--- Log opened Fri Jun 01 00:00:04 2018
--- Day changed Fri Jun 01 2018
00:00 < jim> annihilator, would you say you need something stable for that?
00:00 < jim> if so, you might try debian
00:02 < annihilator> partially stable.  basically as debian testy repo
00:02 < annihilator> something that will work but may need me to fix a few bugs or lines of code myself.
00:02 < bls> probably the best approach since the criteria appears to consist of "give me something to entertain myself" more than anything else
00:02 < annihilator> throwing myself into the deep in of coding
00:03 < annihilator> i used to debug software a few decades ago but since got out and want to re up my skills but im bored if im not challenged to learn something i need such as debuging the OS i need to use to do anything.
00:08 < bls> then rather than throwing random poorly made distros onto a USB, why not work through LFS?
00:09 < annihilator> ubuntu is not that bad and gentoo is good but i will try LFS.  linux will be better than WSL on windows.  LMAO
00:10 < chomwitt> how can i find the IRQ number assigned to my usb keyboard?
00:11 < bls> your keyboard isn't going to have one
00:13 < jim> annihilator, for LFS, you'll need some experience with linux, which you seem to have
00:14 < jim> if it turns out you're somewhat weaker, I think probably you'd still be able to do it
00:15 < jim> chomwitt, I dunno... have you looked at /proc/interrupts?
00:15 < annihilator> i dont have a lot as a lot in here but i have enough to get me into trouble and fix it.
00:15 < jim> the "and fix it" part will be key
00:16 < l__q> https://serverfault.com/questions/914677/gnutls-handshake-failed-error-in-the-push-function
00:16 < l__q> maybe someone knows
00:17 < annihilator> i have google lfs but cant find how to get access to it
00:17 < bls> http://www.linuxfromscratch.org/lfs/download.html
00:17 < Psi-Jack> annihilator: LFS is not a distro.
00:18 < l__q> how to fix gnutls handshake?
00:19 < Psi-Jack> Best way to fix gnutls is not to use gnutls.
00:19 < l__q> or is this issue with openssl?
00:20 < annihilator> this will be interesting but looks like fun
00:20 < Psi-Jack> If you're using gnutls, your not using openssl.
00:21 < l__q> SSL routines:ssl23_write:ssl handshake failure:s23_lib.c:177:
00:21 < l__q> https://github.com/Microsoft/WSL/issues/3265
00:21 < bls> heh, and not even on linux
00:21 < Psi-Jack> l__q: WSL is not supported here.
00:22 < l__q> okay, sorry
00:25 < annihilator> wsl is supported in a windows/ubuntu channel i forgot which one cause im using linux instead of wsl lol
00:26 < chomwitt> jim: yep , i see an xhci_hcd entry,  but i think that must be for all usb devices?
00:27 < bls> chomwitt: do you understand how interrupts work?
00:27 < bls> at the physical level?
00:28 < chomwitt> bls: no, just the broad picture
00:29 < chomwitt> bls: kernel get interrupted and according to the IRQ line the appropriate driver-hanlder will take over
00:30 < chomwitt> kernel gets..
00:30 < bls> chomwitt: they're essentially lines tied to pins on the processor that indicate there's a piece of hardware that needs attention. you don't wire your USB keyboard directly to your processor, your keyboard talks to the USB chip and the USB chip is what has the interrupt interface with the processor
00:32 < T-Rog> https://pastebin.com/eawDdkBd
00:32 < Psi-Jack> pastebin.com is frowned upon due to many issues they themselves have caused. Pastes being reformatted, malvertising, adblock blocking, being blocked due to many reasons. See /topic for the channel's official pastebin.
00:33 < Psi-Jack> And, a random pastebin link with no context.... Not very useful. :)
00:33 < chomwitt> bls: so i can only found the sum of all usb interrupts , not for each usb device?
00:34 < T-Rog> Psi-Jack: I broke KDEinit5 again upgrading to a newer version of Mesa. I also changed vendors. I decided to upgrade because one of the games I want to play fails, and I compiled a debug build and used gdb and did a backtrace, and it's fucking up in Mesa
00:35 < T-Rog> so I sleuthed around on google until I found a forum topic on OpenSUSE's forms and a guy was having the same problem with KDEinit5 not working, and he did a objdump -p /usr/bin/kdeinit5 | grep NEEDED and got the same results that I got
00:35 < T-Rog> they told him to forcibly reinstall a bunch of packages and it worked for him
00:35 < T-Rog> it didn't work for me
00:35 < T-Rog> I don't know what to do and I'd much appreciate some help
00:35  * Psi-Jack sighs. "And that language!"
00:37 < evanesoteric> How can I prepare microcode using iucode_tool? I have extracted an archive containing newer microcode, but can't seem to compile into .dat, or whatever needs to be done. What needs to be done?
00:40 < Psi-Jack> evanesoteric: What are you actually trying to do, what distro, etc?
00:41 < deego> i know extensions can, but can chrome themes also contain exploits/malicious code?
00:41 < T-Rog> should I just give up, is it impossible?
00:41 < bls> trying to get someone to hand hold through applying processor u-code for meltdown/spectre
00:41 < Psi-Jack> T-Rog: Tried asking in #suse?
00:43 < jim> evanesoteric, , ok, if you could prepare this microcode you got, what problem would that solve? what do you want to apply the microcode to?
00:43 < T-Rog> Psi-Jack: last time this happened I tried asking in suse no one helped me forever and then they eventually just told me to roll back the changes using snapper
00:43 < Psi-Jack> And.. you don't use btrfs do you?
00:44 < T-Rog> no
00:44 < Psi-Jack> And did you say that? :p
00:44 < Psi-Jack> What I recommend is.... Backups. Make them. Have them. Use them when you screw things up royally.
00:45 < T-Rog> reverting changes isn't fixing the issue.
00:45 < Mibix> anyone ever set up a mysql server for kodi on ubuntu? I cant seem to connect to the one I just made using this https://kodi.wiki/view/MySQL/Setting_up_MySQL#tab=Ubuntu_Linux
00:45 < cookmod>  /quit
00:45 < evanesoteric> jim: I'm just trying to patch spectre-meltdown on an older computer, this part is dreadful.
00:45 < evanesoteric> https://serverfault.com/questions/914680/spectre-meltdown-i5-2410m-update-microcode
00:45 < T-Rog> I have a problem with the graphics driver suse provides
00:45 < T-Rog> and doing nothing isn't going to solve it
00:46 < kerframil> evanesoteric: the answer is in the article I linked to earlier
00:48 < Psi-Jack> evanesoteric: Linux has already mitigated this.
00:48 < Psi-Jack> With retpoline.
00:48 < kerframil> evanesoteric: in short, there's a package that installs the microcode, whereas intel-ucode bundles up one or more of these payloads into a cpio archive that can be referenced as in initrd by grub.
00:49 < kerframil> evanesoteric: if all this seems like hard work, consider what I said earlier - which is that you could very likely just apply a firmware update from your laptop vendor
00:50 < evanesoteric> intel-ucode
00:56 < kerframil> also, generic retpolines ain't gonna work unless you build your kernel with at least gcc-7.3.0. this matters if you use gentoo.
00:57 < bls> weren't there also some mitigations needed in glibc as well?
00:59 < kerframil> you can go the whole hog and build your entire userspace with them, if so inclined
01:05 < evanesoteric> thx kerframil - I took notes. I'll find other clues and eventually get something figured out. Those gentoo links will come in handy soon.
01:10 < domhnall> I'm troubleshooting a live-usb with persistence setup. The system respects all changes made after a reboot EXCEPT the root password. Where should I be looking to understand why the root password isn't persistent across reboots while other changes are?
01:11 < domhnall>  - /etc/shadow shows the changes before and after reboot and it's clear that the password isnt persistent.
01:16 < jim> domhnall, curious, first, what do you think of your experience with linux?
01:17 < domhnall> jim: I have a fair understanding of how to install, update and admin a linux system.
01:17 < BlueProtoman> My laptop isn't detecting my (built-in) webcam.  How can I tell if this is a hardware or a software problem?
01:18 < jim> domhnall, ok, my curiousity is more around whether you like linux, whether you find it useful
01:19 < domhnall> I've also moved from Linux to OBSD as my main system. But I've always tried a Distro that interested me. I.e Gentoo, Arch, Slackware, OpenSuse, Ubuntu, Debian, CentOS, Kali, Tails, Pentoo and Sabyon...to name a few.
01:19 < evanesoteric> kerframil: If I download the latest microcode from Intel, do I want to replace the /lib/firmware/intel-ucode  or cp to the existing /lib/firmware/intel-ucode? Replace with new or append?
01:20 < domhnall> jim: absolutely...
01:22 < domhnall> jim: my adventure in Linux started in 2012 or so after taking a Unix course in college. I've been hooked on CLI ever since.
01:22 < jim> ok, you moving to obsd, did you not have enough disk space to have a full install of linux?
01:23 < domhnall> obsd is using all diskspace right now, but I've dual-booting Linux/Linux before, so yes.
01:23 < jim> ok, I see. sec, checking something
01:24 < domhnall> most recent full-disk install of linux was Arch, Gentoo and Slackware.
01:24 < domhnall> that'd be a cronological order...first being most recent.
01:24 < jim> slackware, that was my first linux that I installed (first one I had was SLS)
01:25 < koala_man> BlueProtoman: does lsusb show it?
01:25 < domhnall> my first was (as many have likely) Ubuntu. Then OpenSuse...(by accident, lol)
01:26 < jim> then being the brillient person that I am, I replaced libc in my linux, not realizing I'd be disabling every binary executable on my system...
01:26 < BlueProtoman> koala_man: It's not USB, it's built-in
01:26  * domhnall found it really cool when slackware had 13.37 ... that being haxor speak for leet.
01:26 < koala_man> BlueProtoman: built-in cameras are often USB
01:26 < Pentode> lol jim
01:27 < Pentode> did you try and fix it with sym links afterwards? ;p
01:27 < jim> no, I had to bail on it... went to redhat (which I think at the time was at vers 2)
01:28 < jim> then I tried to upgrade it... that was a failure, so I bailed on that too, and went to debian
01:29 < BlueProtoman> koala_man: Really?  Didn't know that.  I don't know what my web cam would be listed as, though
01:29 < BlueProtoman> koala_man: http://paste.ubuntu.com/p/Csh3Hbfctc/ sudo lsusb -v
01:30 < jim> at the time, we had a.out executables and libc4... then it came time to upgrade the debian, and it happened that the upgrade included moving to elf executable from a.out ones, and moving from libc4 to libc5
01:30 < koala_man> BlueProtoman: doesn't look like any of those. does it have a kill switch?
01:30 < jim> to my amazement, the upgrade went smooth as silk, and the result worked like a champ
01:31 < Pentode> good times
01:31 < jim> never looked back :)
01:31 < Pentode> the minute i realized linux existed i knew i was in trouble
01:32 < Pentode> 50 or so floppy disks later...
01:32 < deepfreez> Hi, how did you monitoring a lot of servers? what software do you use... and how?
01:32 < jim> Pentode, did you do slackware too?
01:32 < BlueProtoman> koala_man: Actually, wait.  Alternative plan.  It turns out that as a feature specific to this laptop, I can turn on my webcam by pressing Fn+F10...however, doing so does nothing.  How can I figure out if the OS even knows what this means?
01:32 < DoctorDick> I use ton cans and string
01:32 < Pentode> yeah
01:32 < DoctorDick> Tin*
01:33 < jim> those ton cans, the weigh a lot
01:33 < Pentode> i think i got it from wallnut creek or something, i cant remember. ;p
01:33 < DoctorDick> About a ton actually
01:33 < jim> yeah I don't remember the exact weight, but I couldn't lift em
01:34 < koala_man> BlueProtoman: check dmesg to see if any new devices connect when doing that
01:34 < DoctorDick> deepfreez: But seriously I use Nagios
01:35 < jim> there's something for monitoring net connectivity, it made these fairly pretty web graphics that showed net usage over time
01:35 < deepfreez> I used a old version of nagios :) 5 years ago
01:39 < BlueProtoman> koala_man: Ah, yes, one does!
01:39 < BlueProtoman> Son of a bitch.  Thanks for the tip.
01:41 < BlueProtoman> koala_man: Now, how can I uniquely identify this device?  I want to set up an indicator applet to tell me whether or not my webcam is on (by looking somewhere in /dev)
01:41 < domhnall> hm..okay, looking into running a live ubuntu for someone and it's not connecting to wifi. 18.04LTS, any tips?
01:43 < domhnall> ah nevermind, they think its plain and dont like it...moving on to the next.
01:44 < Loshki> domhnall: run a more stable, older LTS. Also, http://dilbert.com/strip/2018-05-29
01:45 < domhnall> "I want it to be like yours", they say. But umm, I put work in on this machine...something they wont be willing to do being a newb.
01:45 < domhnall> Loshki: hmm...ok.
01:47 < koala_man> BlueProtoman: if you don't plan on having multiple cameras you could just look at the video0 device or whatever it's called these days
01:47 < domhnall> Loshki: question, unrelated to Ubuntu, I'm having trouble with usb-persistence and root password not persisting across boot after changing...besides /etc/shadow..where else can I look to troubleshoot this issue.
01:47 < koala_man> BlueProtoman: alternatively, you can have udev recognize it by device id or serial number and create a symlink of your choice
01:47 < DoctorDick> You could use the lsusb bus and device ids
01:48 < koala_man> or parse lsusb, yes
01:49 < domhnall> at least the touchpad worked for them out-of-box. thats sometimes rare.
01:52 < Loshki> domhnall: it's a good question. I don't know if there's any standard to what gets saved on a persistent usb stick. Have you poked around it? Is it a regular filesystem? Do they overlay it, or copy, or just mount it?
01:52 < domhnall> yes, it does overlay
01:54 < domhnall> overlay and /dev/sdb3 show same diskspace usage. example, If I install a new application, both are update to reflect the new install.
01:55 < domhnall> overlay         27978836 7427072  19107464  28% /
01:55 < domhnall> /dev/sdb3       27978836 7427072  19107464  28% /lib/live/mount/persistence/sdb3
01:57 < domhnall> /dev/sdb1        2709088 2709088         0 100% /lib/live/mount/persistence/sdb1
01:57 < epicmetal> domhnall: use a pastebin
01:57 < domhnall> not spamming channel, im done posting pastes....sorry if it was annoying.
01:57 < Loshki> domhnall: and are /etc/passwd and /etc/shadow part of the overlay? Are the dates reasonably recent?
01:57 < domhnall> Loshki: let me check on that
01:57 < Loshki> domhnall: it was annoying and no one asked
01:58 < domhnall> noted
02:04 < HappyHobo> How can I tell which usb stick is which without having to use gparted?  I don't want to format my harddrive by accident.
02:05 < Pentode> fdisk -l?
02:05 < HappyHobo> isn't fdisk format?
02:05 < Pentode> it's a lot of things
02:05 < domhnall> '-l'
02:05 < esselfe> HappyHobo: lsblk -o NAME,LABEL,UUID
02:06 < HappyHobo> I don't know any of that.
02:06 < Wharncliffe> dmesg   might be helpful.
02:07 < esselfe> HappyHobo: lsblk is what I always use
02:07 < HappyHobo> Now that suonds famiair
02:08 < HappyHobo> lsblk rocks
02:08 < domhnall> so does diskutil
02:08 < esselfe> HappyHobo: or lsusb, but I'm not sure it shows the device naming
02:10 < HappyHobo> diskutil doesn't exist domhnall
02:10 < domhnall> not for Linux
02:10 < HappyHobo> What's it for?
02:10 < domhnall> OSX
02:11 < domhnall> does the same thing as lsblk
02:11 < HappyHobo> oh brb  must switch units
02:11 < annihilator> i know this is off topic but what do yall think about linux on dex (samsung phones)?
02:12 < esselfe> HappyHobo: you can also add FSTYPE in the options to see whether it's vfat or ntfs
02:12 < masuberu> good morning, I am quite new to c++, I am trying to run a tool in linux which gives me this error http://dpaste.com/2Q7Q1V2 any thoughts?
02:12 < domhnall> annihilator: arm devices support linux fairly well
02:13 < domhnall> soon as I find my sdcard, I'll start my first arm install.
02:15 < annihilator> its not out yet but its suppsed to be something like wsl but for ubuntu when S8 and newer are in dex mode
02:15 < ayecee> and soon after, a leg
02:15 < Pentode> masuberu, it depends on a version of glibc you don't have. have you compiled the utility from source?
02:15 < masuberu> Pentode, no, someone just sent it to me
02:16 < annihilator> i need to find a way to install linux on my phone without rooting or screw it up lol
02:16 < domhnall> ayecee: heh, nice.
02:16 < Loshki> HappyHobo: you start with a full backup. Even the most careful have a bad day and wipe the wrong disk eventually...
02:16 < masuberu> Pentode, so I guess I can either compile it or install dependencies?
02:16 < Pentode> masuberu, you need to build it on your machine for it to run properly
02:16 < domhnall> annihilator: not sure you can without rooting it...might be possible though.
02:16 < Pentode> you really don't want to go changing glibc out for other versions
02:17 < annihilator> there are apps but from what i can see its a little slugish i might be working on it if my son is admited to the hospital....LMAO
02:17 < annihilator> i wont have my computer...lol
02:18 < masuberu> Pentode, will it compromise the stability of the system?
02:18 < Pentode> it will break _everything_
02:18 < Pentode> unless you do a side by side install, which you don't sound ready for yet.
02:19 < masuberu> Pentode, ok, I guess it is because glibc is a system core library?
02:19 < Pentode> that sortof thing really isnt recommended at all anyway, i wouldn't even do it myself unless it were dire circumstances, lol
02:19 < Pentode> yeah
02:19 < inthl> I am running fsck on a disk with a few TB and lots of files, having ext4. at 70% the check seems frozen, althow CPU is still consumed by fsck. any ideas what is going on at 70% ?
02:20 < masuberu> Pentode, ok I see, how can I differentiate when a dependency is a core library or not? I just want to be able to identify those type of things myself to avoid problems in the future
02:21 < Pentode> just experience really but libc is really the only one you need to be concerned with for now
02:21 < Pentode> most libraries can be upgraded with your package manager safely
02:24 < masuberu> Pentode, ok, thank you this is very helpful!
02:43 < inthl> I have a serious problem, can one help me please?
02:43 < inthl> what does it mean if bash says "Bad message" on cd somedir ?
02:43 < inthl> or ls somedir
02:43 < ayecee> i've never seen that. could you take a picture?
02:44 < inthl> it simply says
02:44 < inthl> ls: cannot access 'dirname': Bad message
02:44 < inthl> I have this on hundreds of directories
02:44 < ayecee> is that a no on the picture?
02:44 < inthl> just performed fsck.ext4 -vftp -C0 /dev/mdX
02:45 < domhnall> inthl: try update-env
02:45 < domhnall> or even source /etc/profile
02:45 < inthl> i just logged in again via ssh, did not help. I guess that would cover that
02:46 < inthl> any ideas..?
02:46 < ayecee> how about that picture?
02:46 < inthl> this is kinda serious, has fsck broken my fs? :/
02:46 < inthl> ...what picture? of the shell output?
02:46 < ayecee> yes
02:47 < ayecee> they say a picture is worth a thousand words
02:48 < inthl> https://ibb.co/hiLZTJ
02:48 < inthl> there you go
02:48 < Pentode> inthl, have you tried remounting?
02:49 < inthl> yes
02:49 < ayecee> what might have led up to this?
02:49 < inthl> fsck was performed
02:49 < inthl> before that everything worked fine
02:49 < ayecee> what led to the fsck?
02:49 < inthl> just a routine check
02:49 < ayecee> what was the output from that?
02:50 < inthl> I executed the command
02:50 < inthl> https://pastebin.ca/4034564
02:51 < inthl> do note there were some broken direktories, some sub-directories with broken indexes or something.
02:51 < Pentode> the plot thickens
02:51 < ayecee> that looks bad. was the filesystem unmounted at the time?
02:51 < inthl> yes
02:51 < inthl> unmounted.
02:51 < inthl> the command was fsck.ext4 -vftp -C0 /dev/mdX
02:51 < ayecee> what do those options do?
02:52 < inthl> -v seems to be a typo? I guess I wanted -V
02:52 < ayecee> that seems to be a lot of options i don't normally use
02:52 < inthl> f force even if reports clean, t is measurements of time, p is prune, to fix possible errors
02:52 < domhnall> -V is version iirc
02:53 < inthl> -C is just progress
02:53 < domhnall> -v is verbose
02:54 < inthl> dmesg shows something
02:54 < inthl> ext4_dirent_csum_verify:354: inode #2: comm touch: No space for directory leaf checksum. Please run e2fsck -D.
02:54 < inthl> and
02:54 < inthl> ext4_find_entry:1448: inode #2: comm touch: checksumming directory block 252
02:54 < inthl> I wanted to touch a file in there, bad message was the error again
02:54 < ayecee> this was after the fsck?
02:54 < inthl> yes
02:54 < inthl> right now
02:54 < ayecee> did you unmount it immediately before the fsck?
02:55 < inthl> yes
02:55 < erizoe> hey guys, has anyone read this (https://www.tldp.org/LDP/tlk/tlk.html) ? I found it a fascinating read
02:55 < inthl> it was not mounted. afaik fsck would complain about that anyway
02:55 < ayecee> inthl: right, but was it mounted immediately before that?
02:56 < ayecee> inthl: i mean, honestly, i've got nothing. i'm fishing for something unusual.
02:56 < inthl> yes, I already said that
02:56 < inthl> i umounted it, then ran fsck
02:56 < inthl> it was not mounted I am sure
02:57 < ayecee> right, it was not clear if it was mounted before that, or if it was freshly attached
02:57 < ayecee> i guess either restore from backups, or get ready for a hairy data recovery adventure.
02:57 < inthl> what about Please run e2fsck -D
02:58 < inthl> and why would that fail this way :(
02:58 < ayecee> you can unmount it and try it. it might help, or it might make things worse.
02:58 < inthl> there is a lot of files
02:58 < ayecee> would be good to take an image of the damaged partition first, so you have a second chance.
02:59 < inthl> theres lots of TB of data, I have no spare space for that
02:59 < inthl> in fact I did this on 2 discs on 2 hosts
02:59 < inthl> both show 100% the same problem, wth is going on :/
03:00 < Psi-Jack> inthl: Not backed up? Then it's not important.
03:00 < ayecee> burn
03:00 < Psi-Jack> So hotly.
03:01 < inthl> it is in raid, I did not assume that fsck would break the entire fs ...?
03:01 < Psi-Jack> Raid is not backup.
03:01 < inthl> as mentioned, I did this on 2 hosts, both show the same errors
03:02 < ayecee> i'm not saying you've done something weird, but most of the time when a user gets weird error messages, it's later discovered that they were doing something weird.
03:02 < inthl> you have seen the command, the fs was umounted
03:02 < inthl> I did it on a raid array, aka /dev/mdX, ..this is not false afaik
03:04 < ayecee> i don't even know what errno Bad message would correspond to
03:04 < Psi-Jack> So you have some data loss due to some kind of corruption.
03:05 < ayecee> if it makes you feel better, it looks like there was damage before fsck ran.
03:05 < Psi-Jack> Yep
03:05 < ayecee> probably doesn't make you feel better though :(
03:05 < inthl> nope
03:06 < Pentode> inthl, find some ext4 whiz or something but in a nutshell the data was corrupted and some remnants may be in lost+found
03:06 < inthl> I really don't understand, it's not that I ran fsck for the first time, nor is it untested software
03:06 < Pentode> but there really isn't any way to reverse what happened afaik
03:06 < Pentode> no magic fix here :|
03:06 < inthl> well cd /lost+found results in bad message as well
03:06 < ayecee> so it's probably not fsck that did it, but rather fsck detected damage that was already there.
03:06 < inthl> on 2 hosts? wth
03:06 < ayecee> what do the two hosts have in common?
03:07 < inthl> same directory and partition structure
03:07 < ayecee> what else?
03:07 < inthl> the data is totally different
03:07 < ayecee> which distribution are they running? how old are they? do they have shared storage?
03:07 < inthl> basically they are the same, just different IPs and differend data in those partitions
03:07 < ayecee> is it a virtual machine, or a physical machine?
03:07 < inthl> one week old
03:07 < inthl> physical
03:08 < inthl> debian, self-compiled latest kernel
03:08 < ayecee> what kind of computers are these?
03:08 < inthl> intel servers, core i stuff, not xeon
03:08 < Psi-Jack> And why self compiled kernels?
03:08 < Psi-Jack> Intel Core is desktop grade not server grade.
03:09 < inthl> well I wanted to have the latest versions for speed and also for trampoline
03:09 < ayecee> probably server in that they weren't being used as desktop machines
03:09 < ayecee> i.e. there wasn't a user in front of them pushing around a mouse
03:09 < ayecee> inthl: that's a very interesting problem. i wish you luck.
03:10 < ayecee> tried rebooting, by chance?
03:10 < Psi-Jack> Kernels don't necessarily mean speed. And debian stretch already has full retpoline done in 8+
03:10 < ayecee> it seems like every command is failing.
03:10 < ayecee> may be memory corruption rather than disk corruption.
03:10  * ayecee wildly guesses
03:10 < inthl> hm
03:12 < inthl> any ideas at all?
03:12 < inthl> have not attempted a restart yet
03:12 < inthl> trying fsck one more time
03:14 < ayecee> maybe boot from a rescue disk and run the fsck from there
03:14 < Psi-Jack> You mentioned also this was on two machines?
03:14 < inthl> yes
03:14 < Psi-Jack> How so?
03:14 < inthl> I took them down for nightly maintenance
03:15 < inthl> I knew there were faulty directories on one, not knowing about the other one for sure if it was clean
03:15 < inthl> so I umounted them, ran fsck, took an hour
03:15 < ayecee> the plot thickens
03:15 < Psi-Jack> heh
03:15 < inthl> mounted again...then this crap
03:15 < Psi-Jack> faulty directories?
03:15 < ayecee> i am also ?
03:16 < inthl> well sometimes it happens that one can not access a directory, for example
03:16 < Psi-Jack> No.
03:16 < ayecee> what happened when you tried?
03:16 < Psi-Jack> That doesn't just "happen"
03:17 < inthl> ..when a directory index is broken or so
03:17 < Psi-Jack> Again.. That doesn't just "happen".
03:17 < inthl> well it does
03:17 < Psi-Jack> No.
03:17 < inthl> https://serverfault.com/questions/65616/question-marks-showing-in-ls-of-directory-io-errors-too?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa
03:17 < rascul> fight!
03:18 < inthl> this is an example
03:18 < ayecee> that describes errors in the presence of io read failures
03:18 < inthl> this does not happen everyday, but in my 10 years of sysadmin it occured from time to time
03:18 < Psi-Jack> Without even following that link, looking at the search string alone. I/O errors are generally caused by hardware communication issues.
03:18 < ayecee> inthl: have you scanned for bad sectors?
03:19 < Psi-Jack> I'd suspect failing HDDs.
03:19 < ayecee> or failing controller. i had one that could get the same md5sum of a disk twice, but never three times
03:19 < inthl> on 2 hosts, on systems that are one week old?
03:19 < Psi-Jack> Or at least, HDD's that are not properly handling error-correcting, which is quite common.
03:19 < ayecee> inthl: is the hardware identical?
03:21 < inthl> yes
03:21 < Tech_8> hi
03:21 < inthl> Directory inode 2, block #252, offset 0: directory corrupted , Salvage<y>?   ... this is the same inode as from the dmesg error
03:22 < inthl> I see that error on both hosts, I was just running fsck on both again
03:22 < ayecee> same inode on both hosts?
03:22 < inthl> same inode, same block #
03:22 < ayecee> weird.
03:22 < inthl> well the partition setup is the same on both
03:24 < Psi-Jack> And how are these in any way related to each other, being completely separate computers?
03:24 < Psi-Jack> Is the same data in both places? If so, how?
03:25 < inthl> no, as mentioned, totally different data, those 2 hosts are both standalone servers, one with one IP the other one with another, serving different content, but the same way, hence the identical setup and hardware
03:26 < SpeakerToMeat> You know...
03:26 < Psi-Jack> So, completely unrelated.
03:26 < SpeakerToMeat> I always hated poninters and references in C, specially when they where stacked.
03:27 < SpeakerToMeat> But for some reason even after a long time of not using them, I can grasp those better than how many backslashes I need to use for anything in bash
03:27 < ayecee> there are two kinds of people
03:28 < inthl> ffs
03:28 < inthl> I am not sure yet
03:28 < inthl> but I think that second run of fsck fixed this sh*t
03:28 < inthl> I swear I will never run fsck again
03:28 < Psi-Jack> ...
03:29 < SpeakerToMeat> inthl: .....
03:29 < ayecee> once got strangled by my seatbelt. never wearing a seatbelt again.
03:29 < inthl> I so fcn swear
03:29 < Psi-Jack> inthl: Yeah. Take a pill and mind the language.
03:30 < ayecee> eh. it's an emotional moment
03:30 < inthl> it is
03:30  * rascul strangles ayecee's seatbelt
03:30 < inthl> TB of data, important enough for RAID, but no moneyz for tape backups
03:30 < ayecee> maybe you could have two independent disks instead of a raid
03:31 < ayecee> and do rsnapshots between them
03:31 < ayecee> at least that way filesystem errors couldn't ruin your day quite so thoroughly.
03:31 < inthl> well there are lots of possibilities, I just made the choice I had 1. money for, 2. time for, 3. the requirements for
03:31 < inthl> the ls commands seem to work
03:31 < ayecee> fair enough
03:31 < inthl> just got a listing, on both hosts
03:32 < inthl> jeeeez... now what in the world caused this I wonder. do you see any problems with the fsck options..? there is no witchcraft there afaik or something that might have caused this
03:33 < ayecee> less options is better options imho
03:33 < inthl> in fact as I'd claim `usual fsck options` left my fs in a broken state
03:33 < SpeakerToMeat> fsck is a fickle beast
03:33 < ayecee> unless you can confidently list the reason for each
03:33 < inthl> well it was just -p ...the rest was just time measurement and progress report?
03:33 < SpeakerToMeat> The more important the data the more you want to do fsck on an image
03:33 < ayecee> inthl: so leave off those options for easier troubleshooting.
03:34 < inthl> I will never run fsck again
03:34 < inthl> never ever
03:34 < ayecee> "i used all these obscure options" is a big crimp in troubleshooting
03:34 < Tech_8> im back
03:34 < ayecee> Tech_8: i didn't even know you were gone!
03:34 < inthl> almost died of a heart attack here for an hour
03:34 < Psi-Jack> inthl: Never say never.
03:35 < ayecee> sometimes say sometimes
03:35 < Psi-Jack> fsck is a file system checking tool. That's why it exists.
03:35 < ayecee> still, interesting problem alright.
03:35 < inthl> it is a I will break your stuff tool and cause a heart attack and let you die
03:35 < inthl> ..tool
03:36 < ayecee> "i will detect stuff that is already broken"
03:36 < inthl> or make it broken
03:36 < inthl> whatever it was
03:36 < ayecee> not so much that
03:36 < inthl> I played around now with the directories, seems fine for now
03:36 < Psi-Jack> It's not fine.
03:36 < ayecee> you know what they say about problems that go away by themselves
03:37 < ayecee> "whew!"
03:37 < inthl> well I have a different saying
03:37 < Psi-Jack> "trouble for another day!"
03:37 < ayecee> day shift's problem
03:37 < inthl> life is boring without problems. you need them or your life is meaningless
03:38 < ayecee> heh. problems distract you from the existential dread for a little bit.
03:42 < ayecee> (too real?)
03:49 < ffejj> hi y'all
03:52 < maxcell_> oi
04:08 < thatpythonguy> so what WMs do you guys use
04:13 < saltlake> 127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4 in /etc/hosts , can anyone explain this?
04:13 < saltlake> what is localhost4 and localhost4.localdomain4 here?
04:15 < Sitri> ip4
04:16 < Sitri> thatpythonguy: ratpoison/wmii
04:20 < Psi-Jack> thatpythonguy: The one I want to use at the time I want to use it.
04:21 < thatpythonguy> Sitri, is that tiling or floating?
04:21 < thatpythonguy> Psi-Jack, haha ok what about right now
04:22 < Sitri> Both are tiling, but wmii has a floating mode
04:23 < thatpythonguy> and why did you pick them over other tilers?
04:23 < Sitri> However, wmii is also completely unmaintained (the group that made it doesn't even list downloads for the code anymore)
04:23 < thatpythonguy> probably why ive never heard of it haha
04:25 < Sitri> I found wmii's workflow to be excellent.  Ratpoison was available, easy to compile (which i3 wasn't), once I figgured it out I could mimic most of the wmii workflow, and it has a few nicities that wmii didn't have (like 1005 keyboard based window-resizing)
04:26 < Sitri> wmii basically made vitual desktops not only trivial but kind of focused on them.  You could even dynamically change the number of them on the fly.
04:27 < thatpythonguy> thats really interesting, havent heard of anything like that
04:28 < Sitri> I've tried dwm (same group as wmii), awesome (fork of dwm), pekwm (pure floating), fluxbox and some others I can't think the names of immediately.
04:28 < Sitri> Oh, also e16
04:28 < thatpythonguy> ive only heard of awesome and fluxbox out of those
04:29 < thatpythonguy> what about i3 or bspwm
04:29 < Sitri> "easy to compile (which i3 wasn't)"
04:30 < thatpythonguy> my bad
04:30 < Sitri> Haven't used bspwm yet, but it's on my to-try list
04:30 < thatpythonguy> i tried it today as my first tiling wm
04:30 < Sitri> i3 is actually a spiritual successor to wmii BTW
04:31 < jim> what makes i3 hard to compile?
04:31 < n-iCe> hi
04:31 < Sitri> The half-dozen perl modules the compile scripts need
04:31 < jim> oh ok
04:31 < Sitri> I generally don't touch perl at all, so I don't have those and trying to install them has yet to succeed
04:32 < jim> I've installed perl modules before
04:33 < jim> sometimes in my dist certain perl modules are packaged
04:34 < Sitri> Yeah, I was using Slackware, then CRUX when I was looking at i3
04:34 < jim> ahh,,, what is the package system on crux like?
04:35 < Sitri> It's literally ports-like.  Which effectively makes it source-based.
04:35 < Sitri> Except the repo is pitifully small.
04:36 < jim> I see, that does sound kinda hard... did you have to get perl modules from cpan?
04:36 < Sitri> Yeah
04:37 < justan0theruser> whats the best way to backup a luks encrypted drive?
04:38 < Sitri> Any file-based back-up
04:38 < bls> same way you'd back up a non-LUKS drive
04:38 < Sitri> That isn't disk imaging*
04:39 < Happyhobo> My harddrive was just as disorganized as my life.
04:39 < Happyhobo> I've been cleaning, moving and straightening for the past 45 minutes and I only had 104G of data and stuff.
04:40 < justan0theruser> bls: well thing is, the entire encyrpted partition is 1tb, but the data is 250GB, and I want to backup to be encrypted. I was hoping there were some tools available to handle that
04:41 < justan0theruser> s/data/volume/
04:42 < Psi-Jack> justan0theruser: Sure there is. borgbackup does encryption, compression, and deduplication.
04:42 < bls> there are tools specifically for that, they just don't care whether or not the drive is encrypted
04:42 < bls> justan0theruser: read up on borg, gitannex, rsnapshot, bup, etc
04:46 < justan0theruser> ok thanks
04:46 < bls> tarsnap is another option if you're willing to pay
04:47 < Psi-Jack> Nah, borgbackup is all you'll ever need for backups. :)
04:57 < supernov3h> why is the find utility so confusing... I want to ignore a directory, and look for a file, why does this give me only the directory I want to ignore: find . -type f -name \*1.9.9140.59\* -o prune -type -d -path './rootfs'
04:58 < supernov3h> Im trying to find occurances of that string outside of that directory
04:58 < supernov3h> apparently prune acts on the arguments that come after it
04:58 < Gerrymandered> hi
04:58 < nekoseam> hi
04:58 < supernov3h> hi
04:58 < Rekonnected> sup gerry
04:59 < Gerrymandered> How are you all doing?
04:59 < nekoseam> as usual
05:00 < thatpythonguy> swell
05:01 < thatpythonguy> what other channels do you guys hang out on? i feel like they're all dead
05:03 < Gerrymandered> #thelounge-spam and #freenode
05:06 < Rekonnected> yeah, we just set up a web irc client and we're partially testing it now
05:12 < supernov3h> why does find make itself so hard to use ow
05:12 < granttrec> can someone help me with my find command:
05:12 < granttrec> find . \(-name '*.pdf' -or -name '*.epub'\) -exec cp {} ~/books/non-fiction/ \;
05:12 < granttrec> find: invalid expression; you have used a binary operator '-or' with nothing before it.
05:13 < supernov3h> try -o?
05:13 < supernov3h> intead of or
05:18 < granttrec> supernov3h: nope same output
05:22 < granttrec> its the whitespace it matters!
05:22 < bls> supernov3h: have you read: https://mywiki.wooledge.org/UsingFind#A-prune ?
05:24 < strive> Currently experiencing "gpg: no valid OpenPGP data found." after running "sudo apt-key add Release.key"
05:25 < strive> Do I have to install a relevant package pertaining to gpg?
05:26 < bls> the error message is pretty straight forward, the key isn't valid
05:27 < strive> Uh oh.
05:27 < strive> :/
05:27 < strive> Thanks.
05:28 < bls> did you inspect the key and verify it has valid pgp data in it?
05:29 < strive> I was following a guide; I didn't know I needed to do that.
05:30 < strive> Doesn't look like it does.
05:30 < bls> pretty standard procedure when you run a command on a file and that command tells you the file isn't valid
05:30 < strive> Gotcha.
05:30 < supernov3h> bls: no I hadn't but I randomly figured it out, legitimately randomly
05:32 < bls> but yeah, there's plenty of "this guide will help you if everything works 100% as expected, the second it doesn't, sorry, no help"
05:32 < supernov3h> ><
05:41 < nekoseam>  
05:41 < nekoseam>  
05:41 < nekoseam> uhh
05:41 < nekoseam> https://www.reddit.com/r/unixporn/comments/8noqn9/pekwm_blue_and_grey_chromeoslookalike_desktop/ Decided to do a pekwm setup :)
05:41 < supernov3h> so does having the same name local variable in a function, and calling another with a variable (called local) with the same name, break that variable...
05:42 < nekoseam> pekwm is becoming my favorite wm. I highly recommend you guys try it out. It's a lot like Openbox
05:43 < bls> depends on what you mean by "break that variable"
05:45 < nekoseam> welp reddit is down
05:45 < nekoseam> so guess you cant view screenshot
05:46 < nekoseam> https://i.imgur.com/sbwx1Hy.png there
05:49 < Happyhobo> brb must reboot to let the computer get used to the new sorted and restructured home directory
05:49 < Happyhobo> It needs that right?
05:50 < sacules> any expert on slitaz?
05:50 < epicmetal> nekoseam: somehow the cute 503 image helps
05:51 < sacules> nekoseam: seems you're still ricing :P
05:51 < nekoseam> lol
05:51 < bls> sacules: maybe in #slitaz ?
05:52 < sacules> bls: lel good point, forgot about that
05:53 < epicmetal> How do I make Materia-dark-compact GTK theme not mess with Firefox UI/form colours under MATE?
05:54 < nekoseam> epicmetal: which distro
05:55 < nekoseam> specifically version of firefox
05:55 < epicmetal> nekoseam: Arch, Firefox 60.0.1
05:55 < nekoseam> hm
05:56 < epicmetal> Or is there a dark theme that doesn't mess with Firefox? I really do like this theme, though.
05:57 < bls> https://askubuntu.com/questions/8336/how-can-one-make-firefox-ignore-my-gtk-theme-entirely
05:57 < epicmetal> nmourey: apparently it's a Firefox bug https://github.com/nana-4/materia-theme/issues/205
05:58 < epicmetal> nekoseam: ^
05:58  * epicmetal looks at bls' link
05:58 < nekoseam> lots of users are reporting bugs with it
05:58 < nekoseam> yeah
05:59 < nekoseam> https://github.com/nana-4/materia-theme/issues/220
05:59 < nekoseam> "first good issue"
05:59 < nekoseam> devs seem a bit incompetent to be honest. they put most things as low priority
06:00 < epicmetal> bls: that "works", but it's not the effect that I'm after
06:05 < epicmetal> nekoseam: what do you mean by "first good issue"?
06:07 < nekoseam> epicmetal: look at the tags
06:07 < epicmetal> "good first issue"
06:08 < epicmetal> I have no idea what that means
06:22 < masber> good afternoon, i would like to know which version of c++ libraries are included in this rpm centos 6 devtoolset-7 (http://mirror.centos.org/centos/6/sclo/x86_64/rh/devtoolset-7/) could someone help me how to find out?
06:25 < pnbeast> masber, I imagine that if you install the RPM, you can check the information in /usr/share/docs/<devtoolset directory>.  You could also extract the contents of the RPM and examine them for version numbers/information.  I guess RH has some web information, someplace, too, but that is a guess.
06:28 < bls> or `rpm -ql`
06:40 < Kremator> guys, i do want to speed up my boot time, i already disabled a couple of service (including one that does waited for the nic to be alive) so, from these what could i disabled safely? : http://dpaste.com/11SN1QC
06:41 < Kremator> also, what does do the gpu-manager.service? i didnt had it back when i had xubuntu 17.10 (now im on 16.04) so can i disabled it or would screw up with the video drivers?
06:46 < cmj> make tea while your computer boots and relax
06:46 < cmj> service --status-all
06:47 < cmj> and filter through things you don't need
06:47 < Kremator> ok ty, gonna check that
06:47 < cmj> google how to disable services at boot
06:48 < Kremator> cmj, i already know, # systemctl disable <servicename>
06:48 < Kremator> cmj, is njust that idk what does most of the service
06:49 < cmj> you need to also use man update-rc.d
06:49 < Kremator> and btw, what does means the + symbols in the "service --status-all" command?
06:49 < cmj> it means it's started
06:51 < Kremator> oh btw, what are CUPS?
06:51 < cmj> it's the print service daemon
06:51 < Kremator> i see it is started, and yesterday checking up the app list with ufw, i saw them but idlk what it is
06:51 < Kremator> ok, gonna cut that
06:52 < cmj> you need to use update-rc.d to disable these services from starting at boot
06:52 < cmj> assuming you're on a debian-based distro
06:54 < kuri0> why is earlyprintk=efi so slow ?
06:54 < kuri0> the printing is so slow and bottlenecks the boot
06:54 < cmj> do you have printk=debug enabled?
06:54 < Kremator> cmj, ok got it, gonna check update-rc.d
06:55 < cmj> sorry, printk.time=1 initcall_debug
06:58 < Kremator> cmj, just checked, using update-rc.d its only neccesary if you are not using systemd (which im using) or in *buntus earlier than 14.10
06:58 < nekoseam> Man I love zsh
07:02 < cmj> Kremator: i know pretty much zero about systemd, sorry :|
07:03 < Kremator> cmj, oh ok, but i though debian since 7 version migrated to it
07:03 < Kremator> cmj, anyways ty for the guidance
07:03 < nekoseam> cmj: nobody does
07:03 < cmj> i choose not to use many methods of operation
07:03 < cmj> try #debian ?
07:15 < ffejj> bop.
07:15 < ffejj> so many join/parts
07:18 < rumpel> ffejj, you could filter them
07:20 < ffejj> yeah, you're right
07:21 < date_night> Friday woohoo, hows everyone doin
07:24 < pingfloyd> not friday here
07:25 < Kremator> pingfloyd, sad
07:27 < cmj> /ignore ##linux JOINS PARTS QUITS MODES
07:31 < Kremator> cmj, wanna know how do i ignore these joins parts?
07:31 < cmj> i already do
07:32 < sauvin> It varies by client.
07:32 < cmj> true
07:32 < Kremator> sauvin, cmj, i do have an universal client agnostic method
07:32 < Kremator> wanna know it?
07:32 < cmj> /quit ?
07:32 < sauvin> There's only one I can think of: /quit
07:32 < cmj> :p
07:33 < Kremator> kek lol
07:33 < cmj> sounds productive
07:34 < Kremator> cmj, you save badnwidth using that emthod ;)
07:39 < pingfloyd> I only see parts and joins of users that have spoken recently enough
07:40 < pingfloyd> https://weechat.org/blog/post/2008/10/25/Smart-IRC-join-part-quit-message-filter
07:40 < ffejj> pingfloyd, that sounds like a sweet script.
07:40 < pingfloyd> yeah, it works great
07:41 < pingfloyd> after all, I like to know if someone that talking has left, but I don't care about the lurkers.
07:41 < ffejj> i am using hexchat, which i am unfamiliar with, and i also have not used irc much in the last 15 years.
07:42 < ffejj> might need to get me some of this weechat! :)
07:42 < pnbeast> You'll be pleasantly surprised, ffejj.  IRC is now a source of enlightenment and good will and pure, loving truth.
07:43 < ffejj> that's exactly what i need <3
07:43 < pingfloyd> an nobody trying to get you to view adds and collect marketing analytics on you
07:43  * pnbeast prepares the doxxing and DOS bots.
07:44 < ffejj> and no memes, eh?  yeah, i feel the nostalgia.
07:47 < pingfloyd> seems like outside of IRC, nobody knows what IRC is.  Anyone I've met in real life, has no idea what it is.  Even people that are supposed to be technically savvy.
07:47 < Sveta> welcome back :-)
07:48 < ffejj> that's my experience, too.
07:50 < pingfloyd> what are the four most important keys in less?
07:51 < pingfloyd> h, /, n, and N
07:54 < stevendale> Hi
07:55 < Kremator> pingfloyd, well that is because irc as mainstream chat died long ago
07:59 < ffejj> Kremator, right.  in my own experience instant messengers in the early 2000s coaxed me and my frands away from irc.
08:00  * ffejj fondly remembers slutty girls with webcams on msn
08:00 < Kremator> ffejj, early 2000's? what did you use back then? aol?
08:00 < ffejj> i had an account on most of the big messengers.
08:01 < pingfloyd> Kremator: which is a great thing
08:01 < Kremator> pingfloyd, idk, i pretty much like irc to any other method
08:01 < pingfloyd> Kremator: I mean it's great it isn't mainstream
08:01 < pingfloyd> keeps the hipsters away
08:01 < ffejj> including aol, but i preferred the moddded yahoo and msn programs.  how old are you guys btw?  i'm 37.
08:02 < Kremator> pingfloyd, why? so we are not flooded with "normies"/normalfags here?
08:02 < Kremator> pingfloyd, the fact that is not mainstream actually attracks hipster dude
08:02 < Kremator> ffejj, im kinda a newfag, im 23 less 1 week
08:02 < pingfloyd> ffejj: 45
08:03 < ffejj> pingfloyd, ur old tee hee
08:03 < Kremator> i only knew MS messenger after it became "live messenger" (a.k.a the visual update back in 2006 i think)
08:03 < pingfloyd> ffejj: but I still look young
08:03 < pingfloyd> I look more like 30
08:03 < Kremator> pingfloyd, that's all what matters really
08:03 < pingfloyd> people are always shocked when they find out my real age
08:04 < pingfloyd> they think I'm pulling their leg when I tell them my age
08:04 < Kremator> and i guess that explains your nickname, back then pinkfloyd was huge
08:04 < pingfloyd> Roger Waters stayed huge
08:05 < ffejj> msn live is around the time i quit using it i think.  people are sometimes shocked to learn my age.  i look old as shit though but most of the folks i know here are a lot younger, early 20s.
08:05 < pingfloyd> ffejj: I never could stand instant messaging
08:05 < Kremator> ffejj, msn and winXP, imo the best MS products of that era :C
08:05 < Kremator> they both died in 2014
08:06 < pingfloyd> it becomes an echo chamber just like social media except happening in real time instead.
08:06 < ffejj> oh, well that was my whole life back in the day lol.
08:06 < cmj> windows me?
08:06 < Kremator> pingfloyd, yes, but the fact that it was a bit more personal meant that it was "organized" by political agendas just like social media and "news" sites are these days
08:06 < stevendale> OS: Microsoft Windows XP Professional 5.1.2600 Uptime: 4hrs 28mins 29secs Total Memory: 503.37MB Available Memory: 214.05MB
08:07 < pingfloyd> Kremator: that's its other downside
08:07 < Kremator> stevendale, dude... my feels
08:07 < cmj> what channel is this?
08:07 < Kremator> cmj, this is ##nostalgia
08:07 < Kremator> :(
08:07 < dqsii> huh xD
08:08 < pingfloyd> irc was created out of the basic need for real time communication over the internet.  IMs were created out of commercial agendas.
08:08 < Kremator> stevendale, let me guess, you have 512 but you have assigned 8 MB to the onboard/integrated gpu
08:08 < sauvin> pingfloyd, I remember exactly where I was and what I was doing when the radio announced JFK had been shot.
08:08 < stevendale> 96 MB actually, Kremator
08:08 < pingfloyd> sauvin: I'm not that old
08:08 < sauvin> Why am I not shocked? :D
08:08 < Kremator> stevendale, then how is that your total memory is 503 mb?
08:09 < Kremator> sauvin, now, you should tell us, what were you doin at that time?
08:09 < stevendale> Kremator It assigns it as it needs it
08:09 < sauvin> I was clipping my nails. :D
08:09 < Kremator> sauvin, ouch
08:10 < Kremator> sauvin, pingfloyd, guys, usually how do you deal with life after so much time? really, how do you avoid being underwhlemed all the time?
08:10 < cmj> i pretend not to be old
08:10 < pingfloyd> Kremator: be active
08:11 < Kremator> pingfloyd, mentally, phisically or both?
08:11 < pingfloyd> Kremator: the older I get, the less material means much of anything.
08:11 < pingfloyd> Kremator: both
08:11 < pingfloyd> ideally
08:11 < stevendale> I'm only 18 \o/
08:11 < pingfloyd> being on your feet all day is good for you health
08:11 < ffejj> Kremator, am thoroughly underwhelmed :(
08:11 < Kremator> stevendale, wait until you are 23 and have no degree, then your country goes to shit as your economical situation
08:12 < ffejj> Kremator, drugs. :)
08:12 < Kremator> ffejj, oh well, sorry, i just feel the same and im suposedly "young"}
08:12 < pingfloyd> also your mental health as sedentary lifestyle has away of eventually affecting the mind too.
08:12 < stevendale> Actually I'm mentally impaired Kremator, so the government supports me economically
08:12 < cmj> super drifting off-topic
08:12 < stevendale> And I am starting my own computer repair shop
08:12 < pingfloyd> *sedentary lifestyle has a way of eventually affecting the mind too.
08:12 < ffejj> Kremator, it gets worse with age.
08:12 < Kremator> stevendale, still, if the goverment goes to crap, your economical help does too
08:13 < epicmetal> You're inspiring me to get active this weekend
08:13 < Kremator> pingfloyd, so, i should go to do some sport?
08:13 < stevendale> Kremator I guess I'd be relying on my parents for food and water then :(
08:13 < pingfloyd> Kremator: also, lots of cannabis
08:13 < Kremator> im nto so sedentary, but i do not practise sports
08:13 < pingfloyd> that helps lower stress
08:13 < ffejj> the mental deterioration stuff is at least a little interesting because you get to learn new ways to live.
08:13 < Kremator> pingfloyd, i would prefer to keep it clean
08:14 < sauvin> Oh, I'm still *overwhelmed* by the pretty young babes. :>
08:14 < Kremator> stevendale, at least in your case that is supposed to be like that, im fucking 23 and still living with my parents, and i either go out of the country or stay here for the rest of my life trapped
08:14 < pingfloyd> Kremator: sports would probably be good in your case
08:15 < pingfloyd> Kremator: there's also weight lifting.  Good one to get into if you're still young.  Just don't get mixed up with steroids.
08:15 < Kremator> pingfloyd, again i would like to keep it clean
08:15 < Kremator> pingfloyd,  how  old would be considered too old to get into weight lifting?
08:15 < pingfloyd> "the candle that burns twice as bright, burns half as long".
08:16 < pingfloyd> Kremator: it really depends on what your body can handle
08:16 < Kremator> because there is no gym nearby, and definitelly dont have the needed mindset to start by own
08:16 < pingfloyd> Kremator: when you start off, go easy on yourself.  Once your body becomes conditioned, push it to as far as you can.
08:16 < Kremator> by my own at house*
08:16 < Kremator> pingfloyd, do you practise anything in specific?
08:16 < pingfloyd> I did weight lifting in my younger years
08:17 < pingfloyd> I pretty much had no body fat in those days
08:18 < Kremator> pingfloyd, and you kept it natural right? or you went full arnodl swazenager?
08:18 < pingfloyd> I kept it natural
08:19 < cmj> …
08:19 < pingfloyd> I had a build like bruce lee then
08:19 < pingfloyd> except I was a little bigger
08:19 < pingfloyd> I never touched roids though
08:19 < pingfloyd> screw that
08:19 < epicmetal> I've got a build like, uh, John Goodman
08:19 < pingfloyd> at a certain point you're in peek shape and should be happy with it.
08:19 < pingfloyd> *peak
08:20 < pingfloyd> then just maintain it from there as long as you can.
08:20 < Kremator> pingfloyd, ok then
08:20 < Kremator> i always wanted to do alpinism :/
08:20 < Kremator> but lol broke
08:21 < pingfloyd> as in mountaineering?
08:21 < pingfloyd> there's still lots of enjoyable things you can do and be broke
08:21 < Kremator> yes
08:21 < pingfloyd> like hiking
08:21 < pingfloyd> swimming
08:21 < pingfloyd> biking
08:21 < pingfloyd> walking
08:22 < Kremator> pingfloyd, i can't do neither of thsoe because im so broke that just the small things are 2 months of my sañary
08:22 < pingfloyd> any woods with trails nearby?
08:22 < Kremator> pingfloyd, i live in a city, and i already have to walk 13 km per day to go to work so nothanks no more walking
08:22 < Kremator> pingfloyd, no...
08:22 < pingfloyd> bummer
08:22 < Kremator> ?
08:22 < pingfloyd> being inside the city is like that though
08:23 < pingfloyd> sit in front of computer all day, and no nature around to enjoy
08:23 < Kremator> i live inside of a city, with neither of the advantages of it and all the cons of it
08:23 < pingfloyd> closest I ever want to live to a city is in the suburbs
08:23 < pingfloyd> ideally close to some nature
08:23 < Kremator> pingfloyd, i would bear city of i could do any of th city things
08:24 < pingfloyd> or close enough, that I can visit it regularly
08:24 < Kremator> but again >lo broke
08:25 < Kremator> pingfloyd, are you living in a cabin in the woods these days?
08:25 < pingfloyd> no
08:25 < Kremator> suburbs¡
08:25 < Kremator> ?
08:25 < pingfloyd> I live near a smaller city
08:25 < Kremator> pingfloyd, sounds nice
08:25 < pingfloyd> it is and it isn't
08:25 < pingfloyd> but there's some decent parks nearby at least
08:26 < pingfloyd> and if I have the gas, I can cruise to a national park within an hour or so drive.
08:26 < Kremator> that's pretty nice
08:27 < pingfloyd> Gentrification has made the dilemma worse
08:27 < pingfloyd> seems like there used to be more things to do back in the day that didn't cost anything
08:27 < Kremator> sadly her, if i do try to do biking or any of thatr fancy outdoor thing i risk to either : a) get shot by some thieve b) get roll over by a car driven by a stupid moron using a smartphone c) get cancer by inhaling all the fucking crap of the vehicles
08:27 < pingfloyd> now everything so consumer driven
08:27 < Kremator> (i already do plenty of c) option anyways)
08:28 < Kremator> pingfloyd, is just the natural progress of a materialist society
08:28 < Kremator> materialistic*
08:28 < pingfloyd> pretty much
08:28 < Kremator> pingfloyd,  but believe me, better to be like that, than in the opther side of the fence (comunism/socialism)
08:29 < epicmetal> Kremator: I think materialism is orthogonal to that
08:30 < Kremator> epicmetal, sorry, but that sentence does have 20 IQ points over what i do understand lol
08:30 < pingfloyd> Russia is pretty materialistic
08:30 < pingfloyd> and so is China
08:30 < pingfloyd> arguably, China even more so than us
08:30 < epicmetal> Kremator: haha. Probably better to stick to GNU plus Linux for us, anyway
08:30 < Kremator> pingfloyd, russia is hardly socialism these days
08:30 < Kremator> russia si more like a corruption dirven capitalism
08:30 < pingfloyd> Hong Kong is like the world mecca of materialism
08:30 < Kremator> is more like corruption driven capitalism*
08:31 < pingfloyd> Kremator: there weren't ever truly socialist to begin with
08:31 < Kremator> pingfloyd, yeah, but china is straight out slavery imo
08:31 < pingfloyd> they're just another Empire under that banner
08:31 < Kremator> pingfloyd, true, but these days they arent even tryng, they all accepted they need to "consume" and shit
08:32 < Kremator> pingfloyd, yes, but socialist states these days while being materialistic a lot, they are also more enslaving and more corrupt than it's western counterparts
08:36 < Oddity> he thinks there are "socialist" countries :')
08:42 < Kremator> guys, what does do libc-bin?
08:44 < pingfloyd> that's the runtime counterpart of one of the most foundational libraries
08:45 < Kremator> pingfloyd, ok then, what is libc? :)
08:45 < barometz> one of the most foundational libraries
08:45 < Kremator> and what does it do
08:45 < Kremator> god dammit
08:45 < pingfloyd> standard C library
08:45 < barometz> approximately every program you run depends on it
08:45 < pingfloyd> basically the most required library for C there is.
08:46 < Kremator> so why exactly it is called after APT read the packages databse?
08:46 < pingfloyd> practically everything on the system depends on it directly or indirectly
08:46 < autopsy> pingfloyd, I thought it was PINK floyd?
08:46 < Kremator> you know, after installing something in debian based distros
08:47 < Kremator> autopsy, what if he config'd a machine to make diferent tones after getting ping'd ?
08:51 < autopsy> Kremator, hey what if he did man thats cool.
08:52 < Kremator> autopsy, i guess we never know, maybe i gave him an idea ofr  afuture project
09:08 < Minnebo> hi
09:09 < jim> hi
09:14 < ziggylazer> hi
09:15 < ziggylazer> So I hate to say it but I "must" run Kali to study for certs. But I love my old systemd in fedora
09:16 < ziggylazer> I read that It could work just fine adding the deamons. But sounds fishy...And ideas?
09:16 < ziggylazer> Any*
09:17 < bashprogfortysix> adaemons for??
09:18 < ziggylazer> the systemd with firewalld and so on is very nice
09:20 < bashprogfortysix> yeah if youre running , kali linux, im not quite understanding why just for the class sounds weird
09:21 < ziggylazer> This is pree certification studies. No class. Just want my system as I want it
09:21 < bashprogfortysix> oh yeah just be making sure you using you'''re most fav distro otherwise checkout some linux podcast they tell everything
09:23 < ziggylazer> Yeah but kali Is damn near a necessity. But I can live with the old ways for this.
09:25 < ziggylazer> Anyhow. Ist fun bell! During the night I broke a weak RSA by having the public key and a encrypted msg
09:25 < pingfloyd> why must you run kali?
09:26 < ziggylazer> All the tools are already confiigure to work. It just seem like more work to redo all that in a dist that I like. And hey maybe Debian is cool. Who knows
09:26 < eb0t> everything on kali you can get on other machines
09:26 < ziggylazer> eb0t, never said you couldent
09:27 < eb0t> plus 100 percent of people who use kali.....only use about 0.1 percent of its tools
09:27 < pingfloyd> installing them will help you understand the tools better
09:27 < ziggylazer> pingfloyd, you got a solid point.
09:27 < aclaivi> Knowing what tools exist is also a good thing
09:27 < ziggylazer> True
09:27 < pingfloyd> especially if you compile them from upstream
09:28 < pingfloyd> then you'll also know exactly how it is integrated into your system (since you integrated it).
09:28 < eb0t> i used to run kali on a laptop until i realised i only used nmap kismet and all the metasploit stuf f and nessus whic you can get anyway
09:28 < ziggylazer> pingfloyd, so true...
09:28 < ziggylazer> eb0t, there are lots and lots of more tools that I will need that can be a pain to conf
09:28 < eb0t> but that said kali is good for proper hack..just get it installed and the tools are just there
09:29 < ziggylazer> But then again thats knowledge
09:29 < aclaivi> Once you figure out the tools you need, then you can make your own system in your own fashion
09:29 < aclaivi> That's my current belief, that is.
09:29 < aclaivi> (and I have very little knowledge)
09:29 < eb0t> plus real hackers specialise
09:30 < ziggylazer> Well you can always check the references to whats installed on Kali to see what they choose to go with
09:30 < ziggylazer> So there is some guidens that way
09:31 < ziggylazer> Yeah. I'll think I remove kali
09:31 < aclaivi> Remove it if you dont think its going to provide anything to you
09:31 < aclaivi> If you think that it's still a good resource for you there's no reason not to keep it
09:31 < ziggylazer> Its the lazy option
09:31 < ziggylazer> M2 discs cost
09:32 < aclaivi> ah
09:32 < aclaivi> I just made a small 90G partition on a HDD for it
09:33 < ziggylazer> I got a new SSD laying around. But ofc I need a Lenovo special SATA connector.
09:33 < ziggylazer> And some kind of rubber rails
09:38 < pingfloyd> you could put it in a VM
09:42 < ziggylazer> Yeah thats what I got now. But either I lack knowledge regarding how to say use aircrack over a bridged card. ... Or wait. I could just buy some usb thing right?
09:45 < aclaivi> I think that problem is less specific to kali itself rather than to the virtualization your run on
09:45 < aclaivi> ie: if you ran kali as a full desktop then this problem wouldn't occur (wifi and VMs), so it's not really kali's fault
09:46 < pingfloyd> you talking about putting your wifi in promiscuous mode?
09:46 < ziggylazer> Yeah
09:46 < ziggylazer> No
09:47 < ziggylazer> Passive
09:47 < ziggylazer> Prom is easy to fix
09:47 < ziggylazer> But yeah its not Kalis fault. Its VM
09:47 < ziggylazer> Need the card to listen
09:47 < pingfloyd> that's a limitation
09:48 < ziggylazer> But If I got a USB stick then I think that would work with a VM as well
09:48 < pingfloyd> it's a dilemma, best fix it probably just run kali from a usb stick when you need it
09:49 < pingfloyd> then you have that env on the go to
09:49 < pingfloyd> too*
09:49 < ziggylazer> Yeah. Or just buy a new laptop. Need windows that I dont normal run at all for AutoCad
09:50 < ziggylazer> And that doesnt work very well in a VM
09:50 < aclaivi> Just stumbled upon this https://deshmukhsuraj.wordpress.com/2015/06/30/run-airmon-ng-from-vm-without-external-usb-wireless-card/
09:50 < pingfloyd> and dual booting is inconvenient
09:50 < aclaivi> not sure if it will help you, but it's a nice concept
09:50 < ziggylazer> *reading*
09:51 < aclaivi> It does add extra layers, but if you don't mind using kali as more of a terminal application, then it shoudl do the trick?
09:53 < sauvin> ziggylazer, I use Solidworks in a VM. Works fine.
09:53 < sauvin> I can very easily promise that Solidworks is a lot heavier than AutoCAD.
09:54 < ziggylazer> sauvin, I installed it like 3-4 times yet it just lags. Almoste as if it was running under wine or something
09:54 < sauvin> Then there's something wrong with your setup.
09:55 < ziggylazer> Must be
09:55 < aclaivi> Is it bad practice to move the xdg folders (like Desktop, Downloads) etc to a hidden folder instead of users home dir
09:55 < aclaivi> I don't like that the Documents etc folders are created in the first place
09:56 < iflema> aclaivi: delete them.... some apps recreate them. What about luks or gpg or some such?
09:59 < pingfloyd> aclaivi: make them symlinks to /dev/null
09:59 < aclaivi> bully xdg into oblivion
10:01 < pingfloyd> xdg has taken over your home
10:02 < iflema> is this about xdg or hiding "stuff"
10:03  * iflema xdg promotion
10:03 < dimm> hello, All! am i right understand that this output say that ntpd choose 127.127.1.0 for sync? https://pastebin.com/psc9CYbb
10:05 < aclaivi> rip the xdg dirs reappeared on startup, even after changing user-dirs.dirs
10:06 < ziggylazer> so that is a sync to localhost
10:06 < ziggylazer> Perhaps sync to an NTP pool instead
10:06 < ziggylazer> Atleast thats what I do
10:07 < deepfreez_> Hi, if I want to create a file in linux with before " -CPnRT6LANM1WYtci7aOhJgL0YkJMoa7FEjFteyPMdo " it is posible?
10:08 < iflema> who can tell?
10:09 < jim> you mean the sync command?
10:10 < ziggylazer> step by step
10:10 < ziggylazer> https://www.howtogeek.com/tips/how-to-sync-your-linux-server-time-with-network-time-servers-ntp/
10:10 < iflema> file name?
10:10 < deepfreez_> Hi, if I want to create a file in linux with '-' before file name? like: " -CPnRT6LANM1WYtci7aOhJgL0YkJMoa7FEjFteyPMdo " it is posible?
10:10 < jim> now I haven't read the source of the sync cmd, and what I remember is that it flushes all the to-be-written buffers back to disk... I'm not sure what it does over any networking protocol
10:10 < iflema> =)
10:10 < jim> if anything
10:11 < jim> but, -maybe- there's a buffer scheme in some of the network protocols, and sync also makes sure such buffers that have been marked to send, get sent
10:12 < jim> the previous theory is possible, but it's also a reach
10:12 < jim> look at the source :)
10:13 < bezaban> deepfreez_: touch -- -CPnRT6LANM1WYtci7aOhJgL0YkJMoa7FEjFteyPMdo
10:13 < jim> and before that just read the man page
10:13 < iflema> deepfreez_: yes. quote maybe or backslash even
10:14 < iflema> depends where you are and where your going
10:15 < deepfreez_> bezaban, and how I edit it using vim / vi ?
10:15 < bezaban> deepfreez_: vi -- -CPnRT6LANM1WYtci7aOhJgL0YkJMoa7FEjFteyPMdo
10:15 < bezaban> see the pattern here? ;)
10:16 < aclaivi> what does the double hyphen signify
10:16 < deepfreez_> thanks bezaban  :P
10:16 < bookworm> stop interpreting options
10:16 < aclaivi> Neat
10:16 < bezaban> aclaivi: end of command options
10:16 < bezaban> mm
10:18 < nicholasBPM> when copying text between two terminal windows the rows break, i am desperate for a solution.
10:19 < bookworm> what do you mean by the rows break?
10:19 < iflema> nicholasBPM: copying what? what you doing?
10:19 < nicholasBPM> iflema, i want to copy text from one nano terminal session to another
10:20  * iflema i no know nano apart from ctrl+k and x is it x?
10:20 < nicholasBPM> iflema, yes in x
10:21 < iflema> nicholasBPM: I dont know nano or whats its doing sorry. No idea .
10:22 < iflema> nicholasBPM: i mean the ctrl+x shortcut. Thats my nano knowledge ctrl+k ctrl+x and im nore about x oh and ctrl + u is it? meh
10:23 < nicholasBPM> iflema, yeah that dosent work between two different sessions
10:23 < iflema> yes this keyboard make me extra retarded
10:24 < SIX10> i just got a new keyboard for my 16th birthday
10:24 < SIX10> a varmillo with RGBY modifiers
10:24 < iflema> nicholasBPM: I have no idea, maybe someone else knows or can google
10:24 < SIX10> very pretty
10:28 < FManTropyx> hm, linux doesn't seem to support unicode
10:29 < iflema> the kernel?
10:30 < FManTropyx> I think it's this text editor called SciTE
10:31 < iflema> FManTropyx: is you distro configured? is this the only application not playing nice?
10:31 < jim> curious about one thing... does your irc client have the ability to see international characters?
10:32 < FManTropyx> the characters displayed correctly in HexChat and yeah, this system is a little bit screwed up (I accept no blame for this)
10:33 < iflema> :D
10:33 < jim> no one's blaming you for anything
10:33 < FManTropyx> thanks
10:33 < jim> welcome
10:33 < DildoSwaggins> i am
10:33 < Armand> I blame jim
10:33 < FManTropyx> aw, crap
10:33 < iflema> thanks jim!
10:34 < jim> DildoSwaggins, what you do here can sustain your welcome and affect it in other ways
10:35 < DildoSwaggins> FManTropyx, before you can properly view the characters in your text editor, you need to make sure you have the valid language/font packages installed
10:35 < jim> just be aware and be cool.
10:35 < jim> now that, is true
10:35 < DildoSwaggins> i understand jim, i was only teasing him a bit because i planned on helping himn
10:35 < mexx> Hello, I'm tethering my iphone while my laptop's nic is connected to the lan but I can't access the internet through the phone's connection. Do I need to use routing tables?
10:35 < DildoSwaggins> :P
10:36 < jim> DildoSwaggins, I remember a time you were here before, and it didn't become any kind of problem (yet still I'm a bit nervous about your nick)
10:37 < nicholasBPM> I have enabled mouse support in tmux, but how to i copy from one pane to another?
10:38 < jim> mexx, you're trying to have two default routes?
10:38 < mexx> jim: dhcp does that I guess
10:38 < mexx> I'm just trying to access a few hosts through the phone
10:39 < jewboy> whats with all the monkeys around these days?
10:39 < jim> ok... so you do want two connections?
10:39 < mexx> I tried removing the default route from the phone's connection
10:39 < mexx> Oh yes
10:39 < nicholasBPM> sorry found the answer
10:40 < DildoSwaggins> jewboy is on ambien
10:40 < mexx> jim: I want to evade the company lan
10:40 < jim> yeah, when you remove that default route, then everything goes through the other default route
10:40 < mexx> But still access it
10:40 < jewboy> no
10:40 < jewboy> im not on ambien
10:40 < jewboy> i have just seen an abnormal amount of monkeys around lately
10:40 < ffejjj> love it.
10:41 < jewboy> even saw one on the street grunting and smoking a joint
10:41 < iflema> nicholasBPM: the man page is usually more indepth and can point in other directions    man tmux
10:41 < Armand> jim...
10:41 < jim> looks like th3ere are problems to deal with
10:41 < DildoSwaggins> https://i.ytimg.com/vi/3Ctky8ZyfJU/hqdefault.jpg
10:41 < jewboy> then it got violent because someone called it the N word
10:41 < Armand> pwnt
10:41 < aclaivi> Quality
10:41 < nicholasBPM> iflema, i used ctrl+b, ] to paste, i am so happy now
10:42 < iflema> \o/
10:43 < dyske> In the last column of "last" commands output, what does the N in (N+HH:MM) mean?
10:44 < ffejjj> oh, see i didn't realize at first that jewboy was being racist.  i thought there were just monkeys where he lives.
10:46 < DildoSwaggins> dyske, nanoseconds
10:46 < iflema> some ship has dumped 80 containers off my beaches
10:46 < iflema> full of crap im sure
10:46 < ffejjj> hope you find something good :)
10:46 < iflema> no... clean it up quick
10:47 < ffejjj> they will clean it up for ya
10:47 < iflema> they who dropped it? doubt it...
10:47 < dyske> DildoSw really? 2 nanoseconds? in (2+01:23) seems weird.
10:48 < mexx> jim: but adding routes through the phone's connection doesn't seem enough, I need to disconnect the lan's nic
10:48 < ffejjj> iflema: i'm being supportive <3
10:48 < iflema> send me some milk
10:49 < ffejjj> human milk?
10:49 < jim> mexx, what you're trying to do is called multihoming, and it's not easy
10:49 < iflema> eww
10:49 < iflema> pass
10:49 < ffejjj> lol
10:49 < mexx> <pun>it's so much easier using OpenBSD</pun>
10:49 < DildoSwaggins> dyske, ya bro, rly; check here: https://serverfault.com/questions/151109/how-do-i-get-current-unix-time-in-milliseconds-using-bash/588705
10:50 < DildoSwaggins> date +%s
10:50 < DildoSwaggins> will return the number of seconds since the epoch.
10:50 < DildoSwaggins> This:
10:50 < DildoSwaggins> date +%s%N
10:50 < DildoSwaggins> returns the seconds and current nanoseconds.
10:50 < mexx> I just want to add a couple of static routes, damn
10:51 < DildoSwaggins> btw mex, I often use the same setup as yours (laptop with nic connected to lan + tethered to phone), why cant you just let it dynamically assign the IP via dhcp?
10:52 < DildoSwaggins> do you know for a fact that dhcp is disabled?
10:52 < jim> mexx, be careful of the legal risks you take by avoiding your company lan (they range anywhere from "we don't care" to "you're going to federal prison. NOW."
10:52 < iflema> ffejjj: ill just close my mouth
10:52  * iflema ducks
10:52 < DildoSwaggins> this guy ducks
10:52 < jim> depending on what your company does and what you do
10:52 < mexx> DildoSwaggins: dhcp is enabled on my laptop
10:52 < dyske> DildoSw: you misunderstood my question :) I was asking about output of $(last) and it's logged in column
10:53 < DildoSwaggins> mexx, it needs to be enabled on your router/gateway
10:53 < mexx> jim: we don't have a federal prison here
10:53 < DildoSwaggins> your laptop cant assign you an IP if it's invalid on the router
10:53 < ffejjj> iflema: do you want some cow milk, or like some goat milk?
10:53 < DildoSwaggins> otherwise you have to setup NAT tables on your laptop
10:53 < mexx> DildoSwaggins: you mean do I have a dhcp server ? yes
10:53  * iflema moo
10:53 < jim> mexx, ok... I'm just saying that it's possible you're taking a risk
10:53 < ffejjj> i have the best cow milk, believe me.
10:53 < DildoSwaggins> use the same gateway that your laptop uses for your phone
10:53 < mexx> jim: thanks mate
10:54 < jim> now having said that... the risk is all yours and we don't have to discuss it any more :)
10:54 < iflema> thanks jim!
10:54 < DildoSwaggins> disclaimer aside, so... your laptop gets an IP from the router via DHCP, correct?
10:55 < dyske> DildoSw: like dyske    pts/3        :0               Thu Mar 15 16:56 - crash (1+02:32)
10:55 < DildoSwaggins> next, next, next, next, accept, continue
10:55 < mexx> DildoSwaggins: yes
10:55 < DildoSwaggins> no where were we
10:55 < ziggylazer> Need a IRC client for my VM. IRSSI still? Used GUI before
10:55 < ffejjj> weechat seems cool
10:55 < aclaivi> IRSSI/Weechat
10:55 < jim> that being dealt with, what would be wrong with just using your phone tether?
10:56 < ziggylazer> Ok Look at wee also
10:56 < DildoSwaggins> jim, that's what I'm still trying to figure out lol
10:56 < DildoSwaggins> I'm not sure if he's going for a specific setup, or just making it overcomplicated
10:56 < jim> I figure if he doesn't have to multihome, so much the better
10:57 < ffejjj> is discord similar to irc?
10:57 < DildoSwaggins> mexx, are you just simply trying to get your tethering working and go online on your phone? or is it necessary for you to assign it a static IP?
10:57 < dimm> Why i was comment some line in /etc/hosts i still see how reverse resolving to this name?
10:58 < DildoSwaggins> the effect only takes place on the machine you edited /etc/hosts on, other machines will still query the DNS server
10:58 < mexx> DildoSwaggins: sometimes the laptop won't get a lease from the phone and I will assign a static IP but aside from that, it works ok as long as I'm not connected to the lan
10:59 < mexx> restarting the phone will allow dhcp to work again
10:59 < mexx> I don't know if it's a problem with the phone or the computer
11:01 < mexx> ffejjj: discord is similar to slack, I guess
11:01 < DildoSwaggins> hmmm, yeah, hard to isolate if it's an issue w/ the phone requesting an IP or the laptop assigning it one
11:01 < DildoSwaggins> i would honestly just stick w/ restarting the phone and move on w/ it
11:01 < jim> mexx, maybe instead of a default route on your phone, you could add specific static routes for each of the machines you want to contact over the phone instead of the lan
11:01 < DildoSwaggins> but i'm sure we could get to the bottom of the issue ^^
11:02 < ffejjj> mexx: thanks, i was just reading about it.  i haven't used either before.  sorry for being off-topic
11:02 < mexx> jim: that's exactly what I'm trying to do
11:02 < DildoSwaggins> mexx, instead of restarting the phone, perhaps you should try restarting the network interface on the phone (make sure wifi is turned off before tethering to your laptop)
11:03 < Triffid_Hunter> mexx: you need to set the default route via the phone to have a lower metric than the LAN
11:03 < Triffid_Hunter> mexx: and unless you set up ip rules and ip route tables and stuff, all non-local outgoing connections will go via a single gateway
11:03 < jim> mexx, so then, your only default route is for the lan, but your phone is still connected and providing a connection, which you can send specific IPs to
11:03 < mexx> DildoSwaggins: I don't really care about that problem, I can live with a static IP
11:04 < DildoSwaggins> then it's not a problem
11:04 < mexx> Triffid_Hunter: lower than 0 ?
11:04 < Triffid_Hunter> mexx: I have an interesting setup here where I'm using a dns resolver hook and geoip to route connections to a certain country outside my VPN, which I suppose might be somewhat similar to what you're doing
11:04 < Triffid_Hunter> mexx: I set my ethernet metric to 500 and wifi to 2000, vpn is 0 of course
11:05 < DildoSwaggins> get a loan balancer while youre at it #overkill
11:05 < mexx> Triffid_Hunter: if I setup a static route through the phone, it needs to have a lower metric than the default route ?
11:06 < Triffid_Hunter> mexx: yep or a more specific subnet mask
11:06 < DildoSwaggins> mexx, do this:
11:06 < DildoSwaggins> restart your phone to get it working on dhcp, and then save the route/network values once it's working and then save them
11:06 < Triffid_Hunter> mexx: I've seen some setups use 0.0.0.0/31 and 128.0.0.0/31 for the preferred route to avoid having to mess with metrics
11:06 < Triffid_Hunter> no, /1 not /31, forgetting how CIDR works :/
11:06 < DildoSwaggins> whatever IP your phone was assigned, reserve it and assign it statically
11:07 < mexx> Triffid_Hunter: a /32 should be ok then ?
11:07 < Triffid_Hunter> mexx: well default route is 0.0.0.0/0 so /1 is more specific therefore gets used rather than default
11:07 < mexx> Funny enough, I just enabled the lan nic to see the default route metric and although I could not type anything, messages would flow here
11:08 < mexx> That would mean that the phone was still connected
11:08 < DildoSwaggins> mexx, so perhaps it's been working all along? ^^
11:08 < Triffid_Hunter> mexx: sure, incoming data keeps coming but apparently your OS isn't smart enough to remember which interface specific connections were using :P
11:08 < mexx> But I couldn't type anything
11:09 < DildoSwaggins> perhaps a hiccup in the app
11:09 < DildoSwaggins> during the disconnect/reconnect to tether
11:10 < DildoSwaggins> perhaps needed to re-identify via nickserv
11:11 < mexx> well this irc client is running on a distant host to which I'm connected through ssh
11:11 < mexx> but I could still see messages going through, that doesn't make sense
11:12 < DildoSwaggins> i agree lol
11:13 < ziggylazer> What is libsemanage why is it there and what does it do and why is it destroying my life?
11:13 < ziggylazer> And what happens if I just dont comment that setting?
11:14 < ziggylazer> in SElinux
11:16 < ziggylazer> in /etc/selinux/ I dont have the normal conf file for SElinux. I have settings for libsemanage instead.
11:18 < mexx> hm so I changed the default route provided by the phone to the lower one by the lan and I can write here but now I can't access any host in the lan because everything goes through the phone. I added a static route to one host on the lan but I get "No route to host" when sshing to it and "Destination Host Unreachable" while pinging it
11:19 < ziggylazer> pastebin settings and switch to correct channel
11:26 < DildoSwaggins> mexx, check your DNS settings. make sure they're the same as your laptop's
11:27 < DildoSwaggins> and instead of hostnames, see if you can atleast ping the IP's of your lan devices
11:27 < AppAraat> hello, I'm trying to record audio and video of my desktop. I'm doing "ffmpeg -y -f alsa -i hw:0 -f x11grab -framerate 30 -video_size 1366x768 -i :0.0+0,0 -c:v libx264 -pix_fmt yuv420p -qp 0 -preset ultrafast output.avi" but it says "cannot open audio device hw:0 (Device or resource busy)", I have ALSA running, JACK connected to it and then PA as bridge.
11:27 < DildoSwaggins> if not, then something is wrong besides DNS
11:27 < mexx> DildoSwaggins: name resolution is ok for that particular host, it was in cache
11:27 < AppAraat> so is there a way to record audio like this at the same time?
11:27 < DildoSwaggins> cool, so you're past that issue?
11:28 < mexx> but it won't go through
11:28 < ziggylazer> mexx, the channel is networking I was just there with an issue and they are good at....networking
11:29 < ziggylazer> And Please provide some documentaion and you problem will be solved within 20 min
11:29 < skoup> can i ask what are the most common security issues in the Linux kernel?
11:29 < mexx> ziggylazer: linux specific ?
11:30 < ziggylazer> what you are running 0.0.0.0/0 that is networking
11:30 < ziggylazer> Not linux
11:30 < ziggylazer> Just saying
11:30 < ziggylazer> I dont really care. But you will get better help there I think
11:30 < mexx> I have a problem with linux routing
11:31 < ziggylazer> What is "linux routing"
11:31 < DildoSwaggins> well it doesn't hurt to try both channels
11:31 < DildoSwaggins> heh
11:31 < mexx> I will, thanks guys
11:31 < DildoSwaggins> skoup, vulnerable applications being installed on the system
11:31 < skoup> alright
11:31 < skoup> any example please
11:32 < mexx> the kernel itself has had a few problems too
11:32 < DildoSwaggins> but I guess specifically for Linux Kernal, it would be race conditions and local escalation of privileges
11:32 < DildoSwaggins> RCE buffer overflows seem to be more rare these days
11:32 < ziggylazer> Thats my topic for next week
11:32 < DildoSwaggins> heap/stack overflow, race conditions, and privilege escalation
11:33 < skoup> thanks
11:33 < ziggylazer> Better brush up on the C and learn assembly ;)
11:33 < DildoSwaggins> skoup, examples: https://www.exploit-db.com/platform/?p=Linux
11:34 < DildoSwaggins> I guess DOS should go in that list too
11:34 < skoup> cheers didloswaggins
11:35 < DildoSwaggins> youre welcome
11:36 < skoup> those links i suppose they contain actual patches exploit code, am i right?
11:36 < skoup> patched*
11:39 < Squall5668> skoup: not always, unfortunately. People are lazy and things are not supported forever
11:43 < horseface> can i not make every application follow through the proxies in proxychain?
11:43 < horseface> without having to call the application separately through the CLI?
11:44 < DildoSwaggins> you mean w/o telling each app to use the proxy settings?
11:45 < DildoSwaggins> if so, perhaps have your router do the proxying
11:46 < autopsy> Why does opening mirage in a terminal give a module import error no module named mirage?
11:47 < autopsy> What package has mirage module for Python?
11:48 < horseface> no it says that to use proxychanins i need to lauch apps to use it through the command line
11:48 < horseface> like proxychains opera
11:48 < horseface> i want proxychains to just be used for all apps like as if it were set in the encironment variables...
11:50 < FManTropyx> I always do everything as root - no need to escalate
12:05 < Night_Elf> Hello all. If I wanted a distro to be used as a site to site ipsec vpn router, what would be a good choice?
12:05 < mexx> OpenBSD
12:06 < AppAraat> to answer my own question: SimpleScreenRecorder did the job. I was hesitant at first because I was used to it being a GTK app (I am running Plasma), but apparently the new interface is in Qt and it only installed simplescreenrecorder and simplescreenrecorder-lib :p
12:06 < AppAraat> just set the record out to JACK and record speakers
12:07 < AppAraat> do note that you'll also get sounds from PA, so if you don't want that then just disable the PA bridge or something like that
12:07 < Night_Elf> mexx: I mean not too complicated to setup. pfsense is a choice, but I was curious what else is out there.
12:07 < DildoSwaggins> Night_Elf, i've never personally used it but it seems MikroTik's RouterOS is built for this purpose: https://mikrotik.com/download
12:07 < DildoSwaggins> Night_Elf, https://wiki.mikrotik.com/wiki/Manual:IP/IPsec
12:08 < mexx> Night_Elf: takes two lines in one config file on OpenBSD
12:09 < DildoSwaggins> yeah but technically, so does getting tethering to work
12:09 < mexx> DildoSwaggins: LOL
12:10 < DildoSwaggins> :)
12:10 < mexx> technically tethering works
12:11 < Night_Elf> DildoSwaggins: I alctually have a RB952UI as a router. But I don't know how much the ipsec would load the processor. So was thinking also to build a VM in the server for the ipsec related traffic.
12:12 < Night_Elf> mexx: To be honest I have never used any of the bsd-s.  (hides)
12:12 < mexx> Noone's perfect
12:13 < DildoSwaggins> Night_Elf, honestly, you want to go w/ something that has the most support and documentation online. you won't go wrong w/ BSD, sure
12:14 < Night_Elf> Wouldn't it be easier on me if I went the pfsense route ?
12:14 < DildoSwaggins> Night_Elf, that's a similar situation to why i originally looked into routerOS, but then I ended up just buying a new router instead, so no idea how it fares
12:15 < DildoSwaggins> Night_Elf, i've never actually done what you're trying to do, so I'm gonna let others try to help instead of just speculating
12:16 < DildoSwaggins> but upon first glance, pfsense looks like a solid path to take
12:17 < Night_Elf> DildoSwaggins: It seems to be a freebsd, with a webbish interface. I guess that could make life easier to manage.
12:18 < DildoSwaggins> a solid foundation
12:19  * iflema milks nasa api
12:19 < DildoSwaggins> Night_Elf, https://lifeoverlinux.com/how-to-setup-ipsec-vpn-on-pfsense-2-3/
12:19 < DildoSwaggins> it's from last year but seems pretty straightforward
12:20 < Night_Elf> DildoSwaggins: thanks for that. am looking at the link right now.
12:25 < TheWild> hello
12:26 < Night_Elf> Any thoughts about IPFire ?
12:26 < TheWild> I want to run windows console application through Wine, with proper stdin, stdout and unaltered stderr.  wineconsole produces unnecessary escape sequences. How to disable that freaking "Application tried to create a window, but no driver could be loaded." warning?
12:26 < DildoSwaggins> Night_Elf, you're welcome
12:29 < DildoSwaggins> Night_Elf, from https://www.reddit.com/r/homelab/comments/577rfl/ipfire_vs_pfsense/:
12:29 < DildoSwaggins> "IPFire is pretty basic but the features that are there are extremely stable and very well tested. pfSense on the other hand brings a lot of features that many professional firewalls don't even have out-of-the-box and even packs all of that into a modern and sleek web interface. "
12:30 < DildoSwaggins> the consensus (of all 3 people there!!!) seems to be that IPFire is easier to use
12:31 < DildoSwaggins> TheWild, not sure
12:33 < jim> what does this particular firewall do exactly?
12:34 < DildoSwaggins> <Night_Elf> Hello all. If I wanted a distro to be used as a site to site ipsec vpn router
12:34 < mawk> you really need a separate machine for that ? can't you use the real router instead ?
12:35 < DildoSwaggins> lol
12:36 < DildoSwaggins> using an opensource software vs buying a piece of hardware... they're different circumstances. let's just stick w/ what he asked for
12:40 < jim> mawk, it gets harder if the machine that does the routing is also the client
12:40 < jim> I have a separate gateway too (it runs debian, and an old debian package called ipmasq
12:40 < jim> )
12:42 < jim> it also runs a dns and a dhcp server which serves two subnets that go on two separate wires
12:43 < mawk> machine could mean VM in my sentence
12:43 < mawk> I wasn't talking about hardware
12:44 < jim> oh ok... in my case it -is- separate hardware
12:46 < jim> thailand, you would need to be logged into an acct in order to speak here... https://freenode.net/kb/answer/registration or https://freenode.linux.community/how-to-connect/
12:48 < BluesKaj> Howdy all
12:48 < jim> hiya blues
12:49 < BluesKaj> Hey jim ;-)
12:57 < Smithe> I have a script called by pam-motd who must find the user who's logging in
12:57 < Smithe> On ssh it works like a charm, but with tty's not
12:57 < Smithe> Suggestions?
13:00 < DLange> Smithe: usually $USER should be set
13:00 < Smithe> DLange, nope
13:01 < Smithe> Scripts called by pam_motd are called in a clean environment, with no variables
13:01 < DLange> Smithe: does  id -un      work for you?
13:02 < jim> Smithe, ...which implies that any variables you want to have set, must be set in the same script
13:02 < DildoSwaggins> is there any relevant error output by the script?
13:02 < DLange> Smithe: PAM should have $PAM_USER btw
13:03 < Smithe> DLange, it says root
13:03 < Smithe> DLange, I will try
13:04 < Smithe> DLange, nope, $PAM_USER is not set
13:05 < jim> Smithe, I'm curious about that too... do the scripts output anything at all?
13:05 < Smithe> jim exactly
13:05 < DLange> that's bad, it should be for PAM modules, so much for consistency :)
13:05 < Smithe> jim what do you mean?
13:06 < jim> do these scripts print anything to the terminal, or are they silent?
13:06 < Smithe> jim they print in the MOTD
13:07 < jim> ok so nothing like error messages from any of them? are they complaining in any way?
13:07 < Smithe> jim no
13:08 < Smithe> scripts by pam_motd are called as root but with no variables set
13:09 < jim> yeah I'm horrible at pam
13:11 < Armand> jim, what did pam ever do to you, ehh ?
13:11 < DildoSwaggins> Smithe, is this a script or an application? if it's a script, perhaps you can edit it to enable debug/logging
13:11 < DildoSwaggins> if it's an app, perhaps look at it's usage to see how to run it w/ debugging enabled
13:12 < DildoSwaggins> by usage, I mean runtime parameters. there must be a way to get an error output of exactly what's going wrong
13:12 < DildoSwaggins> or even something
13:12 < DildoSwaggins> saying the error prints in the MOTD doesnt really provide much context for us
13:12 < Smithe> DildoSwaggins, I already set it to know what happens when I call my commands
13:13 < DildoSwaggins> ok well, make those commands echo the output and see where things are going wrong
13:13 < Smithe> DildoSwaggins, there isn't an error
13:13 < Smithe> Simply my script doesn't know what user pam is giving access to
13:13 < alipoor90> how can i make a rule for both udp and tcp in nftables? something like: {tcp, udp} dport 53 counter accept comment "accept DNS"	
13:14 < Smithe> Because it's executed as root instead of the pam user and without variables set
13:17 < Smithe> DildoSwaggins, jim, DLange https://superuser.com/questions/981897/trying-to-print-username-at-message-of-the-day-motd
13:18 < Smithe> The problem is that I'm trying to do that in a tty, so cannot search back in the processes to get the current user from the SSH daemon
13:20 < DildoSwaggins> i'll be honest, i haven't worked w/ PAM/kerberos in a couple years... not sure if i'll be too helpful here
13:21 < DildoSwaggins> I did see a couple things online that said you can edit your sudoer file to disable "requiretty"
13:21 < DildoSwaggins> but i'm not sure how relevant that is to your issue
13:29 < DildoSwaggins> Smithe, see if this helps any: https://en.wikipedia.org/wiki/Pointing_stick
13:29 < DildoSwaggins> oops,
13:29 < DildoSwaggins> https://superuser.com/questions/981897/trying-to-print-username-at-message-of-the-day-motd
13:34 < Sitri> DildoSwaggins: try pam_exec with stdout
13:34 < Sitri> Err
13:34 < Sitri> Smithe: ^
13:34 < Sitri> http://www.linux-pam.org/Linux-PAM-html/sag-pam_exec.html
13:36 < Smithe> DildoSwaggins, that's exactly what I've written
13:39 < vim_user_19> hey guys! i'm on an arch linux system right now and i'm asking myself if there is a way to see which commands are running? For example, which command is passed when maximizing a window,changing a workspace,... in real time.
13:39 < autopsy> vim_user_19, I know there's ps.
13:40 < DildoSwaggins> Smithe, sorry I'm not really sure what else to tell you. hopefully someone else can
13:40 < jhodrien> 'Command' ?
13:41 < jhodrien> xev?
13:41 < DildoSwaggins> yeah, just pipe your command to ps to find what's running (including the current command)
13:41 < vim_user_19> nice thanks guys!
13:42 < jhodrien> You probably don't mean processes though.
13:44 < vim_user_19> i give you a clear example. i'd like to change my shortcuts for cycling through windows and i don't know which command to assign to my new key-combination.
13:45 < jhodrien> See, this is where I figured you'd be going.
13:45 < jhodrien> Can you not poke at the documentation for your desktop environment?
13:46 < vim_user_19> yeah right now i'm looking through the stuff, but i thought it would be quicker to ask here.
13:46 < TheWild> "#!/usr/bin/env whatever". Whatever language I test it on, it seems to ignore this line, even when # is not a comment character. Is that because the languages I tested were kind enough, or that the kernel strips the offending line?
13:47 < TheWild> tested: php, python, perl, lua
13:47 < azarus> TheWild: https://stackoverflow.com/questions/1122778/what-does-usr-bin-at-the-start-of-a-file-mean
13:48 < jhodrien> Try it with something obvious.
13:48 < jhodrien> #!/bin/env cat
13:48 < jhodrien> test
13:48 < jhodrien> Run it.  What gets output?
13:49 < TheWild> (I think) I know what does this mean, but how does the language interpreter not get confused with this line.
13:50 < TheWild> jhodrien: that was smart. cat indeed shows nothing magically did strip the line.
13:50 < djph> TheWild: you mean how does the shell interpret the shebang line?
13:52 < TheWild> I mean rather: how doesn't "#!/usr/bin/env" confuse the language interpreter of language in which # is not considered a start of a comment, lua for example. I could assume an edge case has been programmed into the interpreter, but there might be some other magic as well - can't say for sure.
13:52 < Smithe> Sitri, the script is called from session
13:53 < Smithe> session    optional   pam_motd.so motd=/run/motd.dynamic
13:55 < Sitri> TheWild: they're all coded to specifically handle that use-case
13:56 < Sitri> Smithe: Not sure why that's a counter-point here?
13:56 < Falc> TheWild: quote from wikipedia about shebangs: "The shebang line is usually ignored by the interpreter, because the "#" character is a comment marker in many scripting languages; some language interpreters that do not use the hash mark to begin comments still may ignore the shebang line in recognition of its purpose."
13:56 < Smithe> Sitri, pam_motd.so isn't pam_exec
13:57 < Sitri> Right, but it does what you want.  Where-as pam_motd is losing the dynamic motd feature (as per the latest docs)
13:57 < TheWild> ok, thank you Sitri and Falc. I didn't know it's on wiki though.
13:57 < TheWild> I have to bear this in mind when ever writing my own interpreter (oh God, like we don't have enough programming languages already ;))
13:57 < sadasaulna> Falc: cool.  TIL.
13:59 < s10gopal> how to rename multiple files ? i have files like file 1 , file 2 ... i want to remove file word from every file
14:00 < jhodrien> rename ?
14:00 < s10gopal> change file 1 to 1
14:00 < jhodrien> rename
14:00 < jhodrien> That was a suggested answer.
14:00 < jhodrien> man rename
14:00 < s10gopal> but how to remove pattern ?
14:01 < ananke> s10gopal: manual explains
14:01 < ananke> different distros provide different rename tools. so look at the one that came with your distro
14:02 < s10gopal> thx
14:05 < s10gopal> ananke, file names are S10E01 - Allahu Akbar.mp3'          'S10E01 - Latthay Di Chaadar.mp3''S10E01 - Chaa Rahi Kali Ghata.mp3'  'S10E01 - Naina Moray.mp3' , i want to remove s10e01 from every file
14:05 < ananke> s10gopal: well then, 'man rename'
14:09 < sadasaulna> s10gopal: you can also use a for loop and ${var#}
14:09 < sadasaulna> but rename probs easier
14:10 < sadasaulna> NSA recording this convo now that Allahu Akbar got mentioned lol
14:10 < ananke> sadasaulna: and you're on the list
14:11 < sadasaulna> i've been on the damn list for years ;)
14:11 < s10gopal> thx
14:17 < za1b1tsu> I need a common partition for macos and linux where I put my projects. What type of partition can it be?
14:18 < Armand> EXT4 should be fine, if you use something like fuse.
14:19 < ananke> za1b1tsu: vfat is the lowest common denominator that doesn't require any third party components
14:19 < Armand> Good hit, ananke.. I don't use Mac, so fuse was the first answer that came to mind.
14:19 < sie> Are there any tools for monitor setup, so I wouldn't've to xrandr?
14:20 < azarus> sie: arandr?
14:20 < sie> "xrandr gui" — why didn't I think of googling that before :D
14:20 < jhodrien> xrandr support is often integrated with your desktop environment.
14:20 < sie> Does xfce have something for that?
14:20 < sie> Yeah, found it.
14:21 < redredhathat> in a makefile can you set environment variables
14:23 < sie> redredhathat: so makefile could use them or so programs that makefile creates could use them?
14:23 < redredhathat> so that the makefile could use them
14:23 < RougeR> any idea whats causing this blurry text effect
14:24 < RougeR> https://i.imgur.com/FzZa9Ev.png
14:24 < za1b1tsu> ananke: Thank you
14:24 < redredhathat> unless theres a more elegant way of doing it, i want to make en env variable for "YYYY_MM_DD" and in all my config files i always set this to $TODAY
14:24 < redredhathat> but the machine im deploying to wont have this variable
14:25 < azarus> export DATE=`date +%Y_%m_%D`
14:25 < azarus> ?
14:26 < azarus> ah no, %d not %D
14:27 < redredhathat> ya
14:27 < redredhathat> but i can just slap that into a makefile
14:27 < redredhathat> ?
14:30 < azarus> redredhathat: sure, somehow
14:30 < redredhathat> i think this would do it
14:31 < redredhathat> TODAY ?= `date '+%Y_%m_%d'`
14:31 < redredhathat> looking at the other entries in the file
14:58 < pankaj> I do not want to use any interface or tool like Network manager to configure my  network. I just wanted to  use text file and some set of basi and lightweight command so that I can get nearer to IP address and networking. The tools make me sick. Is their any good tutorial or guide that applies to all linu distributions for basic networking via cmd and mosly text files.
15:01 < sadasaulna> pankaj: get a dist without systemd ;)
15:03 < pankaj> sadasaulna: I am fed up by using so called tools that can do it for me. Last week I installed ubuntu server. It was unable to identify my network so I clicked on option that I will do it manually when the system is installed. And after that when I installed I have no idea what to do. I know I could take help with tool but I cannot download the tool because their is no net connection. Therefore I will love
15:03 < pankaj> it the way old guys used to do via some lightweight command and text file.
15:03 < mynameisdebian> I have a command that I am trying to run as root on reboot, and I do not want the command to prevent the rest of the crontab script from executing, as the command does not close on its own.  Is it appropriate to use nohup in the root crontab here?
15:04 < pankaj> sadasaulna: How to get that? Is it possible. How can my system work without systemd?
15:05 < sadasaulna> pankaj: maybe you can achieve what you want on Ubuntu with systemd, i steer clear of it myself.  Last I remembered trying to manage network by hand it wasn't worth it as systemd was tramping over everything..
15:06 < sadasaulna> if you run a distro that doesn't get in the way, then you can just script your own stuff and use the tools like "ip", "ifconfig", and such and do it all yourself, the proper way
15:07 < pankaj> sadasaulna: I want a standard procedure that applies across all linux distributions. Using any tool is just does not help me across using different distributions.
15:07 < pankaj> sadasaulna: You got it. I understood. I want to do the same way.
15:07 < BCMM> pankaj: do you specifically want to do it in a distro-agnostic way? or do you just not want to use the network manager GUI?
15:08 < BCMM> pankaj: debian's /etc/network/interfaces file isn't "not using any tool", but it is a simple text-only config file for setting up networks, and it's available on debian-derived distros too
15:09 < pankaj> BCMM: The GUI can change any time. I do not trust on them. I want to do the way the geek's do. Because many times GUI are not available or anothe problem like no network connection so what should I do.
15:09 < BCMM> pankaj: have a look at `man interfaces`
15:09 < sadasaulna> there is a cli to network manager pankaj (it was ugly as sin iirc)
15:10 < BCMM> honestly, network manager doesn't make much sense on a server anyway. it has some nice features for, like, a laptop. a machine that might move around different wireless networks, tether to a mobile phone sometimes, etc.
15:10 < jhodrien> If I am a member of a group, why would newgrp prompt for a password when I attempt to change my primary group to be it?
15:10 < jhodrien> BCMM: It's the Redhat advocated default on servers.
15:10 < BCMM> but a server usually wants the sort of totally static network config that can be achieved with more reliable tools
15:11 < BCMM> jhodrien: the redhat advocated default is not always something that makes sense
15:11 < sadasaulna> BCCM, damn straight
15:12 < pankaj> BCMM: Is their any resource on these old way of network configuration when people used to do all my hand and text files configuring ip addresses. Because I want to understand things and using tools does not make me understand how it works.
15:12 < pankaj> BCMM: I think I should see a good video on this.
15:13 < sadasaulna> pankaj: read a tutorial on "iproute2"
15:13 < BCMM> pankaj: unless you're talking about literally a shell script that calls ifconfig, the "old way" of configuring networks was distribution-specific
15:13 < sadasaulna> pankaj: also look at man pages for like ifconfig, iw
15:13 < BCMM> pankaj: /etc/network/interfaces is the way to manually configure networks on ubuntu, though
15:14 < BCMM> pankaj: on debian-based distros like ubuntu, networkmanager should ignore any interface that is specified in /etc/network/interfaces. you can also just remove nm once you're sure your network is working with your manual config.
15:14 < BCMM> https://wiki.debian.org/NetworkManager#Wired_Networks_are_Unmanaged
15:15 < pankaj> sadasaulna: ok
15:15 < pankaj> BCMM: OK
15:16 < BCMM> pankaj: and here's a sample of a manually-configured ethernet card so you can see if this looks low-level enough for you https://wiki.debian.org/NetworkConfiguration#Configuring_the_interface_manually
15:16 < sadasaulna> pankaj: and have a read up on iptables if you're interested in firewalls
15:21 < pankaj> BCMM: Very much thanks for that.
15:21 < sanjibukai> Hi.. Can anyone help me customizing my keyboard layout using xkb?
15:21 < sanjibukai> https://unix.stackexchange.com/questions/447253/how-to-have-multipl-level-3-modifier-using-xkb
15:21 < pankaj> sadasaulna: I am recording what you all guys are saying. I will surely read each topic clearly.
15:22 < sadasaulna> good stuff pankaj you will enjoy learning it, linux networking is fun
15:29 < fury> anyone familiar with hidraw? is there a way to block a write until the associated packet goes out the physical USB? i'm talking to a device that wants to be relatively precise about how often i can send certain messages, e.g. 500ms, and my current method results in failing to wait the full 500ms in some cases (i.e. at T0 I write() a packet, T+16ms the packet actually gets written to the device over USB, T+500ms I write() another)
15:29 < Zephranoid> networking in general in fun
15:30 < fury> i'd add an arbitrary +16ms delay onto my wait period except it isn't always 16 and it doesn't always happen, so i don't want to add extra delay where there doesn't have to be any
15:31 < kurahaupo> fury: so have you set O_SYNC on the filedescriptor?
15:32 < kurahaupo> (I would at least start there.)
15:32 < ttyX> what config option do I need to add/change for dnsmasq to allow queries from non-native subnet?
15:32 < ttyX> dns queries*
15:33 < fury> kurahaupo: oh god. i haven't set that. i'm going to feel like such an idiot if i haven't and it works. let me check. thanks :)
15:35 < lng> Hi! I have added user to a group, re-logged in a user, but group is not listed. `grep docker /etc/group` > docker:x:999:roman . What am I doing wrong?
15:36 < Zephranoid> lng: what command did you run to add the user to the group?
15:39 < CrazyTux> hello..
15:46 < CrazyTux> is there any way besides using tlp to enhance laptop's battery life on linux?
15:47 < mexx> shut it down ?
15:48 < lng> Zephranoid: usermod -aG docker $USER
15:48 < saltlake> Hi, does kwrite has any file browser support as leftpanel?
15:49 < lng> docker run hello-world
15:49 < lng> oops
15:51 < Zephranoid> lng: try seperating your -aG to -a -G
16:02 < fofalee> hi
16:02 < fofalee> there is a google chrome history file, how do I make sure it's always empty
16:02 < fofalee> there was a trick to always keep this file empty
16:02 < Zephranoid> CrazyTux: Powertop is an alternative, I believe you have to do the settings manually though
16:03 < CrazyTux> Zephranoid, ok
16:04 < ayecee> fofalee: create an empty file in its place and make the file immutable, maybe
16:04 < fofalee> immutable?
16:05 < mexx> chattr +i
16:05 < ayecee> yes, chattr +i
16:06 < fofalee> what chattr, why can't I just use chown to change the permission bits
16:06 < fofalee> so you are implying there are other bits too apart from permission bits...
16:06 < ayecee> it's bits all the way down
16:07 < CrazyTux> could anyone suggest a linux distro which can be used entirely through GUI tools. One should never have the necessity to use the terminal and cli.
16:08 < Trifecta> hi, i wanted to know if it is possible to create a zpool with glabel e.g. zpool create test raidz3 /dev/label/<generic label>
16:09 < fofalee> does find ~ -iname ledger ; find in the home directory recursively, when I use GUI I can open the chrome apps, or by the chromium browser, but without it directly from cli how do I open those chrome apps
16:09 < Trifecta> since the /dev/ada<x> values change when the system is restarted
16:10 < oiaohm> CrazyTux: windows with printer lock-ups you have to go to command line.   So its not only Linux where you have to use command line every now and again.
16:10 < triceratux> CrazyTux: linux lite 4.0 ftw https://sourceforge.net/projects/linuxlite/files/?source=navbar
16:10 < DoctorDick> If you use Linux, you're going to need to use the terminal
16:11 < fofalee> ayecee: what's the equivalent of chattr +i using permission bits chown
16:11 < ayecee> there is not an equivalent
16:12 < bipul> DoctorDick, I use linux graphically.
16:12 < DoctorDick> bipul: So do I
16:13 < Sitri> CrazyTux: whatever Mandriva is now.
16:13 < bipul> Humm..
16:15 < Zephranoid> fofalee: If you really want to avoid chattr just removing write permissions for everyone might work
16:17 < jhodrien> What are you actually trying to do?
16:20 < kuahara> I seriously don't know if this is at all #linux related other than that the problem we're having is occurring on a linux server and may or may not have anything to do with the OS.  Just looking for a thought or idea we haven't had yet that might get us pointed in the right direction.
16:20 < kuahara> We have case documents that get stored on a Centos server in .htm format.  They contain variables that are populated with case specific information when the user goes to print them.
16:20 < kuahara> Two counties in particular have a super frequently recurring issue where they will go to print these documents and heading will just be a bunch of garbage and various parts of the document aren't generated properly.  To fix this, I literally download the document from the server, open it in notepad++, save it without making any changes, then upload it back again, overwriting the existing copy.
16:20 < hendrix> mandriva... haven't heard that name for a long time..
16:20 < kuahara> That's it.  Everything works great again, but usually only for a day before someone else has to repeat that process.
16:21 < jhodrien> You need to diff it with the original copy, and stop just overwriting it.
16:23 < kuahara> going to hunt down a different document real quick that has this same problem.  What do you use to do the comparison?
16:23 < JimBuntu> kuahara, Are you able to audit the doc, verify no one else is making changes to it?
16:23 < kuahara> The contents of the document is not what is changing
16:23 < JimBuntu> kuahara, `man diff`
16:23 < kuahara> We know no one else is.  It's possible some automated task might be doing something.
16:23 < CrazyTux> Sitri, I didn't get you.
16:24 < CrazyTux> triceratux, are you still using Linux Lite?
16:24 < CrazyTux> triceratux, and what about MX 17?
16:26 < triceratux> CrazyTux: im running Lite 4.0 right now. its only the 3rd distro to ship a 18.04 lts version. all of them go down in flames over webmedia playback. probably ff60 is a little too fresh rofl
16:26 < Sitri> <CrazyTux> could anyone suggest a linux distro which can be used entirely through GUI tools. One should never have the necessity to use the terminal and cli.  <Sitri> CrazyTux: whatever Mandriva is now.
16:27 < Sitri> It's their main goal.
16:27 < triceratux> CrazyTux: mx-17 is still the reference standard. perfect team, perfect community, perfect branding, perfect base. shows what debian can be without bringing canonical into the mix :)
16:29 < CrazyTux> Sitri, you mean OpenMandriva?
16:30 < CrazyTux> Sitri, https://www.pcworld.com/article/2930369/mandriva-linux-is-dead-but-these-3-forked-distros-carry-on-its-legacy.html
16:30 < Sitri> Mageia was the one I was thinking of
16:30 < BCMM> i don't think an operating system you can use entirely without the command line exists at all, tbh
16:30 < CrazyTux> Sitri, ok. Of the three Mageia is what you suggest?
16:30 < BCMM> one you can use for a narrow set of everyday tasks without the command line, sure
16:31 < triceratux> BCMM: android ?
16:31 < CrazyTux> BCMM, ok
16:31 < jhodrien> Something with a crappy web panel front end?
16:31 < Sitri> I haven't used any.  Mandrake (what Mandriva used to be) is where I started.  I'm just giving another option to try
16:31 < BCMM> triceratux: hmm good point. should have said "general-purpose desktop OS" i guess
16:31 < Zephranoid> triceratux: I have a cli on my android
16:32 < BCMM> triceratux: also, on android i personally have to drop to a command line to get something done every few days
16:32 < jhodrien> You definitely don't *have* to use the CLI on Windows or Android.  Or indeed any old desktop linux distro.
16:33 < hendrix> you don't have to use cli on chrome os!
16:33 < BCMM> i dunno, seems like a couple of times a year, windows gets in a state where you just absolutely have to do ipconfig /release if you want things to start working again
16:33 < BCMM> for no reason whatsoever
16:33 < triceratux> Zephranoid: me too. ive got termux & material terminal on android. but i dont have to. i dont need them to administer the system
16:35 < mexx> l
16:35 < mexx> /c~.~.
16:35 < mexx> sorry
16:35 < triceratux> hrm what about a cli version of regedit that runs on the wsl using ubuntu & ncurses ? that would bring the linux cli firmly into the windows sysadmin mix
16:36 < CrazyTux> triceratux, your suggestion?
16:37 < djph> triceratux: maybe for the good ones ... but they seem to be pretty rare
16:39 < triceratux> CrazyTux: stick with mx-17. dont boot windows. learn your cli. run altlinux sisyphus xfce instead of mageia
16:39 < triceratux> standard stuff
16:39 < CrazyTux> ok
16:41 < jhodrien> BCMM: Although isn't ipconfig /release similar to telling the NIC to repair itself via the GUI?
16:41 < BCMM> jhodrien: it should be. somehow, sometimes, it isn't...
16:41 < BCMM> actually i have no idea what the repair connection thing in windows is supposed to do
16:41 < jhodrien> I find myself using netstat and find in a cmd prompt because task manager can't do what I need.
16:42 < BCMM> jhodrien: process explorer
16:43 < BCMM> i know i'm a mile offtopic now, but sysinternals procexp is basically would task manager would be if it wasn't dumbed-down
16:43 < autopsy> BCMM it deletes then reinitializes the connection and sets it's rout ing tables.
16:43 < BCMM> also answers questions like "which actual service is making services.exe eat CPU"
16:44 < Juesto> svchost not exactly services.exe
16:44 < Juesto> and yes, thread view too
16:45 < CrazyTux> triceratux, what is your opinion on OpenSuse Leap 15? which one would you recommend between MX 17.1 and OpenSuse Leap 15?
16:45 < JimBuntu> release should be "release and don't renew", where as repair should be "release and renew"
16:47 < triceratux> CrazyTux: if youre an enterprise customer planning on keeping a conventional install administered the new leap is the way to jump. im just a hacker running liveisos to keep my imap up so ill go with mx-17 every time. or tumbleweed rescue
16:48 < CrazyTux> triceratux, ok
16:49 < ananke> I highly doubt anybody in enterprise would seriously consider Leap
16:52 < Psi-Jack> heh
16:54 < Psi-Jack> opensuse has some serious initial config issues to content with as well, especially with the default openfiles limit on users.. Not even enough to maintain a modern day browser.
16:54 < Psi-Jack> contend*
16:58 < triceratux> CrazyTux: ok i misspoke then. just stick with mx-17 & tumblweed until the bugs get ironed out of leap ;)
16:58 < Psi-Jack> Tumbleweed has the same issue. :p
16:59 < CrazyTux> triceratux, for general purpose desktop use?
16:59 < Psi-Jack> As for CrazyTux ... Well, he's here every few days asking the same questions over and over and over again.
16:59 < Pentode> well he is a crazytux ;)
17:00 < Psi-Jack> Well, he's been doing this for the past year. Same pattern.
17:00 < triceratux> CrazyTux: nope, to keep an eye on opensuse labs. in fact, check out gecko if youre leaning that way. its essentially the only leap based linux http://geckolinux.github.io/
17:01 < CrazyTux> some new releases have happened. Based on Ubuntu 18.04. Just wanted to ask which one of them is more convenient.
17:02 < CrazyTux> triceratux, I think MX 17.1 is patched for Spectre and Meltdown now.
17:02 < fury> kurahaupo: it indeed magically worked when i set O_SYNC in the open command. i have no idea why my brain glossed over that particular bit. i was fixing to write my own replacement for the hiddev driver because i was sure it wasn't that simple :P thank you!
17:03 < kurahaupo> Glad to be of service 😙
17:03 < triceratux> CrazyTux: when lite 4.0 comes up systemd has to timeout 30sec over vboxadd.service waaaaa. so voyager 18.04 is the best of the lts lot so far. not impressive
17:03 < Psi-Jack> CrazyTux: Here
17:03 < Psi-Jack> CrazyTux: Here's the thing. Only YOU can answer that. Nobody else.
17:03 < fury> i almost did my own driver anyway until i figured out that's what the usbhid stuff does. i shall forever be a linux noob even though i've been messing around with it for 16 years :P
17:04 < CrazyTux> triceratux, ok
17:07 < triceratux> CrazyTux: right now im orbiting between siduction, mx-17, extix, swagarch, & the remaining 16.04 clones while im waiting for 18.04 to get in shape
17:08 < CrazyTux> triceratux, ok
17:09 < Psi-Jack> CrazyTux: When are you never in "orbit" around X*Y distros? Seriously.
17:09 < Psi-Jack> heh
17:09 < n-iCe> hi
17:12 < azarus> All of those distros are like... only made to look pretty and be easy for the user
17:15 < promach_> Could anyone help with https://unix.stackexchange.com/questions/447340/nautilus-file-manager-starts-only-after-an-hour-or-more ?
17:15 < CrazyTux> ok
17:17 < wizzi> for system engineering and security, what should i use...fedora,kali ...?
17:17 < ayecee> ubuntu
17:17 < JimBuntu> promach_, if this is an issue right after boot... have you tried something like `sudo killall nautilus` in shell, then try opening nautillus?
17:19 < promach_> JimBuntu: sudo killall nautilus    does not help even though it happens right after boot
17:20 < JimBuntu> promach_, so I understand, you kill nautilus... then when you tried to open nautilus, it is still hanging, correct?
17:20 < promach_> yes
17:20 < wizzi> for system engineering and security, what should i use...fedora,kali ...?
17:21 < JimBuntu> promach_, please disconnect from any networks (pull USB dongles, disconnect CAT) and reboot.
17:21 < promach_> I have no USB dongle, ethernet cable
17:21 < promach_> I have also tried rebooting before, also same issue right after boot
17:21 < JimBuntu> promach_, in-case I am forgetting to list something, it's safe to say that computer is on no network whatsoever?
17:22 < promach_> I am using Wifi on that computer to communicate with you NOW
17:22 < JimBuntu> Ok. Disable the wifi however you see fit and reboot and see if nautilus is more responsive right from boot.
17:23 < promach_> JimBuntu: ok
17:23 < promach_> see u later then
17:23 < atif5> Hi
17:23 < Psi-Jack> "you"
17:23 < atif5> Hi
17:23 < atif5> Hi
17:23 < atif5> Hi
17:23 < Psi-Jack> atif5: No need to flood it.
17:24 < triceratux> wizzi: whats wrong with original vanilla debian ?
17:26 < promach_> JimBuntu: I am just back from a reboot
17:26 < Psi-Jack> wizzi: Kali? Why is that even in your list?
17:27 < promach_> with WiFi OFF just now
17:27 < promach_> however, ....
17:27 < promach_> :(
17:27 < Samian> bitcoin whale comin through
17:27 < Samian> everyone bow at my presence
17:27 < promach_> JimBuntu: nautilus still does not become responsive
17:28 < wizzi> Psi-Jack so what should i use?
17:29 < Psi-Jack> wizzi: As I'd tell anyone. Whatever the heck YOU want.
17:30 < wizzi> Psi-Jack are all similar ?
17:30 < Psi-Jack> Linux is Linux is Linux.
17:30 < JimBuntu> promach_, Ok, thanks. So, we have roughly eliminated it being due to network connections attempting to be restored... What distro? Has it been like this since install or suddenly started?
17:31 < promach_> JimBuntu: arch linux
17:31 < Psi-Jack> However, Kali, is a specialized distribution, locked down, and designed for a single minded purpose, not one to be used casually.
17:31 < promach_> it just started becoming not very responsive in a week or two
17:31 < Psi-Jack> And you will get negative karma for kali usage. :p
17:33 < wizzi> Psi-Jack what do you suggest ?
17:33 < Psi-Jack> I suggest you choose for yourself.
17:33 < bashprogfortysix> wizzi: no use slackware linux, check out gnuworldorder.info use arch linux arch wiki docs are top of the line ~ tis the linux education for a good start then use ever distro once you got your base figured and stop asking this since i just told u..
17:34 < Psi-Jack> Use whatever YOU want to use. The only thing I'll recommend is the major distros. S
17:34 < JimBuntu> promach_, I see your trace log is pretty short, does it simply stop resonding for that long? Hmm, I'm not sure how much more help I will be. I would be trying things like removing any config files and cache, thinking about the other programs nautilus calls for things like image thumbnails... where it's starting by default (which directory)...
17:34 < JimBuntu> promach_, So, same thing if you do something wicked like start it from the command line with sudo?
17:35 < wizzi> Thnx all
17:35 < Psi-Jack> "Thanks"
17:35 < wizzi> Thanks* :D
17:36 < promach_> JimBuntu: I will post the full strace log when it finally becomes responsive after a long wait
17:36 < JimBuntu> Pedantic Silly Intellectual - Jack... that's how I will think of you now :)
17:37 < bashprogfortysix> also wizzi i recommending making a 20 g partition just for the rootfile system then the rest for home and maybe a partition for var then also a 1 gig swap space this should give u a good start and experiemnt while u have a home folder that you do not reformat while you learn things
17:38 < bashprogfortysix> at least note this mentally then u can know later on :)
17:39 < Psi-Jack> bashprogfortysix: "you"
17:39 < kurahaupo> Make two root partitions, so you can roll back when an upgrade goes sour. And swap might as well be 2~10× physical RAM, since it's cheap
17:39 < promach_> JimBuntu: see https://paste.ubuntu.com/p/y3QZFTkvcZ/
17:39 < wizzi> bashprogfortysix i did
17:39 < bashprogfortysix> oh word
17:40 < wizzi> bashprogfortysix and what do you suggest about system learning  ?
17:43 < RayTracer> wizzi: what's your goal? if you just want to learn without a specific direction, try to understand why the system booted that you just installed. ;)
17:43 < JimBuntu> promach_, I don't know how to properly interpret the results, but there looks to be gold in them hills. Sadly, I see others reporting the same issue, I have not stumbled across one where they also fixed it
17:44 < thatpythonguy> is it wrong if I copy + paste README.md's from someone's dotfiles on github
17:44 < bashprogfortysix> nothing i cant really tell you anymore of the aspect you just gotta do a 'full install' is really all ig ot now and just get use to run level three look up stuff like using a home theatre system, just get multiple boxes and learn how computer interact with each other, theres a lot of second hand computers for cheap so , look into ported linux video games not just ones that use 'wine'
17:45 < RayTracer> thatpythonguy: that depends on the licenses that apply to those files
17:45 < JimBuntu> promach_, `nautilus -n ~`
17:45 < Pentode> thatpythonguy, does it feel wrong? and how generic is the readme, and what license is involved?
17:46 < JimBuntu> promach_, or perhaps `nautilus -n --no-desktop ~`
17:46 < promach_> JimBuntu: same, still no response
17:46 < promach_> --no-desktop also does not help
17:47 < thatpythonguy> the readme's are pretty generic, but nicely worded + formatted. I don't see a License, looking now
17:47 < RayTracer> thatpythonguy: it's also always good style to mention where it's copied from if you copy it
17:47 < JimBuntu> Are there any dbus aficionados out there that want to take a look at promach_ paste?
17:49 < Pentode> thatpythonguy, typically whatever license applies to the software itself is going to apply to the rest of the data unless otherwise noted.
17:49 < Polylith> Hey everyone, I have a question. If you already have one physical partition for "/boot" that the main bootloader resides, can you have a logical partition where another "/boot" (for another distro) that the main bootloader can chainload to? If anybody knows, please mention my nick.
17:50 < jhodrien> As long as grub can read it, you can use it.
17:50 < promach_> JimBuntu: dbus ?
17:50 < Polylith> Hmm. But have you seen it work like that before?
17:50 < jhodrien> But if you're talking chainloading, then you can have as many as you like.
17:50 < jhodrien> You can chainload onto another disk.
17:50 < RayTracer> JimBuntu, promach_: cannot open display :0 sounds like you don't have DISPLAY (or XAUTHORITY) set correctly in sudo.. does "sudo xterm" work?
17:50 < JimBuntu> promach_, yeah, your issue seems to be related to nautilus failing to communicate
17:51 < jhodrien> You'd install grub onto the other partition, and chainload onto the second grub.
17:51 < jhodrien> It's not really different to chainloading onto a windows boot loader.
17:51 < Polylith> It's more that I want to use the current SSD I have, which I unfortunately formatted as MBR instead of GPT.
17:51 < JimBuntu> promach_, you may want to run the same trace, without the sudo and paste the results
17:53 < Pentode> Polylith, if you are unsure why not test it and see if it works?
17:53 < Pentode> you can always boot from something else and restore your original grub configuration
17:53 < promach_> RayTracer , JimBuntu: I am using wayland
17:54 < Dirkos> What is a good way to read a specific value from a config file? I have a file with: MY.HOST.IP = '172.12.20.222'
17:54 < Dirkos> How can i get the raw IP
17:54 < promach_> see https://paste.ubuntu.com/p/ndjfxt3b72/
17:54 < Polylith> Yeah, I will. But it also requires that Linux installers supports installation to logical partitions. Those I've seen so far may only install to physical partitions.
17:54 < uplime> Dirkos: you want just the ip value?
17:54 < RayTracer> promach_: then I'm out.. the usual troubleshooting I read on the channels on anything related to wayland is: try with X
17:54 < Dirkos> uplime: indeed
17:55 < JimBuntu> Sorry promach_ I don't have any Wayland experience :( That may help to explain the issue to some degree.
17:55 < Pentode> Polylith, is there a particular reason you want to do it that way?
17:56 < promach_> let me try in X later and see if nautilus could become responsive
17:56 < RayTracer> promach_: as sudo xterm fails as well, check "sudo printenv | grep -E 'DISPLAY|X'" - it tells DISPLAY=:0 and XAUTHORITY=/home/myuser/.Xauthority here and sudo xterm works
17:57 < Polylith> Pentode: As said, I made the mistake of using MBR instead of GPT. I currently have one physical "/boot" partition and an encrypted "/" partition which is used by Debian, and then I have another physical partition with Windows.
17:57 < Pentode> oh i see
17:57 < RayTracer> promach_: or take the approach to avoid using sudo, I didn't read what the details are of what you want to do/achieve
17:58 < Polylith> So I can only have one physical partition more or an extended partition where I can add logical partitions.
17:58 < Dirkos> uplime: have it till awk -F "=" '/VAGRANT.SERVER.IP/ {print $2}' .env
17:58 < Dirkos> but it gives me the quotes as well
17:58 < promach_> RayTracer: it returns DISPLAY=:0
17:58 < promach_> there is nothing on XAUTHORITY
17:59 < OtakuSenpai> /join ##ChatterZ
17:59 < promach_> RayTracer: see https://unix.stackexchange.com/questions/447340/nautilus-file-manager-starts-only-after-an-hour-or-more
17:59 < RayTracer> promach_: ok.. no idea if wayland needs something like XAUTHORITY, but X programs usually need that to authenticate to the X server
17:59 < OtakuSenpai> ok
18:05 < Happyhobo> ayecee:   You around?
18:05 < Happyhobo> Hi folks
18:05 < dgurney> hi
18:09 < RayTracer> Dirkos: the naming is bad in .env - usually you write something like VAGRANT_SERVER_IP='whatever' in .env, then in a script just ". .env" and you have the var defined
18:10 < RayTracer> but no vars with dots in bash.
18:10 < bipul> Hello
18:15 < Trifecta> hello i am trying to export a zfs pool, but it says it is busy constantly, i don't know what is making it busy. how can i identify what is accessing it? lsof | grep poolname doesn't give anyhting
18:18 < Psi-Jack> heh, zfs/
18:19 < Trifecta> Psi-Jack: I'm sorry ?
18:20 < Trifecta> ah got it
18:20 < Happyhobo> I'm going to have the world's goofiest notebook, two antennas mounted to the lid poking out the top.  It's going to be badass.  It's a shame I can't paint it green like an OLPC
18:20 < Trifecta> somehow the subdirectory in the volume was mnt even though mount didnt show it
18:20 < Psi-Jack> Huh?
18:20 < Trifecta> s/mnt/mounted
18:21 < Psi-Jack> Ahh, yes, English preferred.
18:21 < Psi-Jack> Despite it being ZFS. :)
18:21 < Trifecta> my bad :)
18:21 < Trifecta> i just noticed zfs list and mount do not respect each other
18:21 < Trifecta> like at all
18:21 < Psi-Jack> That's because well, ZFS,
18:21 < revolve> I've got a bunch of kernels installed but grub isn't booting from any newly defined defaults (defined via /etc/default/grub). Does anyone know of any other options for booting an alternative kernel?
18:22 < Yardanico> revolve did you regenerate grub config?
18:23 < revolve> I think so - I ran update-grub as root. Is there another step for regenerating the config?
18:23 < Yardanico> no, but update-grub should've worked
18:24 < revolve> Yardanico: thanks. any idea what might be enforcing the stock kernel?
18:27 < RayTracer> /boot/vmlinuz and initrd symlink maybe
18:29 < revolve> Thanks. There's no symlinks in /boot in this case though.
18:35 < bipul> Do chroot share the same interference?
18:36 < revolve> bipul: ?
18:36 < ayecee> it's a trap!
18:38 < bipul> I mean if we deploy chroot environment then i guess it uses the same network interference.
18:39 < revolve> oh, this probably pertains to a conversation from before i joined the channel, sorry
18:39 < bipul> I mean if we deploy chroot environment then i guess it uses the same network interference.
18:39 < RayTracer> what's a network interference?
18:40 < Trifecta> Is it possible to offline a zpool with a sinlge disk?
18:41 < bipul> Network interference you don't know? without of it we can not connect with network.
18:41 < Trifecta> I get the messag "no valid replicas" which is fine, since this is a pool to just replicate to and extract the disk when the files are needed
18:41  * RayTracer hits the trap
18:42 < bipul> Trifecta, What are you doing? be more specific?
18:43 < revolve> bipul: interference != interface
18:44 < bipul> Yes sorry typo mistake.
18:44 < Trifecta> bipul: basically, i have a zpool consisting of 10 disks in raidz3. the server has 4 additional disks, i want to replicate parts of the pool into the disks (works) be able to remove the disk and set a new one in so to have a backup of a part. this should be done periodically. In the end i have a part of the fileserver on one disk e.g. monthly backed
18:44 < Trifecta>  up
18:45 < Trifecta> I can replicate into these separate pools, but i am not able to set it offline, so i can remove the disk and pop a new one in and let the replication start over
18:46 < Trifecta> does it make sense?
18:46 < Sitri> bipul: by default chroots don't have access to /dev /sys or /proc, so they don't see the network interfaces at all.  A process could chroot itself after setting up network connections though (some daemons will do this, usualy ftpd's).
18:46 < bipul> Trifecta, And How your replicating it?
18:46 < Trifecta> bipul: I use a freenas local peridiodic snapshot which is then replicated to one of the local drives
18:46 < bipul> What do you mean by set it offline? which tools your using? And what are you replicating?
18:47 < bipul> So please ask at #freenas
18:47 < Trifecta> zpool offline a volume so i can remove the drive and pop a new one in
18:47 < Trifecta> bipul: i get no response at freenase
18:47 < Trifecta> and i dont solely use the gui
18:47 < Trifecta> it's supposed to be automatied
18:47 < bipul> Join the mailing list. Ask there
18:47 < Trifecta> am there
18:47 < Trifecta> no response or not in the scope of freenas solely
18:48 < Trifecta> it's not i tried every official outlet
18:48 < bipul> I 'm sorry i could not help you out. Since i have worked on Backup and disaster recovery but not on freenas
18:48 < Trifecta> well, if you are used in zfs i can still need some help
18:48 < Trifecta> basically i dont want to use the freenas function beside intially creating the volumes
18:49 < Trifecta> it's clean zfs
18:49 < bipul> Try asking inside #freeBSD there you will find minded people.
18:49 < Trifecta> i can't for example set zpool offline <Pool> label/<label> because it's just one disk
18:51 < Trifecta> bipul: I asked there a few days ago. but the channel seems completely dead
18:53 < bipul> Then hired me to work on your problem.
18:54 < Trifecta> bipul: would gladly accept, but i am not authorized to confirm/deny that
18:57 < elim_garak> I have an odd question. I have a locally networked hard drive that is accessed by a few people. Writing to it is very slow. Something like fast dot com will test my connection to the outside world and doesnt really ehlp here. How can i, via the command line, test to see what my speed is like connecting to and writing to this drive?
18:58 < bls> iotop, sar, iostat
19:11 < bipul> ayecee, Hello
19:15 < ziggylazer> If thé resolve.conf file gets a rred square around it. What does that mean?
19:16 < twainwek> what
19:17 < ziggylazer> Yes. The file has neem taged with a red box around it.
19:17 < bipul> share on pastebin what exactly are you seeing. ziggylazer
19:17 < ziggylazer> A very clear marking that something is wrong
19:17 < ziggylazer> Not on my machine.
19:18 < bipul> Where it's being marked?
19:18 < eb0t> hey elim_garak create a 1 gig file on the drive and then use time i think it is whilst you coppy the file fromt he device to your harddrive
19:18 < ziggylazer> We have no DNS. After installed OpenVPN
19:18 < ziggylazer> in /etc
19:19 < ziggylazer> Can it be some cert issue?
19:19 < eb0t> the time command will take how long it takes for the file to copy form the slow drive to your machine
19:19 < bipul> Yes, so please share it with http://pastebin.com/
19:19 < eb0t> then time copying from a fast drive
19:19 < eb0t> and you will have a good idea of speed
19:19 < twainwek> why pastebin.com? that's literally the worst pastebin out there right now
19:19 < ziggylazer> Ok the markings means that a link file is broken
19:20 < bipul> Then you could suggest something to him. I don't see anything worst in it.
19:20 < twainwek> 90% of the site is ads. The remaining 10% is 90% whitspace
19:21 < bipul> Oh okay, actually i never paid any attention to those ads, just stick with my work,
19:22 < twainwek> how about http://ix.io/ https://paste.debian.net/ https://paste.opensuse.org/
19:27 < bipul> Hi Sitri , I would like to know more about deploying separate network connectivity inside chroot
19:28 < Sitri> Use a VM
19:28 < ananke> bipul: chroot has nothing to do with networks
19:30 < prussian> network namespaces
19:30 < bipul> Well i would like to make a chroot type container, where i can assign a different network inside it.
19:30 < prussian> start learning linux network namespaces then
19:31 < bipul> prussian, Where to start?
19:31 < prussian> uh idk. man ip-netns?
19:33 < lukey> bipul: prussian Ehy not use lxc?
19:34 < ananke> bipul: you can try docker for example. however, there's no such thing as 'chroot type container'
19:34 < lukey> bipul: Why not lxc? *
19:35 < bipul> lukey, Actually i 'm trying to understand how things works inside chroot and how it's different from container, And is it possible to extend it like a container.
19:35 < sadasaulna> bipul: chroot is just change root - it does what it says on the tin
19:36 < bipul> Even i'm thinking to deploy some application like databases and all.
19:36 < sadasaulna> bipul: then use lxc or docker
19:36 < ananke> bipul: you shouldn't even think in terms of chroot. nor use it.
19:36 < bipul> I will but first  i would like to do an experiment with chroot.
19:37 < bls> you've picked a solution before you've identified what problem you want to solve
19:37 < sadasaulna> bipul: i can see how this experiment will go... "did it change the root?" "why yes it did".  experiment over.
19:38 < sadasaulna> bls is right, don't fit a problem to a solution
19:39 < bipul> Just to assign a network interface to my chroot, And i believe it's possible.
19:39 < Sitri> Only real reason to use chroot now is part of install or repair.  Containers or VMs are much better at anything else you'd use one for.
19:40 < sadasaulna> bipul: chroot has zero to do with networking
19:40 < bls> chroots are also useful in building packages
19:40 < Sitri> There are two options to networking in a chroot: 1) it has none 2) it uses the host's.
19:40 < bipul> let see.
19:40 < ananke> bipul: it's as if you kept ignoring everything people tell you
19:40 < bipul> Okay i agree.
19:43 < Psi-Jack> sadasaulna: Or systemd-nspawn. :)
19:43 < Dagmar> This is a bit like assigning an electrical outlet to your cat.
19:43 < bls> does this cat run at 110 or 220?
19:43 < JimBuntu> I assigned the electrical outlet that isn't hooked up, to my cat.
19:43 < Psi-Jack> piiiikaa chuuuuuuu!
19:44 < bls> and how do I mount a transformer to said cat for travel purposes?
19:44 < Dagmar> Precisely
19:44 < Dagmar> Nothing about it makes any sense
19:44 < JimBuntu> I suggest Bumble-Bee, every loves Bumble-Bee
19:44 < sadasaulna> my cats anus did not fit the DC jack, should I use a barrel connector type
19:45 < JimBuntu> over the line ;-)
19:45 < sadasaulna> and will i need a defibrilator accessory?
19:47 < Pentode> o_O
19:55 < Dagmar> ANyhoo, to be more explicit... chroot() is a call for radically changing the filesystem abstraction for a process.  It has zilch to do with networking.  There is SElinux and Linux capabilities for that sort of thing (i.e, altering rights to the network, per-process)
19:55 < Dagmar> It's _their job_
19:56 < Dagmar> Don't even try to use your cat to recharge your phone.  It won't go well.
19:56 < Dagmar> ...and if you can't tell it doesn't make sense, it's because you don't quite understand cats.
19:57 < lnnb> what if we can fasten some sort of static electricity collector on the cat, make it run around a carpet for a few hours with a phone plugged in?
19:57 < Dagmar> Nope
19:58 < Dagmar> Slightly changing the question doesn't make it suddenly not be the act of an unwell person
19:58 < sadasaulna> pfft lnnb everybody knows you attach buttered toast to the cat then dropping it, creating a perputal motion machine which you run a generator with to charge your phone
19:58 < Dagmar> heh
19:59 < TheWild> I just bound that old SD card to /home/thewild and my profile works on new laptop. Yay!
19:59 < TheWild> How the hell the battery could lose capacity from not being used?
19:59 < lnnb> sadasaulna: intereseting theory, lets fast track this one for peer review
20:00 < sadasaulna> :)
20:00 < Sitri> TheWild: chemisty.  The reaction that lets it store and release power breaks down over time.
20:01 < dgurney> in unrelated news, just ordered storage and a power supply for a new PC build
20:02 < TheWild> well, I didn't know it goes that fast. The battery has 6 months already, it was loaded once... maybe twice. And the capacity dropped from 100% to 65%.
20:03 < sadasaulna> did you store it dead? cos that is not good for them
20:03 < sadasaulna> ideally a lithium should sit around 80% for storage
20:03 < TheWild> it was fully loaded I think
20:04 < sadasaulna> yeah but things like some software switch might be gradually draining it when left for long periods
20:04 < sadasaulna> so best to pop the battery out of laptop after charging it if you're not using it for a good while
20:05 < HyP3r> Hello, does anyone use the Windows BCD bootloader to boot into Linux? I just follow this guide: https://pw999.wordpress.com/2013/12/21/windows-7-linux-dualboot-using-bcd-instead-of-grub/ Unfortunately, I have the Grub Bootloader directly in the MBR installed and overwritten by the windows installation. How can I get this again?
20:06 < HyP3r> The tutorial creates a backup of the Bootloader partition: sudo dd if=/dev/sda4 of=/Windows7/linux.bin bs=512 count=1
20:07 < Sitri> That dd command didn't grab the bootloader, unless GPT does something I'm unfamiliar with there.
20:07 < HyP3r> At the moment I just started a Live Boot CD and followed this tutorial: https://wiki.ubuntuusers.de/GRUB_2/Reparatur/#Reparatur-mittels-Desktop-CD
20:07 < eb0t> not sure but can you not just run grubmkconfig from linux ...and use something called osprober to make sure it can see the windws drive
20:08 < eb0t> prior to doing the grub mkconfig
20:08 < HyP3r> This manual also describes the boot loader: grub-install /dev/sdX
20:08 < Sitri> Bootloader isn't normally on any of the partitions, its in the space before them.  So if=/dev/sda is what you would've wanted.
20:08 < HyP3r> Where should I write the bootloader now?
20:09 < sbrothy> Hi! Sorry for my many "stupid" questions, but you're a goldmine of information. Let's say "ldconfig" is not installed by default, will the cache files it uses (e.g.: /etc/ld.so.cache, /etc/ld.so.conf) be there anyway?
20:09 < JimBuntu> sadasaulna, depends on the tech, last I was told by a manufacturer was store at 60%, worst thing for the battery when idle is fully charged and when in use is too high of a current draw.
20:09 < eb0t> install grub ...and the bootloader will be written to the mbr
20:09 < HyP3r> I thought I would write the bootloader somewhere again and then create a backup of it with dd again
20:10 < HyP3r> eb0t: But I want to use the Windows Bootloader
20:10 < eb0t> then do a grub mkconfig ...reboot and tada you should have grub bootloader on boot and it should have the windows and slinux
20:10 < Sitri> sbrothy: ldconfig is part of glibc, as is ld.so.  ld.so will read from those files even if ldconfig isn't present.
20:10 < eb0t> ah ok sorry i wouldnt think anyone would want the windows bootloader
20:10 < eb0t> thought you wanted the grub bootloader
20:10 < sadasaulna> JimBuntu: yeah, different opinions of the ideal storage percentage, but certainly flat is bad, and 100% charged not ideal.
20:10 < eb0t> ignore everything i said
20:10 < HyP3r> Sitri: but at this posistion is at the moment the Windows Bootloader
20:11 < HyP3r> eb0t: I want to boot into the Grub Bootloader FROM the Winodws Bootloader
20:11 < eb0t> ah not sure how to do that
20:11 < Sitri> HyP3r: so... currently it's using the windows bootloader... and you want to use the windows bootloader?
20:11 < Sitri> Ah.
20:11 < Muimi1> Is there a way to reboot my computer into memtest from ubuntu?
20:11 < Muimi1> I want to test the memory to make sure that the memory isn't broken
20:11 < HyP3r> Sitri: no I need this 512 Byte Block of PBR to boot from Windows Bootloader to GRUB Bootloader
20:12 < Sitri> Look up on how to install GRUB to a partition.
20:12 < Muimi1> When the computer starts, it loads so fast that I don't see grub
20:12 < Sitri> Muimi1: https://askubuntu.com/questions/574295/how-can-i-get-grub2-to-boot-a-different-option-only-on-the-next-boot
20:12 < k3rn3lpan1c> Pardon me if this has been discussed already. When using DHCP, dhclient usually rewrites resolv.conf and I know I don't have my "nameserver" pointing to my gateway IP including "search" directive to "lan". Where is this coming from and how can I modify this? Updating /etc/resolv.conf gets re-written after reboot.
20:12 < HyP3r> Sitri: well I'm actually trying that: grub-install /dev/sda4 (Bootloader Partition of my Linux Installation)
20:12 < sbrothy> Sitri: Thx.
20:13 < k3rn3lpan1c> I'm using Debain 9 stretch
20:13 < k3rn3lpan1c> Debian :)
20:13 < HyP3r> Sitri: but I get this error/warning: https://pastebin.com/bGuf1sQR
20:14 < Sitri> HyP3r: Right... but you should follow a guide to see if that's the correct procedure.  I don't use GRUB so I don't know how to advise you beyond that.
20:14 < HyP3r> Sitri: oh :(
20:15 < Muimi1> It doesn't say the answer is too generic.
20:15 < HyP3r> https://wiki.ubuntuusers.de/GRUB_2/Problembehebung/#Embedding-is-not-possible-GRUB-can-only-be-installed-in-this-setup-by-using-blocklists
20:15 < Muimi1> Sitri: how do you know which option to use?
20:15 < HyP3r> Maybe I have to use a ext partition instead of fat32?
20:15 < Sitri> Muimi1: read the config.  Because it depends on the config.
20:15 < Muimi1> but it is grub
20:16 < Muimi1> so i'll try to stop the computer at grub somehow if I can.  Is there a key to make it ffreeze at grub?
20:16 < Muimi1> like f11?
20:16 < Muimi1> The computer I'm dealing with has some issues like sometimes when I turn it on it just spins the fan really hot then turns off then turns on then off then on then off etc.
20:16 < Sitri> Yes.  I don't know the key, but all bootloaders have some key press you can use to force a menu to come up.
20:16 < Muimi1> (and so on)*
20:17 < Muimi1> And then after running the computer for like 5-20 minutes it starts to slow down a lot.
20:17 < Muimi1> I replaced the power supply, and that helped with other issues (spontaneous shutdown, locking up...)
20:17 < Sitri> Possibly overheating and thus thermal throttling
20:17 < BluesKaj> k3rn3lpan1c, are you trying for a static IP then you have to disable network-manager
20:18 < Bashing-om> Muimi1: EFI machine? , then grubs looks for the escape key . 3 second window of oportunity .. try spamming the escape key .
20:18 < Muimi1> I don't understand why the heck it skips grub
20:18 < Muimi1> that's retarded
20:19 < Muimi1> okay i'll try esc. thank you
20:19 < Pentode> HyP3r, if you want to use blocklists you must use ext3 or 4. then you can use --force to avoid that message.
20:19 < Pentode> why are you doing this anyway?
20:20 < geckos> Hi everybody. I need to tune linux for VoIP, I was told to use RT kernel, but I'm refusing to simply accept it, I read some RHEL tuning articles and no ones talk about RT kernels
20:20 < HyP3r> Pentode: I want to install the bootloader on a partition instead of the MBR because the MBR is allready in use with the windowws Bootloader :)
20:20 < Muimi1> Yay
20:20 < Pentode> why not just use grub to load both windows and linux?
20:21 < Muimi1> It's running
20:21 < geckos> as far as I read, the problem is network bounded, so I need to decreace network latency... using RT kernel from the begginning seems wrong to me
20:21 < Pentode> you are in for a world of hurt trying to make this work....
20:22 < HyP3r> Pentode: I want to test it... I know that this could be painful
20:22 < Pentode> oh ok
20:23 < Whiskey> wget https://sourceforge.net/projects/tcl/files/latest/download?source=typ_redirect don´t find the file
20:23 < Whiskey> Do you need to give it any parameters to cache it?
20:23 < Muimi1> So, if the memtest turns out okay, then what else could be the problem with that pc? Any ideas?
20:24 < Muimi1> After about 20 minutes, it starts to freeze up.
20:24 < Muimi1> It locks, and then it runs fine, and then it locks, and then it runs fine.  It locks up in like 5-15 second intervals
20:25 < SuperSeriousCat> Muimi1, try "wget https://downloads.sourceforge.net/project/tcl/Tcl/8.6.8/tcl8.6.8-src.tar.gz"
20:25 < Muimi1> I thought some process was using too many resources, and then I realized it even happens when I'm only using nano
20:26 < Muimi1> SuperSeriousCat: I'm waking up in 5 hours.
20:26 < HyP3r> Pentode: I have now created a ext4 parition where I want to install the grub bootloader but I stell get the warnings
20:26 < Whiskey> SuperSeriousCat: i was hoping it would find that itself
20:27 < Pentode> HyP3r, use try using --force
20:27 < Muimi1> SuperSeriousCat: is that a sripting language for linux?
20:27 < SuperSeriousCat> Ah. I highlightet wrong :p It was not to you, Muimi1
20:27 < Bashing-om> HyP3r: While installing the boot code to a partition is discouraged, it can be done. The manual: https://wiki.archlinux.org/index.php/GRUB#Install_to_partition_or_partitionless_disk .
20:30 < Whiskey> SuperSeriousCat: wget don´t manage to figure that out, strange if a webbrowser can but not wget :D
20:30  * Muimi1 wrestles with the linux gods
20:31 < SuperSeriousCat> Whiskey, I believe curl can follow and download redirects. wget is "dumb"
20:32 < phogg> Whiskey: wget figures it out for me, but gets the wrong file name
20:32 < phogg> which is not too surprising since the remote webserver doesn't properly annotate the file name
20:32 < SuperSeriousCat> "man wget" say it follow redirects. Probably some arguments needed
20:32 < HyP3r> Is there a way to switch sda7 and sda6?
20:33 < Whiskey> phogg: how do you use curl then neve ruse it
20:33 < Bashing-om> Muimi1: Overheating ? how about following the journal log and see if you get any hints ' journalctl -f ' .
20:34 < truthadjustr> can we extract the ipv4 address using the `ip` command? or do we have to parse it with grep etc!?
20:34 < phogg> Whiskey: I don't use curl much, just wget. If you look at the headers wget of your original URL redirects 3 times. The 2nd one respondss with the right Content-Disposition header to provide the right file name, but it is also a redirect and the final download of the actual file does not include any Content-Disposition
20:34 < phogg> wget *could* guess that a previously seen disposition for the same redirect chain should be used in that case but it does not
20:34 < phogg> truthadjustr: you have to parse it
20:35 < truthadjustr> damn...
20:40 < phogg> truthadjustr: start with: ip -4 -o a
20:41 < kerframil> truthadjustr: for instance, to get the primary IPv4 address for a given interface: read -r _ _ _ addr _ < <(ip -o -f inet a s dev lo); echo "$addr"
20:41 < kerframil> -f inet and -4 are equivalent
20:43 < phogg> I would use this for getting the address: IFS='/ ' read -r _ _ _ addr _ < <(ip -4 -o a s dev eth0 scope global)
20:43 < phogg> presuming you don't want localhost or the like
20:44 < kerframil> using scope is a good idea, yeah
20:44 < phogg> kerframil: if I wanted to do something like that but omit the interface name and obtain a list of IPs is there a way to do it with read -a or is a loop necessary?
20:45 < phogg> e.g. ip -4 -o a s scope global and have a list of each IP in the end
20:46 < kerframil> phogg: you'd need a loop. there's mapfile, but that would require the parsing to be done in the process substitution, which would defeat the (logical) way in which we use read here.
20:46 < phogg> thought so; too bad
20:48 < phogg> if only you could say something like read --table -a var and split rows by the -d and columns by IFS into a multidimensional array
20:52 < kerframil> multidimensional indexed arrays can be simulated with associative arrays, though it's kind of ugly
20:52 < phogg> this is definitely "just use Perl" territory
20:53 < kerframil> yes. though it does bother me that this problem remains so difficult to solve reliably - at least by way of parsing.
20:53 < phogg> kerframil: what, specifically obtaining the IP?
20:53 < kerframil> yeah
20:54 < phogg> you'd think that anyone writing a fancy new ifconfig replacement would make machine parseability a core requirement
20:54 < kerframil> one only needs to look at the evident shortcomings of some of the CPAN modules that attempt to do this
20:54 < kerframil> yes, I couldn't agree more
20:54 < truthadjustr> any way to intercept (ie get) the assigned ip addr of an interface? (in wpa_supplicant, wpa_cli)
20:54 < phogg> the best option is to use the C APIs directly, everything else sucks
20:54 < truthadjustr> i don't want to pull
20:54 < vim_user_19> hey! i'm using xfce4 and i'm trying to get the window tiling to work. if i put my terminal to the left half of the screen it doesn't fit the whole half, there is left a little space at the bottom. anyone know how to disable this behaviour and fill out half of the screen?
20:55 < phogg> the fact that UNIX hasn't traditionally had command line tools for returning basic networking information is part of how you know networking was not part of the original design
20:55 < kerframil> phogg: I forget the details just now, but there's a file in either proc or sysfs that has what I understand to be a dependable format. obviously, that would only cover Linux.
20:55 < phogg> kerframil: I've looked but never seen such in /proc. In /sys maybe somewhere, but getting lost is easy
20:56 < lnnb> where does this break, if you have multiple IP's available?  ip -f inet -o a | sed 's/.*inet //' | cut -d ' ' -f 1
20:56 < phogg> truthadjustr: you have to tell wpa_supplicant what you want to use from outside, it doesn't embed a scripting language
20:56 < phogg> lnnb: you can't tell which ones are important
20:57 < truthadjustr> phogg: it's a dhcp assigned ipaddress...
20:57 < truthadjustr> phogg: i need to able to know the ip addr
20:57 < truthadjustr> phogg: from ubuntu, thetre are scripts being invoked everytime the ip address changes
20:58 < phogg> lnnb: try this instead (way more efficient): while IFS='/ ' read -r _ _ _ ip _ ; do printf '%s\n' "$ip" ; done < <(ip -4 -o a s scope global)
20:58 < truthadjustr> i suspect, then.. it is coming from dhcp subsystem
20:58 < bn_work> hi, what is the recommended way to take a (preferably live) file-level snapshot of a root EXT4 FS on CentOS7 w/o having to go offline?  I've looked into Snapper but it seems to require btrfs or EXT4 on top of LVM, which is a bit difficult to change to after the fact..
20:58 < phogg> truthadjustr: you can get that information from udev, I believe. Just write a rule.
20:59 < phogg> truthadjustr: do you need to know "What is my IP?" or do you need to *detect when the IP changes*?
20:59 < truthadjustr> phogg: i need to know value of ipaddr
20:59 < lnnb> it's too confusing for me, might save a few dozen milliseconds though
21:00 < truthadjustr> real time... during when it changes.. so i don't need to pull...
21:00 < phogg> lnnb: vs. your version mine will not return the loopback IP (because I filtered it out with scope global) and omits the /8 or whatever.
21:00 < truthadjustr> phogg: will display it in my PS1
21:00 < phogg> truthadjustr: so you want to get it each time. Just write it to a file when the interface comes up; it won't change.
21:01 < phogg> lnnb: if you want *simple* use: ip -4 -o a s scope global | awk '{print $4}'
21:01 < truthadjustr> phogg: i'm moving on different networks
21:02 < truthadjustr> will print it in my PS1
21:02 < phogg> truthadjustr: so you do want an event when it changes... write a udev rule which writes the changed value to a file, then always read that fikle
21:02 < phogg> s/fikle/file/
21:02 < lnnb> phogg: i was also worried about ip command adding more indents for other addresses
21:02 < phogg> lnnb: man ip, look at -o
21:03 < lnnb> or maybe the devs one dday say oh no we're wasting three spaces here when two is more efficient!
21:03 < phogg> awk operates by *field*, not character offset, and -o assures a single line
21:03 < lnnb> we must correct this mistake!
21:03 < kerframil> doesn't matter
21:03 < kerframil> the default value of IFS splits on all forms of whitespace, with repeated instances effectively being folded
21:04 < kerframil> (help read)
21:04 < kerframil> so, as long as the fields aren't re-ordered ...
21:04 < lnnb> same with awk?
21:04 < kerframil> yes
21:04 < lnnb> neat
21:04 < phogg> kerframil: given that ip is part of util-linux and kernel devs are super conservative I expect that won't happen
21:05 < kerframil> phogg: I just remembered - there's /proc/net/if_inet6. I could have sworn there was something similar that covers inet addresses, but I'm drawing a blank right now.
21:05 < kerframil> phogg: in any case, you're right. using APIs is always the better thing.
21:05 < phogg> lnnb: in any case if you want to save typing you should use my while/read version and wrap it in a shell function
21:06 < phogg> kerframil: easy fix: just switch to ipv6, stop using v4. Problem solved?
21:06 < kerframil> phogg: :)
21:07 < lnnb> phogg: i will look into this read thing more
21:07  * rpgio holds breath until everyone starts using IPv6
21:07 < lnnb> RIP
21:08 < kerframil> lnnb: https://mywiki.wooledge.org/BashFAQ/001
21:10 < phogg> Yesterday I used tac in order to help get some actual work done. That's the first time I've seen it used for a non-toy purpose.
21:12 < nurdie_> A thumb tac?
21:12 < phogg> no; tac as in the opposite of cat
21:13 < nurdie> well that's enough IRC for me
21:13 < revel> I've used rev a whole lot more than tac.
21:18 < lnnb> i use `ls -lhat | tac` a lot
21:18 < esselfe> tac is from coreutils and rev from util-linux... I would have put both in coreutils
21:19 < kerframil> lnnb: there is -r
21:19 < lnnb> kerframil: heh learning lots today
21:21 < esselfe> actually I would merge util-linux into coreutils :P
21:21 < gronke> So I'm trying to write a script to run some commands in qiime2, but it looks like it has to be launched through conda, is there any way I can set it up globally so "qiime2" can be ran by all users without having to launch a conda environment? https://docs.qiime2.org/2018.4/install/native/#install-miniconda
21:22 < phogg> esselfe: the two packages have different maintainers so the choice is not arbitrary
21:23 < phogg> gronke: maybe, but you also do not need to
21:24 < gronke> phogg, why not?
21:24 < phogg> gronke: you can just write a wrapper script which first sources qiime2-2018.4 or whatever your env file is
21:24 < gronke> phogg, do you have a guide to that somewhere? is that like a bash script?
21:25 < phogg> gronke: http://paste.debian.net/plain/1027628
21:25 < kerframil> gronke: if you don't know what source does, run `help source`
21:26 < phogg> copy that file to some bin dir in your PATH and chmod a+x it, then you have a command which when run invokes your real binary in the right environment
21:26 < gronke> phogg, so that third line, that's the path to my conda install?
21:26 < phogg> gronke: it's the path to whatever command you normally run after you source the env file
21:26 < gronke> phogg, okay, thanks, I'll give it a shot
21:26 < phogg> gronke: if you do not know the path but know just the command name use: which yourcommand
21:27 < kerframil> also, if your sourced code sets up PATH appropriately, you may just be able to invoke conda without qualifying the path
21:27 < gronke> yeah it didnt :-/ I had to run conda from the install location
21:27 < kerframil> depending on exactly how the "environment" gets set up for this software
21:29 < Psi-Jack> felda: Multiple nick changes in large channels is just noice. Mind not doing that, please?
21:29 < Psi-Jack> noise*
21:29 < felda> sure sure sure
21:30 < felda> but the memes are important
21:30 < Psi-Jack> Much appreciated. :)
21:30 < twainwek> Psi-Jack: you can turn off nick change notifications
21:30 < gronke> phogg, so this command they have here, "conda env create -n qiime2-2018.4 --file qiime2-2018.4-py35-linux-conda.yml", does that need to be run each time, or is it set up for all users now?
21:30 < Psi-Jack> twainwek: I can /ignore you too. :)
21:30 < felda> surprised you didn't kick me for being in here with the pfsense nickname.  Technically it runs on FreeBSD >_>
21:30 < twainwek> Psi-Jack: i couldn't care less
21:30 < Psi-Jack> felda: You didn't feel that knife in your back earlier?
21:31 < uplime> whats wrong with changing your nick in a lot of channels?
21:31 < uplime> its not like they're doing it multiple times
21:31 < Psi-Jack> uplime: He was.
21:31 < felda> pfsense fears nothing (except Gonzo posting stupid drama)
21:31 < uplime> guess my client isn't showing me that then
21:31 < felda> I had really important story to tell tho
21:31 < ReScO> thank got for WWAN :/
21:31 < felda> have you ever lived with 20+ cats for 10 years doing all litterboxes each day?
21:32 < ReScO> god*
21:32 < felda> it's terrible
21:32 < ReScO> Damn ISP fucked their shit up
21:32 < dviola> watch your language
21:33 < gronke> my ears!
21:34 < Psi-Jack> twainwek: Besides. You already know my thoughts on that matter. So, next time you feel like stating that, bugger off instead.
21:35 < twainwek> i didn't even bother reading that completely
21:35 < sauvin> Looks to me like somebody needs to get a room.
21:35 < dviola> rofl
21:35 < ReScO> okay, my ISP went FUBAR and tooted, my internet is down until 11pm, has been down since 2pm, it's 9pm now and i'm thanking the LTE network right now
21:36 < Shazam1203> I am having a problem with centos getting a weird IP on boot. Anyone able to help?
21:36 < sauvin> What's a "weird IP"?
21:36 < Psi-Jack> Shazam1203: To help requires information.
21:36 < phogg> gronke: that just creates the file qiime2-2018.4. After that you don't need to run it as long as you have that file.
21:37 < gronke> phogg so here is what I have so far, I am not sure it's correct: https://paste.pound-python.org/show/WTCikiGafuch9pq35xW6/
21:38 < Shazam1203> <sauvin> It is not in any of the ifcfg-eth* files and I have no idea where the IP comes from.. but it is trying to get a 12.*.*.* IP on boot and then later in boot gets the correct IP at 172.16.192.*
21:38 < phogg> gronke: no
21:39 < sauvin> Got some kind of VPN thing going on?
21:39 < phogg> gronke: you should have an executable called qiime (per the docs you linked)
21:39 < phogg> gronke: it is *that* command you should be running on the last line
21:39 < phogg> gronke: you also need to list the full path to each file you are sourcing
21:39 < Shazam1203> <sauvin> no.. I cloned this VM from opennebula to esxi and now have this weird IP address showing up on boot
21:40 < gronke> phogg, oh, apparently that is just accessible by typing "qiime" after I run that source command, okay, thanks
21:40 < phogg> gronke: perhaps like this: to/real/binary/conda "$@"
21:40 < phogg> gronke: it *may* be accessible that way. After you run the source command do this: which qiime
21:41 < phogg> gronke: you will want the path which is emitted as the full path to use in the script. It's better not to rely on a PATH lookup for this so you can call your wrapper by the same name.
21:41 < gronke> yep I see it, thanks
21:42 < sadasaulna> Shazam1203: where exactly does it show up in boot?
21:44 < paul___> <sadasaulna> towards the beginning - Bringing up interface eth0:  Determining if ip address 41.43.233.233 is already in use for device eth0...
21:45 < bls> is this some weird client nick complete or are you guys having a conversation by just pasting what other people have written?
21:46 < Psi-Jack> Lol
21:47 < Shazam1203> <sadasaulna> and then the same line later in boot but with the correct IP. After boot there is no sign of this weird IP but it is messing up my NFS mounts and slowing boot process as it isnt a valid IP
21:47 < Psi-Jack> Shazam1203: How is it getting the correct IP?
21:48 < eb0t> Shazam1203: the original ip is a temporary ip that the interface is named before dhcp issues it with a proper ip
21:48 < Psi-Jack> Shazam1203: also addressing people is usually like so.
21:48 < eb0t> once the machiens is started and dhcp kicks in the machine will get a proper ip
21:48 < eb0t> dont worry about it
21:48 < gronke> phogg, so I just quit out of the sesssion and tried again, source activate doesn't see it, do I need to run the conda env command again?
21:49 < Shazam1203> eb0t: is there any way to make NFS shares mount after machine gets the correct IP?
21:49 < Psi-Jack> When you put nick in <>'s it looks more like you copy/pasted someone
21:49 < bls> Shazam1203: the convention you're using to refer to others <nick> is generally reserved for quoting what someone else has written, directing something to someone's attention is nick:
21:49 < Shazam1203> Psi-Jack: thanks.. lol new to irc
21:49 < eb0t> the filesystem is onthe same machine is it
21:50 < eb0t> if or do you mean samba fileserver
21:50 < Shazam1203> eb0t: yes it is now.. although I dont believe it was before as it was on opennebula kvm shared between two hosts
21:51 < Shazam1203> eb0t: I used vmware converter to convert the machine to esxi
21:51 < eb0t> ah ok...
21:51 < Psi-Jack> One question. Why esxi?
21:52 < kerframil> gronke: a distinct instance of bash is spawned to process your executable script. once it exits, its environment dies with it - if that's what you mean.
21:52 < Shazam1203> Psi-Jack: I took over as linux admin here and I prefer ESXI as it is easier to manager
21:53 < cluelessperson> hey there
21:53 < gronke> kerframil, so if another user is doing this they have to run the conda environment command each time to create it?
21:53 < kerframil> gronke: put another way, environmental changes don't propagate upwards through a process heirarchy
21:53 < cluelessperson> is there a way to load a file into ram so that you can increase read speeds for multiple processes?
21:54 < JimBuntu> cluelessperson, is your /tmp actually RAM?
21:54 < gronke> kerframil, just wondering the right command I need to run before I run "source activate," because it appears its not working at the moment
21:54 < kerframil> gronke: I don't know anything about this software, but if you must source some shell code in order to configure your shell environment appropriately, that must be carried out at least once in either your interactive shell or, where applicable, the non-interactive shell that is launched to run your script.
21:54 < gronke> kerframil, using this guide: https://docs.qiime2.org/2018.4/install/native/#install-miniconda
21:55 < gronke> I ran that 'conda env create -n' command, then the source activate command, it worked. then I quit out of the shell and want to run it again
21:55 < cluelessperson> JimBuntu: no?
21:55 < gronke> I'm trying to set up a bash script to run so this can be accessed by other users
21:56 < cluelessperson> gronke: does it require root?
21:56 < TheWild> "cp -an source/* destination"
21:56 < TheWild> Argument list too long
21:56 < TheWild> now what?
21:56 < kerframil> gronke: if you read `help source`, you'll see that source looks for the filenames to be sourced based on the value of $PATH, if the names are not fully-qualified. so it's possible that your conda env setup does actually modify PATH for the following source command to work as expected.
21:56 < bls> TheWild: use find or a for loop
21:56 < eb0t> try rsync TheWild
21:56 < gronke> kerframil, so can I specify a path for activate to get it to work?
21:57 < kerframil> gronke: in any case, if you alter the environment from within any given shell instance, it is gone once that instance exits.
21:57 < Psi-Jack> Shazam1203: Hmm, tis a shame.
21:57 < ayecee> ashes to ashes, bytes to bytes
21:57 < kerframil> gronke: in so far as source can be given fully qualified pathnames, yes
21:58 < gronke> kerframil, I'm guessing since actiavet says "no such file or directory" that that is the problem
21:58 < TheWild> okay, I used find and -exec argument
21:59 < JimBuntu> cluelessperson, Well, that's one way you can achieve something like what you are looking for. Basically set a line in your fstab to mount a tmpfs directory for storing that file. It will come with all the caveats of any RAM drive, but it will also come with the pros.
21:59 < kerframil> gronke: if the issue really is that PATH needs to be modified, you are free to observe any such change that is effected by conda and replicate it yourself - if that's what it takes.
22:00 < kerframil> gronke: the value of PATH is entirely at your discretion. you may export it at the top of your script, for instance.
22:00 < gronke> thanks!
22:01 < cluelessperson> JimBuntu: ah.
22:03 < Evidlo> I'm trying to debug /usr/lib/systemd/system/acpid.service.  `env -i acpid` works fine when I invoke it from the command line, but it doesnt work correctly when systemd starts it
22:06 < TheWild> omg, 33551 files in a directory :O
22:07 < Dagmar> Evidlo: You probably need to specify the full path to acpid
22:08 < bls> also, what do the log files say when it fails to start
22:08 < Evidlo> it's selinux.  I should have known
22:08 < Evidlo> I wonder how many hours of my life have been wasted because of selinux
22:09 < Dagmar> Only the ones you wasted from not learning how to deploy selinux rules
22:09 < bls> or disabling as part of installation if it provides you no value
22:11 < Evidlo> its not my fault that package maintainers are confused by selinux
22:11 < Dagmar> Hmm... let's see... which is probably more important to security... SELinux or systemd
22:11 < Dagmar> I'm actually 3/4 with you on that one
22:11 < Dagmar> They should have included rules for it in the acpid package
22:14  * JimBuntu hereby declares (without any authority at all) that if devs don't understand SELinux, it's Evidlo's fault.
22:21 < ayecee> *crickets*
22:24 < Dagmar> Nah.
22:24 < iHmD>  Hi, I am runing archlinux i3, my screen randomly freezes I can't type or move the mouse even after I unplug and replug it! same thing occured while installing gentoo from the DVD, is it software or hardware related problem and thank you!
22:24 < Dagmar> Maybe only 10% blame for not having seen that problem coming
22:24 < Dagmar> To say it's his "fault" is really pushing it
22:24 < Evidlo> so if selinux is in targeted mode, does that mean if I delete all references to acpid in /etc/selinux, then it should work?
22:25 < Psi-Jack> iHmD: Generally: hardware
22:25 < Psi-Jack> Evidlo: No
22:25 < iHmD> Psi-Jack: hhhmmm which one?
22:25 < Psi-Jack> iHmD: Pick one.
22:26 < Sitri> iHmD: likely hardware.  Run some diagnostics, like memtest and smartctl.  Try and see if you're overheating.
22:26 < iHmD> ok thank you guys!
22:37 < bn_work> so no one has any ideas regarding live snapshotting EXT4 systems?  I'm looking at rsnapshot now too, as snapper seems to need LVM or btrfs
22:38 < koala_man> bn_work: ext4 doesn't support snapshotting by itself. rsnapshot doesn't snapshot in the lvm/btrfs sense
22:38 < Psi-Jack> snapper does lvm snapshots now?
22:39 < bn_work> koala_man: what do you mean?
22:39 < Dominian> it does?
22:39 < Dominian> since when?
22:39 < Psi-Jack> Yeah, exactly. LOL\
22:39 < Dominian> I thought it was only btrfs
22:39 < Psi-Jack> Me too.
22:39 < koala_man> bn_work: rsnapshot just syncs files one by one. it doesn't try to be atomic
22:40 < Psi-Jack> bn_work: What do you /really/ want?
22:41 < Psi-Jack> Ugh, why does gpg have to be such a PITA?
22:41 < Psi-Jack> Non-stop. LOl
22:41 < triceratux> woo woo todays swagarch monthly has the new xorg [    78.428] X.Org X Server 1.20.0 X Protocol Version 11, Revision 0
22:42 < Psi-Jack> But, does swagarch actually have any good swag?
22:42 < bn_work> Psi-Jack / Dominian :  if you check the docs it says it supports btrfs, EXT4, and "LVM logical volumes with thin provisioning" (although I suspect EXT4 support needs to be on top of LVM, unless you have a modified kernel from the NEXT4 project)
22:42  * triceratux is only in it for the arch, not for the swag per se
22:42 < bn_work> psi-jack:  live snapshotting of root on EXT4
22:42 < Psi-Jack> bn_work: Why live?
22:42 < bn_work> koala_man:  ah, so it's much slower?
22:44 < Dominian> bn_work: ahh didn't realize that had been updated.. that's interesting
22:44 < bn_work> Psi-Jack: because rebooting to go offline to backup is kind of a pain with cloud boxes?
22:44 < Psi-Jack> bn_work: you don't need to do that in most cases.
22:44 < koala_man> bn_work: it's more that if you snapshot a dir full of IRC logs, they'll all end at exactly UTC 20:44. if you rsnapshot, the first log will end at 20:44 and the last will end at 20:57, because time passed between them. whether that matters to you is another question
22:44 < Psi-Jack> So, what is your real /actual/ goals, hence why I asked. :)
22:46 < bn_work> Psi-Jack:   ah, I see what you're saying, snapshotting a moving target.  I could live with going into single-user-target mode
22:46 < Psi-Jack> bn_work: You can backup most things live.
22:46 < bn_work> just don't really want to have to resort to booting into serial over IP or (G|L)ISH or so
22:46 < Psi-Jack> Technically everything can be, some not as reliably.
22:46 < bn_work> *shudder*
22:48 < bn_work> Psi-Jack:  I'm about to install a pretty big piece of software so I want to snapshot the system before I make that change (or any subsequent one potentially), Snapper seemed like the right tool but it seems to have the caveats mentioned earlier
22:48 < Psi-Jack> bn_work: I usually recommend borgbackup these days.
22:48 < bn_work> hence I'm looking at rsnapshot now
22:48 < Psi-Jack> It can do backups with compression, encryption, and deduplication.
22:49 < Psi-Jack> And is impressively fast.
22:49 < lukey> bn_work: If you just want snapshots (with easy rollback), then LVM is the way to go for ext4, just remember that snapshots are not an backup solution and if an snapshots runs full, you wil lose it
22:50 < Psi-Jack> Also, this is a "cloud" instance? Why not use the udnerlying backup solutions available?
22:51 < bn_work> lukey:  may be a bit of a pain to convert to LVM now after the fact.  what do you mean by "lose it"?
22:51 < Psi-Jack> Literally, lose it.
22:51 < Psi-Jack> The snapshots, of course.
22:52 < bn_work> Psi-Jack:  I've already looked into that, it's overkill for I'm trying to do and not granular enough
22:52  * Psi-Jack scratches his head and shrugs.
22:52 < koala_man> I use GCE's snapshot feature before large upgrades. It's very convenient
22:53 < bn_work> Psi-Jack: right, obviously :)  I meant like *how* do you lose it?  because it may eventually overwrite the old ones?
22:53  * bn_work reads up about borg
22:53 < Psi-Jack> borg is nice. :)
22:54 < Psi-Jack> You will be assimilated.
22:54 < bn_work> koala_man: Google Compute Engine snapshot?
22:55 < Psi-Jack> I've got 3 months of backups of my whole desktop system, including / and  /home, and it's still only 35GB
22:55 < Psi-Jack> That also includes my laptop. :)
22:56 < koala_man> bn_work: yes. since my box is hosted on GCE, I can just trigger a snapshot from bash or web. from the machine point of view, it's just running ext4 straight on a partition
22:56 < Psi-Jack> I might have to take a look at GCE some day.
22:57 < koala_man> I did it because I worked there at the time and got a hefty discount. Now I just do it out of habit
22:57 < bn_work> koala_man: does that only work with stuff in Google Cloud?  I'm on a different cloud provider...
22:57 < koala_man> bn_work: it only works with GCE hosted things, yes. the point is that your cloud provider may have similar features and they're pretty convenient
22:58 < Psi-Jack> bn_work: Who is your provider?
22:58 < bn_work> they do, but it's like an entire backup solution that one needs to tack on a monthly charge for
22:58 < koala_man> blerk
22:58 < bn_work> it's overkill for me
22:58 < Psi-Jack> Should've gone with AWS or similar. :)
22:58 < toogley> hm.. when i do ssh -t $server "tmux attach", i cannot type german umlauts. however, when i normally ssh into $server (such, that i receive a shell), and start "tmux attach" afterwards, my umlauts work again. however, ssh -t $server and starting tmux attach inside that also makes my umlauts work. any idea what i could do?
22:58 < bn_work> or at least for this scenario of what I'm looking to do
22:58 < Psi-Jack> AWS snapshots are differential backups and costs pennies.
22:59 < lukey> bn_work: If you have no backups anyways I guess it's not important ;)
22:59 < bn_work> yeah, we have some stuff on AWS too, I'm debating rebuilding on that *sigh*.. because they have some other benefits (ie: ENI, etc.)
22:59 < bn_work> but I like the simplicity of this cloud provider (Linode)
23:00 < Psi-Jack> bn_work: Yep. AWS is pretty awesome, especially if you know how to use it.
23:00 < Psi-Jack> Linode, ewww.
23:00 < Psi-Jack> Just eww...
23:00 < Exagone313> Hi, it seems that the page cache fills my whole RAM when running some applications that do a lot of disk reading and my whole computer is slowed down until I reboot.  Is there anything I can do about this (except buying more RAM)?  Using latest mainline.  Thanks for your help.
23:00 < bn_work> lukey: it's not really, it's a test box, but the software being installed is a beast and I just want to be able to rollback if it gets f-ed up or misconfigured
23:00 < Psi-Jack> Linode refuses to even implement private networking, because,,, Why should they?
23:00 < bn_work> lukey: BTW, what did you mean earlier?
23:01 < Psi-Jack> Anyway. See ya. Headin home from ye olde grindstones.
23:01 < Exagone313> I didn't that that issue with my previous CPU with the same amount of RAM (8GB)
23:01 < toogley> my TERM is btw. tmux-256color
23:01 < bn_work> Psi-Jack: AWS is kind of overly-complex though, and don't get me started on their pricing model
23:01 < Exagone313> have that*
23:01 < bls> what kind of setup is this that a test environment is too high risk and needs to be backed up?
23:02 < TheWild> what's the easy way to fix Windows line endings in place?
23:02 < bn_work> Psi-Jack: I think you can have private networking with Linode
23:02 < bls> TheWild: dos2unix
23:02 < TheWild> gonna check soon, thanks bls.
23:02 < bn_work> Psi-Jack:  does borg have a Freenode channel?
23:03 < bls> bn_work: /msg alis list yoursearchtermhere
23:03 < TheWild> I copied some git repos from my old dead Windows partition. Updating them all through automated scripts, and git won't touch some, claiming I have uncommited changes.
23:03 < TheWild> I suspect the line endings.
23:03 < bn_work> bls: I'm trying to avoid stumbling into Star Trek channels :)
23:03 < bls> could also be filename case sensitivity depending on which version of windows it is
23:04 < bn_work> oh here we go:  #borgbackup :)
23:09 < TheWild> bls: well, I did "find -type f -exec dos2unix {} \;". Now let's check that git repos didn't get corrupted.
23:09 < lukey> bn_work: Well LVM-snapshots happen at the block level, so to fit an full snapshot of your filesystem, it would need double the space to store it. You can make a snapshot with less than double the space, because LVM only duplicates the blocks that are actually changing, but if there's no space avivable anymore, the snapshot can't keep up with the changing blocks.
23:09 < kerframil> toogley: when you specify a command to ssh, the remote shell is launched as a non-interactive one. almost certainly, the explanation is that the resulting environment is different and that the LANG (and/or LC_*) env vars are not what they should be prior to "tmux attach" being executed.
23:11 < bn_work> lukey:  you're differentiating between a "full snapshot" and "snapshot"?
23:12 < kerframil> toogley: to verify this possibility, try: ssh ... -t 'LANG=de_DE.UTF-8; tmux a'
23:12 < kerframil> toogley: rather: -t 'export LANG=de_DE.UTF-8; tmux a'
23:12 < bn_work> lukey: but if I only want a single snapshot  in time, I should be ok, no?
23:13 < Dagmar> LVM snapshots are generally meant to be temporary unless you want to keep your volume group only 50% allocated
23:14 < kerframil> toogley: in that case, a potential workaround is to add "AcceptEnv LANG LC_*" to your sshd_config and - similarly - to ensure that your ssh client is sending these variables.
23:16 < lukey> bn_work: http://tldp.org/HOWTO/LVM-HOWTO/snapshotintro.html
23:19 < vlt> Hello. What is one of the smaller install iso files that I can put on a USB drive that will install a minimal linux system including GRUB to an empty hard disk?
23:20 < bls> vlt: debian netinstall?
23:20 < bn_work> lukey: thanks, the problem is converting an already existing box with an EXT4 partition may be a pain
23:21 < vlt> bls: I'm looking for something that works offline.
23:21 < bls> heh, what's your real goal then, because your two requirements appear contradictory
23:21 < bn_work> vlt: alpine-linux is pretty small, not sure if they have an iso though
23:22 < toogley> kerframil: lol, that is really crazy. my locale LANG is set to en_US.UTF-8 (and has always been), and still ssh -t $server  'export LANG=en_US.UTF-8; tmux attach' works
23:23 < kerframil> toogley: it's not crazy if you consider that openssh, by default, does not allow for any propagation of the clients environment to the server. and then there is the issue of the shell behaving differently depending on whether it's a login/interactive/non-interactive shell.
23:23 < bls> toogley: that's because that's set when you execute a login shell. when you give ssh a command to run, it does that instead of launching said login shell
23:24 < lukey> bn_work: Oh ok, I recomend to use LVM for every linux-installation as it makes many things easier
23:24 < kerframil> toogley: see the INVOCATION section of the bash manpage for the gory details
23:24 < toogley> kerframil: bls ah, makes sense
23:24 < toogley> kerframil: and bls: thanks
23:24 < vlt> bls: My real goal is what I failed to describe properly: I have an empty notebook, a usb drive and a very bad internet connection. I want to download an iso that will boot and then install linux to the target machine.
23:25 < kerframil> toogley: either way, your shell is being launched, but it's done so in a different way if you specify a command parameter to ssh
23:25 < lukey> vlt: Many distros have netinstall-ISOs
23:25 < ioerror88> in irssi is there a way to replay log entries which triggered highlight (by saying my name)
23:25 < ioerror88> I forgot i left IRC open :)
23:25 < kerframil> toogley: try ssh -t 'env', then compare it to what env reports if you just ssh in 'normally'
23:25 < bls> lukey: he wants as small of an ISO as possible that requires no internet connection
23:26 < bn_work> lukey: yeah, I unfortunately I didn't realize Linode's images didn't use LVM :/  apparently it's incompatible with their backup solution :(
23:26 < toogley> kerframil: aah, interesting
23:27 < kerframil> toogley: the lack of a sane LC_CTYPE is probably what specifically triggered your issue
23:27 < vlt> lukey: But netinstall means it needs a network connection, right?
23:27 < kerframil> toogley: (LC_CTYPE will inherit from LANG if not explicitly defined, so an empty LANG would also be a culprit)
23:27 < BSODjunkie> I installed Ubuntu 18.04 on an external SSD that I boot from, but it feels a little "laggy" compared to my Win10, what troubleshooting / stuff do I need to do?
23:27 < djph> BSODjunkie: not use USB
23:28 < BSODjunkie> djph: It's through USB3.1 I would have thought it wouldn't be this laggy :(
23:28 < djph> usb is slow :)
23:28 < BSODjunkie> djph: Well if intel weren't so annoying about TB3 I could have a direct PCIE connection :(
23:28 < djph> well, usb3 is snappy, but its still not as snappy as sata6GB/S
23:29 < BSODjunkie> speeds are fine it's a latency issue :(
23:31 < djph> usb is all channeled thru the CPU (errr, i think so still).  not as snappy as other things
23:31 < kerframil> toogley: in summary, you could try something like ssh -t '. /etc/profile; tmux attach' so that the behaviour of a login shell is approximated. or you could set things up so that your client's LANG variable is respected by the server.
23:34 < BSODjunkie> What is the Linux equivalent of CrystalDiskMark that I can run in 1 - 2 mins
23:34 < _KaszpiR_> fio
23:35 < _KaszpiR_> https://github.com/axboe/fio/blob/master/examples/fio-rand-RW.job
23:35 < BSODjunkie> is it part of the phoronix-test-suite?
23:35 < _KaszpiR_> I don't remember, maybe
23:35 < _KaszpiR_> but phoronix is pretty big
23:35 < BSODjunkie> okay, I'm really knew I don't know how to run anything haha. What performance optimizations do I need to do to a new linux install. I have 0 drivers
23:36 < _KaszpiR_> none
23:36 < _KaszpiR_> just use it.
23:36 < _KaszpiR_> it works in 95% of use cases
23:36 < BSODjunkie> Dang it feels quite slow
23:36 < _KaszpiR_> usually disks are the slowest part
23:37 < _KaszpiR_> also by default linux cares about disk writes so that they are safe
23:37 < BSODjunkie> Is it a bad sign that ubuntu takes like 3 seconds after login to get to desktop, my install feels really buggy, is it usually instant?
23:38 < _KaszpiR_> 3s? that's normal
23:38 < _KaszpiR_> mine takes about 10
23:38 < kerframil> BSODjunkie: for a quick filesystem benchmark, `iozone -a` is pretty good
23:38 < _KaszpiR_> I have a lot of shit underneath
23:38 < michael2> network is usually slower than disks
23:38 < michael2> just saying...
23:39 < _KaszpiR_> especially on 10Mbit card ;D
23:39 < BSODjunkie> I had to add "nomodeset" on e for it to even boot, is that okay or was that a bad fix
23:40 < BSODjunkie> kerframil: thanks, I'll use it I think it's part of the benchmark suite thin
23:41 < _KaszpiR_> Dondald Knuth would disagree ;)
23:42 < BSODjunkie> Okay 1 last question, does ubuntu use "themes" or how do people customize it? What's a good resource for a beginner to learn this stuff
23:43 < Psi-Jack> bn_work: You will be assimilated.
23:43 < Psi-Jack> :)
23:44 < wilbert> Does anyone know the magic incantation I should put in Makefile.am to build a library without having to link it to anything?
23:44 < bn_work> Psi-Jack: ha, hopefully this is the droid, err... solution I'm looking for :)
23:44 < Psi-Jack> Silly humanoid. The Borg are not droids.
23:45 < BSODjunkie> We are the Borg, Help me figure out how to theme ubuntu
23:45 < Psi-Jack> Just like you'd theme ANY other distro. :p
23:46 < BSODjunkie> I've never used linux before, it feels kind of lonely without the NSA to talk to
23:46 < djph> with a lobotomy and the terminal?
23:46 < Psi-Jack> NSA?
23:47 < maxxe> BSODjunkie, sorry, but what do NSA have to do with your choosing a theme?
23:47 < maxxe> *you
23:47 < BSODjunkie> maxxe: It was a reply to Psi-Jack saying to use any distro, it was my way of saying ubuntu is my first distro
23:47 < Psi-Jack> Lack of good funny detected.
23:48 < BSODjunkie> I did preface it with we are the borg, not the funniest chaps
23:48 < Shazam1203> any ideas how to make nfs mount after network is availiable?
23:48 < Psi-Jack> Shazam1203: Reboot.
23:48 < djph> Psi-Jack: i thunk he means he was a windows user
23:48  * Psi-Jack prepares an hourglass.
23:49 < djph> s/thunk/think/
23:49 < Shazam1203> Psi-Jack: same problem as before... when I reboot nfs shares try to mount before valid IP address is obtained
23:49 < djph> Psi-Jack: bourbon will be quicker.
23:49 < Psi-Jack> Ahh.
23:49 < Psi-Jack> Shazam1203: Fix the real problem, not try to bandaid the problem.
23:51 < Shazam1203> Psi-Jack: but how to fix the real problem :P .. I have looked through sysconfig scripts and rc scripts and cannot find any reason why it is getting a non valid IP before mounting nfs
23:51 < BSODjunkie> On Windows, my SSD required 2 different drivers (Intel Chipset and a Thunderbolt driver) to work at full speed, does that mean on linux I can never have it at full speed?
23:51 < Psi-Jack> Well, you're the engineer that decided to convert an existing, what was it, gnebula or something, to ESXi? YOu know what you're doing, right? ;)
23:52 < Psi-Jack> BSODjunkie: That makes no sense.
23:52 < BSODjunkie> Psi-Jack: my write speed was 100mb, I installed intel chipset and thunderbolt driver and then it hit 420mbps. not sure which one it was.
23:52 < Psi-Jack> Intel chipset, I can understand. Linux already has that generally native to the kernel. Thunderbolt, however... Where does that even come into play at all?
23:53 < Psi-Jack> Mbps, you mean?
23:53 < Shazam1203> Psi-Jack: It was essentially a linux kvm hypervisor on top of centos 4.5.... Weird thing is the other vms that are still on that hypervisor dont have this issue.. just the ones I have cloned
23:53 < BSODjunkie> Psi-Jack: I connect the SSD through my thunderbolt port (USB-C)
23:53 < qrvpzvb> do I specifically have to enable iommu in the kernel line?
23:53 < Psi-Jack> Shazam1203: Yes, and kvm is definitely far far better than vmware. :)
23:53 < Psi-Jack> BSODjunkie: Oh... Why?
23:54 < Shazam1203> Psi-Jack: maybe so.. but easier to manage and replicate with veeam on esxi
23:54 < BSODjunkie> Psi-Jack: Not sure which part you're asking why. Because USB3.1 > USB3.0?
23:54 < Psi-Jack> No, why USB at all?
23:55 < BSODjunkie> Psi-Jack: My laptop has only 1 drive, which has windows installed (which I need). I didn't like virtualmachines so I wanted to see if it was a better experience having ubuntu installed on my external SSD, so far it's not going great.
23:55 < Psi-Jack> Shazam1203: Not really. Proxmox VE is a fine example of good you can do.
23:56 < Psi-Jack> BSODjunkie: Why external? LOL
23:56 < rebelmoon> Why can't you partition the drive and do a dual boot?
23:56 < Dagmar> If that's not a USB3 drive on a USB3 bus, the performance will be underwhelming
23:56 < BSODjunkie> Psi-Jack: I'm not sure what you mean, it only has 1 drive slot, how would I make it internal
23:56 < BSODjunkie> rebelmoon: I could but wanted to try and avoid having to split up my drive and lose more space
23:57 < Psi-Jack> BSODjunkie: A computer... With only 1 drive slot? wut?
23:57 < Shazam1203> Psi-Jack: true.. proxmox is very good. Maybe this issue is from remnants of an old network adapter?? Too bad kudzu is not in centos 6...
23:57 < BSODjunkie> Dagmar: It's an M.2 SSD, through USB3.1 connector connected to Thunderbolt 3 (USB-C)
23:57 < Dagmar> BSODjunkie: You realize that Linux will do pretty well in just 16-24Gb of space, right?
23:57 < BSODjunkie> Psi-Jack: A laptop, most laptops only have 1 drive
23:57 < Dagmar> Actually you can go rather lower than that
23:57 < Psi-Jack> Shazam1203: Couldn't say, but something wierd is definitely going on. Does that 12.x network have any resemblances from the previous hypervisor at all, or any relation to anything?
23:58 < Dagmar> I wouldn't really try to go below 12Gb without a capacity test first
23:58 < Psi-Jack> Ahh, laptop.
23:58 < BSODjunkie> Dagmar: I heard some nightmare stuff with getting my laptop to dual boot, but I guess it could be a better option
23:58 < Psi-Jack> Why didn't you just say so from the start? heh
23:58 < BSODjunkie> Psi-Jack: yea sorry, I can't really have a desktop I move so much
23:59 < BSODjunkie> Dagmar: I'm very new and don't really want to start partitioning my drive without having some ubuntu experience
23:59 < kerframil> qrvpzvb: if you're referring to intel_iommu, it depends on whether the kernel was built with INTEL_IOMMU_DEFAULT=y.
23:59 < Shazam1203> Psi-Jack: Thats the strange thing... I have no clue where this 12x subnet is coming from.. I cant find anything on the existing kvm vms... the only subnet we have here is 172x.. ifconfig on existing vms just shows the one adapter with the correct IP
23:59 < kerframil> qrvpzvb: if it wasn't then yes
23:59 < BSODjunkie> Dagmar: And I think that was a good thing, considering when installing ubuntu it had a big issue with my SSD being in 512byte blocks
23:59 < Psi-Jack> Shazam1203: Well, 12.x would be a public IP.
23:59 < Oxyz> using sed or similar... how do I replace a value between two strings??  i.e if I want to replace the number 232 with $NB  ?      "lst";a:232:   so like
23:59 < Dagmar> BSODjunkie: You misread something then
23:59 < qrvpzvb> do I have to enable iommu in the kernel boot line?
--- Log closed Sat Jun 02 00:00:06 2018