--- Log opened Tue Jun 12 00:00:00 2018
--- Day changed Tue Jun 12 2018
00:00 < Dagmar> diverdude: There's a large difference between the size of a disk and the space utilized by the filesystem on i
00:00 < Dagmar> t
00:00 < diverdude> Dagmar: ok but how can i then find out where space is being used?
00:03 < diverdude> Dagmar: ok but how can i then find out where space is being used?
00:04 < diverdude> How can i see why all space is used on my system disk?
00:09 < mgolisch> how big is the filesystem for /?
00:09 < mgolisch> df -h /
00:10  * TheSHAD0W notes kdirstat and its descendents are handy tools
00:11 < riotz> ncdu is my tool of choice for that
00:11 < TheSHAD0W> https://wiki.debian.org/SynapticsTouchpad#Determining_the_type_of_Touchpad <- ran this, came up with bupkus
00:12 < searedvandal> if my math is correct the total in that paste sums up to somewhere around 7G, so pretty close to 8G if that's the size of /
00:14 < Pentode> searedvandal, yeah it looks about right. diverdude, if you want some kind of detailed analysis of data usage in more detail other than partition wide you are going to have to find a utility or write a script to do something like that.
00:16 < candidat> apt-get install knowledge
00:17 < GunqqerFriithian> >using apt not aptitude
00:29 < pressure679> Have anyone succeded in finding a program which can get a certain number of characters from an offset from a file? I found the sed and awk (answers from askubuntu) are quite slow on a HDD. - In any case: I made a source file using Go version 1.10.1, amd64, compiled with Ubuntu 18.04 was faster, > 1 sec on a file several GB's large. Here is the link: https://code.launchpad.net/~pressure679/+junk/ofo ,
00:29 < pressure679> I am unsure how to notify Canonical or any Linux/GNU distributor, so I write here.
00:31 < diverdude> Pentode: yeah, thx
00:31 < scampbell> pressure679: see the 'dd' command.
00:32 < pressure679> scampbell: Thanks, but I found dd requires knowledge of byte speed if required, which is not the case for every user (e.g it made my favorite USB unusable).
00:32 < pressure679> s/speed if/speed if/g
00:33 < scampbell> pressure679: dd if=something of=outfile bs=1m skip=2 count 3
00:34 < pressure679> - It was a bs value too low which made it unusable, and a USB should have a bs oser several MB's to my knowledge.
00:34 < scampbell> pressure679: great command, sorry you don't like it.
00:34 < pressure679> s/bs oser/bs over/g
00:35 < koala_man> pressure679: tail/head will do that
00:35 < pressure679> koala_man: tail and head takes lines as argument, not the character count.
00:36 < scampbell> use --bytes
00:36 < koala_man> pressure679: it supports bytes
00:36 < pressure679> koala_man, scampbell: ah, thanks.
00:45 < Dagmar> USB-connected drive or not the native block size is still going to be 512 bytes or 4kb
00:46 < Dagmar> The size of the data transfer packets for the USB bus is *not* the largest source of inefficiency if you get it a bit off
00:47 < koala_man> though of course, the dd buffer size is unrelated to the device's native block size
00:47 < candidat> he koala_man are you from australia ?
00:47 < candidat> y
00:47 < koala_man> nope
00:47 < candidat> :-)
00:48 < Dagmar> They get a little angry in AU about why he's calling himself 'koala man'
00:48 < Dagmar> ;)
00:48 < koala_man> yep, they banished me once I was caught getting this nick
00:50 < phogg> banished from Australia? Isn't that a bit backwards?
00:50 < ace510> ah, Australia doesn't like it when you use stuff from there country, guess i'll remove all that merino wool and aluminum from my Amazon cart
00:50 < ace510> their*
00:51 < koala_man> you have a very strange shopping list
00:51 < oerheks> The koala does not drink ..
01:05 < strixdio> anyone know of a good foss/floss budget software/site? preferably something that doesn't mine your data ;)
01:07 < bls> do you want software or a website? the latter isn't really on topic for here
01:08 < strixdio> either one, really.
01:08 < strixdio> I'm installing gnucash right now.
01:08 < strixdio> if there were something I could host locally, that might be awesome.
01:09 < bls> gnucash is a popular one if you want a GUI, ledger is a really awesome CLI with GUI/web wrappers
01:10 < strixdio> hmmmm, ledger sounds interesting.
01:12 < strixdio> bls: thanks :)
01:15 < bls> also see http://plaintextaccounting.org
01:33 < tony-> anyone know if chromeos will function as any other linux distro?
01:34 < koala_man> what does it mean to function as a linux distro?
01:35 < revel> It's running the Linux kernel and Google distributes it.
01:36 < tony-> im not sure how to ask the question. will linux stuff run on it
01:36 < revel> Define "linux stuff"
01:36 < tony-> ya i get the linux kernal thing. just like android
01:36 < tony-> revel like the steam gaming client for example
01:36 < tony-> if the right prerequits are met etc
01:37 < tony-> how ever that is spelled
01:37 < tony-> or if the right dependincies i guess is the right word to use
01:37 < meyou_> chromeos tends to have a lot of things stripped out
01:38 < tony-> ok so it is like android then
01:38 < b1101> tony chromeos isn't linux
01:38 < koala_man> yes
01:38 < meyou_> i recall some uproar about removing ext2 usb drive support
01:38 < koala_man> b1101: what is it?
01:38 < b1101> chromeos
01:38 < b1101> :P
01:38 < revel> I've not heard much about ChromeOS being very great for doing anything Google doesn't want you to do. i.e read-only rootfs by default, little storage space, I'm not 100% sure if it's running something X.org-ish.
01:38 < tony-> i get it now. chromeos is like android. they both use the linux kernal and both are stripped down
01:38 < koala_man> yes
01:38 < tony-> that answers my question then. thanks
01:41 < phogg> revel: chromeOS does not use X
01:41 < koala_man> doesn't it? what does it use?
01:42 < revel> Something Wayland-ish or something Google-specific?
01:42 < revel> SurfaceFlinger? :P
01:42 < phogg> koala_man: their own graphics system on top of fb, I don't even think it's the same as android
01:42 < phogg> not wayland, not surfaceflinger. Hang on
01:42 < koala_man> huh. I was looking at https://www.chromium.org/chromium-os/chromiumos-design-docs/software-architecture and it does list X
01:42 < phogg> https://tech.slashdot.org/story/15/03/09/0220205/google-introduces-freon-a-replacement-for-x11-on-chrome-os
01:43 < phogg> that's the article I remember
01:43 < triceratux> https://www.zdnet.com/article/linux-comes-to-chromebooks/
01:45 < koala_man> looks like that one is through KVM
01:46 < phogg> koala_man: if you look at the link you posted all of the info is from 2009. It's not current.
01:46 < koala_man> yes
02:13 < pantato> hi
02:14 < winst123> hi
02:15 < Valeyard> hello
02:18 < Dexx1_> Whenever I try to do "gem update jekyll", I get: ERROR:  While executing gem ... (Gem::FilePermissionError)     You don't have write permissions for the /usr/local/bin directory. -- any ideas how I could fix this?
02:19 < Dexx1_> Also, autopsy if you are there, I think my problem maybe fixed... hard to say, but things seem normal.
02:21 < bls> Dexx1_: that's a privileged location, if you want to install there, you need to do it as root/via sudo
02:21 < Dexx1_> bls: I upgraded jekyll succesfully using root/su, but as regular user, it shows an outdated version
02:22 < bls> Dexx1_: are you using rvm?
02:23 < Thedarkb> Is there any way to restart i8042?
02:23 < Thedarkb> Oh, saying that made it come back to life apparently.
02:24 < Thedarkb> Anyway, while I'm here, I'm having suspend issues.
02:24 < Dexx1_> bls: I don't think so..
02:24 < Thedarkb> if I type sudo pm-suspend, it turns off my screen and then it immediately comes back on.
02:25 < winst123> I have the same problem with suspend
02:25 < winst123> It only seem to happen for me when firefox is open
02:26 < Thedarkb> [34556.958483] PM: Suspending system (mem)
02:26 < Thedarkb> [34556.958568] Suspending console(s) (use no_console_suspend to debug)
02:26 < Thedarkb> [34556.959022] wls1: deauthenticating from a0:e4:cb:42:91:04 by local choice (Reason: 3=DEAUTH_LEAVING)
02:26 < Thedarkb> [34556.974494] sd 0:0:0:0: [sda] Synchronizing SCSI cache
02:26 < Thedarkb> [34556.974773] sd 0:0:0:0: [sda] Stopping disk
02:26 < Thedarkb> [34556.975215] e1000e: EEE TX LPI TIMER: 00000000
02:26 < Thedarkb> [34556.978935] xhci_hcd 0000:05:00.0: WARN: xHC save state timeout
02:26 < Thedarkb> [34557.016146] suspend_common(): xhci_pci_suspend+0x0/0xd0 returns -110
02:26 < Thedarkb> [34557.016150] pci_pm_suspend(): hcd_pci_suspend+0x0/0x30 returns -110
02:26 < Thedarkb> [34557.016153] dpm_run_callback(): pci_pm_suspend+0x0/0x130 returns -110
02:26 < Thedarkb> [34557.016157] PM: Device 0000:05:00.0 failed to suspend async: error -110
02:26 < Thedarkb> [34557.486166] PM: Some devices failed to suspend, or early wake event detected
02:26 < Thedarkb> oops.
02:26 < Thedarkb> Sorry.
02:26 < Thedarkb> It only happens to me when Pale Moon is open.
02:26 < Thedarkb> Which is a cousin of modern firefox.
02:26 < Thedarkb> forked off in 2008 I think.
02:32 < Thedarkb> winst123, I found out my problem device
02:32 < Thedarkb> Device 0000:05:00.0 failed to suspend async: error -110
02:32 < Thedarkb> 05:00.0 USB controller: Renesas Technology Corp. uPD720202 USB 3.0 Host Controller (rev 02)
02:32 < Thedarkb> It's my sketchy USB 3.0 ExpressCard.
02:36 < Thedarkb> winst123, Try putting it to sleep and give me your dmesgs
02:38 < redredhathat> hey can someone help me write to the terminal on my other machine
02:39 < MikeFromIT> What do you mean by write to the terminal redredhathat
02:39 < redredhathat> im ssh'd in and write doesnt exist
02:39 < redredhathat> like i want to display text to an open terminal
02:39 < redredhathat> and i cant echo > tty7
02:39 < bls> is wall installed?
02:40 < MikeFromIT> Is write installed on it
02:40 < redredhathat> wall works
02:40 < redredhathat> write says no command, oddly
02:40 < winst123> Thedarkb, it's working fine right now, it's really random, can't figure out how to reproduce it
02:41 < Thedarkb> That's how it was for me too.
02:41 < Thedarkb> When it happens again, get your dmesgs.
02:42 < winst123> It's like a 1 in 50 chance it happens, i'll make sure to get my dmesgs next time
02:43 < bls> if it's isoloated to a single device, you can often add a hook to modrobe -r on suspend and modprobe on resume
02:46 < Thedarkb> Nah, I'll just throw the card in the bag in case I need it.
02:46 < cr0w3> having problems with scp from a remote server. What is the proper way. scp <file> user@host: /remote/     is what I've tried
02:46 < Thedarkb> Not worth the hassle.
02:47 < bls> cr0w3: no space between the user@host spec and the remote directory
02:47 < ayecee> cr0w3: scp user@host:/dir/file localfile
02:47 < cr0w3> ayecee:thank you
02:48 < ayecee> :)
02:48 < cr0w3> bls:you too
02:48 < MikeFromIT> Anyone here set up a samba server with windows clients before?
02:48 < bls> with a space between the nick and the message :P
02:49 < ayecee> MikeFromIT: one way to find out is to ask a question about it
02:49 < cr0w3> lol
02:49 < bls> MikeFromIT: that's too generic of a question to get anything but a useless response
02:49 < ayecee> q e d
02:49 < cr0w3> bls: thanks again
02:50 < MikeFromIT> So I have samba working to where each user has their own share that they can access, but can't get any to access the public share that should be open for anyone
02:50 < ayecee> that sounds frustrating
02:51 < bls> MikeFromIT: being able to provide verbose logging from both ends would go a long way in triaging this
02:51 < ayecee> might even solve it outright
02:51 < bls> hey, the first step is admitting you have a problem, so at least we're there
02:52 < ayecee> which step submitting yourself to a higher power?
02:52 < ayecee> ah, that might be a different program
02:52 < bls> too many steps in that one
02:54 < MikeFromIT> Annnnnnnnnnnnd I found my typo
02:54 < MikeFromIT> Nevermind, false
02:54 < ayecee> \o/
02:54 < MikeFromIT>  alarm
02:54 < ayecee> /o\
02:55 < dotcomboom> \o\
02:55 < dotcomboom>  /o/
02:55 < ayecee> get down, party time
02:56  * dotcomboom parties
02:56 < ayecee> party like it's 1999
02:57 < ayecee> you know, because nick
02:57 < dotcomboom> xd
02:57 < dotcomboom> unfortunately all bubbles burst at some point e.e
02:57 < ayecee> but what a ride until they do
02:58 < dotcomboom> let's invest in some pets.com stock eh?
02:59 < ayecee> i'm down with the c o m
02:59 < ayecee> shut up and take my money!
03:03  * Pentode makes it rain
03:07 < storge> i read that bane as Penetrode
03:07 < ayecee> missed opportunity
03:07 < Pentode> nice
03:09 < ayecee> i wonder what the cultural touchstones are in computer culture in other languages.
03:09 < Dexx1_> Whenever I try to do "gem update jekyll", I get: ERROR:  While executing gem ... (Gem::FilePermissionError)     You don't have write permissions for the /usr/local/bin directory. -- any ideas how I could fix this?
03:10 < Dexx1_> Also, storge if you are there, I think my problem maybe fixed... hard to say, but things seem normal.
03:10 < ayecee> Dexx1_: do you have write permission for the /usr/local/bin directory?
03:10 < Dexx1_> ayecee: it shows that error because I don't right?
03:10 < ayecee> stands to reason, yes
03:10 < ayecee> normally only root would have write access there
03:11 < winst123> run it as root?
03:13 < storge> Dexx1_: what was the problem? bad initrd's or something? --oh right the loop
03:14 < storge> Dexx1_: what did you do
03:14 < leopard> test
03:14 < storge> passed
03:14 < leopard> thanks :)
03:16 < Dexx1_> storge: I just changed it to "no" from https://ubuntugenius.wordpress.com/2010/05/24/fix-a-failed-initramfs-update-do-it-before-you-reboot/ -- then, tried a bunch of stuff, changed it back and it SEEMS to have done the trick. Can't confirm.. is there a way to check?
03:16 < Dexx1_> ayecee: so, what should I do?
03:16 < ayecee> Dexx1_: perhaps run the command as a user that has write access to that directory
03:16 < ayecee> or change the permissions of that directory, but probably not that.
03:17 < Dexx1_> :/
03:17 < Dexx1_> That doesn't really help.
03:17 < ayecee> well i'm sorry
03:17 < ayecee> you're not giving me much to go on.
03:17 < storge> sudo gem update foo   doesn't work?
03:19 < ayecee> i'm also sorry about saying i'm sorry when i wasn't really sorry
03:24 < ayecee> which i suppose means i was sorry after all
03:24 < ayecee> sorry for bringing it up
03:24  * storge doesn't forgive ayecee 
03:25 < ayecee> i can accept that
03:34 < storge> ok i forgive you
03:34 < ayecee> whew
03:53 < pyro_> what does echo $! resolve to?
03:54 < ayecee> i don't remember offhand, but i know the bash manpage has it
03:55 < ayecee> manpage says "expands to the process id of the job most recently placed into the background"
03:55 < iflema> what about $$
03:55 < Dagmar> Current pid
03:56 < pyro_> Dagmar: says the current pid of the bash
03:56 < Dagmar> Your point being?
03:56 < pyro_> ayecee: what does that mean..  a job ran with command &.. its process id?
03:56 < ayecee> pyro_: exactly!
03:56 < pyro_> ok
03:57 < ayecee> i often use this with dd to get status, e.g. dd etc.etc.etc. & ,then, while kill -0 $!; do kill -USR1 $!; sleep 5; done
03:58 < ayecee> kill -0 returns true if the process exists, with no other effects. kill -USR1 to dd tells it to report status.
03:58 < ayecee> with a 5 second interval in the loop
04:01 < ayecee> i'm sure gnu dd has some clever thing that makes this unnecessary now, though
04:01 < Dagmar> conv=progress I think
04:01 < pyro_> i want to demonstrate bg and fg jobs working to my student. i use vagrant.. i want to run a command that takes sometime to finish.. could someone suggest me a command
04:02 < Dagmar> Building a kernel from source
04:02 < Dagmar> Granted, that's going to jibber a lot
04:02 < Dagmar> updatedb would be a good candidate
04:02 < Dagmar> It should take at least a few minutes
04:03 < Dagmar> ...or if you have a couple of very small hard drives you can mdraid them together and force an explicit sync check
04:03 < Dagmar> (4Tb could easily be 14+ hours, so... SMALL HARD DRIVES)
04:03 < storge> i noticed updatedb on my ssd is lightning compared to on my hdd. it's not even close.
04:03 < Dagmar> Random seek times make a huge difference there
04:04 < ayecee> pyro_: sleep 60 &
04:04 < Dagmar> You can cheat those a bit with something like, find /some/paths /moar/paths -exec cat {} ]; 2>/dev/null 1>/dev/null
04:04 < Dagmar> It results in everything being in the filesystem cache
04:05 < Dagmar> (kernel builds run almost horrifyingly fast after a stunt like that)
04:05 < pyro_> sure Dagmar ayecee
04:08 < ayecee> also dd if=/dev/zero of=/dev/null for various sizes
04:09 < storge> between using ccache and localmodconifg, my kernel compiles on this old laptop are rather quick
04:09 < joe_z> or, you know...
04:09 < joe_z> sleep 30
04:09 < joe_z> sure, it doesn't do anything useful...
04:09 < swift110> storge: which laptop
04:10 < storge> T520
04:10 < joe_z> oh, ayecee suggested that already
04:10 < storge> i just compiled 4.17.1 in the last 5 minutes
04:11 < storge> i guess it was more like 7
04:11 < storge> but a generic kernel and i'd be walking away for supper
04:12  * storge boots it
04:14 < swift110> storge: oh nice I have several thinkpads myself
04:14 < swift110> storge: why did you compile the kernel
04:14 < storge> swift110: just because
04:14 < storge> 4.17.1-storge
04:15 < storge> it's a weekly hobby habit
04:15 < pyro_> ayecee: just a curiousity
04:15 < pyro_> [2]-  Running                 sleep 46 &
04:15 < pyro_>  [3]+  Running                 sleep 56 &
04:16 < pyro_> all i did was Ctrl+z and then bg %2 ; bg %3
04:16 < pyro_> but why for [2]- Running and [3] + Running
04:16 < pyro_> what are those - and+
04:16 < ayecee> what should you see instead?
04:16 < ayecee> oh
04:16 < ayecee> i don't know offhand
04:17 < pyro_> aah ok..
04:18 < swift110> storge: I mean no special reason at all?
04:19 < storge> swift110: not really beyond hobby interest. i do like to see what's changed between versions, even if i'm usually saying N to everything newly introduced (nothing new for my old machinery)
04:22 < storge> whenever i get a different computer i like to strip the kernel of everything that machine doesn't need. just for the hell of it really. my computers are entirely inconsequential anymore. my work machine is employer provided and runs Microsoft, so it's purely hobby intent
04:23 < Pentode> storge, i used to enjoy doing that back in the day.
04:23 < storge> still, over time i've seen the kernel images grow inexorably over time
04:23 < Pentode> when i used to run slackware on my p120, i'd trim it down as much as humanly possible.
04:23 < storge> in the 3.x era my kernels were all around 3.5M, now they're around 4.4M
04:24 < storge> Pentode: that's what got me into it actually. i was running sourcemage on an old IBM 390 (pre-"thinkpad") and if i didn't want the compile to take days, it was all about localmodconfig
04:26 < storge> i still have that laptop, i should get it out some day. smaller and yet much heavier than my larger/newer thinkpad T520. but also seemingly immortal; it's old case is like a tank
04:26 < Pentode> i dont remember them taking _too_ long to compile. this was back in the 2.2.x era
04:26 < Pentode> ha
04:27 < Pentode> the oldest thinkpad i have is a 760cd
04:27 < storge> not in the toughbook category but the hinges on it seem very well built
04:27 < Pentode> sadly the battery leaked and took out a _single_ line on the keyboard's multiplexer
04:27 < Pentode> so the i k and , keys dont work :(
04:28 < swift110> goodness
04:28 < storge> oh that's a point, i should get it out just to examine the battery
04:28 < Pentode> oh yeah and the battery _still_ holds a charge for a while, too. believe it or not.
04:28 < Pentode> i pull it out to charge it every once and a while..
04:29 < swift110> storge: sorry about that. my x201 overheated and shut off
04:29 < storge> brb
04:29 < Pentode> i think it's NiMh though. if it were lithium ion im sure it'd be stone dead by now
04:30 < Pentode> if i could trace the pins out it may be possible to solder a jumper wire on there somehow.
04:33 < swift110> ok cool
04:33 < swift110> Pentode: how old is the 760cd
04:34 < Pentode> i think it was released in 94 or 95
04:35 < swift110> Pentode: does it still boot up?
04:35 < swift110> goodness thats old
04:35 < Pentode> yeah still works fine otherwise
04:35 < swift110> My oldest thinkpad is a t40 with a dead graphics card
04:35 < swift110> My actual oldest laptop is a year older and is a compaq presario 2500 with 704 mb ram
04:35 < Pentode> i have a t43. i actually still use it almost daily.
04:36 < demio> welp
04:36 < swift110> Pentode: what draws you to it
04:36 < demio> just caught a randome fucker with a dev blog
04:36 < swift110> what DeMiNe0
04:36 < demio> trying to cryptomine with my CPU
04:37 < swift110> oh
04:37 < demio> -_-
04:37 < demio> wtf
04:37 < demio> a fucking dev blog
04:37 < swift110> wow
04:37 < demio> did he not expect other devs to notice
04:37 < pyro_> is it possible to connect my storage from aws to vagrant using nfs and setting the entry in fstab file?
04:37 < Pentode> i always liked thinkpads. almost everything about it. the keyboard, the way it looks, feels.
04:38 < Pentode> but IMO it has one of the best laptop keyboards i've ever used.
04:39 < swift110> yes I agree Pentode classic thinkpads are really something indeed
04:40 < Pentode> yeah. i grabbed this one up when i was working at a repair shop. it's one of the last true thinkpads and my boss bought it from some corporation in a bulk pile
04:40 < Pentode> this one seems like it was setup and then just put on a shelf and _never_ used
04:41 < Pentode> i can't find a single spot where it looks like an appreciable number of fingers have come in contact with it ;p
04:41 < cmj> ibm keyboards?
04:41 < Pentode> thinkpad keyboards
04:42 < cmj> with clitmouse?
04:42 < Pentode> clitmouse and trackpad
04:42 < cmj> yeah we had a bunch
04:42 < cmj> used them with old pcmcia radios for hotspots
04:43 < cmj> some had uptimes for 4-5 years
04:43 < Pentode> my only complaint is the LCD. the resolution is fine but the contrast ratio is pretty bad, even for the era.
04:44 < cmj> yeah this is back when embedded boards where just coming out. so laptops were used. old (large) microtik routers replaced them
04:46 < Pentode> i could see them making nice little servers
04:46 < pyro_> what is the difference between inode and file descriptor? i know inode is a datastructure used by files and inode is just a number in the memory
04:46 < Pentode> and theyre neatly stackable
04:46 < pyro_> what is fd? and also if we do 2>/dev/null.. is this someway connected to fd?
04:48 < Pentode> a descriptor is an _identifier_ to open a file.
04:49 < Pentode> a handle basically, in other words.
04:53 < Pentode> pyro_, im not sure what you are asking
05:07 < z3t0> hey all I am trying to debug random crashes on my system and haven't gotten much luck so far
05:07 < z3t0> I am trying to find a log that shows the crash
05:07 < z3t0> using journalctl -xe --boot -1
05:07 < z3t0> but it seems that the logs stop about 2 minutes or so before the crash occurs
05:07 < Pentode> z3t0, try dmesg
05:07 < Pentode> hmm
05:08 < Pentode> what kind of crash behavior?
05:08 < z3t0> and the crashes do really feel random as they happen about once a day?
05:08 < z3t0> I just had two crashes about four minutes a part
05:08 < z3t0> Pentode, so the entire pc seems to freeze, even the caps lock light doesn't repsonde
05:08 < z3t0> respond*
05:08 < Pentode> sounds framebuffery
05:09 < Pentode> check dmesg, see if anything is in there at all
05:10 < z3t0> okay
05:10 < Pentode> also check /var/log/Xorg.0.log
05:12 < Dagmar> If the caps lock key doesn't work anymore, the machine is basically dead
05:12 < Pentode> yeah
05:12 < z3t0> I am seeing an acpi error that is something which appears on every boot
05:12 < Pentode> that there be a hard lockup
05:12 < z3t0> It may be relate
05:12 < Dagmar> Doubtful
05:12 < Dagmar> Still, you may wind up having to add kernel params like noaspm or somesuch
05:13 < z3t0> https://pastebin.com/7V2wcvF1
05:13 < z3t0> Not really sure what that error is but I'll see what I can figure out
05:13 < Pentode> _usually_ hard lockups like that are graphics related, though it could be disk I/O, power supply ripple, DC-DC converter failure on the main bus / processor supply, or even RAM
05:13 < swift110> anyone use solus?
05:14 < z3t0> I'll go ahead and do a system upgrade and see if I have the same issue again and investigate further
05:14 < z3t0> It's fairly difficult to debug as it is random
05:14 < Dagmar> That's a bug in the kernel, but it's not what's killing the machine
05:14 < Dagmar> Numerous people are reporting it, but no one else is reporting hard lockups
05:16 < Dagmar> Pentode's right about random hard crashes being a giant PITA to debug
05:16 < Pentode> sucks :/
05:17 < Dagmar> All things being equal, running a yum update or whatever isn't likely to make things worse (well, without a very different problem manifesting)
05:17 < Pentode> yeah.
05:17 < Dagmar> ...which reminds me that I need to update my emby box in preparation for burning some stupid out of it with a blowtorch
05:18 < Dagmar> I've had just about enough of the media server just sitting around in gc-land for whole minutes at a time
05:28 < Pentode> well we can't have that
05:28 < Dagmar> I suspect whoever built the packages for it did something not-so-smart with the mono binaries
05:29 < Pentode> well packaging is the least favorable job maintaining
05:29 < Dagmar> I don't care for mono, but it doesn't normally just sit around soaking up 99% of the CPU like I see it doing with emby
05:29 < Pentode> eesh, doing what?
05:37 < Dagmar> As best I can tell, it almost has to be garbage collection
05:38 < Dagmar> Disk IO is nothing, and memory usage drops when it's over
05:39 < Dagmar> I'm about to just migrate the entire setup over onto a Slackware VM where I've built and tested each of the packages myself
05:39 < Happyhobo> I can't remember my syndaemon command syntax to make my mouse inactive for three minutes while I write.
05:40 < Dagmar> However, their idea of documenting dependencies apparently involves psychic visions.
05:40 < Pentode> haha
05:40 < Dagmar> ...why I went with CentOS for that VM to begin withh
05:40 < Happyhobo> syndaemon -v -t 3?
05:41 < Pentode> ugh. i have to reboot. it's been 35 days, i hate rebooting.
05:41 < SuperSeriousCat> I use Docker for stuff that require mono. It always used all my cores at max speed on VM's whenever needed, but inside Docker it sits nicely
05:41 < Pentode> everything bad happens when you reboot
05:42 < ayecee> at least you get to choose when you reboot
05:42 < ayecee> unlike when other bad things happen
05:42 < Pentode> your glass is always half full ayecee
05:43 < ayecee> with a silver lining
05:43 < Dagmar> Docker won't help this, since emby itself is navel-gazing while mono-sgen is off doing whatever
05:44 < Dagmar> The VM seerver has plenty of clockcycles and RAM to go 'round
05:44 < Dagmar> Like, enough to transcode h264 at 1080 to two different TVs at once
05:46 < Dexx1_> Whenever I try to do "gem update jekyll", I get: ERROR:  While executing gem ... (Gem::FilePermissionError)     You don't have write permissions for the /usr/local/bin directory. -- any ideas how I could fix this?
05:47 < Dagmar> You generally need to be root to update the system
05:47 < Dagmar> ...especially if it's trying to write to /usr/local/bin
05:47 < ayecee> Dexx1_: run as a user with write permissions to that directory
05:47 < ayecee> Dexx1_: i mean honestly. you're still at the same point, with no progress?
05:47 < Pentode> we told you earlier Dexx1_
05:47 < swift110> oh
05:48 < ayecee> Dexx1_: maybe you could tell us what you've tried since, and what happened when you did
05:50 < ayecee> (my bet is that this is macos and he's being coy)
05:52 < swift110> hmm
05:52 < swift110> https://www.youtube.com/watch?v=AUaJ8pDlxi8
05:52 < ayecee> swift110: no.
05:53 < ayecee> this is not your dumping ground for youtube urls.
05:56 < Dexx1_> ayecee: how do I safely change permissions to that directory so I can write to it?
05:56 < ayecee> Dexx1_: you normally don't.
05:57 < ayecee> Dexx1_: you would run the update process as root.
05:57 < Dexx1_> Right. I mentioned that I already did that succesfully. Root shows 3.8, but my regular user shows 3.0 (old version)
05:57 < Happyhobo> ayecee:  what is the syndaemon command I need?  It's frustrating to be writing and have the curse move somewhere else on the ss
05:57 < ayecee> Happyhobo: no idea
05:57 < Happyhobo> screen
05:58 < Happyhobo> wow
05:58 < ayecee> Dexx1_: i guess you must have said that earlier some time before you repeated the question without that detail.
05:58 < ayecee> Dexx1_: maybe uninstall the user-level version of jekyll.
05:59 < Happyhobo> Why is it Manx cats like to have their stubby little tails up when they turn around in your lap and put their bubtts in your face.
05:59 < ayecee> Happyhobo: because you touch yourself at night
05:59 < Happyhobo> Yes, yes, my day is made.  I have seen a cat sphincter up close
06:00 < swift110> hmm
06:02 < dvvdv> which of these woudl you use as your main Linux OS?
06:02 < dvvdv> https://i.imgur.com/m9dgae5.jpg
06:02 < Pentode> o_O
06:02 < searedvandal> Dexx1_, if you follow the documentation for Jekyll, it's explained how to set up Jekyll without the need to install it as root.
06:03 < dvvdv> i cant decide
06:03 < Pentode> well they are both hideous but unity sucks worse than anything known to man, even more than the giant vacuum cleaner in space balls, so...\
06:03 < dvvdv> the lastest ubuntu doesnt use unity
06:04 < Pentode> oh, well the one of the bottom looks like it.
06:04 < searedvandal> which of those? for all we know they can be the same linux distro.
06:04 < dvvdv> ubuntu vs mint?
06:04 < Dexx1_> searedvandal: I already had jekyll installed and I am following the docs, but have gotten stuck in this step.
06:04 < Happyhobo> Pentode do you know syndaemon?
06:04 < Dexx1_> (when attempting to upgrade)
06:04 < Pentode> Happyhobo, nope.
06:05 < Happyhobo> Oh jim jim jim\
06:05 < Happyhobo> katie are you in Chesapeake or Virginia Beach?
06:06 < searedvandal> Dexx1_, if you set up GEM_HOME as the docs explain, gem update jekyll should update the jekyll in your GEM_HOME
06:06 < Dexx1_> ayecee: even if I remove and re-install, it keeps installing the old 3.0 version
06:11 < Dexx1_> searedvandal: I think it works now. BTW, do you use Jekyll?
06:11 < Dexx1_> (sounds like you do)
06:12 < searedvandal> not currently. but yeah, I've used Jekyll before.
06:12 < storge> jekyll - Simple, blog aware, static site generator
06:39 < mephux__> anyone running linux on their pixelbook using containers? Wondering if anyone else is having tty issues with gpg
06:45 < lord|> what's an easy way to intentionally congest a network
06:45 < lord|> i.e. send a shitton of udp packets
06:45 < moony> How many CPU cores can Linux handle at once
06:45 < lord|> moony: 4096?
06:46 < [R]> a million
06:46 < moony> wew
06:46 < moony> yea, 2048 in practice
06:47 < moony> thanks
06:52 < Triffid_Hunter> lord|: ping -f
07:01 < killown> https://www.reddit.com/r/linux/comments/8qdkd2/microsoft_already_knows_the_solution_switch_to/ lol
07:01 < killown> microsoft linux distro is comming any time soon
07:15 < Pentode> whew. glad thats over with. hopefully i can go without rebooting for atleast another month or two.
07:15 < [R]> yeah, screw security updates
07:15 < [R]> who needs em
07:16 < DildoSwaggins> windows
07:17 < Pentode> bah i don't have anything to hax0r anyway.
07:18 < jim> find a hackerspace near you
07:32 < foxmask> o/
07:35 < foxmask> i'm on a remote connection to my office desktop and i would like to activate an usb stick I plugged in it but can't find the device and mount point I made yesterday. How can I enable it again as I dont have physical acces to remove and plug the stick again ?
07:36 < [R]> what do you mean activate?
07:36 < foxmask> i mean, as if I just plug in
07:37 < [R]> if the device isn't showing up in /dev
07:37 < [R]> then you broke somethign
07:37 < [R]> and it has to be removed
07:37 < searedvandal> and if it's showing up but not mounted, you'll have to manually mount it
07:38 < foxmask> ok so it's dead ... in the log I have "FAT: Directory bread(block xxxx) failed"
07:38 < foxmask> thanks for your replies
07:52 < za1b1tsu> can someone recommend me a nice CLI pdf reader?
07:52 < DildoSwaggins> hmmm, not sure if such a thing exists?
07:53 < Triffid_Hunter> za1b1tsu: strings :P
07:53 < DildoSwaggins> you would need to convert the PDF to text
07:53 < Triffid_Hunter> or at least html then feed it to links
07:53 < za1b1tsu> I don't mean voice reader
07:54 < za1b1tsu> like viewer, with vim-like bindings?
07:54 < Triffid_Hunter> who mentioned a voice reader?
07:54 < za1b1tsu> ah ok, so there is no such thing
07:55 < za1b1tsu> ok, so out of curiosity, why is it hard to read pdfs from the terminal?
07:56 < Triffid_Hunter> za1b1tsu: because they're basically a list of individual characters in specific locations, it's kinda hard to convert that to linear readable text in a sensible fashion
07:57 < searedvandal> you can use a tool like pdftotext
07:57 < za1b1tsu> I see, damn you learn something new everyday. And the text converters are accurate or buggy?
07:58 < DildoSwaggins> searedvandal +1
07:58 < DildoSwaggins> but za1b1tsu, the quality of the conversion will depend on the PDF
07:58 < searedvandal> or if you wanna see how a .pdf looks raw, do less file.pdf
07:58 < DildoSwaggins> some PDF's are comprised of images
07:58 < DildoSwaggins> whereas others are properly compiled with text, you will have better success w/ these
07:58 < DildoSwaggins> but the formatting may not be ideal
07:59 < za1b1tsu> I never used Latex but heard good things about it, can I convert it to latex?
07:59 < DildoSwaggins> like, if your PDF has 2 columns at the top, the text will return both columns as if it was 1 line
07:59 < DildoSwaggins> hmmmm
07:59 < DildoSwaggins> i dont know about from PDF to LaTeX, but if you get the PDF to Text, then you can do Text -> LaTeX
08:00 < Triffid_Hunter> za1b1tsu: and many PDFs are somewhere in between, should be able to get text out but it may be somewhat scrambled
08:06 < za1b1tsu_> yeah... I guess gui is the choice then :D
08:06 < za1b1tsu_> too bad
08:50 < Dagmar> That's a pretty big 'if'
08:51 < Dagmar> There is no requirement for the internal structure of PDF files to make anything resembling human "sense"
08:51 < Dagmar> It's almost more reliable to re-OCR them
08:55 < hexnewbie> Buy Acrobat (version supported by Wine), convert them to DOCX, use LibreOffice to export as .tex
08:56 < hexnewbie> Although for plain text with no hyphenation, the output of pdftotext is almost valid LaTeX
09:02 < sinatrablue> http://tinypic.com/r/2435h5j/9
09:02 < sinatrablue> am i doing it right?
09:13 < chomwitt> with evtest i can see the usb usage ids of my keyboard, also i've found the keycode->keysum related files. But where is the mapping scancode(usb usage id) -> keycode ?
09:21 < psyk3d> hi
09:22 < psyk3d> Does anyone know how we can change wallpaper in linux manually
09:22 < psyk3d> I mean without using feh or nitrogen etc
09:22 < sinatrablue> what de?
09:23 < kirk781> psyk3d, won't it depend on the desktop environment?
09:23 < psyk3d> I use i3wm
09:23 < Triffid_Hunter> psyk3d: xsetroot
09:23 < Triffid_Hunter> psyk3d: unless you have a WM/DE with its own background thing that overrides it
09:24 < psyk3d> I think i3wm doesn't provide a background
09:24 < well_laid_lawn> most tilers don't
09:24 < sinatrablue> Ah i3 doesnt provide it
09:24 < sinatrablue> you need to edit the x11 root window
09:25 < psyk3d> Ok, Ill check xsetroot out
09:26 < sinatrablue> feh or xsetroot
09:26 < psyk3d> sinatrablue: hows arch treating you?
09:26 < sinatrablue> psyk3d, Good so far, i actually really like it
09:27 < sinatrablue> KDE is a little buggy, but thats not arch's fault
09:27 < psyk3d> nice
09:45 < system16> hi. i have installed xubuntu on a usb drive. it boots fine but its super slowww it takes like 10 min for it just to load desktop icons. is this because of the usb drive ? it uses usb 2.0
09:46 < system16> its not live cd
09:46 < DildoSwaggins> yes, it's slower because of the USB
09:46 < paulcarroty> system16, usb2 is slow like shit
09:47 < system16> so if i install xubuntu on a usb 3.0 drive, things will go faster ?
09:47 < paulcarroty> of course
09:47 < DildoSwaggins> yes, as long as you plug that usb 3.0 drive into a usb 3.0 port
09:47 < system16> and btw how can i see disk activity in linux (im a windows user so im not a linux pro)
09:47 < paulcarroty> iotop
09:48 < system16> i want to avoid terminal as much as possible
09:48 < paulcarroty> not in Linux, buddy
09:49 < system16> i know some really basic stuff tho like cd,cp,pwd etc.
09:49 < system16> so sudo -i iotop ?
09:50 < paulcarroty> yep
09:50 < system16> k thanks 4 ur help
09:50 < paulcarroty> nvm
09:55 < system16> if linux is free then how they make money ? linux doesnt show ads on desktop.
09:56 < Pentode> there is a foundation that takes donations. companies that use it in their product charge for services and support.
09:56 < DildoSwaggins> because many big corporations depend on open-source software, such as Linux, they often fund development of it
09:57 < system16> so linux doesnt sell private info to advertisers like most other companies ?
09:57 < Pentode> ^ most donations come from said corporations
09:57 < DildoSwaggins> and yes, like Pentode said, there are enterprise Linux distros that come w/ support
09:57 < notmike> Um, no
09:58 < Pentode> no thats google and facebooks job
09:58 < Pentode> they have that whole market cornered
09:59 < system16> yup. reality is sad
10:01 < Pentode> and the more people use their services and the more money they make the better they get at figuring out "who you are, and what you like" and ever more valuable that data becomes.
10:01 < Pentode> vampires
10:02 < hackinghorn> what distro should I start with when using linux
10:02 < mutante> Debian. /me closes case
10:02 < Pentode> debian or debian based is a good start
10:03 < mutante> except if you do "debian based" you dont get support in #debian. so... Debian
10:03 < Pentode> you'll eventually figure out what you like, though.
10:03 < hackinghorn> ah, thanks
10:03 < Pentode> just don't tell them, then. ;)
10:04 < mutante> they will see it as soon as you start asking about PPAs and synaptic and pastebin stuff
10:04 < Pentode> hah
10:05 < oerheks> or use a fork of a fork, like mint, much more fun
10:06 < system16> fork ? i thought fork is for eating stuff
10:06 < Pentode> a fork of a fork is a fork ofork. but everyone knows a fork's a fork..
10:06 < mutante> or.. realize that you can install different window managers on Debian.. and there is no need to have a separate distribution to replace the window manager
10:06 < oneko> ofork ? We have a word like that ?
10:07  * system16 .exe (.deb) has stopped responding
10:07 < hackinghorn> mmm
10:07 < mutante> what exacly is more "fun" about using something that copies all the packages from Debian and then changes a few random settings.. not sure :)
10:07 < hackinghorn> sweet mutante
10:08 < system16> ubuntu is the best
10:08 < psyk3d> system16: depends on who you ask
10:08 < system16> looks good. feels good. and its EZ to use
10:08 < Pentode> everything depends on who you ask. that's the beauty of it.
10:08 < psyk3d> looks good - any distro can look however you wan
10:08 < psyk3d> want*
10:09 < psyk3d> feels good - I dont know how you can feel a distro? you run you finger over a live cd and feel for grooves?
10:09 < Pentode> well in that case we should all still be using SLS.
10:10 < mutante> just install the WM you like.. no point in creating all those distros. imho it's more for the ego of people who like to say they made their own distro
10:11 < psyk3d> lol use lfs :P
10:11 < Pentode> lol
10:11 < system16> i meant its not that hard to use
10:11 < mutante> flawed comparison. all you have to do is select a damn package name..
10:11 < psyk3d> I think the major difference between distros is their package manager, and init system
10:12 < Pentode> let's not get into one of those arguments.
10:12 < psyk3d> yea :D
10:12 < Pentode> it makes everyone grouchy.
10:12 < mutante> ok, keep spending time on trying to find the "best distro" and use something obscure and then live without the support.. that's what a newbie needs, right
10:12 < DildoSwaggins> fuck you, no it doesn't
10:13 < psyk3d> someone's triggered already :D
10:13 < deniska> The major difference between distro is their default wallpaper
10:13 < sauvin> DildoSwaggins, mind the language and the attitude.
10:14 < shrdlu68> How am I supposed to heat my house if I don't use a source-based distro?
10:14 < mutante> hackinghorn: best you can do is idle in this channel and watch this discussion repeated over and over again.. then you will see .. and meanhwile ..just install any of the major distros.. it doesnt matter for now
10:14 < psyk3d> I say if you're new to linux, just install ubuntu! after you've had some experience, get rid of ubuntu and use another distro
10:14 < Pentode> I won't disagree that there are too many "xerox distributions" out there, though they don't _all_ fall in this area.
10:15 < sauvin> shrdlu68, do what I do, use a retro computer with a 21" CRT.
10:15 < mutante> psyk3d: or just.. install Debian  and <the end>
10:15 < deniska> shrdlu68: cryptocurrencies \o/
10:15 < psyk3d> Ubuntu has a large n00b friendly community
10:15 < deniska> If you are new to linux install arch or gentoo. Once you are experienced, install ubuntu
10:16 < Pentode> apparently on linus tech tips they connected an entire radiator to a graphics card cooler and managed to use it as a heater effectively, lol.
10:16 < psyk3d> shrdlu68: lol
10:16 < mutante> If you are new to Linux, spend the first 6 months in tryign to find the "best". dont install anything
10:16 < DildoSwaggins> sauvin, i apologize, it was meant to be a joke that just came out wrong. I was trying to be funny with "<Pentode> it makes everyone grouchy."
10:16 < Pentode> DildoSwaggins, dry sarcasm is easy to miss on IRC. ;)
10:17 < dob1> hi, I would like to monitor if there is some disconnection of my internet connection, right now I am just using a "ping google.com",  but there is something more efficent?    I would like to see if the site doesn't respond in 10 seconds for example
10:17 < psyk3d> mutante: thats kinda counter intutive, don't spend 6 months finding a good distro, get one now and start using it
10:17 < psyk3d> and distro will teach you about linux basics
10:17 < storge_> whatever distro you do use, my advice is to focus on the commandline. it will make all other linux distros make more sense, and you'll see more easily what makes them alike and what makes them different
10:18 < sauvin> What storge_ said. You'll furthermore have a LOT more power under your fingertips.
10:18 < mutante> psyk3d: yes, i switched to sarcasm to make him stop obessesing over the #1 newbie question and just start
10:18 < psyk3d> lol sry then
10:19 < sauvin> psyk3d, "sorry", not "sry".
10:19 < mutante> dob1: you should use the TCP protocol instead of ICMP (ping). it's possible that firewalls drop ICMP while TCP works fine
10:19 < storge_> laughing out loud, not "lol"
10:19 < mutante> dob1: so that would mean "mtr" for example
10:20 < mutante> dob1: "mtr google.com" . if you have a GUI it is a GUI window
10:21 < dob1> mutante, I take a look
10:22 < heftig> mtr uses icmp by default, you need to specify --tcp (and probably also run as root)
10:22 < mutante> oh, sorry, yes. ^ use tcp
10:23 < mutante> i was thinking of "tcptraceroute" . you can use that too
10:25 < dob1> ok !
10:28 < password2> hi , what does linux use as a newline marker again?
10:29 < EriC^> \n ?
10:30 < password2> i copied batch files fromwindon't and nw when i run them it says '\r' command not found
10:30 < EriC^> there are some tools that convert windows to linux
10:30 < EriC^> dos2unix
10:30 < Armand> Hahahaa
10:31 < shrdlu68> tr -d \\r
10:31 < sauvin> I think Linux uses 0x0a.
10:31 < Armand> EriC^: I understand what you're trying to say, but that came out SO wrong! :P
10:31 < password2> shrdlu68: and the filename?
10:32 < shrdlu68> Filename?
10:32 < EriC^> password2: yeah add that to the end
10:32 < Pentode> password2, as in *.bat files? you realize they wont work, right?
10:32 < password2> eh ,sh files
10:32 < Pentode> oh ok
10:33 < password2> *.sh so shell , soryy my mind autocompleted the wrong word
10:33 < Pentode> i thought you went mad
10:33 < password2> tr does not seem to take a file?
10:33 < sauvin> So pipe it!
10:33 < shrdlu68> tr -d \\r < file
10:34 < password2> and tehn >file to pipe it out again?
10:35 < shrdlu68> Yep, or use sed and edit in-place.
10:35 < password2> wait no my file got emptied
10:35 < password2> script runs faster than ever , also does nothing
10:36 < pankaj_> I want to insert some text above the login prompt so that whenever I login before that above the '$USER Login ' that text appears. What is the text file that controls the behaviour?
10:36 < Pentode> pankaj_, /etc/motd
10:37 < EriC^> pankaj_: which os?
10:37 < j1745-26> pankaj_ /etc/issue
10:38 < pankaj_> EriC^: Archlinux
10:38 < Pentode> oh, that's been changed _too_?
10:38 < j1745-26>  /etc/motd prints after login
10:38 < Pentode> oh yeah
10:38 < Pentode> forgot, thanks
10:39 < pankaj_> j1745-26: Yes, it is what I want.
10:40 < password2> ok , so somehow tr don't like it when i pipe from the same file I['m piping to
10:40 < password2> i guess it clears teh outfile first
10:40 < pankaj_> j1745-26: But just if I want to run a command for that view I want. how to do that?
10:40 < shrdlu68> password2: Not just tr, piping doesn't work like that.
10:41 < Pentode> password2, yeah that's not gonna work.
10:41 < shrdlu68> Piping doesn't work sequentially.
10:42 < markus_d> If files have 744 rights and I am root should I be able to change them (they do not belong to root)? Is there a difference between sudo to change them and being root?
10:42 < j1745-26> pankaj_ you want to run a command before login?
10:43 < EriC^> markus_d: yeah you should be able to change them
10:43 < pankaj_> j1745-26: Yes, like the user name is printed
10:43 < pankaj_> j1745-26: Sorry, my name is printed
10:43 < EriC^> markus_d: if sudo gives you root privileges then it's not that different, possibly some environment variables and stuff like that are different depending on the sudo options you used
10:44 < j1745-26> pankaj_ there is some escape sequence to put in /etc/issue that resolves to your name i think
10:44 < j1745-26> but i don't think you can run a command
10:45 < pankaj_> j1745-26: OK. I have echoed the text (which was a pictures made from ascii characters) appended to /etc/issue
10:46 < pankaj_> j1745-26: But I now want it like: After I login the name of the user gets printed like "Welcome $USER"
10:46 < j1745-26> pankaj_ here's a little guide: http://linuxfromscratch.org/blfs/view/svn/postlfs/logon.html
10:46 < Pentode> pankaj_, mine has a \l at the end of it. maybe that's it?
10:47 < Pentode> it looks like this: Ubuntu 16.04 LTS \n \l
10:47 < markus_d> EriC^: I get the error that I only have the right to read the directory
10:48 < j1745-26> pankaj_ if you want it after the login you have to put it /etc/motd
10:48 < password2> yeah , i got it to work
10:48 < password2> i just piped to a new file
10:48 < password2> my script is now runnign fine
10:49 < pankaj_> j1745-26: OK. Worked
10:49 < crankslider> https://www.preining.info/blog/2018/06/microsofts-failed-attempt-on-debian-packaging/
10:50 < pankaj_> j1745-26: What is the way to add color to text to console or tty.
10:51 < j1745-26> pankaj_ do you mean on the prompt?
10:51 < j1745-26> or like writing in a text editor?
10:52 < pankaj_> j1745-26: Yes
10:52 < pankaj_> j1745-26: Like decoration so that it looks good.
10:52 < shrdlu68> pankaj_: https://en.wikipedia.org/wiki/ANSI_escape_code
10:53 < DildoSwaggins> <password2> script runs faster than ever , also does nothing
10:53 < DildoSwaggins> lol
10:53 < j1745-26> pankaj_ https://www.cyberciti.biz/faq/bash-shell-change-the-color-of-my-shell-prompt-under-linux-or-unix/ scroll down to Add colors to the prompt
10:54 < pankaj_> j1745-26: OK
10:54 < jason85> Has anyone managed to run WoW with wine?
10:54 < pankaj_> j1745-26: It would be nice if I could say 'Welcome $USER' in /etc/motd based the username that logs in.
10:55 < j1745-26> pankaj_ you can add that to the .profile or .zprofile
10:56 < j1745-26> I think there's some env variable you can use
10:57 < pankaj_> j1745-26: Yes, but how
10:57 < j1745-26> open .profile in a text editor and add in the last line echo 'Welcome $USER'
10:58 < j1745-26> edit .zprofile if you use zsh instead of bash
10:58 < j1745-26> put it in .bashrc to display the message in the terminal too
10:59 < uselessPos734> so linux users hate installation wizards. but why ? i cant see a problem with wizards
11:00 < j1745-26> uselessPos734 what installation wizards?
11:00 < Pentode> package managers are wizard enough
11:01 < uselessPos734> j1745-26, its a windows thing. u go to a website>>click on the download link>>a wizard pops up
11:02 < Dagmar> "Would you like to install our malware for free?  Choose Y)es! or S)ure!
11:02 < shrdlu68> uselessPos734:  https://www.reddit.com/r/linux/comments/8q92hs/microsofts_failed_attempt_on_debian_packaging/
11:04 < uselessPos734> yes. microsoft tried this linux idea. they made an app called windows store (i cant remember the name) but it sucks
11:05 < j1745-26> uselessPos734 we don't talk of the devil here
11:07 < uselessPos734> oh
11:12 < MarkusDBX> is there any DIY selfhosted "serverless/lambda" project out there? Just want a serverless-style interface for a docker cluster I guess.
11:17 < phob0s> hello
11:18 < phob0s> ip netns add 123 && ip netns exec 123 iptables-save && ip netns del 123
11:18 < deim0s> Hello
11:18 < phob0s> Why I dont have output of iptables-save there?
11:18 < phob0s> if I do "ip netns add 123 && ip netns exec 123 iptables -L &>/dev/null && ip netns exec 123 iptables-save && ip netns del 123"
11:18 < phob0s> I have output ?
11:21 < Skunky> phob0s: but... you won't have any iptables rules, you just created the ns...
11:28 < phob0s> but I have some default values like: :INPUT ACCEPT [0:0]
11:28 < mcdnl> those are counters
11:29 < phob0s> why they appear if I add  "ip netns exec 123 iptables -L &>/dev/null"
11:29 < phob0s> and there is nothing if I ship that command?
11:29 < phob0s> s/ship/skip
11:30 < mcdnl> because netspace 123 doesnt have rules?
11:30 < dbolser> am I being dumb?
11:30 < dbolser> The file is g+rw, but I can't change permissions or ownership
11:30 < phob0s> in both cases it does not have any rules
11:30 < dbolser> (I'm in the same group)
11:31 < mcdnl> only root changes ownership
11:31 < dbolser> weird
11:31 < phob0s> but the longer command provides some output
11:31 < mcdnl> and only owner changes perms
11:31 < dbolser> can I change group level permission?
11:31 < dbolser> so I can delete and re-create the file, but not change ownership?
11:31 < mcdnl> phob0s: iptables-save prints the current rules
11:31 < mcdnl> in "save" format
11:31 < mcdnl> you have to save them in a file
11:32 < mcdnl> for the "standard" ns its stored in /etc/iptables/rules.v4
11:32 < mcdnl> havent worked with netspaces in linux so i dont know how does it work for netspaces
11:32 < dbolser> mcdnl, thanks for help
11:33 < mcdnl> yes dbolser, even if it doesnt make sense in this situation, it has a reason
11:33 < phob0s> when I did strace, iptables-save had nothing to read from socket, so maybe some delay is needed to get iptables-save state after ns creation ....
11:34 < mcdnl> no problem btw
11:34 < mcdnl> socket? why would iptables-save read a socket?
11:34 < mcdnl> iptables-save extracts the in-memory rules and prints them in "save" format
11:35 < mcdnl> "ip netns add 123 && ip netns exec 123 iptables -L &>/dev/null && ip netns exec 123 iptables-save && ip netns del 123"
11:35 < mcdnl> shouldnt have output as the ruleset for the newly created namespace will be empty
11:36 < mcdnl> maybe not empty but no rules at all, just the counters
11:36 < phob0s> yes
11:36 < phob0s> exactly
11:36 < phob0s> but when I do without "ip netns exec 123 iptables -L &>/dev/null"  I do not see even counters. No output
11:36 < phob0s> that is my question
11:36 < mcdnl> well, you create a new namespace "123", list the rules and send output to null (dunno why), then export them to save format and then delete the namespace
11:37 < mcdnl> because you are redirecting output to /dev/null
11:37 < mcdnl> remove the redirection
11:37 < mcdnl> and you'll see output
11:37 < phob0s> you dont get my point
11:38 < phob0s> can you run both commands
11:38 < phob0s> and see difference ?
11:40 < mcdnl> dont do && to chain commands
11:40 < mcdnl> cmd1; cmd2; cmd3
11:40 < phob0s> that doesn't change anything in this case
11:40 < phob0s> :)
11:41 < Armand> Chaining via && is perfectly acceptable in most cases.
11:42 < Armand> Cheatsheet:
11:42 < Armand> A; B    # Run A and then B, regardless of success of A
11:42 < Armand> A && B  # Run B if A succeeded
11:42 < Armand> A || B  # Run B if A failed
11:42 < Armand> A &     # Run A in background.
11:42 < mcdnl> unless a bad or inusual return code from one of the commands stops the chaining
11:42 < Armand> Context is for Kings.
11:43 < mcdnl> yeah, circumstantial
11:43 < mcdnl> phob0s: i dont get any output either, dunno why
11:44 < mcdnl> haha, i found your question on unix stackexchange
11:44 < Armand> <mcdnl> yeah, circumstantial < Very much so.
11:45 < mcdnl> phob0s: either it takes a while or until you dont add or query rules, iptables-save doesnt output anything
11:45 < mcdnl> seems so
11:46 < mcdnl> (probably until you dont use the firewall chains they don't get created as there are no default rules)
11:48 < DildoSwaggins> <Armand> Chaining via && is perfectly acceptable in most cases.
11:49 < DildoSwaggins> Only when each sequentual command depends on the one before it returning 'true'
11:49 < DildoSwaggins> in `A && B && C`, B will only run if A completed successfully
11:49 < DildoSwaggins> if you dont care about what each command returns, then you need to chain then w/ a ;
11:49 < DildoSwaggins> A; B; C
11:50 < searedvandal> nice little cheatsheet you shared there Armand
11:50 < DildoSwaggins> oh
11:50 < DildoSwaggins> i guess i should have scrolled up a bit ^^
11:51 < DildoSwaggins> but yes, that cheatsheet is perfect
11:52 < Armand> :)
11:52 < Armand> Google ftw
11:53 < mcdnl> </dev/tcp/localhost/11211 && echo Port open. || echo Port closed.
11:53 < mcdnl> this is one good example of the use of both && and ||
11:53 < mcdnl> also, nifty way to check if a port is open without netcat/nc/telnet
11:54 < shrdlu68> (useless use of nc)
11:54 < DildoSwaggins> <mcdnl> phob0s: either it takes a while or until you dont add or query rules, iptables-save doesnt output anything
11:54 < DildoSwaggins> <mcdnl> (probably until you dont use the firewall chains they don't get created as there are no default rules)
11:54 < DildoSwaggins> that's correct
11:54 < DildoSwaggins> if you do `man iptables-save` you will see that:
11:54 < mcdnl> hah, it was a wild guess
11:55 < DildoSwaggins>        iptables-save — dump iptables rules to stdout
11:55 < DildoSwaggins>        ip6tables-save — dump iptables rules to stdout
11:55 < DildoSwaggins> so if it doesn't output anything, that means there are not any iptable rules to output
11:56 < phob0s> default policy is not considered as rule ?
11:56 < mcdnl> unless you change it from accept
11:56 < TyrfingMjolnir> For openssl How do I do mulitple IPs( one public and 3 local IPs ) for the same device cert?
11:57 < TyrfingMjolnir> CN=?
11:57 < TyrfingMjolnir> Multiple NICs on the same node, i e
11:58 < mcdnl> why use ips?
11:58 < mcdnl> use a domain name
11:58 < Armand> Because an FQDN still needs to resolve to an IP, no ?
11:58 < mcdnl> so what? certificates are issued for fqdns
11:59 < tds> you can use an IP as a SAN, it's pretty unusual though
11:59 < Armand> Indeed... Just seemed a little rhetorical. lol
11:59 < shrdlu68> IPs are perfectly legitimate.
11:59 < DildoSwaggins> it needs to resolve to an IP, yes, but dont use IPs in the certificate
11:59 < tds> ie cloudflare did it for https://1.1.1.1
11:59 < mcdnl> that was a bad idea
11:59 < mcdnl> well not
11:59 < mcdnl> but thats a special case
11:59 < mcdnl> xD
12:00 < shrdlu68> TyrfingMjolnir: Not sure how to do it with openssl, try cfssl?
12:00 < inthl> I have hosts with lots of files on their filesystems, all ext4 and I do have the issue that the systems I/O causes slowdowns. this is nothing to wonder about, but I noticed that the buff/cache is not exhausted as it should be. The hosts have 64G of ram, software running on it uses say 10G, the cache is only like 5G. now there are times when it becomes higher and the overall perforamance is really better, but then sometimes the caches dr
12:00 < inthl> op to just 5G. I wonder about this, since saying that, about 40G of RAM are free all the time. Now I followed most of the instructions given say here: https://unix.stackexchange.com/questions/30286/can-i-configure-my-linux-system-for-more-aggressive-file-system-caching?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa ...still I do not see that much caching happening, or when the cache raises to say 10-15G, it alm
12:00 < inthl> ost like gets truncated down to 5G automatically. can one help me please ajust these settings? I would prefer as much caching as possible since as mentioned I clearly see that this helps (a lot)
12:00 < mcdnl> anyway, if you need to use the same certificate for connections from outside (public ip) and for inside (private ips) just make the dns resolve to local ips when queried from inside
12:01 < mcdnl> inthl: take a look on btrfs
12:01 < shrdlu68> IIRC, Openssl asks for such details when generating a CRL.
12:01 < inthl> I can not switch TB of data on those fs that easily now. how do you think btfs would help here btw?
12:03 < mcdnl> i read something about it being good for that kind of situations
12:03 < mcdnl> just a suggestion though
12:10 < djph> inthl: what's happening when the caches "drop"?
12:11 < inthl> djph, hard to say, since the I/O is not constant, the access is very random, usually several hundred files per second out of a xyz million pool of files on the fs
12:12 < inthl> but it happens that I monitor lots of io-wait in say top
12:12 < inthl> then again: seeing on such a use case just 5G in buff/cache seems a little low to me
12:12 < djph> inthl: I meant that is is shortly after a reboot; or early in the day; or something else that'd indicate perhaps a cache-cleaning operation
12:13 < inthl> to my knowledge there is no cache-cleaning operation set up
12:13 < djph> I mean, I don't know of *nix "automatically" flushing the cache (unless there's some kind of a timer on it, that if you're not using it ... )
12:15 < inthl> well that's what I am thinking of. look at the first answer of that stackechange discussion. I took over most of that, except vfs_chache_pressure (still 100, the default), and dirty_*_centisecs
12:16 < inthl> for *that* many I/O the hosts have, I see like 5-8G of caches at the moment, also clearly that it is decreases sometimes. Example: right now I see 8.02G, a feew seconds later it is 7.96, ..and stays there for a while. but again: over 40G of ram are availble
12:17 < inthl> I would expect this to constantly rise and rise until there is no more free ram
12:18 < inthl> ...where the caches are freed when other applications request more RAM
12:21 < djph> except that they also fall as data gets written to disk and "ages out" (that's all I can figure) -- ooh, found what might be the documentation
12:21 < djph> sysctl/vm.txt.gz
12:21 < djph> in the linux kernel documentation (part of the sourcecode)
12:22 < inthl> do you know perhaps where that data is stored, by means of a directory mapped in / somewhere? I am just thinking, maybe there is a privileges issue since I did change some chmods here and there for different reasons
12:23 < djph> in debian, linux-doc ... but that was an ancient forum post, so IDK if it's still there (still downloading)
12:25 < djph> yep, it's in there
12:26 < djph> /usr/share/doc/linux-doc-4.9/Documentation/sysctl/vm.txt.gz <-- after installation of the "linux-doc" package
12:28 < inthl> can you pastebin that somewhere perhaps?
12:33 < Daesh> kernel.org/Documentation
12:33 < felix_vs> Hi, I want to rename multiple files with one command e.g. I have directory with file1.txt and file2.txt , want to rename suffix from .txt to .md . I tried doing this with 'mv *.txt *.md' and wasn't able to. Figuring out how to make 'mv' do this: I read on unix.SE that 'mv' doesn't rename multiple files at once, and I should use 'rename' instead. Since I'd rather not learn more commands than necessary, I want verification: can 'mv'
12:33 < felix_vs> be used to do this or not?
12:33 < Daesh> err https://www.kernel.org/doc/Documentation/
12:33 < Daesh> no need to pastebin something that is already online
12:33 < ananke> felix_vs: you can't use mv for it that way. you can use 'rename' tool instead though
12:34 < Daesh> note: `rename` can be one of many tools, all with vastly different behaviour
12:34 < Daesh> some distroes ships the util-linux rename, some ship perl-rename, some ships one of the bsd variants
12:35 < Daesh> so you have to read the manual for the one your distro ships
12:35 < felix_vs> Daesh: ah yes. I'm using Linux Mint and it ships with perl-rename
12:36 < felix_vs> Daesh: I wanted to use 'mv' for this, because it is universal command: GNU mv software AFAIK is ubiquitous
12:36 < Daesh> for loop then
12:36 < ananke> felix_vs: except you can't. not without additional steps
12:36 < felix_vs> ananke: yes. that's why I ask here; for verification!
12:36 < Daesh> for f in herpaderp; do mv "${f#//...}; done
12:37 < felix_vs> Daesh: hmm that's in bash script yes?
12:37 < Daesh> in whatever shell you are using
12:37 < Daesh> replace the ${} with the string processing required
12:40 < Daesh> ${f//.txt/.md}
12:40 < Daesh> see https://mywiki.wooledge.org/BashGuide/Parameters
12:40 < Daesh> er.. you want /% for obvious reasons
12:41 < felix_vs> Daesh:  for FILE in DIR_PATH; do mv "${%f//.txt/.md};
12:41 < felix_vs> Daesh: like that you mean?
12:42 < felix_vs> ops
12:42 < Daesh> don't use caps
12:42 < felix_vs> Daesh: for FILE in DIR_PATH; do mv "${%FILE//.txt/.md};
12:42 < Daesh> caps is for session variables, use lowercase
12:42 < felix_vs> Daesh: ok
12:43 < felix_vs> Daesh: didn't want to use 'file' for fear maybe it is keyword in bash
12:43 < felix_vs> Daesh: and didn't want to shorten to 'f' because i never used this before so I won't shorten it right away
12:44 < Daesh> for f in *; do mv "$f" "${f/%\.txt/\.md}; done    test with printf before you run with mv
12:44 < Daesh> (missing one ")
12:44 < felix_vs> Daesh: thanks for the suggestion
12:44 < felix_vs> Daesh: i'm not ready to dive into bash yet :-)
12:45 < felix_vs> Daesh: also not familiar with regex yet
12:45 < Daesh> see the guide I linked to, it is somewhat decent
12:45 < Daesh> https://mywiki.wooledge.org/BashGuide  the full thing
12:45 < felix_vs> Daesh: alright i bookmarked and will revisit sometime hopefully :p
12:45 < felix_vs> Daesh: it's nice to know that i CAN do this...
12:46 < jim> here's some regexp, in which we're matching patterns:
12:47 < Daesh> if this is for some script then I would consider some other language than shellscript
12:47 < jim> the regexp: a just matches itself, and also abcd matches itself
12:48 < jim> why does it only match itself? because the chars: a b c and d have no special meaning in regexp
12:48 < inthl> well after reading all that I still do not know why caching is kinda like disabled. I also just played around with vm.dirty_ratio = 5 and vm.dirty_background_ratio = 5 just to see what happens, after sysctl -p and a few minutes of waiting, obviously nothing has changed with the buff/cached values
12:48 < felix_vs> Daesh: heh ye I remember doing something similar in non-shellscript (python) in the past...  can't remember the syntax for the life of me ;-)
12:48 < jim> a{4} matches aaaa
12:49 < jim> felix_vs, with me so far/
12:49 < jim> ?
12:49 < felix_vs> jim: wasn't following, sorry
12:49 < felix_vs> jim: didn't understand your 2nd sentence
12:49 < jim> which one?
12:50 < felix_vs> jim:  'the regexp: a just matches itself, and also abcd matches itself'
12:50 < jim> oh the one that starts "why....")
12:51 < jim> felix_vs, ok... the goal of regexps (for our purposes right now) is to match some text
12:51 < redredhathat> can I compile a wm myself locally and login into it all without sudo?
12:51 < Daesh> felix_vs: python pathlib is quite nice
12:51 < Daesh> https://docs.python.org/3/library/pathlib.html#module-pathlib
12:52 < jim> felix_vs, there are some chars in regexp that have special meaning, and rather than matching themselves, they have other effects
12:52 < felix_vs> jim: yea i know a bit about regex, but I hate memorizing syntax rules so i only do it when i have no other choice ;-) so i'd rather steal somebody else's ready-made regex after google-searching with me not using it often enough to warrant memorizing...
12:54 < jim> felix_vs, ok, so should I assume you don't want to do this? :)
12:57 < jim> felix_vs, lisp and scheme are said to have no syntax, other than the ones with combinations
12:57 < felix_vs> jim: yea sorry. i can see you're interested in communicating this information, but i'm the lazy type who can't stand listening to other people's normal-speech-pattern explanations. I can't sit through a mere 5 minutes lecture. I cope with it by creating my own resources
12:58 < jim> ok, thanks for letting me know
12:58 < felix_vs> jim: hopefully you're ok with no sugar-coating. :-s
13:00 < felix_vs> jim: i probably came across as anti-social on second read. tried to put it as neutral.
13:00 < jim> well what I'm ok with right now, is that you don't want to deal with syntax rules and lectures :)
13:00 < felix_vs> jim: yep!
13:00 < jim> not sure where the commas are in your last sentence :)
13:01 < felix_vs> i don't see commas in my last sentence
13:01 < felix_vs> which one?
13:01 < gotcha> hey guys, i have a harddrive connected to a pc and i booted using systemrescuecd, how can i access the internal harddrive?
13:01 < BluesKaj> Howdy folks
13:01 < jim> the one where you say "hopefully you're ok with no sugar-coating"
13:02 < jim> generally speaking, I'm ok :)
13:02 < jim> heya blues
13:02 < BluesKaj> Hi jim
13:04 < felix_vs> jim: "hopefully you're ok with no sugar-coating. :-s" no comma here
13:04 < jim> ok ok... what was the sugar coating?
13:07 < Yamakaja> Is there a good way to set the ip that an openvpn servers will use as the gateway for its clients?
13:09 < Yamakaja> Maybe my wording wasn't optimal, so lets describe it differently: An openvpn server will usually take one address in the range which is to be used for clients that is then sent to the clients as their peer address. How can i make that address fixed and/or set it the the first one in the subnet?
13:10 < spear2> what dotfile/script would i normally put the command 'setxkbmap -opion "ctrl:swapcaps"' if i want that to be the default for the system?
13:12 < spear2> -option
13:13 < shrdlu68> Yamakaja: Give it a one-address range?
13:13 < Yamakaja> Then i don't have space for clients - but seems that i've missunderstood something here
13:13 < Yamakaja> The server interface actually has the first ip in the range assigned, which is what i want
13:14 < spear2> i guess it needs to go in x windows configuration somewhere
13:14 < Yamakaja> The issue is: The server and client each have another (and different) ip out of that range assigned as the peer of the interface, which is confusing me
13:17 < tds> Yamakaja: can you upload a copy of the openvpn config?
13:18 < Yamakaja> it's just a normal "server 172.20.20.72 255.255.255.248" (172.20.20.72/29)
13:19 < tds> so does openvpn add an additional ip to the tun interface on the server? (as well as the first one in the subnet)
13:19 < redredhathat> how can i add the command "source activate base" to my zshrc?
13:19 < Yamakaja> tds nope
13:19 < redredhathat> When I add it to the top of my config, it says activate is not  valid
13:20 < searedvandal> spear2, .xprofile or .xinitrc
13:20 < Yamakaja> On the server: inet 172.20.20.73 peer 172.20.20.74/32 scope global
13:20 < Yamakaja> Client: inet 172.20.20.78 peer 172.20.20.77/32 scope global tun0
13:22 < tds> hmm, is openvpn in net30 mode or whatever?
13:23 < Yamakaja> net30 mode? 0.o
13:24 < searedvandal> redredhathat, what should that command do for you?
13:25 < tds> oh yeah, apparently that's the default still, I see the same behaviour on one of my servers, adding topology subnet sorts it
13:25 < spear2> searedvandal: so i should not use /etc/X11/xorg.conf.d/00-keyboard.conf? https://wiki.archlinux.org/index.php/Keyboard_configuration_in_Xorg#Using_X_configuration_files (i'm not actually using archlinux)
13:26 < redredhathat> searedvandal start my python virtual env with conda
13:26 < searedvandal> spear2, that should work too.
13:27 < searedvandal> redredhathat, try conda activate base
13:30 < redredhathat> thanks searedvandal
13:30 < tds> Yamakaja: did switching the topology sort that?
13:30 < tds> I guess there's no harm in using net30 anyway, you just waste some address space
13:45 < pankaj_> How to decorate the login prompt in linux?
13:45 < EriC^^> the login prompt or command prompt?
13:45 < mcdnl> login prompt? the password: thingie?
13:46 < pankaj_> EriC^^: I am using linux. So, I do not know whether using 'command prompt' is good choice.
13:46 < stane1215> the terminal prompt or login manager?
13:46 < pankaj_> EriC^^: But yes, tty or console as you say it.
13:46 < mcdnl> depends on shell
13:47 < pankaj_> stane1215: That in tty. Like coloring and making it good looking like PS1 variable and other stuff
13:47 < mcdnl> take a look on zsh and powerline for example
13:47 < mcdnl> pretty and shiny and does the work
13:48 < pankaj_> mcdnl: OK.
13:49 < pankaj_> mcdnl: It looks good
13:49 < mcdnl> :)
13:52 < hackinghorn> Linus doesn't use Ubuntu or Debian
13:52 < hackinghorn> should I
13:54 < pxfgod> on *nix, there is a signal called EINTR. If a thread is blocked by a syscall, and an interupt occurs. The thread fall into kernel mode to deal with the interupt. and WHY should that syscall returned with EINTR? I mean that after dealing with interupt, it only needs to keep waiting until done. WHY return with EINTR to notify you to retry the syscall???
14:01 < mawk> pxfgod: it only happens with long syscalls
14:01 < mawk> like read(), write(), etc
14:01 < mawk> and it's facultative, you can tell the signal handler to restart long syscalls instead of making them fail
14:01 < mawk> it's a feature to be able to interrupt long syscalls
14:01 < mawk> well not maybe the first intended use but it's a possible use
14:02 < mawk> sigalarm(42); if (long_syscall() >= 0) sigalarm(0) else puts("syscall took too long");
14:06 < pxfgod> mawk, so it's a chance for user-mode signal-handler to run?  Why not after signal-handler run, then, the thread automatically returns to kernel-mode and continue to wait syscall to be done???
14:06 < mawk> I just said that's it's optional
14:06 < mawk> you can ask for that behavior if you want
14:06 < mawk> nobody's preventing you from doing so
14:07 < varshitbhat> what
14:07 < mawk> it's just that some people like their long syscalls to be interruptible by signals
14:07 < mawk> like read(), write() and all the I/O family
14:09 < pxfgod> mawk, I know. It's up to you to decide whether to restart. OS just provides you an oppotunity, right?
14:11 < mawk> yes pxfgod
14:12 < jozefk> o/
14:12 < jim> hi
14:13 < jozefk> anybody knows a live distro which can boot with secure boot UEFI turned on, and with ddrescue package available, also that it can mount exfat formatted drives?
14:14 < mawk> ubuntu/fedora
14:14 < mawk> these are the only two I know with secure boot support
14:14 < mawk> because they got their keys signed by MS
14:14 < mawk> once inside just install ddrescue or whatever
14:15 < jozefk> then I would need wifi connection or cable to be able to install additional packages
14:16 < mawk> or put the package on a usb key, or anything
14:16 < gotcha> hey guys, booted using systemrescuecd, opened up GParted and one of the partitions has a key icon, i assume it means its mounted but i cannot unmount it
14:16 < gotcha> any ideas?
14:17 < jozefk> gotcha, what message do you get when you try to umount it?
14:18 < gotcha> jozefk, GParted is showing that partition /dev/sda5 Mount Point is /dev/md2 .. when i try to umount it just says "not mounted"
14:18 < jozefk> try to do it in terminal
14:19 < jozefk> close gparted first
14:19 < gotcha> jozefk, yes, im doing it in terminal, GParted is not giving me an option to unmount it.. its greyed out
14:19 < gotcha> ah ok
14:19 < gotcha> let me try again
14:19 < goodafternoon> hi there
14:19 < gotcha> jozefk, same thing
14:20 < goodafternoon> someone can explain me the differences between VIRT, RES and SHR columns of htop command please ?
14:20 < jbit> gotcha: if /dev/md2 is what's using the partition, it means it's part of a RAID array or an LVM group
14:20 < jozefk> first type "mount" and see what is mounted and where
14:21 < gotcha> jbit, it was part of a raid, yes
14:21 < gotcha> im trying to format it to use it on another system
14:22 < NetTerminalGene> i learnt that ISPs and the internet sites knows how long you are staying that site. i sthis measured with java script? if i disable js, can they still know how long i stayed the site?
14:22 < NetTerminalGene> is this*
14:22 < gotcha> jozefk, when i run mount its not listed there
14:22 < jbit> gotcha: you'll need to tell linux to disolve the raid array before you can resize it
14:23 < gotcha> jbit, really? :s i have no idea how to do that
14:23 < jozefk> if it was part of RAID then I don't know how to do it
14:23 < jbit> gotcha:  mdadm --remove /dev/md2 /dev/sda5   should do i think
14:23 < greenit> hi, is it possible to convert ttf-fonts to psf-fonts? I want to use hack ( https://sourcefoundry.org/hack/ ) on a headless server
14:23 < jbit> gotcha: i'm not sure if this will preserve data on the disk... so if you have data on it that you care a bout.. please backup!
14:24 < gotcha> jbit, yeah, the data already has been backed up to a larger drive. I do not mind completely removing all the data from the drive
14:26 < jbit> gotcha: if you want to start from scratch you can do something drastic like: sgdisk --zap-all /dev/sda
14:26 < jbit> gotcha: that will remove all partitions for the drive
14:26 < jbit> you probably want to reboot after running that
14:26 < gotcha> jbit, mdadm did not work, says hot remove failed cuz the device or resource busy
14:28 < jbit> gotcha: hrm.... maybe:   mdadm --fail /dev/md2 /dev/sda5    and then try the remove? (if you don't want to do --zap-all)
14:28 < jbit> (it's been a while since i've touched mdadm... i usually use ZFS for my raid needs these days :) )
14:30 < gotcha> same
14:30 < gotcha> ill just try the zap-all
14:31 < gotcha> "GPT data structures destroyed!"
14:31 < gotcha> always scary even when done on purpose
14:31 < jozefk> haha
14:33 < gotcha> thanks for the help guys
14:33 < gotcha> i guess this is why when i tried to clone another drive to this drive using Clonezilla it failed
14:34 < jbit> gotcha: make sure you reboot after zapping, not sure how the raid system reacts to things disappearing under it's feet :)
14:35 < gotcha> yeah, i rebooted and now GParted sees the full 6tb as unallocated
14:35 < gotcha> well, 5.46tb
14:36 < ALowther_> My Ubuntu OS won't boot. It keeps booting into grub prompt. I link the appropriate files and 'boot', then it sends me into initramfs prompt. I think from reading around the issue has something to do with a repartition I did last week where I took some of the space from the end of my main Linux FS partition....I have been trying to follow guides to fix it, such as 'resize2fs /dev/sda2', but that command tells me to run 'e2fsck -f /
14:37 < ALowther_> But that command forces me to abort because superblock reports a different size than the physical size.
14:37 < mawk> boot into the system, change the UUIDs in grub, in fstab, in the initramfs
14:38 < mawk> first is done with grub-mkconfig -o /boot/grub/grub.cfg, second is done in /etc/fstab, third is distro-dependant but in ubuntu I guess that you just rebuild the initramfs and it automatically detects the uuid, so you run update-initramfs -k all
14:38 < section1> i think that resizing the partition don't change the uuid.
14:38 < mawk> no, but the problem looks like an uuid that has changed
14:39 < ALowther_> I am on the same computer right now. I've booted into a Kali Live instance from a USB drive. Am I able to make these modifications from the terminal? Or do I need to boot into the grub prompt?
14:41 < section1> ALowther_, yes you can...check how to moutn your main linux part.. and chroot'ing to that mount.
14:43 < ALowther_> My fstab doesn't have much information in it. Is it supposed to?
14:44 < mawk> you're looking at the kali fstab ALowther_ ? or you've chrooted into your main system
14:44 < ALowther_> Yeah, whoops. Thanks. I'll look into how to do that now
14:47 < mawk> you mount the main system in some directory, mount the boot partition into the $system_mount_dir/boot directory, then you can start checking stuff
14:48 < ALowther_> The issue I am having is with /dev/sda2
14:48 < mawk> it's asking for a fsck right ?
14:48 < mawk> try to do it then
14:48 < mawk> if you have backups and all
14:49 < EvilRoey> <mawk> it's asking for a fsck right ? <-- all. the. fsckin'. time.
14:49  * EvilRoey melts back into the wallpaper
14:49 < mawk> lol
14:49 < EvilRoey> :)
14:49 < GunqqerFriithian> instead of swearing just tell someone to go repiar their file system
14:49 < GunqqerFriithian> repair*
14:49 < ALowther_> mawk: I tried to run 'resize2fs' and it keeps telling me to run 'e2fsck', yes.
14:50 < mawk> then run it
14:50 < ALowther_> I did
14:50 < mawk> if you have backups of your data
14:50 < mawk> ah
14:50 < mawk> what did it say ?
14:50 < ALowther_> e2fsck 1.43.7 (16-Oct-2017) The filesystem size (according to the superblock) is 120934656 blocks The physical size of the device is 109731953 blocks Either the superblock or the partition table is likely to be corrupt! Abort<y>?
14:51 < section1> do you use gparted for resizing ALowther_  ?
14:52 < ALowther_> I used parted from the command line, yes.
14:52 < nothos> ALowther_ Are you running e2fsck with the -f flag?
14:52 < ALowther_> But it's more of a "I used", the way you phrased the question would imply that I frequently do such things. :p. I am learning and don't know a lot about the specifics for partitions and partition tables, etc.
14:53 < ALowther_> nothos: Yes, the exact command I ran was 'e2fsck -f /dev/sda2'
14:53 < c06> can i ask libvirt (vm) related questions here
14:54 < c06> i enabled network boot for my libvirt vm i am unable to take the remote conole
14:54 < hackinghorn> Linus Torvalds doesn't use Ubuntu or Debian
14:54 < nothos> c06 don't ask to ask, just ask :P
14:54 < hackinghorn> should I use them
14:55 < section1> ALowther_, do you remmember what commands do you sue to resize the partition ?
14:55 < nothos> hackinghorn If they're going to fulfil your needs best use them
14:55 < section1> use*
14:55 < nothos> Just because Torvalds doesn't use them doesn't mean they're bad distros
14:55 < nothos> c06 what hypervisor are you using?
14:55 < c06> qemu-kvm
14:56 < c06> nothos: i am able to deploy vm and i installed the os
14:56 < ALowther_> section1: parted resizepart
14:56 < c06> nothos: using ssh i am able to login to the vm. i am making network changes so network is getting disconnected so i am trying for virsh console
14:59 < nothos> c06 Do you have the serial devices added to your VM?
15:00 < hackinghorn> nothos, alright, thanksss
15:00 < c06> nothos: yes, here u can find my xml file  <domain type='kvm' id='39'>   <name>cmp001.vcee.local</name>   <uuid>33109eb7-5ae9-4af4-76ce-ffd31df1893f</uuid>   <memory unit='KiB'>42360832</memory>   <currentMemory unit='KiB'>42360128</currentMemory>   <vcpu placement='static'>12</vcpu>   <resource>     <partition>/machine</partition>   </resource>   <os>     <type arch='x86_64' machine='pc-i440fx-xenial'>hvm</type>   <!--  <boot dev='network
15:01 < c06> sorry
15:01 < nothos> c06 please use a pastebin :D
15:01 < c06> nothos: https://paste.linux.community/view/e5e325ed
15:01 < c06> i sent the older paste.. :(
15:02 < jim> [[pastebin]]
15:02 < jim> hmm, it didn't start
15:03 < jim> [[pastebin]]
15:03 < section1> c06, try the vnc port
15:03 < jim> still didn't work :/ sorry
15:04 < jim> oh I remember, I renamed it
15:04 < jim> if you have nc installed, you can pastebin the output of an arbitrary command, for example ls -CF if you run it like this: ls -CF | nc termbin.com 9999
15:04 < nothos> c06 Try installing qemu-ga
15:04 < nothos> https://wiki.libvirt.org/page/Qemu_guest_agent
15:05 < nothos> in its xml and the guest itself
15:08 < ALowther_> I have absolutely no idea what I just did, but it changed the partition size to, I believe, what it used to be. Time to reboot and test! Thanks for all thoughts and help.
15:09 < c06> section1: i need to try after this
15:09 < c06> section1: but other vms are running fine only this network boot vm making issue
15:10 < section1> c06, qemu/kvm make a vnc port for you...so you can try something vncviewer 127.0.0.1 5902
15:10 < section1> c06, from your  config <graphics type='vnc' port='5902' autoport='yes' listen='127.0.0.1'>
15:13 < c06> i cant take console all the times its running on ubuntu server some remote machine.. :(
15:14 < c06> section1: so i am trying to fix this one
15:14 < pankaj_> What are the main advantages of sed and awk. WHere it is most used?
15:15 < pankaj_> I have used them while building linux from scratch. Is it worth learning and why?
15:16  * Dan39 yawns
15:17 < rumpel> pankaj_, it's maybe enough to know what they can be used for and where to look up examples.
15:17 < ananke> pankaj_: they're often used in shell scripting. 'worth' is very much subjective
15:23 < jnor> so, what is diff on intel_iommu and just iommu kernel parameter?
15:26 < jnor> have to bounce, nvm question
15:31 < bipul> Hi
15:35 < CrazyTux> which one is more recommendable for a linux newbie Linux Mint Cinnamon, Kubuntu or OpenSuse/GeckoLinux KDE?
15:35 < dgurney> how many times do we need to give distro recommendations to you?!
15:35 < dgurney> that said, Linux Mint is quite newbie-friendly
15:35 < CrazyTux> dgurney, not for me.
15:36 < CrazyTux> I have been asked to install a linux OS on some computers.
15:36 < no26> Hello the community. I created a module for my Linux who exchange data via the parallel port with my own card. When I use the physical parallel port I have no problem but when I use a parallel port emulated with PCI board, it does the job but got a complet Linux freeze after 5-10min. I debug my code and see any infinite loop or something. I research some tools or indication of "how can I debug Linux when Linux freeze". Any idea ? Than
15:36 < no26> ( also sorry for my bad english, it's not my native )
15:37 < CrazyTux> dgurney, btw, what's the difference between KDE Neon and Kubuntu?
15:37 < dgurney> KDE Neon is basically Ubuntu LTS with the latest, rolling KDE components
15:38 < CrazyTux> dgurney, ok. Would you not prefer Kubuntu over Linux Mint?
15:39 < Armand> I prefer Mint over... well, anything else.
15:39 < Armand> But, subjective.
15:39 < CrazyTux> Armand, could you state the reasons?
15:39 < dgurney> CrazyTux: I'm indifferent
15:39 < Armand> CrazyTux: Yes.
15:40 < CrazyTux> Armand, what are they?
15:40 < Armand> Numerous and belligerent.
15:40  * Armand coughs loudly
15:41 < Armand> I mean.... It's like a comfortable pair of shoes. It feels familiar and I've never had any issues with it.
15:41 < CrazyTux> Armand, ok.
15:41 < CrazyTux> Armand, I'm waiting for Mint 19 to be released.
15:41 < Armand> It's already on pre, no ?
15:41 < CrazyTux> Armand, which DE on Mint you prefer?
15:41 < Armand> Cinnamon
15:42 < CrazyTux> ok
15:42 < Armand> Ahh, beta
15:42 < Armand> :)
15:42 < Armand> I wonder if I should try it out on the Via Eden box..
15:42 < CrazyTux> I have been asked to install a user friendly linux distro on some computers for daily use. I think I will consider your suggestion.
15:43 < Armand> I love Mint 17/18
15:43 < CrazyTux> ok.
15:44 < Armand> Also... Junos is EFFING STRESSFUL!
15:44 < CrazyTux> what is that Junos?
15:45 < CrazyTux> this is interesting. https://distrochooser.de/en
15:46 < Armand> CrazyTux: Juniper SRX OS
15:46 < CrazyTux> Armand, ok
15:49 < CrazyTux> https://distrochooser.de/en.  I got Kubuntu as a recommendation here.
15:49 < CrazyTux> followed by Mageia and then Mint.
16:00 < CrazyTux> Armand, can we install Xfce on Linux Mint Cinnamon without creating any inconsistencies?
16:03 < c06> section1: thank you i took vnc something messed up in vm
16:16 < Armand> CrazyTux: I wouldn't know.. never tried
16:16 < Armand> Damnit
16:48 < searedvandal> if you want xfce on Linux Mint, just grab the xfce iso
16:49 < gotcha> when clonezilla says its copying data at the "Rate of 7.68GB/min", this is the speed its copying the data at?
16:50 < Thuryn> yes
16:51 < gotcha> is this fast or slow?
16:53 < Thuryn> well let's see
16:53 < Thuryn> 7.68 GB/min * 8 = 61.44 Gb/min
16:53 < Thuryn> divide by 60 to get Gb/sec... you're running at just over 1 Gb/s.
16:55 < Thuryn> whether that's "good" or not depends on what sort of hardware you have.  if you're running over a 1 Gb Ethernet network, you're pretty much going wire speed.  pretty good.
17:02 < GunqqerFriithian> what OS flashing software do you guys use, I use etcher but are there better ones to use?
17:04 < jelly> /bin/cp
17:04 < Siva_Machina> https://www.gog.com/game/ziggurat
17:04 < Siva_Machina> Aperantly that is being given away
17:05 < pengui> sed -i 's@(?!^host.*)(ident)@md5@' pg_hba.conf
17:05 < pengui> anyone know why that doesnt work?
17:06 < theos> hi
17:07 < theos> does anyone have a link to opera 32bit? i cant find it online
17:09 < copart> pengui: cannot have ^ in middle of regex unless inside []
17:09 < Fulgen> The getdents() syscall "returns" (?) EIO, how can I find out which directory entry caused this error?
17:10 < pengui> copart: how could i make that work ?
17:10 < Siva_Machina> theos: Opera doesn't do 32bit Linux anymore
17:11 < JimBuntu> From what I am reading, opera is now based on Chromium and no longer supports 32bit
17:12 < JimBuntu> If you really want it... ftp://ftp.opera.com/pub/opera/desktop/45.0.2552.898/linux/
17:12 < JimBuntu> theos, ^^
17:32 < emx> Any recommendations for applications to create activity diagrams (UML)?
17:38 < inthl> assuming I have 100GB of RAM and tons of I/O operations on local disks, with applications running that consume say 10GB of RAM, which means 90GB are free. usually, when the host is running for a while, the kernel would cache the opened files in RAM, so that sooner or later, those 90GB would serve as buff/cached, is that true?
17:39 < FreeFull> inthl: Yeah, the file cache will fill up with stuff as programs access the disks
17:40 < inthl> well then. what is a possible reason that this is simply not happening?
17:40 < inthl> on a real host I have 64GB or RAM, about 10 are used by apps, I have just a few GB used as buff/cache, most of the RAM, about 40GB+ is totally unassigned aka free
17:41 < emx> inthl, open more/bigger files?
17:41 < Dagmar> You've got to actually read the disk before it's going to be cached
17:42 < Dagmar> It doesn't just go and start reading the disks ahead of time for thatt
17:42 < inthl> this host opens and closes several hundred files per second, ram never gets used up to 100%, where I mean 10GB apps and the rest caches
17:42 < Dagmar> That doesn't change anything much
17:42 < Dagmar> Ooo several hundred
17:42 < inthl> well I thought that `file descriptors`, not file contents are cached as well..?
17:42 < Dagmar> Let us know when you get into the 20k range
17:43 < Dagmar> ...and it still won't change the fact that only the parts of a disk that have been read or written to will be in filesystem cache.
17:43 < Dagmar> Just opening and closing a filehandle won't do it
17:44 < Dagmar> If you want something in cache, use find and some something stupid like cat to /dev/null to force the files to be read in their entirety
17:45 < inthl> Dagmar, so file descriptors are not being held in RAM?
17:45 < Dagmar> Try again when you're asking a non silly question
17:45 < inthl> I am serious about this question. say you have 100k files in a direcotry
17:45 < Dagmar> Doesn't matter
17:46 < Dagmar> NOt even in the slightest
17:46 < inthl> now a listing on an average disk takes say 5 seconds
17:46 < Dagmar> The SOLE thing that matters is whether or not the kernel has a reason to have those pieces of the filesystem in memory
17:46 < Dagmar> That only happens two ways
17:46 < Dagmar> 1. You write something to the filesystem.
17:46 < Dagmar> 2. You read something from the filesystem.
17:47 < jim> figure out what he doesn't know and help him
17:47 < inthl> so nothing to cache there when I just do ls directorywith100kfiles ?
17:47 < Dagmar> When one of those two things happens, whatever was read or written will somewhat obviously _remain_ in memory after the buffered data has been written
17:47 < Dagmar> _whatever was read or written_
17:47 < Dagmar> So, looking at a directory gets the filenames in cache, NOT THE FILE CONTENTS
17:48 < Dan39> well lets do a quick google search
17:48 < Dan39> https://unix.stackexchange.com/questions/76746/how-to-configure-linux-to-cache-file-metadata-in-preference-to-contents
17:48 < Dan39> bam
17:48 < Dan39> kthnxbye
17:48 < inthl> Dagmar, that's exactly what I meant, the metadata
17:49 < Dan39> Controls the tendency of the kernel to reclaim the memory which is used for
17:49 < inthl> I did not mean to talk about the file contents
17:49 < Dan39> caching of directory and inode objects
17:49 < ksk> if you have spinning disks its kind of normal for a ls to be slow in a directory with many files.
17:49 < Dagmar> 100k files is so far from even possibly consuming 90Gb of RAM it's not even funny
17:49 < Dagmar> ksk: Spinning disk has nothing to do with the fact that everyone has ls doing a bloody alpha sort
17:50 < GunqqerFriithian> `du -d 1` best way to list in directory
17:50 < jim> remember too, that ls sorts the filenames, so it's gonna take some time, even to start (because the printing of the filenames doesn't happen till the sort is finished)
17:50 < ksk> Dagmar: I dont really understand, you try to tell me its rather slow because of ls, not because of the disks? (I have no machine here right now with said "configuration" ;))
17:50 < inthl> I have been playing around with vfs_cache_pressure as well, setting this to low values did not reveal noticeable changes, where as again the question is whether I was able to measure them anyway
17:50 < Dagmar> The data can be on a SAN with dual-redundant 10Gb fiber channel backplanes and a ridiculous amount of cache RAM in the cabinets and 100k files in one directory is still going to appear to hang the prompt for well over a minute while ls does an alpha sort of the filenames if you're not careful
17:51 < ksk> okay, thanks for sharing Dagmar.
17:51 < Dagmar> This I have seen in person and even bounced the support tickets
17:51 < Dan39> well there you have it, the damn sorting slows it down :o
17:51 < ksk> my assumption so far was "it takes time", now I even know why ;)
17:51 < Dagmar> People being foolish and attempting to use the filesystem like a high-volume database isn't something new
17:52 < Dan39> try ls -U ?
17:52 < Dagmar> It's just something we're not allowed to slap sense into people over
17:52 < Dagmar> No sorting and ls will start spewing filenames immediately
17:52 < ReScO> hey folks
17:52 < jim> hi
17:52 < ReScO> What's the best FS for an NVMe SSD? EXT4?
17:53 < inthl> well I wonder just about the fact why I have so much free ram. I was not sure about this since I am talking about a new host with very latest linux stable, 4.16 or so, I do recall that on the older host (I do not have those in operation any more to compare) which had 32G of ram, the RAM was always full - the remains where taken by the buff/caches, this is not the case right now any more
17:53 < idodeisuke> Anyone knows a Telegram client for Linux, that has the feature to group chats into categories?
17:53 < ksk> ReScO: that question is kind of provoking a filesystem-war. ext4 is reasonable.
17:53 < Dagmar> inthl; If you want to watch it get eaten up, do what I said.  Pick a directory with at least 90gb of data in it, use find and cat to actually force the system to read the contents of those files
17:53 < ksk> inthl: uptime of that system is high enough?
17:53 < Dagmar> RAM will be all gone
17:54 < ReScO> ksk: i've been looking at some benchmarks but they're benchmarks, y'know?
17:54 < jim> ReScO, what things about the ssd are you considering in your choice of fs?
17:54 < paulcarroty> phoronix has a tonf of benches
17:54 < paulcarroty> *tons
17:54 < inthl> Dagmar, with cat, yes, I was referring to my belief that the metadata was stored in ram
17:54 < ksk> inthl: I think it would be safe to assume the linux guys got this caching thingy right and keep doing.
17:54 < Dagmar> There is very little metadata
17:54 < inthl> uptime is about 20 days
17:55 < ReScO> paulcarroty: already looked at them
17:55 < ksk> like, many people fiddle around with disk caching, its seldomly needed though.
17:55 < ReScO> XFS seems to be a strong contender
17:55 < Dagmar> ReScO: You need to define a use case before you can make those kinds of determination
17:55 < Dagmar> Absent a use case, there's f2fs
17:55 < ksk> ReScO: xfs is reasonable, too. Id argue if you are not netflix/doing something crazy just go for the FS you are most familiar with
17:56 < jim> ReScO, what has you thinking about the ssd as something that needs to be handled in some special way by a filesystem?
17:58 < inthl> Dagmar, okay, then one theory question: there is buffers with 10GB of size, 40GB as mentioned are free. should those 10GB not grow until all additional ram is used up anyway? so that those read files (cat directorywithlotsofGBcontents) remain in RAM because they may possibly used in the near future?
17:59 < Dagmar> Generally they do remain in ram, except programmers like to call sync()
17:59 < inthl> ah
17:59 < Dagmar> This doesn't evict anything from RAM
17:59 < inthl> so what does sync do then? afaik it does not free the contents in RAM, does it?
17:59 < theos> thanks JimBuntu
18:00 < Dagmar> It just makes them valid for discarding the moment something that matters needs some free RAM
18:00 < inthl> good - then why is my cache usage increasing slightly and decreasing slighttly all the time, spinning around a few GB only
18:00 < Dagmar> ...because it explicitly forces the caches to be written to disk, meaning the data doesn't need to be cached anymore
18:00 < inthl> that's my point - the discarding happens before the ram is even filled up
18:00 < oerheks> JimBuntu
18:01 < Dagmar> There's lots of reasons this could happen, almost all are things a fully-aware admin would know about
18:02 < Dagmar> The hypervisor telling the VM to free up the RAM being a huge one that's probably quite appropriate considering the conspicuously-not-a-power-of-two volume of RAM that keeps being cited
18:03 < Dagmar> It's one of those things that ESX loves to do and the reason people have to keep reinserting those darn kernel modules
18:03 < mkjl> I have an issue, I installed Xanmod 4.14.44-1 kernel and after booting it when I move horizontally on a trackpad cursor moves vertically, when I move vertically it doesn't move at all, Cinnamon 3.8, macBook Pro 15" Core i7 2.66 GHz mid-2010, can anyone help?
18:06 < Dagmar> Reinstall the kernel that ships with Cinnamon and your problem will probably disappear
18:08 < Yogui> hi there, i have a question about file descriptors in Bash. Say i have the line "TMPFILE=$(mktemp /tmp/XXXXXX.log) && exec 3>$TMPFILE && rm $TMPFILE" and after that, I redirect the standard output from a program to that same filename; what happens to the file descriptor then, is it lost?
18:09 < mawk> no Yogui
18:09 < mawk> it still points to a file on the filesystem
18:09 < mawk> it's just an anonymous inode
18:10 < Yogui> ok cool, then the file is deleted as soon as the script ends
18:10 < mawk> as soon as the last file descriptor is closed yes
18:10 < theos> hmm so chrome doesnt support 32bit either
18:10 < Dagmar> Yes.
18:11 < Yogui> that's perfect
18:11 < Dagmar> It's a "sneaky trick" commonly used by unruly users and people running things they shouldn't on a common-use machine
18:11 < mkjl> Dagmar: but how to get this one fixed?
18:11 < Yogui> now say i automate a script like this, and every one of them opens the 3rd file descriptor: are they independent from each other?
18:11 < mawk> if you're in C you can do the same thing without having to create a file first Yogui
18:11 < Dagmar> mkjl: Start with the kernel Cinnamon actually ships with
18:11 < mawk> you use open() with the O_TMPFILE flag
18:12 < Yogui> mawk nah it's just bash, not c
18:12 < mawk> the file will never show up in the directory listing, even for a fraction of a second
18:12 < mawk> yeah I know
18:12 < mawk> it's just for further reference
18:12 < Dagmar> mkjl: Problems you have with some third-party kernel are things you should take up with whoever built that kernel
18:13 < JackPeralta99> I am getting segmentation fault message when I try to do apt-get update
18:13 < JackPeralta99> Any suggestions?
18:13 < Yogui> so, can "exec 3>newTmpFile" be run simultaneously by several scripts? assuming all the temp filenames are different
18:13 < Dagmar> Yogui; Assuming they're all launched while the file still exists
18:13 < Yogui> they each create their own tmp file yes
18:13 < mkjl> Dagmar: yeah but are you aware of any Xanmod channel?
18:14 < Dagmar> The rule is simple... Resources on a disk are never freed, even when "deleted", until after there are no more file descriptors open to them
18:14 < Yogui> i'm trying to understand if "3" is shared between scripts
18:14 < Dagmar> mkjl: No.
18:14 < searedvandal> mkjl, don't think xanmod has any irc channel. try their forums maybe?
18:14 < JackPeralta99> I am getting segmentation fault message when I try to do apt-get update; distro is LinuxLite 4.0
18:14 < Dagmar> mkjl: Also I'll mention you're clearly still taking a rather mystical approach towards software management
18:15 < Dagmar> It's ricer tech of dubious merit
18:15 < ksk> JackPeralta99: Is it a fresh install? ask linuxlite. if not, what did you do?
18:16 < JackPeralta99> ksk, I am not aware of IRC for LL.
18:16 < ksk> still, if they ship broken images, how(or why) should we try to fix that?
18:17 < JackPeralta99> I am sure I've goofed something up, after install all was OK.
18:17 < mkjl> Dagmar: default kernel will work, I don't understand what you mean
18:17 < Dagmar> So undo the last few things you used sudo to "fix"
18:17 < Dagmar> mkjl: You've installed a third-party kernel of dubious merit, and it broke things
18:18 < Dagmar> How is this confusing
18:19 < pxfgod> Q: Is it possible that a non-blocking syscall is interrupted by signal and returns with EINTR, WHY or WHY NOT?
18:19 < Psi-Jack> A: You do not need to prefix a question with Q:.
18:19 < Dagmar> It's possible.  Because it's non-blocking.
18:19 < searedvandal> JackPeralta99, a full paste of your command and output would probably give people a bit more to go on and have a better chance to help you
18:20 < bipul> pxfgod, Q. What do you mean "non-blocking syscall" ?
18:20 < searedvandal> use the pastebin in topic
18:20 < GunqqerFriithian> You don't have to preface with Q, but personally I like it, makes it easier to differentiate between stuff
18:21 < bipul> Yes, Agree.
18:21 < pxfgod> bipul, e.g. , recv on a non-blocking socket.
18:21 < searedvandal> JackPeralta99, and have you checked the Linux Lite forums?
18:22 < bipul> And put "Ans" previous To give the solution.
18:24 < Psi-Jack> The ? isn't enough to distinguish a question for you, GunqqerFriithian ? :p
18:25 < GunqqerFriithian> well we're speaking in english so we don't have something denoting a question at the begninning
18:25 < bipul> Psi-Jack, Yes that would be awesome to represent symbolically a question.
18:27 < GunqqerFriithian> Q. Is there any way to sync music from my computer running ubuntu to my iDevice?
18:27 < Psi-Jack> A: Prefixing questions with Q doesn't help much. :p
18:27 < GunqqerFriithian> but.... that isn't an answer
18:27 < Dagmar> WHy not warn people ahead of time you're about to ask a question
18:28 < Dagmar> That way people will know that they should pay attention, rather than just sit there and read the screen
18:28 < GunqqerFriithian> it helps denote if something is general chatter or someone wanting help
18:28 < Dagmar> ...oh and then you could warn people you're about to warn people that you're going to ask a question.
18:28 < Psi-Jack> This is already a Linux Help & Discussion channel. Questions are already expected. heh
18:28 < Dagmar> Precisely.
18:29 < GunqqerFriithian> but a lot of discussion also happens
18:29 < pxfgod> a prefix Q does no harm, isn't it???
18:29 < Dagmar> Yeah, screw that.  People should only be required to read specific answers to their own questions without being asked to wade through explanations
18:30 < braindead_> hey peeps
18:31 < GunqqerFriithian> the prefix doesn't hurt you in any way, you don't have to use it nor have to preface what you answer with A.
18:31 < braindead_> linux noob here. is it recommend to store backups to /var/lib/psa/dumps or what is best practice?
18:32 < GunqqerFriithian> backups? An external drive
18:32 < Dagmar> An endless rabbit hole of pedantics
18:33 < braindead_> of course you are right GunqqerFriithian but to store them in the first place
18:33 < Dagmar> Anywhere that can't run out of space
18:33 < Yogui> hey i have another fd-related question; say i open a file descriptor on a temporary file, call rm on it, and then i call mv on that temp file; what happens to the file at the end, is it present at the other location?
18:34 < Dagmar> No.  You'd get an error from the mv
18:34 < mkjl> Dagmar: it was recommended to me as a good one, many good stuff arrives first as 3rd party
18:34 < braindead_> and is dump a folder or some kind of special file?
18:34 < Yogui> huh ok
18:34 < Dagmar> You don't get to open new fds to something after you've destroyed it's directory entry
18:34 < Dagmar> mkjl: You're going to need a really large bucket to hold all the lack of concern I have about that
18:35 < Dagmar> mkjl: ...not to mention the complete disbelief.
18:35 < Yogui> i mean consider this line: "TMPFILE=$(mktemp /tmp/XXXXXX.log) && exec 3>$TMPFILE && rm $TMPFILE && mv $TMPFILE /root/final.log"
18:35 < Psi-Jack> braindead_: Backups are generally not kept on the same host as the thing being backed up.\
18:35 < Dagmar> mkjl: You literally linked an article written by someone who apparently doesn't have a clue what the features he's touting _actually do_
18:36 < Dagmar> mkjl: There's a phrase that describes this, and that phrase is "Endless September" and has been a thing for so long it was likely named _before you were born_
18:36 < braindead_> yes Psi-Jack. my concern right now is, that i cant cd into /dump and i wonder if it because of user rights of that user or because that isnt really a folder but more some kind of container
18:37 < Psi-Jack> braindead_: So.... Ask what you really need help with? heh
18:37 < Yogui> yeah; it errored out
18:37 < Dagmar> mkjl: It should be a great big warning sign that a third-party kernel can't manage to maintain feature parity with the mainstream kernel that has all of it's source readily available
18:37 < GunqqerFriithian> but yeah if im running ubuntu and want to sync music with an iDevice how would I do so
18:38 < Psi-Jack> GunqqerFriithian: You don't.
18:38 < Psi-Jack> GunqqerFriithian: There is no iTunes for Linux.
18:38 < GunqqerFriithian> no one's made an alternative?
18:38 < Psi-Jack> No.
18:38 < Dagmar> Apple is very stingy with the specs
18:38 < GunqqerFriithian> That's surprising
18:39 < GunqqerFriithian> wonder what $SearchEngine has to say about it
18:39 < Dagmar> YOu can try getting iTunes to run with WINE, but otherwise, just shove your music on a microSD card
18:40 < GunqqerFriithian> looks like im better off googling my questions :P https://www.groovypost.com/howto/howto/sync-your-iphone-or-ipod-touch-in-ubuntu/
18:40 < Psi-Jack> There is ifuse, which is the closest thing, however, there are limits to what it can do and it's not pretty.
18:40 < Psi-Jack> iTunes requires USB, wine itself doesn't quiiiiite yet support USB, though that's "in devel"
18:41 < bipul> Q. How to distinguish cold backup and hot backup?
18:41 < akd> Hi
18:42 < ananke> bipul: try using google
18:42 < akd> I cant write proper accent with US International Layout
18:42 < Dagmar> Just reach out and touch them
18:42 < akd> ´el`eve is not working
18:42 < braindead_> or hover your hand above
18:42 < Dagmar> ...or stick 'em in the microwave for about a minute on 50% power
18:42 < akd> How do I change my ctrl to alt-gr ?
18:43 < mkjl> Dagmar: no man, I didn't quote or link anything, I think it's too late for you
18:43 < GunqqerFriithian> bipul: I don't think there really is a way to figure out if files are from a cold or hot backup as it's just based on how the backups were taken
18:43 < Dagmar> mkjl: Doesn't change the fact that it's a ricer kernel of dubious utiliity
18:43 < Yogui> without mv, is there a better wah to duplicate a file than cp + touch + chown + chmod?
18:43 < bipul> ananke, Yes, might get the answer there, but i would love to discuss here.
18:43 < ananke> bipul: you provided absolutely no context
18:43 < Dagmar> Why would you ever follow cp with touch?
18:44 < searedvandal> hot or cold, plugged in or not
18:44 < Yogui> with --reference
18:44 < bipul> GunqqerFriithian, What you mean "How the backups were taken" ?
18:44 < ananke> bipul: what's the context of that question?
18:44 < Psi-Jack> Yogui: Why do you need touch, chown, chmod?
18:44 < Yogui> Psi-Jack to make sure all attributes are the same as a reference file
18:45 < Psi-Jack> Just cp.
18:45 < GunqqerFriithian> well a cold backup requires all users to be logged out and the system to not be active, while hot backups (I think) can be done at any time, so the only different is how they were taken
18:45 < Psi-Jack> Yogui: man cp, for further details.
18:45 < GunqqerFriithian> cold backups are done for safety as no data can be changed while the backup is in progress
18:45 < GunqqerFriithian> but they do require downtime
18:45 < Yogui> huh why didn't i think of that
18:45 < Yogui> thanks
18:46 < Dagmar> wat
18:46 < ALowther_> Am I on the right track of thinking that all of these different Linux distributions are based on the same Linux kernel?..The main difference is in the packages those distributions choose to ship with the kernel as part of the core OS?
18:46 < ananke> ALowther_: that's partially correct. many distros add their own kernel patches
18:46 < Dagmar> ALowther_: Yes
18:47 < bipul> ananke, I mean here to know a real time example when cold backup and hot backups were taken.
18:47 < ananke> bipul: unfortunately your poor grammar results in a nonsesical question
18:48 < ALowther_> ananke: Okay, but I assume those patches are also applied to the main Linux kernel, the distro just might not be up-to-date with the very latest kernel release because they are making sure everything is testing and compatible?
18:48 < Dagmar> bipul: The thing that should be your major concern is that someone slipped up and allowed Oracle "admins" to contact you in the first place
18:48 < ananke> ALowther_: no, many of those patches are NOT applied to the main linux kernel.
18:49 < Dagmar> Generlaly it's only Oracle DBAs who use those terms with respect to backup data
18:49 < bipul> Dagmar, I don't know :D But happy to work for Oracle.
18:49 < Dagmar> A sign that they've been allowed to run free for too long
18:49 < ALowther_> ananke: Am I interpreting the use of the word 'patches' correctly in thinking it is put in place to eliminate a bug(s)?..If so, why wouldn't the main kernel also seek to eliminate that bug?
18:49 < rud0lf> bipul: the one from Matrix?
18:49 < bipul> What is VSS and how it take backup?
18:49 < Dagmar> ALowther_: They do
18:50 < Dagmar> bipul: Because we have better options
18:50 < pankaj_> I have a dual boot system. Sometimes I want to have to work with another OS but I do not want to reboot again. Is it possible to do something like chroot so that use it as a virtual machine? I have used 'systemd-nspawn' but it does not work as expected.
18:50 < ananke> ALowther_: not all patches 'eliminate bugs'. they alter things: sometimes they add things, sometimes they change things, sometimes they remove things. not all of them are accepted by the main kernel.
18:51 < ananke> ALowther_: sometimes they backport bug fixes or enhancements
18:51 < mnemon> there's also the configuration of the kernel which might differ even if the code is the same.
18:51 < bipul> pankaj_, Yes try it out. But don't install windows in chroot.
18:52 < bipul> :D
18:52 < Dagmar> You can, in theory, tell virtd to present your actual disk to a VM as a virtual disk
18:53 < Dagmar> This is generally not recommended
18:53 < ALowther_> Okay, I think I am on board. But generally, it's about: configuration settings, default settings, fs structure & default packages that is separating these distros, while the kernel itself is the same, or very similiar with some custom tweaks?
18:54 < Dagmar> So, everything is different.  Yes.
18:54 < Dagmar> No Linux kernel generally means it's called BSD or Hurd or Windows or something else.
18:55 < mnemon> ALowther_: the distros choose some kernel version, compile it with their own configuration and sometimes apply custom patches to tweak things/add features(often backporting for the ones that use older kernel lines)
18:55 < morf> nice
18:55 < morf> what's going on
18:57 < mnemon> ALowther_: but yes, the results are generally somewhat similiar across most distros(on the same kernel version) in comparison to what it's possible to compile yourself.
18:57 < ALowther_> Dagmar, mnemon: Cool, thank you.
18:58 < hans_> hrm, apparently screen's "-X stuff" command is vulnerable to timing.. when doing screen -S foo -dm ; screen -S foo -X stuff "echo test";  the 2nd command may be completely ignored on a busy system if you call them too fast, but seems to be safe if you add a sleep 1 in between
18:59 < hans_> ofc, sleep 1 isn't an ideal solution but..
19:00 < hans_> it randomly broke 1 of my scripts on a busy (nearly overloaded at the time of command execution) system
19:01 < hans_> but adding a sleep 1   seems to have fixed it
19:01  * bipul thinks Dagmar has kung fu linux skill set
19:05 < pankaj_> bipul: Sorry for late response. But I asked how to work with 'systemd-nspawn' when I start it with the partition mounted and run some commands like installing something it refers to the current system I am using. As you know it is not what I want. So, how to deal with it.
19:09 < pankaj_> Dagmar: I came to know that you have kung fu linux skill set.
19:11 < GunqqerFriithian> Q what was the command that shows how long the system took to boot and what process was taking what time
19:14 < pankaj_> Guest48888: I forgot. Tell me when you come to know
19:14 < bipul> GunqqerFriithian, Okay so you wants to know the command that shows you how long the system was up and running? And what about the process? will you be clear?
19:14 < GunqqerFriithian> not uptime, but how long it took to boot
19:14 < ksk> GunqqerFriithian: systemd-something iirc ;)
19:14 < searedvandal> GunqqerFriithian, systemd-analyze you're thinking of?
19:15 < GunqqerFriithian> yes that sounds right
19:15 < bipul> GunqqerFriithian, May be you should find in system log.
19:15 < GunqqerFriithian> `systemd-analyze blame` is what I was looking for
19:15 < ksk> bipul: your approach so 1990 :P
19:15 < GunqqerFriithian> shows what took how much time
19:15 < GunqqerFriithian> 17 sec boot time HNGGGGGG
19:15 < searedvandal> ah yeah, you wanna know who to blame
19:16 < bipul> ksk, Yes. That's why i said "May be" :)
19:16 < GunqqerFriithian> network manager wait online taking 6 sec smh
19:16 < pxfgod> GunqqerFriithian just wanna figure out which step take longest time to launch a kernel.
19:16 < GunqqerFriithian> before I got my SSD solid 1:30 boot time
19:17 < searedvandal> I blame man-db.service
19:18 < bipul> GunqqerFriithian, Do you able to find it out?
19:22 < GunqqerFriithian> bipul: yeah I was looking for the comman `systemd-analyze blame`
19:23 < hans_> you have to reboot the system for the changes to take effect: yes/no
19:23 < marko_> GunqqerFriithian: 'Why the 'blame' at the end of command?
19:24 < GunqqerFriithian> instead of just showing `17 seconds` and a bit more, it says which process took what time exactly
19:24 < marko_> GunqqerFriithian: Oh. Now I know when I used it. Very important command.
19:24 < GunqqerFriithian> yea
19:24 < bipul> GunqqerFriithian, And your about to search booting time? right?
19:25 < GunqqerFriithian> ?
19:25 < ALowther_> Everytime I login on my Ubuntu machine, I open a terminal to run 'sudo openvpn vpn.openvpn'.....What is the best way to automate this and allow it to run as a background process?
19:25 < marko_> GunqqerFriithian: Can all these services be handled by systemctl?
19:25 < hans_> ALowther_, you mean everytime you reboot and login, or literally every time you login, regardless of reboot?
19:25 < GunqqerFriithian> I have no idea
19:26 < ALowther_> hans_: I run it in the terminal, so anytime the terminal running the process is closed, I have to run it.
19:26 < SuperSeriousCat> Your network manager dont have an option to save and autologin, ALowther_?
19:26 < hans_> ALowther_, if you just need it to always run in the background, i'm leaning towards crontab -e
19:27 < ALowther_> SuperSeriousCat: No, it's a personal VPN.
19:27 < bipul> GunqqerFriithian, I see your intelligent.
19:27 < GunqqerFriithian> I hope that's sarcastic, as I'm... yeah I kinda stumble through
19:27 < marko_> GunqqerFriithian: But I ask are all the services listed via 'systemd-analyze blame' can be handled by systemctl?
19:27 < ALowther_> hans_: I
19:28 < GunqqerFriithian> I couldn't even remember the command :|
19:28 < ALowther_> I'll check that out
19:28 < bipul> No it's not.
19:28 < ALowther_> hans_: Thanks
19:29 < hans_> ALowther_, make a script that runs it, make sure to chmod +x, and tell chron run it at @reboot /bin/sh /path/to/script.sh
19:29 < bipul> marko_,  Do you use Archlinux?
19:29 < searedvandal> ALowther_, if you want it to start at system boot, systemd can handle it. https://wiki.archlinux.org/index.php/OpenVPN#systemd_service_configuration
19:29 < ALowther_> hans_: I have to run it with sudo, though. Will it still work?
19:30 < ALowther_> searedvandal: Cool, I'll look into that too.
19:30 < ALowther_> searedvandal: Thanks
19:30 < bls> ALowther_: you're way better off figuring out how to properly have your init or network management system handle this instead of hacking around it with cron and shell scripts
19:30 < SuperSeriousCat> Dont see why it is relevant where you connect to, ALowther_. VPN is VPN. Your network manager can handle auto connect
19:31 < marko_> bipul: Yes.
19:31 < ALowther_> SuperSeriousCat: Are you referring to the network manager on my system, or a human network manager providing these abilities with their software/config?
19:31 < SuperSeriousCat> On your system
19:32 < ALowther_> bls: Such as the suggestion above with systemd, right?
19:32 < bls> ALowther_: yes
19:33 < bipul> marko_, oh then you know advance linux.
19:34 < bls> ALowther_: that'll be way more resilient and won't required you to maintain something unique to just you that you can't google or ask for help with
19:34 < ALowther_> bls: I'll figure out how to get that going then, thank you.
19:40 < GunqqerFriithian> Startup finished in 12.768s (kernel) + 2.543s (userspace) = 15.311s
19:40 < GunqqerFriithian> HNNNGGGGG
19:41 < searedvandal> 15s is alright
19:41 < GunqqerFriithian> I uesd to have a solid 1 min 30
19:42 < bls> how often do you reboot that the time relative to the machine being on isn't just a statistical irrelevancy?
19:42 < GunqqerFriithian> I have a laptop and take it with me
19:42 < GunqqerFriithian> to save battery I turn off
19:46 < marko_> bipul: Do, you know some linux administrators in India. I can have a lot of help by contacting them and learning more about linux. Please reply.
19:47 < TECFALL> trying to tftp from a cisco switch (192.168.102.4) to my linux box (192.168.100.1). If i specify port 69 in my iptables rule, the tftp transfer fails, but if don't specify a port, the transfer is successful. Here are my rules: https://pastebin.com/6xBfaXsE
19:47 < TECFALL> This makes me believe that the tftp is using some other port other than 69
19:47 < bipul> marko_, join #archlinux-in,#ubuntu-in,##linux-india You will find them
19:48 < marko_> bipul: Sometimes I am too stupid. You are absolutely right. Why did not it popped into my head earlier.
19:51 < bls> TECFALL: the initial tftp request is on that port, it randomly switches to another after connection establishment
19:53 < bls> for certain meanings of "connection"
19:58 < marko_> What does that 'Type' field in 'fdisk -l PARTITION_NAME' signify. I always confuse it with the type of partiton like 'ntfs, ext4' or other partition but it is not the case. What is it?
20:01 < akd> I was able to use accent with my "English (US Alternative Layout", I was able to write "é" when pressing (ALT-GR + ") + e , now I am writing "'e". I want to restore my ALT-GR . I dont remember where is the kebyoard configuration. I am on Linux Debian Stretch With a Cinnamon desktop. Any Help ?/
20:02 < searedvandal> marko_, the "Disklabel type:" field?
20:04 < absolutejam> hey guys
20:04 < absolutejam> just wondering if somebody can steer me straight with a multi boot setup
20:04 < absolutejam> I have kubuntu installed in LVM1, Arch in LVM2
20:05 < absolutejam> Grub is sda2, ESP is sda1
20:05 < absolutejam> And my LVMs are within a LUKS container, the sdas aren't
20:05 < marko_> marself: Yes, May be this is the one that is at the last of all the 5th (I think 5th) column
20:06 < absolutejam> In Kubuntu, I've ran os-prober and updated my Grub config so it detects Arch at the boot menu.
20:07 < bls> marko_: old operating systems used to require a flag be set in the partition table so it could know what filesystem was going to be on the partition
20:07 < absolutejam> you know what
20:07 < absolutejam> explained all that and I think I got it
20:07 < absolutejam> thanks ;)
20:07 < bls> marko_: modern operating systems know how to read the FS superblocks/do FS fingerprinting to discover what the filesystem actually is
20:08 < absolutejam> But out of curiosity, does each partition have its own initramfs?
20:08 < fyrp> what is the use of file descriptors?
20:08 < absolutejam> or is that a feature of the bootloader?
20:09 < bls> fyrp: it's a unique ID for open files
20:09 < marko_> bls: Well, It is something hard to grasp and make sense (atleast for me).
20:09 < marko_> bls: For me the problem becomes more when it shows three of them in 'Type field'
20:10 < fyrp> bls: is the fd's are used in case of troubleshooting or analyzing what went wrong?
20:10 < fyrp> bls: could you tell the use case
20:10 < bls> fyrp: no, they're used by programs to read and write to files
20:10 < marko_> bls: Like I attached ntfs 1.4 TB ssd and in 'Type' field it said 'HPFS/NTFS/exFAT'. Why and how?
20:10 < fyrp> bls: programs like?
20:11 < bls> fyrp: programs like any program that opens, reads, and writes files
20:11 < redredhathat> would like some input on what lang i should write my upcoming irc bot in
20:11 < redredhathat> https://goo.gl/forms/dh52xCQWQAq4Z0of1 any input helps
20:11 < trafaret1> hi there
20:11 < redredhathat> if this isnt allowed please let me know
20:11 < fyrp> bls: vi or emacs?
20:11 < redredhathat> i already know some python but am looking for something new
20:12 < DolphinDream> howdy
20:12 < bls> marko_: because it's up to the OS to interpret what that type means, an so either different OSs take that flag and decide which FS it means, or it could mean any of the three if they all work in a similar enough way for the OS
20:12 < bls> redredhathat: this channel doesn't really do generic, open ended polls
20:13 < redredhathat> yeah, sorry
20:13 < redredhathat> just want some opinions
20:13 < bls> redredhathat: try #friendly-coders
20:13 < redredhathat> im thinking maybe D
20:13 < redredhathat> will do
20:13 < DolphinDream> if i added a user to a group A.. and have a folder having the group name set to the group A and the owner and the group permissions on that folder is rw-rw- for both the owner and the group. shouldn't anotehr user beloging to the same group have write access ?
20:14 < bls> DolphinDream: directories also require execute permissions
20:15 < DolphinDream> bls: the folders have it : drwxrwxr-x 12 someUser someGroup 4096 May  8 04:08 sandbox/
20:15 < DolphinDream> another user beloging to the someGroup seems to not have access to write (create a file ) in the sandbox folder
20:17 < bls> DolphinDream: how are you determining that the other user is in the group?
20:17 < DolphinDream> do i need to do some setafcl magic  to enable write access for the group ?
20:18 < DolphinDream> I created a group first.. then added a user to that group: sudo usermod -aG someGroup  someUser
20:18 < Daesh> did you log said user out/in?
20:18 < DolphinDream> bls: then groups someUser gives me the list of the groups.. which includes the someGroup group
20:18 < bls> DolphinDream: because the thing we see over and over in here is 1) user logs in 2) admin adds user to new group 3) admin changes group permissions on something 4) user can't access file because group permissions are applied at login/process creation time
20:19 < DolphinDream> do i need to log out/in to have the group changes take effect?
20:19 < DolphinDream> ahh.
20:19 < bls> DolphinDream: run the id command as the user
20:19 < bls> DolphinDream: if you don't see the group listed, yes
20:19 < DolphinDream> yeah
20:19 < DolphinDream> id doesn't list the gruop.. so i need to login/out
20:20 < DolphinDream> nice.. now it works
20:20 < DolphinDream> thx bls
20:24 < balletjebal> i'm getting a Warning: Pycurl is not compiled against Openssl while executing a script,does anyone know how to solve this ?
20:24 < djph> compile it against openssl
20:24 < djph> ?
20:25 < bls> complain to the person that wrote the script?
20:25 < balletjebal> djph, eh! well hows that im new to linux
20:26 < Daesh> balletjebal: how did you install pycurl?
20:26 < balletjebal> Daesh,  it came with parrot linux
20:26 < Daesh> go talk to the maintainers of parrot then
20:26 < bls> balletjebal: then ask the parrot linux people
20:27 < Daesh> them building it without linking to openssl is quite strange
20:27 < balletjebal> i've read that uninstalling and reinstalling pycurl might do it but it wont let me
20:28 < balletjebal> i'm just trying stuff out since im new to linux what does pycurl do ? is it like an python compiler ?
20:29 < bls> balletjebal: curl is a CLI tool for interacting with HTTP server, pycurl is python code to use that tool/its libraries
20:29 < Daesh> balletjebal: https://en.wikipedia.org/wiki/Curl
20:29 < Daesh> er.. wrong link
20:29 < phogg> balletjebal: talk to whoever packaged your pycurl
20:29 < Daesh> https://en.wikipedia.org/wiki/CURL
20:29 < Daesh> there, case sensetive
20:29 < phogg> balletjebal: this is also known as "Use a distribution that has better packages"
20:30 < bls> so this means a "security" oriented distro has packaged tools than can't do HTTPS
20:30 < balletjebal> phogg, well the script was written for kali backtrack that might be it, ill dig into it
20:31 < Daesh> just use a normal distro
20:31 < phogg> balletjebal: Why are you using Kali?
20:31 < revel> He's not, he's using Parrot.
20:31 < bls> yeah, if you don't know much about linux, running a pentesting distro as your daily driver doesn't make much sense
20:31 < phogg> bls: Are you at all surprised? A distribution built by amateur distro makers for *any* purpose invariably ends badly, even if the intended purpose somehow relates to security.
20:32 < balletjebal> bls, i figured it would be a broader way to look at linux :P and i was curious about pentesting in the 1st place
20:33 < bls> not at all surprised. unfortunately there's a style over substance element to a lot of these endeavors
20:33 < phogg> balletjebal: you don't need a "pentesting distribution" to learn about pentesting. It's just a convenience thing for professionals who, one hopes, know when NOT to use it.
20:34 < bls> balletjebal: understandable, but the pentesting distros are generally suited to a single purpose: booting them, running some pentesting tools, then moving on to something more suited to everyday tasks
20:35 < phogg> Think of it like a you would a liveCD used for data recovery. You boot it to run a specific task on a system, then you shut it down. Anything else is doing it wrong.
20:35 < ace510> be careful, you're gonna make the one person that uses Kali as a daily driver come harrangue you
20:35 < searedvandal> use a "regular" distro and read up on tools used in pentesting, install them yourself and learn about it that way.
20:35 < Daesh> phogg: s/professionals/skript kiddies/ in 99% of the cases
20:35 < phogg> Daesh: I cannot account for those guys.
20:36 < phogg> ace510: sadly it is *not* one person. We get them in here VERY frequently.
20:36 < djph> phogg: dumb kids who're finally allowed on the internet
20:36 < phogg> ace510: Mostly newbies who can't tell the difference between "Security distribution" and "Secure distribution" and think they're both safe AND cool.
20:36 < ace510> hey, i'm technically a dumb adult now
20:36 < searedvandal> everyone wants to be a hacker
20:37 < balletjebal> i think its because these distros are like sugarcandy, thats why i installed it in the 1st place,just to see what its like.
20:37 < phogg> balletjebal: What it's like is a trainwreck. When  you need it you'll know, if you're not sure just stay away.
20:38 < Psi-Jack> Do hackers get hatches?
20:38 < Psi-Jack> hachets*
20:38 < phogg> I want to know if they get hatches, too.
20:38 < revel> I have one. Does that make me a hacker? :D
20:39 < phogg> revel: do you carve lawn furniture from tree trunks using it?
20:39 < ace510> I hack, but only when I whack and smack as well
20:39 < revel> No, I just chop firewood, sadly.
20:39 < revel> I guess I'm a chopper then.
20:39 < phogg> revel: then no, not a hacker. At least not for that reason.
20:40 < searedvandal> revel, start hacking firewood and you'll be there
20:40 < balletjebal> ^^
20:40 < revel> Maybe when winter gets closer.
20:41 < phogg> A hacker is someone who takes the wrong tool and does the job anyway, just to see if it can be done.
20:41 < phogg> Or for the pleasure of figuring out how it can be done.
20:41 < phogg> Or for the fun of doing it faster than the last guy.
20:41 < revel> So... Using shell scripts for $something ?
20:42 < searedvandal> so if I'm using nano instead of vim for my coding I'm a hacker?
20:42 < balletjebal> nothing wrong in being curious when no harm is done i guess
20:42 < phogg> A hacker is someone who revels in exploiting intimate domain knowledge to do something unexpected.
20:42 < balletjebal> ethical
20:42 < phogg> searedvandal: No. Using nano is still using the right tool.
20:42 < ace510> woo, Nano, represent!
20:42 < revel> phogg: Using `cat` then?
20:42 < phogg> balletjebal: there's nothing wrong with being curious, there is something wrong with being insecure due to ignorance.
20:43 < phogg> revel: programming with cat, now that's a real man's game
20:43 < searedvandal> phogg, I know. it was a bad attempt at a stab at everyone saying 'use vim' whenever someone uses anything else.
20:43 < djph> searedvandal: use emacs, you philistine.
20:43 < phogg> if you've never written a script on a system which had cat but no editor... You will, eventually
20:43 < yaldak> theyre all tools in a box
20:43 < searedvandal> djph, I use emacs for everything else, but I need a text editor as well.
20:43 < balletjebal> phogg, true but not knowing and doing stuff is not my way of doing stuff so to speak
20:43 < yaldak> use the one that works best for you and your situation
20:44 < bls> as long as you acknowledge your hacks for what they are and don't advocate them and refuse to learn best practice where it exists
20:45 < acresearch> people where can i go to connect to a wifi in xfce?
20:45 < phogg> yaldak: Or use ed, because ed is the standard text editor.
20:45 < phogg> acresearch: I don't know, but you can always use the command line.
20:45 < acresearch> phogg: command line?
20:45 < hexnewbie> Be sure to tell script kiddies you are not a real hacker unless you use ed
20:46 < phogg> bls: a real hacker by definition already knows how to do it right
20:46 < acresearch> phogg: i think i tried that before and it was so complicated no one was able to help me figure it out
20:46 < phogg> acresearch: you know, that thing you get when you open a terminal
20:47 < phogg> acresearch: I cannot comment on complicated. You write a wpa_supplicant.conf for your AP and then run the daemon. Or, you run the daemon and then manually configure with wpa_cli.
20:47 < phogg> Is it complicated? Probably a bit.
20:47 < acresearch> phogg: i think i will stick with the default application rather than write my own .config file, especially if i don't know what wifi si available
20:48 < phogg> acresearch: you can discover what is available using wpa_cli.
20:48 < phogg> or wpa_gui if you prefer graphical feedback
20:50 < bls> agreed, but I've worked with and seen on here plenty that think their quirky, unmaintainable abuse of some tool wrapped in worst practice bash/perl script should be used by everyone when a first class, maintained tool/library/parser already exists because "it works for me" or "tools to parse XML/json/directory structure are bloat, use cut/sed/awk/raw perl instead"
20:50 < searedvandal> acresearch, xfce menu - settings - network connections (or something like that). or the network manager applet icon on your panel
20:50 < akd> I was able to use accent with my "English (US Alternative Layout", I was able to write "é" when pressing (ALT-GR + ") + e , now I am writing "'e". I want to restore my ALT-GR . I dont remember where is the kebyoard configuration. I am on Linux Debian Stretch With a Cinnamon desktop. Any Help ?/
20:52 < MarkusDBX> If I got a fast network and I want to forward gpu instructions to another machine, is X11 forwarding and iglx still the way to go, or is there faster ways?
20:52 < MarkusDBX> faster/newer?
20:53 < phogg> bls: I do not disagree in any way
20:54 < triceratux> akd: on siduction the alt-gr is controlled via /etc/default/keyboard. i got mine back by setting it to pc105
21:00 < pikaro> what's the best practice workflow when my ssl cert has renewed in the background and I want apache to serve the new one? do I really have to restart the server? I'm assuming a reload will do as well. are there any more specific commands? using debian stable.
21:01 < hexnewbie> pikaro: ‘systemctl reload apache2’, under the scenes that calls apachectl graceful
21:02 < pikaro> "This  differs  from  a normal  restart  in that currently open connections are not aborted." ah nice, thank you!
21:08 < slojanko> Hello, I'm trying to echo something to a fd in /proc/ID
21:08 < slojanko> echo "test" >> 1, was hoping it to appear on my console output but it doesn't happen
21:08 < slojanko> I'm already in /proc/id/fd directory
21:10 < phogg> slojanko: 1 is stdout, try 0
21:10 < slojanko> Also doesn't but why would it be 0?
21:10 < phogg> actually I doubt either should work
21:10 < searedvandal> echo "test" > &1
21:11 < phogg> slojanko: both work for me
21:11 < phogg> at least to some extent
21:12 < akd> triceratux, I have pc105 but it still do not have 'e accent
21:12 < slojanko> I should be in my shell PID not process?
21:12 < akd> I was able to use accent with my "English (US Alternative Layout", I was able to write "é" when pressing (ALT-GR + ") + e , now I am writing "'e". I want to restore my ALT-GR . I dont remember where is the kebyoard configuration. I am on Linux Debian Stretch With a Cinnamon desktop. Any Help ?/
21:12 < phogg> slojanko: I cannot tell what you are asking.
21:13 < slojanko> Difference between ppid and bashpid
21:13 < slojanko> Okay I'm in the wrong dir
21:14 < slojanko> phogg: thanks
21:15 < searedvandal> wouldn't >&N be easier than /proc/self/fd/N in most cases?
21:16 < phogg> searedvandal: I assume he's actually in a different process
21:17 < searedvandal> of course, brainfart.
21:17 < searedvandal> sorry
21:29 < tonsofpcs> Looking for suggestions on an NMS that can handle snmp from non-standard (not HP, not Cisco, not anything you've ever heard of) systems and automated logging on a specific timetable (say every n hours).  Suggestions?  Web-based interface preferred but I'll run it on any platform imaginable.
21:30 < ayecee> what is a nms
21:30 < ayecee> network monitoring system?
21:31 < Psi-Jack> Nano Medical Services
21:31 < ayecee> nuclear magnetic spin
21:31 < Psi-Jack> Negative Management Syndrome
21:32 < ayecee> i guess he must be busy copying and pasting that into other channels
21:35 < Psi-Jack> Yep. Already did in ##networking
21:35 < Psi-Jack> tonsofpcs: It's bad practice to cross-post and ignore.
21:39 < searedvandal> tonsofpcs, opennms maybe?
21:49 < Dagmar> If they're "non-standard" the answer is simple.  Don't use them, or go ahead and call up the vendor who sold you them and ask how much they charge for more equipment that *will* work with it
21:49 < Dagmar> "Non-standard" SNMP from a vendor is another way of saying "crap"
21:50 < Dagmar> Otherwise, stop being lazy and import the MIB documentation from the vendor so your NMS *can* talk to the equipment.
21:50 < Dagmar> <-- has quite a bit of experience with Spectrum and Business Objects
22:15 < AndroidKitKat> is it "illegal" to take someone elses work on github that isnt licensed, fork it and add a license?
22:16 < AndroidKitKat> nv
22:16 < scampbell> It would demonstrable prior art which could be used to invalidate the license.  That could leave you liable.
22:17 < AndroidKitKat> it has a "UNILICENSE"
22:17 < AndroidKitKat> can I change that to the MIT License?
22:17 < DLange> yes as that work is under personal copyright as it has not been licensed differently
22:17 < AndroidKitKat> I didn't see the license prior
22:17 < scampbell> but more than that, one must realize that take another persons work and placing any claim of ownership or control is improper.
22:17 < DLange> yes, public domain -> MIT is o.k.
22:18 < DLange> (IANAL, etc, yadda, yadda)
22:18 < AndroidKitKat> ya
22:18 < AndroidKitKat> https://hastebin.com/foqiyusiwe heres what the license included says
22:18 < scampbell> yup, public domain specifically grants that right.
22:18 < AndroidKitKat> let me read more about this
22:19 < AndroidKitKat> So i think unilicense is public domain
22:19 < AndroidKitKat> ofc id still give original owner credit
22:19 < scampbell> What is it that you are trying to accomplish?
22:19 < AndroidKitKat> just change the license of a project i forked to something im more comfortable with
22:19 < AndroidKitKat> more knowledge-able about too
22:20 < scampbell> Be aware, the license holder is still free to work out another license with you.   That doesn't happen to much in cooperative works because you'd have to get a lot of sign off.
22:21 < scampbell> If I wrote some cool software myself, gave it gnu license. I can still sell it to you with another license that doesn't restrict you.
22:21 < AndroidKitKat> ya
22:21 < scampbell> Until Dlange, contributes to it, then he's got some say.
22:21 < AndroidKitKat> I might just work with what was there
22:21 < AndroidKitKat> as to not potentially muck something up for myself down the road
22:22 < scampbell> You are wise to do so.
22:27 < kryptynasium> I just found out - system-config-lvm is inactive. Any alternatives (RHEL)?
22:28 < kryptynasium> RHEL 7 to be precise
22:29 < kryptynasium> #SoftwareRetirement
22:34 < searedvandal> kryptynasium, maybe try asking in #rhel also
22:35 < MarkusDBX> If I want a process to use ram on another machine, whilst still using the cpu locally, anyway to do that?
22:35 < kryptynasium> searedvandal: Thanks - Ok I will
22:35 < MarkusDBX> My network card supports rdma
22:37 < fryguy> lol mount a ramdisk over the network
22:42 < Dagmar> He's right.
22:47 < Abbott> openvpn suggests using easy-rsa to generate certificates and private keys, but isn't it just a wrapper for tools that I have already? what would I use to generate the certificates on my own?
22:47 < bls> Abbott: it's just a wrapper around the openssl command
22:47 < DLange> openssl but it is so hard to use that easy-ca is a very good idea
22:48 < bls> Abbott: but there's no real reason to avoid easy-rsa
22:50 < xamithan> Woah matrix died
22:57 < omento> Has anyone ever experienced an LD_LIBRARY_PATH being completely ignored?
22:58 < koala_man> omento: yes, for static binaries and things called through sudo
22:58 < Dominian> !3
22:58 < Dominian> oops
22:58 < ayecee> that's my password too!
22:59 < omento> koala_man: I've built Python with --enable-shared so it wouldn't build statically and it's linking to the system libpython in /usr/lib64 rather than my location.
22:59 < dgurney> Root#123 is the best and most secure root password
22:59 < dgurney> trust me!
22:59 < omento> Hmm, the sudo part is interesting
22:59 < omento> I'm currently root
23:01 < omento> No difference.
23:01 < koala_man> does 'env' show your variable?
23:01 < omento> Yup. `LD_LIBRARY_PATH=/apps2/python/2.7.15/lib:/apps2/ncurses/6.1/lib:/apps2/openssl/1.0.2o/lib:/apps2/zlib/1.2.11/lib:/apps2/gcc/5.4.0-alt/lib64:/apps2/gcc/5.4.0-alt/lib:/gpfs/gpfs1/lsf/7.0/linux2.6-glibc2.3-x86_64/lib:/usr/lib64:/usr/local/lib`
23:02 < koala_man> omento: what's "your location" in there?
23:03 < omento> As in PWD? A completely unrelated directory to anything in that path. A NetCDF build parent directory.,
23:03 < koala_man> you said it's picking libraries from /usr/lib64 instead of your location
23:04 < omento> $ ldd `which python`: libpython2.7.so.1.0 => /usr/lib64/libpython2.7.so.1.0
23:04 < koala_man> what did you expect it to say?
23:04 < omento> It's bypassing the LD_LIBRARY_PATH. It doesn't do this on my local machine, but it does on our cluster. It should be /apps2/python/2.7.15/lib/libpython2.7.so.1.0
23:05 < koala_man> there's a cluster? does it have a job scheduling system?
23:06 < omento> SLURM. But it's not related at the moment. I'm building a Python module and the build system uses an environment system from our modules, which sets the environment correctly, but it doesn't seem to be utilized properly by python.
23:07 < omento> If I were to load up any other library/binary I've built they link correctly, but Python doesn't. Which feels weird because I could have sworn it was when I built it a few days ago.
23:07 < koala_man> are the files available on whichever node the thing runs on? are they the right arch?
23:08 < omento> I just flipped to another node and things work as expected. Something's up with this node
23:08 < frwlkkaw> HELLO
23:08 < frwlkkaw> someone help me im trying to build kernel module to ati driver its giving me this error
23:08 < frwlkkaw> https://paste.linux.community/view/e9362eac
23:09 < omento> koala_man: It's something for me to look at tomorrow. Thanks for replying.
23:17 < kurahaupo> frwlkkaw: do you have the xsave file anywhere on you system?
23:18 < frwlkkaw> kurahaupo: no
23:19 < kurahaupo> frwlkkaw: have previous module builds worked?
23:19 < frwlkkaw> kurahaupo: i dont know i have ubuntu 16 it already came with driver installed i guess but i want to install the ati one
23:20 < kurahaupo> frwlkkaw: what have you figured out so far?
23:21 < frwlkkaw> kurahaupo: i just wanted to get opencl runing
23:21 < Dagmar> He's never built much of anything on that machine before is my guess
23:22 < kurahaupo> frwlkkaw: do you have experience doing software development in C?
23:22 < frwlkkaw> kurahaupo: just a litle bit why
23:23 < kurahaupo> frwlkkaw: so I don't waste time telling you stuff you already know
23:24 < frwlkkaw> kurahaupo: what do you wanna tell me? i think im just going to try install on ubuntu 14 since it says on ati its tested , and i dont want to waste too mutch time with this
23:25 < kurahaupo> Firstly, does Ubuntu have a pre-built ATI driver in a package? Secondly, you probably need to install the kernel headers package (that matches you kernel version)
23:25 < Dagmar> That you should try Google once in a while: https://laanwj.github.io/2016/05/06/opencl-ubuntu1604.html
23:28 < searedvandal> "NOTE: For Ubuntu 16.04 LTS and above, the AMD Catalyst or fglrx driver is no longer supported by AMD, or in Ubuntu."
--- Log closed Wed Jun 13 00:00:56 2018