--- Log opened Wed Apr 25 00:00:22 2018
--- Day changed Wed Apr 25 2018
00:00 < Lope> Psi-Jack: thanks but that's not a great search. I prefer `acs linux-image`
00:00 < notmike> Always include the grep
00:00 < notmike> So important
00:00 < Lope> then I grep the version number I'm interested to get the extra and so on for that version
00:01 < Lope> Basically I can install older kernels, not newer. Which is fine.
00:01 < Lope> As long as it works.
00:02 < Psi-Jack> You should be able to install up to 4.10, as from 17.10, since they backport newer kernels from newer releases to LTS.
00:03 < Psi-Jack> LTS is the only one that do such things with.
00:03 < Psi-Jack> That and the x server stack as well.
00:03 < Lope> Psi-Jack: I see I've already got backport stuff loaded
00:03 < Psi-Jack> But, on that note, I'm heading home.
00:03 < Lope> I'm running 4.13 kernel which is quite new probably.
00:03 < Lope> So I'm going down to 4.11
00:03 < Lope> Cool, safe travels
00:03 < Lope> Thanks
00:06 < TyrfingMjolnir> Psi-Jack: Thanks! http://download.celtx.com/source/celtx-2-9-5-src.tar.bz2
00:06 < _KaszpiR_> Lope https://github.com/magnific0/wondershaper/issues/17
00:07 < Lope> _KaszpiR_: Aha! that's my shithouse kernel.
00:07 < Lope> I'm going to reboot to 4.11 kernel quick. BRB
00:08 < TyrfingMjolnir> Downloads/celtx-2-9-1-src.tar.bz2
00:08 < TyrfingMjolnir> Downloads/celtx-2-9-5-src.tar.bz2
00:08 < TyrfingMjolnir> Downloads/celtx-2-9-6-src.tar.bz2
00:08 < TyrfingMjolnir> Downloads/celtx-2-9-7-src.tar.bz2
00:09 < dviola> cool story
00:09 < Lope> _KaszpiR_: LOL, i just read more carefully. I see the old version of wondershaper that shipped with distros is broken, not the kernel (it looks like)
00:09 < Johnjay_> does anybody know why wget has to have --no-check-certificate for everything?
00:09 < Lope> Anyway, I'm going to reboot to the latest kernel quickly.
00:10 < _KaszpiR_> I suiggest trying on vms
00:10 < nbah> hello
00:10 < nbah> how to install missing recommended pkgs from already installed one?
00:11 < mawk> I heard wondershaper is outdated Loshki
00:11 < mawk> lope*
00:11 < mawk> which isn't here
00:12 < debkad> nbah: there are difference between "missing" pkgs and "recommended" one
00:13 < nbah> How to install "missing" RECOMMENDED pkgs from already installed one?
00:14 < mawk> copy paste output of apt show in apt install ?
00:14 < nbah> such lack of words happening now
00:14 < debkad> ._.'
00:14 < nbah> all right
00:17 < nbah> well, the case is, i can't hp-setup -i *net* option. Returns: 'error: Failed to delete None Print/Fax queue. Error : client-error-not-found'. Tried hp-setup -i <ipaddress> *unleased one.
00:17 < nbah> actually not this error message
00:18 < nbah> 'error: No device selected/specified or that supports this functionality.'  better.
00:18 < nbah> attempt some search engine... couldn't find relevant results.
00:19 < nbah> maybe it is the lack of words issue
00:25 < luxio> Is it possible to burn in subtitles using ffmpeg?
00:25 < debkad> yes
00:26 < debkad> luxio: https://trac.ffmpeg.org/wiki/HowToBurnSubtitlesIntoVideo
00:26 < nbah> hello?
00:27 < debkad> nbah: from where you get that *net* in your command?
00:27 < ayee> 4.15.0-19-generic .. I'm using the latest version of ubuntu, this is in a vagrant VM. for some reason the hidraw kernel module doesn't exist anywhere. Is this module deprecated?
00:30  * debkad ==> bed
00:32 < Lope> if I systemd-nspawn a container, can I bind mount a directory rw into it, so it can access a dir that's on the host?
00:32 < Lope> or ro for that matter, of course.
00:33 < triceratux> ayee: strange. im running 4.15.0-15-generic on an 18.04 daily & theres evidence of hidraw: [    7.897939] hidraw: raw HID events driver (C) Jiri Kosina
00:35 < nbah> too late!
00:35 < nbah> hp-setup -i for interactive mode
00:35 < nbah> 0 usb, 1 net, 2 paralel
00:42 < ayee> triceratux: hmm, weird. maybe my vagrant vm is turning that off or something? or are you running it on baremetal?
00:42 < triceratux> ayee: yep
00:43 < ayee> maybe it has something to do with being a VM?
00:43 < ayee> ➜  /boot cat config-4.15.0-19-generic|grep -i hidraw; uname -r CONFIG_HIDRAW=y 4.15.0-19-generic
00:43 < ayee> I can see it's compiled in on the config
01:11 < mawk> what's the MTU of openvpn over tls ?
01:12 < ayecee> 1400 is a common number to use
01:12 < mawk> 1400 ? that's very low
01:12 < mawk> I'd have expected 1480 or something like that
01:13 < mawk> well, not that low
01:13 < mawk> but still
01:13 < ayecee> you can probably get away with more, but that's a value that works in most situations.
01:14 < mawk> the real underlying question is what is the size of the TLS headers
01:26 < dannylee> ...
01:41 < mawk> when I do struct ifreq ifr = {}; ifr is always initialized to zero ? in C++
01:47 < danieldg> mawk: modulo constructors, I believe so
01:47 < Psi-Jack> 4.15? Pfft old! Hehe
01:49 < Psi-Jack> Yeah openvpn varies with also size of the tls cert as well iirc.
01:50 < Psi-Jack> So a cert that's 4096 bits. That's 512 bytes.
01:51 < mawk> it's a C structure danieldg
01:51 < mawk> so just default constructor
01:51 < mawk> but that's just sent during the handshake Psi-Jack no ?
01:51 < mawk> not with every TCP packet ?
02:01 < Celmor> I'm getting bluetooth related messages in dmesg although I don't have a bluetoot interface https://ptpb.pw/CIx9
02:02 < revel> Celmor: Might be from a compiled-in module. Shouldn't really matter much.
02:03 < Celmor> it kind of would if there's suddenly some sort of radio online searching and open for pairing with other bluetooth devices
02:04 < revel> I thought you said you didn't have a bluetooth interface...?
02:04 < Celmor> from what I know
02:04 < revel> Well, figure out if you have one then.
02:04 < Celmor> who knows what sort of bluetooth device or w/e has a radio thing hidden
02:04 < Celmor> well, how do I know what those messages are caused by?
02:05 < revel> The kernel.
02:05 < Celmor> what module though
02:06 < revel> Does lsmod list anything that starts with bt?
02:07 < revel> Also, if the bluetooth thing was supposed to be hidden, then talking to and depending on the kernel isn't a great idea. Neither does the fact that bluetooth is pretty short-range.
02:07 < Celmor> nope
02:07 < revel> Then maybe it's compiled in, idk
02:08 < Celmor> could I get a trace for future "Bluetooth:" messages?
02:17 < revel> I'd try to figure out how you have a bluetooth interface instead. Maybe it's bundled into your wireless NIC.
02:54 < Psi-Jack> Snoopy: Seriously, please stop constantly nick-flooding us.
02:54 < dviola> hey Psi-Jack
02:54 < dviola> still on i3?
02:54 < Psi-Jack> Yep
02:54 < dviola> nice
02:55 < Psi-Jack> dviola: Was it you that suggested me towards bcache or no?
02:55 < Spawndemonic> There is a proccess called sleep that's been slowing taking up more and more of my cpu. I can't seem to find any info on it,any advice?
02:55 < dviola> Psi-Jack: I don't honestly remember, probably not :P
02:55 < revel> Spawndemonic: It may or may not be malware.
02:55 < dviola> Psi-Jack: are you using it?
02:55 < Psi-Jack> Okay. It was probably triceratux
02:55 < Psi-Jack> I was. :)
02:55 < dviola> how is it?
02:56 < Psi-Jack> It's good, and dangerous at the same time. :)
02:56 < dviola> heh
02:56 < Psi-Jack> But, definitely has potential. It definitely uses more RAM memory than I'd like it to.
02:56 < dviola> I see
02:57 < Celmor> revel, I don't have wireless
02:57 < Psi-Jack> Now I'm using ceph purely with Bluestore on spinning disks, and getting pretty phenominal performance. Better than I was with bcache on spinning disks backed by SSD.
02:57 < Snoopy> My nick changing didn’t flood you shut the fuck up
02:57 < Snoopy> :|
02:57 < Psi-Jack> !ops Snoopy Language, Nick flooding again
02:58 < Snoopy> :|
02:58 < revel> Snoopy: It sent $((n_nicks * n_nickchanges)) more messages than it had to.
02:58 < Snoopy> Oh
02:58 < revel> s/n_nicks/n_usersinalljoinedchannels/
02:59 < revel> Well, uhh, you get the idea.
02:59 < Celmor> what driver handles NDIS/tethering?
02:59 < triceratux> Psi-Jack: nope bcache wasnt me either. im barely getting thru the dnsmasq reading ;)
02:59 < Snoopy> Idk if I do or not but ok
02:59 < Psi-Jack> Hmmm
03:00 < revel> Well, the nick change message gets sent to every user in every channel you're on every time you change your nick.
03:00 < Psi-Jack> Snoopy: https://paste.linux-help.org/view/62a3ee2d Just earlier. And you change nicks like 10+ times/day
03:00 < Psi-Jack> Often times multiples in a row.
03:08 < aaa_> :|
03:10 < aaa_> i also hate pi-jack
03:10 < aaa_> this guy is old pig
03:12 < Aph3x-WL> better that than what ever you are
03:28 < ayecee> a war of words between unarmed combatants
03:35 < djph> ayecee: IRC?
03:36 < ayecee> heh
03:42 < funksh0n> Hi all.
03:42 < illkitten> hi
03:42 < sh4z> hi
03:43 < funksh0n> Can someone jog my memory on the 'clipboard' buffers in Xorg?  More specifically in vim I can use "6p to paste one of them, but I can't remember the command for the other common one.
03:46 < funksh0n> Hm, okay so it's "+p or "*p that I'm thinking of
03:46 < funksh0n> but for some reason I cannot get the content yanked from qutebrowser
03:46 < funksh0n> I guess I'll ask in #qutebrowser
03:47 < jim> funksh0n, usually in a terminal that displays itself in X, it's ctrl-shift-X to cut, ctrl-shift-C to copy and ctrl-shift-V to paste... if you have a qwerty layout, you'll see x c and v at the bottom left
03:48 < jim> those aren't dependent on running vi
03:48 < funksh0n> jim but you have to highlight the content for that functionality.
03:48 < jim> yes
03:48 < funksh0n> I am aware of how to do that, but it's not what I'm asking about.
03:49 < nai> funksh0n: if your version of vim wasn't built with clipboard support, you won't be able to access X's clipboard via registers
03:49 < funksh0n> ahh
03:49 < funksh0n> Yes I think that's it nai
03:50 < jim> emacs has some kind of cut/copy/paste thing that also isn't X's copy buffer
03:50 < nai> run vim --version, look for -clipboard or +clipboard
03:52 < funksh0n> Hm there's mention of clipboard eitherway.  It's nvim and `:help feature-compile` suggests that it would have clipboard, although doesn't explicitly reference clipboard.
03:52 < funksh0n> it's neovim
03:52 < funksh0n> there's no mention*
03:52 < triceratux> hrm thisll work https://askubuntu.com/questions/907246/how-to-disable-systemd-resolved-in-ubuntu
03:52 < nai> ah. you might wanna ask in #neovim or whatever the channel is called
03:52  * triceratux only needs a few more days
03:52 < triceratux> or weeks
03:52 < funksh0n> yeah will do, thanks for the pointers jim and nai
03:57 < jim> funksh0n, is it that you want a vim that was built with clipboard support?
03:58 < funksh0n> jim no
03:58 < jim> hmm. then I never did understand what you wanted
03:58 < funksh0n> I've just got a new computer so I'm setting things up, and I stumbled accross this problem.  I don't recall it being a problem on my previous setup also using qutebrowser and neovim.
03:59 < funksh0n> The content exists in the clipboard as I can C+A+v it, but for some reason neovim cannot access the buffers through "+p or "*p it seems.  I think nai is correct that it's a neovim issue so I've moved my question there.
04:00 < funksh0n> or C+v it*
04:01 < jim> yeah, C-A-v is more global than just vi/vim
04:01 < funksh0n> So you're correct in your assumption that I do want a vim that has clipboard support, but my goal is not to look for a vim with clipboard support because I know neovim already has it... or should... it's just not working.
04:02 < jim> funksh0n, have you ever built vim before?
04:02 < funksh0n> jim am I correct in assuming C-A-v is just a shortcut for my terminal emulator, to avoid conflict e.g. C-c and C-A-c?
04:02 < jim> yeah, and, it's been long-running
04:02 < funksh0n> jim I did when I used vim, for clipboard support.  I think that's one of the reasons I switched to neovim anyway.
04:04 < jim> I'm personally not great at vi or vim, couldn't get past the modalness
04:05 < jim> so I got used to a few nonmodal editors, nano, emacs, kwrite, a few others
04:05 < stevendale> Hiya
04:06 < jim> hi.
04:06 < revel> Hi, Windows guy.
04:06 < funksh0n> maybe it's time for me to bite the emacs bullet
04:07 < jim> I dunno, maybe you'll like it, maybe not
04:07 < stevendale> Windows isn't that bad
04:07 < stevendale> :P
04:07 < funksh0n> I like to be agnostic jim ;)
04:08 < jim> ok well if you want to run the emacs tutorial, from inside emacs, ctrl-h t
04:08 < SporkWitch> funksh0n: how many hands do you have?
04:09 < jamesarch> hi
04:09 < jim> hi
04:09 < funksh0n> hi jamesarch
04:10 < funksh0n> jim I figured it out.  I don't have xorg-xclipboard installed :P
04:10 < jamesarch> i have very strange problem
04:10 < SporkWitch> If you have a question, just ask! For example: "I have a problem with ___; I'm running Debian version ___. When I try to do ___ I get the following output ___. I expected it to do ___." Don't ask if you can ask, if anyone uses it, or pick one person to ask. We're all volunteers; make it easy for us to help you. If you don't get an answer try a few hours later
04:10 < revel> SporkWitch: I wonder how many people that have one hand there are that use emacs.
04:10 < SporkWitch> revel: the reason i ask is because two is insufficient for emacs use
04:11 < revel> Yeah, yeah, but if you have less than two?
04:11 < SporkWitch> revel: SOL
04:11 < jim> or less than one
04:11 < revel> jim: A stump?
04:11 < jim> or two stumps
04:12 < revel> Is emacs why this "pair coding" thing exists?
04:12 < SporkWitch> pair coding?
04:12 < ayecee> no, that was because web 2.0
04:12 < revel> https://en.wikipedia.org/wiki/Pair_programming
04:13 < jim> pair coding is where you got one person with a keyboard and the other running out ideas
04:13 < SporkWitch> i would kill the guy in the back seat...
04:13 < revel> " an online survey of pair programmers, 96% of them stated that they enjoyed their work more than when they programmed alone"
04:13 < jim> so pair programming isn'/t for you....
04:14 < SporkWitch> you can review my pull request; leave me a lone while i'm writing it
04:14 < SporkWitch> s/a\s/a/
04:14 < djph> ^
04:14 < funksh0n> SporkWitch: two
04:14 < funksh0n> two hands and zero mices
04:15 < revel> Mices.
04:15 < funksh0n> mices
04:15 < revel> Do you prefer menses or womenses?
04:15 < funksh0n> can I have both?
04:15 < jim> for what?
04:15 < revel> Do you like the book "Of mices and menses"?
04:15 < funksh0n> funsies
04:15 < revel> jim: Fun.
04:15 < nai> boths*
04:15 < revel> Yes, funsies.
04:15 < SporkWitch> i wouldn't think you'd want either; most women complain about their menses...
04:15 < revel> lol
04:16 < jim> oh, that's probably a different thing :)
04:16 < SporkWitch> (go away, aunt flow!)
04:16 < revel> SporkWitch: Find oneses that don't have menses.
04:16 < SporkWitch> revel: those are called "men"
04:16 < jamesarch> I have a problem with use PPTP Multiple dialing using the same account; I'm running Centos  version 7. When I try to do use mp or multilink param I get the following outputhttps://pastebin.com/hB6m4n4V
04:16 < funksh0n> the dam menopauses
04:16 < revel> All womenses have menses?
04:17 < funksh0n> huh so maybe menopauses is a suspension of activity with menses
04:17 < SporkWitch> revel: all biological ones of child-bearing age and no relevant birth defect do
04:17 < SporkWitch> funksh0n: you have insufficient hands for emacs use
04:18 < funksh0n> SporkWitch: I got my clipboard working so growing extra limbs is on holdsies
04:18 < revel> Hmm... I'd try to refute that, but I can think of very few womenses to whom this doesn't apply.
04:19 < SporkWitch> revel: male-to-female trans women, those that have hit menopause, those who have no yet reached puberty, and those suffering some form of birth defect.  suppose injury could potentially cause it, but that covers the normal exceptions
04:20 < revel> What about the occasional introvert?
04:20 < revel> I suppose they're more career-oriented instead, maybe/
04:21 < jamesarch> Does PPTP support multiple dialing of the same account?
04:22 < jamesarch> I Try use RouterOS test same thing Routeros is support ,but why centos or debian is not support?
04:23 < djph> huh?
04:23 < jamesarch> Or my operation is wrong?
04:23 < djph> well, using PPTP is a bad idea, yes
04:24 < jamesarch> yew i konw...
04:24 < jamesarch> but Business needs _(:з)∠)_
04:26 < ayecee> pptp isn't as bad as it once was
04:26 < jamesarch> So is there any way to this is this function?Multiple dialing for one account use PPTP
04:26 < funksh0n> jamesarch: are they both the same version?
04:26 < funksh0n> on deb and routerOS?
04:27 < ayecee> jamesarch: doesn't seem like it would be a problem, if the server permits it
04:27 < djph> ayecee: pptp is worse than ever.  at least use ipsec for vpn
04:28 < jamesarch> i think is same version Routeros is  Commercial version, do not see the version number
04:28 < ayecee> djph: saying it again more emphatically doesn't make it true.
04:28 < djph> ayecee: pptp has been obsolete for years ...
04:28 < ayecee> no, it hasn't
04:29 < jim> just so I know, what's pptp?
04:29 < jamesarch> yes... Business needs boss needs
04:29 < ayecee> point to point tunneling protocol, a vpn protocol that uses a virtual ppp connection
04:29 < djph> ayecee: please provide links to Point to Point Tunneling Protocol *not* being obsolete
04:30 < jamesarch> And ... is not use ROuteros
04:30 < ayecee> djph: please provide a link to it being obsolete
04:30 < jamesarch> l0l
04:31 < triceratux> https://www.ovpn.com/en/blog/pptp-has-become-obsolete/
04:31 < jim> okok, please provide a link to how to end arguments
04:31  * triceratux was tired of dnsmasq anyway :P
04:31 < funksh0n> jamesarch: this is probably a horrible suggestion, but if you cannot get it running on debian you could virtualise routeros on debian :/
04:31 < jamesarch> Routeros test success i think is not pptp Server problem
04:31 < funksh0n> iptables?
04:32 < RaptorJesus> hey guys
04:32 < mikez> hi WodenCafe
04:32 < jim> hi
04:32 < RaptorJesus> what is devops
04:32 < jamesarch> hi
04:32 < funksh0n> hi
04:32 < RaptorJesus> and how do?
04:32 < triceratux> RaptorJesus: admins that know how programmers think
04:32 < ayecee> RaptorJesus: software development in the support of operations
04:32 < funksh0n> RaptorJesus: do you enjoy sanity?
04:33 < Psi-Jack> RaptorJesus: https://www.redhat.com/en/command-line-heroes episode 3
04:33 < RaptorJesus> uhm
04:33 < djph> dammit triceratux ... beat me to it...
04:33 < RaptorJesus> i enjoy scripting things
04:33 < Psi-Jack> ayecee: Incorrect. :)
04:33 < jamesarch> em...  if use ROuteros Project Architecture It will be very  complicated
04:33 < jim> ok
04:33 < jamesarch> very
04:33 < RaptorJesus> episode 3 or 4
04:34 < ayecee> oh ok
04:34 < Psi-Jack> I said episode 3.
04:34 < mikez> hi OtakuSenpai
04:34 < Psi-Jack> 4 covers Agile.
04:34 < RaptorJesus> cuz 4 has the word devops in it
04:34 < jamesarch> and  my server is cloud server
04:34 < RaptorJesus> it's in order 6,5,4,3
04:34 < Psi-Jack> Oh, I'm backwards, sorry.
04:34 < Psi-Jack> Episode 4 is DevOps, 3 is Agile. heh
04:35 < Psi-Jack> I recommend of course listening to both.
04:35 < Psi-Jack> Heck, the whole series is awesome, Listen to all.
04:36 < WodenCafe> what
04:36 < WodenCafe> oh hi mikez
04:37 < funksh0n> djph: kinky
04:37 < ayecee> djph: fwiw, it's mschapv2 that's the problem, not pptp itself.
04:37 < ayecee> pptp doesn't require mschapv2.
04:41 < stevendale> :o
04:41 < stevendale> People refer to me as 'Windows guy' for some reason now o/
04:41 < SporkWitch> could it be because it's literally the only thing you talk about?
04:41 < SporkWitch> in a LINUX channel
04:45 < jamesarch> so... PPTP protocol is support one account one public Dial several times like 2 or 3 time?
04:45 < ayecee> jamesarch: there's nothing about pptp as a protocol that would prohibit that.
04:46 < ayecee> though the vpn server may not be configured to allow it
04:46 < jamesarch> and  what is mean?
04:47 < ayecee> what
04:47 < jamesarch> anon log[decaps_gre:pptp_gre.c:418]: discarding duplicate or old packet 1
04:47 < jamesarch> this  anon log[decaps_gre:pptp_gre.c:418]: discarding duplicate or old packet 1
04:47 < ayecee> means what it says. it discarded a duplicate or old packet.
04:48 < jamesarch> so... if i have connect suceess and i try agian  Will this message be prompted?
04:48 < ayecee> no
04:50 < jamesarch> well... thanks
04:50 < jaggz> is there a way I can stop my hd from writing for longer periods?
04:51 < ayecee> disconnect it
04:51 < jaggz> I'm trying to record audio and the mic is picking up the hd chatter
04:51 < jamesarch> i just stock there don't know how to achieve this function
04:53 < ayecee> jaggz: interesting question. i don't know of a way offhand to suspend writes.
04:53 < ayecee> or suspend io more generally\
04:54 < Ox600df00d> remount everything as rdonly, sync disks and record to tmpfs?  or get a proper microphone + isolation booth
04:54 < ayecee> in most cases it would be the amplifier in the sound card rather than the microphone itself
04:55 < ayecee> if it's electronic noise and not actual audible hard drive chatter
04:55 < ayecee> maybe this is good reason to get ssd? :)
05:00 < jaggz> https://unix.stackexchange.com/questions/150211/how-to-reduce-disk-access
05:01 < boblamont> I have this script running in crontab https://paste.linux.community/view/52062726 but though the files are being created, they are coming up empty (the mp3 files are 834 bytes and the oggs are 3.9 KiB). I have successfully recorded manually in the terminal. Any idea why it isn't working?
05:03 < ayecee> boblamont: there's no # on line 1
05:04 < ayecee> more generally though, when a script works when you run it manually but doesn't work in cron, it's almost always because your PATH is different.
05:04 < ayecee> to fix it, set your path in the script accordingly.
05:04 < littel5t> what is context switching
05:05 < littel5t> lets say i hav 2 jobs, cpu schedules 1 job to 1 core and after the 1st job finishes it goes to 2nd job?
05:05 < littel5t> now the 2nd job runs on 2nd core. is this called as context switch?
05:06 < ayecee> context switch would be when a core switches from one task to another, sure
05:06 < littel5t> ayecee: is my explanation right?
05:06 < ayecee> but not just because job 1 finished - more often because job 1 used its allocated time
05:06 < littel5t> also is there a way to control this
05:06 < littel5t> how can we know the allocated time?
05:07 < ayecee> the allocated time is the scheduling interval, configured at compile time to be anywhere from 100 to 1000 times per second.
05:08 < jim> littel5t, program A is running, program B is running... only one program can run at a time... how to make it appear like both programs run at the same time? answer: stop A, run B for awhile, stop B, run A for awhile, repeat
05:08 < ayecee> it's how often the scheduler decides which process gets the cpu
05:08 < boblamont> ayecee: thanks, it must be running part of the script if it's creating the files, though, right?
05:08 < littel5t> ayecee: where is this configured?
05:08 < littel5t> jim: sure
05:08 < ayecee> boblamont: makes sense
05:09 < ayecee> littel5t: it's an option when compiling a kernel.
05:09 < littel5t> aah ok
05:10 < ayecee> the "context" is the bit of info that describes what a task was doing when it was interrupted, so it can start from the same place again
05:11 < jaggz> is there a way to suspend an X11 program?  I have a bunch of pdf's loaded in pdf viewers and they write session info out
05:11 < jaggz> okular's writing out stuff
05:12 < jim> littel5t, "stop A, run B for awhile" is a context switch... "stop B, run A for awhile" is another one
05:12 < jaggz> (okular's a pdf viewer)
05:12 < jaggz> I don't want to close them all 'cuz I don't think it'll keep my session
05:13 < boblamont> ayecee: the filenames are defined after being piped to the encoders...so would that mean it gets to the encoders, or no? It has to be either that arecord isn't recording the raw audio (so the encoders have nothing to encode) or the raw audio isn't being encoded (so the raw audio has nowhere to go)
05:13 < ayecee> boblamont: my guess is that arecord couldn't be found, so lame gets no input.
05:14 < jim> littel5t, and all the tiny pieces of info you have to save about running A, gets saved, so that A can start later, right where it left off
05:14 < littel5t> jim: ayecee is there a way to know what is the scheduling interval currently used for a context switch?
05:14 < ayecee> boblamont: or that arecord threw an error
05:14 < ayecee> littel5t: probably, but i don't know it offhand.
05:14 < jim> I never found out about that piece
05:14 < ayecee> i'd have to google it.
05:16 < WishBoy> can I consider Android a linux distribution for portable devices?
05:16 < ayecee> WishBoy: in a sense yes, but not in the sense that we'd discuss it here.
05:17 < ayecee> i guess inasmuch as osx is a freebsd distribution
05:18 < WishBoy> ayecee oh, okay... was just that doubt :)
05:18 < jim> WishBoy, yes, but the programs you can run on android is not the GNU tools, so for -this- channel, it's considered off topic
05:19 < boblamont> ayecee: thanks, that makes sense. When I I use arecord in terminal, I don't need a path, so I assumed it would be fine. So should I substitute " /user/bin/arecord" where I just have "arecord" in the script?
05:19 < jim> WishBoy, #android would be a start for that
05:20 < ayecee> boblamont: the ideal thing would be to catch whatever error is being thrown in the script, perhaps with "exec 2>/tmp/myscript.log" as the second line. If not that, then set the PATH in the script rather than expanding locations like that.
05:21 < jim> boblamont, a long time ago, unix devs misspelled "user" as "usr", for the directory name... so it would be /usr/bin/arecord
05:22 < WishBoy> jim my question was just that doubt, i have no questions about Android, i had just that question of how it is seen Android that uses linux kernel in the linux world, no more than that ... i wanted to see just the point of view, not i have questions about Android itself
05:23 < boblamont> jim: thanks, I do know that, I just sometimes inadvertently "autocorrect" it
05:23 < stephen> I have multiple wireless clients from which I'd like to gather their logs on the main server. These clients sometimes can't connect, so I want their logs to queue, but ultimately I just want them on the server. What's the most efficient and reliable way to get them there?
05:23 < funksh0n> ayecee: does cron not respect the users PATH?
05:25 < ayecee> funksh0n: cron normally doesn't parse the same files that an interactive login would.
05:25 < funksh0n> is that the reason dmenu_run wont run scripts in my ~/bin folder, despite it being in my path?  I can run the scripts from rofi or terminal.
05:25 < ayecee> could be
05:28 < WishBoy> jim ayecee thanks, i need to sleep now, to late here! goodnight
05:28 < funksh0n> hm, from the manpage "dmenu_run is a script used by dwm(1) which lists programs in the user's $PATH and runs the result in their $SHELL."
05:28 < funksh0n> It does not exhibit this behavior for me :(
05:30 < nai> in what man page is it documented that "." is the current directory and ".." is the parent directory?
05:31 < triceratux> funksh0n: dmenu_run sometimes need some parms passed to run correctly like a font &/or a number of lines to display
05:32 < ayecee> nai: i wouldn't think that would be spelled out explicitly.
05:32 < littel5t> can i also know what are interrupts?
05:33 < ayecee> littel5t: someone taps you on your shoulder to get your attention. that's an interrupt.
05:33 < funksh0n> triceratux: dmenu_run seems to be functioning correctly, except for respecting my PATH...
05:33 < littel5t> i know that hardware uses interrupts to get cpu attention.. but can someone brief me more a little :)
05:33 < nai> ayecee: what makes you think that?
05:33 < funksh0n> Does it actually look at my PATH every time I invoke it, or does it cache?  I've rebooted the machine a few times and the problem has persisted.
05:34 < funksh0n> nai: is it a POSIX thing?
05:34 < ayecee> nai: nothing in particular, just experience
05:34 < triceratux> funksh0n: i havent run into that exactly but it does keep a persistent cache that may have to be deleted over time
05:35 < funksh0n> My path was actually set before I installed it :/
05:36 < funksh0n> I see a file ~/.cache/dmenu_run though, I'll try remove that
05:37 < wr> installed a debian lxde then i put xfce, i wanna remove lxde but without removing wicd and damaging dependencies, first try did ruined my system, viable?
05:37 < funksh0n> Nope, it regenerated the file and still does not respect my PATH
05:37 < funksh0n> I'll see if I can find where to report a bug
05:37 < Triffid_Hunter> funksh0n: cron runs stuff with a minimal sanitized path
05:37 < Triffid_Hunter> funksh0n: if you want your scripts to use your path, include your ~/.bashrc or so in the script
05:37 < funksh0n> Maybe my googlefu is weak but I can't see others with this issue or an explanation as to why.
05:38 < funksh0n> Triffid_Hunter: I see, good to know.
05:38 < Triffid_Hunter> funksh0n: because those of us who use cron get to know this pretty quickly, and don't think about it much thereafter
05:38 < Triffid_Hunter> funksh0n: alternatively, use full paths to the things you're calling
05:38 < funksh0n> My issue isn't regarding cron Triffid_Hunter
05:39 < triceratux> wr: ewww. just start with the xfce liveiso. if debian ships it that way there must be a reason https://cdimage.debian.org/debian-cd/current-live/amd64/iso-hybrid/
05:39 < Triffid_Hunter> funksh0n: ah must have misread something
05:39 < funksh0n> Triffid_Hunter: I just piggybacked on someone elses cron issue, mine is about dmenu_run not respecting my $PATH
05:40 < wr> triceratux, already at another config, just wanna do what asked cause dont wanna restore my system all over and files
05:40 < Triffid_Hunter> funksh0n: does it know about your path?
05:41 < wr> triceratux, by the way already have xfce installed
05:41 < funksh0n> Triffid_Hunter: "dmenu will look for executables in the directories defined in your $PATH. " and "If certain entries are missing from dmenu, the cache may be malformed. Delete it and restart dmenu.
05:42 < funksh0n> both from https://wiki.archlinux.org/index.php/Dmenu
05:42 < funksh0n> the man page suggests the same
05:42 < funksh0n> rofi respects my PATH
05:42 < funksh0n> I can launch the scripts from my terminal
05:42 < funksh0n> but dmenu_run does not see them
05:42 < funksh0n> I suppose I could manually add them to the cache, but that's atypical usage
05:43 < triceratux> funksh0n: how are you invoking it ? the installed hotkey ? sounds like its coming up in a shell with a summarised ${PATH}
05:45 < funksh0n> triceratux: I'm using i3 so $mod+d
05:46 < funksh0n> `bindsym $mod+d exec dmenu_run -nb '$black' -sf '$white' -sb '$light' -nf '$light'`
05:47 < jab416171> funksh0n, what are you trying to do?
05:47 < funksh0n> jab416171: dmenu_run is not respecting my PATH
05:48 < jab416171> oh right
05:48 < jab416171> I had that same problem
05:48 < funksh0n> I have scripts in /home/user/bin, that is in my PATH, rofi can launch them and they run from terminal from any directory.
05:48 < funksh0n> oh not just me then!
05:48 < jab416171> I forget how I fixed it
05:48 < funksh0n> Think damit!  Tell me everything!
05:49 < jab416171> where do you set your path?
05:49 < jab416171> .bash_profile?
05:49 < funksh0n> Show me your config file (if such a thing exists for dmenu) and how you invoke?
05:49 < funksh0n> ohhh
05:49 < funksh0n> zshrc
05:49 < funksh0n> I think you've cracked it
05:49 < jab416171> it has to be in .profile
05:50 < jab416171> or it has to be read in from .profile (for bash anyway)
05:50 < funksh0n> my dude this feels like the right path ;)
05:50 < jab416171> so like, my .profile sources .bash_profile
05:50 < Triffid_Hunter> isn't profile only used for login shells?
05:50 < triceratux> http://pastebin.centos.org/704296/raw/
05:51 < LissajousPattern> Elon?
05:51 < LissajousPattern> Where are you?
05:51 < stevendale> OwO
05:52 < funksh0n> triceratux: so have my $mod+d execute a script which invokes dmenu_run instead?
05:52 < funksh0n> I like the rm, toomanysecrets
05:53 < triceratux> funksh0n: the thing is somewhat brittle. youll have to adjust your strategy to the exact bug ypure seeing
05:54 < rangergord> 1 more day to Ubuntu LTS!
05:54 < rangergord> I moved to Australia so I'd get it one day faster than back in Canada
05:54 < funksh0n> Well it's a decent suggestion if it works, but maybe jab416171 is onto a better solution of setting my PATH in a more appropriate place.
05:55 < ayecee> one more day to 18.04. I think it becomes LTS on the .1
05:55 < LissajousPattern> nice
05:56 < rangergord> ayecee: oh, that's right, 16.04 will only update to 18.04.1
05:56 < funksh0n> are LTS releases forwards compatable?
05:56 < rangergord> yeah
05:56 < ayecee> funksh0n: in that you can update from one LTS to another, yes
05:57 < rangergord> I've never done it myself, I just create a new VM and run personal scripts to get everything back to the way it was
05:57 < rangergord> putting my dotfiles on bitbucket has been great
05:57 < littel5t> what does curl ifconfig.me and ifconfig show.. what is my ip address out of these
05:58 < rangergord> ifconfig shows your local IP address (on your home network). ifconfig.me will show you the one you have on the Internet, ie the one of your ISP's modem
05:59 < funksh0n> triceratux jab416171 my /usr/bin/dmenu_run it turns out is just a shell script, and it specifies the shell as /bin/sh.  However, if I run sh and `echo $PATH`, my /home/user/bin is there.  Is this because I've run sh from within zsh, so the PATH is inherited?
05:59 < rangergord> the MPAA sees the ifconfig.me address when you torrent something
06:00 < rangergord> so you and your family/roommates will all have the same ifconfig.me IP usually (WAN IP), but each have a different ifconfig IP (LAN IP)
06:00 < ReddyTeddy> hey guys, my pip / python is all very strange ... I can't seem to run certbot (even in certonly)
06:00 < ReddyTeddy> and i can't figure out if it's b/c of python2.7, python3.5, or something wrong with pip
06:00 < ayecee> what happens when you try
06:02 < ReddyTeddy> nginx: [emerg] BIO_new_file("/etc/letsencrypt/live/here.ismy.domain/fullchain.pem") failed (SSL: error:02001002:system library:fopen:No such file or directory:fopen('/etc/letsencrypt/live/here.ismy.domain/fullchain.pem','r') error:2006D080:BIO routines:BIO_new_file:no such file)
06:02 < ReddyTeddy> nginx: configuration file /etc/nginx/nginx.conf test failed
06:02 < ReddyTeddy> (which makes sense, because i let my cert expire without renewing it i think)
06:02 < ayecee> doesn't sound like a pip or python problem
06:02 < ayecee> sounds like the file is missing
06:03 < ReddyTeddy> https://paste2.org/46ahXdZw
06:03 < ReddyTeddy> maybe, but the thing is, all results about how to debug this issue point to python/pip
06:03 < ReddyTeddy> (the openssl issues)
06:04 < ayecee> take a look in /var/log/letsencrypt for more details
06:06 < ayecee> it seems like it's failing to verify the certificate for acme-v01.api.letsencrypt.org
06:06 < jeffreylevesque> does this trace indicate that something in ubuntu repo is broken - https://github.com/jeff1evesque/machine-learning/issues/2935#issuecomment-384149635?
06:06 < jeffreylevesque> some of the error links do not exist
06:07 < ayecee> jeffreylevesque: run apt-get update first
06:08 < ReddyTeddy> ayecee, do you know what exactly i should look for?
06:08 < ayecee> no
06:12 < jeffreylevesque> ayecee: ah dang you're right, that's awesome
06:13 < ayecee> :)
06:13 < rangergord> I'm trying to do something a bit weird: I have a bunch of patch files generated from git commits. I want to apply them to a NON-GIT codebase. It's a hacky manual merge job. So I'm relying on the GNU patch command. It works, but is there a way to get some visual assistance so I can more easily process failed hunks? Using CLI is a pain.
06:15 < ayecee> ReddyTeddy: possibly old version of certbot?
06:21 < wr> wanna use a network manager on xfce, what would be better to use? wicd, network manager, other?
06:22 < revel> I'd personally use NetworkManager. I'm on xfce as well.
06:22 < thebigj> I am trying to configure networking of QEMU VM.
06:23 < thebigj> I am on Gentoo distro.
06:24 < ReddyTeddy> ayecee, I just did a full reinstall of certbot from https://certbot.eff.org/lets-encrypt/ubuntuxenial-nginx
06:24 < thebigj> I have less knowledge of TUN/TAP. I have skimed through the Wiki page and this discription of kernel https://www.kernel.org/doc/Documentation/networking/tuntap.txt
06:25 < thebigj> I am trying to add tap device from https://wiki.gentoo.org/wiki/QEMU section "Networking"
06:26 < thebigj> When I am running "ip tuntap add dev tap0 mode tap group kvm" command, it is exiting with "1" status and "open: No such file or directory" as a error message.
06:26 < thebigj> I can assume, it is unable to find "tap0" may be.
06:26 < ayecee> ReddyTeddy: did that help?
06:27 < ReddyTeddy> no, still getting the same issue I pastebinned
06:27 < thebigj> Can anyone help me to add a tap device?
06:27 < thebigj> Apologies if I am asking something stupid. Thanks!
06:29 < ReddyTeddy> when trying to do stuff with pip, getting a lot of : AttributeError: module 'lib' has no attribute 'X509_up_ref'
06:29 < ayecee> would be helpful to see what you're seeing
06:31 < ayecee> seeing some suggestion that that might be due to out of date python cryptography module
06:31 < litt5li> rangergord: what is the difference between WAN ip and LAN ip
06:31 < ayecee> i'm not sure how to check the version though
06:32 < litt5li> lets say im configuring a webserver on my home computer.. so the website does the dns translation to ifconfig's ip or curl ifconfig.me's ip?
06:32 < Psi-Jack> Alllllrigty! Finally got a functional HTTPS/SSH haproxy setup.
06:32 < ayecee> \o/
06:32 < ReddyTeddy> ayecee, I just think my python / pip versions are all confused
06:32 < ReddyTeddy> dependencies and whatnot
06:34 < Psi-Jack> Hmmm, now the question is, what host gets logged from this method, hmmm
06:35 < funksh0n> Psi-Jack: congratulations and condolences
06:35 < ayecee> worthwhile setting up the forwardedfor option
06:35 < ayecee> otherwise everything looks like it comes from the proxy
06:35 < Psi-Jack> funksh0n: condolences?
06:35 < funksh0n> Psi-Jack: webdev
06:35 < Psi-Jack> ayecee: Yeah.
06:36 < litt5li> anyway to check the logs for this channel?
06:36 < Psi-Jack> I use forwardfor for the https backend.
06:36 < Psi-Jack> litt5li: Yes, check your logs your client makes.
06:36 < litt5li> i use webchat.freenode.net
06:37 < ayecee> then the webserver must be configured to use that as well. in apache it's the remoteip module, and you must tell it to trust what the proxy says
06:37 < ayecee> (i just set this up recently)
06:37 < Psi-Jack> Yeah, but for mode tcp, forwardfor, doesn't work.
06:38 < Psi-Jack> For ssh anyway. ;)
06:38 < ayecee> ah, yeah, it wouldn't
06:38 < stevendale> OwO
06:38 < Psi-Jack> So, logins appear as-if 127.0.0.1 :/
06:38 < ayecee> odd to put haproxy in front of ssh on same host
06:39 < Psi-Jack> ayecee: It's an interesting way to provide ssh from a well-known port that's not blocked by things like hotel wifi.
06:39 < Psi-Jack> It's possible to combine https, ssh, and openvpn all via this method.
06:39 < ayecee> it's interesting alright :)
06:40 < ayecee> all on the same port?
06:40 < Psi-Jack> Yes
06:40 < ayecee> huh
06:40 < ayecee> how does it tell which service to respond with?
06:41 < Psi-Jack> OpenSSH clients always have a string that the client sends on connection, which you can pattern match. For https, you can match the http protocol with the ssl hello, and openvpn being the default backend.
06:41 < ayecee> ah, i thought it was the server that sent the greeting on ssh connections
06:41 < ayecee> does that when i telnet to the port anyhow
06:42 < Psi-Jack> Yep, they both do. :)
06:42 < ayecee> i see
06:43 < Psi-Jack> https://limbenjamin.com/articles/running-https-ssh-vpn-on-port-443.html
06:44 < Psi-Jack> Pretty good concept. For actual https, it sends it to the localhost and passes it back into haproxy to terminate the ssl connection itself.
06:45 < ayecee> as a bonus the ssh server doesn't advertise itself
06:45 < ayecee> i like it :)
06:46 < Psi-Jack> hehe
06:46 < Psi-Jack> Yep. Client never sees the openssh server until after the client sends the appropriate signature first.
06:46 < Psi-Jack> But, the openssh server only sees 127.0.0.1 for connections, itself.
06:56 < Psi-Jack> Can still get the logs from the haproxy logs to some extent though, but correlating that with the ssh auth logs adds complexity.
06:57 < ReddyTeddy> hm, now I'm getting 'Problem binding to port 80: Could not bind to IPv4 or IPv6.
06:57 < ReddyTeddy> '
06:58 < Happyhobo> Good morning folks.
06:59 < Happyhobo> Hi skizzy ciscam
07:00 < Happyhobo> I'm receiving a message activation of network connection failed then the internet goes down and I have to restart the computer for it to start back up because network manager just searches and searches.
07:01 < stevendale> How is my Intel Core 2 Duo P8700 going to 2660.0 MHz
07:02 < loganlee> hello
07:02 < jim>  hi
07:02 < loganlee> :)
07:03 < jim> Happyhobo, what kind of nic is it?
07:03 < Ox600df00d> how is a P4 going to 3400.0MHz
07:04 < stevendale> The maximum clock for a P8700 is meant to be 2533 MHz
07:05 < jim> stevendale, maybe you should check in your bios what rate your cpu is being clocked at
07:05 < ReddyTeddy> I run 'sudo service nginx stop' but still I see listening to nginx (from nginx.conf)
07:05 < ReddyTeddy> so I can't bind to port 80 with certbot
07:06 < ReddyTeddy> is there a way to actually kill nginx for a bit?
07:06 < jim> does certbot need 80?
07:06 < Psi-Jack> Need? No. certbot itself doesn't bind to any port itself.
07:06 < Happyhobo> Hi Jim
07:06 < ayecee> it does in certonly mode
07:07 < jim> hi
07:07 < ayecee> ReddyTeddy: service nginx stop ?
07:07 < Psi-Jack> No, it doesn't. It uses something else to bind to the port that it configures and runs.
07:07 < Psi-Jack> Unless that's changed.
07:07 < ReddyTeddy> oh wait
07:07 < ReddyTeddy> sorry
07:07 < ayecee> right, a separate program, but still part of certbot
07:07 < ReddyTeddy> after I do fully stop it
07:07 < ReddyTeddy> 'The server could not connect to the client to verify the domain '
07:08 < Psi-Jack> It starts apache or nginx to provide temporary service on port 80.
07:08 < Happyhobo> there is so much information on my issue and none of it makes sense because for every different flavor it's a different method.  I'm on Antergos, an Arch deviant, and the message is Activation of network connection failed.
07:08 < nshire> what do the major:minor device names mean in lsblk?
07:08 < ayecee> but not in certonly mode
07:09 < ayecee> nshire: they're how the kernel knows which device you're talking about.
07:09 < nshire> is that like how in windows disks are accessed with PhysicalDisk[number]?
07:09 < ayecee> kind of, sure
07:09 < Happyhobo> sorry
07:09 < jim> Happyhobo, well let's start with the basics... what kind of nic do you have?
07:09 < Happyhobo> wireless
07:10 < jim> ok, let's start with the device/driver/firmware side
07:11 < jim> when you run lspci -nn | grep -i net, how many lines do you get?
07:11 < Happyhobo> okay sir
07:12 < jim> umm brb
07:13 < jim> (keep working on that... maybe try it this way: lspci -nn | grep -i net | wc -l )
07:15 < nai> ayecee: about my earlier question, i did some very dirty man page grepping, and found results in man path_resolution :)
07:16 < ayecee> nice
07:16 < nai> everything is documented
07:16 < ayecee> if only they'd document where to find the documents
07:17 < nai> true
07:18 < Happyhobo> jim?
07:18 < jim> Happyhobo, you're back... is there a problem in running that line?
07:18 < Happyhobo> I think it was coincidence.
07:19 < Happyhobo> I just ran the lspci command again.
07:19 < jim> not sure I understand... ok, with the wc -l at the end?
07:20 < Happyhobo> lspci -nn | grep -i net
07:20 < Happyhobo> That was the last command I did before I disappeared.
07:20 < jim> ok, how many lines did you get as output?
07:20 < Happyhobo> two two about controllers
07:21 < jim> ok... probably one's an ethernet card and the other is your wireless...
07:21 < jim> and so,
07:22 < Happyhobo> the one that says Network controller is a Centrino Ultimate-N 6300
07:22 < jim> you should see two things that look like this: [1234:abcd]
07:22 < jim> I can look those up
07:22 < Happyhobo> 8086.422b     and 8086.1502
07:22 < jim> ok, very good... let me look those up
07:24  * luke-jr works around GPU problems with the fbdev Xorg driver, yay
07:24 < ayecee> intel wifi
07:24 < ayecee> using iwlwifi module
07:24 < jim> yep
07:24 < Happyhobo> Is that real bad?
07:24 < jim> let's see if it's already inserted
07:24 < ayecee> Happyhobo: no, it's actually pretty good
07:25 < jim> can you run: lsmod | grep iwlwifi
07:25 < jim> do you get any output from that at all?
07:25 < Happyhobo> two lines
07:25 < jim> ok... what dist are you running?
07:26 < ayecee> djph: pptp discussion was fruitful. i realized my vpn concentrator was using bare mschapv2 and not PEAP as i'd assumed. Looks like it also supports l2tp, so i'll give that a whirl.
07:26 < stevendale> OwO
07:26 < Happyhobo> iwlwifi 323584 1 iwl dvm and cfg80211 749568 iwlwifi mac80211 iwldvm  Antergos
07:26 < CrazyTux> what are the pros and cons of Ubuntu LTS and OpenSuse Leap for a beginner for casual home and office use? I intend to install either of them for some beginner end users and perhaps make one of them as my primary OS.
07:26 < jim> that's probably a good indication that the module is in...
07:27 < jim> CrazyTux, I'd go with ubuntu or debian
07:27 < jim> (for that particular audience)
07:27 < CrazyTux> jim, probably I should include Mint also in that list.
07:27 < Happyhobo> if it's in why is it failing
07:27 < ayecee> CrazyTux: could probably go with either. they both have large user bases.
07:28 < Happyhobo> do I have a rootkit?
07:28 < CrazyTux> and I also intend to install some windows softwares through wine or playonlinux.
07:28 < jim> Happyhobo, don't know yet... could you (probably have to do this as root) run: dmesg | grep iwl
07:28 < jim> that's probably going to be a lot of output
07:29 < CrazyTux> some essential packages like MS Office and some accounting packages that are necessary for office use.
07:29 < jim> do you see anything that looks like a failure (especially failure to load firmware)?
07:29 < ayecee> djph: that'll also allow me to support the macs in our environment without additional software.
07:29 < Happyhobo> it's about 15 pages
07:29 < Happyhobo> please give me a second
07:29 < storge> MS Office is an essential package
07:29 < jim> ok
07:30 < jim> back in a few mins
07:31 < Happyhobo> Microcode Swerror detected
07:33 < KaylinStapleton> CrazyTux: Well... I highly recommend Ubuntu or Mint.
07:33 < jim> and I guess the string "iwl" would be in the message that reports that?
07:33 < Happyhobo> yes sir
07:33 < CrazyTux> ok
07:34 < nshire> is it possible to block a certain port above 1024 from being listened to by non-root users? I've moved my ssh port above 1024
07:34 < jim> and which dist are you running?"
07:34 < Happyhobo> Antergos not sure what version it is now because I upgrade when it wants.
07:35 < Happyhobo> Heh I'm missing Mepis.
07:35 < jim> do you get any output from ifconfig wlan0?
07:35 < stevendale> Ms Office is amazing storge
07:35 < Happyhobo> device not found
07:36 < jim> ok.... then: can you run ifconfig -a
07:36 < jim> and look at that output...
07:36 < Happyhobo> wlp3s0
07:36 < revel> nshire: Is this for yourself? Since if you have the server's ssh pubkeys available somewhere else or already on any clients you plan on connecting with to the server, then the client will complain really loudly about the server's keys not matching.
07:36 < jim> you found a likely interface
07:37 < jim> can you try: iwlist scan
07:37 < Happyhobo> but when that message comes up and it is dropped that line isn't in the ifconfig
07:37 < nshire> revel, yeah its just for a home server.
07:37 < CrazyTux> is there too much difference between Mint and Ubuntu?
07:37 < nshire> I can add the new key to the cache
07:38 < stevendale> CrazyTux, Mint pulls packages from lots of different third party repos and that's bad
07:38 < jim> CrazyTux, I think mint takes ubuntu and adds stuff to it
07:38 < revel> nshire: Then yeah, if you've connected with that client before already, then it'll be fairly secure as is since the server keys are only accessible to root.
07:38 < Happyhobo> A bunch of stuff was found jim
07:38 < revel> And known keys are stored in ~/.ssh/known_hosts
07:39 < jim> Happyhobo, do you think you see the neighbor's networks?
07:39 < nshire> yeah the thing I want though it to prevent non-root processes from listening on this port, since apparently non-root processes can listen on anything above 1024
07:39 < Happyhobo> I'm on my neighbor's network
07:40 < CrazyTux> hello..
07:40 < Happyhobo> I'm not stealing net, he knows, I pay him 20 bucks a month
07:40 < jim> Happyhobo, ok then, do you think you see your -other- neighbor's networks?
07:40 < jim> at least the names of them
07:40 < Happyhobo> yes
07:41 < jim> ok... so here's where we are:
07:41 < Happyhobo> ok
07:41 < jim>  your card is working, it sees networks, and so it should work for connecting
07:42 < Happyhobo> but it fails then sees nothing
07:42 < jim> especially, the radio in it is working
07:42 < jim> sees nothing? what are you looking at that shows you that?
07:42 < Happyhobo> it won't see anything again until I reboot, sometimes it's up for hours other times minutes.
07:42 < Happyhobo> network manager in gnome
07:43 < Happyhobo> I lo ok at the ifconfig then and see that in the ifconfig there is a listing for the ethernet and lo but not the wireless
07:43 < jim> that's weird... how fast was the iwlist scan?
07:43 < Happyhobo> pretty fast
07:44 < Happyhobo> I swear I've got a rootkit and someone is screwing with me
07:44 < jim> ok, so something about network manager is blocking it
07:44 < jim> cause your card is -working-
07:45 < Happyhobo> weird huh?
07:45 < jim> driver's working, I guess the right firmware is loaded
07:45 < nshire> every google result for "prevent non privileged processes from accessing port" is the exact opposite of what I'm looking to do..
07:45 < Happyhobo> we've been up about half hour this time
07:45 < jim> that's all "below" (and required by) network manager
07:47 < Happyhobo> if I open up several pages at once it fails.
07:47 < jim> something about that network manager or maybe "rfkill" is blocking it
07:48 < Happyhobo> do I have a rootkit?
07:48 < jim> I don;'t know
07:48 < jim> do you have one of those rootkit checkers?
07:49 < Happyhobo> no I don't
07:49 < maarhart> Hi, I have built a newer version of gcc so it would support c++ 14. Then I could compile my application, but when I tried to launch it, it says libstdc++.so.6 version 'CXXABI-1.3.8' not found
07:49 < maarhart> how can I fix this?
07:49 < jim> maybe you could find and get one?
07:50 < jim> maarhart, what did you do exactly to get that message?
07:51 < maarhart> jim: I just tried to launch an app called mpd, which I just built
07:51 < maarhart> using a newer version of gcc
07:51 < maarhart> 5.3
07:51 < maarhart> newer to the one that there is in this machine
07:51 < jim> mpd? does that launch gcc or g++?
07:51 < maarhart> g++
07:52 < maarhart> anyway, I could build mpd using export PATH=/usr/bin/gcc-5.3/bin:$PATH
07:53 < maarhart> now I want to be able to launch it
07:53 < jim> Happyhobo, look for a root kit checker that is installable from your version of antergos
07:53 < maarhart> maybe it has to do with https://stackoverflow.com/questions/23494103/version-cxxabi-1-3-8-not-found-required-by#23494975
07:53 < maarhart> ?
07:54 < Happyhobo> rkhunter and uhide are what I'm getting
07:54 < Happyhobo> I have them now
07:54 < jim> ok, see if you can install one of those, then run it and see what it does
07:55 < jim> maarhart, it -could- be that the compiler you built is too new to use the libs you still have
07:56 < jim> which might mean you'd have to go back to the earlier version of gcc/g++
07:57 < jim> then, to get the version of gcc you want, you should look for a version of your dist (or maybe other dists) that have it
07:58 < maarhart> no, it work when using export LD_LIBRARY_PATH=/usr/bin/gcc-5.3/lib:$LD_LIBRARY_PATH, how can I make this permanent?
07:58 < Ox600df00d> might be able to rebuild gcc and add ./configure --with-default-libstdcxx-abi=gcc4-compatible   but really just a stab in the darkness, i don't really know much about the issue just obscure options
07:58 < jim> maarhart, you could add the path to /etc/ld.so.conf
07:59 < maarhart> I add that line to /etc/ld.so.conf ?
07:59 < jim> which line?
07:59 < jim> look at what's already in the file
07:59 < Happyhobo> I have suspicious shared memory segments, large ones.
08:00 < maarhart> this app will launch from .config/autostart, is it enough if it is in /etc/profile ?
08:00 < maarhart> I mean, do I add the line export LD_LIBRARY_PATH=/usr/bin/gcc-5.3/lib:$LD_LIBRARY_PATH ?
08:01 < Happyhobo> it says I have passwd and group file changes I'm warned that ssh root access and protocol vl is allowed.  I was warned that I may have hidden files and directories.
08:01 < jim> you canj either do that or add this: /usr/bin/gcc-5.3/lib on a line by itself to /etc/ld.so.conf
08:01 < Happyhobo> I have 4 possible rootkits
08:01 < maarhart> jim: what do you mean by 'do that'?
08:02 < jim> maarhart, do as you suggested
08:02 < maarhart> okay, so I add the line '/usr/bin/gcc-5.3/lib' to the end of /etc/ld.so.conf?
08:02 < stevendale> Back ^_^
08:03 < jim> on a line by itself, yes
08:03 < Happyhobo> now I'm running tiger
08:03 < jim> order might matter
08:03 < maarhart> jim: do I need to use include?
08:03 < justanotheruser> what are your thoughts on evrcam? Is it stable and commonly used?
08:03 < revel> Never heard of it.
08:03 < maarhart> because what I see in /etc/ld.so.conf is a line that says include /etc/ld.so.conf.d/*.conf
08:04 < revel> I don't see any search results either...
08:04 < maarhart> okay, I will try adding it to /etc/profile and see if this is enough.
08:04 < jim> maarhart, I'd look at that dir for files, and read the comments in them to see what they did
08:05 < revel> Oh, evercam.
08:05 < Happyhobo> Security report is in `/var/log/tiger/security.report.PCIN.180425-02:01'.
08:05 < Happyhobo>    jim how do I look at this?
08:06 < jim> Happyhobo, look for that file in that dir
08:06 < revel> `less`, image editor of choice, cat...
08:06 < revel> Print it out and read it on paper...
08:06 < stevendale> https://en.wikipedia.org/wiki/Far_Cry_5 Ugh... It's out
08:07 < revel> For about a month, it seems.
08:08 < revel> It seems to be Windows/PS4/XB1-exclusive though, and I don't think this channel is for outdated gaming news.
08:08 < stevendale> https://www.systemrequirementslab.com/cyri/requirements/far-cry-5/15464 https://www.systemrequirementslab.com/cyri/requirements/far-cry-4/12380 System requirements took a big jump
08:08 < jim> Happyhobo, do you see the file?
08:10 < jim> stevendale, maybe they want you to upgrade your cpu, and your ram (maybe not only is it not enough ram, but it's too slow, meaning you'd have to replace ALL of it), and get a nice new $700 graphics card
08:11 < jim> I stopped playing all games for this reason
08:11 < Happyhobo> it's confusing jim
08:11 < Ox600df00d> soon graphics cards will become self aware and build the games for us
08:12 < Happyhobo> there is a user for avahi and nobody, networkma is listening to a few ports, this thing is screwed.
08:12 < jim> maybe also the graphics cards might steal ram and cpu :)
08:12 < Ox600df00d> they can use the "host" ram and cpu as slower backup ram and cpu
08:13 < Happyhobo> how do I list users
08:13 < Happyhobo> lsusers didn't work
08:13 < jim> Happyhobo, look at the file /etc/passwd
08:14 < revel> avahi and nobody having users is fine, afaik.
08:14 < ||JD||> moore's law is dead, hardware upgrade is in most cases a waste of money nowadays
08:15 < revel> Not so sure why NetworkManager should be listening on any ports though.
08:15 < jim> moore's law made hardware upgrades every 2 years or so, worth it
08:15 < revel> Doesn't for me.
08:16 < Happyhobo> https://pastebin.com/aYkw5ZwP
08:16 < jim> Happyhobo, it seems like the things you mention are normal
08:16 < LissajousPattern> even moore new his "law" would plateau
08:17 < Happyhobo> So why is it doing this then?
08:17 < LissajousPattern> isn't it called entropy?
08:17 < Happyhobo> I can chat just fine but as soon as I go to open a couple of pages it goes down and I have to reboot.
08:17 < jim> I don't know... maybe your neighbor is pretty far?
08:18 < jim> more than a couple houses over
08:21 < Happyhobo> townhouse next door
08:21 < ||JD||> chip manufacturers can't keep sinking transistors size forever, with manufacturing process of 10nm they are about to reach a dead end
08:21 < maarhart> I want to install a software, should it be 32 bit or 64 bit? the output of lscpu says the architecture is i686 and the cpu op-mode(s) are 32-bit, 64-bit
08:22 < LissajousPattern> ||JD||, yup entropy
08:22 < Happyhobo> No more ewes gone wild for me, no more shear sheep delight
08:22 < LissajousPattern> sciences best kept secret
08:22 < revel> maarhart: If I understand this right, then you've got an x86_64 processor and you've installed the x86_32 (32-bit) version of whatever distro you have.
08:23 < revel> Unless you plan on reinstalling it as 64-bit, go with 32-bit.
08:25 < trae32566[w]> x86_32 or i686?
08:25 < trae32566[w]> because they _are_ both a thing.
08:26 < maarhart> okay, so what is the veredict? I should install 32-bit apps?
08:27 < revel> trae32566[w]: Did I claim otherwise? Though i386-i586 is also x86_32, aren't they?
08:27 < trae32566[w]> no
08:27 < revel> No?
08:27 < Ox600df00d> nope
08:27 < Ox600df00d> that's "x86"
08:27 < maarhart> uname -a says i686 athlon i686.
08:27 < trae32566[w]> https://en.wikipedia.org/wiki/X32_ABI
08:27 < maarhart> should i install 32 bit apps or 64 bit apps?
08:28 < revel> What? X32 is something else entirely and I doubt he has that.
08:28 < trae32566[w]> revel: x86_32 would be considered x32 afaik, that's why I pointed it out.
08:28 < revel> Since when?
08:28 < revel> https://en.wikipedia.org/wiki/x86_32 points to IA-32 aka i686
08:28 < revel> s/686/386/
08:28 < emr> Hello, i'm getting error when i restart netfilter service here is the log https://paste.ubuntu.com/p/FqYbxW4xPd/
08:29 < emr> any ideas?
08:29 < Happyhobo> jim should i buy a dongle wifi card?
08:30 < [R]> emr: what does it say when you run the failing command manually
08:31 < Ox600df00d> revel: wikpedia is dumb, x86_32 is a special ABI
08:31 < emr> [R] i'm trying iptables-restore /etc/iptables/rules.v4 its not throwing any error
08:31 < trae32566[w]> revel: ...odd. Everything I've seen has said x86 / i686 or x86_64. The only time I've ever seen anything referred to as '32' is with x32
08:31 < Ox600df00d> so you can use x86_64 assembly with 32 bit pointers
08:31 < [R]> emr: where does it say that command failed?
08:31 < revel> Ox600df00d: I know about x32 and I'm pretty sure x86_32 just means 32-bit x86, which is not x32
08:31 < toothe> Is there a command-line json parser?
08:31 < trae32566[w]> x86_32 is not a thing, is basically the bottom line.
08:31 < [R]> toothe: and this parser would do what?
08:32 < emr> in syslog
08:32 < toothe> [R]: search, filter, sort, etc
08:32 < [R]> toothe: python
08:32 < revel> Funny, I'm finding plenty of results for x86_32 online...
08:33 < revel> And they all seem to be referring to it as just regular ol' 32-bit x86
08:33 < trae32566[w]> yeah, not x86_32 :D
08:33 < revel> Search results specifically for x86_32
08:33 < poopBot> hi i want to make backup of rpiHD , so am using dd if=/dev/sdb  /home/user/somefile.img
08:33 < trae32566[w]> just because Google returns a result does not mean that's the equivalent. Perfect example, search for an issue, and then "RHEL 7" after it. You know what you get often times? RHEL 6.
08:34 < Ox600df00d> i guess people do refer to x86 as x86_32 soemtimes, wth
08:34 < trae32566[w]> It doesn't mean RHEL 6 = RHEL 7, it just means Google is trying to figure out what you wanted.
08:34 < bigbike82> How would I go about customizing .conkyrc? I want to change the text color to red
08:34 < revel> trae32566[w]: That definitely means that people have talked about something claled x86_32.
08:34 < [R]> bigbike82: with a text editor
08:34 < Ox600df00d> doesn't help that theres also IA64
08:34 < poopBot> rpiHD has 200gb of wich only 5gb is used, will it make 200gb file or 5-6gb file?
08:34 < [R]> poopBot: dd copies everythign
08:34 < revel> Ox600df00d: Can you find anyone talking about x86_32 as x32? I sure haven't.
08:34 < [R]> poopBot: you probably want rsync
08:34 < bigbike82> [R]: I know that. How do I change the text color? DOesn't seem to be hexcode
08:35 < Ox600df00d> revel: well calling x86 something other than x86 is just stupid and annoying
08:35 < [R]> bigbike82: i'm sure they have documentation
08:35 < poopBot> so it will make 200gb  mybackup.img ?
08:35 < trae32566[w]> ^ agreed
08:35 < [R]> poopBot: it will make an image of the device
08:35 < revel> Okay, but that doesn't mean people don't do it.
08:35 < wyseguy> I added a ssh key to my ubuntu server, I can connect with terminal  and then it asks for my passphrase, but when i try and use a ssh app (gui version) it just wants a password which does not work..., have no option to put in the passphrase
08:36 < revel> poopBot: It'll be as big as the whole disk image. Keep in mind that dd is NOT a good backup tool.
08:36 < poopBot> i prefere not to use rsync cuz that way i need to fdisk create new partritions evrytime i need to reinstall
08:36 < [R]> wyseguy: what "ssh app"
08:36 < revel> If the rest of the disk is just billions and billions of zeroes, then it'll copy all o those to the image as well.
08:36 < [R]> poopBot: you could tyr clonezilla
08:37 < poopBot> that sux , ok will check conzilla
08:37 < wyseguy> [R] Termius
08:37 < trae32566[w]> it's not going to do what they want afaik
08:37 < [R]> wyseguy: sounds like its not using yoru key
08:37 < trae32566[w]> I believe they're wanting a bare metal backup
08:37 < trae32566[w]> ie: block level and immediately restorable
08:37 < [R]> wyseguy: if you insist on using stupid guis... you'll have to learn how to configure them
08:37 < trae32566[w]> what you _could_ do is create a template, then do backups using something like rsync, and if you have to restore, spin the template up and put the data in.
08:38 < wyseguy> [R] nothing makes them stupid lol, its a convenience factor as I have multiple that i connect to
08:38 < [R]> wyseguy: "multiple"?
08:38 < wyseguy> [R] yes
08:38 < [R]> huh?
08:38 < wyseguy> servers
08:38 < [R]> ssh command line can connet to more than one thing...
08:38 < [R]> lol
08:39 < wyseguy> i know
08:39 < [R]> apparently not
08:39 < wyseguy> and this is easier
08:39 < [R]> apparently not as well
08:39 < wyseguy> cool story bro
08:39 < wyseguy> going way off my original question
08:40 < bigbike82> what's a good dark openbox theme?
08:40 < poopBot> beh dont have clonzila in repo , anyone used ccd2iso
08:40 < poopBot> can that be used for hd too
08:40 < wyseguy> figured it out, was under keys section
08:40 < [R]> ccd2iso has nothing to do with imaging hardd rives...
08:41 < poopBot> ok
08:48 < maarhart> how to know if an app, called mpd, is currently running? It should have been launched by ~/.config/autostart
08:49 < [R]> ps
08:49 < poopBot> can partclone clone whole disk or only partritions?
08:49 < Triffid_Hunter> maarhart: ps faux | grep mpd
08:50 < storge> the autostart for mpd might only work if mpd config has been set to allow user to start it
08:50 < [R]> poopBot: sounds like you should read the documentation
08:50 < poopBot> i am reading it mentions only partrition , and i was hoping :)
08:50 < maarhart> Triffid_Hunter: what should it show?
08:50 < poopBot> guess can try and see
08:51 < maarhart> I guess it shows that I'm doing grep
08:52 < Triffid_Hunter> maarhart: if you only see grep, it's not running. if it is running you'll see grep plus the process
08:52 < storge> you can eliminate it showing the grep by bracketing the letter in the search term, like: ps aux | grep [m]pd  ...i think
08:53 < geirha> see if ~/.xsession-errors mentions why it failed
08:53 < maarhart> okay, again, mpd only works if I do export LD_LIBRARY_PATH=/usr/bin/gcc-5.3/lib:$LD_LIBRARY_PATH before running it. so how can I ensure that this happens before ~/.config/autostart is called?
08:54 < maarhart> I added export LD_LIBRARY_PATH=/usr/bin/gcc-5.3/lib:$LD_LIBRARY_PATH to /etc/profile but that didn't seem to do the trick
08:54 < [R]> maarhart: that seems like your system is extremely broken...
08:54 < storge> i seem to recall mpd not letting me start it as a user unless i specifically set it in the config, i'm assuming you did that?
08:54 < poopBot> yep partclone is only for partritions
08:54 < geirha> Exec=env LD_LIBRARY_PATH=/usr/bin/gcc-5.3/lib mpd
08:54 < maarhart> nothing is broken in my system. I just compiled it with a newer version of gcc than the one that came from apt
08:55 < [R]>  /usr/bin/gcc-5.3 is just wrong on so many levels
08:55 < maarhart> geirha: do I need to add line that somewhere or just run it?
08:55 < [R]> plus, if 5.3 is the "new" version... how the hell old is your system!?
08:55 < [R]> 20 years old?
08:55 < maarhart> it's linux mint 17.1
08:55 < geirha> maarhart: you'd put it in the desktop file
08:56 < sauvin> maarhart, what distro are you using?
08:56 < maarhart> thanks geirha
08:56 < sauvin> Argh... where did you install your newer gcc?
08:56 < poopBot> guess will just boot clonzila from usb
08:56 < maarhart> geirha, you always have the right answer.
08:58 < sauvin> maarhart, please answer: do I understand correctly that you installed a newer version of gcc than what your revision level of Mint supplies from repos?
08:59 < maarhart> sauvin: that's right
08:59 < maarhart> I needed a version that would work with C++14
08:59 < sauvin> maarhart, where did you install it?
09:00 < maarhart> geirha: I have to say that I admire you. This is not the first time that you understand the question properly and give the right answer, instead of telling me how much I suck, how crappy my system is, and so forth.
09:00 < maarhart> sauvin: I just built it in any random folder
09:00 < Sitri> maarhart: what was the ./configure line you used to compile gcc?
09:01 < sauvin> maarhart, you didn't install it to /usr ?
09:01 < maarhart> Sitri: I guess that the line was ./configure
09:01 < maarhart> I couldn't remember
09:01 < sauvin> o.O
09:01 < sauvin> If you DID install it to anywhere in /usr, you may very well have broken your system.
09:01 < maarhart> I couldn't care less. I just needed this version of gcc for that specific app. Nothing else will ever be installed in this computer
09:02 < sauvin> Ah. In that case, it doesn't matter. :D
09:02 < stevendale> ./configure --prefix=/usr
09:02 < Sitri> If you're only using it for the one program, you may as well setup your ld.so properly then...
09:03 < Sitri> Add /usr/bin/gcc-5.3/lib/ to /etc/ld.so.conf, then run ldconfig
09:05 < [R]> yeah, thats a great way to break the system
09:06 < storge> you suck and your system is crappy, and so forth.
09:06 < storge> i never saw anyone say that
09:06 < storge> well not yet today anyway
09:07 < maarhart> storge: well, no, nobody said that, but you know, it sometimes feel like that
09:07 < maarhart> feels
09:08 < Mo> Hi, is anybody on 4.16.x able to modprobe bcache? I got could not insert 'bcache': Cannot allocate memory
09:08 < maarhart> and I get a sea of questions and comments coming from others... but always geirha comes up with the answer <3
09:08 < maarhart> okay bye
09:08 < maarhart> thanks guys
09:09 < storge> if only there was a digital currency you could tip geirha with that wasn't ruined by speculators
09:09 < sauvin> Your digital currency is broken! You suck!
09:09 < storge> my system sucks so bad i haven't been able to mine it since 2011
09:10 < storge> and so forth
09:24 < jpsollie> hello everyone, I am looking for an IRC channel to discuss with the linux developers about a possible PCI bug (which shows up in my rather unusual system, and as such may require some explanations).  Am I at the right place?
09:27 < oneko> jpsollie: Not so sure a lot of people here are kernel developers :-P. I guess you could ask but you'd have better luck asking in a more kernel related channel
09:27 < Sitri> jpsollie: I think ##Kernel exists for that
09:28 < jpsollie> okay, thanks :)
09:37 < Triffid_Hunter> jpsollie: kernel mailing list would give best results
09:39 < jpsollie> yes, but I have no urgency or whatever, it's just that the setup is complex, and talking to someone is easier for answering 10 questions about how he understands the situation
09:40 < jpsollie> with the mailing list, I have the risk that I need to send 50 mails before my bug is clear
09:40 < jpsollie> it's easier to try chan at this point :)
09:40 < jpsollie> sorry, chat
09:41 < ||JD||> the only risk you have at lkml.org is that you get insulted to death, besides that you won't get better help in any other place
09:42 < toffe> Whats up guys :)
09:43 < toffe> Anyone know how i could easily parse out XX from the following string? VERSION="XX"
09:43 < toffe>  sed -n 's/VERSION="/\\1/p'
09:43 < toffe> I tried that but that gives me \1XX"
09:43 < sauvin> toffe, is "XX" constant?
09:44 < toffe> sauvin: No, its a version number
09:45 < pepermuntjes> sauvin, tr -d X
09:45 < sauvin> I'm not a sed guy, I'm a perl guy, so take me with a grain: I'd be tempted to try something like s/(VERSION=)\"[^"]+\"/$1/
09:46 < sauvin> If pepermuntjes' idea is relevant (I don't know command line tr either), it's simpler.
09:46 < toffe> I also tried TR now. that worked. echo "VERSION=\"1.2\"" | tr -d "VERSION=\""
09:46 < toffe> it removed VERSION and = and all "
09:48 < toffe> Thanks guys :)
09:51 < Triffid_Hunter> toffe: cut -d\" -f2
09:57 < Li> sleeping channel
09:58 < Li> trying to play an alert every specific minute of every hour, added this line into crontab 52 * * * * mplayer /absolutePath/alarm.mp3 , then #systemctl restart cron.service, tried # service cron reload ,, it doesn't work .. any suggestions/corrections to what I'm doing wrong?
09:59 < toffe> Triffid_Hunter: care to explain? I like it, short and sweat, just like my girl
10:00 < toffe> oh its like a csv, using " as delimiter, selecting field 2
10:01 < toffe> thanks Triffid_Hunter  :)
10:02 < pppingme> Li What do the logs say? cron logs everything
10:05 < Triffid_Hunter> toffe: well 'cut' is a tool for splitting lines into fields delimited by some character, then returning a specific field or group of fields.. I just told it that " is your delimiter and to emit field 2
10:05 < Li> pppingme: can't find any /var/log/cron
10:06 < Li> however the service status shows the new changes
10:06 < Triffid_Hunter> toffe: heh yeah somehow missed your last line there
10:06 < pppingme> Li where did you add it?  in /etc/crontab? or somewhere else?
10:06 < Li> pppingme: crontab -e
10:07 < pppingme> Li ok, were you a normal user or root when you did that?
10:09 < Li> pppingme: normal user
10:10 < Li> it allows to edit as a user
10:10 < pppingme> Li what linux distrib is this?
10:11 < TheDcoder> Hello guys, I am trying to install Windows in a virtual machine. The problem is that the main hard disk image for the virtual machine is on a mounted NTFS partition (my large storage partition) and it looks like qemu has troubles accessing the qcow2 file, any recommendations?
10:12 < TheDcoder> Not sure if I can directly modify the permissions of the qcow2 file since it is in a NTFS partition
10:13 < Li> pppingme: debian .. now I've added 1 * * * * root mplayer <paht>/alarm.mp3
10:14 < pppingme> TheDcoder are you getting a specific error or what?
10:14 < Li> into /etc/crontab && reload but nothing is happening again
10:14 < forgon> Could you recommend me utilities for GUI testing in Linux?
10:14 < TheDcoder> "Could not open '/home/TheDcoder/Data/Storage/Fedora/VM/win10.qcow2': Permission denied"
10:14 < revel> Li: What about if you use your own user?
10:14 < forgon> I know xdotool, but writing scripts with it is rathre hackish.
10:14 < pppingme> Li I don't think mplayer is going to work for you, I'm about 75% sure it requires a graphical environment, which is not available from cron..
10:15 < TheDcoder> That is the error I am getting pppingme
10:15 < revel> Oh, and use something that can work without graphics maybe, yeah, like mpv
10:15 < revel> If mplayer does indeed require them.
10:15 < forgon> pppingme: Does it? I never use mplayer with GUI.
10:15 < revel> If it works without graphics, then it should be fine, I got mpv working with cron.
10:15 < TheDcoder> mpv should be able to function without GUI
10:15 < pppingme> you'll probably need to use aplay or something else
10:16 < revel> Didn't try with root though, just my own user.
10:16 < Li> pppingme: what command should I put instead to just have a pop message?
10:17 < forgon> pppingme: `mplayer music.ogg -novideo -really-quiet < /dev/null &` works well for me (assuming something is to be automated)
10:17 < revel> Li: libnotify-notify-send
10:17 < pppingme> if you can convert it to a raw .wav file, you could get real simple and just do "cat audiofile > /dev/pcsp" or something along those lines (won't work for .mp3)
10:18 < revel> Though I'm not so sure if that would work.
10:18 < pppingme> forgon just gave a non-gui example of mplayer it looks like..
10:18 < revel> I'm guessing it's just because Li tried to have root run it.
10:19 < Li> 1 * * * *       root     mplayer /usr/share/sounds/speech-dispatcher/test.wav
10:19 < Li> pppingme: I've added xmessage HELLO
10:19 < revel> Li: pls try your own user.
10:22 < Li> Ok I think I got my mistake .. I'm supposed to use */1 * * * * to play that every minute, instead I was using 1 * * * * which means every first minute of every hour
10:22 < Li> lesson learnt
10:23 < pppingme> */1 would be redundant, just plain * will work if you intend every minute
10:23 < Li> however it doesn't send the xmessage
10:23 < Li> pppingme: thanks for the explanation that's even better
10:28 < revel> I thought you were replacing the 1 with the next minute, i.e "29" if you'd do it now.
10:28 < revel> Since I guess I'd do that instead of *
10:28 < pppingme> Li the /X means every increment, so /2 would run at (assuming minutes) 00, 02, 04, and so on..
10:29 < pppingme> don't confuse with division..  for month /2 would do 1, 3, 5, not 2, 4, 6...
10:31 < Li> pppingme: yes I get it, but using * is not playing the alert at all
10:31 < pppingme> then for some reason the command probably isn't executing..
10:31 < pppingme> so back to check cron logs
10:32 < pppingme> typical problems are pathing (path EVERYTHING), need for X, ncurses, etc (won't run from cron), or permissions
10:34 < Li> pppingme: checked again and there are no /var/log/cron logs
10:34 < forgon> Could you recommend me utilities for GUI testing in Linux?
10:34 < forgon> I know xdotool, but writing scripts with it is rathre hackish.
10:34 < Li> the same permission/command/path give an alert when I use */1 * * * *
10:40 < rangergord> I have a bunch of patch files (from diff) I'd like to apply to a directory. Is there any tool to help me do this interactively, i.e. let me visualize and confirm the changes line by line?
10:40 < rangergord> similar to git add -p, or a GUI tool
10:41 < rangergord> I tried Kompare and couldn't make sense of it
10:41 < Li> well it works now, but it takes 4 seconds before playing the alarm
10:42 < oneko> I'm not sure it's possible to apply a patch line by line but I thought you'd be able to edit hunks with --dry-run option
10:43 < pppingme> Li cron isn't guaranteed to the second
10:46 < acetakwas> How can I check the used RAM for a process if I know its process name?
10:46 < acetakwas> Using a script^? (I know I can manually inspect this with `top`.)
10:48 < pppingme> acetakwas you're just wanting to know if something is running?
10:48 < acetakwas> pppingme::  I want to know the amount (percentage) of RAM used by the process.
10:49 < acetakwas> The ultimate goal is to run a CRON job that does something when the RAM usage gets to a certain threshold.
10:49 < pppingme> acetakwas look at "free"
10:56 < toffe> Quick question, I'm creating a RPM package which installs a bash script as an executable on /usr/local/sbin/myscript i also add /etc/cront.d/myscript.cron and /var/log/myscript.log
10:56 < toffe> But I've come to the need for a configuration file.
10:56 < toffe> What would be the preffered location for it?
10:58 < acetakwas> pppingme::  That doesn't do what I want.
11:02 < djph> toffe: either /etc or $HOME/.config
11:02 < toffe> /etc/myscript.conf ?
11:02 < djph> acetakwas: that's not what cron does.
11:02 < toffe> would be acceptable?
11:02 < djph> toffe: yup
11:02 < acetakwas> djph::  How do you mean?
11:02 < acetakwas> My question is not about what a CRON job does.
11:03 < toffe> djph: Thanks mate
11:04 < acetakwas> I am asking about how to programmatically determine how much RAM a process has used up.
11:04 < dogbert2> hey djph
11:04 < djph> acetakwas: cron is a scheduler based on *time*.  It will not kick off a job based on RAM usage.
11:04 < acetakwas> djph::  I know that. And I know clearly what I want to do.
11:04 < acetakwas> Please address the question if you can.
11:04 < acetakwas> Thanks.
11:04 < toffe> another quick one, my config is also a bash script with variables, when sourcing it give : command not found for every blank line i have
11:05 < Ben64> <acetakwas> The ultimate goal is to run a CRON job that does something when the RAM usage gets to a certain threshold.
11:05 < djph> toffe: err
11:05 < Ben64> well that's not a thing that can be done
11:05 < djph> ^
11:05 < Ben64> because cron works off of time
11:05 < toffe> Is that because it executes _every_ line as a function with source command?
11:05 < acetakwas> Ben64::  That is based on my `if` condition in my script.
11:06 < acetakwas> Guys please address the question. I know what CRON jobs do.
11:06 < dogbert2> yawn...
11:07 < djph> toffe: generally a config file is something that'd be read, not sourced.
11:07 < sexystud> cock penis weiner dick dong
11:07 < LissajousPattern> wha wha what?
11:08 < toffe> djph: yeah, found the error trough it was some hidden characters on every newline ^m . I'll look onto a read command for configurations
11:08 < sexystud> LissajousPattern: do you like nigger vagina?
11:08 < TheDcoder> !ops sexystud too horny
11:08 < sexystud> !ops thedog3catch3r nigger faggit
11:08 < djph> acetakwas: anyway, if you want to see how much ram a process is using, ps aux (RSS column)
11:09 < sauvin> TheDcoder, good yell.
11:09 < TheDcoder> :P
11:09 < dogbert2> he hit the upright going out the door
11:09 < sauvin> Yup. Freenode got him before I did.
11:09 < sauvin> I *assume* that was a "him", anyway.
11:10 < Ben64> fairly safe
11:10 < LissajousPattern> yeah he was just *poof*
11:10 < uptime> sauvin: he was your old friend geeknerd!
11:10 < sauvin> Oh? He's gotten over his fascination with fecal matter?
11:11 < LissajousPattern> dang all the odlies crawlin out the woodwork
11:18 < pepsi_power> hello I am listening with netcat on port 80 and when I connect from lan from the browser an http request is sent (as expected) but if I connect from my external IP address it doesn't work.
11:20 < Ben64> pepsi_power: check firewall / nat / port forwarding
11:22 < pppingme> pepsi_power define "connect from my external IP"  do you mean you hit your wan ip from a device on your lan, or you hit y our wan ip from a device completely outside your network?
11:23 < pepsi_power> pppingme: I hit the external IP address of the machine running netcat with another device (on a completely another network)
11:24 < pppingme> then you either don't have it forwarded correctly, or you don't really know your wan ip..  this is assuming you are testing from 100% completely outside of your network and not just hitting your wan/outside ip from another pc on  your lan
11:24 < pepsi_power> Ben64: port forwarding should be ok, going to check nat and firewall
11:25 < pepsi_power> pppingme: I am using mobile data so it should be on another network, no?
11:25 < pppingme> yeah
11:25 < Ben64> i wouldn't assume port forwarding is ok then
11:26 < pppingme> yep, either port forwarding, or you don't really know your wan/outside ip
11:27 < CrazyTux> what is the difference between BSD and Linux?
11:27 < pepsi_power> pppingme: I googled my external IP address
11:27 < pepsi_power> Ben64: ok rechecking port forwarding
11:27 < pepermuntjes> CrazyTux, no hugs with BSD
11:27 < pppingme> pepsi_power you need to see what your ROUTER thinks your wan/outside ip is, don't use some check site..
11:27 < djph> ^
11:27 < CrazyTux> are BSD based distros as stable and userfriendly as those based on Linux?
11:28 < pepermuntjes> CrazyTux, no
11:28 < pppingme> CrazyTux no, yes
11:28 < djph> pepsi_power: if you're on a mobile data, chances are fairly high that it's CGNAT
11:28 < pepsi_power> djph: no only the other device is on mobile data
11:28 < pepsi_power> pppingme: how do I do this then?
11:28 < CrazyTux> I presume there aren't many well supported BSD based distros.
11:29 < pppingme> pepsi_power like I just asked, what IP does your ROUTER report?
11:29 < pepsi_power> pppingme: how do I check what IP address router reports?
11:29 < djph> pepsi_power: so you're trying to connect from mobile_data -> internet -> your router / whatever's behind it.
11:29 < pppingme> you look at your routers status page
11:30 < pepsi_power> djph:  yes
11:30 < pepsi_power> pppingme: the router reports the same IP address
11:30 < sauvin> BSDs aren't really "distros". The BSDs use a BSD kernel, which is internally very different from a Linux kernel, and the userlands between the BSDs and Linux distros are also very different.
11:31 < pppingme> pepsi_power is that IP 89.44.92.166 or something else?
11:31 < pepsi_power> yes the one you just wrote
11:31 < CrazyTux> ok
11:31 < sauvin> I forget what the different BSDs are called, but they're not "distros".
11:31 < djph> and what service are you trying to host, on what port?
11:31 < CrazyTux> ok
11:31 < pppingme> pepsi_power then either your isp is blocking port 80, or you simply don't have it forwarded properly
11:32 < CrazyTux> are BSDs as usable and userfriendly as Windows?
11:32 < anzipex> How to fix "Open with" in nautilus? gedit program doesn't show there
11:32 < anzipex>  And i also can't delete LibreOffice Writer and notepad from that menu
11:33 < pepsi_power> on port forwarding there is an option "wan host IP address" which is by default 0.0.0.0 should I change this?
11:33 < djph> that *should* be understood by the router as "anything"
11:33 < anzipex> I have Ubuntu 14.04 x86
11:34 < pppingme> pepsi_power probably ok, whats the rest of the info?
11:34 < pepsi_power> lan host: here I have the lan IP address of the machine running netcat
11:35 < pepsi_power> wan port / lan port here I have the same port that netcat is listening to
11:35 < pepsi_power> protocol TCP
11:36 < pppingme> pepsi_power are you *SURE* the server (the one running netcat) doesn't have a firewall blocking/dropping the traffic?
11:37 < djph> pppingme: not to mention "the ISP allows inbound port80"
11:37 < stevendale> Hey
11:37 < sauvin> CrazyTux, the BSDs' usability and friendliness depend largely on who you ask, but I doubt Windows users would be comfortable with it at first.
11:38 < sauvin> *them, anyway.
11:38 < CrazyTux> sauvin, ok
11:38 < stevendale> I put an NVIDIA GeForce 8400 GS 256 MB in my desktop, using it now instead of the onboard VGA Intel HD Graphics Ironlake
11:38 < stevendale> So much faster
11:40 < pepsi_power> pppingme: stopped the firewall service and it worked i am dumb
11:40 < pepermuntjes> CrazyTux, the only reason i personally would use BSD is FOMO (fear of missing out).
11:41 < CrazyTux> ok
11:41 < pepsi_power> note: netcat dies if a request is sent a second time is there a way to keep it alive?
11:41 < pepermuntjes> i think BSD you are limited to thinkpad laptops, 'becaue thats what their devs use'
11:41 < pepermuntjes> *with
11:48 < pepsi_power> nevermind, I thought I was typing -k but I wasn't
12:12 < lopid> apps that place themselves into the system menu using the company name first, instead of the product name
12:13 < lopid> spent three days looking for "virtualbox" when i should have been looking for "oracle"
12:14 < LissajousPattern> what?
12:17 < CoJaBo> Is there any way to get nmi_watchdog option working? It does not appear to be generating NMI interrupts, and I can't find anything saying why this would be the case; system is Ryzen, X370 chipset
12:18 < Alexander-47u> jim, you there bro ;)?
12:18 < CrazyTux> is the current version of lxqt stable and usable by end users? can it be safely installed on a distrol like Kubuntu?
12:19 < CrazyTux> Kubuntu LTS.
12:27 < alxtb> How do you find where an alias is set?
12:27 < alxtb> I've looked in all the .bashrc files and can't find where an alias is set
12:27 < alxtb> and unalias isn't working
12:30 < Ben_1> because of my dumbness I moved the /etc/pam.d/system-auth-ac file. Is there a way to move it back to that dir? At the moment I can not use sudo anymore
12:30 < ruby4diam> hi all! please, i could use a hint. i have an rsync command inside bash script with a variable inside containing files names. when expanded, it is wrapped inside single quotes, which prevents the rsync from functioning (no such file or directory). when removed, all works perfectly. how can i avoid them? i used bash debugging options and echoed the command inside the script before actual triggering of
12:31 < ruby4diam> the command and echo outputs without quotes, rsync with quotes. same with cp. thank you very much
12:31 < rypervenche> alxtb: grep -R 'aliasname' /etc/ (use the actual alias name)
12:31 < Sitri> Ben_1: Can you `su -`?  Otherwise you'll have to boot into a rescue disk.
12:32 < alxtb> rypervenche, thanks
12:32 < rypervenche> alxtb: Depending on what it is it might be in /etc/profile or /etc/profile.d, or in your home directory in ~/.bashrc or ~/.bash_aliases
12:32 < alxtb> rypervenche, exactly what's been confusing me :)
12:32 < ruby4diam> example: echo 'rsync files dir' whereas rsync 'files' dir
12:33 < stevendale> :3
12:33 < Ben_1> Sitri:  no access denies
12:34 < Ben_1> *denied
12:35 < rypervenche> ruby4diam: Can you give us an example of the bash script? Why is it all wrapped in quotes?
12:36 < ruby4diam> for i in $files; do
12:36 < ruby4diam> 	current_media_len=`mediainfo "$i" | grep Duration | awk 'NR==1 { print $3/60 + $5/3600 }'`
12:36 < ruby4diam> 	media_len_sum=`echo $media_len_sum + $current_media_len | bc`
12:36 < ruby4diam> 	if (( $( echo "${media_len_sum}>=$1" | bc -l ) )); then
12:36 < ruby4diam> 		files_to_move+=`printf '%q' "$i"`
12:36 < ruby4diam> 		break
12:36 < ruby4diam> 	else
12:36 < ruby4diam> 		files_to_move+=`printf '%q ' "$i"`
12:36 < ruby4diam> 	fi
12:36 < ruby4diam> done
12:36 < ruby4diam> echo $files_to_move
12:36 < ruby4diam> echo ""
12:37 < ruby4diam> echo "rsync -vurn --delete $files_to_move $sd_card"
12:37 < ruby4diam> rsync -vurn --delete $files_to_move $sd_card
12:37 < CoolerY> whats a good textbook for these kind of things? https://screenshots.firefox.com/C5HRFRE9mKl2lPex/www.cse.iitk.ac.in
12:38 < ruby4diam> rypervenche: thats what im trynna find out - why its wrapped in quotes
12:38 < CoolerY> does a process have to enter the ready queue only for CPU bursts? or is there a ready queue for I/O bursts as well?
12:40 < ruby4diam> any ideas?
12:41 < rypervenche> ruby4diam: And you give an example of the output from the files_to_move variable?
12:41 < rypervenche> ruby4diam: Also, next time please use a pastebin website for more than 2 lines of code.
12:43 < ruby4diam> rypervenche: here https://pastebin.com/h44yLBr7
12:44 < ruby4diam> rypervenche: that is a sample of bash -x script
12:46 < rypervenche> ruby4diam: Can you try adding -s to your rsync command?
12:46 < geirha> ruby4diam: you're assuming bash expands quotes recursively. It doesn't
12:47 < ruby4diam> rypervenche: same
12:47 < geirha> Use arrays to hold multiple filenames. Don't try to mash them into strings
12:47 < ruby4diam> geirha: that is a good idea, thought i assumed this should be possible
12:48 < geirha> for file in "${files[@]}"; do ... files_to_move+=( "$file" )
12:49 < geirha> rsync ... "${files_to_move[@]}" "$sd_card"   # including all the quotes
12:50 < rypervenche> That's much better. I need to use arrays more often.
12:50 < geirha> it's either the clean way with arrays, or the eval route, which could easily turn bad
12:56 < Rembo> hello, i want to see in octal/numeric value of drwxr-x---  , can someone help?
12:56 < rypervenche> Rembo: 750
12:58 < ruby4diam> geirha: can "${files[@]}" be used on files=`find $cwd -type f | shuf` ?
12:59 < ruby4diam> this is the whole altered code: https://pastebin.com/Uw5T9tF3
12:59 < geirha> no, that's broken
13:00 < geirha> mapfile -d '' -t files < <(find "$cwd" -type f -print0 | shuf -z)
13:00 < ruby4diam> in what way?
13:00 < geirha> assuming you have GNU utilities
13:01 < ruby4diam> sorry too late :)
13:01 < ruby4diam> this transforms it into array?
13:01 < geirha> and mapfile -d requires the latest version of bash, unfortunately. For older versions you can use a while read loop
13:02 < geirha> while IFS= read -rd '' file; do ...; done < <(find ... -print0 | shuf -z)
13:02 < geirha> mapfile creates an array, yes
13:02 < ruby4diam> i do have mapfile
13:02 < geirha> yeah, but -d was added in 4.4
13:03 < ruby4diam> how do i get the version?
13:04 < geirha> printf '%s\n' "$BASH_VERSION"
13:04 < revel> Or just `echo $BASH_VERSION`, or `bash --version`
13:04 < ruby4diam> 4.4.12
13:05 < geirha> ok then, you can use the mapfile version
13:06 < geirha> and you can run   declare -p files   to see that each element of the array is a single filename
13:07 < ruby4diam> i have an error somewhere...
13:09 < ruby4diam> files declaration doesnt work
13:10 < geirha> by that you mean ...?
13:10 < ruby4diam> no value assigned. is this ok? files=`mapfile -d '' -t files < <(find "$cwd" -type f -print0 | shuf -z)`
13:11 < geirha> no
13:11 < geirha> remove the surrounding files=`...`
13:11 < ruby4diam> do i need to substitude them?
13:11 < geirha> just remove them
13:11 < ruby4diam> files=mapfile
13:12 < ruby4diam> bash -x says that
13:12 < geirha> no, mapfile is a command
13:12 < ruby4diam> i understand that
13:12 < geirha> mapfile -d '' -t files < <(find "$cwd" -type f -print0 | shuf -z)
13:12 < geirha> it has no output
13:12 < ruby4diam> oh
13:13 < ruby4diam> so the -t files is the vriable right?
13:13 < ruby4diam> variable
13:13 < pepermuntjes> doesnt -t stands for type?
13:13 < geirha> ''help mapfile''
13:15 < ruby4diam> alright so w/out the -t
13:15 < ruby4diam> just files is the var
13:15 < ruby4diam> i get it now
13:16 < geirha> the -t isn't strictly necessary when you have -d '', but you usually want -t
13:17 < ruby4diam> thank you very much, it works perfectly. i appreciate your help. will have a closer look at the logic so i will be able to use it in the future. thank y'all :)
13:22 < BluesKaj> Howdy folks
13:22 < badboyjer> howdy bud
13:34 < ruby4diam> there's just one more thing, the files array begins with an empty element. any ide why?
13:35 < ruby4diam> probably know why :P
13:36 < ruby4diam> yep. the variable was initiated with an empty value. thanks anyway
14:15 < noodlepie> MIDI is wonderful. I love linux.
14:17  * noodlepie has a General MIDI 2.0 synth - a Roland GAIA SH-01. Its fully Linux compatible and sounds boss due to its high quality DSP and decent soundfonts. A wonderful peice of kit.
14:17 < noodlepie> Now, it only Sony MiniDisc support was of a similar level
14:17 < noodlepie> NetMD should be open tech
14:24 < mawk> hi
14:24 < noodlepie> Hi mawk
14:24 < mawk> my MawkVPN© is making great progress
14:25 < noodlepie> How has your day been? I hope it's looking up.
14:25 < mawk> I just woke up
14:25 < noodlepie> hehe, sleepy head
14:25 < mawk> it's 2:25 PM here
14:26 < mawk> and it's been productive for you ?
14:27 < noodlepie> Yeah, productive is the word I was looking for (: <grin>
14:27 < mawk> lol
14:28 < noodlepie> I can't wait for the HURD to be stable. The user filesystem mount apps wound let you create network pipes really easily
14:28 < mawk> like unix sockets ?
14:31 < artiken> Hi noodlepie, I hear you.
14:33 < artiken> Anyone know how to set the nice(priority) level from -20 for system threads to something more....ah...less hoggy, like -10 or -5. Of course I would like the settings to remain between boots.
14:38 < catphish> is there any way in linux i can have 2 servers, each with a block device, and have both serve the same block device by iscsi, and sync reads + writes between the 2? it seems like it should be simple, but never found a way :(
14:38 < DLange> catphish: https://en.wikipedia.org/wiki/Distributed_Replicated_Block_Device
14:39 < ananke> drbd
14:39 < DLange> (obviously don't cache on iscsi layer :))
14:39 < ananke> ohh, I see DLange already linked it
14:39 < DLange> o/
14:39 < CoJaBo> how do I figure out why nmi_watchdog doesn't work; is it yet another ryzen issue, or something else?
14:39 < catphish> DLange: interesting, for some reason i was convinces drbd couldn't do this
14:40 < ananke> catphish: problem is not in syncing the underlying block device. problem is coordinating it on whatever laid on top of it
14:40 < catphish> although, i think there might be a problem with IO ordering in this setup
14:41 < ananke> catphish: what's going on top of the iscsi lun?
14:41 < catphish> multipath, then VMs
14:41 < ananke> catphish: define 'VMs'
14:41 < catphish> well, multipath, then lvm, then VMs
14:41 < ananke> so for starters, you'd need clvm
14:42 < catphish> LVM, where each volume is presented to a virtual machine as its disk
14:42 < catphish> not necessarily, but that part isn't the problem
14:42 < ananke> catphish: problem is with your expectations
14:42 < catphish> how so?
14:43 < catphish> i believe the problem is the active-active nature of multipath and the order of the operations
14:43 < ananke> catphish: you're expecting the underlying layers to take care of synchronization and replication, without the top layers being aware of it
14:44 < catphish> i don't understand why the top layers would care, the layers that matter are the multipath, iscsi and drbd, right?
14:44 < ananke> catphish: perhaps I'm missing something from your picture. are those two separate iscsi servers providing the same LUN to a _single_ initiator?
14:44 < catphish> ananke: yes
14:44 < ananke> catphish: ahh, then drbd should be sufficient
14:44 < catphish> there may be multiple initiators, but they wouldn't be accessing the same blocks, so i don't think that's important to the conversation
14:45 < ananke> catphish: it's not, but it wasn't explicitly stated in your initial description, so assumptions were made to the contrary
14:46 < catphish> ananke: would iscsi and drbd properly handle the ordering of the requests?
14:46 < ananke> I thought you had multiple hypervisors accessing a pair of iscsi systems
14:46 < rypervenche> Relevant: https://xkcd.com/1339/
14:47 < jken> hello, I am not sure where to ask this question so direct me to the proper channel if this is not applicable here. I am bootstrapping a raw disk image with a tool called mkosi. The image is for a device with a 30GB ssd. However, I don't want my image file to be 30GB. Does anyone know if there is a way to "shrink" raw disk images but have them "grow" again when written to a disk? Or perhaps another method for acheiving similar results. I am trying to avoid
14:47 < jken> having to expand the / partition after imaging the disk.
14:47 < ananke> catphish: not sure why you'd even bother doing active/active multipathing in this scenario, since it won't gain you performance
14:47 < catphish> ananke: well i do, but the multiple clients part works fine :)
14:47 < mawk> having the image sparse jken
14:48 < mawk> ls will tell the image is 30 GiB, but du will tell you the real underlying size
14:48 < rypervenche> jken: https://wiki.archlinux.org/index.php/sparse_file
14:48 < mawk> copy it around with cp --sparse at all times, to not lose that property
14:48 < mawk> also burn it to the drive with the sparse option of du
14:48 < mawk> of dd*
14:48 < catphish> ananke: that's a good question, and you're right, active-active is totally pointless, the only reason was that i wanted clients to maintain consistent connections to both endpoints, because that's what i'm used to
14:49 < jken> mawk, thank you! "sparse" was the term I needed to make google give me results about what I wanted. Thanks for the info.
14:50 < catphish> ananke: should i be considering dropping multipath for the failover and just using vrrp and making the client reconnect to the backup host/
14:51 < catphish> i think that makes planned failover rather more painful
14:51 < rypervenche> jken: You can either dd the image to the SSD and convert it to sparse on the way, or convert to sparse first and then copy and keep it sparse. Note that if you use dd, you'll want to keep your bs=512
14:51 < ananke> catphish: not sure how iscsi initiator in linux works with vrrp, you'd have to do some research on that. you could just use active/passive multipath setup
14:52 < catphish> does multipath support that? who decides which path is active?
14:52 < rypervenche> jken: You can also just cp --sparse=always to do it all in one command though. But yeah, you might want to fallocate it first on your source machine if you want to save some space.
14:53 < ananke> catphish: there are a few algorithms
14:53 < catphish> seems there are options, i do wonder if drbd+iscsi+multipath would work, or whether it would mess up the order of the IOs
14:54 < mawk> why doesn't sudo put the previous UID in the saved UID ?
14:54 < ananke> mawk: define 'saved UID'
14:54 < catphish> maybe i should actually have a look at the (i)SCSI protocol and understand it a bit better
14:54 < catphish> ananke: thanks, i'll have a look at ways to do active-passive too
14:54 < jken> rypervenche, thanks!
14:55 < kazdax> if i install an application as root
14:55 < kazdax> does it get installed on every other user also ?
14:55 < rypervenche> jken: I think I may have said that wrong. I don't think cp will convert it, but it will keep it. But you have all of the commands in that wiki page.
14:56 < kazdax> and what about the other ways
14:56 < rypervenche> kazdax: What kind of application? From the distribution's repositories?
14:56 < kazdax> yes
14:56 < jken> reviewing the arch wiki page now, seems mkosi already generates sparse images and I just didn't know how to use them properly.
14:56 < catphish> what happened to drbd :|
14:56 < kazdax> this is a debian distro
14:57 < rypervenche> kazdax: It will usually be installed in a location that can be run by all users.
14:57 < kazdax> okay i see
14:57 < mawk> ananke: there is the uid, the effective uid, and the saved uid
14:57 < mawk> it's just a number intrinsic to each process
14:57 < kazdax> what if i make my current user part of the sudo group
14:57 < kazdax> is that bad practice ?
14:57 < rypervenche> kazdax: Sometimes the binary will not be in your PATH variable, so you either have to be root or use sudo, or use the full path.
14:57 < mawk> see getresuid()
14:57 < catphish> ah, found the docs now, it went a bit corporate
14:57 < mawk> ah no sorry, I misunderstood that
14:57 < rypervenche> kazdax: What are you ultimately trying to accomplish here?
14:58 < mawk> or maybe not, the docs are confusing
14:58 < kazdax> just that i dont install always by logging into root
14:58 < mawk> it's something about an UID that is saved, that's for sure
14:58 < kazdax> that i can do it with a normal user with root privilages
14:58 < ananke> mawk: the docs seem to be very sparse
14:58 < rypervenche> kazdax: Yes, that is a good practice. You should only use root/sudo for when you need it.
14:58 < mawk> yeah
14:58 < ananke> mawk: what are you trying to accomplish?
14:58 < kazdax> ahh i see ..like for installing apps for example
14:58 < kazdax> i see
14:58 < rypervenche> kazdax: Correct.
14:59 < mawk> get the uid before sudo took place
14:59 < rypervenche> kazdax: I personally like to become root for such things with "sudo -i" or "su -", depending on which system I'm using, but others like to use "sudo" before each command. It's up to you how you wish to do it. :)
14:59 < mawk> but that's in the SUDO_UID env variable
15:00 < ananke> mawk: yep
15:01 < mawk> some docs talk about "saved set-user-ID", some other talk about "saved UID"
15:02 < mawk> ok, the saved UID is used with setuid programs to save a copy of the EUID, to be able to lose and gain privileges at will
15:02 < mawk> but what's the point of losing privileges if you can gain them immediately after
15:02 < mawk> where is the security in that
15:02 < mawk> credentials(7) talk about it
15:09 < xubunto> is there a way that I could combine these ssh config settings? https://paste.linux.community/view/c4e7f580
15:09 < mawk> yes
15:09 < mawk> using wildcards
15:10 < xubunto> what would that look like?
15:11 < JimBuntu> Make a new Host entry, as 'Host *' and apply any settings you want to use for all other entries to it
15:11 < JimBuntu> xubunto, ^^
15:11 < Psi-Jack> xubunto: Also, look into ssh's -J option, which would be so much better than -W
15:12 < xubunto> jim, I have that at the top of my config file already
15:12 < mawk> Host spl-as-* something something
15:12 < xubunto> i found canonical domain which seems promising
15:13 < mawk> then use %h and it will be replaced by spl-as-foobar
15:13 < mawk> hence factoring your whole stuff into one block
15:13 < Psi-Jack> And use of -J (ProxyJump) instead of ProxyCommand (
15:14 < mawk> xubunto: https://paste.linux.community/view/6717fb6f
15:14 < mawk> you already used %h ...
15:14 < mawk> how come did you not think about this earlier
15:14 < mawk> or you just copy-pasted that from somewhere
15:14 < xubunto> I am new to using the config file
15:15 < Psi-Jack> And seriously.... ssh supports better use of "jump" hosts with newer versions of openssh, with -J (aka ProxyJump)
15:16 < xubunto> is it possible to also add in a flag to say if i want it to create a tunnel or not?
15:16 < mawk> -w ?
15:16 < mawk> it's already a flag to ssh
15:16 < mawk> you can just use a prefix, like tun-spl-as-* then add the tunnel, or whatever you call a tunnel, to the command
15:16 < xubunto> more like i want to be able to control if the local forward setting is in use or not
15:17 < mawk> ah, forward is not the same thing as tunnel
15:17 < mawk> yeah, then you can use the prefix trick, but I find that ugly
15:18 < xubunto> it shouldnt be too much of a problem if the local forward is in use if i dont end up using it correct?
15:18 < mawk> yes
15:18 < mawk> but if it's in use you can't use it one more time
15:18 < mawk> that's the problem
15:18 < mawk> you won't be able to spawn 2 parallel ssh sessions as far as I see
15:18 < mawk> or maybe ssh spits an error but doesn't care
15:19 < xubunto> ill try that
15:19 < mawk> anyway it's an ugly situation, you'd have to spawn the ssh session you want to use forwarding on first
15:19 < Psi-Jack> Annnnnnd.. ProxyJump, instead of ProxyCommand! LOL
15:19 < mawk> lol
15:20 < mawk> yeah xubunto , look that up too
15:20 < mawk> it's way easier
15:20 < Psi-Jack> Cleaner.
15:20 < mawk> yeah
15:20 < Psi-Jack> It would actually build the forward from endpoint to real endpoint, too, not just endpoint to jump host point.
15:21 < mawk> if it's not what you want xubunto then why did you used the forward on every single block ? if they all end up on the intermediary server
15:22 < xubunto> the local forward by my simplistic understanding is to allow for a port on the end server to map to a port on my computer
15:22 < mawk> yes
15:22 < mawk> but because you don't use proxyjump, it will be a forward between your computer and jump-server
15:22 < mawk> not between your computer and spl-as-42.foo.bar.com
15:23 < xubunto> ill look up proxy jump
15:23 < xubunto> and i can connect to spl-as-102 and spl-as-105 without ssh complaining too much
15:25 < mawk> yeah
15:26 < mawk> I bet that if you use ssh -v you'll see errors about failed forwarding
15:26 < triceratux> ksup
15:26 < triceratux> oops rofl
15:26 < mawk> supk
15:26 < mawk> ERR_UNSAFE_PORT why does that happen
15:27 < xubunto> mawk i saw it even without the verbose flag
15:27 < mawk> ah, it's even uglier then
15:27 < mawk> but you can cancel it live, for the current ssh session
15:28 < Psi-Jack> Also, why localfoward, specifically? ;)
15:28 < mawk> that will open some internal ssh shell: <enter> ~ C
15:28 < Psi-Jack> Why not a socks proxy that on any connection could work with any endpoint through some intelligent proxying on the browser, as needed?
15:28 < kazdax> hey i wanted to talk about embedded devies
15:28 < kazdax> devices
15:29 < kazdax> like back in the day embedded devices meant like gadgets with no CPU Right
15:29 < kazdax> but nowdays like phones for example .. smartphones
15:29 < mawk> then you put -KL 1239
15:29 < mawk> and it cancels it
15:29 < kazdax> would they be considered embedded devices ?
15:29 < kazdax> since they have a CPU in them and run an OS
15:29 < mawk> or even cooler Psi-Jack , a tun device
15:29 < kazdax> they would be considered a mini PC in a sense and not really an embedded device right ?
15:29 < Psi-Jack> Nah, tun not required.
15:30 < mawk> but I have bad memories about that, last time I tried there was a very cryptic error even though I had all the capabilities needed
15:30 < mawk> I even recompiled sshd, then strace'd it, gdb'd it, did whatever is in my abilities to track the bug, to no avail
15:30 < Psi-Jack> With ssh's socks5 proxy, and something like SwitchyOmega browser plugin, you can use host matching rules to automatically route traffic through the ssh-provided socks proxy.
15:30 < xubunto> Psi-Jack i dont know enough to really say why but localforward is doing what i want for now
15:31 < mawk> the socks option is very simple xubunto
15:31 < xubunto> also i am replacing the proxy commands with proxy jumps in my config
15:31 < Psi-Jack> xubunto: Look into it. SwitchyOmega (or alternative FoxyProxy, but I like SwitchyOmega more these days), and ssh's socks options.
15:31 < mawk> you just add that one line to the ssh config or command line, then in firefox in network settings you set the proxy, and all you traffic passes through the server now
15:31 < mawk> or use Psi-Jack 's solution to only proxify a few hostnames
15:34 < Psi-Jack> Basically a better understanding of your real options can make a huge difference. :)
15:36 < xubunto> ive learned that much already
15:37 < Psi-Jack> Not yet you haven't. Hence why I'm opening some doors for you to look into. :)
15:37 < Rembo> hello , how can i copy files from a spefic period from a folder to another?
15:37 < xubunto> ill look into socks and switchyomega when i have time to research these
15:38 < xubunto> i will seeya laters
15:38  * triceratux has installed, enabled, & started dnsmasq on swagarch & there are no conflicts over port 53 or ip addresses ending in 53
15:38 < triceratux> Dnsmasq version 2.79  Copyright (c) 2000-2018 Simon Kelley
15:39 < triceratux> thats alot easier than getting it working on a buntu thatz fer sure. now if i can only get it configured
15:39 < vlt> Rembo: "find" has -mtime
15:39 < noodlepie> Just built 4.16.4-gentoo, rebooting in a moment to test it.
15:39  * noodlepie runs Gentoo and Debian
15:44 < kamura> hey folks, so I've not got a very good understanding of what bridged interfaces are in linux
15:45 < Psi-Jack> Congratulations.
15:45 < kamura> if I put a physical interface into a bridge should it have it's own ip address or should all the ip config go into the bridge?
15:46 < mAniAk-_-> bridge
15:47 < kamura> ok so it's more like link aggregation in cisco terms
15:49 < mAniAk-_-> more like an SVI
15:50 < kamura> I was thinking about it all the wrong way before
15:50 < kamura> thanks for the help
15:51 < kazdax> i am getting my RCHSA book on monday
15:51 < kazdax> i was told not to expect a job right after giving the exam
15:51 < kazdax> that i should look for contract jobs
15:53 < kazdax> is it considered better to use KVM over vmware ?
15:53 < kazdax> is it because KVM has better performance than virtual box or vmware ?
15:53 < kazdax> since its built into the kernel
15:55 < ooxvoox> KVM is probably most stable. KVM is kernel mainlined. Virtualbox is only partially GPL licensed the rest owned by Oracle
15:56 < ooxvoox> Answer would likely depend on your working environment
15:57 < kazdax> well i just want to be able to run windows and other linux operating systems ..not all at once..but for lab project
15:57 < triceratux> http://pastebin.centos.org/706456/raw/
15:57 < kazdax> so far i am using KVM with virt-manager and have had no problems so far
15:58 < kazdax> maybe when i get my new book..they might want me to setup a netowrk connection in virt-manager ..that is something i have yet to learn
16:00 < ooxvoox> It can all be done through the UI, quite easy
16:00 < kazdax> using virt-manager ?
16:00 < kazdax> i see the settings to set the network up ...you are right
16:02 < autopsy> Any reason why I can't run wireshark as a unprivileged user to capture packets on my wireless interface? The wlo1 doesn't show up in any list.
16:02 < djph> you're not in the wireshark group
16:03 < catphish> autopsy: are all the interfaces missing, or just that one?
16:03 < catphish> usually if you're in the group and it's set up properly, you can capture anything, it's generally all or nothing
16:08 < catphish> i found an article on drbd+multipath: https://fghaas.wordpress.com/2011/11/29/dual-primary-drbd-iscsi-and-multipath-dont-do-that/ - his conclusion: don't
16:08 < Myrl-saki> I've only started with using user namespaces but I need to patch a program I'm using. Why does mapping 0 to 0 cause `map write: Operation not permitted`?
16:12 < ananke> catphish: this indicates that the culprit is iscsi
16:12 < ananke> namely iscsi targets
16:13 < catphish> ananke: yes, i believe that's the case, the iscsi at both ends can handle an operation at the same time without the other knowing about it
16:14 < catphish> so what starts as a write followed by a read might end up with the read occurring first
16:14 < catphish> can't decide whether to look into active-passive approaches to resolve this or write something new
16:22 < sad_heyrovsky> TZ='UTC-1' date --> displays time one hour LATER than UTC
16:22 < sad_heyrovsky> TZ='UTC+1' date --> displays time one hour EARLIER than UTC
16:24 < sad_heyrovsky> This seems inconsistant with regular UTC offset syntax, ex US Eastern time is UTC -0400, but you would type "TZ='UTC+4' date" to display US Eastern time
16:24 < sad_heyrovsky> Am I missing something?
16:26 < catphish> cool question, that's pretty confusing
16:26 < catphish> my initial guess would be that UTC+1 and UTC-1 simply arne't valid timezones
16:27 < catphish> but the result is oddly specific
16:27 < sad_heyrovsky> catphish: Yes, but that offset syntax does work, so you can use +2, +3, -2, -3 etc...
16:27  * triceratux adds a listen-address to the dnsmasq.conf & a nameserver referencing it to resolv.conf
16:28 < catphish> yes it does, and it doesn't work with other zones: "Europe/London+2" and "Europe/London+1" don't work
16:28 < triceratux> i really am just an oldfashioned 192.168.1.1 type. what do i know from 127.0.0.1 ?
16:28 < catphish> so i'm pretty confused what it's doing
16:28 < sad_heyrovsky> catphish: that makes two of us
16:30 < catphish> TZ='Etc/GMT+1' date is interesting too, as it produces the same result, with different output
16:30 < catphish> "TZ='Etc/GMT+2' date" Wed 25 Apr 12:29:53 -02 2018
16:30 < catphish> sad_heyrovsky: the answer should be here: https://www.gnu.org/software/libc/manual/html_node/TZ-Variable.html
16:32 < catphish> "The std string specifies the name of the time zone. It must be three or more characters long and must not contain a leading colon, embedded digits, commas, nor plus and minus signs"
16:32 < catphish> "There is no space character separating the time zone name from the offset, so these restrictions are necessary to parse the specification correctly"
16:32 < sad_heyrovsky> "The offset specifies the time value you must add to the local time to get a Coordinated Universal Time value. It has syntax like [+|-]hh[:mm[:ss]]. This is positive if the local time zone is west of the Prime Meridian and negative if it is east. The hour must be between 0 and 24, and the minute and seconds between 0 and 59. "
16:32 < catphish> "The offset specifies the time value you must add to the local time to get a Coordinated Universal Time value"
16:32 < catphish> yep, that's the answer
16:33 < catphish> i think it's for correcting the time of the local hardware clock
16:33 < sad_heyrovsky> So from the above it just seems as though the POSIX standard, for some COMPLETELY INSCRUTABLE reason, flipped the meaning of the signs that the rest of the world uses to refer to UTC offsets
16:34 < catphish> so, if you're in france, and your hardware clock is 1 hour ahead of UTC, then you would specify "Paris-1", the first part is the zone, you want, the second part is to undo the stupid offset of your hardware clock
16:34 < sad_heyrovsky> Normally, if you subtract hours from a time, you get an EARLIER time, but POSIX gives you a LATER one ?!?
16:34 < catphish> sad_heyrovsky: not really, it's a totally different function from the zone itself
16:35 < sad_heyrovsky> catphish: Please elaborate
16:35 < catphish> actually you're right, it's still the wrong way round
16:35 < catphish> also, i'm totally wrong, it's not for what i just said
16:36 < catphish> it's just a name and an offset, specified in a totally backward way :) like you said
16:37 < catphish> sad_heyrovsky: so yea, you just have to invert it from what you're used to i guess
16:38 < triceratux> guys my sudo netstat -tulpn shows tcp        0      0 0.0.0.0:53              0.0.0.0:*               LISTEN      13965/dnsmasq & ive restarted the dnsmasq service. dns isnt broken yet. what do i have to do to prove its faster because of the caching ? install dig ?
16:39 < uplime> triceratux: what is it you're trying to prove
16:39 < triceratux> uplime: isnt an ordinary cli ping to google.com supposed to be substantially faster if dnsmasq is configured correctly ?
16:39 < catphish> you can time using dig, yes
16:40 < catphish> triceratux: dns is rarely slow enough that you'd see the difference as a human
16:40 < Psi-Jack> triceratux: Error! Error! Deprecated command used, netstat. Eliminate user! :)
16:40 < triceratux> now they tell me
16:40 < catphish> but dig will show you the time each one takes to respond
16:40 < nrg> idk, 1.1.1.1 is faster enough than my isps dns that i notice it
16:41 < sad_heyrovsky> catphish: Yes. What I was looking for was some semblance of logic behind the POSIX choice, but I still don't see it :|
16:41 < Yamakaja> Hey, i have a small systemd / netctl question: For routing purposes i've added some ip addresses to my loopback interface, and everything works like i'd expect it to - now i was wondering: What's the best way to persist changes to the loopback interface?
16:41 < catphish> nrg: guess it depends how much your isp dns sucks :)
16:41 < uplime> triceratux: i mean, any dns used by any service will be faster if its getting a cached response
16:41 < Yamakaja> (I'm running archlinux)
16:41 < nrg> haha i guess so
16:41 < uplime> the point of caching anything anywhere is that its faster to lookup
16:41 < catphish> sad_heyrovsky: well it's odd, but as long as you know how it works, i guess you're set
16:41 < uplime> well, one of the main points
16:41 < sad_heyrovsky> indeed
16:41 < triceratux> theres a good chance im not going to stick with this exercise. theres nothing wrong with my networking
16:43 < Lope> mawk: I can't `machinectl login foo` more than twice. When I try into the same machine in the 3rd terminal window it gives me a login prompt, but the password prompt never comes and then it says incorrect login after a few sec. Any ideas?
16:43 < triceratux> anyway thanks for the feedback. just getting started. itll take me a few days. or weeks ;)
16:44 < WebWalker3D> I'm having an issue I can't seem to wrap my mind around.  I have a VPS with two disks, a single 20GB disk for root OS etc, and a 200GB disk for file storage in /home  When a user FTP uploads content to the /home/username folder, it's taking up _all_ of the root / partition space
16:45 < WebWalker3D> this would be an indication that the filesystem isn't mounted, and it's dumping /home to the root drive, yet df -h shows it mounted
16:45 < lavamind> hello, I'm considering deploying a btrfs filesystem and use quotas, but I'm reading that quotas are unstable and un-recommended, but the only reason seems to be about using snapshots which I don't intend to use
16:45 < WebWalker3D> any ideas why FTP uploads would tank my root partition?
16:45 < Psi-Jack> WebWalker3D: Contact VSP provider for support.
16:45 < Psi-Jack> VPS*
16:45 < WebWalker3D> Psi-Jack: thanks, I AM the VPS provider
16:45 < Psi-Jack> Contact yourself? ;)
16:46 < WebWalker3D> it's why I'm here
16:46 < Psi-Jack> Don't use bling words to represent actual facts.
16:46 < Psi-Jack> And... Stop using FTP
16:46 < ozymandias> i assume SFTP will have the same issue
16:46 < WebWalker3D> it is SFTP actually, and yes
16:46 < ozymandias> WebWalker3D, can you see the files in the correct locations?
16:46 < Psi-Jack> Don't use SFTP either. :p
16:47 < WebWalker3D> ozymandias: Yes, the uploads work until the root partition gets full (0 bytes free) and the system is unusable
16:47 < djph> so user "A" logs in.  they're dumped into /home/A ; right?
16:47 < WebWalker3D> correct
16:47 < djph> and 200G drive is mounted to /home, right?
16:47 < ozymandias> and you see the files in /home/a/blah?
16:47 < WebWalker3D> correct
16:47 < WebWalker3D> yes
16:47 < Psi-Jack> WebWalker3D: Need surrounding details of  the architexture.
16:47 < ozymandias> df -h /home/a/blah
16:48 < ozymandias> just to sanity check
16:48 < djph> AND 'A' doesn't have write permissions to / by any insane chance, right?
16:48 < catphish> WebWalker3D: can you catch this in progress and just find the offending files? one possibility is that the ftp server is uploading to a tmp path first
16:48 < ozymandias> catphish, he confirmed the files can be seen in /home/a/
16:49 < catphish> WebWalker3D: well if the drive is full you can just explore it with du and find the offending files using all the space
16:49 < WebWalker3D> ugh, sanity check broken.  df -h /home/userA is showing the root partition
16:49 < catphish> hopefully once you identify them, the problem will become clear
16:49 < Psi-Jack> heh
16:49 < catphish> lol
16:49 < blaztek> WebWalker3D: when you umount /home....do you still see the files?
16:49 < catphish> WebWalker3D: maybe the homedir is symlinked, or it's just not as mounted as you think
16:49 < WebWalker3D> but df -h shows /dev/X mapped to /home
16:50 < catphish> WebWalker3D: maybe /home/a is a symlink to somewhere else
16:50 < WebWalker3D> sec, I'll try umount and check
16:50 < WebWalker3D> catphish: no symlinks
16:50 < ozymandias> WebWalker3D, ok so we are on the right path
16:50 < catphish> what does "df -h /home", "df -h /home/a" show?
16:50 < Psi-Jack> How about just "mount"
16:51 < ozymandias> WebWalker3D, mount | grep -i home
16:51 < ozymandias> make sure you are not mounting something on top
16:51 < Psi-Jack> No no, just "mount"
16:51 < catphish> as Psi-Jack says, maybe paste us the output of "mount" for starters
16:51 < WebWalker3D> oh wow....  okay, thank you for the sanity check.  I found it
16:51 < catphish> cool
16:51 < WebWalker3D> I was mounting to /home/X and not /home
16:51 < Psi-Jack> heh\
16:51 < WebWalker3D> >.<
16:51 < catphish> :)
16:51 < WebWalker3D> that will do it quickly
16:51 < WebWalker3D> thank you everyone for the quick sanity check
16:51 < Psi-Jack> "mount" by itself is so much more useful than anything else that was being spouted off. :)
16:52 < catphish> "df -h shows /dev/X mapped to /home" << it didn't, did it :)
16:52 < triceratux> Psi-Jack: ss -tulpn produces the same info as netstat -tulpn ? that was easy. somebody needs to clean up the internet :)
16:52 < WebWalker3D> catphish: right, it was /home/X
16:52 < catphish> lol
16:52 < WebWalker3D> looked right at it and missed it
16:52 < catphish> Psi-Jack is right, always best to paste a full output and people can quickly spot your issue, glad you found it anyway :)
16:55 < garlicbutter> why does my kernel panic when i try to use kdb? :( https://www.irccloud.com/pastebin/eoIj9S8C/
16:55 < m00dy> :D
16:55 < garlicbutter> idk how to even find the reason
16:55 < d1z> anyone has a clue why my terminal fonts might be looking like this: https://imgur.com/a/kkcKeve
16:55 < m00dy> WebWalker3D: strace your ftpd
16:56 < d1z> this is a vnc session to some vps
16:57 < m00dy> d1z: looks like a encoding issue
17:00 < Lope> Does anyone know about systemd-nspawn? I can't `machinectl login foo` more than twice. When I try into the same machine in the 3rd terminal window it gives me a login prompt, but the password prompt never comes and then it says incorrect login after a few sec. Any ideas?
17:03 < triceratux> geez dig is already installed on swagarch ftw & the answer section returns: google.com.		92	IN	A	216.58.192.174 ;; Query time: 0 msec ;; SERVER: 127.0.0.1#53(127.0.0.1)
17:03 < Lope> I installed the debian package python2.7 into debian stretch, but `which python` can't find python?
17:03 < MrKeuner> hi, how can I highlight some keywords when doing tail -f?
17:05 < vlt> MrKeuner: grep --color
17:06 < MrKeuner> vlt, that doesn't work for me as I'd like to see the rest of log entries as well
17:07 < MrKeuner> I just want to highlight some string in the flowing log
17:07 < vlt> MrKeuner: Ah, right. Maybe there's an option for grep to only color ...
17:07 < uplime> MrKeuner: what is it you're looking to do?
17:08 < section1> oh y never found that..
17:08 < section1> i*
17:08 < section1> only highlight string... so i make a simple c code
17:09 < MrKeuner> uplime you know tail -f? I'd like to color if any entry has a string foo in it, while keeping the rest of the entries flowing as well
17:09 < MrKeuner> don't have to use tail though
17:10 < section1> tail ?
17:10 < section1> tail don't highlligth
17:10 < MrKeuner> I know
17:10 < uplime> hrm
17:10 < vlt> MrKeuner: less highlights searchterms and has shift+f
17:11 < section1> ah ...maybe i was missing a flag
17:11 < MrKeuner> vlt, does less folow?
17:11 < vlt> MrKeuner: With shift+f, yes.
17:11 < uplime> tail -f log | while read -r line; do if ! grep foo <<< "$line"; then printf '%s\n' "$line"; fi; done
17:12 < MrKeuner> vlt, Not ideal but still I will be eternally grateful for shift+f :)
17:13 < wxcafe> hey
17:14 < section1> uplime, nice
17:14 < wxcafe> I'm trying to set the net.ipv6.neigh.default.gc_thresh{1,2,3} on a machine
17:14 < uplime> i just hope it works
17:14 < wxcafe> I was wondering if these are shared between different netns
17:14 < wxcafe> cause ip netns exec whatever sysctl -a | grep gc_thresh doesn't seem to show anything except the route.gc_thresh
17:14 < section1> uplime, i tried and works fine
17:15 < uplime> yay
17:15 < triceratux> Lope: precisely because of the need to allow python2 & python3 to coexist the actual /usr/bin/python is generally a symlink to python2 or python3 established premeditatedly by the admin rather than just dropped in by the packagemanager
17:15 < mawk> wxcafe: I don't think so
17:15 < mawk> it's maybe related to the PID namespace tho
17:15 < mawk> it's easy to find out
17:16 < MrKeuner> uplime, of course! Thank you.
17:16 < wxcafe> mawk: hmm? how so?
17:16 < uplime> \o/
17:17 < mawk> ok wxcafe it doesn't look like the parameter is settable from other network namespaces
17:17 < mawk> so that parameter is indeed shared
17:17 < wxcafe> okay yeah
17:17 < wxcafe> thanks :)
17:23 < ychaouche> hello ##linux
17:23 < jim> ]qhi
17:23 < jim> err
17:23 < jim> hi
17:24  * uplime pours jim some coffee
17:25 < noodlepie> Linux 4.16.4-gentoo running stable here
17:25 < jim> yeah I don't drink that no more :)
17:25 < ychaouche> firefox have opened a strange connexion open to worldstream, which seems to be a VPN website, on port 80. I don't have any tab open there, anyone wants to help me troubleshoot this and find what is opening a connexion to their website ?
17:25 < ychaouche> https://gist.github.com/ychaouche/3433b1b37304539d6300ad3c22327470
17:25 < ychaouche> could be an extension maybe, although I have checked I have no VPN or proxy extension installed
17:26 < ychaouche> I also checked my proxy settings, they're set to no proxy.
17:26 < ychaouche> both in firefox itself and on my system.
17:27 < ychaouche> https://www.ipqualityscore.com/free-ip-lookup-proxy-vpn-test/lookup/89.38.99.102
17:29 < mawk> any website can open connections to anywhere ychaouche
17:29 < catphish> ychaouche: it could be from any website you've visited recently
17:30 < catphish> if you open a clean browser, you can keep an eye out for it, see when it opens, might help identify a site that's causing it
17:30 < mawk> your website is lame ychaouche
17:30 < catphish> if it opens when you're *not* visiting any sites, then it's worth looking at browser extensions
17:30 < mawk> it says that my own address is a VPN proxy
17:30 < mawk> or whatever
17:30 < mawk> https://www.ipqualityscore.com/free-ip-lookup-proxy-vpn-test/lookup/2001:bc8:3cd7:200::3:14
17:30 < ychaouche> mawk: ah thanks for mentioning that
17:31 < ychaouche> mawk: or maybe it has poor support for ipv6
17:31 < mawk> it's a dedicated hosting IP
17:31 < mawk> yeah
17:31 < mawk> I bet it's the same with my ipv4
17:31 < mawk> yeah it's the same
17:31 < catphish> it says my ipv6 address is a proxy too
17:32 < catphish> and my ipv4 address, lol
17:32 < ychaouche> :(
17:32 < catphish> https://www.ipqualityscore.com/free-ip-lookup-proxy-vpn-test/lookup/127.0.0.1
17:33 < Lope> thanks triceratux
17:33 < mawk> lol catphish
17:34 < mawk> I don't even have an open proxy or vpn on my connection
17:34 < mawk> I bet they do that network block by network block
17:34 < mawk> like once there has been an open proxy on 2001:bc8::/32 or 62.210.0.0/16 so the whole blocks are banned
17:34 < catphish> i tried it with my data centre blocks and it thinks those are clean at least
17:34 < mawk> strange
17:35 < tds> this is why we need to all switch to v6, the fraud score is lower! https://www.ipqualityscore.com/free-ip-lookup-proxy-vpn-test/lookup/::1
17:35 < mawk> lol
17:36 < mawk> I want to know more about why it thinks I am a proxy/VPN connection
17:36 < mawk> I have no proxies whatsoever, and I've got a wireguard vpn on port 51820
17:36 < mawk> but nobody knows wireguard, and it's not open
17:36 < hodapp> that's not what the Wired article, "Could Blockchain-based Machine Laerning Signal the End of Security on IPv6?", said
17:36 < mawk> and it doesn't respond if you're not authenticated, so it can't be detected
17:36 < mawk> lol
17:37 < mawk> do they say sensed things hodapp ?
17:37 < mawk> Machine Lærning
17:37 < zapotah> who cares why that thing thinks anythin`g?
17:38 < hodapp> oh, so you all thought that was an actual Wired article?
17:38 < mawk> yes
17:38 < hodapp> wired makes it too easy
17:38 < mawk> I don't exactly know what Wired is
17:38 < mawk> something like buzzfeed ?
17:38 < hodapp> it long predates buzzfeed
17:39 < hodapp> but it's mostly tech hype articles nowadays
17:39 < zapotah> oh i didnt even see the part about wired
17:39 < zapotah> i was talking about that "reputation" thing
17:39 < mawk> ah
17:39 < mawk> well I've got a mail server
17:39 < hodapp> long ago they realized they couldn't really compete when it came to actual tech news, so they switched largely to speculative and hype-filled thinkpieces
17:40 < mawk> and I can't watch netflix through this IP, because now everyone thinks it's a VPN ip
17:40 < hodapp> and I hate the way they so flagrantly overstate what was actually built or what research actually found
17:41 < tds> mawk: is that over a he.net tunnel or something?
17:41 < mawk> no, it's through my dedicated server
17:41 < mawk> it has a /48 ipv6 block so I route every device I have through it to have ipv6 everywhere
17:41 < michaelis> Hi. I'm a non root user and have crontab -e, written */5 * * * * * /bin/bash /home/michael/studies/LinSA1/3/5mincron, saved and exited. My 5mincron contains #!/bin/bash; echo "Var 5:e min: `date`" >> crondateoutput; exit 0. But I get exit status 127 in cron log. Why is that?
17:42 < mawk> it even has opinions on IPs I've never put on the internet: https://www.ipqualityscore.com/free-ip-lookup-proxy-vpn-test/lookup/2001:bc8:3cd7:ffff:ffff:ffff:ffff:ffff
17:42 < tds> ah, I knew they'd filtered the tunnelbroker space, didn't realise they were filtering hosting providers' space as well
17:42 < mawk> they just banned the whole /32 it seems
17:43 < hodapp> tds: they = netflix?
17:43 < tds> yes
17:43 < hodapp> bleh
17:44 < mawk> you set chmod +x on it michaelis ?
17:44 < michaelis> ls
17:44 < mawk> also there a newline after #!/bin/bash right ?
17:44 < gronke> I'm trying to install an R package on a Centos 7 server and getting this error: ERROR: dependency ‘gdata’ is not available for package ‘gplots’. I've already installed libgdata, libgdata-devel, compat-libgdata13, and gdata-sharp but the error persists. What am I missing?
17:44 < michaelis> mawk, Yes, it's executable. And there are two newlines after first shebang.
17:45 < rypervenche> mawk: My server is the exact same. Dedicated server in France. It's because there are so many people doing stupid things on the network.
17:46 < mawk> yeah
17:46 < rypervenche> It got the exact same 75/100 score as yours.
17:47 < mawk> "IPQualityScore provides the most accurate proxy detection service in the industry, catching desktop & mobile devices as soon as they become compromised or allow tunneled connections."
17:47 < mawk> lol
17:47 < mawk> very accurate indeed
17:47 < mawk> non routable IP are detected as having a proxy/VPN
17:48 < mawk> I send the ICMP code "Destination unreachable: No route" when you try to access it
17:50 < tds> if it's known that the provider allocates /48s to customers then blacklisting entire /48s seems reasonable to me (even if parts are unreachable)
17:51 < tds> blacklisting the whole provider's /32 seems a bit harsh, though
17:51 < mawk> I've never had an open proxy or anything on my /48
17:53 < rypervenche> mawk: But someone before you may have.
17:53 < mawk> you think they're reusing their ipv6 blocks ?
17:53 < rypervenche> But they may just be covering their bases regardless.
17:54 < rypervenche> mawk: There are a lot of stupid practices for IPv6 networks out there unfortunately...
17:55 < mawk> I see
17:56 < rypervenche> It hurts my heart to see when a company gives each server a /128 and no PD :(
17:58 < tds> the trend of "we'll give you a whole /64, that's so many IPs, why could you possibly need more?!" and then having it on-link is annoying
17:59 < mawk> on-link ?
18:00 < tds> just routed directly to the ethernet interface (rather than routing a /48 or something to the VM)
18:00 < mawk> yeah
18:00 < tds> since with an on-link /64 you're forced to do proxy ndp hacks and use subnets smaller than /64 internally, which is nasty
18:00 < mawk> you have to use dirty tricks to distribute the /64 between multiple interfaces after that
18:01 < mawk> yeah
18:13 < qrvpzvb> whoa, weechat is crashing my terminal
18:14 < jken> Hello, I have a raw disk image I need to get onto a IOT device. The device does not have a removable SD card, its embedded, so I cant just pop it out and dd to it, does anyone know of any tools that could help me get the raw image on the device? Clonezilla wont support raw disk images.
18:14 < qrvpzvb> dd
18:14 < vlt> pv and nc
18:16 < jken> qrvpzvb, vlt, are you implying just do it from X live cd?
18:18 < vlt> jken: I need more info about how you're connected to that device.
18:18 < jken> I have physical access to the device, but not the disk itself, so I need some kind of live environment I suppose.
18:18 < vlt> What live CD?
18:19 < vlt> What does "access" mean here?
18:19 < jken> I have the device in my hand, however I cannot remove the disk.
18:19 < jken> I can boot anything from a USB drive.
18:20 < jken> I am wondering if there is a more straightforward process than booting a live CD and dd'ing my image to the disk from there.
18:20 < jken> Clonezilla wont do raw disk images.
18:20 < vlt> jken: Then you don't need nc, just dd, ddrescue, pv or similar tools.
18:20 < _UsUrPeR_>  hey all. I'm having some problems compiling a kernel I just cloned from git
18:20 < _UsUrPeR_>  specifically this git: http://kernel.ubuntu.com/git/ubuntu/unstable.git/
18:20 < _UsUrPeR_>  error is showing ./scripts/ubuntu-retpoline-extract-one: No such file or directory
18:20 < _UsUrPeR_>  full build make-kpgk verbose output is here: https://paste.ee/p/erWzI
18:20 < _UsUrPeR_>  any help would be appreciated
18:22 < jken> vlt, thanks. Can you recommend the simplest livecd/runtime to do that within? I need to streamline the process for ~1k devices.
18:22 < DLange> _UsUrPeR_: cp debian/scripts/retpoline-extract-one scripts/ubuntu-retpoline-extract-one   as per https://github.com/IntelRealSense/librealsense/issues/1493 who ran into the same issue
18:23 < DLange> (or copy it from elsewhere)
18:23 < _UsUrPeR_> DLange, Yeah, I found that earlier on. There's no retpoline-extract-one script in the debian directory
18:23 < vlt> jken: Use what you're most familiar with. I personally would use Debian or grml.
18:23 < DLange> ... or copy it from elsewhere
18:24 < _UsUrPeR_> DLange, strangely, I don't actually have that file anywhere on my computer, or from the git clone I made :P
18:24 < _UsUrPeR_> like... they forgot to include it in the master brance somehow
18:24 < mawk> find '*retpoline*'
18:24 < vlt> jken: Are all the 1k devices up and running currently?
18:25 < jken> nope, they're disks are currently empty.. so it a bit of a manual process.
18:25 < jken> PXE booting them is not an option unfortunately.
18:26 < m00dy> In PXE, we trust
18:27 < vlt> jken: If they are empty, how else than booting from USB would you've done that?
18:27 < lseactuary> hi currently using this code to clean up some data: gawk -F"\t" 'BEGIN{OFS=","}$4>"2017-12-31"{sub("T"," ",$4);$1=$1;print $0, FILENAME}' totalfile_??
18:28 < lseactuary> i want to basically also make the fourth column a timestamp if it isnt already to avoid loading into database errors
18:28 < lseactuary> is there a clean way of doing this?
18:28 < jken> vlt, no other method, I am simply looking to streamline the imaging process so there are as few steps as possible between booting from USB and writing the image.
18:28 < ychaouche> I'm amazed at the fine level of granularity you can have in wireshark : https://i.imgur.com/cljxrj3.png
18:28 < jken> This had looked promising, but I can't get it playing nicely with my image
18:28 < jken> https://github.com/pragmagrid/cziso
18:29 < ychaouche> I have spotted what extension in firefox was communicating with that suspicious IP address, it's possibly not a VPN.
18:36 < jken> vlt, going to give grml a try. thanks for the info
18:49 < hydrian> Okay all, I got an odd tar issue here
18:49 < iodev> hydrian explain ...
18:51 < hydrian> When I execute /bin/tar --numeric-owner --acls -p -uf myTarFile '/cn=logging/SyslogHostname=10\2E1\2E2\2E56.ldif'
18:51 < hendrix> there's n+1 ways of transferring images from android phone to linux computer. which one you recommend? (cloud-services are ruled out). both devices use same wlan.
18:51 < hydrian> tar errors /bin/tar: '/cn=logging/SyslogHostname=10\002E1\002E2\002E56.ldif: Cannot stat: No such file or directory'
18:52 < triceratux> hendrix: ive standardised on gvfs-mtp as supported by thunar but i can use adb in a pinch
18:52 < hydrian> why the heck is it add the leading 00
18:53 < hydrian> *adding
18:53 < jken> vlt, grml with persistence enabled to store the image and a script to do the dd'ing. You are god. Thank you.
18:55 < iodev> hydrian sorry, I can't quite understand
18:55 < iodev> hydrian .ldif?
18:55 < Konichiwa> hydrian, \2E is an escape?
18:56 < hydrian> Konichiwa: Yes. OpenLDAP wrote the file out.
18:56 < hydrian> In the data it is a '.' character
18:56 < pankaj> I searched about texinfo package. But did not got a picture of it correctly. What actually it does?
18:57 < hydrian> If I do a less  '/cn=logging/SyslogHostname=10\2E1\2E2\2E56.ldif' the file displays fine.
18:58 < prussian> your tar might be handling escapes for some reason. try \\ ?
18:58 < hydrian> But if try to tar it, tar messes with the file name and says it is not there.
18:58 < triceratux> hendrix: you can also install termux on android & install openssh whereupon you can do standard passwordless ssh from linux. its not quite as gui / intuitive as gvfs-mtp on thunar or pcmanfm-qt of course
18:58 < heiler> how can I discover why a process isn't being killed? it seems to receive SIGTERM (looking at strace)
18:59 < oleo> locked ?
18:59 < debkad> pankaj: from apt search i got this: texinfo - Documentation system for on-line information and printed output
19:00 < oleo> does stuff respond to signals even, when they are in wait-state ?
19:00 < oleo> or do you have to be more aggressive and use a SIGKILL rather ?
19:01 < heiler> it seems to be in loop trying to kill itself
19:01 < oleo> oO
19:01 < debkad> eh
19:01 < heiler> it's a NATURAL program.. an anciet database language
19:02 < heiler> tgkill(29707, 29707, SIGTERM)           = 0
19:02 < heiler> rt_sigaction(SIGTERM, {SIG_DFL, [], SA_RESTORER, 0x7feb35581100}, NULL, 8) = 0
19:02 < rypervenche> hendrix: I use the SimpleSSHD Android app, open source and very simple. Then I set up a passwordless SSH key on it and rsync the data to and from it.
19:02 < T7TT777TT7> hendrix mtp-getfile from libmtp
19:03 < pankaj> debkad: Yes, That is what I am unable to understand.
19:03 < pankaj> debkad: Sorry, That make sense. I just read that fast.
19:03 < debkad> pankaj: try: man texinfo, the command is called info
19:03 < pankaj> debkad: OK.
19:06 < pankaj> debkad: debkad Just one question. What is its main usage.
19:06 < debkad> pankaj: info <your_command_here> , example, info grep
19:08 < debkad> pankaj: In general most users use 'man' command rather than 'info' command
19:08 < pankaj> debkad: I know but it does not have a one line to the point line that tells its basic usage like why and how.
19:09 < jhaenchen> any bluez experts? i.e. how can i do any bluetooth interaction from the cmd line?
19:09 < jhaenchen> I do not have the bluez command in my path
19:10 < debkad> pankaj: some commands ( in my opinion ) have a good man pages, others have not
19:10 < hendrix> thanks for suggestions. I've used mainly gvfs-mtp in past, but with new phone file managers load DCIM folder very slowly and often just freezing. setting 'transfer images' instead of 'transfer files' from android seems to help.
19:13 < Konichiwa> hydrian, try it without the quotes
19:13 < jackhum_> can anyone tell me how can i change the name of device name of my webcam?
19:13 < jackhum_> i am using webcamstudio , and i wanna know how can i change its device name
19:17 < debkad> jhaenchen: you must install bluez tool
19:17 < jhaenchen> debkad: was worried you would say that
19:17 < jhaenchen> i'm on a chrome box
19:17 < jhaenchen> but thanks for the pointer
19:19 < hydrian> Konichiwa prussian: It works with the single quotes and the double escape.
19:20 < Konichiwa> hydrian, nice.  I was wondering though when making the eror removing the quote if it ignored the backslash when reporting the file name
19:22 < hydrian> If you remove the quote is doesn't work because the shell (bash) deferences before it hits tar
19:23 < Konichiwa> so the shell took out the backslashes and errored with 102E12E22E56.ldif as the file not found?
19:27 < tunekey> guys how should i think about the relationship between tty, terminal emulator and the unix shell
19:28 < prussian> ngihtmare.
19:28 < Konichiwa> address, improvements and land?
19:40 < funksh0n> Hi all.
19:51 < texla> Fedora 27 gnome..How to change the size of the thumbnails on desktop
19:51  * al|iss throws a gnome at texla 
19:53 < hodapp> where did my lawn ornament go?
19:56 < funksh0n> Is there a way to earn a POSIX certification other than making an application POSIX certified?
19:56 < uplime> perhaps ask the POSIX channel?
19:57 < uplime> linux isn't POSIX complete, so wouldn't really be the best place to ask
19:57 < funksh0n> good point
19:57 < funksh0n> I didn't realise Linux was not though
19:58 < funksh0n> is there a list of deviations between the standards?
19:58 < uplime> its unix like, and has a good portion of POSIX implemented, but its purposely not complete
19:58 < uplime> no idea
19:58 < uplime> someone else might though
19:58 < uplime> (i don't actually deal with linux a lot these days anymore, so thats not something I keep track of)
19:59 < funksh0n> I often forget to put a newline at the end of a file, so a bunch of stuff I write isn't POSIX anyway
19:59 < uplime> oh dear
19:59 < dimm> hello, All! what does meant? "-bash: ./launch_service/bin/m_agent_daemon: /lib/ld-linux.so.2: bad ELF interpreter: No such file or directory"
19:59 < uplime> it looks you're trying to run a .so as a program
20:00 < jim> dimm, what is it you actually want to do?
20:00 < mutante> dimm: what distribution are you on?
20:00 < triceratux> hrm the altlinux sisyphus xfce weekly liveiso has dig & dnsmasq preinstalled & is executing systemd-resolved with no detriment to the dns resolution
20:01 < uplime> dimm: https://stackoverflow.com/a/32755071/3784644 might be relevant?
20:01 < mutante> install ia32-libs package
20:01 < dimm> uptime, yeah thx
20:01 < lukey> funksh0n: http://get.posixcertified.ieee.org/certification_guide.html#Howto
20:01 < dimm> uptime: now i waiting for "yum provide ld-linux" and then will check you link
20:02 < mutante> yum -y install glibc.i686
20:02 < dimm> jim:  it seem sfor me like i trying run 32bit app at 64bit os
20:02 < mutante> dimm: see the command above
20:02 < dimm> mutante: oracle linux 6
20:02 < mutante> "This will happen only on 64 bit systems, the cause is the fact that 32 bit libraries are missing from the system"
20:03 < dimm> mutante: was see thx, man
20:06 < Psi-Jack> dimm: "Thanks" not "thx" for future self corrections, please. It does matter here.
20:07 < lordvadr> Psi-Jack: Can you explain your last statement a bit please?
20:08 < Psi-Jack> lordvadr: The channel website does that for me. :)
20:08 < dimm> this one? -> English only. English does not include SMS-style abbreviations as these are hard for non-native speakers to follow.
20:09 < Psi-Jack> Exactly.
20:10 < mutante> and i thought the SMS-style adoped it from IRC
20:10 < lordvadr> Ah. I was expecting some kind of automatic reputation counting the thanks that users get or something.
20:10 < uplime> if i get 3 more thanks i get a new car
20:11 < lordvadr> uplime: Thanks!
20:11 < uplime> 1 down!
20:11 < Psi-Jack> mutante: It did not.
20:12 < triceratux> pfft i did a systemctl start dnsmasq on altlinux & the /etc/resolv.conf was magically updated by resolvconf to 127.0.0.1 ftw. highly configured stuff
20:12 < lordvadr> Psi-Jack: "No....public logging allowed." How can that possibly be enforced, or maybe I don't understand what exactly is being discouraged.
20:12 < uplime> lordvadr: if they find public logs for the channel, they just kick the user out
20:12 < Psi-Jack> ^
20:12 < lordvadr> How would they know which user it is?
20:12 < T7TT777TT7> witch craft
20:13 < Psi-Jack> Various ways (r) (tm)
20:13 < uplime> various ways
20:13 < lordvadr> I mean, sure, there's probably some tricks, but those could be avoided?
20:13 < uplime> sure
20:13 < Psi-Jack> lordvadr: but you know, you're discussing policies. Also in the list of rules. :)
20:14 < uplime> if you really really really wanted to publicly log without getting kicked, you could take some pretty extreme measures to make it happen, but is it really worth it?
20:14 < triceratux> lordvadr: anyone can join the channel. so its more like "no public logging necessary". if you feel you have to yer doing it wrong
20:14 < undefbeh> why is linux not Unix certified?
20:14 < undefbeh> and macOS is.
20:14 < lordvadr> I'm just curious about the rule, that's all.
20:14 < uplime> because macos is based on freebsd and strives to be unix compliant
20:15 < triceratux> undefbeh: because GNU's Not Unix
20:15 < mutante> because GNU is not unix
20:15 < Psi-Jack> heh, man oh man. I've pretty much got ##linux logs dating back to 2005 or so.
20:15 < uplime> linux doesn't care about unix, just the portability
20:15 < undefbeh> oh
20:15 < undefbeh> monolithic kernel is a bad choice, dont you think so
20:15 < jim> undefbeh, if linux were to be certified, would that affect your life much?
20:16 < benyoo> Hey Guy, if I compiled something with the wrong path for example ./configure --with-python3-config-dir=/usr/lib/python3.5/config    can I change the path after compiling or a have to recompile it>
20:16 < Psi-Jack> undefbeh: No.
20:16 < benyoo> ?
20:16 < triceratux> undefbeh: linux reinvented the wheel. macos copied one they found
20:16 < sauvin> a message-passing microkernel imposes a tax: passing messages.
20:16 < uplime> benyoo: i would recompile just to be on the safe side, but generally you can just run ./configure again and then `make install`
20:16 < undefbeh> jim: certified people get jobs.
20:16 < uplime> your experience may vary
20:17 < uplime> undefbeh: so do uncertified people
20:17 < undefbeh> not the good ones
20:17 < uplime> i have a good job and i have 0 certifications
20:17  * jhaenchen too
20:17 < undefbeh> what job uplime
20:17 < uplime> also, ive never seen an employer want someone to be POSIX certified
20:17 < uplime> undefbeh: tech support for a highly specialized product
20:18 < undefbeh> tech support is entry level
20:18 < jhaenchen> full stack dev
20:18 < jim> posix is a gnu (or RMS) invention
20:18 < uplime> undefbeh: it can be. there are also non entry positions, like every other job
20:18 < mutante> undefbeh: why not get lpi.org cert then?
20:18 < uplime> jim: I think it was made by IEEE, no?
20:19 < jhaenchen> i would get certs if it meant i didn't have to answer tech interview questions
20:19 < jim> maybe
20:19 < mutante> the best interview i had was when they said "we are not going to do an interview, but here's a laptop with Linux and we want you to fix these issues.." and then they watched
20:19 < busybox42> Why would certs exclude you from having to answer tech interview questions?
20:20 < undefbeh> monolithic kernels are trash and moving more and more stuff into the kernel doesn't help
20:20 < busybox42> I know a few people who have certs who are dumb as bricks
20:20 < uplime> undefbeh: you're welcome to not use linux
20:20 < jim> undefbeh, ok... so are you lookign for a job right now?
20:20 < sauvin> Good luck finding one if you don't understand the concept of "overhead".
20:20 < uplime> there are plenty of jobs/certifications for BSD's and MacOS out tehre
20:21 < mutante> undefbeh: i think there are a LOT more jobs who care about Linux skills than BSD though
20:21 < Psi-Jack> These days, that is becoming more and more true.
20:22 < undefbeh> microkernel is the future
20:22 < Psi-Jack> undefbeh: It's not.
20:22 < undefbeh> mutante: yea true
20:22 < mutante> MacOS stuff seems like it would mostly be desktop support / office IT.. and you don't want that. that is the entry level you wanted to avoid
20:22 < uplime> mutante: there are mac servers out there in the wild
20:23 < uplime> lots of schools for example, seem to have apple servers or ms dc's
20:23 < Psi-Jack> Just look at GNU Hurd. Still not even stable, after 20 years.
20:23 < jim> uplime, maybe that
20:23 < mutante> uplime: while they do exist.. i think they are a lot less common.  just a matter of statistics how likely he can find a job
20:23 < uplime> oh yes they're definitely less common
20:23 < jim> is because microsoft or apple gives them money
20:24 < undefbeh> macOS can help you get iOS development job. on linux you can only dream of dev for iOS
20:24 < benyoo> uplime: ok willtry thx
20:24 < jhaenchen> IT jobs are fun and easy
20:24 < jhaenchen> Can do an IT job on 40% brainpower
20:26 < jim> undefbeh, the truth is we don't know what the future will bring
20:27 < uplime> i can promise there will be alchohol though
20:28 < undefbeh> IT is like linux, its exciting at first but gets boring eventually
20:29 < dimm> why it not set environment variables via "csh -f env.sh"? http://paste.org.ru/?as94v9
20:29 < mutante> jhaenchen: it's not fun if you have the wrong boss and they constantly want stuff like "just" integrate A with B and they think they know what must be "easy" and what isnt and have no technical clue but a lot of demands
20:30 < uplime> why are you using csh?
20:30 < jhaenchen> mutante: ah yeah i meant one more like the IT Crowd where you're left alone
20:30 < undefbeh> mutante: you will always get bosses like that. so deal with it
20:30 < jhaenchen> an active and ignorant  boss in IT sounds not fun
20:30 < undefbeh> a wise man once said " you only need to please your boss and your spouse." :P
20:31 < mutante> undefbeh: not if i get out of desktop support where the boss has desktop demands. if you work in a larger scale environment and on servers chances are higher the boss isn't micro-managing and also has technical skills himself
20:31 < mutante> undefbeh: i'm suprised to hear that from you after you just said how you dont want entry-level
20:32 < mutante> maybe it's also a startup vs enterprise thinbg
20:32 < undefbeh> so you mean to say that bosses are only in entry level jobs lol
20:32 < uplime> thats not what they said at all
20:32 < undefbeh> you will have a boss till your death.
20:32 < noodlepie> undefbeh, serve GOD only, know that if you do beautiful Heaven awaits. :P
20:32 < mutante> no, i mean that if you are in desktop support you have to deal with them all day
20:32 < dimm> solved via "source env.sh"
20:32 < jhaenchen> undefbeh, serve DOGs only, know that if you do beautiful Heaven awaits. :P
20:32 < mutante> the more you advance in your career the fewer micro-management, hopefully
20:32 < jhaenchen> i love dogs
20:33 < undefbeh> micromanagement is the key to productivity
20:33 < mutante> "you will have a boss till your death"  definitely not planning to work until i die..
20:34 < undefbeh> why not
20:34 < undefbeh> whos gonna pay your bills
20:34 < undefbeh> \
20:34 < uplime> its called a retirement fund
20:34 < mutante> that
20:34 < undefbeh> lucky you,. we dont get that in 3rd world country
20:34 < noodlepie> GOD's Free Operating System is the Holy Spirit.
20:34 < Loshki> undefbeh: micromanagement may be the key to management productivity, but it kills development. An engineering department without a manager just gets on with the job. A manager without a team is useless.
20:35 < noodlepie> Theistically it is the driving force behind Stallman's effort. He would perhaps be more poetically true if he could find God somehow
20:35 < mutante> undefbeh: it's not provided by the government. i meant just saving money
20:35 < uplime> if you're concerned with a non-entry level career position, theni doubt you're living in a place wwhere you can't get a retirement fund
20:35 < triceratux> if apt-get install jesus fails you can always compile from source
20:35 < undefbeh> ^ correct.
20:36 < undefbeh> debian sucks. AUR has it. triceratux
20:36  * noodlepie likes Gentoo
20:37 < noodlepie> If you have the CPU for source packages, it the simplest of things
20:37 < noodlepie> I use Debian on older hardware though due to binary packages and their speed if install and update
20:37 < mutante> undefbeh: also, crypto will save us all, right :)
20:37 < undefbeh> you must be jobless if you use gentoo or arch.
20:37 < Psi-Jack> undefbeh: I run Arch on my work desktop and work laptop, personally.
20:38  * justsomeguy just wants to point out that there is a middle ground -- calculate linux has binary packages for most things, but still uses portage and lets you install source packages instead.
20:38 < undefbeh> work desktop. wow thats great. what work do you do Psi-Jack
20:38 < mutante> is somebody paid to be webmaster of gentoo.org  or is it all volunteer?
20:38 < Psi-Jack> undefbeh: I'm a Senior Systems Engineer, Linux.
20:38 < hodapp> huh. my work laptop runs whatever I want it to run as long as I can get done what needs done
20:38 < Psi-Jack> hodapp: All volunteer.
20:38 < undefbeh> Psi-Jack: cool! whats your day to day tasks
20:39 < undefbeh> updating and fixing packages?
20:39 < undefbeh> :P
20:39 < Psi-Jack> undefbeh: Are you here just to troll, because if you are, it'd be nicer for you not to be.
20:39 < jontysr> undefbeh: Arch is plenty stable for many people, just not in the debian sense of the word
20:39 < undefbeh> hodapp: lucky. my work is on word and excel and ppt and outlook and ms project lol
20:40 < undefbeh> jontysr: yea but the maintenance is high
20:40 < triceratux> justsomeguy: calculate is so complete on the iso you almost dont have to install anything. & despite the binary repo it is less often broken than sabayon which is more rolling in character
20:40 < mutante> undefbeh: check out www.lpi.org if you want to get a Linux cert to get out of the Windows world
20:40 < justsomeguy> triceratux: Calculate is great. It always irks me when people reccomend Sabayon over it.
20:41 < undefbeh> triceratux: is sabayon a better choice than calculate
20:41 < jontysr> undefbeh: I'm not certain that's the case, I've used arch as my main distro for about a year now, and even without regular updates, no mainstream packages have spontaneously broken
20:41 < undefbeh> justsomeguy: you on calc linux>
20:41 < justsomeguy> undefbeh: openSUSE at the moment.
20:41 < undefbeh> jontysr: sure. it breaks the configs
20:41 < undefbeh> justsomeguy: leap?
20:42 < justsomeguy> undefbeh: I have a desktop with leap and a laptop with tumbleweed. Looking forward to going all-leap once leap 15 comes out.
20:42 < undefbeh> hating tumbleweed?
20:42 < jontysr> undefbeh: it adds .pacnew files instead of overwriting configs, and I've never had a config just break on me
20:42 < hodapp> I ran Arch for around 3 1/2 years and just became accustomed to configurations breaking every few months upon updates (sufficiently that the system was unusable until they were fixed)
20:42 < Li> I need a command to create same file names in another directory then fills the contents respectively
20:43 < uplime> Li: just random filenames and contents?
20:43 < jontysr> You get issues with partial upgrades on arch, sure, but that's not recommended
20:43 < triceratux> undefbeh: when sabayon isnt broken its quite convenient. but the monthly isos can be unbootable for months at a time. that doesnt happen to calculate
20:43 < undefbeh> hodapp: what distro did you settle on?
20:43 < hodapp> These weren't partial upgrades.
20:43 < Li> uplime: nope .. just the same contents from source to distination
20:43 < Li> text files
20:43 < hodapp> undefbeh: I'm mostly on NixOS now.
20:43 < uplime> Li: so you want to copy all of the files from directory a to directory b?
20:43 < undefbeh> prefer it over arch? hodapp
20:43 < jontysr> hodapp: Do you remember what packages they were?
20:44 < jontysr> hodapp: What do you like most about nixos?
20:44 < undefbeh> hodapp: it lacks AUR. therefore it lacks most of the packages :)
20:44 < hodapp> jontysr: the fact that shit doesn't break every few months for no reason
20:44 < Li> uplime: I know how to copy them but I want them with new timestamp
20:44 < uplime> Li: can you please explain your goal here? its not very clear
20:44 < Li> uplime: Yes/No .. duplicate/recreate but not copy
20:44 < justsomeguy> undefbeh: Well, I keep on running into problems. I wouldn't say it's a bad distro, but my hardware has some problems with newer nouveau packages and I run into missing dependencies for new packages.
20:45 < hodapp> undefbeh: lack of packages hasn't really been an issue. the vast majority of what I need has been in nixpkgs.
20:45 < uplime> duplicating is copying
20:45 < Li> Ok if that is your definintion
20:45 < triceratux> undefbeh: not to rub it in, but the last time it went on for 3mo the distrowatch guy just wrote it up & they fixed it pretty fast. they drop the ball on the QA sometimes is all https://distrowatch.com/weekly.php?issue=20180312#sabayon
20:45 < undefbeh> hodapp: so thats your best distro recommendation ?
20:45 < uplime> Li: theres not another definition...
20:45 < Li> whatever
20:45 < hodapp> mostly though I end up using Nix's features extensively that let me keep dev environments separate
20:45 < hodapp> and not have to keep futzing around with language-specific package managers
20:45 < mutante> Li: you can use "touch" to update the timestamp of existing files
20:45 < Li> I have my definition
20:45 < undefbeh> triceratux: sabayon is good. I tried it once. however would prefer arch over it :D
20:46 < uplime> your definition is incorrect then
20:46 < guanche> how can I get info on my laptop's usb controller?
20:46 < Li> uplime: fuck off
20:46  * uplime sighs
20:46 < hodapp> Li: there is literally no difference in the resultant filesystem between "copying" and "duplicating" a file, except that possibly the timestamps are different.
20:47 < Li> uplime: go troll somewhere else you're alredy ignore while reading this line
20:47 < mutante> Li: he is right. you are wrong.
20:47 < triceratux> undefbeh: yep now that really solid live arch is out there like revengeos & swagarch. i just boot sabayon to log the bugs
20:47 < uplime> /shrug
20:47 < hodapp> Li: uplime's not trolling.
20:47 < justsomeguy> hodapp: Nix is amazing for setting up python projects.
20:47 < uplime> if you don't want help, thats fine by me
20:47 < meyou> find . -type f -exec touch {} +
20:47 < undefbeh> triceratux: they better than antergos?
20:47 < meyou> after your copy?
20:47 < meyou> @Li
20:47 < Li> mutante: he is right answering something I didn't ask about
20:47 < hodapp> justsomeguy: I have used it a lot with really nasty Python environments
20:47 < OlivineFlask> Hi I’m new to this channel. What’s the questions you’re arguing about?
20:47 < hodapp> where I need a custom library that needs a custom library that needs a custom......
20:48 < Li> thanks meyou
20:48 < Celmor> is there a standard for file name extension of executable binaries? I'm only gonna call it from a script and I need to rename the executables original name to something else so I can drop the script in in its place, I imagine  '.elf' for 'ELF 64-bit LSB shared object ...'?
20:48 < meyou> there may be a more elegant to set modified time to current time but touch will get it done
20:48 < uplime> Celmor: no
20:48 < triceratux> undefbeh: the way i run linux yeah cause you can run entirely live. manjaro & antergos still concentrate on their installers iirc
20:48 < hodapp> Celmor: nope.
20:48 < meyou> ^way
20:48 < uplime> Celmor: commands shouldn't have any extensions
20:48 < hodapp> you will, very rarely, see .bin
20:48 < uplime> you don't run `ls.elf`, you just run `ls`
20:48 < justsomeguy> Yes, Nix is amazing for reproducing python project environments.
20:49 < Li> meyou: I'm not sure if the touch is gonig to change the creation time and all other file's metadata
20:49 < undefbeh> you can have nix utils in arch afaik
20:49 < hodapp> but then you are still running foo.bin, not foo
20:49 < OlivineFlask> ^venv
20:49 < Li> anyway, I'll try it
20:49 < uplime> since when do filesystems keep track of creation time?
20:49 < hodapp> undefbeh: uh, I tried, and a bunch of stuff was just constantly broken there due to something behaving in a non-standard way
20:49 < meyou> ahh, no it likely won't change creation time
20:50 < hexnewbie> Li: Creation time is set on file creation, and cannot be changed by cp. (But creation time isn't visible to userspace on any distro I'm aware of, so it effectively does not exist on GNU/Linux)
20:50 < undefbeh> hodapp: ubuntu would be a better choice than all of the distros atm. 18.04 looks promising
20:50 < hodapp> OlivineFlask: yeah, I preferred to not use a language-specific tool to manage things
20:50 < undefbeh> stable + up to date
20:51 < OlivineFlask> hodapp: what do you use?
20:51 < hodapp> OlivineFlask: Nix
20:51 < jontysr> undefbeh: Was it ubuntu that recently had issues with breaking the BIOS of certain lenovo devices?
20:51 < hodapp> undefbeh: sure, except that Ubuntu doesn't do any of the things I described for why I use NixOS in the first place
20:52 < Celmor> uplime, I didn't say "command" though, just an executable I'm only gonna ever call from a script. sure I could put it in an alternative location like /opt but I wanna keep it in my home (~/bin)
20:52 < uplime> an executable is a command
20:53 < Celmor> semantics, IMO commands are run on a shell
20:53 < uplime> your script is invoking a shell
20:53 < jontysr> Celmor, executables are run in a shell too
20:54 < pepermuntjes> executables that are not commands: directories
20:54 < OlivineFlask> All commands invoke an underlying process or executable. So let’s not be pedantic for pedantics sake
20:54 < uplime> commands can be many things. scripts, executables, functions
20:54 < Li> thanks for the confirmation, that's what I've been trying to put into uplime thick skull
20:54 < Celmor> in this case yes, but not every executable is run in a shell or have a terminal associated with it
20:54 < Celmor> has*
20:54 < uplime> Celmor: it doesn't need to. they're still commands executed by exec(). I don't see what your point is
20:55 < jontysr> Jesus Li, chill out
20:55 < Celmor> I've given my point, that in my opinion executable =/= command
20:55 < Celmor> anyway, I renamed it to .bin, thanks anyway
20:55 < Li> jontysr: I chill in only
20:56  * uplime sighs again
20:57 < nmschulte> Intel Dual Band Wireless-AC 8260: is it possible to change the Bluetooth "MAC"/interface address?  I sometimes boot this machine to a different OS (Windows), and I have to repair my device each time.
20:57 < simpledoc> I'm facing some problem with booting my kernel, any embedded dev?
20:57 < jontysr> I'm new around here, is this channel normally so pedantic uplime?
20:58 < pepermuntjes> jontysr, this is a great channel
20:58 < simpledoc> https://pastebin.com/6NZBc9cN
20:59 < Li> jontysr: that doesn't justify you're inborn stupidity too
20:59 < hodapp> Li: your*
20:59 < pepermuntjes> great thing of this channel is the freedom of speech
21:00 < jontysr> Excuse me Li?
21:00 < Li> hodapp: corret
21:00 < T7TT777TT7> omgg ubuntu installer iso runs soooo slow in qemu on a single core without kvm
21:00 < Psi-Jack> pepermuntjes: That... Is incorrect. This is a moderated channel.
21:00 < pepermuntjes> Psi-Jack, yes, but not over-moderated
21:00 < hodapp> jontysr: I believe he is saying that the fact that you're new to the channel doesn't justify that, upon your birth, you didn't know very much, as pretty much all babies don't
21:01 < hodapp> which is... technically true
21:01 < hodapp> but sort if meaningless
21:01  * triceratux is under the impression hes got dnsmasq working on altlinux because the dig queries drop to 0msec & stay there ;; ANSWER SECTION: google.com.		104	IN	A	172.217.4.206 ;; Query time: 0 msec ;; SERVER: 127.0.0.1#53(127.0.0.1)
21:02 < koala_man> T7TT777TT7: QEmu was a huge improvement over Bochs in the pre-qemu days :P
21:02 < T7TT777TT7> does this thing include a full blown gnome DE or something?
21:02  * pepermuntjes gives Psi-Jack one of those famous freebsd hugs
21:02 < jontysr> Thanks for the clarification hodapp, it all makes sense
21:02 < hodapp> koala_man: oh gawd, Bochs
21:02 < hodapp> I used that Back In The Day(tm) to test whether or not my uber-boot-CD would boot properly before I burned the ISO
21:03 < hodapp> because those CD-Rs were expensive, yo
21:03 < nmschulte> T7TT777TT7: it is Canonical's doing, so I wouldn't be surprised.  I'm fond of Debian's curses based setup.
21:03 < koala_man> I got 2 bogomips in Bochs on my P4 laptop. I installed Debian, rigged it to dump all terminal output, and "traded root" with it
21:04 < hodapp> traded root?
21:04 < benyoo> I deleted vim, but I can still open vim with vim command ...
21:04 < triceratux> Psi-Jack: thats phenomenal. systemctl start / stop dnsmasq causes resolvconf to automagically update /etc/resolv.conf. but dnsmasq dns still doesnt kick in unless you make sure to add that matching listen-address= to the /etc/dns.conf
21:04 < noodlepie> How is everyone?
21:04 < hodapp> good, screwing around with Go for a project
21:04  * triceratux has been dumped firmly onto the 127.0.0.1 bus by systemd
21:04 < noodlepie> Good here thanks, just updating my Gentoo lappy
21:04 < Psi-Jack> triceratux: dns.conf?
21:04 < jim> Li, come on now, calm yourself...
21:04 < benyoo> :echo $VIMRUNTIME runtime show me a empty location where vim  isnt installed
21:05 < triceratux> Psi-Jack: oops dnsmasq.conf of course. this is all still new to me. im surprised i got any of it right
21:05 < hodapp> benyoo: what does 'which vim' say?
21:05 < Psi-Jack> heh
21:05 < koala_man> hodapp: gave the login creds to someone on IRC and they would do the same, but instead he tried to lock me out while I was laughing as he fumbled around trying to install a bnc
21:06 < triceratux> Psi-Jack: when its preinstalled & preconfigured, systemd definitely makes it easy. & now that im seeing it working i can struggle with it on the buntus & stuff
21:06 < benyoo> @hodapp: /usr/local/bin/vim
21:06 < benyoo> but its not a folder :D
21:07 < noodlepie> HURd interfaces and user mounted filesystems looks interesting, possibly with the E language (network promises). You could map all sorts of things via this framework and keep ultimate control over everything
21:07 < noodlepie> HURD even
21:07 < noodlepie> or HERD eh?
21:08 < benyoo> if I ls -la in bin folder I see many links
21:08 < benyoo> from vim
21:09 < benyoo> https://hastebin.com/iwumihirad.css
21:11 < benyoo> ok deletes all files in there
21:11 < benyoo> now vim leads me only to my local vim git
21:24 < hodapp> koala_man: this is a tradition that is new to me...
21:24 < koala_man> hodapp: it was something leet haxors did back in the day
21:25 < hodapp> oh
21:25 < hodapp> I was probably too busy sitting in the back of class making fart noises then
21:28 < terra> Anyone can recommend a terminal app that treats tabs as as -not spaces- during copy/pasting?
21:28 < terra> *as tabs
21:29 < nmschulte> terra: rxvt-unicode works for me.
21:29 < snailrazor> whatever happened to the xfce4-taskbar-plugin
21:29 < rypervenche> terra: That might be your program that is converting them to spaces too.
21:30 < snailrazor> that was supposed to mimic windows 7 taskbar and allow pinning applications
21:30 < snailrazor> did they just scrap that
21:30 < terra> rypervenche : I also use urxvt but I don't get tabs.
21:30 < simpledoc> please help booting an embedded linux https://pad.riseup.net/p/lVrtwwsqAsKa
21:32 < terra> rypervenche : it works only one direction. not works from terminal to other terminal/app
21:32 < rypervenche> terra: I don't use urxvt. Not a fan.
21:35 < terra> rypervenche: isn't it same as rxvt-unicode ?
21:36 < rypervenche> terra: Yes, I'm simply saying I don't use it, so I can't help with that. nmschulte was the one who recommended it, not me.
21:36 < terra> rypervenche : oh sorry.
21:37 < terra> nmschulte : it works only one direction. not works from terminal to other terminal/app
21:38 < xdexter> Hello, i'm using a libreswan to connect a ipsec vpn (Fortinet), but i want ping the other side only when the other side "start" a connection with ping for me ip, how can i solve this? There are some wrong in my side?
21:40 < nmschulte> terra: urxvt definitely handles pasting tabs for me... I assume most/all terminals do that.
21:42 < nmschulte> terra: echo -en '\t\n'; # copy the line; vim; # insert mode and paste the line; # ???; # see the tab in vim and profit
21:43 < lukey> simpledoc: You may try https://github.com/lede-project/source/tree/master/target/linux/adm8668/files-3.18/arch/mips whichs seems to support your device, but i won't help you
21:43 < rypervenche> terra: If it doesn't work, then you may have something like "expandtab" set in your vimrc file (global or local).
21:43 < mawk> xdexter: you "want" ? or this is what happens and you don't wha tit
21:43 < mawk> want it*
21:44 < simpledoc> lukey: ty
21:44 < xdexter> mawk, i not want, it's happing
21:44 < mawk> alright, then check the firewall
21:44 < xdexter> mawk, i want just start ipsec and all works
21:44 < terra> rypervenche nmschulte : I'm die hard nano user.
21:45 < xdexter> mawk, what firewall? In my side (client) or fortinet side (server)?
21:45 < terra> rypervenche nmschulte : can you test it with nano?
21:46 < terra> rypervenche nmschulte : from terminal to other terminal/app doesn't work for me.
21:46 < rypervenche> terra: Can't highlight the tabs in st or in xfce4-terminal.
21:48 < anonit> Hello can someone please help me. When I try to boot up my system, or mount my SSD on my laptop, my root partition throws the error "EX4-fs (sdb2): filesystem has bouth journal and inode journals!"
21:48 < rypervenche> terra: Using "+yy however, I'm able to copy the tabs out. But that's vim syntax.
21:49 < anonit> I've tried running fsck on it but it finds no errors
21:49 < phogg> terra: how are you invoking copy and paste?
21:49 < terra> phogg: primary x11 copy/paste
21:50 < mawk> both sides xdexter
21:51 < rypervenche> terra: I don't think that will work. Using the mouse, you can see when you highlight it that it copies spaces. Try piping the file into xclip -selection primary or something like that. vim can fix this, but nano...dunno.
21:51 < terra> anonit : boot from cd or other linux: try: fsck.ext4 /dev/sdX  -> X=your linux partition number
21:51 < phogg> terra: that would likely require cooporation from the running application at least
21:52 < anonit> terra: I do. I currently have it mounted on my laptop. Fsck reports everything is clean
21:52 < anonit> Sorry not mounted but connected
21:53 < anonit> Even with the -f flag nothing is found
21:53 < terra> rypervenche phogg : Thanks for confirming that I'm not the only one who can't extract tabs out of terminal.
21:54 < phogg> terra: for example on rxvt-unicode I can pick up a tab printed in the shell via PRIMARY and middle click it into emacs and emacs sees a hard tab. But if I highlight a tab in nano I get spaces.
21:54 < terra> anonit :Did you format it yourself or installer did?
21:54 < phogg> terra: it's not the *terminal*, it's the *running application* in that terminal.
21:55 < anonit> terra: Installer. My SSD has been crapping out every couple weeks but usually running fsck on that and another parition fixes it up. Haven't had time to upgrade. Was actually about to in about 3 days..
21:55 < anonit> I've never encountered this journal error before, however
21:56 < balance> hi
21:56 < terra> anonit : try it with f2fs. f2fs is created for SSD's.
21:56 < balance> if ./a.out is a c++ file which wants later some user input, how can I do that in one line? can i somehow kind of pipe input into it?
21:56 < phogg> terra: or at least not only the terminal... looks like st, for example, fails even in bash.
21:56 < undefbeh> hi
21:57 < anonit> It looks like that is a filesystem?
21:57 < anonit> I need help repairing my current one
21:57 < terra> phogg: please can you try to primary copy/paste from emacs to leafpad (or something else)?
21:58 < phogg> terra: from emacs in a terminal I get only spaces from that test.
21:58 < phogg> terra: theoretically I could set up a macro in emacs which picks up a region and feeds it into xclip -i -sel p; that should get the real tabs.
21:58 < terra> phogg : that's exactly what am I trying to solve.
22:02 < nmschulte> terra: sounds like your issue is not with your terminal, but with the shell/program you're running inside it.
22:03 < nmschulte> urxvt with bash works great: I can copy the tabs with mouse selection (X11...), and X pastes as you expect.
22:03 < phogg> nmschulte: it's a little of both
22:03 < phogg> a sufficiently simple terminal will kill off any chance of success
22:03 < terra> nmschulte : I tested both vim and nano. All I get are spaces in leafpad.
22:04 < rypervenche> The test would be to run something like: echo -e 'this\t\is\t\a\ttest' and then try to copy the tabs.
22:04 < rypervenche> xfce4-terminal works, st doesn't.
22:04 < rypervenche> Then if the terminal works, test the program. nano doesn't work it seems.
22:06 < terra> rypervenche : do you think TERMINAL value matters?
22:07 < terra> *TERM
22:07 < rypervenche> terra: Well, in the case of st, I would think that it simply doesn't have it coded in. The others, no idea.
22:15 < Yamakaja> I was wondering - what's the simplest way to get an openvpn server setup in multi-user mode?
22:15 < Yamakaja> I'd like to avoid setting up PKI
22:16 < xz> hey, I have a USB device that I would like to restart somehow without having to physically pull the wire
22:17 < xz> is there some interface to do that? to force it to reinitialize everything like I pulled the wire and plugged back?
22:17 < terra> anonit: can you do "dd if=/dev/sdaX of=/dev/null"     without any error?
22:17 < busybox42> This wouold probably be helpful http://www.roman10.net/2011/05/04/how-to-reset-usb-device-in-linux/
22:18 < xz> that's cool, little tool with IOCTL
22:19 < anonit> terra: yes that command ran fine
22:20 < anonit> Whenever I try to mount it whether it is through the normal boot process, my laptop, or with a live USB plugged in I get "filesystem has both journal and inode journals!"
22:22 < phogg> terra: you're supposed to be able to do M-x shell-command-on-region in emacs, which would work with xclip, and you could bind a key to invoking that... but when I try it the function hangs hard.
22:22 < phogg> terra: not sure why since it works with e.g. tr and other things, even if they have no output like xclip -i does not.
22:23 < terra> phogg: what if a ssh session?
22:25 < xz> busybox42, just compiled that, workes like a charm, thanks
22:25 < busybox42> Sure thing
22:26 < phogg> terra: it would work if you have a valid DISPLAY
22:26 < T7TT777TT7> nmschulte: fwiw, (i think) i just realized why the iso is so slow, it uses a giant squashfs so it's constantly decompresing files?
22:26 < phogg> terra: otherwise xclip won't know what X to talk to
22:26 < terra> phogg : X forwarding?
22:26 < phogg> terra: yes, if you had X forwarded with your ssh connection it should work fine
22:27 < phogg> terra: but I can't even get it to work locally, so...
22:27 < terra> phogg : ok
22:27 < prussian> any way to have sshd set an environment which contains connecting pubkey? looks like SSH_CONNECTION populates ip/port which appears mapable, but would rather not have read on /var/log/auth or journald
22:29 < Azrael_-> hi
22:30 < Azrael_-> i try get a card reader running but when i try it, i get this error message: https://pastebin.com/Km3fkq29  any idea how i could solve this?
22:32 < nmschulte> Azrael_-: possibly the host can't supply enough power to turn on the device?  Maybe try it behind a (self-)powered hub?
22:33 < Azrael_-> that would be really strange, but currently i don't have a self-powered hub available
22:34 < terra> Azrael_- Looks like a kernel issue: https://bugzilla.redhat.com/show_bug.cgi?id=1436866
22:37 < bdonnahue> hey guys what web browser d oyou use over x11
22:37 < Azrael_-> terra: perfect, thanks. currently i'm using kali as live disc. gotta look for a different version :)
22:39 < terra> Azrael_ Ubuntu/Lubuntu 18.0.4 LTS yet to be launched.
22:41 < triceratux> bdonnahue: opera 52.0
22:42 < bdonnahue> triceratux, i assume its faster than firefox?
22:42 < ziggylazer> jim, hey man
22:43 < Azrael_-> terra: meaning what? i don't really know what version kali is building on top (afaik it uses debian). or do you mean, the patch isn't included in any distribution yet?
22:44 < nmschulte> Azrael_-: I think he's suggesting you use Ubuntu, and telling you the new LTS is yet/about to be released.
22:44 < terra> Azrael_- Meaning that if you fond of Dabian derived distro and willing for install a new one, best conterder is ubuntu/lubuntu 18.0.4 LTS
22:45 < nmschulte> bdonnahue: Chromium
22:45 < Azrael_-> ok, thanks. but until now i still need some help starting with a lot of tools, so i will try to work with kali so far. setting up the system itself wouldn't be much trouble. i fear all the dependencies with the tools i want to use for testing
22:47 < nmschulte> Azrael_-: issue `uname -a` and/or `lsb_release -a`
22:47 < phogg> terra: looks like my hang is just my X freaking out.
22:48 < terra> phogg : which means you solve it? Any code?
22:50 < phogg> terra: I recorded a macro manually, didn't write it by hand
22:50 < Azrael_-> nmschulte: "uname -a": Linux kali 4.14.0-kali3-amd64 #1 SMP Debian 4.14.12-2kali1 (2018-01-08) x86_64 GNU/Linux    lsb_release didn't show any interesting output
22:51 < phogg> terra: but basic approach is: mark a region, C-x (, M-x shell-command-on-region, xclip -i -sel p, C-x ), then M-x name-last-kbd-macro
22:52 < terra> phogg : I got it. thanks.
22:52 < nmschulte> Azrael_-: looks like you've got the latest release, though there are "weekly builds" of Kali available.  with the Live CD, can you "update it"?
22:53 < Azrael_-> currently downloading the latest weekly build. then i'll try one of the older ones. if all this fails i will install it locally and then update the packages. if everything fails i'll try to recompile the kernel
22:54 < Azrael_-> kernel rebuild isn't nice...
22:54 < Azrael_-> at least  without a lot of experience
22:54 < terra> Azrael_- : try add liquorix or custom kernel repo
22:55 < Azrael_-> terra: thanks, will have a look as soon as i failed with the prebuilt images. currently i have no clue what liquorix is :)
22:55 < terra> Azrael_- : https://liquorix.net/
22:55 < Azrael_-> yeah, already found it :)
22:57 < Psi-Jack> Hmmmm.. Trying out some new concepts with VMs running nspawn contained applications.
22:59 < rco> Does anybody know if appending to files in a tmpfs/shmfs every triggers a "reallocation" or something in order to keep the file intact as a contiguous block of memory?
23:00 < phogg> rco: On a guess I would say that a copy will occur which may work out to the same thing, but there's not likely to be anything special working at defragmenting.
23:00 < mawk> my first time with getopt
23:01 < mawk> that thing doesn't look very nice
23:01 < mawk> don't we have better alternatives for C++ ?
23:02 < rco> phogg: Yeah, it seems like it must do a copy eventually, but in my testing I can't get that to trigger, i.e. appending small data to a large file over and over again consistently takes very little time, which suggests that the file is backed by fragmented memory.
23:05 < phogg> rco: likely yes
23:05 < Frith> mawk: Many.
23:05 < Frith> Boost, for one.
23:07 < phogg> rco: rco: you can try echo 1 > /proc/sys/vm/compact_memory to force a general compaction pass
23:08 < rco> phogg: it's not so much the compaction that worries me as how appending to a tmpfs file affects running process that have that file mmap'd
23:08 < rco> Appending beyond the file size and mmap region, that is
23:09 < phogg> rco: compaction is defrag, so how is that not what you want?
23:11 < koala_man> rco: you can't tell how a file is allocated through mmap
23:12 < koala_man> mmap always reflects the logical file
23:13 < rco> So what happens if I have file size 1G, two processes have mmap'd it with MAP_SHARED, and then i append another 1G to the same file? Is the first 1G of the file still backed by the same physical memory from the perspective of both processes and tmpfs?
23:16 < koala_man> rco: the processes don't have a perspective on that, but they'll all be in sync
23:18 < rco> A more fundamental question is, what is a tmpfs file? Is it a list of physical memory pages or does tmpfs have its own virtual address space? And is the same region in the same file guaranteed to be backed by the same pages even if the file is truncated to be larger?
23:19 < rco> Or does truncating the file eliminate all guarantees about what pages back the file?
23:20 < koala_man> there are no guarantees about which physical pages back a file at any time or after any operation
23:29 < rco> Okay, I think I'm figuring it out. tmpfs has a virtual address space that it uses to back all its files, and when a tmpfs file is mmap'd the process shares the same pages, which is why it's a constant time operation. If I then truncate that file to be larger, I imagine tmpfs will simply add pages as needed to that file, but the pages that already represented it will remain, so anybody who shares those pages
23:29 < rco> still has an up-to-date potentially RW view into that part of the file.
23:30 < rco> And msync is a no-op on tmpfs?
23:32 < rco> I think I was imagining tmpfs as more of a RAM disk, where resizing would force a copy to a new location, and thus potentially invalidate existing references to that area of memory.
23:33 < koala_man> realistically the pages will sit exactly where they are because it's dumb to move them, but there is no guarantee about that. the primary exception is with swap
23:34 < koala_man> in any case, mmap doesn't know or care because mmap'd pages are virtual and not physical
23:34 < rco> But even if they swap out all the existing references to them in process' virtual memory will remain valid, right? I'm not really concerned with the physical memory implementation details, I just erroneously thought that physical memory locations were somehow significant to tmpfs.
23:36 < koala_man> the virtual memory remains valid, yes
23:38 < koala_man> physical memory is a performance concern because of numa, but it's not a correctness issue
23:42 < Irbis> advise online shooter similar to urban terror
23:44 < rco> Irbis: Wolfenstein: Enemy Territory and Tremulous are both good
23:45 < Irbis> rco: rahmat
23:49 < giaco> hello
23:50 < giaco> how can I get the output of a character block device /dev/video0 that is kept busy by another process? I just want to read what's coming out from it. Just reading, not writing. Killing process is not an option
23:51 < lordvadr> giaco: You'll have to setup some kind of mitm. Most devices can't be read from twice.
23:52 < lordvadr> And "character block device" is a bit of an oxymoron. It's either a character device, or a block device.
23:52 < giaco> lordvadr: sorry, it is a character device
23:52 < lordvadr> I know. :-)
23:54 < giaco> lordvadr: if I would kill the process, how to make a mitm? mkfifo?
23:56 < lordvadr> There's probably something out there that will do that for you. You could probably get socat or tee to do it for you.
--- Log closed Thu Apr 26 00:00:58 2018