?
14:01 < hassoon> pingfloyd: yeah i can run things on vim and do things the linux-way, but oh well..i need to do this one task very very fast, so a graphical dumb editor would help better
14:01 < phogg> I don't usually do raid with LVM, so I'd have to look it up.
14:01 < fiter> jim: I have two volumes of 4T and pvs shows [dellemc@localhost ~]$ sudo pvs   PV         VG     Fmt  Attr PSize  PFree   /dev/sda3  centos lvm2 a--  <3.64t    0
14:02 < hassoon> pingfloyd: around ~400M is required at least for this one editor ahah
14:02 < fiter> jim: currently I want to extend root
14:03 < pingfloyd> hassoon: yeah, it usually comes down to a few programs like that in the environment.  All those features comes with their resource requirements.
14:03 < jim> fiter, could you do this: (pvs ; echo '===========' ; vgs) | nc termbin.com 9999
14:03 < phogg> hassoon: I consider 8G to be a minimum amount of RAM for anyone these days, and 16G a minimum if you plan to play games or write software.
14:03 < fiter> jim: [dellemc@localhost bin]$ sudo (pvs ; echo '===========' ; vgs) | nc termbin.com 9999 bash: syntax error near unexpected token `pvs'
14:03 < hassoon> pingfloyd: been woring with silly windows users, i blame them for making me get farther from vim/cli world
14:04 < hassoon> phogg: agreed on that one, 8G is pretty interesting
14:04 < jim> did I screw it up...
14:04 < jim> let's see
14:04 < pingfloyd> there's really no way around the ram requirements if the programs you like have steep requirements.
14:04 < phogg> jim: no
14:04 < blackbeard420> theres always swap
14:04 < jim> oh he put sudo
14:04 < phogg> fiter: you can't just pass that to sudo. Use sudo -s to obtain a root shell first, then run it.
14:05 < phogg> fiter: or you can do: { sudo pvs ; echo '========' ; sudo vgs ; } | nc termbin.com 9999
14:05 < jim> lemme check something
14:05 < fiter> jim: PV         VG     Fmt  Attr PSize  PFree   /dev/sda3  centos lvm2 a--  <3.64t    0  ===========   VG     #PV #LV #SN Attr   VSize  VFree   centos   1   3   0 wz--n- <3.64t    0
14:05 < sh1ro> there's ram compression
14:06 < sh1ro> but it costs a perf hit and has limits
14:06 < fiter> output of (pvs ; echo '===========' ; vgs) | nc termbin.com 9999
14:06 < phogg> RAM compression is great if you have extra CPU. Does wonders.
14:06 < jim> ack :)
14:06 < phogg> jim: it writes to stderr )-:
14:06 < fiter> sh1ro:  where is ram compression?
14:07 < phogg> except that for me it does not...
14:07 < fiter> phogg: it gave a link that I opened in browser
14:07 < phogg> fiter: why didn't you paste the link, then?
14:07 < jim> what does? pvs?
14:07 < sh1ro> it's a kernel option
14:07 < fiter> phogg: http://termbin.com/5sbu
14:07 < phogg> fiter: that's the point of using a pastebin. So you don't spam the channel with garbage.
14:07 < sh1ro> iirc there's no way to do it from userland
14:07 < fiter> phogg: ok
14:07 < hassoon> lmao getting random tens of dumb friendship invitations in my fake female FB profile
14:08 < jim> ok
14:08  * hassoon laughs at the worold
14:08 < hassoon> * world
14:08 < jim> theres nothign free
14:08 < sh1ro> hassoon: but how many of them are advertiser accounts or other fake accounts
14:08 < hassoon> sh1ro: meh who knows, mostly are male users according the their pictures
14:08 < pingfloyd> hassoon: there's a sucker born every minute
14:09 < fiter> jim: https://paste.linux.community/view/3df7ffbf
14:09 < jim> fiter, there's a problem... the problem is, there's no space in your vg to allocate to add to your root lv
14:09 < phogg> hassoon: you should maybe look in to using zram https://wiki.gentoo.org/wiki/Zram
14:09 < pingfloyd> fiter: you need to add another PV to your VG
14:10 < hassoon> phogg: will give it a shot, thanks
14:10 < pingfloyd> or reduce another lv to make room in that vg
14:10 < jim> yeah, youve got no space in your lvm setup right now... the good thing about lvm is it's easy to add if you have some space somewhere
14:10 < phogg> hassoon: tgere's also https://wiki.debian.org/ZRam but it's a bit dated
14:10 < pingfloyd> reducing an lv, is going to require taking the fs offline that you'll be manipulating
14:11 < fiter> pingfloyd: your command https://paste.linux.community/view/cd5314c5
14:11 < phogg> fiter: do you have any disks which have unused space on them?
14:11 < pingfloyd> unless the underlying fs actually support online shrinking
14:11 < fiter> phogg: let me check
14:12 < phogg> fiter: and when I say "unused" I mean "not part of any partition", preferably an entirely empty disk.
14:12 < hassoon> phogg: meh it's not unusual to see something dated in gnu/linux debian ;)
14:12 < fiter> phogg: see if it helps https://paste.linux.community/view/5739f3da
14:13 < jim> well if he's got room on that disk that would probably be preferable (at least the way I'm used to doing things)
14:13 < fiter> phogg: or can you tell me how to check for the free disk?
14:13 < chedi1> join #haskell
14:13 < pingfloyd> fiter: when you first setup lvm, it's best to extend LVs as required and leave excess storage unallocated in the vg
14:13 < phogg> hassoon: and/or also zswap https://wiki.archlinux.org/index.php/zswap
14:14 < fiter> pingfloyd: ok, but what can be done now?
14:14 < pingfloyd> fiter: there's two reasons.  1. so you can utilize lvm snapshot and 2. it's less intrusive to extend FSes and lv than to shrink them.
14:14 < jim> fiter, could you run: df -h | nc termbin.com 9999
14:15 < phogg> fiter: you would need to examine each disk's partition table
14:15 < jim> let's see what he's got :)
14:15 < pingfloyd> fiter: since you don't have any free extents left in your current VG, you need to either add more storage (PV) to your VG, or reduce the existing LV(s) to make room.
14:15 < fiter> jim: https://paste.linux.community/view/e3b7cc2e
14:15 < fiter> pingfloyd: I can free existing LVs
14:16 < fiter> phogg: how ?
14:16 < pingfloyd> you could reduce centos-home
14:16 < jim> fiter, maybe you can reduce your /home?
14:17 < jim> by (say) half a tb
14:17 < pingfloyd> you'll likely have to boot into a live env to do it though, since you'll need to shrink the underlying filesystem
14:17 < fiter> jim: half tb?
14:17 < jim> pingfloyd, maybe he can unmount /home if he boots into single user mode?
14:17 < pingfloyd> many filesystems support online expanding, but not shrinking
14:18 < fiter> jim: /home is already more TB free
14:18 < fiter> *more than
14:18 < jim> fiter, well you've got like 2.5t free
14:18 < fiter> jim: yes
14:18 < jim> you could go the other way, and reduce /home by 2 tb
14:19 < fiter> jim:  yes I could do this
14:19 < jim> that would leave half a tb free in /home
14:19 < fiter> jim:  right
14:20 < fiter> jim: and from that 2TB i can extend root?
14:21 < mustu> guys, where can I generate the complete list of all a0, a1, a2, a3 hex human readable form in the auditd log
14:21 < jim> fiter, yes, for example you could add 20g
14:21 < fiter> jim: it does not require to go offline ?
14:21 < jim> then you'll have 2tb - 20g free in the vg
14:22 < jim> wanna do that?
14:22 < fiter> jim: right
14:22 < fiter> jim: yes
14:23 < jim> ok, I don't know what centos offers as boots, but maybe you have a maintainance mode? (aka single-user mode)
14:23 < jim> are you ircing from that machine?
14:24 < pingfloyd> the bottom line is you have to be able to unmount /home somehow to shrink its FS
14:24 < jim> yeah
14:24 < fiter> jim:  yes
14:24 < jim> ok, is there another machine available that you can irc to here from?
14:24 < fiter> jim:  but let me go offline and then I will connect irc from another machine
14:25 < jim> ok see you when you get back
14:25 < fiter> pingfloyd: unmounting /home will lose data on /home ?
14:26 < pingfloyd> fiter: if you don't have a single user mode to be able to get to a point where you can unmount /home and end up having to boot off a usb stick, I'd recommend grml (as it will cover your emergency/admin needs really well out of the box)
14:26 < pingfloyd> fiter: you can't unmount /home when it's in use
14:26 < pingfloyd> fiter: that's the crux of it
14:26 < jim> (now I don't know exactly how I'm gonna have him umount his /home... maybe some one else should decide?
14:26 < iflema> single
14:26 < pingfloyd> unless the FS on it, supports online shrinking (which I don't know of any that do)
14:26 < iflema> s
14:27 < fiter> pingfloyd: what do you mean by single user mode ? currently I am only one using this machine
14:27 < fiter> so its single user mode?
14:27 < pingfloyd> fiter: single user mode is a special init mode
14:27 < iflema> kill X and log in as root
14:27 < jim> iflema, you can probably do that if you log out from any plain user logins you have and just have a textual root login
14:27 < pingfloyd> meant for recovery
14:27 < iflema> boot into single
14:28 < fiter> iflema: so i options for single user mode comes while booting?
14:28 < pingfloyd> it's basically a minimal emergency enviroment.  I don't even bother single user mode though.  I just boot grml and do whatever needs to be done.
14:29 < jim> fiter, it's a maintainance mode, you would put in root password then you get a root shell
14:29 < pingfloyd> the idea with single user mode, is that /home won't be mounted and freed up to make the changes you need.
14:29 < fiter> thats ok
14:30 < fiter> going offline will join soon from another machine
14:30 < iflema> hola
14:31 < jim> well on many systems, it will mount stuff from the fstab... since root is the only one logged in, /home is not busy and can be unmounted
14:31 < fiter> pingfloyd: does it guide is right for single user mode ? https://www.centos.org/docs/5/html/5.2/Installation_Guide/s1-rescuemode-booting-single.html
14:31 < pingfloyd> jim: that's also a good example of how ubuntu's logic is flawed in how they disable root account and have the user use sudo for everything requiring root access.
14:31 < jim> fiter, ok cool
14:31 < fiter> jim: thanks
14:32 < pingfloyd> the typical ubuntu user would be screwed here since they wouldn't know what the root password is.
14:32 < jim> fiter, that wasn't an answer to your question though :)
14:32 < pingfloyd> since it's not set (disabled)
14:32 < fiter> jim:  which question?
14:33 < pingfloyd> well, screwed until they gave root a password
14:33 < jim> this question: "does it guide is right for single user mode ?"
14:33 < iflema> huh
14:33 < fiter> jim: oh
14:33 < fiter> jim: then how to login as single use mode?
14:34 < pingfloyd> fiter: that's going to be in your boot menu
14:34 < fiter> pingfloyd: ok
14:34 < pingfloyd> fiter: e.g., a menu entry you can choose from the grub menu
14:34 < fiter> ok
14:34 < jim> fiter, best thing right now, is leave the machine running in the mode it's in right now, and come back from the other machine
14:34 < fiter> jim: ok
14:34 < iflema> press  e  at the grub menu on the kernel you booting and add    single    to the end and then    ctrl + x
14:35 < jim> pingfloyd, oh, ubuntu's sudonly logic
14:35 < iflema> end of kernels line
14:36 < jim> iflema, let him come back first, then he will be able to hear us all the time
14:36 < jim> then he can reboot the machine and tell us what's in the menu
14:37 < iflema> the default...
14:37 < iflema> probably
14:37 < iflema> so what do ubunutards do
14:37 < iflema> set a root pAssword first?
14:37 < iflema> wat
14:38 < iflema> can you not just use the password?
14:39 < iflema> *Ubuntards
14:39 < jim> come on, doin't be mean
14:40 < hassoon> 'Ubuntards'
14:40 < hassoon> lmfao
14:40 < jim> that's mean though
14:41 < iflema> dont you hate it when you have been using sudo and for what ever you log in as root and end up walking away from it...
14:42 < jim> well if I can lock my apartment and no one's in it...
14:42 < iflema> pro pain?
14:42 < iflema> m
14:43 < pingfloyd> in a case where you have to worry about walking away, you probably want to set your sudoers so it will require authentication each time, or the timer is a lot shorter.
14:44 < pingfloyd> though at that point, you might want to not bother with sudo and use su anyway.
14:44 < iflema> defaults are ok there
14:44 < pingfloyd> ubuntu has trained users to fear su
14:45 < iflema> pingfloyd: can you not just use _the_ password in case of    single    ?
14:45 < MrElendig> pingfloyd: no, it has trained users in doing sudo su
14:45 < ElinKattunge> Has anyone here done their own distro? Again for fun?
14:46 < pingfloyd> MrElendig: actually ubuntu tells them to use sudo -i
14:46 < MrElendig> pingfloyd: 90% of the guides says sudo su, (their official docs also do in places)
14:46 < iflema> wat
14:46 < MrElendig> though the official docs have slowly been getting fixed
14:46 < rozick> Hey
14:46 < rozick> tar -C /var/www/html/servers/tmp/ -cvzf test.tar.gz /var/www/html/servers/tmp/user3245
14:46 < rozick> I'm running that command
14:46 < pingfloyd> that's pretty sad
14:46 < rozick> it's working, but the problem is test.tar.gz is ending up in the servers directory
14:47 < rozick> and not in the tmp directory
14:47 < Hych> There's a certain video file (.mp4, .avi..etc) that I'd like to download through a direct link. However, I don't want to download the entire file but only that part of the video which goes from 05:20 to 07:00 for example. Is this possible?
14:47 < ElinKattunge> What is the difference between sudo su and sudo -i?
14:47 < jim> iflema, ubuntu I don't think it even asks for a root password when you install, and #ubuntu will kickban you if you sugest setting a root password
14:47 < rozick> sudo su changes you to root
14:47 < Hych> jim: do you have any idea?
14:47 < MrElendig> ElinKattunge: you become root to become root
14:47 < Hych> it seems that this is impossible
14:47 < rozick> sudo -i will just allow you to keep using sudo for as long as the terminal is open
14:47 < MrElendig> rozick: no
14:47 < rozick> without having to keep retyping your password
14:47 < iflema> jim: install?
14:47 < ElinKattunge> I don't normally see a reason to change to root anyway.
14:47 < Hych> anyone?
14:47 < MrElendig> rozick: you do become root when you use sudo -i
14:47 < pingfloyd> ElinKattunge: sudo so is superfluous
14:48 < rozick> oh
14:48 < pingfloyd> *sudo su
14:48 < rozick> well anyone know an answer to my tar question?
14:48 < ElinKattunge> sudo is nice
14:48 < rumpel> is "su" still a thing???
14:48 < ElinKattunge> But it gotta be configured properly
14:48 < MrElendig> rozick: sudo su == is like crossing the stream to get water, only worse
14:48 < pingfloyd> rumpel: when did it stop being?
14:48 < rumpel> pingfloyd, in the 90s, I think
14:48 < KINUKYASHA_> hello i am trying to root my lg g6 and i am running linux on oracle vm and am having trouble finding the usb drive connected to my pc
14:48 < ElinKattunge> I like to use it to permit people access to specific administration commands, but not all.
14:48 < jim> iflema, I think when you install ubuntu it does not set up root as an account you can ever log in as (with some work you can change that)
14:48 < KINUKYASHA_> [android@fwul lglaf]$ ./partitions.py --dump misc.img misc Traceback (most recent call last):   File "./partitions.py", line 459, in <module>     main()   File "./partitions.py", line 406, in main     comm = lglaf.autodetect_device()   File "/home/android/lglaf/lglaf.py", line 441, in autodetect_device     return USBCommunication()   File "/home/android/lglaf/lglaf.py", line 299, in __init__     raise RuntimeError("USB device not
14:49 < pingfloyd> rumpel: you've been drinking the ubuntu koolaid it sounds like
14:49 < MrElendig> KINUKYASHA_: install the non-free gust addons
14:49 < MrElendig> guest
14:49 < iflema> yeah... but im curious by default, when you add    single    what happens?
14:49 < MrElendig> KINUKYASHA_: #vbox can help you with doing so
14:49 < pingfloyd> rumpel: this whole idea that it's bad to use su or that is obsolete is ubuntu propaganda
14:49 < KINUKYASHA_> thank you
14:49 < iflema> computer sez no
14:50 < rumpel> pingfloyd, don't believe the fake facts. Their fake.
14:50 < iflema> anyway whatever
14:50 < MrElendig> KINUKYASHA_: also get a better irc client, one that knows that irc has a max message lenght
14:50 < iflema> where is the centos customer
14:50 < MrElendig> KINUKYASHA_: also: use a pastebin to post things (not pastebin.com)
14:51 < pingfloyd> ElinKattunge: that's really the reason for having sudo at all
14:51 < ElinKattunge> Plus, I like to have root access disabled
14:51 < ElinKattunge> I'd rather someone have to log into someones account before they can access the root account.
14:51 < pingfloyd> ElinKattunge: that you don't have give anyone that needs a few superuser abilities, the root password.
14:51 < jim> iflema, hmm, I dunno
14:52 < iflema> :D
14:52 < ElinKattunge> Agreed
14:52 < MrElendig> ElinKattunge: uhm, that is how sudo works
14:52 < MrElendig> you have to be logged in to use it
14:52 < ElinKattunge> Even I don't like having access to the root account, I just understand the reality that if I was to do away with root, I wont be able to administer my system.
14:52 < ElinKattunge> MrElendig: Exactly my point
14:52 < iflema> single?
14:53 < pingfloyd> ElinKattunge: that's going to backfire when you run into a real emergency
14:53 < ElinKattunge> I really do think sudo is a security feature
14:53 < iflema> S
14:53 < pingfloyd> ElinKattunge: sudo isn't more secure than using su
14:53 < ElinKattunge> There is always a way to gain root access to a system, even if I've to bruteforce it myself.
14:54 < pingfloyd> ElinKattunge: why put yourself through that?
14:54 < ElinKattunge> Though what?
14:54 < triceratux> ah yes. sudoperfluous ubuntuganda ;)
14:54 < ElinKattunge> I rarely have to bruteforce access to a system, infact almost never.
14:55 < pingfloyd> bruteforcing your own system
14:56 < ElinKattunge> On rare occasions, I've lost root access to my system. I've used Linux since 2007, and that's enough time to make even the most silliest of mistakes during learning about the system and how it works.
14:56 < fiter> jim: hi
14:56 < ElinKattunge> So yeah, I've had to bruteforce into systems which I own, due to past error, or experimentation.
14:56 < iflema> centos customer
14:56 < pingfloyd> ElinKattunge: seems like a lot of extra pointless work
14:57 < pingfloyd> ElinKattunge: why not just keep backups and not fear screwing up your system?
14:57 < ElinKattunge> Because why would I restore from a backup when there is no need?
14:57 < hexnewbie> sudo is more secure than su - it provides use_pty for secure privilege dropping to avoid TTY hijacking vulnerabilities, and provides safe execution of  files with arguments, not shell commands like su which require variable quoting and inevitably lead to insecure substitutions in the process
14:57 < MrElendig> he is a M
14:58 < ElinKattunge> When the solution is to break your way into the system?
14:58 < pingfloyd> I once spaced, and removed half of my home by accident.  Oops. had to restore some files.
14:58 < ElinKattunge> That's different.
14:58 < ElinKattunge> That's missing files.
14:58 < pingfloyd> making a requirement to break into your own system is so counter productive.
14:59 < ElinKattunge> Like I said, I've used Linux since 2007.
14:59 < ElinKattunge> I've gone from no-experience to efficient
14:59 < pingfloyd> I've been using it even longer than that, so
14:59 < ElinKattunge> Bruteforcing into systems was something I did in the early days of using Linux.
14:59 < ElinKattunge> When I was learning, and prone to mistakes.
15:00 < pingfloyd> humans are always prone to mistakes
15:00 < pingfloyd> if anything, the big mistakes you make later because of muscle memory.
15:00 < ElinKattunge> I still always am learning, you can't learn an entire Linux system, but atleast these days I've efficient and I don't have to bruteforce my systems, I don't fail to backup, etc.
15:00 < pingfloyd> it's that task you do a 1000 times that you're bound to have a brain fart with
15:00 < sensibel> pingfloyd: I am going to access the machine using single user mode
15:01 < pingfloyd> sensibel: yeah, I'd try that first as it is more convenient
15:01 < hassoon> azei'm gonna punch the heck out of my laptop so hard that my fist would probably get through the screen
15:01 < hassoon> i'm gonna *
15:02 < hassoon> why whould apach2 consume around 2G ?
15:02 < ElinKattunge> I usually write a lot of bash scripts to automate things, I have a script called webctl which allows me to manage nginx
15:02 < hassoon> now that i service-stopped it things went back fine
15:02 < pingfloyd> ElinKattunge: so you do backup?
15:02 < ElinKattunge> It sets up vhosts for me too
15:02 < ElinKattunge> Yeah, I'm smarter
15:02 < pingfloyd> ElinKattunge: I was going to say, you've been using it since 2007 and you still haven't learned that lesson?
15:02 < ElinKattunge> I gone from naive linux user to smarter linux user
15:02 < hassoon> ^ rofl
15:03 < hassoon> ElinKattunge: prove that
15:03 < sensibel> pingfloyd: I pressed F11 and now it have entered to boot manager
15:03 < ElinKattunge> I mean some systems I'll just let go, but only out of lack of money for backup storage.
15:03 < ElinKattunge> hassoon: No need
15:03 < pingfloyd> sensibel: you get to single user from the grub menu (after post).
15:03 < ElinKattunge> I'm not competing
15:04 < ElinKattunge> Nor am I trying to promote myself
15:04 < sensibel> pingfloyd: should continue with normal boot?
15:04 < sensibel> pingfloyd: after post ? what is meant by this ?
15:04 < ElinKattunge> hassoon: Though, I am going to be setting up a portfolio soon.
15:04 < ElinKattunge> As I want to enhance my prospects.
15:04 < hassoon> ElinKattunge: yeah good for oyu
15:05 < ElinKattunge> I'll give you a link when the blog is up and running, if you want something to read.
15:05 < pingfloyd> sensibel: it's the grub menu where can do single user mode, like outlined earlier 'hit e and append single to the line and ctrl+x to run it'
15:05 < hexnewbie> There are many more stages to go, from smarter user to God mode, where you change all your root passwords to iddqd and forgo any backups because you feel invincible
15:05 < jim> sensibel, ok, you're back
15:05 < sensibel> jim: yes
15:05 < birdbolt1> with git, is there a convenient way to `git add *` while ignoring everything thats in my gitignore file?
15:05 < sensibel> pingfloyd: oh
15:06 < birdbolt1> does * require -f or will it override the stuff in my gitignore command?
15:06 < jim> birdbolt1, git add . I think... also there's a #git here
15:06 < birdbolt1> oh nice thanks
15:07 < ElinKattunge> hexnewbie: lel
15:07 < jim> that might go deepo into the dir structure
15:08 < ElinKattunge> I'm really loving this qutebrowser.
15:08 < ElinKattunge> I had an issue with it last night, however I managed to fix it.
15:09 < FightingFalcon> Google analytics say that there are currently 100 active users in my website. But netstat -anp says there are 15 connections to port 80. whats up?
15:09 < MrElendig> keepalive is not that long
15:10 < MrElendig> sidenote: google analytics is evil
15:10 < MrElendig> you are giving out all your visitors to google for free
15:10 < FightingFalcon> hmm
15:10 < FightingFalcon> and whats wrong with giving all my visitors free?
15:10 < pingfloyd> such a good cog
15:11 < MrElendig> atleast earn some money from it if you want to screw over the privacy of your visitors!
15:11 < ElinKattunge> My net is slugging due to nasty download.
15:11 < FightingFalcon> huhhhhhhhhhh
15:11 < MrElendig> <insert evil laughter here>
15:11 < FightingFalcon> MrElendig, so i should INCREASE the keepalive of MY Varnish RIGHT?
15:11 < sensibel> pingfloyd: error can't find command single
15:12 < MrElendig> no
15:12 < pingfloyd> sensibel: here's an example https://www.cyberciti.biz/faq/grub-boot-into-single-user-mode/
15:13 < jim> sensibel. does your phone let you send picture files to your computer?
15:14 < FightingFalcon> MrElendig go on, then what?
15:14 < MrElendig> why care about it?
15:16 < pingfloyd> why don't you care about your visitor's privacy is the real question.
15:17 < sensibel> jim:  I am going to take and send it via imgur
15:18 < FightingFalcon> Most of my visitors are stupid idiots
15:18 < hassoon> FightingFalcon: and what are they visiting you for ?
15:18 < pingfloyd> for visiting your site?
15:18 < hexnewbie> FightingFalcon: That's like the second reason you've cited so far that people should stop visiting your site
15:20 < FightingFalcon> hexnewbie you again huh
15:20 < lostfile> hexnewbie: i had  stupid people vist my site all the time and it was quite amuseing
15:21 < birdbolt1> what license do i use on github if I dont want my code shared or distributed?
15:22 < birdbolt1> I working in my private repository for a startup, and all these licenses seem to have something allowing others to share somehow
15:23 < FightingFalcon> hexnewbie, how long have you been using this nickname?
15:23 < cxc99> does anybody know how to get a self-signed cert for ldap/tls to be "trusted" for nslcd?
15:23 < cxc99> in RH7 ;)
15:23 < hexnewbie> birdbolt1: That's the point of free and open source software licenses. Do private repositories require you to choose one? If not, simply don't. If yes, that's surprising.
15:24 < sensibel> jim: see if it helps
15:24 < sensibel> jim:  https://imgur.com/a/TTY0XPP
15:24 < birdbolt1> no they dont, first time actually trying to choose a license that PREVENTS sharing, so i had no clue what to do lol
15:25 < sensibel> pingfloyd: here is the screen comes when I press e at grub https://imgur.com/a/TTY0XPP
15:26 < hexnewbie> FightingFalcon: Possibly since forever
15:28 < jim> sensibel, maybe you put the word single at the end of the linuxefi line?
15:29 < sensibel> jim: ok
15:30 < jim> sensibel, it's worth a try... boot it that way... what should happen is at some point it asks for the root password
15:30 < MrElendig> birdbolt1: don't put your code in a public repo in the first place
15:30 < MrElendig> problem solved
15:30 < MrElendig> birdbolt1: sitenote: no license == no one is allowed to use your code
15:30 < MrElendig> copyright is restrictive by default
15:31 < sensibel> jim: it asked and I gave it the password
15:31 < jim> ok, what did it do next?
15:32 < sensibel> root@localhost cursor blinking
15:32 < MrElendig> so you are logged in
15:32 < jim> ok, that's your root shell prompt
15:32 < sensibel> jim: yes
15:32 < jim> try: umount /home
15:32 < birdbolt1> MrElendig, its private. First private repo. By habit, i was treating it like a public one lol
15:33 < hexnewbie> birdbolt1: You give license to people to allow them to do things, not to prevent them from doing them. In the past, copyright notice or copyright registration (earlier) prevented people from doing so, but those are automatic now in many/most? countries. This differs from EULA (‘license agreement’) which is an agreement you need to agree to to be allowed/licensed to use the software, the legality of which is disputed.
15:33 < hexnewbie> And probably varies wildly
15:34 < jim> sensibel, try: umount /home
15:34 < sensibel> jim: ok
15:34 < birdbolt1> I just wrote up a license with explicitly states ownership and rights to all the code, no modifications, etc etc
15:34 < birdbolt1> now not sure if i should just remove or leave it
15:35 < jim> sensibel, what happens when you do that?
15:35 < hexnewbie> birdbolt1: For that you just need ‘ProgramName, Copyright (C) 2018 PersonOrCompanyName. All rights reserved.’
15:35 < hexnewbie> s/need/may use/
15:35 < sensibel> jim:  the target is busy
15:36 < sensibel> jim:  let me share the pic
15:36 < jim> ok
15:36 < birdbolt1> aahh nice thanks @hexnewbie
15:37 < jim> back in a bit...
15:37 < birdbolt1> It's a repo with various docker services, not one "program". Does using the name of the repo suffice?
15:37 < sensibel> jim: here it is https://imgur.com/a/cfR4Adb
15:38 < sensibel> pingfloyd:                          ^^
15:38 < ddoobb> So I deleted the boot partition accidentally, and more it says Operating system wasn't found. What does this mean?
15:39 < hexnewbie> birdbolt1: I'm not a copyright lawyer, can't say, but from what I can tell, the name or the entire notice isn't needed at all. I do include a copyright notice in *each* of my files, but that's because of GNU/FSF's recommendations on how to apply free software licenses (kind of opposite of what you're after)
15:39 < triceratux> ddoobb: sounds like you know what it means
15:40 < ddoobb> No I mean what needs to be fixed? MBR?
15:40 < ddoobb> There are other OSs on it. I just deleted the first
15:41 < birdbolt1> thank you very much hexnewbie
15:42 < SaEeDIRHA> hey guys , i was wondering if i mix usb3 slot and usb2 slot for lvm partition groups would that cause any problem (other than maybe slows down the read/write) ? because i have 2 slots with usb3 and 2 slots with usb2 on my pc, and i want to create lvm group using them all as one partition
15:43 < pingfloyd> SaEeDIRHA: in other words, you want to make some usb stick your PVs for a single VG?
15:43 < pingfloyd> *sticks
15:45 < SaEeDIRHA> pingfloyd, they are external usb hard drives, but yes , so i have 3 external usb hard drives that support usb3 , but only 2 usb3 slot on the pc and others are usb2 slots
15:46 < pingfloyd> SaEeDIRHA: in theory it should work
15:46 < pingfloyd> as far as caveats and/or bugs, you're probably just going to have to try it and see.
15:46 < SaEeDIRHA> pingfloyd, yeah i guessed that as well but was wondering if it would cause any problem , guess i wouldn
15:47 < SaEeDIRHA> pingfloyd, yeah i guessed that as well but was wondering if it would cause any problem , guess i wouldn't know untill i try :)
15:47 < pingfloyd> SaEeDIRHA: maybe you'll get lucky and stumble across someone that has tried such a setup before
15:47 < SaEeDIRHA> i just bought a cheap mini pc that has only 2 usb-3 slots which i want to use as a file server
15:48 < SaEeDIRHA> i am going to try it to so if any problems would happen :)
15:48 < pingfloyd> SaEeDIRHA: you might consider letting everyone know how it went
15:48 < jim> sensibel. ok, I think we have to resort to booting some kind of rescue disk or something like that, that thing that pingfloyd mentioned is probably a good choice... do you have a usb memory stick handy that you can overwrite?
15:48 < pingfloyd> I'm kind of interested to find out
15:49 < sensibel> jim:  yes I have
15:49 < peetaur> what if the usb3 goes too fast and due to backpressure, some of the files go back in the usb2 disks and corrupt them
15:49 < SaEeDIRHA> pingfloyd, will let you know , and thanks :)
15:49 < pingfloyd> sensibel: jim: https://grml.org/
15:49 < pingfloyd> SaEeDIRHA: good luck
15:50 < sensibel> pingfloyd: I should download it and copy it in the usb and attach the usb to the system and reboot from the usb?
15:51 < jim> sensibel, yeah. did you say you have an i5?
15:51 < pingfloyd> sensibel: yeah, you'll use dd though to copy it to the usb stick
15:51 < pingfloyd> sensibel: it has instructions on the site too
15:52 < sensibel> jim:  i5 ? i5 phone?
15:52 < jim> cpu
15:53 < sh1ro> also there is no such phone
15:53 < jim> is it intel based, and 64 bit?
15:53 < sensibel> sh1ro: there is i6
15:54 < sensibel> jim:  yes I have the the machine from which I am ircing is i7
15:54 < ddoobb> Can you make a live usb from Windows iso on Linux? How?
15:54 < jim> sensibel, ok, what about the machine you want to move some storage of?
15:55 < sensibel> jim: its a server with 32 cores on two processors and 127 GB ram
15:55 < sh1ro> ddoobb: dd if=/path/to/filename.iso of=/dev/sdX bs=512k
15:55 < jim> which processor?
15:55 < sh1ro> as root
15:56 < ddoobb> sh1ro this will make it bootable too?
15:56 < jim> ddoobb, probably the image is already bootable
15:56 < sh1ro> yes, replace the path name and device names accordingly
15:56 < sensibel> jim:  it dell poweredgeT440
15:56 < pingfloyd> ddoobb: makes sure you get the of= arg right
15:56 < triceratux> ddoobb: thats the eternal question. the answer seems to be: https://itsfoss.com/bootable-windows-usb-linux/ http://multibootusb.org/ never tried it myself
15:56 < sh1ro> this is how i make windows install sticks all the time
15:57 < jim> sensibel, which processor does it have? two i7s or somethjing like that?
15:58 < sensibel> 16 cores on each processor
15:58 < jim> got it... what is the processor called?
15:58 < sensibel> jim: can I check it now for your convenience through a command?
15:58 < sh1ro> if you're aiming to live boot and rescure a system though i reccomend hiren's boot cd or trinity rescue kit for windows; bitdefender or kaspersky for viruses; linux usually just use your install iso
15:59 < jim> yes... you can do: cat /proc/cpuinfo | nc termbin.com 9999
16:00 < pingfloyd> or run lscpu
16:00 < jim> sh1ro, all we're doing is resizing a filesystem, then the lv it's in... that's all we need to do with a rescue cd
16:01 < sensibel> jim intel R XEon(R)  4110 @ 2.10 Ghz
16:01 < ddoobb> I already tried Ultimate boot cd but nothing worked so Windows startup repair seems like the best option
16:02 < jim> sensibel, I'd like to see the whole cpuinfo output
16:02 < purplex88> how can i access usb pen drive in ubuntu?
16:02 < jim> I guess that must be an amd64 based
16:02 < purplex88> where does it show?
16:02 < sensibel> jim: can a picture suffice ?
16:03 < jim> sure
16:03 < ziggylazer> Hey all
16:03 < jim> purplex88, look at dmesg | tail right after plugging in the usb
16:03 < jim> hi
16:04 < purplex88> where is that?
16:04 < purplex88> actually what is that?
16:04 < jim> it's a command
16:04 < jim> I don't know if your linux will let yuou see it as nonroot
16:04 < purplex88> i copy "dmesg | tail" into terminal?
16:04 < ziggylazer> Anyone familiar with the finger user@host enumeration exploit on solaris?
16:05 < phogg> who still runs fingerd?
16:05 < pingfloyd> purplex88: run dmesg -w in a terminal and then plug in the usb device
16:05 < jim> ziggylazer, are you running solaris?
16:06 < ziggylazer> Not personaly no, but I need to understand it in order to try to get on the system
16:06 < purplex88> invalid option -- 'w'
16:06 < qzhong0605> Is there any method to build a vpn tunnel from a pfx file?
16:07 < jim> purplex88, you should man dmesg, it might be different on your system
16:07 < purplex88> i give up
16:07 < sh1ro> jim: there was someone else asking too figured i'd shotgun the answer and hit it all
16:08 < purplex88> in windows its so easy, in ubuntu its hell
16:08 < sensibel> jim: here it is https://imgur.com/a/aemjFE6
16:09 < ananke> ziggylazer: you're in the wrong place asking for help with that
16:09 < ziggylazer> jim if you got some experience and time and if you want to I'd love to pick your brain when/if you would get a min over
16:09 < sensibel> jim:  there are 31 such processors
16:09 < ziggylazer> ananke, yeah both Sec and metasploit is dead :/
16:09 < ziggylazer> But if its not ok I will not mention it
16:10 < eblip> hey if any of you guys are professional developers using python possibly postgresql or django...what distro do you use for development
16:10 < eblip> that offers proper stability and cutting edge packages for devs
16:10 < ziggylazer> eblip, use Ubuntu
16:10 < ananke> eblip: those two things are mutually exclusive
16:11 < ziggylazer> That too
16:11 < eblip> yes ananke i was thinking that
16:11 < triceratux> lubuntu 18.10 lxqt ftw http://i.imgur.com/x00megh.jpg
16:11 < eblip> debian seems great but using the repo is a bit outdated
16:11 < ananke> eblip: pick a distro you like that has tools you need. for cutting edge (which is silly imho), you can always use virtual environments, containers, etc
16:12 < jim> sensibel, ok... I can't tell what cpu that is :) best I guess is to try it... did you download the image yet
16:12 < eblip> yes not really cutting edge im after but quite modern would be good
16:12 < jim> ?
16:12 < ananke> base distro does not necessarily need to provide you with the cutting edge stuff
16:12 < eblip> i was thinking arch linux but i worry about the stability as i think its too cutting edge
16:12 < sensibel> jim: 3 min left
16:12 < jim> oh ok
16:12 < ziggylazer> eblip, if you can configure it then go for it
16:12 < ananke> eblip: nobody sane uses archlinux for 'professional' work
16:13 < eblip> yes thats what i thought ananke
16:13 < eblip> just too risky
16:13 < ziggylazer> But he will have his own system
16:13 < eblip> im currently using gentoo to be honest....but over past month i have had a couple of kernel upgrades...and a major profile switch which actually interfered with my development
16:14 < ziggylazer> eblip, the to be honest, it doesnt matter that much what you choose
16:14 < sensibel> jim: I think I should be back after some time as it will take some time downloading because downloading speed fluctuates, is it ok?
16:14 < ananke> eblip: archlinux would be even worse in that aspect
16:14 < eblip> i mean im totally happy with everything ...except the time it took to rebuild and downgrade the kernel the switch profiles and then 2 major kernel upgrades
16:14 < jim> sensibel, before you go...
16:14 < sensibel> jim: yes ?
16:14 < ananke> eblip: pick a _normal_ mainstream distro, eg ubuntu/fedora/etc. don't dork around with stupid rolling release distros
16:14 < jim> the next few steps...
16:15 < jim> you'll want to see what is its filename
16:15 < ddoobb> How long does this take usually?
16:15 < eblip> i do love gentoo...but like stated the past couple months did interfere with development
16:15 < sensibel> jim:  of the image ?
16:15 < jim> yes
16:16 < jim> do you know how to write it to the usb device?
16:16 < sensibel> jim: I can use rufus?
16:17 < longxia> purplex88: are you on ubuntu?
16:17 < jim> yeah, on windows?
16:17 < sensibel> jim:  yes
16:17 < pingfloyd> sensibel: I'd avoid those types of utils
16:17 < pingfloyd> sensibel: dd is the best option because it will make an exact copy
16:18 < ddoobb> how long does dd takke? theres no progres bar either?
16:18 < jim> pingfloyd, but he's running windows on that pc,
16:18 < sensibel> pingfloyd: then ? how use dd?
16:18 < ElinKattunge> SHOOT ME IN THE FOOT!
16:18 < pingfloyd> whatever util you use, you want to use one that an exact bit for bit copy
16:18 < sensibel> jim: no i have linux here
16:18 < pingfloyd> and doesn't add or change anything (like what unetbootin and many others do)
16:19 < sensibel> by your question on windows I meant using that software on windows
16:19 < pingfloyd> win32diskimager will make an exact copy
16:20 < jim> sensibel, rufus should work fine, if you're used to that. (pingfloyd, it works fine with netinstall images, it's not like unetbootin)
16:20 < pingfloyd> jim: does it do a plain raw copy?
16:20 < jim> I believe so
16:20 < ddoobb> rufus is a fine piece of software. Never failed for me
16:20 < sensibel> jim: completed whats next?
16:21 < sensibel> how to do this using dd pingfloyd ?
16:21 < sensibel> jim: image donwload completed
16:21 < jim> sensibel, you have to find out what linux has named your usb device
16:22 < sensibel> jim: ok
16:22 < jim> do you know how to do that?
16:24 < jim> sensibel, would you mind paying closer attention to the conversation?
16:25 < ddoobb> Will something happen if i terminate dd in pocess?
16:25 < sensibel> jim: i lost the usb then found it
16:25 < sensibel> jim: now i attached it to my system
16:26 < jim> ok, now look at dmesg | tail as root
16:26 < sensibel> jim:  it is /dev/sdb1
16:26 < jim>  /dev/sdb actually (the partition table on the device isn't relevent
16:27 < jim> good,
16:27 < jim> what is the filename of the image?
16:27 < sensibel> grml64-full_2017.05.iso
16:27 < jim> ok, the command is:
16:28 < jim> dd if=grml64-full_2017.05.iso of=/dev/sdb bs=1m
16:28 < sensibel> jim:  here it is https://paste.linux.community/view/3e3c5a1c
16:29 < jim> oh, wait, you want a progress indicator... how do you do that...
16:29 < sensibel> jim:  of=/dev/sdb1? or just of=/dev/sdb?
16:29 < sensibel> yes waiting
16:29 < jim> don't do that yet
16:29 < sensibel> jim: i am not so fast *
16:30 < sensibel> jim:  so haven't done it
16:30 < ElinKattunge> I want a program that will open and close the front door for my cat
16:30 < ddoobb> guys it's still not finished
16:31 < jim> how do you get a progress indicator using dd?
16:31 < longxia> jim: status=progress
16:31 < jim> thanks
16:32 < ddoobb> longxia what if it's already going on?
16:32 < jim> dd if=grml64-full_2017.05.iso of=/dev/sdb bs=1m status=progress
16:32 < jim> sensibel, ^^
16:33 < ddoobb> Usually 4 gb iso takes how long?
16:33 < sensibel> jim: ok
16:33 < jim> no idea... half hour?
16:33 < sensibel> jim: dd: invalid number: ‘1m’
16:34 < jim> ok one sec
16:34 < longxia> ddoobb: when its already going on you just wait without progress indicator and have faith
16:35 < jim> I think the M has to be a cap
16:35 < jim> so
16:35 < sensibel> ok
16:35 < jim> sensibel, dd if=grml64-full_2017.05.iso of=/dev/sdb bs=1M status=progress
16:35 < hans_> any distro that has optimized boot speed?
16:36 < hans_> i need to reboot a lot
16:36 < ddoobb> do you have to format the drive before dd?
16:36 < triceratux> tinycore
16:36 < jim> no I don't think so
16:36 < purplex88> longxia: yes, vbox
16:36 < sh1ro> gentoo, kernel mod your boot time to your hearts content
16:36 < jim> sensibel, is it working?
16:37 < longxia> purplex88: dmesg -w (not tail -w)
16:37 < purplex88> -w is invalid
16:37 < sensibel> jim: 590+0 records in 590+0 records out 618659840 bytes (619 MB, 590 MiB) copied, 0.368937 s, 1.7 GB/s
16:38 < purplex88> ubuntu 14.4 lts
16:38 < jim> so it's done?
16:38 < sensibel> jim: i think so
16:38 < jim> you have a prompt again?
16:38 < longxia> purplex88: 14??? ok so just do a dmesg before, and a dmesg after you plugged in the usb stick
16:38 < jim> before you take it out, type: sync
16:39 < sensibel> jim: yes
16:39 < longxia> purplex88: and see what got logged extra about a new device
16:39 < sensibel> just give me 15 minutes will be back after minutes jim
16:39 < purplex88> its ridiculous to use command lines just to access pen drive
16:39 < jim> sensibel, actually it's not going to be me
16:40 < longxia> purplex88: do dmesg | tail if you only want to see the last few lines
16:40 < jim> see if pingfloyd is still here when you get back, and if he still has time
16:41 < purplex88> why am i doing that
16:41 < purplex88> seeing lines and tails et.
16:41 < longxia> purplex88: well, i have to agree with you there but i also remember distro's which have automount for these things, just can't remember which ones.
16:41 < purplex88> etc.
16:41 < jim> what's the that which you're doing?
16:42 < triceratux> hrm thats weird http://toybox.landley.narkive.com/oCevuvwo/patch-implement-dmesg-w
16:42 < purplex88> ok so it needs mount
16:43 < longxia> purplex88: your are looking for a line similar "Attached SCSI removable disk" and line above that with sdb: sdb1
16:44 < longxia> purplex88: very much like what jim told sensibel the past few minutes
16:44 < longxia> purplex88: but the drive letters may very depending on your configuration
16:45 < purplex88> i'm getting an error "can't attach usb to VM ubuntu"
16:45 < longxia> purplex88: yes, it needs to be mounted. If you know the device, see man mount to have it mounted at your favourite location
16:46 < ddoobb> will aborting dd damage usb?
16:46 < jim> physically? probably not, but until you write it completely it won't work
16:47 < purplex88> my goal is to usb to transfer a file from ubuntu vm to windows
16:47 < purplex88> use usb*
16:47 < purplex88> but i guess i'll just re-download the file
16:47 < mawk> use usb ?
16:47 < longxia> purplex88: maybe you should ask #vbox about that, i have no experience wih vbox
16:48 < purplex88> instead of wasting 4 hours
16:48 < purplex88> to figure it out..
16:48 < mawk> why not use the file sharing capabilities of virtualbox ?
16:49 < purplex88> thats too i suppose some mumble jumble i'll look at that option too.
16:50 < purplex88> i never even got shared clipboard to work
16:50 < purplex88> nothing works..
16:52 < jim> purplex88, you sound pretty stressed right now... do I have that right?
16:53 < triceratux> ubuntu 14.04 is getting a little old by todays standards. theres 2 lts releases that are far better atm
16:53 < purplex88> luckily i still have workarounds i can use
16:54 < purplex88> not that stressed but now for i'll try to do things in simple way
16:54 < sensibel> jim: yes
16:54 < sensibel> its not you it was me
16:54 < kuri0> how can i convert hex codes to a real file ?
16:55 < pingfloyd> real file?
16:56 < sensibel> jim: live usb has been prepared should I do next?
16:56 < sensibel> pingfloyd:   ^^
17:00 < naan> hi I'm trying to get palm rejection working on my surface pro 4 on ubuntu and I know that this rule could help but struggling to find my hardware id to make the right entry. The rule: https://github.com/systemd/systemd/blob/master/hwdb/70-touchpad.hwdb
17:01 < justsomeguy> I'm curious. Why does everyone run KVM guest machines on the qemu:///system domain, instead of as a user session qemu:///session ?
17:02 < ddoobb> It's been an hour. Is this normal for dd?
17:03 < justsomeguy> ddoobb: What are you dd'ing? Is it a very large file/device?
17:03 < ddoobb> 4gb windows iso on a 16gb flash drive
17:03 < RayTracer> ddoobb: if you forgot eg. status=progress with dd, you can kill -USR1 it and it will print current status
17:04 < ddoobb> RayTracera type "kill -USR1 it" in terminal?
17:04 < ddoobb> RayTracer
17:05 < RayTracer> ddoobb: more like "pkill -USR1 dd"
17:05 < justsomeguy> Where "it" is the process id of the running instance of dd.
17:05 < jim> ddoobb, 4gb is big... maybe, go out for a sandwich
17:06 < justsomeguy> Which you can find with ⍘ pgrep dd ⍘ , btw.
17:06 < ddoobb> I typed it but nothing?
17:06 < RayTracer> ddoobb: if something goes really wrong, like defective flash, it should also complain in "dmesg -T"
17:06 < RayTracer> ddoobb: the output is expected in the shell where you have the dd process running
17:07 < ddoobb> yeah nothing
17:07 < sensibel> I want to extend space for root partition , I have created live usb containing https://grml.org/download/, how can I extend root using this live usb?
17:08 < triceratux> sensibel: boot the flashdrive & run gparted ?
17:08 < RayTracer> ddoobb: if there is something really wrong, like defective flash, the dd process might be waiting for i/o to settle, and probably doesn't get to the status printout until the current i/o request has returned
17:09 < sensibel> triceratux: ok
17:09 < ddoobb> I terminated it. It has mad zero changes to the drive
17:09 < Dan39> ha
17:10 < ddoobb> What to do
17:10 < Dan39> what was the command you ran ddoobb ?
17:10 < sensibel> triceratux: the flash has grml64-full_2017.05.iso
17:10 < triceratux> thats a good thing
17:10 < sensibel> triceratux: ok let me boot
17:11 < ddoobb> dd if=/path/to/Windows10/Windows 10 iso.iso of=/dev/sdc bs=512k
17:11 < MrElendig> spaces matters
17:11 < MrElendig> also, that won't work anyway
17:11 < MrElendig> (won't boot)
17:12 < MrElendig> if that is a win10 install image
17:12 < ddoobb> MrElendig no spaces allowed in iso filename?
17:12 < MrElendig> sure there are, buy you have to quote them
17:13 < ddoobb> And wha's the other problem?
17:15 < ddoobb> It says dd: invalid status flag: ‘progress’
17:16 < jim> you're doin it again?
17:16 < fuzzy-logic> paste the command you are entering in here please ddoobb
17:16 < ddoobb> dd if="/media/KK/Imp. setup files/Windows10/Windows 10.iso" of=/dev/sdc bs=512k status=progress
17:17 < RayTracer> maybe your dd is too old for status=progress
17:17 < fuzzy-logic> and are you trying to create a bootable Win image ?
17:17 < Dan39> didn't MrElendig just say a windows 10 iso won't be bootable from a USB drive?
17:17 < ddoobb> Yes. I'm using ultimate boot cd so maybe that's why it's old?
17:17 < fuzzy-logic> It will - but you will need to install and run ms-sys.
17:18 < naan> oh crap I think I got it working alone
17:18 < naan> i am god
17:18 < jim> RayTracer, in my case for example, man dd doesn't show progress as a possibility for status=
17:19 < ddoobb> fuzzy-logic: I have ms-sys here, what do I need to do?
17:19 < sensibel> triceratux: boot failed please ensure a compatible media is found?
17:19 < jim> you have a weird space in the filename
17:19 < revel> jim: Mine does. The date at the bottom appears to be "July 2018" and from "GNU coreutils 8.30"
17:20 < Dan39> status=progress is relatively new option i think
17:20 < RayTracer> jim: mine has it http://dpaste.com/2HW2VFD
17:20 < jim> I'm running debian stretch
17:20 < RayTracer> and this is on fedora 24 which is pretty outdated
17:21 < triceratux> my dd manpage shows status=progress dd (coreutils) 8.28
17:21 < Dan39> ok, everyone in the channel now!
17:21 < Dan39> wow, mine with dd 8.29 does too!
17:22 < Dan39> and even 8.26 from February 2017!
17:22 < revel> My Pi on Raspbian Stretch shows it as well, 8.26, February 2017
17:22 < Dan39> Item posted by Pádraig Brady <pixelbeat> on Fri 03 Jul 2015 10:13:45 PM UTC.
17:22 < Dan39> coreutils-8.24 released
17:23 < kerframil> it was introduced in 8.24. everyone can stop checking now.
17:23 < Dan39>  dd accepts a new status=progress level to print data transfer statistics
17:23 < Dan39>   on stderr approximately every second.
17:23 < revel> kerframil: No, we must continue.
17:23 < kerframil> revel: oh, ok. one sec then.
17:23 < triceratux> everyone upgrade their ancient buntus & debians
17:23 < kerframil> revel: I have it in 8.29!
17:23 < revel> Dan39 already confirmed that.
17:24 < kerframil> but I had to be sure
17:24 < RayTracer> question is, why would jim's debian stretch don't have it
17:24 < revel> True.
17:24 < sensibel> jim:  the usb was formed correctly but now it is giving error please ensure compatible media, why this is so?
17:24 < Dan39> maybe he has some non coreutils version of dd?
17:24 < jim> mine (version 8.26), it does have it, but they must not have put it in the man page
17:25 < triceratux> pfft
17:25 < triceratux> https://packages.debian.org/stretch/coreutils
17:25 < lostfile> any one here know how to use jircii
17:26 < Dan39> knowing debian they probably made their own wrapper around dd :P
17:26 < jim> I must have screwed up in my search of the man page... I checked again, and it's there
17:26 < lostfile> im trying to hide joins and parts and ive tried every command
17:26 < Dan39> lostfile: switch to weechat
17:26 < kerframil> lostfile: http://wiki.xkcd.com/irc/hide_join_part_messages
17:26 < Dan39> xkcd wiki? lol what?
17:26 < lostfile> ive tried that to
17:27 < justsomeguy> It's pretty easy to configure in Konversation and weechat.
17:27 < kerframil> Dan39: do you have a better suggestion?
17:27 < lostfile> ignore * quit
17:27 < Dan39> no, that looks good, i just never knew that wiki existed, and find it kind of funny since isn't xkcd a web comic?
17:28 < kerframil> oh, I see. yes, it is somewhat amusing.
17:28 < jim> sensibel, did you get past that problem?
17:29 < sensibel> jim: yes it happened in past also?
17:29 < jim> it's booting?
17:30 < jim> I dunno, it's possible
17:30 < sensibel> jim: it started booting then it gave errror no please ensure compatible media
17:31 < jim> could you show the dd command you used?
17:31 < Dan39> the guy hasn't talked in here for like 10+ minutes
17:31 < Dan39> we just talking to ourselves at this point
17:32 < sensibel> jim: https://paste.linux.community/view/298b1ea9
17:33 < jim> found it
17:33 < jim> you need to write the usb again
17:33 < sensibel> jim: ok , how this time ?
17:33 < rico_> is .bashrc read for every subshell?
17:33 < jim> sensibel, dd if=grml64-full_2017.05.iso of=/dev/sdb bs=1M status=progress
17:34 < jim> make -sure- of=/dev/sdb NOT /dev/sdb1
17:34 < sensibel> jim: ok
17:34 < RayTracer> I'm regularly using oflag=direct to not pester the ram (and get plausible stats)
17:34 < jim> rico_, I think so
17:35 < purplex88> jim: on ubuntu server, i want to set unlimited lines.
17:35 < jim> on what, an irc client?
17:36 < sensibel> jim: here is progress https://paste.linux.community/view/d46b97cf
17:36 < purplex88> ?
17:36 < sensibel> jim: does it seems fine now ?
17:36 < purplex88> terminal
17:37 < jim> sensibel, yeah, looks good,,, try booting
17:37 < sensibel> jim: ok
17:37 < purplex88> like this: https://i.stack.imgur.com/GRZWD.png
17:37 < purplex88> i don't have a GUI
17:38 < jim> sensibel, does it boot?
17:39 < sensibel> jim: yes it works
17:40 < jim> purplex88, the pic looks graphical to me
17:40 < purplex88> oh man
17:40 < purplex88> i googled it obviously
17:40 < sensibel> jim: its asking to press a key
17:40 < rico_> jim: is it possible then to determine which bash shell is interepting a script, I have different versions of bash, and I want to ignore a variable in a certain case
17:40 < purplex88> i'm trying to do in ubuntu server if you know what it is lol
17:40 < jim> ok, see what happens when you do
17:41 < TR2990X> In Jenkins I use sh 'kill $(cat /opt/app/pid.file)' to kill the app before deploying the new version. But that command fails if the pid file doesn't exist. How do I ignore that error?
17:41 < jim> purplex88, what are you trying to configure?
17:41 < TR2990X> Right now my deployment stops because that command fails when deploying for the first time
17:42 < rico_> hmm I think probably but i'd have to give the full name of the binary
17:42 < RayTracer> TR2990X: check for existence before.. [ -f /opt/app/pid.file ] && kill $(cat ...)
17:43 < purplex88> the same thing via command line that is shown in the picture, sir
17:43 < RayTracer> purplex88: that setting is specific to the gui program "terminal"
17:44 < RayTracer> purplex88: so the question doesn't make sense to us
17:44 < sensibel> jim: here it is https://imgur.com/a/4cK68FC
17:44 < jim> purplex88, maybe I'm not looking in the right place... but I see the dialog box, but no program name
17:44 < sensibel> how to grml for adding space to root?
17:44 < purplex88> forget it
17:44 < TR2990X> RayTracer: I still get ERROR: script returned exit code 1
17:44 < Dan39> purplex88: in a console?
17:45 < Dan39> purplex88: or in some other terminal emulator?
17:45 < purplex88> ubuntu server's terminal
17:45 < triceratux> sensibel: woo woo welcome to grml !
17:45 < purplex88> it only has a terminal
17:45 < Dan39> purplex88: are you accessing it from a physical monitor and keyboard?
17:45 < Dan39> or over ssh...?
17:45 < RayTracer> purplex88: if you want to transfer settings of gnome-terminal, you can use "dconf dump /org/gnome/terminal/ > file" and "dconf load /org/gnome/terminal/ < file"
17:46 < purplex88> monitor and keyboard
17:46 < sensibel> triceratux: just waaaao?
17:46 < Dan39> purplex88: https://askubuntu.com/questions/790804/how-to-set-unlimited-scrolling-on-a-tty
17:46 < jim> sensibel, what country and language do you speak?
17:46 < triceratux> sensibel: you can run that grml-x thing & it will start x11 & gparted will be available. otherwise you can use the cli tools
17:46 < purplex88> i don't want to transfer
17:46 < purplex88> i want to only increase the number of lines of terminal
17:47 < purplex88> via command line
17:47 < Dan39> transfer? who said transfer?
17:47 < sensibel> jim: Pakistan and language is urdu
17:47 < RayTracer> Dan39: I did, in an attempt to interpolate the question
17:47 < sensibel> jim: any reason for this question?
17:47 < Dan39> oh i see
17:48 < jim> sensibel, ok, but you obviously understand english... reason is, to see how to configure the keyboard
17:48 < jim> sensibel, is english ok?
17:48 < sensibel> jim: yes english is ok
17:49 < Dan39> purplex88: the key point here is that you are talking about a tty console, not a GUI terminal emulator
17:49 < jim> ok, hit u then n
17:49 < sensibel> jim: ok
17:49 < purplex88> unbuntu server doesn't have GUI
17:49 < Dan39> ffs
17:50 < iodev> I found something worse than systemd
17:50 < Dan39> iodev: purplex88 ?
17:50 < iodev> it's called netplan, and I hate it!
17:50 < jim> sensibel, any time you don't understand something, take a picture and show us
17:50 < iodev> absolutely!
17:50 < sensibel> jim: network card config menu
17:50 < sensibel> let me share a picture
17:50 < salshabila> ehy guys, i'd like to install debian buster but i desperately need a program that was designed for debian stretch. is there any way to install it in buster? it's ROS
17:50 < jim> ok
17:51 < jim> salshabila. do you have stretch installed now?
17:52 < jim> salshabila, and yes, it should be possible to run the program you want
17:52 < salshabila> jim: i had ubuntu 18 but (as usual) it caused me only a great load of troubles so i'm switching to debian definetely
17:52 < jim> ok
17:52 < sensibel> jim:  here it looks like now https://imgur.com/a/g5I0GsC
17:53 < jim> sec
17:53 < triceratux> salshabila: i know how you feel. i finally settled on siduction
17:54 < jim> the highlighted choice is correct... (you have an ethernet card, right?)
17:54 < RayTracer> TR2990X: so the file exists but its pid does not? you could maybe set +e / set -e after that part you know it can fail, to be able to continue the script after non-zero exit code of a command
17:54 < sensibel> I have wired connection
17:55 < jim> sensibel, ok, I -guess- that's ethernet
17:55 < sensibel> jim: university internet so highlighted is correct?
17:55 < neozaga> any of you guys have experience with dsh?
17:55 < salshabila> triceratux: would u rather take siduction or q4os? i'd go with q4os, it never failed me
17:55 < jim> oh, yeah, should be correct sensibel
17:55 < salshabila> triceratux: but i never actually tried siduction
17:55 < sensibel> jim: so I should go with highlighted?
17:56 < triceratux> salshabila: definitely q4os. i even run it on my raspi3b. its no nonsense clean debian
17:56 < jim> salshabila, Please spell out u as you... it would help folks who are here and are new english speakers, some don't hear u as a rhyme for you
17:56 < salshabila> jim: ok!
17:57 < jim> thanks :)
17:57 < salshabila> triceratux: ok, thank YOU!
17:57 < salshabila> going to use q4os as usual
17:57 < sensibel> jim: its now asking for network device
17:58 < jim> ok, what are your choices?
17:58 < sensibel> I there a way I can check the info using current maching , what about ifconfig?
17:58 < sensibel> jim: let me share the pic
17:58 < jim> you might have ifconfig
17:59 < jim> ok
17:59 < salshabila> jim: the "u" thing it's actually pretty much funny. I'm italian, and the first time someone used "u" with me i immediately understood (even if at the time i had really big problems with english eheh). So i never got why some people do not understand it!
17:59 < kerframil> TR2990X: how is the shell code launched from jenkins exactly? is it executed as a script, inline with sh -c or something else?
17:59 < RayTracer> ifconfig is deprecated, please move on to ip
17:59 < jim> salshabila, it's because some don't hear it the same way
18:01 < sensibel> jim: here it is https://imgur.com/a/lroC2CJ
18:02 < jim> hmm according to that you have two ethernets'
18:02 < SpeakerToMeat> Do you guys know any linux rsync based (or similar) GUI backup app that has a) graphical exclude select from a dry run built list b) a static build? (or is statically buildable)
18:02  * triceratux wishes he were in a uni with that kind of internet access
18:02 < jim> SpeakerToMeat, you mean a backup program that does file-by-file copying?
18:03 < SpeakerToMeat> yes
18:03 < sensibel> jim: one is near, how could I see its the nearer one?
18:03 < SpeakerToMeat> I basically want something like an rsync gui that I can tell "you're gonna copy from this dir to this dir", it compiles a list of files to copy, I can deselect files, and then run it.
18:04 < SpeakerToMeat> ANd it uses normal rsync tactics (deltas, timestamp, etc)
18:04 < jim> sensibel, is one connected and the other not? (it occurs to me, why are we configuring the net....)
18:04 < jim> it could be handy
18:05 < sensibel> jim: I just want extra space for root
18:05 < jim> are both of them connected with a cable? two wired cables coming out?
18:05 < sensibel> jim: can i skip this?
18:05 < jim> ok, yeah
18:06 < sensibel> so i will cancel this
18:06 < jim> e for exit
18:06 < jim> yeah
18:07 < sensibel> again press any key prompt
18:07 < jim> try: x
18:07 < sensibel> small?
18:08 < jim> umm :) not sure that matters
18:08 < jim> I'm trying to start your x env
18:08 < kerframil> SpeakerToMeat: FreeFileSync
18:09 < triceratux> yeah that one post https://imgur.com/a/g5I0GsC shows hes connected & via eth1 so the config part is largely unnecessary
18:09 < sensibel> jim: its showing 2017.OS logo
18:09 < sensibel> jim: let me share the pic
18:09 < jim> sensibelif by "small?" you mean something different than shift X versus unshifted x, share a pic
18:09 < SpeakerToMeat> kerframil: Oh my
18:09 < sensibel> jim: ok
18:10 < system16> who has tried vmware ? is it better than virtualbox
18:10 < SpeakerToMeat> kerframil: Thanks, this looks very promissing, not onyl for this but for replacing an ailing CCC on a mac at home as well, thanks
18:10 < SpeakerToMeat> system16: It is different, and similar to virtualbox...
18:10 < kerframil> SpeakerToMeat: I haven't actually tried a Linux build yet but I've tried it on Windows and it is excellent
18:10 < SpeakerToMeat> system16: You'd have to start by mentioning which vmware product, and for what use
18:10  * triceratux is a huge fan of grml 2017.05
18:11 < system16> workstation pro 14
18:11 < sensibel> jim: here it is https://imgur.com/a/69zcr05
18:11 < sensibel> triceratux: interesting
18:11 < system16> i wanna ditch vbox because i have performance issues
18:12 < triceratux> sensibel: thats it. yer up. thats fluxbox on top of a minimalist debian
18:13 < sensibel> triceratux: whats should I do next?
18:13 < jim> the pic must have got damaged... it's all black
18:14 < sensibel> jim: again check it was uploading at that moment https://imgur.com/a/69zcr05
18:14 < lessthan0> I need to unmount my boot drive to shrink my ext4 over LVM. would it be easier to use a debian live cd, use ramdisk, or install the physical hard drive in my other working linux box?
18:14 < triceratux> sensibel: dig around the menus looking for gparted or bring up a terminal & do the cli resizing thing
18:15 < lessthan0> I am not sure if the live cd/usb has all the packages for LVM resize and file system resize
18:15 < lessthan0> I have the tutorials
18:15 < lessthan0> read through it multiple times
18:16 < lessthan0> they say not to use the gui
18:16 < lessthan0> but before I can do anything through the CLI I need to decide what I will boot from and how I will unmount mount the physical disk
18:16 < jim> is pingfloyd here still?
18:17 < jim> sensibel, for some reason I'm not getting the image
18:17 < lessthan0> the ramdisk option could go wrong therefor I am probably going to move a hard drive into another machine or boot from usb recover console debian 9.4
18:18 < triceratux> http://i.imgur.com/EMe9PVa.jpg ?
18:19 < triceratux> sensibel: i dont think theres gparted on grml. youll want to do the cli resize stuff from a terminal
18:19 < system16> its so f ing strange that thepiratebay is in google
18:19 < sensibel> triceratux: theres just resizing screen and other things
18:19 < jim> triceratux, pingfloyd says there is
18:19 < lessthan0> *solved* they told me in #debian that the live cd has all the packages for resize LVM
18:20 < jim> lessthan0, so grml is fine?
18:20 < triceratux> jim: ah *thats* how grml got into this rofl
18:20 < jim> yep
18:21 < jim> ubuntu blocked at every turn pretty much
18:21 < sensibel> jim: I am waiting that parties agree on a decision
18:21 < sensibel> jim: what do you say about it?
18:21 < SpeakerToMeat> system16: I doubt you'll see any performance difference between vbox and vmware desktop
18:21 < jim> I think you should try running gparted
18:22 < SpeakerToMeat> system16: btw be aware vmware desktop is a paid product.... there are instances in which you can work with vmware player, but not always
18:22 < lessthan0> I don't need to download grml if I can use the debian usb I have already
18:22 < SpeakerToMeat> kerframil: Marvelous tool, thank you
18:22 < system16> SpeakerToMeat, i will pay for it if it would run better
18:22 < sensibel> jim: but how to get rid of this black screen and run gparted?
18:22 < lessthan0> it should work fine
18:23 < SpeakerToMeat> system16: Desktop virt is normally heavy, you're running a whole desktop enviroment inside another one
18:23 < jim> sensibel, what happens when you press any key
18:23 < jim> ?
18:24 < system16> SpeakerToMeat, im pretty sure mylappy can handle it
18:24 < sensibel> jim:  it is there still
18:24 < SpeakerToMeat> system16: It can but you're sharing resources between two whole systems, it's like having half a pc
18:24 < sensibel> jim:  which key should I press and see if it moves away or not?
18:24 < SpeakerToMeat> system16: What host and guest are you running?
18:25 < system16> windows 10 is my host
18:25 < system16> ubuntu is my guest
18:25 < jim> unfortunately I can't see your screen, I don't kjnow how to work it
18:25 < system16> i can even see screen tear when i drag things
18:25 < sensibel> jim: have you seen the picture ?
18:25 < system16> in my guest
18:25 < jim> nope
18:25 < jim> I can try again
18:26 < SpeakerToMeat> system16: Yeah full virt is probably your best option.
18:26 < system16> you mean vmware ?
18:27 < SpeakerToMeat> I mean full virtualization rather than containers or at least paravirt.
18:27 < sensibel> jim:  I saw a menu for xterm on it and pressed it , now xterm is running , can i do something grml-xterm?
18:27 < jim> wait I think I do see it
18:27 < sensibel> jim: let me share the pic
18:28 < jim> does it say 2017.05?
18:28 < system16> SpeakerToMeat, what is "full" virtulization ?
18:28 < SpeakerToMeat> system16: Full virtualization is the virtualization of computer hardware. it's what vbox and vmware desktop do.
18:29 < jim> sensibel, go in #debian
18:29 < system16> oh. i have virtualbox. as i said it just preforms slow and luggish
18:29  * triceratux has booted grml 2017.05
18:29 < sensibel> I should type #debian?
18:29 < triceratux> yep looks like its got gparted
18:30 < SpeakerToMeat> system16: and vmware will do the same
18:30 < SpeakerToMeat> most probably
18:30 < system16> so they both run OSs like crap ?
18:30 < jim> sensibel, /join #debian
18:31 < sensibel> jim: see if it helps https://imgur.com/a/2IFF9GD
18:31 < SpeakerToMeat> system16: They both add some overhead to the process.
18:32 < sensibel> triceratux: how start gparted?
18:32 < system16> plz speak english :)
18:32 < SpeakerToMeat> system16: you're running two systems on one machine
18:32 < day> am i violating any grep rules here? the filter is according to regex101.com fine. grep "href=\"[^\"]+" index.html
18:32 < jim> type: gparted at the prompt
18:33 < triceratux> sensibel: get to a terminal or the fluxbox run prompt & type gparted
18:33 < ashleyk_> system16, you need to enable VT extensions in your system bios
18:33 < jim> looks like he's at a terminal
18:34 < sensibel> triceratux: I have opened gparted
18:34 < sensibel> now?
18:34 < SpeakerToMeat> system16: You're running two whole OS on a single machine, two kernels, two gui environments, and all the programs. if you have two cpu cores each system runs like it has only one core worth of performance, if you have 16Gb each system will run liek it has 8Gb worth of memory, etc (not really but it's an aproximation of how things are). appart from this, you have a program that makes this possible... by the fact that this program is usi
18:34 < jim> good, share a pic
18:34 < SpeakerToMeat> system16: to achieve this, this program itself is consuming resources
18:34 < system16> ashleyk_, its enabled.
18:34 < sensibel> jim: w8
18:34 < triceratux> sensibel: now nothing can go wrong. do what jim says ;)
18:34 < sensibel> triceratux: ok
18:35 < jnewt> at what point is it not worth trying to fix a linux installation and just reinstall the whole thing?  i've got several issues after upgrading from ubuntu 16.04 lts which was working perfectly to 18.04 lts  now my printer, sound, desktop environment, rdp server are all not working.  should i just start over or will i end up in the same place if i reinstall 18.04?
18:35 < SpeakerToMeat> system16: But nowadays with VT extensions, the ammount of resources the virtualization consumes is reduced as well, and not hat much
18:35 < mroutis> what's a good resource to learn about disk partitions + LVM + RAIDs, etc. ?
18:35 < ashleyk_> system16, and how did youconfigure the machines?  make sure they have enough ram and CPUs or they will be slow
18:35 < ashleyk_> especially ram
18:35 < SpeakerToMeat> specially ram
18:36 < mroutis> I've read the man pages for mkfs.ext4, gdisk, but I'm having a hard time to figure out the basics
18:36 < ashleyk_> mroutis, do you really want to do software raid?
18:36 < system16> ashleyk_, i just enabled this in bios : enable intel virtualization technology
18:36 < Psi-Jack> Heh.. Finally. My Ceph rebalancing... Is finally almost done. LOL
18:36 < ashleyk_> system16, thats good, sometimes theres a seperate one for VT-d
18:37 < ashleyk_> w/e that is
18:37 < jim> triceratux, fairly dangerous operation. he has an lvm setup with 3.6 tb in a logical volume, and we want to reduce its size by 2 t
18:37 < zapotah> mroutis: you want to learn the difference between block devices filesystems, volumes, etc
18:37 < Psi-Jack> w/e != word.
18:37 < zapotah> Psi-Jack: how much data?
18:37 < mroutis> I don't, ashleyk_ , currently doing a fresh install and my curiosity bit me
18:37 < sensibel> jim: https://imgur.com/a/qo7nxFH
18:37 < sensibel> have a look ^^
18:37 < zapotah> Psi-Jack: you added osds or...?
18:38 < mroutis> yep, zapotah, that would work for my purposes
18:38 < Psi-Jack> zapotah: 9.69 TiB, I replaced 3 1TiB HDDs with 1 4TiB.
18:38 < zapotah> ah
18:38 < ashleyk_> mroutis, https://www.thomas-krenn.com/en/wiki/Linux_Software_RAID
18:38 < mroutis> basicaly, I don't want to follow mindlessly installation guides
18:38 < mroutis> thanks ashleyk_
18:39 < ashleyk_> i guess, or anything with basic information...  then lvm goes on top of that
18:39 < SpeakerToMeat> ashleyk_: There are many good reasons for doing software raid
18:39 < RayTracer> mroutis: there is also https://raid.wiki.kernel.org
18:39 < jim> Psi-Jack let me do that part
18:39 < ashleyk_> mroutis, https://siva2009.wordpress.com/2010/08/26/how-to-create-lvm-using-pvcreate-vgcreate-lvcreate-and-lvextend-commands/
18:39 < ashleyk_> mroutis, i just startpage'd it
18:40 < SpeakerToMeat> ashleyk_: And they can be combined anyhow, I once dealt with a machine that used md over lvm over md
18:40 < SpeakerToMeat> or somethign horrible like that
18:40 < mroutis> ashleyk_, the last one seems pretty good (content, obviously, design is a little bit awful)
18:40 < ashleyk_> SpeakerToMeat, i know
18:40 < ashleyk_> mroutis, well once you know the commands you can read man pages
18:40 < sensibel> jim: its dangerous ?
18:40 < mroutis> thanks!
18:41 < jim> ashleyk_, hi... I request you not use abbreviated words (example w/e), because I'm not sure much newer english speakers can understand
18:41 < ashleyk_> jim, w/e
18:41 < ashleyk_> good luck with that
18:42 < lessthan0> I think I figured out one of my problems. the WD 1TB uses https://en.wikipedia.org/wiki/Advanced_Format
18:42 < lessthan0> that is why the debian installer guided partitioning fails to align all the partitions to 4096 byte sectors
18:43 < triceratux> jim: is gparted up to that provided he can find the right device ?
18:43 <@jim> sensibel, now, not sure gparted is the right tool?
18:43 < sensibel> jim: ah
18:43 <@jim> triceratux, I see the partition that's the pv
18:43 < lessthan0> my problems are like an onion with many layers
18:44 < lessthan0> there is no way to backup the 1TB unless I pipe dd to gz maybe
18:45 <@jim> sensibel, what happens if you double click on /dev/sda3
18:45 < lessthan0> I can probably get it down to 10GB but I don't see how I would be able to mount the .gz to edit things inside without decompressing the entire thing
18:45 < sensibel> jim: let me see
19:11 < ExtremeFMan> help! I runned a command called exifprobe and now my terminal is invisible
19:11 < epitamizor> what brand laptops are most compatible for linux
19:11 < c0mrade> Can anyone help me with a db query issue that I am having. I can pay in Bitcoin something like ~20$.
19:12 < ExtremeFMan> ok, after resetting the terminal I can see it again...
19:12 < ExtremeFMan> thanks for the help!
19:14 < pnbeast> select INSTABILITY,NOT_ACCEPTED from PROBLEMS_WITH_BITCOIN where SITUATION = "in irc";
19:14 < akk> epitamizor: Lenovo is fairly popular, but you should google on a specific model before choosing one.
19:14 < akk> epitamizor: There's no one brand where you can say that all models will work well with Linux.
19:15 < akk> epitamizor: Acer is probably the worst.
19:15 < pnbeast> akk, I'm pretty sure there are two companies that build nothing *but* Linux laptops.
19:15 < lessthan0> lenovo 2 years old
19:16 < akk> Oh, true: if you buy from a company like ZaReason or Company 7, they're actually the best.
19:16 < lessthan0> always 2 years old laptop when using linux
19:16 < akk> If they happen to make a laptop of the specs you're looking for.
19:16 < akk> And Dell and Asus sometimes make models that come with Linux, too. (My last two laptops.)
19:17 < lessthan0> check connonical for official laptops
19:17 < lessthan0> that was a thing 2 years ago
19:17 < akk> It'll cost more to buy a laptop that comes with linux, though, sadly; you have to pay for the privilege of not having a non-free OS.
19:17 < lessthan0> not sure if it is still a thing
19:18 < akk> (Usually; my current Asus was a happy exception, a model that shipped with Ubuntu and was still very inexpensive. But that was years ago.)
19:19 < lessthan0> here https://certification.ubuntu.com/desktop/
19:19 < lessthan0> ubuntu is really just the second beast so
19:20 < c0mrade> No one?
19:20 < ayecee> apparently not
19:21 < lessthan0> maybe no one knows the answer in here
19:21 < lessthan0> you didn't even mention what kind of DB
19:21 < ayecee> it's happened before
19:21 < lessthan0> SQL?
19:22 < pnbeast> He didn't even ask *the question*.
19:22 < lessthan0> maybe he wants an NDA before he lets you in root to the DB
19:23 < rrman> When I try to start firewalld on my Centos, I get:
19:23 < rrman> Failed to start firewalld.service: Unit is masked.
19:24 < rrman> What should I do?
19:27 < pnbeast> rrman, what happened when you unmasked the "unit file"?
19:33 < rrman_> pnbeast, I started firewalld after unmasking and I got disconnected. The state is "degraded"...?
19:34 < rrman_> I didn't know there was an unmask command by the way :))
19:36 < zaltekk> anyone around able and willing to help work through an issue with mutt + PGP signature validation?
19:36 < zaltekk> tl;dr is is mutt verifies the message recorded in my sent folder, but not the one in my inbox.
19:36 < zaltekk> (while sending an email to myself)
19:41 < system16> how can i access that tty thing ?
19:42 < sensibel> jim: hi
19:42 < rrman_> system16, CTRL-ALT-F1
19:42 < system16> thanks
19:42 < rrman_> np
19:42 <@jim> hi
19:43 < system16> rrman_, its the same in all distros ?
19:43 < sensibel> jim: what's the next solution?
19:44 <@jim> to copy to a new ext4 partition which we will put on a separate pv and vg
19:45 <@jim> then we can move the pv
19:45 < sensibel> jim: ok
19:45 <@jim> all this time I thought we were stuck without extra space
19:46 < sensibel> jim: I previously said that I have two 4T drives
19:46 <@jim> I bet you did... for some reason I didn't hear that when you said it
19:46 < sensibel> jim: but it does not appears in df -Th so I don't where it is
19:48 <@jim> let's do this... reboot your machine back into the centos
19:48 < rrman> pnbeast, it's fixed, thanks for the tip.
19:48 <@jim> we might not need pics for awhile
19:48 <@jim> if we do that
19:49 < sensibel> jim:now I should just run shutdown now ?
19:49 <@jim> shutdown -r now
19:49 < sensibel> how to boot into centos now ?
19:49 <@jim> or just reboot
19:50 <@jim> we didn't take it apart that I know of, so it should be ready to boot
19:50 < rrman> system16, yes, alternatively you can use `chvt 1` (as root).
19:51 < sensibel> jim: now single user mode or normal boot ?
19:51 <@jim> normal boot
19:51 < sensibel> ok so I will be able to login from there
19:52 <@jim> yeah, just like before
19:52 < sensibel> thats nice its more that 10 hours sitting in the lab :)
19:54 < sensibel> jim: let me join from there now
19:56 < fiter> jim: hi
19:56 < dannylee> lkl
19:56 <@jim> hi
19:56 < lnnb> hay
19:56 < dannylee> hi jim
19:56 <@jim> hey dannylee, lnnb
19:58 <@jim> fiter, ok, could you do... gdisk -l | nc termbin.com 9999
19:58 < fiter> jim: https://paste.linux.community/view/1b563ed6
19:59 < fiter> jim: [root@localhost dellemc]# gdisk -l | nc termbin.com 9999 Problem opening -l for reading! Error is 2. The specified file does not exist! http://termbin.com/6vd1
20:00 < fiter> jim: and the link shows GPT fdisk (gdisk) version 0.8.6
20:01 <@jim> I couldn't see your other drive
20:01 <@jim> you say it's an empty 4tb drive?
20:01 < fiter> jim: here it shows two 3.7T https://paste.linux.community/view/1b563ed6
20:02 < fiter> i told that I dont know where the disk is or how to access it.
20:03 <@jim> I can't see it there.... how many drives do you have connected?
20:04 < fiter> jim: there are two devices connected , what if the other is on RAID1 ?
20:04 <@jim> maybe that's why
20:05 <@jim> listen, it's late and I have to rest... can we pick this up tomorrow?
20:06 < fiter> jim:  yeah , why not
20:07 < fiter> jim: tomorrow at which time ?
20:07 < fiter> jim: before going see the output of this command https://paste.linux.community/view/1fee5840
20:07 <@jim> fiter, not sure right now... one thing you can do that would be very helpful, can you get a third drive?
20:08 < fiter> How much would it cost ?
20:08 < system16> rrman, and how can i quit that tty ?
20:09 <@jim> dunno, as far as size, maybe 2g is enough, or maybe 3
20:09 <@jim> it looks like we can't even see the second drive
20:10 < rrman> system16, ctrl alt f7 or 8... is usually the graphical, sometimes it could be other ones.
20:10 <@jim> anyway,,, I'm out
20:10 < triceratux> jim: np. nobodys keeping you here. turn yer ops off before you crash
20:10 < fiter> jim: ok
20:10 < triceratux> preciated ;)
20:10 < jim> thanks
20:13 < system16> thanks
20:13 < fiter> triceratux: I am also leaving before I also crash
20:23 < jnewt> can someone help me figure out how to switch from mate to cinnamon desktop?   i boot and end up at the mate login.  there's no selection for choosing desktop, it's like it's already chosen.
20:25 < system16> how can i forcefully umount a partition>
20:26 < mawk> by using force
20:26 < mawk> but force is for network filesystems
20:26 < mawk> for other filesystems there is no such thing
20:26 < Elladan> You can use lazy unmount
20:26 < mawk> but you can do a lazy unmount
20:26 < mawk> too late :(
20:26 < Elladan> However keep in mind that it's still mounted.
20:27 < system16> im trying to umount /dev/sda1
20:27 < Elladan> This is relevant if you want to unplug a device or something.
20:27 < Elladan> system16, use lsof or similar to find out what processes are using the mount
20:28 < system16> the OS
20:28 < Elladan> system16, I assume it's mounted as /boot, so you would do something like "sudo lsof /boot"
20:28 < system16>  /dev/sda1 is my system partition
20:29 < Elladan> Do you mean root?
20:29 < lostfile> having problems with mounting stuff
20:30 < system16>  /dev/sda is the boot partition
20:30 < system16> sda1*
20:30 < kerframil> partitions are not filesystems. where is the filesystem currently mounted?
20:31 < system16>  /dev/sda1
20:31 < kerframil> that's not a mount point
20:31 < lostfile> yes it is unless your using a usb or some other drive
20:31 < lostfile> i think
20:31 < system16> idk
20:31 < system16> disks says this
20:31 < system16> mounted at filesystem root
20:31 < kerframil> system16: findmnt -R /
20:32 < zenix_2k2> ok so i have a nooby question, are stdout, stdin and stderror 3 "files" ?
20:32 < kerframil> lostfile: block device nodes are not mount points
20:32 < system16> it did a bunch of things
20:32 < system16> it looks like the tree command in windows
20:32 < kerframil> system16: findmnt /dev/sda1
20:32 < lostfile> kerframil: oh sorry i derped
20:33 < kerframil> lostfile: no worries
20:33 < Elladan> zenix_2k2, they're file descriptors 0, 1, 2, which can be attached to files or pipes depending on how the process was started.
20:33 < system16> FSTYPE options ?
20:33 < kerframil> system16: the TARGET is what is of interest.
20:33 < system16> target is /
20:34 < Elladan> zenix_2k2, for example, "ls > file" would start ls with stdout directed to a file, while "ls | cat" would start it directed to a pipe.
20:34 < kerframil> system16: you can't umount /
20:34 < system16> TARGET SOURCE
20:34 < system16>  /       /dev/sda1
20:34 < system16> it looks something ;ike that
20:34 < system16> like*
20:34 < seven-eleven> hi, if I add "auth    [success=1 default=ignore]  pam_winbind.so debug" to /etc/pam.d/common-auth, should I see in syslog the domain login attempt via winbind? i don't see it, thats why i worry pam isn't even trying
20:35 < kerframil> you can't umount the root (/) filesystem, because it's in use - by definition
20:35 < system16> im trying to destroy this linux mint ( i have a snapshot)
20:35 < zenix_2k2> Elladan: and what is so different between "file descriptors" and normal files ?
20:35 < kerframil> system16: boot from something other than linux mint and perform your destruction from there
20:35 < Elladan> zenix_2k2, a file descriptor is a programming concept.
20:36 < system16> i cant do it from teerminal ?
20:36 < Elladan> zenix_2k2, when you open a file, the open() call returns a file descriptor.
20:36 < system16> terminal*
20:37 < Elladan> zenix_2k2, in unix, there is a concept where files, pipes, and sockets are (somewhat) interchangeable, and each is accessed via a file descriptor.
20:37 < kerframil> system16: you can't unmount the / filesystem because it's the root filesystem of the operating system you are using
20:37 < kerframil> system16: and you're currently using it
20:37 < system16> can i do something else ?
20:38 < kerframil> system16: yes. boot another linux environment and do whatever it is you intend to do.
20:38 < saderror256> am i logged in?
20:38 < system16> windows has system32. what about linux ?
20:38 < saderror256> im setting up an irc client (like always lol)
20:38 < Elladan> If your only goal is to wipe the system, you can dd over your hard disk while the partitions are mounted.
20:38 < system16> dd ?
20:39 < Elladan> However, if you don't know what that means, don't do it. Boot from some other environment.
20:39 < kerframil> what is the actual goal here? why do you want to destroy the operating system that you are currently in?
20:39 < system16> yes. its in a vm. i have a snapshot.
20:39 < system16> its doing something
20:40 < Elladan> If it's a VM, use the VM tools to restore the snapshot or whatever.
20:40 < kerframil> system16: ok. so what, you're testing how it responds to destructive activity?
20:40 < nomenon> ls -l /dev/serial/by-id tells me my serial device is on USB0 but using putty to connect to /dev/ttyUSB0 fails to come up with a serial connection    any suggestions?
20:40 < wizzi> hello, when i test ports with nmap all ports look open what does that mean and should i close them
20:40 < wizzi> ?
20:40 < system16> Elladan, i will. i just wanna have funb
20:40 < system16> fun*
20:40 < mroutis> wizzi, what command are you using to query the ports?
20:41 < saderror256> hello, am i logged in?
20:41 < Elladan> system16, well then have at it. man dd ;-)
20:41 < wizzi> nmap "ip add"
20:41 < mroutis> saderror256: `whoami`
20:41 < saderror256> mroutis, okay, thanks
20:41 < kerframil> system16: if you want to break it, a very simple way is to run "wipefs -a /dev/sda1" as the root user. it will fail upon rebooting. I can't really say much more (channel rules).
20:41 < saderror256> im reffering to irc lol
20:41 < system16> oh it converts files
20:41 < saderror256> not linux
20:41 < system16> ok thanks
20:42 < saderror256> giving hexchat a shot, im on a raspberry pi 3
20:42 < system16> does it wipe filesystem ?
20:42 < kerframil> system16: in any case, you *cannot* unmount a filesystem that is in use. you can mess up the data at block device level, sure.
20:42 < kerframil> system16: not exactly. it identifies the location of the filesystem "signature" and wipes just that.
20:42 < kerframil> system16: for all intents and purposes, it's as if there were never a filesystem there at all
20:43  * nomenon sighs
20:43 < system16> probing failed. nice i think i screwed it up :)
20:44 < mroutis> wizzi: here is a list of common ports https://www.digitalocean.com/community/tutorials/how-to-use-nmap-to-scan-for-open-ports-on-your-vps
20:44 < mroutis> you decide if you want to close them or not
20:44 < system16> nope. still boots :(
20:45 < system16> wait a sec. its stuck
20:45 < wizzi> mroutis, and can these common ports be open ?
20:46 < kerframil> system16: it is not possible to mount /dev/sda1 at this point, because there is no identifiable filesystem header/signature
20:46 < system16> it did
20:46 < kerframil> system16: it appears as an empty block device at this point, even though the data blocks are technically still there
20:46 < kerframil> system16: you didn't do it right then
20:47 < kerframil> system16: anyway, the point is, if you have root, you can spray any block device with whatever data, anywhere. you don't actually have to unmount the filesystem to do that.
20:48 < TheWild> hello
20:48 < mroutis> yeah, wizzi, bro, it doesn't matter that much if you have a firewall outside your vpc or something like that; for example, if you specify in an EC2 instance a security group (firewall) that just allows TCP/22 to ip 123.456.789 it will only allow that ip to connect to SSH, no more
20:49 < mroutis> even if you have all ports open on your machine
20:49 < system16> oh
20:49 < system16> i found a command in youtube
20:49 < kerframil> mroutis: you don't have "all ports open". ports are only "open" if they are in a listening state or otherwise involved in an active connection.
20:49 < system16> it removed my desktop icons
20:49 < kerframil> mroutis: to be in a listening state means that you have some application that is running that has bound to the port in question, and on the interface that is relevant
20:49 < mroutis> that's true, kerframil, thanks for pointing it out
20:51 < system16> yup. i screwed things up
20:51 < system16> its entering grub rescue
20:52 < wizzi> mroutis, so when the firewall is running no worries about ports ?
20:53 < system16> restoring snapshot
20:54 < mroutis> yep, wizzi, however, what kerframil said is important; if you have an application bound to a port, they will not have access from the outside world
20:55 < kerframil> if you don't have, rather
20:55 < system16> here u go. every thing is back to normal
20:55 < wizzi> mroutis, i got it thank you !
20:56 < system16> i wish REAL computers had this snapshot thing
20:57 < kerframil> system16: they can ... sort of. but nothing directly protects against metadata loss.
20:58 < sauvin> system16, "you", not "u".
20:58 < system16> oh sorry
20:58 < kerframil> system16: for instance, you could use btrfs for / and take snapshots. you can boot into any snapshot.
20:58 < mroutis> what about nixos, kerframil ? or what are refering to as metadata?
20:58 < TheWild> just typed "wireshark scripting" and found "Even if the code you write in Lua does not need to be GPL'ed. The code written in Lua that uses bindings to Wireshark must be distributed under the GPL terms. see the GPL FAQ for more info" https://wiki.wireshark.org/Lua
20:58 < TheWild> seriously?
20:58 < kerframil> mroutis: in this case, I'm referring to partition tables, identifiable filesystem superblocks/signatures, LVM headers, dm-crypt headers etc ...
20:59 < sauvin> TheWild, yes, seriously. This is how GPL works.
20:59 < system16> btrfs ?
21:00 < system16> https://en.wikipedia.org/wiki/Btrfs
21:00 < TheWild> holy crap, GPL is a mess then
21:00 < sauvin> It has its moments. :\
21:01 < system16> oh i think clonezilla does the same thing
21:01 < TheWild> I will release my scripts on MIT or WTFPL. Is someone going to takedown this or whatever?
21:02 < kerframil> system16: clonezilla backs up the raw sectors of a block device. it's not the same thing.
21:02 < mustu> any nxlog user here?
21:04 < kerframil> TheWild: given a license violation, someone so inclined could do exactly that
21:04 < sauvin> TheWild, the probability of being taken to court for not honouring the terms of the GPL are low, but not nonexistent.
21:05 < TheWild> da hell
21:05 < TheWild> I haven't even read the Wireshark code, maybe just the documentation.
21:08 < lnnb> TheWild: wireshark API uses GPL'd libs and not LGPL ?
21:10 < Sitri> You're distributing scripts in LUA... it's not like the GPL causes you any grief here.
21:10 < lnnb> how would LUA interpreter copyright make a difference here?
21:11 < Sitri> It doesn't
21:11 < kerframil> doesn't matter
21:11 < Sitri> The fact that what is needed to run is also the raw source...
21:12 < zenix_2k2> also one question, can 2 file descriptors be the same in Linux ?
21:12 < zenix_2k2> cause i am still confused between FDs with Inodes
21:12 < Sitri> zenix_2k2: what do you mean?
21:13 < Sitri> The same in what way?
21:13 < zenix_2k2> their values ?
21:13 < zenix_2k2> cause "i think" that file descriptors can only be different in the same process, correct me if i am wrong
21:14 < zenix_2k2> like in one program with 2 open()
21:14 < Sitri> Every process starts with three file descriptors: 0, 1 and 2.  Those are stdin, stdout and stderr respectively.
21:14 < Sitri> Does that answer your question?
21:14 < TheWild> but if the script for Wireshark is my code from very beginning to the very end and I didn't look into anyone's elses code and I based only on what's in the documentation, then it's purely my code and IMHO GPL has nothing to it.
21:15 < rrman> Where is the gpu device in /dev?
21:15 < TheWild> what if someone does the clean-room implementation of Wireshark?
21:15 < TheWild> I mean Wireshark's API
21:15 < zenix_2k2> Sitri: but i thought that the open() call which was the thing that will return 3 file descriptors
21:15 < kerframil> TheWild: and release the results under a more permissive license? you could that if you really wanted to.
21:15 < Sitri> zenix_2k2: open() returns either 1 file descriptor or an error.
21:16 < kerframil> lnnb: as far as the GPL is concerned, whatever makes use of the APIs is effectively linking. such derivative works must also be GPL.
21:16 < Sitri> TheWild: look at libpcap
21:16 < lnnb> thats why i asked if wireshark libs use GPL and not LGPL
21:17 < lnnb> if they intentionally separated the API it's arguable
21:17 < lnnb> from main program copyright
21:18 < TheWild> imagine the BIOS was GPL'd
21:18 < kerframil> lnnb: wireshark is GPL2
21:18 < Sitri> zenix_2k2: If FDs still confuse you, they aren't hugely important unless you're coding sockets in C or C++.
21:19 < Sitri> So you could just safely ignore them.
21:19 < zenix_2k2> Sitri: actually i am trying to understand them because i am learning C
21:19 < zenix_2k2> so what between inode and FDs, which one the kernel uses to read/write/append to the opened files ?
21:23 < zenix_2k2> it is just confusing when each process have 3 PDs and an opened file have 1 PD and those PDs are only limited to 0,1,2,3... so how can the kernel know which is which ?
21:23 < Loshki> zenix_2k2: a file descriptor is a small integer you use in a program. An inode is a relatively low-level kernel data structure. Both are involved in file operations, but under normal circumstances (i.e. non-kernel programming) you are unlikely to even see an inode...
21:24 < Elladan> zenix_2k2, each process has a set of open files/pipes/sockets/etc. associated with it, which are referred to via system calls using a file descriptor.
21:24 < Elladan> zenix_2k2, the file descriptors are integers whole meaning is specific to that process.
21:24 < rukusza>       ln Loshki Loshki-not-an-inode
21:25 < zenix_2k2> Loshki: so my entire system can only have 1 inode ?
21:25 < Elladan> er, "whose meaning"
21:26 < zenix_2k2> well i am just thinking, not from anyone
21:26 < Sitri> zenix_2k2: inodes are how a filesystem finds a file's data.  FDs are how a process interacts with files, each file or socket it interacts with ultimately has an FD associated with it.
21:26 < Elladan> zenix_2k2, an inode is a concept specific to file systems. You don't directly interact with file systems when programming (except inside the kernel). Instead, you interact via system calls like "open" "read" "stat" etc.
21:27 < Sitri> The FILE* handles that libc provide are actually wrappers around FDs.
21:28 < Sitri> zenix_2k2: Perhaps it'd be better to explain what you're trying to do and this can all be explained to you in a more detailed context.
21:28 < Sitri> Ultimately though, it is very unlikely that you need to do anything with the inodes directly.
21:28 < zenix_2k2> well i am just trying to understand what happens behind the scene when i open a file
21:28 < Sitri> With which function?
21:29 < zenix_2k2> open()
21:29 < zenix_2k2> in C
21:29 < Elladan> The full description would be quite long.
21:29 < zenix_2k2> well do you have any link ?
21:29 < Elladan> ... so you really need to give us some hints about what kind of things you're interested in.
21:29 < Sitri> zenix_2k2: man 2 open
21:30 < zenix_2k2> Elladan: like how can my process find my file and then open it ? cause i know that it won't seek for its name of cause but there has to be something, at least in number
21:30 < Sitri> zenix_2k2: open() handles that for oyu
21:30 < Sitri> open() is a system call, which means everything it does is something the kernel does.
21:31 < Loshki> zenix_2k2: well, that fd is used to look up housekeeping information in a table which contains all the info your process needs to read/write a file. Things like: how many bytes you've read, stuff like that. The kernel keeps it's own version of this information, but the info it needs (kept in an inode) is different e.g. how many blocks are in the file, and where on the disk are they kept. see e.g. https://stackoverflow.com/questions/25819226/what-is-th
21:31 < Loshki> e-difference-between-inode-number-and-file-descriptor
21:31 < Sitri> You don't need to worry too much about the specifics, just that it acts in a well-defined way.
21:31 < zenix_2k2> actually i am not learning all of these for any of the real life practices, i am just curious
21:31 < zenix_2k2> like how things behind commands and buttons work :P
21:32 < Sitri> How much have you coded so far?
21:32 < Loshki> For the moment, take inodes on trust and concentrate on how to make open/read/write/close work
21:32 < zenix_2k2> Sitri: which language ?
21:32 < Sitri> Any language
21:32 < zenix_2k2> well... PHP, JS, MySQL, Python and C
21:32 < zenix_2k2> but i mostly focus on Python and C
21:32 < zenix_2k2> especially Python
21:32 < Sitri> I asked how much, not which
21:33 < zenix_2k2> oh quite a lot ?
21:33 < zenix_2k2> i can't define what do you mean by "how much"
21:33 < Sitri> So... 2 years?
21:33 < zenix_2k2> like how much lines of codes ?
21:33 < zenix_2k2> Oh, well 1 and a half year ?
21:33 < Sitri> I'm trying to gauge your experience here
21:33 < zenix_2k2> not even close to a half
21:35 < zenix_2k2> Loshki: so the inode keeps the data inside my files and the fds will tell my kernel where to look for the file ?
21:35 < zenix_2k2> oh wait you pasted a link
21:35 < zenix_2k2> let's me see
21:38 < Loshki> zenix_2k2: more like the inode has a list of disk blocks which contain the data from your file, almost certainly sitting in a disk cache somewhere. Someone has to keep track of all those buffers. The kernel handles all that, so it's almost machine-independent, and as long as you can find the data when you need it, your program doesn't care exactly where the kernel has buffered it, which makes programming much easier.
21:39 < Elladan> To make things more confusing, "the inode" has different meanings depending on what code you might look at. For example, the inode as stored on disk might be some particular disk block in a particular format, or it might be a value in a b-tree structure, or something else.
21:40 < Elladan> But when the kernel accesses the file, it will read the disk structure (whatever it is) and translate it into an in-memory inode structure of some form, which in the code is also called the inode.
21:40 < one_roOt> does this come through?
21:41 < zenix_2k2> oh wait, so file descriptors values aren't limited to 0,1,2 and 3 ?
21:41 < Sitri> No, I answered that yesterday, they are an int
21:41 < SpaceAce> what's the best partition format for an SSD in an external USB 3.0 chassis, that's going to be plugged into both linux and windows 10 boxes?
21:41 < Sitri> They can theoretically go the full range of a signed int.
21:41 < Elladan> zenix_2k2, file descriptors are small integers. When you open a file/pipe/socket/etc., the kernel will select a number which isn't currently in use and give it to you.
21:41 < SpaceAce> thinking i'll just keep it on NTFS
21:41 < zenix_2k2> Sitri: but i think you said something before about file descriptors which only have 0,1,2 and 3 as the values
21:42 < zenix_2k2> let's me scroll back
21:42 < Loshki> zenix_2k2: file descriptors are small integers, and the rules for their allocation are well defined. This leads to a number of idioms in code that you just have to know about to use them properly.
21:42 < Sitri> zenix_2k2: read the two times I said that very carefully.
21:42 < zenix_2k2> Sitri: "Every process starts with three file descriptors: 0, 1 and 2.  Those are stdin, stdout and stderr respectively."... well you did say that
21:43 < zenix_2k2> and there were file descriptors in it
21:43 < Sitri> Right, but I never said that was the limit.
21:43 < Sitri> By the time you enter main(), those three file descriptors are already open.
21:43 < zenix_2k2> OOOhhhh,,,
21:44 < Loshki> zenix_2k2: by convention, file descriptors 0, 1 and 2 are allocated automatically when a program starts. If you do an open() at this point, a successful open will return the next file descriptor in sequence, so it's value will 3.
21:44 < Loshki> its value will be 3
21:44 < zenix_2k2> so Process's file descriptors (stdin, stdout and stderr) aren't stored in the same place as pds of opened files ?
21:44 < Sitri> "pds of open files"?
21:44 < zenix_2k2> of open() syscall... let's just say that
21:45 < zenix_2k2> when you open a file, it will return a value as a pd ?
21:45 < Sitri> Also file descriptors aren't "stored" anywhere.
21:45 < zenix_2k2> you mean the process's ones or the file's ones ?
21:45 < Loshki> zenix_2k2: in fact, they are the same as other file descriptors, it's just that the system opens them automatically for you, you don't need to call open() to write to 1 or 2, or to read from 0
21:45 < zenix_2k2> and sorry i am quite bad at terms :P
21:45 < Sitri> Where are you getting the term "pd" from?
21:45 < zenix_2k2> oh i mean fd
21:45 < zenix_2k2> opps
21:46 < Sitri> FD == File Descriptor
21:46 < zenix_2k2> yea i know
21:46 < Sitri> When you open a file with open(), you will either get an error (-1) or a new open file descriptor.
21:47 < pnbeast> open() is like a loot box, pre-online-gaming era.
21:47 < Sitri> open() returns the numeric value of that file descriptor.  To your program it's effectively a pointer or a handle.  Only the kernel knows how to handle it properly.
21:47 < Elladan> zenix_2k2, you should understand that inside the kernel, there are data structures and so forth associated with an open file. Because of the user/kernel boundary, you don't access these things directly.
21:48 < Elladan> zenix_2k2, instead, the kernel puts these things into a table associated with your process. Table entry 0 will be stdin, 1 stdout, and so forth.
21:48 < Elladan> zenix_2k2, when you make a system call like read or write, you pass in a number called the file descriptor. This number is nothing more than a table index.
21:48 < zenix_2k2> oh so stdin, stdout and stderr aren't limited to 0,1 and 2
21:49 < Elladan> stdin/out/err are defined as 0, 1, and 2 by convention.
21:49 < Elladan> Those are names for the first 3 table entries in a process.
21:49 < zenix_2k2> but by what you said "Table entry 0 will be stdin, 1 stdout, and so forth"
21:49 < zenix_2k2> let's say i open another process for opening another file
21:49 < Sitri> ...
21:49 < zenix_2k2> well then its stdin,out,err will be 3,4,5 ?
21:50 < Elladan> Yes, the table entry 0 will point to the kernel's data structures for whatever your process's stdin is. When you try to use it, it will access that thing.
21:50 < Sitri> That new process is also going to have 0,1 and 2 be stdin, stdout and stderr.
21:50 < pnbeast> zenix_2k2, no.
21:50 < ziggylazer> Could wget call a local terminal?
21:50 < Elladan> zenix_2k2, the table is per process. Each process has its own table, starting from 0.
21:50 < MrSleepy> does anybody know how I can get my control left and right arrow key to jump by word outside of when I start my window manager?
21:50 < zenix_2k2> oohh... ok get it now :P
21:50 < pnbeast> ziggylazer, wget "calls" web servers.  Unless you have a very unusual terminal, your question doesn't make sense.
21:51 < zenix_2k2> you should have said "each process have its own table" earlier :P
21:51 < Elladan> zenix_2k2, I did.
21:51 < zenix_2k2> wait you did ?
21:51 < Sitri> "into a table associated with your process"
21:51 < Sitri> He did
21:51 < zenix_2k2> oh right... :P
21:52 < zenix_2k2> so where are those tables gonna be stored ?
21:52 < Elladan> <Elladan> zenix_2k2, the file descriptors are integers [whose] meaning is specific to that process.
21:52 < Sitri> They're not stored anywhere
21:52 < ziggylazer> pnbeast, wget can do a lot of things. And I want it to execute a command
21:52 < Sitri> They're in kernel memory.
21:52 < zenix_2k2> well, that is still kinda a place to me
21:52 < zenix_2k2> anyway thank
21:52 < Sitri> Stored implies written to disk.
21:52 < Elladan> They're stored in kernel memory in a place associated with your process somehow.
21:53 < zenix_2k2> well fair enough
21:59 < wizzi> hello, when i click on "Capture" and "interfaces" i get "There are no interfaces on which a capture can be done." on wireshark
21:59 < uruk7> hello i want disable the process vm -> ps auxw | grep vm -> uruk7   6439  0.0  0.0  21536  1036 pts/0    S+   21:58   0:00 grep --color=auto vm
21:59 < TheWild> how crazy, suspending the computer doesn't make the IRC client know the communication has been disrupted
21:59 < lopid> run as root, wizzi?
22:00 < Elladan> TheWild, suspending a computer does not, per se, disrupt a TCP communication channel.
22:01 < Elladan> TheWild, you can, for example, have an ssh connection, suspend your computer, resume it later, and still be connected so long as your IP address is the same.
22:01 < TheWild> how doesn't it timeout?
22:04 < Sitri> A TCP stream isn't a constant stream of packets or data.  Instead it's occasionally transmitted packets with control flags and time-out information.  If you're suspended during a time where no-packets would be transmitted, the other side has no way of knowing that.
22:06 < Elladan> TheWild: TCP timeouts are quite long. IRC has much shorter ones, but you can still suspend and keep the connection if it's brief.
22:23 < sauvin> If it's brief ENOUGH. The bulk if the ping timeouts I see come out to about six minutes. What's the default tcp timeout?
22:24 < jnewt> when i try to log in with cinnamon de selected, i get a black screen with an arrow and then it goes back to the login screen.  can't find any help on this one.   when i try cinnamon --replace, it says it can't find the file for theme Adwaita.   this is supposedly in the gnome-themes-standard, which i have installed.   then i'm unable to use my mate desktop (no response to mouse).  help?
22:25 < akk> It would be nice if IRC clients actually noticed suspensions. IME, they reconnect if the connection is wired or WPA, but not with an open wifi.
22:26 < TheWild> back to Wireshark and its GPL. If I gonna release my script under MIT or WTFPL, does the GPL really bug me?
22:28 < jnewt> Adwaita is in /usr/share/themes.  permissions are 755 root root for everything i can see.   what gives?
22:31 < kerframil> TheWild: basically, either you comply or you don't distribute. if you choose not to comply, the consequence could range somewhere between nothing and being dragged into court, in the extreme case. if so, that's on you.
22:31 < pnbeast> TheWild, what's this script?  I don't use wireshark.  It supports "plugins" or something like that?
22:32 < TheWild> me neither
22:32 < triceratux> jnewt: this will not turn out well. what distro are you attempting this on ?
22:33 < TheWild> Wireshark supports scripts written in Lua, e.g. ones that decode captured packets into something more readable for humans.
22:33 < TheWild> I didn't wrote anything for Wireshark, but the statement really pissed me off.
22:34 < jnewt> triceratux: ubuntu-mate
22:35 < triceratux> jnewt: all i can say is thats why they make a cinnamon spin. in the event anything goes wrong its not trivial to just swap DEs in & out
22:35 < jnewt> why will this not turn out well?   i used to have cinnamon running without issues on ubuntu prior to upgrading from 16.04 lts to 18.04 lts.
22:36 < triceratux> jnewt: did you take the step of actually deinstalling mate ?
22:37 < jnewt> triceratux: not until I have something else running.  i did install xfce, but i couldn't click anything on the desktop or right click and select anything from the menu (it was pretty much useless).
22:38 < jnewt> should i remove them all (mate, cinnamon and xfce and then just install cinnamon)?
22:38 < TheWild> being in the court for that would be really dumb
22:39 < triceratux> after youve tried everything else, yeah
22:39 < kerframil> TheWild: and?
22:39 < kerframil> TheWild: your feelings have no bearing on the standing of the license and its capacity to be enforced
22:39 < kerframil> TheWild: if you don't like the status quo, argue with the wireshark people for change
22:40 < jnewt> triceratux: i don't know what everything else is, i don't have enough information on why it can't find the theme that is where i think it should be.  i'm guessing permission or location is wrong, but i don't know what is right.  and the xfce issue, who knows.
22:42 < jnewt> there seems to be a pretty big knowledge gap here (or resource gap for how all the de stuff interacts with Xorg and window managers, etc) either that or i'm just not sharp enough to understand what is going on here.
22:42 < CyberManifest> how do I feed a text file list into apt to install? Like this: sudo apt install < (cat elementary-5.0-beta1.txt)
22:42 < CyberManifest>  ?
22:43 < triceratux> jnewt: what youre seeing is normal for Adwaita, & i had no problems with 18.04 xubuntu & stuff based on it like voyager. DE config is a process often beyond the abilities of ordinary packagemanagers. im sure its documented somewhere, because the distros manage to do it
22:44 < kerframil> TheWild: if you think it can't happen, read up on Patrick McHardy. he made himself a tidy sum in identifying violators and filing cases against them. his activities were frowned upon by the community at large and he paid a price. the point is, he had the law on his side.
22:44 < triceratux> jnewt: canonical may not even be aware that you cant drop cinnamon in over mate because they essentially never do it. it borders on being unsupported
22:45 < MtotheM> Anyone here using calcurse?
22:45 < compdoc> I was cursed once, does that count?
22:46 < Loshki> MtotheM: I'm guessing a curses interface to a calculator?
22:46 < one_roOt> join #offtopic
22:47 < pnbeast> MtotheM, it works best if you ask your real question.
22:47 < jnewt> triceratux, i installed it over mate on 16.04 without any problems.  the upgrade did something that set me back to mate and broke some part of the process of installing, selecting and logging into cinnamon that worked on 16.04
22:47 < Evidlo> what does the '[-]' button next to people's names on hackernews do?
22:48 < MLarabel> Evidlo: not what you think it does!
22:48 < jnewt> triceratux, ok, i've removed cinnamon, mate, and xfce4.  i'm going to reboot and I expect to get no de at all.  anything i should check before rebooting?
22:49 < Evidlo> what do I think it does?
22:49 < triceratux> jnewt: backup all your data of course
22:49 < MLarabel> dispense bubba gump shrimp cakes
22:49 < jnewt> triceratux: nothing on that partiton that isn't already backed up.  (i'm paranoid)
22:50 < triceratux> youll have a nice runlevel 3 prompt & an apt-get that can get cinnamon up in no time
22:51 < jnewt> nope, i missed something, i got the login, but obviously can't login to anything, just goes to blank screen
22:52 < wizzi> hello, i want to make an owncloud with raspberry to use it like a server ... any Suggestions
22:52 < Dan39> why raspberry pi?
22:52 < Dan39> seems like a silly use of it, imo
22:53 < triceratux> jnewt: wut you removed all the DEs & left the loginmanager & youre still booting into runlevel 4 ?
22:53 < wizzi> Dan39: small and without risks
22:54 < MtotheM> Calcurse is a calender program, my question is basicly if you can have repeatable events while excluding the weekend?  pnbeast
22:54 < jnewt> triceratux: it was the mate-session stuff i think.   trying again
22:57 < jnewt> i've still got the lightdm greeter at boot.   remove that too?
22:58 < triceratux> not much sense having it there until theres at least one DE
23:00 < MLarabel> Let's all send our love down the well for the Thai Soccer Team trapped in a cave~!~!
23:00 < MLarabel> All the way down!!
23:01 < jnewt> ok, third time was the charm, booted and directly into login prompt.
23:03 < pnbeast> MLarabel, I saw some news thing that said they brought four of them out, somehow.
23:03 < MLarabel> pnbeast: oh sweet merciful heavens to mergatroid!
23:04 < triceratux> this is the whole reason i boot from unaltered isos & dont run installers or upgraders. i only have faith in the delivered system as shipped & would never rely on dpkg to effect a recovery of this magnitude from an ubuntu "upgrade"
23:06 < jnewt> triceratux, lesson learned, it was my first upgrade.   my system wouldn't even boot the kernel it came with.   had to use an old one to go get a newer one to get it booted on something recent
23:06 < jnewt> had no sound, no printer, messed up graphics.
23:10 < kubast2> Hey I connected two devices via ethernet ,how do I force them to see/ping each other/acknowledge that they are infact reciveing a ping over ethernet?
23:10 < triceratux> get cinnamon reinstalled & itll be all better
23:10 < kubast2> I looked at the back of my pc and my switch
23:11 < kubast2> and the orange lights indicating that there is a data transfer going on indicates that a desktop is reciving data during a ping command on the other pc
23:11 < Loshki> triceratux: it's simple economics. The 'unaltered isos' are pretty much the only things that have a chance of being tested. After that, no two starting points for an upgrade are likely to be identical. That becomes impossible to QA really quickly.
23:11 < kubast2> ip addr add 192.168.2.2/24 dev enp4s0 ; the other one got 192.168.2.1/24 via a GUI utility from kde
23:11 < Loshki> or even slowly
23:12 < kubast2> both don't ping back
23:12 < [R]> kubast2: do they both think there is a link?
23:12 < kubast2> how do I check for that [R]
23:12 < [R]> dmesg
23:13 < [R]> or mii-tool
23:13 < triceratux> Loshki: exactly. the best thing that ever happened was the advent of the liveiso because theres finally the entire intended product in one place subject to at least some minimal QA. successful linux installatiions of the classical sort were often accidents at best & could quickly degrade if one followed the supported upgrade path
23:13 < triceratux> *happened to linux
23:13 < kubast2> link up on desktop no2
23:13 < TheWild> Github: "You have triggered an abuse detection mechanism." :( :( :( :(
23:13 < Loshki> triceratux: agree completely
23:14 < kubast2> [16103.345780] NET: Registered protocol family 3 on the other pc ? I guess
23:14 < kubast2> that counts as link up?
23:14 < [R]> that has nothign to do with the link
23:14 < [R]> TheWild: stop abusing github!
23:15 < [R]> TheWild: it had a rough childhood
23:15 < [R]> was that too dark?
23:15 < TheWild> I think I've been clicking too fast ;)
23:15 < kubast2> I don't have a network indicator on the other machine but yeh it's connected on the other one
23:15 < jnewt> triceratux, got it reinstalled, enabled lightdm and rebooted.  got the login screen, won't take my password.  i think i've been in this situation before, was i supposed to enable lightdm as root or my user?  i was root when i enabled with systemctl and rebooted.
23:15 < [R]> kubast2: i just told you 2 ways to check...
23:15 < kubast2> I see
23:15 < kubast2> gotta check the package name brb
23:17 < kubast2> SIOCGMIIPHY on 'enp4s0' failed: No such device
23:17 < kubast2> well that's a bigger issue
23:18 < kubast2> ip link returns enp4s0
23:18 < kubast2> do I take th at as link down? [R]
23:18 < [R]> no, you're doing something wrong
23:18 < jnewt> i can drop into a terminal, and startx cinnamon and it works but can't login from the lightdm login screen
23:18 < triceratux> jnewt: totally into the depths. i genuinely have no idea. & if i were to give you ann honest answer id say: "boot the ubuntu 18.04 cinnamon ISO & take a good look at how its supposed to work" rofl
23:18 < [R]> it will say link down if the link was down
23:19 < kubast2> the device was down 0.0
23:19 < kubast2> *interface
23:19 < jnewt> triceratux, i'm close, i can feel it.  i can get to cinnamon now, which i couldn't before.   i also notice it still can't find that Adwaita file it wanted.
23:19 < kubast2> yeah that was only a matter of an interface that was disabled by default [R]
23:20 < triceratux> cinnamon vibration yeah ... positive
23:21 < kubast2> there it goes well I will have to get a gigabit going on my desktop but other than that I'm set
23:21 < kubast2> it run on 100Mbs mode rn I think
23:23 < kubast2> let's see if I can hit a bigger file anytime soon
23:23 < kubast2> hopefully I set it to 1Gb/s
23:26 < [R]> that's what she said
23:26 < kubast2> yeh I have no idea how to do it
23:26 < kubast2> txqueuelen ?
23:28 < kubast2> 10.6MB/s is all I get :(
23:28 < kubast2> over sftp
23:29 < [R]> sounds like 100mbit/s
23:30 < esselfe> [R] you're a mbit/s
23:30 < [R]> esselfe: zoom zoom
23:30 < saderror256> o/
23:33 < Dagmar> If it's a file transfer, the bottleneck is probably the disk.  :/
23:34 < Dagmar> I actually bought 7300rpm disks to mirror so that they could at least have a decent chance at saturating the gigabit here
23:34 < Dagmar> er 7200
23:35 < Dagmar> Now if I can just get the wife to use the network share instead of fussing that her local disks are full
23:35 < Dagmar> ...although that might take stabbing a few people at Adobe, I have to admit I'm up for it.
23:38 < giaco> hello!
23:38 < iltutmus> Hi, I do "... | awk '{print $1}'", how can i print all arguments as they goes. For example from 7 up to last one.
23:39 < giaco> a string expansion went wild and now ls -la says thay I have a file named ''$'\303'. I am trying to delete it, but can't find how to write it
23:39 < giaco> nevermind, just solved
23:40 < giaco> been trying for minutes, the idea came while writing the question
23:41 < Dagmar> It'll require a loop and the use of the NF variable
23:41 < Dagmar> See also https://stackoverflow.com/questions/18457486/print-rest-of-the-fields-in-awk
23:43  * [R] puts Dagmar in a loop
23:43 < esselfe> glaco try 'ls \$\303'
23:44 < esselfe> [R] you're a loop
23:44 < [R]> i'm a roller coaster!
23:44 < [R]> loop de loop
23:46 < DLange> tool is the work you've been looking for, esselfe
23:47 < esselfe> tool?
23:48  * aBound want to be a baller
23:48 < DLange> it's a play on words
23:48 < esselfe> ok, omg a rustc built is 23GB
23:49 < esselfe> I found myself stuck with a full filesystem, unable to create $\303
23:51 < [R]> rust is bigger tahn gcc and glibc and binutils combined?
23:51 < [R]> wtf...
23:51 < [R]> that's just wrong
23:51 < [R]> i tell you what
23:51 < esselfe> what?
23:51 < pnbeast> You're not going to be a trendy, hipster dev with that kind of attitude, [R].
23:52 < [R]> haha
23:52 < [R]> this one guy at my work loves rust and go and all that stuff
23:52 < esselfe> I know the install process requires some 2.5GB using /tmp
23:53 < DLange> [R]: it's web technology!!1! Node is probably bigger :)
23:53 < esselfe> I went out of memory having to remount with 3GB tmpfs, jeez this is a 5hours build
23:53 < phinxy> Is it easy to create a dir that maps to the RAM?
23:54 < [R]> DLange: my node is biggger than yours...
23:54 < [R]> phinxy: tmpfs
23:55 < [R]> esselfe: i guess t hat's why they provide precompiled...
23:55 < esselfe> lol I'm a source-based guy
23:55 < [R]> i kept bluffing the guy at work that i was going to start using rust or go or whatever
23:55 < [R]> one or both of them added arm support
23:58 < aBound> Find the sources.
23:58 < phinxy> could the reason for never having any program occupy more than 50% RAM be because of /dev/shm hogging that space?
23:59 < [R]>  /dev/shm doesn't take up space unless there are files in it
23:59 < revel> Isn't it always supposed to be tmpfs?
23:59 < revel> Oh, he mentioned RAM anyway.
--- Log closed Mon Jul 09 00:00:31 2018