--- Log opened Wed Jun 20 00:00:00 2018 --- Day changed Wed Jun 20 2018 00:00 < simbalion> bls: I just realized I had to run a command to swap the default Fn status, maybe that command didn't run on my last reboot 00:00 < simbalion> that was it, sorry to bother! 00:00 < birkoff> kexec try googling some, or ask in github of WSL 00:01 < bls> there's already a ##windows-wsl channel on here 00:01 < V7> jada: #hardware 00:01 < bls> although I imagine it's like ##windows or ##macosx 00:01 < simbalion> there.. added that Fn swap to my openbox autostart, should prevent confusion in the future 00:01 < jada> V7 people on hardware don't like chromebooks 00:01 < kexec> birkoff: thanks will try 00:01 < jada> V7 also, I'm not sure if all chromebooks allow for Linux installation 00:02 < simbalion> jade: try E-bay, you can get amazing prices just don't leap at the first decent looking machine, keep your eyes open for good bargains and use a sniping tool 00:02 < jada> V7 I'm hearing there are some problems. depending on particular model 00:02 < V7> jada: Each hardware supports linux so choose by your budget 00:02 < simbalion> jade: You should probably not use Skype btw if you're looking to embrace the linux lifestyle. Skype is cancer. 00:03 < jada> simbalion well, google hangouts to be precise 00:03 < jada> simbalion I just said 'skype' meaning I need a webcam in it 00:03 < simbalion> jade: Still bad, but not as bad I suppose... Google used to be a respectable organization. 00:03 < simbalion> jada: understood, and sorry for misspelling your name 00:05 < V7> Could anyone suggest a linux with i586 and below support ? 00:06 < sharp15> anyone know where screen windows store recent command history (.bash_history)? 00:10 < kerframil> V7: slackware, bodhi (legacy edition), gentoo (doable if you have a more powerful system on which to perform the actual compilation) 00:10 < twainwek> sharp15: probably in memory 00:11 < jim> sharp15, I don't personally know myself, and maybe they might know on ##windows? 00:11 < twainwek> jim: i think he means the screen sessions 00:12 < twainwek> sharp15: https://stackoverflow.com/questions/14914570/bash-how-do-i-get-the-command-history-in-a-screen-session 00:12 < sharp15> twainwek: you are correct. and thank you. 00:13 < bls> V7: have you tried porteus or debian? 00:13 < jim> wouldn't that command history be in ~/.bash_history? I don't think that file will get updated until the shell exits normally 00:14 < V7> bls: I din't fidn these for i586 and below 00:14 < jim> (screen might know it's running a program called bash, but mighjt not know it's a shell, or where its files are) 00:14 < rany> V7: Puppy Linux would be great for that 00:14 < V7> kerframil: bodhi gives kernel panic 00:14 < rany> I'd recommend that 00:16 < V7> Which package manager puppy uses ? 00:17 < kerframil> yeah, puppy should work as it advertises the absence of PAE 00:17 < bls> V7: https://cdimage.debian.org/debian-cd/current/i386/iso-cd/ http://ftp.vim.org/ftp/os/Linux/distr/porteus/i586/Porteus-v4.0/ 00:17 < V7> kerframil: Not only pae, but cx8 and cmov 00:18 < V7> bls: Doesn't this i386 mean i686 ? 00:18 < rany> V7: dpkg/apt 00:19 < kerframil> indeed, debian will not work 00:21 < rany> V7: You can install multiple spins of Puppy Linux... ( http://puppylinux.org/main/Download%20Latest%20Release.htm ) 00:21 < bls> hmm, looks like i386 might mean i686 for debian as well 00:21 < rany> So packages manager depends on your spin 00:22 < kerframil> debian will absolutely not work, unless you use <= jessie. 00:23 < V7> Thank you very much ayecee bls kerframil rany, really ... I'll try some of them now 00:23 < bls> porteus at least calls out i486 and i586 support 00:24 < V7> So, btw there's no way to check which instructions ( which really CPU ) needs specific OS ? 00:24 < kerframil> V7: as you imply the need to support anything going back as far as the i386, keep in mind that the Linux kernel itself doesn't support i386 as of version 3.8. 00:24 < rany> V7: What do you mean? 00:24 < V7> kerframil: Roger that, so I need below 3.8 00:24 < kerframil> V7: for i386 only, yes 00:25 < V7> kerframil: What's for 486, 586, 686 ? 00:25 < kerframil> V7: also, according to a prominent debian developer, modern versions of gcc are having a tough time producing i386-safe code, even when the target triplet is set correctly to i386-pc-linux-gnu 00:25 < V7> I mean, is there any place to check which CPU needs specific OS ? 00:25 < kerframil> V7: even the current kernel still supports 486+ just fine 00:26 < V7> kerframil: What dou mean by pre previous message ? 00:26 < V7> do you * 00:26 < kerframil> V7: exactly what it says. which part was unclear? 00:27 < V7> target triplet 00:27 < rany> V7: It's basically the specific operating system 00:27 < kerframil> V7: oh. that's used to inform gcc of the targetted host (CHOST, --host et al) 00:28 < rany> You'd specify what computer the software should run 00:28 < rany> On SUSE Linux, that would be x86_64-suse-linux 00:28 < kerframil> V7: for example, slackware would theoretically work on an >=i486 because they use a target triple of i486-pc-linux-gnu for building (in the 32-bit edition) 00:28 < kerframil> triplet* 00:29 < V7> Oh, roger that, so gcc can compile code for a specific CPU 00:29 < kerframil> V7: and because the kernel itself still supports i486, though it would still need to forgo PAE (as you know) 00:29 < balletjebal> i'm experimenting with linux on my phone whilst trying the ifconfig wont work ,not even from sbin. anyone some suggestions? 00:29 < V7> Even if it compiles on newer 00:29 < kerframil> V7: in short, both the way in which the kernel and the userspace is built matters 00:29 < Psi-Jack> Well, ifconfig is obsolete and broken. 00:30 < balletjebal> harr yeah it says command not found 00:30 < rany> balletjebal: What's the output? 00:30 < balletjebal> ^^ 00:30 < V7> balletjebal: Try ip addr 00:30 < rany> balletjebal: Do you have busybox? It includes ifconfig 00:30 < balletjebal> V7, yay :P 00:31 < balletjebal> rany, its running on a deploy 00:31 < balletjebal> ip addr works .. thx 00:31 < Psi-Jack> "thanks" not "thx". Standard English is preferred here. 00:32 < Psi-Jack> You could shorten that to just "ip a" too. 00:32 < balletjebal> ok sry about that, im dutch and rude :P 00:32 < Psi-Jack> ... 00:32 < Psi-Jack> Again, STANDARD English preferred. "sry" is incorrect. 00:33 < balletjebal> Psi-Jack, LOL or is it laughing out loud ? 00:33 < Psi-Jack> LOL is fine, it represents multiple words. :p 00:33 < V7> So, porteus gives a deal: http://ftp.vim.org/ftp/os/Linux/distr/porteus/ that it supports 486 00:33 < rany> Acronyms are okay 00:33 < twainwek> what about ok? is it ok? 00:34 < Psi-Jack> "ok" is standard English. 00:34 < balletjebal> and lmao ? 00:34 * Psi-Jack rolls his eyes. 00:34 < rany> V7: šŸŽŠ 00:34 < bls> ...and the language rules server as a filter once again 00:34 < V7> Interesting, what's the main difference between them except CPU instructions. Wil 586 work faster than 486 except a bit more features 00:34 < balletjebal> hehe okay thank you very much for the information thus far 00:37 < kerframil> V7: anything from the P54C onwards (Pentium at a minimum of 75Hz) will pretty much leave any i486 in the dust, performance-wise 00:37 < kerframil> V7: which is not saying much in view of today's needs 00:38 < rany> V7: I think i486 is not Pentium while i586 is 00:38 < bls> i586 is for variants of the Pentium-I 00:38 < Psi-Jack> rany: Correct. 00:39 < Psi-Jack> It's all about the Pentium, baby. Wierd Al said. 00:39 < kerframil> V7: if you want to know of the improvements made, they are well documented and easy to research 00:39 < rany> šŸ˜† 00:40 < kerframil> V7: as far as linux is concerned, userspace requiring cmov and the kernel requiring pae are the usual sticking points 00:42 < V7> Thank you very much, really 00:43 < dr4ken> guys, do you know if the kernel version that comes with android 2.3 does support external SD encryption just like home folder encryption in desktop distros? 00:44 < bls> dr4ken: you'd need to ask an android channel that 00:44 < kerframil> V7: several of the BSDs still work, too 00:45 < dr4ken> bls, i already did at #android, but i'd expect people there to be less "tech-ish" so idk if they would understand what im requiring + the #android channels seems pretty dead 00:46 < bls> well android heavily modifies its kernel, so there's not really much support or knowledge about it in here. the people that would know also hang out in the android channels 00:46 < rany> bls: Google only modifies the kernel for power management, etc. 00:46 < rany> Their modifications aren't very complex... 00:47 < dr4ken> gotcha, just in any case, do you know any app that does provide the same functionality? 00:47 < V7> Interestiong 00:47 < V7> s/tio/ti 00:47 < dr4ken> i do have to keep using a 2.3 androird based smartphone and while i like the unbloatness android were back then, it is very unsafe and it doesnt (at least visually) have the external sd encryption option in this version 00:47 < V7> slackware with huge.s kernel gives kernel panic 00:49 < micrex22> V7: are you running linux on a 486? 00:49 < V7> Yes 00:49 < rany> V7: Weren't you running it on 586? 00:49 < V7> https://i.imgur.com/TzMErzj.jpg 00:49 < bls> heh, things have gotten better...the android kernel only has 56k lines different now, it used to be closer to 70k 00:50 < micrex22> V7: oh neat, I wonder how well it would run on a blue lightning... (I say that because the one I have is technically a souped up 386SX) 00:50 < V7> When I've tried to run it with hugesmp.s it told that it needs cx8, pae and cmov 00:50 < micrex22> and 386 support is dropped, but the bluelightning tacked extra stuff on 00:50 < spreeuw> if you buy an intel 8086K (5GHz) you can exchange it for a threadripper 00:50 < kerframil> V7: that's correct. huge.s should work, in theory. one of the issues here is that i486 doesn't get an awful lot of testing these days ... 00:50 < rany> spreeuw: Isn't that the special edition one? 00:50 < spreeuw> but that intel is a better game cpu though (pre hardware hole patches) 00:51 < spreeuw> it is 00:51 < kerframil> V7: honestly, as much as I like Linux, you might be better off with a BSD 00:51 < V7> kerframil: https://i.imgur.com/TzMErzj.jpg as you can see. This is huge.s 00:51 < rany> spreeuw: Well, I wouldn't buy it because I doubt it will keep getting upgrades or support.. 00:51 < rany> s/upgrades/updates/ 00:51 < bls> I've had a similar experience where netbsd tends to work better for obscure hardware 00:52 < spreeuw> its a generic cpu 00:52 < morf> no way... when i was using bsd it was like frack this doesn't have a kernel module ... ok here is something ... ok just couple of those patches ... and it doesn't work anyway 00:54 < micrex22> yeah... Linux can fully work on a 486 with no issues 00:54 < micrex22> BSD runs less software 00:54 < micrex22> so, it'd make sense just to use linux imo 00:54 < bls> unless linux doesn't work because no distro has compiled it in a functional way 00:55 < micrex22> but it can work, so it's moot 00:55 < spreeuw> didnt debian drop old cpus too? 00:55 < kerframil> it's not really moot if he just wants a binary distro that works 00:55 < rany> spreeuw: I mean fixing bugs like meltdown, etc. 00:55 < kerframil> spreeuw: yes 00:55 < spreeuw> imo just get amd64 00:55 < sharp15> is there a program that can identify duplicate directory structures? 00:55 < bls> yeah, debian dropped i486 and i586 00:55 < spreeuw> and save yourself alot of problems 00:55 < bls> sharp15: fdupes 00:56 < bls> sharp15: or just `diff -r` 00:56 < micrex22> bls: huh... well the debian pre-boot environment sitll runs on a i586 00:56 < spreeuw> hardware is so cheap now its wonderful 00:56 < rany> cheap labor from china is so amazing 00:57 < micrex22> spreeuw well sometimes folks want specific old hardware for one reason or another 00:57 < V7> spreeuw: No a videocrad, really. 00:57 < V7> videocard * 00:57 < rany> also for geeky reasons 00:57 < V7> not a video card * 00:58 * V7 damn this keyboard 00:58 < micrex22> V7 yeah I believe most modern linux distros can't run the ThinkPad T4x Radeon GPUs anymore since the aged proprietary driver is retired or something 00:58 < bls> micrex22: https://www.debian.org/releases/stable/i386/ch02s01.html.en "However, Debian GNU/Linux stretch will not run on 586 (Pentium) or earlier processors." 00:58 < micrex22> (and FireGL, but they're virtually identical for that era) 01:00 < micrex22> bls: yeah I am aware 01:00 < rany> Are Nvidia's on Linux getting better? 01:00 < micrex22> rany: kind of? 01:00 < micrex22> my system can't run on nouveau 01:01 < micrex22> which is necessary for a lot of pre-installation scenarios 01:01 < micrex22> so then you have to use a different video card to get it installed and then inject the proprietary drivers 01:01 < rany> I mean is it worth buying a laptop with an Nvidia graphics card on bumblebee 01:01 < micrex22> sucks 01:01 < micrex22> rany depends, you could run into the same issue as I do 01:01 < micrex22> I'd go with intel, honestly... 01:01 < micrex22> save yourself the hell 01:01 < rany> intel is pretty friendly with oss 01:01 < bls> I never had any luck with my bumblebee/optimus/primus/whatever laptop and have just stuck with Intel GPUs since 01:01 < micrex22> unless you needed some high-end specs for gaming or CAD 01:02 < micrex22> although the latest intel integrated GPUs run a lot of 3D games pretty well 01:02 < micrex22> even on my X1 Carbon 01:02 < micrex22> which of course is slimmed down a bit 01:02 < sharp15> bls: thanks. 01:02 < rany> How does Intel integrated GPUs work? 01:02 < micrex22> it's baked int othe CPU 01:03 < rany> Like a different CPU altogether ? 01:03 < Psi-Jack> You mean, they actually cook it? ;) 01:03 < rany> GPU * 01:03 < Sapphirus> bls, bumblebee works fine for me. 01:03 < rascul> i'm cooking 01:03 < Sapphirus> Though, I do admit, it's hit or miss. 01:03 < Psi-Jack> rascul: Hmmm... I thought you were drinking? 01:03 < Sapphirus> Depends on the hardware. 01:03 < bls> Sapphirus: I had it work on on driver release that had some tearing issues, so I upgraded the driver, which killed bumblebee 01:04 < rascul> Psi-Jack i can multi task 01:04 < Psi-Jack> rascul: Nonsense! 01:04 < rascul> you're a nonsense 01:04 < micrex22> psi-jack yes, they bake it in special intel ovens 01:04 < micrex22> like fresh bread 01:04 < kerframil> sharp15: rsync -ain /dir1/. /dir2/. | wc -l 01:04 < rascul> does intel make good bread? 01:04 < rascul> seems like it would be crunchy 01:05 < kerframil> sharp15: add -c if you want it to be thorough to the point of comparing file checksums (not just sizes/times) 01:05 < micrex22> I think so, Paul Otellini is known for his special Italian family recipe 01:05 < micrex22> that he passed on to Intel in their manufacturing processes 01:05 < micrex22> that's why intel fabs have the most delicious sell coming out of their smoke stacks, it'll smell of fresh bread and vague burned electronics 01:05 < micrex22> mmmmmmmmm 01:05 < kerframil> sharp15: you might also want to verify the options implied by -a. for instance, if you don't care about timestamp differences. in any case, counting the number of itemized changes in a dry-run is a legit strategy. 01:06 < micrex22> smell* 01:06 < kerframil> sharp15: er, modify the options, not verify the options 01:06 < sharp15> kerframil: i'll give it a shot. i'll probably drop -a in favor of -c. none of the permissions on this stuff has been preserved over the years. 01:07 < kerframil> sharp15: you can't drop -a entirely without at least including some of the options that it implied (-r being an important one) 01:07 < sharp15> kerframil: i'll look. 01:07 < rany> rascul: I have nothing to intel about their cooking but their genius 01:07 < rany> that was a horrible pun 01:08 < kerframil> sharp15: from -ain, changing the 'a' to 'rlc' should give you a good starting point 01:13 < kerframil> sharp15: almost forgot. you'd want --delete in there too for a proper comparison. just don't forget the all-important -n, or write it out as --dry-run if you prefer. 01:24 < sharp15> kerframil: thank you. 02:01 < phinxy> I typo:d timer.sh, how do I know when my eggs are well done? 02:05 < amigojapan> HI, a long time ago I used a version of Linux which had a main menu which listed all the applications installed on the system, I have been trying to remember the name of this version of Linux but I cant rememberā€¦ for example, f you installed a new application it would automatically be added to the enu 02:06 < amigojapan> it was not a start button, but a menu on the main screen 02:12 < Amm0n> amigojapan, linux is just a kernel. What you are looking for is a Desktop Environment. There are a lot linux-distributions which all could use different desktop environments. What you discribe is nothing unusual, try google picture search with desktop environment and look if you find what you used before. XFCE4, Gnome, etc. all come with that menu. 02:13 < amigojapan> Amm0n: ok, well, it was quite a while ago, so I think the menue was text based not in X, but you are right it is not the kernel I am talking about 02:13 < Amm0n> amigojapan, you could try your luck at https://www.distrowatch.com/ too 02:14 < amigojapan> thanks 02:14 < jeffree> anyone know how to switch to a newly installe desktop environment in sparky linux? 02:22 < dannylee> become a member of the free software foundation...but if i dont pay will they break my legs...but i allready gave them $25.last night 02:23 < MNav> hello again 02:25 < Loshki> amigojapan: all the DEs do that, in one form or another, sometimes using names which differ appreciably from the actual filenames. 02:25 < amigojapan> Loshki: it was on teh main screen in a huge list, not under a start menu, and had also an ā€œadd applicationā€ button 02:26 < phogg> that could describe almost any WM if properly configured 02:26 < Loshki> amigojapan: not one I'm familiar with, sorry, but perhaps someone else will recognize it 02:26 < phogg> What year was this? Can you provide any details? How about a rough sketch of the screen layout? 02:27 < jim> Sveta, hi 02:27 < chey> hi 02:27 < amigojapan> phogg: ok, I need to go now, but i will be bakc later to try to provide that 02:27 < dannylee> https://www.youtube.com/watch?v=K5jvUXij7nU 02:27 < jim> hi 02:27 < dannylee> hi jim 02:28 < phogg> everyone is saying hi 02:28 < phogg> it's weird 02:28 < jim> for one thing, it's a way to make sure people who just joined know there's someone awake 02:28 < chey> thankyou 02:29 < chey> my point 02:29 < phogg> jim: sounds like a lot of work. Maybe there should be a greeter bot 02:29 < nemesys> hi 02:30 < jim> then it would get blamed for onjoin spam :) 02:30 < Sveta> jim: hello :) 02:30 < chey> yo nemesis 02:30 < chey> nemesis youve been here years right? 02:31 < jim> phogg, most times I'll say hi if they do, and especially if they just joined 02:31 < phogg> jim: I've noticed. What's weird is all of the other people doing it at the same time. Quite unusual. 02:31 < jim> or musical :) 02:32 < MNav> i'm gonna kit 02:32 < phogg> jim: https://www.youtube.com/watch?v=tdTRq3lxnXA 02:32 < MNav> quit 02:32 < MNav> bye 02:34 < jim> phogg, those are the same guys who hear "calling do howard, dr fine, dr howard" 02:34 < Sveta> hi StathisA 02:34 < phogg> jim: among many other things, yes 02:36 < chey> my laptop broke this morning so im on my phone which is better tbf 02:36 < Sveta> hi :) what happened with your laptop, chey? 02:37 < chey> hey um the main battery is broke 02:37 < chey> its an acer 02:37 < chey> it wont charge 02:38 < chey> so i bined it lol 02:39 < chey> binned 02:40 < chey> never quit 02:42 < Sveta> chey, what continent are you in? 02:42 < Sveta> northern america right? 02:42 < chey> sunny ol england 02:42 < Sveta> what make was it? 02:43 < chey> acer 02:43 < Sveta> acer what? was it an acer travelmate? 02:43 < chey> 5535 02:43 < Sveta> hmm 02:43 < chey> its old 02:44 < chey> i just installed linux mint on it 02:45 < Dreaman> what is the problem i use acer and ubuntu but dual boot 10 win 02:45 < chey> it was blinking orange 5 times 02:45 < jeffree> just curious if any of you guys use Common Desktop Environment? 02:45 < chey> its fooked 02:46 < chey> yo jeffree 02:46 < Dreaman> mint channel ask 02:46 < jeffree> hi 02:46 < Dreaman> i no bprob;ems 02:46 < Sveta> chey, you can have one for Ā£7.00... https://www.ebay.co.uk/itm/361208332231 02:46 < Dreaman> probem 02:46 < chey> oh lol 02:46 < chey> thankyou sveta 02:47 < Sveta> no worries, if you're lucky you may get it before monday 02:47 < chey> cool awsome ill do that 02:48 < Sveta> chey: the battery is Ā£10.55, https://www.ebay.co.uk/itm/323218496231 02:48 < Sveta> chey: you'd have to figure out whether it's the charger or the battery, I'm not sure how to do that properly.. if you plug it into the mains and it works, maybe you don't need a new charger, maybe you need a new battery 02:49 < TRS-80> evening fine people 02:49 < chey> i was gonna get a windows one, expensive though 02:49 < chey> not battery laptop 02:50 < BenderRodriguez> dear linux 02:50 < BenderRodriguez> I need help 02:50 < chey> go f it 02:50 < BenderRodriguez> i have a linux md raid array setup, it's marking a hard drive as faulty, but smartctl shows it has having passed the self-assessment 02:50 < BenderRodriguez> so...what's going on 02:50 < BenderRodriguez> what should I do 02:50 < TRS-80> OK so following this http://docs.python-guide.org/en/latest/dev/virtualenvs/ python --version returns 2.7.15, although this guide says I should have 3, doing apt-cache rdepends python2.7 gives a quite long list of things, so ??? PLS HALP 02:51 < crazyblob> what is your guys favorite distros? 02:51 < Sveta> hi TRS-80 :) 02:51 < chey> use disks 02:51 < TRS-80> crazyblob: Debian 02:51 < TRS-80> Hi Sveta! :) 02:51 < chey> bender 02:51 < Sveta> crazyblob i have a few peeps here with different needs. they're met by different distros :) 02:51 < Sveta> crazyblob what's your needs? 02:51 < St3ak> linux mint 02:51 < crazyblob> nothing, i use manjaro. i was just wondering 02:51 < chey> i use mint and kali 02:52 < crazyblob> new to linux, so i chose manjaro 02:52 < TRS-80> l33t 02:52 < chey> mint for noobs 02:52 < triceratux> crazyblob: mx-17, swagarch, lubuntu 18.10 lxqt 02:52 < BenderRodriguez> chey: what do you mean 02:52 < chey> or dabian 02:52 < BenderRodriguez> "use disks"? 02:52 < TRS-80> kids these days kek 02:52 < toothe> If I want to know what encoding the terminal is in, what variable should I check? 02:52 < chey> its an app 02:52 < toothe> I presume LANG? Or is there something else? 02:53 < Loshki> BenderRodriguez: use gsmartmontools and run a long/full test. Better yet, boot and run the manufacturer's own diagnostics package, as they generally know things about their own hardware no one else is privy to. 02:53 < chey> partition manager 02:53 < Sveta> chey: if it doesn't boot when you plug it into mains, start with ordering the AC adapter 02:53 < chey> ok sveta blessed 02:54 < chey> sveta: whats your chosen setup 02:55 < Sveta> chey: internet suggests trying to boot without the battery to narrow it down 02:56 < Sveta> chey: i use fujitsu s7220, today morning i broke my fan while cleaning it, it had layers of dust 02:56 < chey> wont boot without 02:56 < Sveta> chey: in the evening i plan to try cleaning it again and put it together more properly 02:56 < Sveta> chey: i apparently missed it in the user manual where it said to clean the fans each year at least once 02:57 < chey> i havnt heard from fugitsu for years 02:57 < Sveta> chey: it's an old laptop model from 2009 or so, i used to run ubuntu for a year and then debian for a few more years on it 02:57 < chey> oh you must clean the cobwebs then 02:58 < chey> haha 02:58 < Sveta> chey: its lcd is blinking so i was using an external monitor for a while, but now i had same symptoms for a friend's laptop that was recovered by cleaning their fan properly, now i went to do the same in my laptop but in the 20 or so minutes in the morning that did not work well 02:58 < chey> oh no 02:59 < chey> its probably so old now 02:59 < Sveta> chey: in our apartment there's lots of dust inside because we have carpets. we ventilate and vacuum regularly, but in the computer it accumulates :P 02:59 < chey> uuum skin 02:59 < Psi-Jack> Does someone here have Debian, Ubuntu, or Fedora and has pinentry-gnome3 installed and could tell me what package that program is owned by? 03:00 < Sveta> chey: yea it is quite old, i'm thinking of getting a lenovo thinkpad x200 but then i realised it is old as well. not sure which of the newer models is reasonably pretty and linux friendly yet 03:00 < chey> freebird .... google 03:00 < TRS-80> Sveta I had an old T60 that I broke the fan blades trying to blow compressed air in there to clean, first it made noise but eventually all the remaining pieces broke off, think kept going for years, I think I still have it around here somewhere... never died even with no fan! 03:00 < chey> lenovo are brill 03:00 < Psi-Jack> Hmm.. Oh.. In Debian it seems to actually BE pinentry-gnome3 package. 03:00 < Sveta> Psi-Jack, take a look at https://packages.debian.org/search?searchon=contents&keywords=pinentry-gnome3&mode=exactfilename&suite=stable&arch=any 03:00 < Sveta> Psi-Jack, yea 03:01 < TRS-80> chey excuse me, sir it was a ThinkPad 03:01 * TRS-80 raises nose 03:01 < BenderRodriguez> crazyblob: RHEL/Centos! 03:01 < BenderRodriguez> there's nothing more stable 03:01 < Sveta> TRS-80: hey that's pure luck, it's like cycling without brakes 03:01 < TRS-80> er, IBM ThinkPad, should have said XD 03:01 < chey> hehe 03:01 < Sveta> TRS-80: i'm kinda impressed :) 03:02 < granttrec> how can I remove garbage ascii characters from a file? 03:02 < TRS-80> legendary reliability, had several of them mounted in trucks in landscape company, drinks spilled, dirty environment, landscapers using them, never a problem! 03:02 < chey> i had a free lenovo in my tv package 03:02 < chey> 32bit 03:03 < triceratux> Psi-Jack: on lubuntu 18.10 its its own package as well: pinentry-gnome3. has its own manpage 03:03 < Psi-Jack> Yeah. Hmmm 03:03 < Psi-Jack> I thought pinentry-gnome3 was part of gnome-keyring... I may be wrong. 03:03 < jim> granttrec, like which garbage chars? 03:03 < toothe> Is there a way to determine if your terminal is capable of displaying unicode characters and/or UTF-8 characters? 03:03 < toothe> err, UTF-8 encoded strings* 03:04 < Sveta> Psi-Jack: same in ubuntu; in fedora also maybe, see https://apps.fedoraproject.org/packages/pinentry-gnome3 03:04 < chey> sveta: iv got a new phone my pride and joy 03:04 < granttrec> toothe: \u201c and such, seem like artifacts picked up from copying from a pdf 03:04 < Sveta> chey: which one? 03:04 < toothe> granttrec: huh? 03:04 < chey> galaxy s3 2016 was my mothers 03:04 < Sveta> chey, cool 03:04 < Sveta> chey, it runs android? 03:05 < chey> its sooo nice and yea 03:05 < allorder> toothe: https://apps.timwhitlock.info/emoji/tables/unicode copy/past 03:05 < Sveta> chey, super. you could install the fdroid app and forget about the ads from apps 03:05 < granttrec> toothe: \u201c is an example char, they came from a pdf I copied pasted into an editor 03:05 < TRS-80> here's the video I was looking for: :) Legends of ThinkPad - "Melted" https://www.youtube.com/watch?v=3ENQ1dUavI8 03:05 < Sveta> chey, it is the same as google play, but it only lists apps which are safe to use 03:05 < toothe> granttrec: pardon? 03:06 < chey> sveta: i get no adds i pay for my apps 03:06 < granttrec> ehh sorry ment to ping jim 03:06 < Psi-Jack> Hmm, yep. Definitely part of pinentry itself. Thanks. :) 03:07 < chey> sveta: have you seen tvplayer 03:07 < chey> its amazingly clear tv on android 03:08 < chey> and samsung are great at no adds apps 03:09 < chey> sveta: once you go samsung you never go wrong lol 03:09 < chey> wharf you cool? 03:09 < Sveta> chey: yea i'll give it a go, i have a phone from 2009. it's working ok (nokia e65). i'll get something new from samsung when i figure out what i'd like it to do 03:10 < TRS-80> Sveta have you heard about Librem phone? 03:10 < jim> granttrec, that's getting into 15 bit chars, and I haven't personally worked with those... 03:10 < chey> sveta: samsung are fir the peopke 03:10 < chey> people 03:10 < Sveta> TRS-80: i'm waiting for them to negotiate ryf certification with fsf, i think they were aiming to get that sometime this year 03:10 < jim> for 8 (and 7) bit chars, there are tools to deal with those 03:11 < chey> * is up all night 03:11 < granttrec> jim: can you suggest me some 03:11 < jim> sed, tr 03:11 < jim> they're basically filters 03:11 < chey> anyone use pacman...i hate it so much 03:11 < granttrec> jim: thanks, going with tr 03:11 < Sveta> TRS-80: they're a bit expensive 03:12 < Sveta> chey, it didn't do something for you today 03:12 < jim> I wonder if you can use python or perl 03:12 < Wharf> hey 03:12 < chey> sveta: oh it just a pain i prefer apt 03:12 < Wharf> sorry i had to identify 03:12 < jim> (but then you'd be forced to learn one of those) 03:13 < Wharf> Hey chey, im cool lol 03:13 < granttrec> jim: might learn pearl tbh seems really good for text processing 03:13 < chey> wharf: good good 03:13 < Wharf> Go easy on me, expected to get slated pmsl 03:14 < Wharf> Any idea's how i can set up an ftp server on a vps im renting. basicly wanted it to point at an apache directory 03:14 < rasknikoff> its a great day tooo be alive..i just really love florida??? 03:14 < jim> granttrec, perl -probably- does too, if so, good choice,,, I know python has tools to deal with other char sets 03:14 < Sveta> chey: hmm, so use apt, i hope that works 03:14 < Wharf> i did try, but to be honest i think it went tis up! 03:14 < rasknikoff> we have beaches 03:14 < Wharf> tits 03:14 < TRS-80> Sveta yes they are but seems like good hardware that is actually free usually is, see also Bunny Huang laptop project. But it's smaller productions runs since (unfortunately) not enough people care about freedom. I'm usually pretty cheap but I will likely pony up for the Librem 5 at some point if they make it a reality. And it looks like they are well on their way. I wish them luck. 03:14 < rasknikoff> ya man 03:14 < Wharf> i been trying to sort it via ssh 03:15 < TRS-80> for any interested lurkers too lazy to search :) https://puri.sm/shop/librem-5/ 03:15 < chey> sveta: yes but i use kali for arm and apt breaks it sometimes 03:15 < chey> raspberry pi it up 03:16 < rasknikoff> richard stallman might really be G0D...ok i just gave him $25.oo..i just like him.. 03:16 < TRS-80> rasknikoff: all hail our lord and savior! XD 03:16 < rasknikoff> ok 03:17 < Sveta> chey: for arm? you mean your arm on raspberry pi? 03:18 < Wharf> also another headache ive had, setting up a red5 server for video and voice chat. 03:18 < chey> sveta: im off now...yes arm for raspberry pi ..im gonna chill...you here tomorrow or tonight? 03:18 < Wharf> simple instructions, edit the flv flash file, add ur own ip. job done apparently. 03:18 < Wharf> seems every program i use to read the file it shows up as gibarish! 03:19 < chey> bye everyone stay healthy 03:19 < Wharf> you too mate 03:20 < TRS-80> Wharf: NextCloud supposedly has a nice video + voice chat in the new version I'm looking forward to testing out but I need more storage to backup so I can upgrade lol 03:20 < Sveta> chey: yea 03:21 < BenderRodriguez> are there any good alternatives to rdiff-backup that's not as slow as rsync 03:21 < BenderRodriguez> rdiff is nice, but those developers seem to be alergic to stdout and printing useful information 03:22 < TRS-80> BenderRodriguez: no more verbose option? 03:23 < BenderRodriguez> the vebrose option seem to have three levels, at level 2, it prints something at the beginning, but no progress of any kind, level 3 just completely covers the screen what seems to be hex or binary characters, possibly the data being transferred? 03:23 < BenderRodriguez> anyway, the tool is absurd -- i need something more sane 03:24 < TRS-80> oh my 03:24 < BenderRodriguez> TRS-80: I guess I'll stick with trusty rsync for now 03:26 < TRS-80> BenderRodriguez: not sure what ultimate goal is but are you aware of oh what's that decentralized sync tool? syncthing 03:28 < BenderRodriguez> hmm..looks interesting 03:29 < BenderRodriguez> Never heard of it 03:32 < Elladan> I use syncthing, it works well. Especially with the inotify extension. 03:32 < Elladan> BenderRodriguez, I like borg for backups. 03:33 < Elladan> It doesn't keep files in an plain format like rdiff-backup does, but you can mount the repo. And it's much, much faster. 03:34 < TRS-80> BenderRodriguez: I didn't know if you were trying to just do a simple a->b backup or what but syncthing is such a cool tool that's still relatively new not everyone has heard of it, but anyone I have spoken to that has used it has had nothing but good things to say. So I thought I'd throw it out there. 03:35 < TRS-80> Elladan: I have heard a lot of good about borg as well, although I need to read up on it like I have read on syncthing so I can understand how it works. 03:36 < Elladan> I can try to give you the hyper-brief version. 03:37 < TRS-80> go on 03:37 < Elladan> It breaks your files up into chunks and then stores them in what amounts to a database, deduplicated, compressed, and encrypted. 03:38 < TRS-80> hmm 03:38 < Elladan> Each backup is a full backup -- with deduplication. The encryption is done on the client side, so it's meant to be suitable for storing your backup on an untrusted machine like a VPS. 03:39 < Elladan> A backup on my machines tends to take a minute or two, over the LAN (more if there's a lot of new data, of course). 03:39 < TRS-80> so "a full backup with deduplication" isn't that ~ a different way of saying "incremental backup" 03:39 < u0m3_> hi. anyone around here have experience with mobile modems and wvdial? I ran into a strange issue: if I connect from windows, all my ports are open. From linux all are closed. iptables -L reports all clear 03:39 < Elladan> It's similar, but different from an incremental in important ways. 03:40 < TRS-80> Elladan: somehow I thought you might say that, lol 03:41 < Elladan> TRS-80, when people talk about incrementals e.g. with tar, they generally mean a backup file(set) which references another backup, but only contains changed files. 03:41 < TRS-80> yesm but functionally... ? 03:42 < Elladan> TRS-80, with borg it all goes into the same repository, conceptually like a database. Also, the deduplication is done on smaller chunks than whole files. 03:42 < TRS-80> aah ok yes chunks 03:42 < Elladan> Functionally they're different. Like with a tar backup, you might put your full on one USB disk, and then some incrementals on a second one and rotate around. 03:43 < Elladan> ... and if you lose the full you lose everything. 03:43 < __Myst__> Is there an utility that lists all configuration files I may want to commit to a dotfiles repo? 03:43 < __Myst__> Or even just a plain ol' text list? 03:44 < Elladan> TRS-80, Also, you can FUSE mount borg repositories (over the network), so restoring is very easy. 03:44 < jim> system-wide config files are usually in /etc and config files that relate to just your user are usually in that user's home dir 03:44 < Elladan> There's a tool written in golang called Restic which is somewhat similar to borg, but seems to be more targeted at cloud backups BTW. 03:45 < TRS-80> Elladan: I have been thinking (actually started buying initial hardware for) a ZFS storage array / file server which also has a lot of super neat features, as my backup / storage solution. 03:46 < TRS-80> including deduplication, etc. 03:46 < Elladan> TRS-80, I'm using btrfs on my storage array, and use borg for backing up machines to it (and also for backups of the NAS, to a USB disk). 03:47 < t5u> hi is anyone using p2pool ? 03:47 < Elladan> I considered ZFS, but I didn't like it for various reasons. I didn't like btrfs either for other reasons, of course. 03:50 < TRS-80> Elladan: yeah impression I got after reading a fair amount was sort of "btrfs may be better in theory and/or some day but ZFS is better now" although as you say lots of pros and cons to both and both seem to be under active development which ultimately will be a Good Thing for all of us 03:50 < jim> Search for pinentry-gnome3 in stretch/amd64: pinentry-gnome3: usr/bin/pinentry-gnome3 04:08 < Elladan> TRS-80, ZFS is a lot more stable, and performs better in some cases. btrfs OTOH has some features ZFS doesn't, including much better expandability. 04:18 < k_sze[work]> Can LVM2 snapshotting cause a machine to gradually become slower and slower? 04:19 < Wixy> how do I remove/clear dns cache so I request the dns server every time with tools like nslookup or dig? 04:21 < jim> usually I just stop and start my dns server, that removes the cache 04:21 < Wixy> but it's not my dns server 04:21 < Wixy> I found a way, it's "dig @dnsserver domain" 04:21 < jim> you could install your own... 04:22 < Psi-Jack> That dig doesn't clear dns cache. :p 04:22 < Wixy> no, I know. but it does what I wanted 04:23 < Wixy> I'm trying to get alternative IPs from a certain domain, which only return 1 for whatever reason 04:23 < Wixy> $ dig domani A <- returns 1 IP, and after some time another, and so on 04:23 < Wixy> it's cloudfront btw, don't know why it works like that 04:24 < qman__> dig and nslookup don't cache the result, the dns server does 04:24 < Wixy> that means I'll still see the same IP over and over? 04:24 < qman__> if you control the next upstream server, you can clear the cache there 04:25 < Wixy> I don't control any of the servers, but I thought using dig @nsserver would actually send the request to the dns server, and thus I would get a fresh record each time 04:25 < qman__> it does, but the server caches its queries 04:26 < kerframil> i.e. there's no guarantee of freshness unless the response is authoritative 04:26 < qman__> per the TTL settings it gets from the upstream server 04:26 < qman__> if you query repeatedly, you'll notice the TTL counting down 04:27 < Wixy> qman__, where's the counter exactly? 04:27 < qman__> in the response 04:27 < Wixy> dig d3h36i1mno13q3.cloudfront.net 04:28 < Wixy> that "id" on the line that says "header"? 04:28 < Psi-Jack> No the TTL 04:28 < qman__> no 04:28 < Psi-Jack> Time To Lilve 04:28 < Psi-Jack> Live* 04:28 < qman__> you'll get a record back, with a TTL value 04:28 < qman__> that's how many seconds that response is good for 04:28 < Dominian> TIME TO LIVE 04:28 < Dominian> AHHHH! 04:28 < qman__> when that time expires, the server will send a new request upstream 04:28 < Wixy> ohh I see 04:28 < Dominian> Sometimes I wish they called TTL YOLO 04:28 < Wixy> it looks like it's 60s by default 04:28 < Wixy> err, remove that "by default". I meant in this case 04:29 < Wixy> is there a way to get more than one A register? 04:29 < Wixy> it would only send me one IP, I know they have probably a hundred 04:29 < Psi-Jack> Wixy: Ask them 04:30 < Wixy> I know some dns servers would send more than one, but this in particular won't 04:30 < qman__> they respond with a specific IP because that's the one they want you to hit 04:30 < Wixy> even with "dig domain any" 04:30 < fryguy> Wixy: what's not how DNS works 04:30 < Psi-Jack> Man, feels REALLY WIERD, having a TENS on the back side of your neck. :) 04:30 < fryguy> Wixy: like we've said at least a half dozen times at this point, it's anycast 04:31 < Psi-Jack> Wixy: CloudFront uses a customized DNS server that's designed to do what it's designed to do. 04:31 < fryguy> also geographic dns 04:31 < fryguy> so you get different results based on where you are in the world 04:32 < Wixy> fryguy, sorry, I still don't know how anycast works, I've added it to my todo list for tonight :) 04:32 < qman__> it's not important how it works internally, what you need to know is that they respond with a specific IP that they want you to hit, based on geographic location, load balancing, and other factors 04:33 < qman__> that's how it's designed to work 04:33 < Wixy> got it 04:33 < Wixy> the problem is it would give me really bad ips for my use case, that's what I'm trying to fix 04:34 < qman__> the IPs aren't the problem 04:34 < Wixy> some of them are really good, latency is low. but then I see traceroutes with 5x that latency and 10x the number of hops 04:34 < qman__> the system is not designed to work the way you want it to, it's designed to work the way they want it to 04:34 < Dominian> Psi-Jack: You host your email internally? 04:34 < fryguy> Wixy: again. like we've said before, hours ago: you need to contact the service provider directly 04:34 < Psi-Jack> Dominian: Yes 04:35 < Dominian> Psi-Jack: oh right.. postfix correct? 04:35 < Psi-Jack> Kolab, which uses Postfix, still currently. 04:35 < qman__> low latency is not only not the top priority, it's actually quite low on the priority 04:35 < Dominian> ah 04:35 < Psi-Jack> Eventually getting away from Kolab, but.. 04:35 < Dominian> I've not used kolab.. hmm 04:35 < Wixy> fryguy, I contacted them already. that was as expected. they won't disclose or provide any direct access 04:35 < qman__> they care more about load and bandwidth 04:35 < Psi-Jack> Dominian: It's postfix+cyrus-imapd+kolab stuff. 04:35 < Dominian> Psi-Jack: I'm debating moving email back in house 04:36 < Wixy> qman__, I know, that's why I have to "fix" it myself. latency is my top 1 priority :) 04:36 < Psi-Jack> Dominian: Heh, yeah. I plan to re-do my entire mail server setup to customized blend of postfix+dovecot+openldap 04:36 < fryguy> Wixy: ok. and you know -- and have been told -- multiple times this is how this load balancing works and given an explanation about why you are seeing the latency you are seeing, so why are you still pursuing this 04:36 < fryguy> there isn't a magic answer here 04:36 < Dominian> Psi-Jack: hehe 04:36 < qman__> what I'm saying is that you can't, not without control over the other end 04:37 < Dominian> Psi-Jack: Most likely I'd do postfix+dovecot with mailscanner 04:37 < Dominian> which of course irritates the postfix guys 04:37 < Psi-Jack> Dominian: Bleh, no mailscanner. amavisd-new 04:37 < Dominian> Psi-Jack: I've used amavisd-new 04:38 < Dominian> Psi-Jack: ever find a decnet web ui for it? 04:38 < Psi-Jack> Not really... LOL 04:38 < Dominian> bingo 04:38 < Wixy> fryguy, I'm still pursuing it because it's the difference between it working or not. like when you must do something, or else your company fails... kind of :P 04:38 < ryouma> when i try to umount dir, i get target is busy. what fuser or lsof command should i run? i tried lsof|grep and fuser on the dir. both revealed nothing. (realized this is maybe a question for this channel not debian) 04:38 < Psi-Jack> Hmmm... Mailscanner has a decent webui? 04:38 < Dominian> Psi-Jack: it's not bad 04:38 < Dominian> Psi-Jack: postfix guys don't like it cause it interacts directly with the postfix queues, but I never had an issue with it 04:38 < Psi-Jack> I'll have to check it out again with more current versions. 04:38 < fryguy> Wixy: how many more people need to tell you before you'll listen? 04:39 < Psi-Jack> We use mailscanner at work too, and I cringed when I found out. LOL 04:39 < Psi-Jack> fryguy: 20 04:39 < Dominian> Psi-Jack: Why did you cringe? 04:39 < Wixy> fryguy, I've learned quite a lot this afternoon, I don't thing it was a waste of time btw 04:39 < Dominian> it works really well 04:39 < Dominian> Psi-Jack: although I might look at using exim instead 04:39 < Psi-Jack> Dominian: mailscanner was horrible, last I tried it. 04:40 < Dominian> weird 04:40 < Dominian> I had no issues with it 04:40 < Dominian> Psi-Jack: I wouldsuggest using Mailwatch with it 04:40 < Dominian> Psi-Jack: https://mailwatch.org/ 04:41 < qman__> I just use scrollout F1 04:41 < qman__> let someone else figure out how to configure all the mail scanning 04:42 < Dominian> Psi-Jack: good th ing is, mailwatch now supports imap auth 04:42 < Dominian> I had to 'hack' something into it before to authenticate against postfix before 04:43 < Psi-Jack> heh 04:45 < Psi-Jack> Dominian: Hmmm, mailscanner is designed for a mail gateway, as in a seperate host? 04:54 < palayoub> Hey, I have a very confusing question, I want to check what websites a process is calling, I have an application and it fetchs data from a server, is there away I can see the website path of the server the app is using to get the data from? 04:54 < Abbott> what's the point of setting system/daemon users shell to /bin/false if you can just choose a shell when using su 05:01 < kerframil> Abbott: to prevent them from having a valid login shell 05:02 < Sitri> palayoub: strace, gdb and/or tcpdump 05:02 < Sitri> Any one of those should work. If it's HTTPS instead, tcpdump won't help you. 05:05 < palayoub> got you, tcpdump doesnt log https calls? 05:05 < Dominian> Psi-Jack: hmm I've always run it on the same host 05:05 < Psi-Jack> Hmmm. OKay. 05:05 < Dominian> Psi-Jack: Where I plan on running it, inbound 25 should work, but I'll have to relay it through google 05:06 < Dominian> relay outbound through google that is 05:06 < Psi-Jack> Heh 05:06 < Sitri> palayoub: it does. But it doesn't tell you the URL 05:06 < Sitri> You could get the domain, since that is sent in the clear 05:07 < palayoub> Sitri: good to know, i am trying to install wireshark but i think it doesn't tell the url too, i am lookin for the full path of the url 05:07 < Sitri> Wireshark has the exact same restrictions as tcpdump 05:07 < kerframil> palayoub: tcpdump is not a tool for conducting MITM attacks on HTTPS connections 05:09 < palayoub> yeah, but i am not trying to get the data sent, i am only lookin for the url which is not a big deal iguess, kerframil 05:09 < kerframil> that is "the data sent", though 05:09 < kerframil> part of it, anyway 05:10 < kerframil> specifically, you need to see the path requested by the user-agent, which is protected by TLS at that point 05:13 < palayoub> hmmm, the full path url is protected too and encrypted? only domains that are open? 05:14 < fryguy> right 05:14 < kerframil> it's transport layer security. the user-agent issues its request within the context of an encrypted session. 05:15 < fryguy> and it's only sort-of a hack that you get the domain name too (SNI) 05:16 < Soni> we don't need better swap, we need this: https://cybre.space/@SoniEx2/100234988015757717 05:16 < kerframil> yeah, the desired host can be leaked via SNI 05:16 < Soni> just give the JVM something to make it free up some of its memory 05:16 < palayoub> that leaves me with the assumption that i can't get the full path anyway, what about the tools Sitri suggested, do they give some sort of hack too? strace, gdb 05:20 < Soni> or uh, with a bit more context https://cybre.space/@SoniEx2/100235002289973544 05:20 < Soni> not sure how much that helps 05:20 < Soni> (in theory this would really help with being required to run 20 electron apps on the same machine) 05:28 < Todden> http://imgur.com/sQj8qXg when reaver cannot get enough signal 05:28 < Todden> So you duct tape a wireless adapter to your balcony 05:32 < Todden> Well im confused,says right link 05:33 < ffejj> hello, what program do you guys use to view your logfiles? 05:33 < lnnb> cat 05:33 < ffejj> no tail? 05:36 * ffejj checkin out that multitail 05:36 < vuex> generic question; I'm thinking about efficiently storing data. So like, lets say I have AAAAAAAAAAAAAA and AAAAAAAAAAAAAB.. only the B differs between these. Anyone know any approach to efficiently store the diffs? So like.. the second entry would somehow link to the first entry and record just B as the change. I'm hoping this saves on space. 05:37 < jim> is this a text file? 05:38 < vuex> can be whatever. 05:38 < jim> if it's a text file, maybe git 05:38 < vuex> I was just thinking one of you guys may know a better approach, something already written.. likely postgresql has something that does this, etc... 05:38 < lnnb> vuex: something like this? https://en.wikipedia.org/wiki/Huffman_coding 05:39 < jim> that's what git does, is store differences between files efficiently, and also in such a way they can be accessed very quickly 05:41 < pxfgod1> in Linux tcp/ip protocol stack, is TCP checksum possible to ACK a wrong packet, (a wrong content & a wrong checksum happen to bypass the detection). Is it rare or absolutely never? 05:41 < vuex> wow, huffman is exactly what I was thinking. Thanks fro that 05:42 * ffejj sigh watching /var/auth.log 05:42 < kerframil> vuex: talk of "diffs" makes it sound like you want delta encoding 05:42 < Psi-Jack> ffejj: What's it doing in /var? ;) 05:42 < kerframil> vuex: RFC 3284 may be of interest (xdelta is an implementation) 05:43 < ffejj> that's where windex stores it 05:43 < Psi-Jack> Where.. What? 05:44 < ffejj> Psi-Jack: :) 05:45 < vuex> thank you 05:47 < jim> Dear irccloud devs, could you find a way to make these joins less intense? 05:48 < ffejj> jim: weechat has a smart join/part script thing, that doesn't show join/part of idle users. 05:48 < Sveta> jim: mailto:team@irccloud.com 05:49 < ffejj> do y'all have any suggestions for servers where there are still a lot of idiots and ansi color spam? 05:50 < ffejj> i need my people 05:50 < Todden> Hmm 05:50 < Todden> You could go to tripsit 05:50 < Todden> Everyone is always high,makes for hilarious stuff 05:50 < Sveta> ffejj: irc server? 05:51 < ffejj> @todden is that a freenode channel? @sveta yes 05:51 < Todden> nAH ffejj 05:51 < Todden> Its own server 05:51 < aaro> ffejj: maybe rizon can fill your expectations 05:51 < Todden> Its a harm reduction community focused on..reducing the harm associated with drug use 05:51 < Todden> Not abstinence 05:51 < Sveta> ffejj: you need more moderators then i guess (they need to identify spam and remove it) 05:52 < Todden> Generally its a bunch of addicts warning people not to get addicted 05:52 < Todden> Oh we dont get ansi color spam 05:52 < Todden> We just get idiots 05:52 < Sveta> ffejj: they may use semi-automated assistants, powered by irc bots that have AI and learn what spam is like 05:52 < Todden> We actually dont have much of a spam issue 05:52 < Todden> We do have a bot,called tripbot..another called DrTripServington 05:52 < Todden> Handles a lot of stuff 05:53 < Sveta> Todden: then more moderators :P they need to praise each improvement in each person's behaviour 05:53 < ffejj> when i said spam, i was talking about those brightly colored chat messages from old irc scripts 05:53 < Todden> Oh thats long gone,every server p much auto bans them 05:53 < Todden> Its much more difficult than moderating most places Sveta 05:53 < ffejj> awww. 05:53 < Todden> You ever had to watch someone die on camera? 05:53 < Todden> Live,who you knew 05:54 < Sveta> i watched something similar in person. although not death 05:57 < ffejj> thanks for the recommendations 06:03 < michaelrose> Todden, if you are going to post snuff videos you can post them elsewhere 06:04 < Sveta> hey michaelrose, how are you? 06:05 < thatpythonguy> that moment when your internet is so bad that even speedtest-cli won't even run but 15% of the time 06:05 < michaelrose> well 06:05 < Todden> Wait snuff videos/ 06:05 < Todden> I didnt post snuff videos? 06:06 < michaelrose> I was kind of wondering why you were asking about watching videos of people dying 06:06 < Todden> Because Sveta asked about moderating that channel 06:06 < Todden> Or needing more 06:07 < Todden> And i gave an example why there arent many volunteers 06:07 < Sveta> michaelrose, yea that was in context of a question about moderation.. a bit strange place to ask, but that conversation seems to have already moved 06:07 < Todden> Yeah its moved on 06:07 < Todden> Wrong channel 06:07 < Todden> I met my girlfriend(LDR) by saving her life via an opiate overdose however,so its not all bad 06:07 < Todden> Anyway as for linux 06:08 < Todden> Can we ask about linux programs or just kernal 06:08 < Todden> Im having issues with reaver and anything below above(below?) -50db 06:10 < [R]> how ridiculous is this... systemd supports socket activation, which is pretty cool, but it DOESN'T support telling a depepdent servie the information on a listening socket 06:10 < Sveta> Todden: linux programs are ok 06:12 < michaelrose> [R], were you expecting good design? 06:12 < [R]> lol 06:12 < [R]> i'm going to be using systemd on my embedded system 06:12 < [R]> am i the devil? 06:13 < kurahaupo> [R]: no, you're Eve who's been convinced to byte the systemd apple 06:13 < [R]> what would you have me use instead? 06:25 < sauvin> Todden, would you mind joining me in ##chops-lounge, please? 07:18 < Psi-Jack> [R]: systemd for embedded? Noice! 07:18 < [R]> Psi-Jack: i'm testing it, and its slow as balls though... 07:18 < [R]> userspace is taking 6 seconds 07:19 < pingfloyd> bloat usually does that 07:19 < [R]> lol 07:19 < pingfloyd> got to bloat ALL the devices! 07:19 < [R]> [09:13:48] <[R]> what would you have me use instead? 07:20 < lnnb> systeme 07:20 < pingfloyd> what happened to the simple old days where it was sysv vs bsd? 07:20 < [R]> well, socket activation and depndencies has me intrigued for my use case 07:21 < [R]> i'm using busybox's version of daemontools right now 07:21 < pingfloyd> runit maybe? 07:21 < [R]> i dont like the fact that i can't ensure my controlling daemon is started before my services 07:21 < [R]> yeah, tahts what im using 07:22 < [R]> i suppose i could have a sub tree of services that my controlling daemon starts 07:23 < [R]> i was surprised that busybox's runit implementation is bsd licesnsed 07:55 < AliSh> hi, I'm working with web app that log in a file continuously. It's large file. I don't have time to change the log way (i know it's dirty way). how can i get to end of file in terminal? I used tail but it show appended data continuously. I want to stop in some section. 07:55 < [R]> large... you say... is it YUGE? 07:56 < [R]> AliSh: "stop"? 07:56 < AliSh> it's absolutely f***ing large for terminal. 07:56 < [R]> what? 07:56 < Happyhobo> HI DUDES 07:56 < AliSh> [R], I mean i want see some section. 07:56 < [R]> "some section"? 07:57 < AliSh> [R], when i use tail it scroll in every data append 07:57 < Happyhobo> sl 07:57 < [R]> then you're running it wrong 07:57 < Happyhobo> sl 07:57 < storge> i think AliSh means some sort of grep or something 07:57 < Happyhobo> dammit 07:57 < pingfloyd> so don't use -f flag with tail 07:57 < AliSh> i didn't use it pingfloyd 07:58 < [R]> Psi-Jack: got it down to 3.7s... still too long 07:58 < [R]> AliSh: tail will just output the last few lines, and exit 07:58 < Happyhobo> Is anyone else seeing all these exits? 07:58 < pingfloyd> alias | grep tail 07:59 < AliSh> fuck,,, how can i filter this ... has quit messages???? 07:59 < micrex22> alish depends on your IRC client 07:59 < Happyhobo> will it ever end 07:59 < AliSh> micrex22 i'm using freenode 07:59 < micrex22> I know quassel has a quick n' easy function to mute them 07:59 < AliSh> webchate 08:00 < micrex22> I mean the IRC client not the IRC server 08:00 < [R]> AliSh: step 1... get a real cclient 08:00 < pingfloyd> freenode is an irc network, not client 08:00 < AliSh> ok, thanks 08:00 < Happyhobo> pidgin doesn't 08:00 < pingfloyd> install weechat and be done with it 08:00 < AliSh> let me search for some irc client in linux 08:01 < kirk781> hexchat is good and simple AliSh 08:01 < kirk781> I tried using irssi but gave up 08:01 < xormo> I prefer HexChat myself. 08:01 < pingfloyd> weechat is easy if you can read a manual 08:01 < AliSh> ok, i'm going to check them 08:01 < xormo> sudo apt install hexchat or choose it from Synaptic and install it. 08:02 < micrex22> kirk781 irssi is awesome if you want a terminal IRC client 08:02 < kirk781> I have used Smuxi but it's just second rate, IMO. Need to check out weechat 08:02 < kirk781> micrex22, I know, but it presumably required a lot of initial configuration. In terms of features, is it par with other standard clients? 08:02 < Happyhobo> night folks 08:03 < kirk781> afternoon folks 08:03 < micrex22> kirk781 initial configuration?! 08:03 < micrex22> it's as instananeous as any other client 08:03 < kirk781> micrex22, frankly, I was just 'scared' of its UI and assumed it would require more tinkering and didn't touch it again. Would give it another try though 08:04 < micrex22> kirk781the only thing you should be scared of, is NTFS :p 08:04 < kirk781> micrex22, the file-system? 08:04 < pingfloyd> weechat is like irssi++ 08:05 < micrex22> yeah :p because it's awful to deal with and is hard-coded with a lot of problems :p 08:06 < alish> ok, I installed hexchat. It seems user friendly for now. Thanks 08:07 < kirk781> Hexchat also has support for themes alish 08:08 < alish> kirk781, how can i turn quit messages off 08:08 < alish> ? 08:08 < micrex22> it's in settings somewhere iirc 08:09 < pingfloyd> here's the best way in weechat https://weechat.org/blog/post/2008/10/25/Smart-IRC-join-part-quit-message-filter 08:09 < alish> ok, i found it. setting -> preferences -> general -> hide join... 08:09 < alish> :D 08:09 < pingfloyd> it's nice because you don't see all the channel join/part/quit spam, but you do catch if people that you're chatting with leave. 08:10 < alish> pingfloyd, yes but better than loosing chats :D 08:10 < alish> so now i'm going to try 'tail' once again 08:12 < alish> 'tail' doesn't stop. It's scrolling to end continuously 08:14 < pingfloyd> alish: alias | grep tail 08:15 < alish> ok 08:18 < alish> pingfloyd, how does it works? 08:19 < alish> nothign shown 08:29 < Dagmar> Sounds like some fool aliased it to `tail -f` tho 08:29 < Dagmar> ...unless you're tailing some special device 08:30 < Dagmar> ...or this file doesn't actually contain _newlines_ and tail is making a guess about the end of the file and showing you the last fine 10-zillion-character long "lines" 08:30 < Dagmar> s/fine/five/ 08:34 < alish> Dagmar, I'm going to see some log that is large and continuously appended. 08:35 < alish> now I came to less +90p logfile 08:35 < Rembo> which syntax is correct with "," or without "," https://hastebin.com/iqoculuzix.bash ? 08:35 < alish> but seems it's not end of file 08:46 < alish> 'less +G logfile' did the work. about 1 minute for 18M file 08:48 < alish> but it's very slow 08:49 < alish> when it reach to EOF and I change lines, it take about 2~6 seconds to change affects. 08:49 < alish> *for each line change 08:57 < alish> neither of ways gave me best result. I went for live check with 'journalctl -u myservice', I think it's best solution for now. 09:02 < pa2_st> is it possible to balance between packages that are not too old and not completely latest? Like running GNOME 3.26, vim 8.0, Linux 4.15 with backported security model like Debian 09:03 < S_Gautam> does the official debian DVD come with a desktop env? 09:03 < S_Gautam> https://cdimage.debian.org/debian-cd/current/amd64/iso-dvd/ <- The first one specifically 09:08 < mcdnl> you can choose desktop task when installing 09:16 < kremator> guys is somewhat possible to compile my own kernel with only the stuff i wanted enabled (like in gentoo) but fopr debian based distros? 09:25 < genewitch> is there a way to safely segfault? throw a segfault error arbitrarily in the code 09:29 < GNU\colossus> genewitch, send yourself a SIGSEGV 09:29 < GNU\colossus> (via kill(2)) 09:30 < toffe> cd 09:30 < genewitch> GNU\colossus: thanks looking 09:31 < post-factum> genewitch: make a sigsegv handler. you can intercept this signal and do what you want 09:35 < Rembo> on Ubuntu 14.04 wich is the correct syntax with "," or without "," between email addresses: https://hastebin.com/olekiduzut.bash , i want to send emails to multiples recipients 09:49 < sauvin> Rembo, that depends on your email client, I should think, and not your OS. 09:49 < genewitch> heh, i closed my screen session 09:49 < genewitch> GNU\colossus: works, thanks 09:53 < winsoff_> how do I find the address of the dhcp server that gave me my ip address in linux? 09:54 < mophed> does anyone have linux installed on a macbook pro? 09:55 < MrElendig> mophed: no, not a single person on the planet has that 09:56 < mophed> MrElendig: yeah. that is clearly what i meant 09:56 < winsoff_> MrElendig: get fucking rekt lol 09:57 < MrElendig> mophed: /msg phrik howtoask 09:57 < mous> winsoff_: ip route 09:57 < sauvin> winsoff_, mind the language. 09:57 < mous> winsoff_: that command should show you what you need to know 09:58 < mous> winsoff_: you'll see DEFAULT via (LAN IP ADDRESS) 09:58 < mous> that's the router/gateway that assigned you an ip 09:58 < nevodka> does anyone know of a vnc server application for android 09:59 < nevodka> google is failing me hard 10:00 < nothos> nevodka I remember using one like 3/4 years ago, but can't for the life of me remember its name 10:00 < nothos> All I can say is it needed root and was awful :) 10:00 < winsoff_> MrElendig: that's not a howtoask problem. That's a "you're a pedantic dumbass on freenode making pedantic dumbass quips to no one's benefit" problem. 10:01 < winsoff_> I know what you're trying to say, by the way, because I'm not a pedantic dumbass, but you're still wrong. What you did doesn't highlight the howtoask response sequence. 10:01 < winsoff_> mous: Ah, thanks, comrade. 10:01 < mous> winsoff_: no problem 10:02 < winsoff_> i really need to learn the ip command 10:02 < winsoff_> instead of using ifconfig 10:02 < MrElendig> winsoff_: no, it is a how to ask 10:02 < winsoff_> MrElendig: I told you, I know what you're trying to say. 10:02 < MrElendig> winsoff_: it is a useless question with only one known answer 10:02 < winsoff_> No, TO YOU it is 10:02 < winsoff_> But to someone who DOES run it in this channel who's active 10:02 < MrElendig> no, to everyone it is 10:02 < winsoff_> THEY have a different answer: their answer would be "yeah" 10:02 < sauvin> winsoff_, MrElendig, get a room. 10:02 < winsoff_> this isn't a service ticket line 10:02 < winsoff_> agreed sauvin 10:02 < MrElendig> says the one who started it 10:03 < winsoff_> i'm actually surprised you're active this late in the day, sauvin 10:03 < winsoff_> do you happen to have a "winsoff is being confrontational" highlight? 10:03 < sauvin> Never know when I'll pop up. :D 10:03 < winsoff_> lol 10:05 < winsoff_> sauvin: what's your favorite distro 10:09 < zamba> is it possible to undelete files on ext4 filesystem? 10:11 < winsoff_> zamba: how were they deleted? 10:11 < winsoff_> use testdisk 10:11 < winsoff_> er, photorec 10:12 < zamba> the problem is that this is the root filesystem 10:12 < V7> Hey all 10:12 < winsoff_> zamba: i'm not certain if that matters, unless i'm missing something here 10:13 < winsoff_> v7: be careful with that kind of friendly banter 10:13 < winsoff_> you might be crucified for not asking a question 10:16 < MrElendig> zamba: short story: no 10:16 < V7> winsoff_: If you're asking :) How to install Porteus properly ? I mean, I've booted from porteus CD; have already been copied both folders to an empty GPT partition with 10G space and change pwd to boot folder on it. After issuing "Porteus-installer-for-Linux.com" it gives ".porteus_installer/installer.sh: line 59: cannot create temp file for here-document: No space left on device". 10:16 < MrElendig> zamba: long story, have fun sifting trough a lot of data trying to find the right bits that may or may not be there anymore depending on the fs flags, time passed and the hardware 10:17 < MrElendig> zamba: what did you delete? 10:17 < BCMM> zamba: yes, if you're lucky 10:18 < BCMM> zamba: i.e. there is no intentional undelete feature in ext4, but you might be able to catch it before it gets overwritten. unmouin 10:18 < bartmon> V7, i don't know what porteus is but "No space left on device" seems pretty obvious 10:18 < BCMM> zamba: (sorry, pressed enter) umount the filesystem now if possible. then look up extundelete. 10:18 < zamba> BCMM: it's the root fs 10:18 < MrElendig> porteus sounds horrible 10:18 < zamba> BCMM: so i'd rather not umount it :) 10:19 < MrElendig> zamba: you could boot the install image 10:19 < bartmon> remount read only then 10:19 < MrElendig> zamba: and as asked, what did you delete? 10:19 < MrElendig> and what hardware is this? 10:19 < BCMM> zamba: well, depends how much the deleted file matters to you. work from a live distro if it does... 10:19 < MrElendig> if it is an ssd.... 10:19 < MrElendig> specially if discard is enabled 10:19 < V7> As they're telling how to install it: http://www.porteus.org/tutorials/37-installing/114-official-porteus-installation-guide-v-10.html#USB-linux 10:19 < zamba> MrElendig: some images from a video surveillance system 10:19 < winsoff_> boot to live 10:19 < winsoff_> and then unmount 10:19 < BCMM> zamba: the space occupied by a deleted file is available for use, and it'd be silly if installing extundelete was what caused your data to be overwritten 10:20 < MrElendig> zamba: restore from your backups then, or send it off to some professional if this is something you are legally required to save 10:20 < winsoff_> zamba: I still recommend photorec. 10:20 < BCMM> winsoff_: are you assuming that mount status persists across reboot, or that a good live distro would automatically mount the disk? 10:20 < BCMM> (when you say to unmount after "boot to live") 10:22 < BCMM> zamba: oh that's a very good point from MrElendig about SSDs. if the region that used to store that file has been DISCARDed then, at a hardware level, the device is not required to retain it (but it still might) 10:23 < MrElendig> really, if this is important data, make a copy using dd and then send the disk off to some professional ship 10:23 < MrElendig> shop* 10:23 < BCMM> zamba: anyway, go to http://extundelete.sourceforge.net/ and read "What to do if you've deleted a file (or multiple files)" 10:23 < BCMM> it will explain how continued use of the system puts your deleted data at additional risk 10:23 < MrElendig> much easier for them to help before you break it even more 10:23 < MrElendig> :) 10:23 < zamba> BCMM: i understand why it does that 10:25 < winsoff_> can linux change into in-ram-mode and take its main disk offline without reboot? 10:25 < zamba> winsoff_: the problem with photorec is that it just restores random files with no context whatsoever 10:25 < zamba> winsoff_: i need to be a bit more targetted than that 10:26 < dw1> how do i stuff a window name change after creating it with e.g. screen -S example -X screen 2 10:26 < winsoff_> zamba: it's true, but you can at least find it through .jpg or .png or .mp4 if they're videos, etc 10:26 < BCMM> winsoff_: photorec is great if you're going to literally just look at data on a disk. but if you can get some additional clues from filesystem structure, then why ignore that? 10:26 < dw1> i want to rename window 2 within screen 10:26 < zamba> photorec is actually not that great.. especially not on ntfs filesystems 10:26 < zamba> but that's a different story 10:29 < winsoff_> BCMM: it's true, but it's just habit for me, now. 10:29 < winsoff_> zamba: how so? works for me 10:31 < dw1> i got it.. screen -S example -p 2 -X title test 10:32 < BCMM> winsoff_: also zamba stated it's a video surveillance system. odds are good that there are a *very* large number of video or image files on there 10:33 < zamba> yup 10:33 < BCMM> so photorec's scattergun approach is going to be more painful than usual 10:33 < zamba> this is zoneminder, so there's a bunch of jpegs there 10:33 < BCMM> zamba: so are you able to power down the system at all? 10:34 < zamba> BCMM: not at the moment, no 10:34 < BCMM> (but before you do, is there any chance the file is still open, as suggested on http://extundelete.sourceforge.net/) 10:34 < zamba> BCMM: i tried running extundelete, both yesterday and today.. and i wasn't able to get any of the files 10:34 < zamba> but that was on a mounted fs 10:34 < BCMM> i have no idea whether it works on mounted filesystems 10:34 < zamba> it stated that it found some recoverable inodes, but it wasn't unable to recover them 10:35 < BCMM> i mean, it isn't *supposed* to, but i don't know if it does anyway 10:35 < winsoff_> I was looking into zoneminder. How many cams do you have, zamba? 10:35 < zamba> could it be possible to just do a 'dd' of the entire disk? 10:35 < zamba> winsoff_: 4 10:35 < winsoff_> what models 10:35 < zamba> winsoff_: but i have been running with lots more 10:35 < zamba> winsoff_: different.. d-link, axis, sony, ubiquiti, wanscam, foscam ++ 10:36 < zamba> BCMM: if i create an image of the disk to a separate disk? 10:36 < zamba> that could be done with the fs still mounted, right? 10:37 < BCMM> zamba: it wouldn't be a *consistent* image, because there will probably be writes while you're doing this 10:37 < zamba> BCMM: well.. that's not that important 10:38 < BCMM> zamba: i take it that this machine is constantly creating files? 10:38 < zamba> BCMM: true 10:38 < BCMM> unless you have a *lot* of free space on the disk it's very likely your target has long since been overwritten, at least partially 10:39 < winsoff_> zamba: they all work with zoneminder? 10:39 < zamba> BCMM: well, there's 120 GB available space on the disk 10:39 < zamba> BCMM: out of a total of 455 GB 10:39 < zamba> winsoff_: yup 10:40 < winsoff_> How many days of footage is that? I've been curious about getting a low-cost setup for some paranoid family members, but eh 10:40 < winsoff_> Seems like it's $300 no matter how you swing it 10:41 < zamba> winsoff_: i'm only doing constant motion detection on one camera.. and then i'm doing motion detection on one camera by demand.. meaning when the house is empty 10:41 < winsoff_> how does the on-demand work? I've been thinking about that 10:41 < zamba> winsoff_: the zm system is set to purge old data whenever disk threshold reaches a certain poit 10:41 < winsoff_> i wish more software i liked was written in rust. 10:42 < zamba> winsoff_: i'm using home-assistant to trigger the camera function to motion detect when noone's at home 10:42 < zamba> winsoff_: previously i just parsed the logs for the different access points looking for assoc and disassoc messages, but now i do all this from the automation in home-assistant 10:44 < winsoff_> oh shit, that's clever 10:44 < winsoff_> i didn't even _think_ about that 10:45 < winsoff_> Also, I just remembered that I'm not supposed to swear in here. No one tell james. 10:45 < winsoff_> zamba: is home-assistant on a pi, or what is your central node 10:46 < escalion> winsoff_ - busted 10:47 < winsoff_> forgive me opers for i have sinned 10:48 < zamba> winsoff_: i'm running h-a on a pi, yes 10:48 < zamba> winsoff_: hassbian 10:48 < winsoff_> it's insane how many tools are openly available these days 10:49 < m2_teknix> I have my laptop connected to internet through wieless network as well as ethernet. I want the OS to route certaing website requests through the wireless and everything else through ethernet. 10:50 < zamba> winsoff_: yeah, i love h-a 10:50 < winsoff_> have you checked out/considered shinobi, by the way? i see that zoneminder seems to be a kind-of unwieldy beast. 10:51 < zamba> winsoff_: yeah, i've flirted a bit with shinobi, but even though they brag about all the advantages over zoneminder, i still find it lacking 10:51 < zamba> zm is, as you say, a piece of crap software.. it's from the 90s and it really, really need a rewrite 10:51 < oxagast> cool, i just finished writing a neural network in C and taught it to XOR values 10:51 < zamba> needs* 10:51 < winsoff_> lacking in what? that's super disappointing. 10:52 < winsoff_> Does zm just poll rstp? 10:52 < zamba> i found it non-intuitive to use.. and also very js-based, so it quickly drags down your browser 10:52 < zamba> winsoff_: or http 10:52 < winsoff_> oh, right, the js part 10:53 < zamba> but, i haven't really used shinobi that much 10:54 < zamba> oh, and yeah, i didn't like the installation process at all.. all that new node.js stuff 10:54 < V7> So, anyone :) Is there any possible wayto install Porteus when booted fom CD ? 10:54 < badboyjer> boom boom in the bed room 10:54 < winsoff_> like i said, we need more rust applications. rust is JS-level simple (except for borrowing ;D) with C++-level performance. 10:55 < FreeFull> I wouldn't call Rust's type system simple, but it's good 10:55 < FreeFull> winsoff_: Best way to have more applications written in Rust is to write them ;) 10:58 < MrElendig> js is more wtf than simple 10:58 < pulsar12> I have a bash interactive script that is used as "jump-to-host" tool to other hosts. It calls ssh as child to connect to other hosts. However i have a problem, when the user disconnects abruptly, the script process receives SIGHUP but before processing it, it waits for child which however never ends. the result is a stuck process that must be cleaned up manually with SIGKILL or the ssh connection is broken 10:59 < pulsar12> is possible to convert sighup to sigkill for example? How to force immediate exit upon receiving SIGHUP? 11:00 < ksk> m2_teknix: you can setup routes for specific targets. not too handy for many targets though- (like ip r a 100.100.100.100 via 10.0.0.1) 11:00 < ksk> (add some dev ethX, too, to that command) 11:05 < pulsar12> i think i found the solution: "shopt -s huponexit" 11:07 < peetaur2> If the huponexit shell option has been set with shopt, bash sends a SIGHUP to all jobs when an interactive login shell exits. 11:07 < peetaur2> isn't that what it's supposed to already do? 11:09 < pulsar12> peetaur2, i didnt know about this option before. by default its set to off 11:09 < reg5> why do we create swap partition using fdisk or why do we create file using swap 11:09 < reg5> what does the swap partition does 11:10 < winsoff_> FreeFull: only if you work on it with me! 11:11 < pulsar12> peetaur2, in addition to that behavior: "Use the huponexit option for killing all jobs upon receiving a SIGHUP signal, using the shopt built-in." 11:11 < pulsar12> so not only when it exits, but also when receives SIGHUP which is my case. 11:12 < badboyjer> i really do updates on windows 10 but how come it deletes the / partition and /boot partition 11:12 < FreeFull> winsoff_: I write my software in rust, but the software is more aimed at myself than at the general public =P 11:12 < peetaur2> reg5: swap is how you tell your computer that you are too cheap to buy enough RAM 11:13 < peetaur2> reg5: swap is where it takes things from RAM and stuffs it on disk when you are low on RAM...and sometimes it does it way ahead of time so you get better performance at the time you need it (only discard from RAM, not write to disk before discard)...which means lots more IO many times when you never actually need that RAM. It's a huge performance waste if you don't need it. 11:15 < peetaur2> pulsar12: but it says it sends SIGHUP to all (which I thought it should already), rather than KILL/TERM 11:15 < peetaur2> test and see what it actually sends 11:15 < peetaur2> I expect if you test withotu the option, you would see SIGHUP, and with that option (assuming docs are wrong and the opt actually does something), you'd get SIGTERM 11:16 < peetaur2> badboyjer: it removes the partition, corrupts it, or just the bootloader? old windows versions would only overwrite the bootloader 11:17 < dgurney> badboyjer, I've only heard of issues with /boot, but /? I've never heard of or seen such thing happen 11:17 < peetaur2> I've seen windows corrupt unrecognized filesystems (including non-linux...like for example take a 3TB NTFS disk and put it in a dock that supports up to 2TB and it will corrupt it even if you don't use the disk, even if it doesn't mount) but not actually delete them purposely. But I have seen the winxp installer wipe out the whole partition table if it crashes. 11:18 < badboyjer> i believe it overwrites the bootlaoder 11:19 < badboyjer> bootloader 11:20 < dgurney> peetaur2, heh, the old i386 install system was garbage in many other ways as well, so I'm not even skeptical of that claim 11:20 < badboyjer> thats why i use rescutex and restore the grub option 11:20 < peetaur2> so if you ask M$, I'm sure they'll say slogans like "We love linux" and say it is because you configured (were forced to accept auto config) it to update the bootloader and assume it's on the whole disk 11:20 < peetaur2> but in reality, they don't give a crap about anything you use other than windows, and they likely know what they are doing is wrong and prefer it that way 11:21 < peetaur2> linux will do the same if you configure it to "install to MBR" (where MBR is a misnomer on GPT) 11:21 < peetaur2> but the difference is that the config is transparent, commonly known how to use and configure, actually is in the installer UI, etc. 11:28 < tempate> Hello. I'm trying to install debian on a new system but it's not detecting the ethernet card for whatever reason. "No Ethernet card was detected. If you know the name of the driver needed by your Ethernet card, you can select it from the list". The thing is ethernet is connected. What should I do? 11:28 < Electroid> open a terminal and type lspci 11:28 < Electroid> then search the device there, when you found it, just google linux drivers for that one 11:30 < tempate> I don't have a terminal, or at least I don't know how to access it 11:30 < rascul> ctrl+alt+f1 11:30 < pulsar12> peetaur2, it sends SIGHUP to all jobs regardless of "huponexit" flag. I happens that the original problem of the script is it was trapping the SIGHUP signal, therefore not forwarding it to the child 11:30 < dgurney> tempate, try pressing ctrl 11:30 < dgurney> +alt+f2 11:32 < tempate> dgurney, it worked 11:33 < tempate> dgurney, how do I go back? 11:33 < iceb0x> ctrl+alt+f7 usually I think 11:34 < tempate> nope 11:34 < iceb0x> try all of them then 11:34 < rascul> what is the ethernet device? 11:35 < tempate> ctrl+alt+f1 11:35 < badboyjer> eth0 11:35 < badboyjer> eth1 11:35 < badboyjer> check ifconfig 11:35 < badboyjer> type ifconfig 11:35 < tempate> Ethernet controller: Intel corporation device 15bc (rev 10) 11:35 < rascul> seems like something debian installer should recognize out of the box 11:36 < tempate> badboyjer: ifconfig not found 11:36 < badboyjer> sudo ifconfig 11:36 < badboyjer> try that 11:36 < badboyjer> or su - 11:36 < tempate> I did 11:36 < badboyjer> enter root password 11:36 < pulsar12> peetaur2, so whats the purpose of "huponexit" ? 11:36 < tempate> there is no password 11:36 < tempate> and it's not asking for one either: "sudo: not found" "su: not found" 11:37 < badboyjer> whats wrong with your systekm thenn 11:37 < rcf> badboyjer: it's the debian installer, I think. 11:37 < tempate> ^ 11:37 < badboyjer> hmm thats weird 11:37 < tempate> it's not recognizing the ethernet card for whatever reason 11:38 < MrElendig> in the installer you have root power to begin with so.... 11:38 < tempate> what's the right driver for the ethernet controller: "Intel corporation device 15bc (rev 10)" 11:38 < MrElendig> also, ifconfig was deprecated a decade and a half ago, use ip 11:38 < MrElendig> tempate: more info needed 11:38 < rcf> tempate: Did it complain about needing firmware earlier in the install? 11:39 < MrElendig> like the vid/pid 11:39 < tempate> rcf: nope 11:39 < MrElendig> and the kernel version 11:39 < tempate> MrElendig: how do I find those out? 11:39 < MrElendig> lspci -vnn 11:40 < MrElendig> probably needs firmware-misc-nonfree 11:41 < MrElendig> or might anyway 11:41 < tempate> a lot of stuff came out but I can't find what you asked for, MrElendig 11:41 < peetaur2> pulsar12: have you found a practical difference via testing? 11:41 < MrElendig> tempate: it is the bit in (:) 11:42 < pulsar12> peetaur2, no. so i wonder what is the purpose of it, maybe missing documentation 11:42 < MrElendig> you could try the unofficial non-free install image 11:42 < tempate> stupid question: how can I scroll up? 11:42 < rcf> MrElendig: as I recall it usually asks for firmware it doesn't include if that's the issue, rather than just not being able to detect the device. 11:42 < MrElendig> or buster 11:42 < pulsar12> peetaur2, but my initial conclusion was wrong about my problem 11:43 < MrElendig> rcf: he hasn't posted the full dmesg so 11:43 < badboyjer> what do you think of the nick crack baby cock blocker 11:43 < rcf> True. 11:43 < MrElendig> nor kernel version 11:43 < tempate> I can't find it! 11:43 < MrElendig> pgup 11:43 < tempate> I mean, I'm just being able to see part of the output, and as I can't scroll up I'm kind of stuck 11:43 < MrElendig> pgup 11:43 < tempate> not found 11:43 < MrElendig> >_> 11:44 < rcf> tempate: kernel version: uname -r 11:44 < MrElendig> tempate: https://www.computerhope.com/cdn/keyboard/page-up.jpg 11:44 < spear2> is there a way to make light colored text in an xterm appear bold? or is it something that can be configured in vim? 11:44 < tempate> 4.9.0-6-amd64 11:44 < MrElendig> shift-pgup even 11:45 < MrElendig> can also pipe to less 11:45 < MrElendig> tempate: dmesg | curl -F c=@- https://ptpb.pw 11:45 < badboyjer> anybody getting fireworks for the 4th of july 11:46 < tempate> oh, that's what pgup meant... 11:47 < tempate> MrElendig, curl not found 11:47 < badboyjer> anybody from the states celebrating the 4th of july 11:47 < tempate> can't I just enter the driver manually? 11:48 < rcf> tempate: actually you won't be able to do that because you don't have a network connection because of this problem. 11:48 < rcf> tempate: the dmesg post, I mean. 11:48 < MrElendig> urgh, debian is so iritating to use 11:48 < tempate> rcf, that's what I was thinking 11:48 < MrElendig> see if it has wget 11:48 < peetaur2> pulsar12: well I don't know what actually sends the normal SIGHUP to the children... tahe process always did it? the compiler added code to do it? the kernel running it did it (likely)? and so maybe this feature is for when the plaform doesn't normally do it, and it's redundant on Linux. 11:48 < tempate> MrElending, yes! 11:48 < rcf> MrElendig: if that's his only NIC, that won't do much good. 11:49 < tempate> how do I find the driver for the controller? 11:49 < MrElendig> that's what a 5ā‚¬ dongle is for 11:49 < MrElendig> or just copy the output to a usb stick and then post it 11:49 < MrElendig> usb stick/phone/whatever he is using to talk to us 11:50 < tempate> alright 11:51 < tempate> I can just give you the info you need though 11:51 < MrElendig> tempate: google says that the buster iso works with your card 11:51 < MrElendig> out of the box 11:52 < tempate> MrElendig, that one is not appearing 11:52 < tempate> how bad is it if I just go with "no ethernet card"? 11:53 < MrElendig> depends on what you want to use the machine for 11:53 < badboyjer> well you can always rely on wireless card 11:53 < tempate> that was a stupid question as I can only connect through ethernet... 11:53 < badboyjer> if you have one 11:56 < rcf> tempate: if you have the full install media you can go ahead with the installation now and fix the problem later on, though you might be better off finding a distribution that actually supports your hardware from the start. 11:56 < tempate> rcf: does that mean I'm not going to be able to run debian on my machine? 11:57 < tempate> I mean, I'm going to be back to the same problem sooner or later if I just pass, aren't I? 11:58 < rcf> Unfortunately yes. Though MrElendig said buster supposedly works fine on your hardware, so if you can put up with testing, that would be the way to go. 11:58 < tempate> rcf, the thing is I'm not finding "buster" under the possible drivers 11:58 < rcf> tempate: No, that's the debian version, not the driver. 11:59 < rcf> https://www.debian.org/devel/debian-installer/ 12:00 < Xatenev> how do i access my smartphone in ubuntu via mtp 12:00 < Xatenev> lsusb lists my device 12:00 < rcf> Whether you want to put up with the potential instability (compared to standard debian) is up to you. 12:01 < MrElendig> Xatenev: gvfs-mount 12:01 < MrElendig> Xatenev: or use the fancy file manager 12:01 < dgurney> as someone who used debian testing for many years, I never noticed any stability problems 12:01 < Xatenev> MrElendig, how 12:01 < tempate> rcf, alright, I think I'll just go with a different distro 12:01 < dgurney> I can't imagine that has changed much since I did so 12:01 < Xatenev> MrElendig, It seems i can access it via something like mtp://[usb:someid,someid]/ 12:02 < Xatenev> MrElendig, no idea where to get the someid,someid tho 12:02 < Xatenev> its obv not the one in lsusb, atleast it doesnt work when i tried it 12:02 < MrElendig> the gvfs tools will tell you 12:02 < MrElendig> to will lsusb 12:03 < MrElendig> gvfs-mount -li | grep -e ^Volume -e activation_root 12:03 < tempate> fedora... here we go 12:03 < dgurney> tempate: great choice 12:03 < MrElendig> gvfs-mtp must be installed, but it most likely is 12:03 < Xatenev> MrElendig, yay works thanks 12:03 < tempate> I was already used to apt-get 12:04 < Xatenev> MrElendig, why did i find 20 threads explaining weird things by editing weird files somewhere 12:04 < Xatenev> and none that states "Use gvfs-mount to get your ID and access it via a file manager" 12:04 < Xatenev> or similar 12:04 < badboyjer> apt works just fine 12:04 < Xatenev> lmao 12:05 < MrElendig> Xatenev: if you use a fancy file manager then you can just click on the icon 12:09 < tempate> would deepin destop environment work on Fedora? 12:11 < dgurney> tempate: possibly, I imagine #fedora would be more helpful with fedora-specific questions 12:11 < tempate> you are right, apologies 12:15 < fire2199> my distro is including gnome-keyring but there is no interface to manage passwords, also i just saw that gpg-agent can suffice for nearly everything, so what's the advantage of using gnome-keyring? 12:17 < revel> fire2199: I think you can manage gnome-keyring with seahorse. 12:19 < fire2199> revel : i know seahorse but wouldn't it be better to get rid of gnome stuff and only rely on gpg and its agent? 12:19 < revel> That's an option as well. 12:19 < fire2199> would gnome-keyring work in a pure text mode session? 12:19 < revel> ĀÆ\_(惄)_/ĀÆ 12:21 < fire2199> revel : is that a CJK character in the middle? 12:21 < revel> Yes. 12:22 < BCMM> it's a tsu katakana 12:28 < djph> fire2199: not really, no. 12:28 < RustyJ> grrrrr Install VPS, setup apache/mysql/php, create and TFA two users, install wordpress and deploy existing data from WP windows.... charge the client 5 hours.... they think i'm ridiculous and it should of taken half as long. 12:29 < djph> no pay = no keys to the kingdom. 12:30 < RustyJ> djph, directed to me? 12:30 < djph> RustyJ: yeah, sorry 12:31 < RustyJ> djph, oh no doubt! just annoying.... thats low/fair for that much work... I have everything TFA and i'm sitting on the time codes till my check clears for sure 12:32 < buoyantair> How do I get read and write access to a directory like /data/db? 12:32 < buoyantair> (I dont want to sudo every time I open mongodb...) 12:32 < djph> TFA? 12:32 < buoyantair> what is that? 12:32 < djph> not at you buoyantair 12:33 < djph> your problem is "give ownership of the db directory to mongo" 12:33 < buoyantair> uh 12:33 < buoyantair> I guess? o_O I thought I need to have ownership over /data/db folder 12:34 < djph> "you" as in "your user"? 12:34 < RustyJ> djph, two factor auth.... 12:34 < djph> RustyJ: ah, I've always seen that as 2FA 12:34 < RustyJ> ok i can do that too 2FA 12:34 < RustyJ> lol 12:36 < djph> RustyJ: it's all good :) 12:38 < buoyantair> djph: No, its just me 12:39 < djph> buoyantair: that makes no sense. 12:39 < buoyantair> >_> 12:39 < djph> buoyantair: wait, did you mean that mongo runs as "your user"? 12:40 < buoyantair> Nooo I just installed mongodb 12:40 < buoyantair> but to start it, I need to sudo because only root has access to /data/db 12:40 < buoyantair> and not my user account 12:41 < djph> so then change the ownership 12:41 < buoyantair> how do I do that? 12:41 < buoyantair> Also, if i changed ownership, will root be able to access it again? 12:41 < buoyantair> o.o 12:43 < u0m3_> hi. anyone ever encountered an issue with ppp0 devices (when used with mobile 3G/4G modems) somehow being dropped although iptables is opened/not used? 12:49 < kbob> u0m3_: maybe that have nothing to do with iptables 12:50 < u0m3_> kbob, tcpdump shows packets incoming... but for some reason none come out. But I can connect through it. 12:53 < kbob> u0m3_: and the connection drop down sometimes? 12:54 < u0m3_> kbob, no, doesn't look like it drops 12:55 < BluesKaj> Hi all 12:55 < kbob> ok, try to install bmon so you can monitor the in and out traffic 12:55 < kbob> hello BluesKaj 12:56 < BluesKaj> hi kbob 12:56 < u0m3_> kbob, the only other (and kind of crazy) ideea I had is that, for some strange reason, packets come in ppp0 and go out eth0, but I didn't check that. 12:56 < u0m3_> kbob, ok, will do and get bak 12:56 < u0m3_> s/bak/back/ 12:56 < kbob> ok 13:13 < VlanX> Hello ppl, in your opinion, what could cause these two server to behave differently with sort -n like in this example? https://hastebin.com/codocijumu.bash 13:14 < djph> locale settings, most likely 13:15 < toffe> VlanX: check `locale` 13:15 < VlanX> djph: so more locale than encoding, right? 13:16 < VlanX> toffe: LANG=en_US.UTF-8 vs LANG=en_US 13:16 < toffe> VlanX: also, are there a whitespace on your server 1 on bototm of the file? unknown character? 13:17 < toffe> probablt that, change to utf-8 on second one 13:17 < V7> Hey all 13:17 < djph> yeah, some locales order lowercase before uppercase (IIRC en_US.UTF-8, C, others), others don't care about case (offhand, don't know any) 13:17 < djph> err *order uppercase before lowercase 13:17 < V7> Waht would be with files if I'd resized a partition with data in GParted with free *preceding* MiBs ? 13:18 < VlanX> toffe: not sure about that. however this problem most likely caused a few million euros worth of damage (they were supposed to name the file with a leading number, not to rely on uppercase/lowercase text) 13:18 < djph> same as if you did it with free space at the end - stuff gets moved. 13:18 < djph> sucks for them. 13:18 < V7> djph: So, https://i.imgur.com/lzabO0V.png no files will be removed ? 13:19 < djph> V7: well, that's the hope anyway. Never not backup. 13:19 < VlanX> oh well ... enough for today :( 13:19 < bartmon> V7, the whole filesystem is copied however many MiBs forward on the partition 13:20 < gbellinoz> anybody use audacious from webupd8? 13:20 < toffe> VlanX: dangit that sucks, dont relay on software who executes based on locale :P 13:20 < V7> bartmon: You mean, files' data will be moved physically ? 13:20 < bartmon> V7, yes. 13:21 < V7> oh, this is awesome if it'd worked 13:23 < djph> V7: don't forget to backup. There is ALWAYS the possibility of things going wrong. 13:24 < kerframil> VlanX: sort -n is an odd thing to do. anyway, in the en_US.UTF-8 case, it's possible that the (LC_COLLATE) looks like this 'aAbBcCdD...' rather than 'abc...ABC...' 13:24 < kerframil> the collation, that is 13:28 < Dagmar> It's not an odd thing to do 13:28 < ice9> how to change the default files manager in Deepin desktop env? 13:28 < Dagmar> It's perfectly reasonable when you don't want a column sorted according to daft left-justified numbers, done entirely wrong 13:32 < kerframil> only that's not what he's doing, but thanks for the non-sequitur 13:34 < V7> uh, ofc, but the backup will consume about ~150GB, so about 5,4 hours and then move ~300GB will costs about 10,3 hours ( 16,7 hours ) 13:36 < djph> V7: 5 hours to move 150G? what're you doing, backing up to 5.25" diskettes? 13:36 < djph> *disks 13:36 < V7> The speed is 28MB/s 13:36 < pingfloyd> would be quicker if you used rsync since it just need to copy changed files 13:36 < V7> pingfloyd: Tar was used 13:37 < pingfloyd> tar is horrible for backups 13:37 < djph> 28 MB/s on a LAN? great lord cthulhu, replace that stuff... 13:37 < pingfloyd> if you need to restore a single file, it has to seek like crazy 13:37 < toffe> We're doing the same thing, using tar for backup, we upload to cloud though 13:38 < toffe> so posting trough https request 13:38 < pingfloyd> tar makes your backup like being on tape without being on tape and missing out on random access to it. 13:38 < toffe> luckily our backups are max 1GB 13:38 < toffe> normally around 100-150MB 13:39 < pingfloyd> linear format on random access media 13:42 < V7> djph: Why LAN, this is USB HDD 13:42 < dbe> If I register an event on an epoll fd, say using but I set epoll data to be, say, "&(struct mystruct){ .val = 3 }"; then... what happens? I guess that structure created on the stack and removed? Will epoll_wait be able to read that structure? 13:43 < V7> So disk's speed is about 28MB/s 13:47 < post-factum> dbe: i don't get what your question is about 13:47 < djph> V7: that's even worse 13:47 < djph> USB should be faster than that 13:49 < toffe> git push 13:51 < dbe> post-factum: me neither. But I think I got it anyway. Thanks though. 13:52 < tomeaton17> ! [remote rejected] master -> master 13:52 < post-factum> This incident will be reported. 13:52 < dimm> hello, All! Is blockchain good choice for save documents into shared storage? 13:52 < dimm> Like google docs but without central storage 13:52 < post-factum> dimm: hi. this is a channel about linux 13:53 < g1itch> I'm not sure if this is related, but I updated my password on my system and had to clear my gnome keyring to prevent the password notice from popping up, but after this, chrome doesn't seem to keep me logged in to websites like it used to. any thoughts? 13:59 < djph> cleared your cookies? 14:00 < zlinux> dimm: what are you doing? 14:02 < mavorsa> does anyone here use newsboat? i downloaded it with snap, but although ive defined my browser in the config file, it doesn't seem to want to work. 14:08 < kbob> browser "browser %u" 14:09 < pankaj_> Sometimes when I am trying to read the output the valuable contents of output are in the upper part and unfortunately the tty gets filled by lot of errors that I am unable to see that valuable part and so I cannot solve the problem. I tried 'less' and outputting the ouput of the error or warning to the file but the errors do not get listed. What to do? 14:14 < mavorsa> kbob: is not working, whether its that, "firefox", "firefox %u" or anything. there are two different config locations for some reason. one in ~/.newsbeuter, and one in ~/snap/newsboat/295/.newsboat. no matter which i edit, no change 14:14 < mavorsa> its almost as if its not recognizing or using the links in any of the feeds 14:17 < kbob> newsbeuter is the old one you must use .newsboat, maybe newsboat have to convert the .newsbeuter in .newsboat backup and try deleting one 14:18 < mavorsa> kbob: i'm inclined to think its meant to be the snap directory as that is its installation and also where the url list file is 14:23 < mavorsa> kbob: ive removed the .newsbeuter folder, no change. then i made a .newsboat directory and copied the snap version over into it, also no change 14:27 < mavorsa> i'm beginning to think its something wrong with this snap version 14:27 < kbob> here newsboat is working fine 14:30 < mavorsa> did you install yours through snap or directly from github? 14:32 < TRS-80> hmm, just installed hg, where is a good location to clone something to? 14:33 * TRS-80 goes off to consult Linux file system documentation 14:33 < TRS-80> or am I overthinking this? just put somewhere in ~? 14:33 < socomm> hello 14:33 < TRS-80> hi socomm 14:33 < post-factum> TRS-80: ~/devel, for instance 14:34 < post-factum> socomm: hi, who are you? 14:34 < TRS-80> post-factum: I'm not devel just using some software where cloning is recommended installation method (it's a Python package, Beancount, specifically) 14:34 < post-factum> TRS-80: what software is that? 14:35 < pingfloyd> does it count beans? 14:35 < TRS-80> text based accounting, like ledger sort of if you've heard of thatt 14:35 < TRS-80> pingfloyd: winrar :) 14:35 < solidfox> woah 14:35 < Pentode> mercurial? 14:35 < solidfox> I've been looking for smth like that 14:36 < post-factum> TRS-80: if your distro doesn't provide a package for that, package it on your own 14:36 < TRS-80> Pentode: main dev is a Googler, docs are actually in Google Docs (blech!) but he's done such a good job with software (and documentation!) that I give him a pass (sort of) 14:37 < Pentode> google docs ugh 14:37 < TRS-80> so I guess what I am saying is that I assume they use hg at Google? I dunno, must be some reason he prefers that... 14:37 < christner> does anyone know what channel I should go to for perf usage questions? 14:38 < lupine> I put all my source code into a hierarchy in ~/dev/ 14:38 < TRS-80> pingfloyd solidfox and others who may be interested: https://plaintextaccounting.org/ 14:39 < post-factum> christner: #perf @ oftc 14:39 < post-factum> not freenode 14:39 < centrix> Can I dynamically link against "*.a" libraries? 14:39 < post-factum> centrix: wut 14:39 < Dagmar> No 14:40 < TRS-80> post-factum: development is pretty active, that's his justification for not creating packages himself, also if you think about who the likely users are of such a software I don't suppose it's a big deal, lol 14:40 < christner> post-factum: thanks 14:42 < centrix> Dagmar, basically I'd have to recompile all *.a libs as shared ones (*.so) and link against those to get what I want. Correct? 14:42 < post-factum> centrix: do you have a src code? 14:43 < pingfloyd> centrix: what do you want? 14:44 < centrix> Basically I have received a project and claims that binaries used to be linked statically. 14:44 < centrix> But I can see a lot of *.a libs there in the Makefile. 14:45 < TRS-80> I just noticed I have a file in ~ called f#@k and I have no recollection of it and contents are not text :/ 14:46 < centrix> Thus I am not certain that my predecessor linked binaries completely dynamically. His claim may be false. 14:46 < centrix> That is why I ask if I can link dynamically against the *.a libs. 14:47 < Pentode> centrix, yeah they are static only. are these libraries part of the source tree itself? 14:47 < post-factum> TRS-80: $ file fuck 14:48 < pingfloyd> sure you didn't make it while drunk? 14:49 < Pentode> post-factum, this is a family channel. :) 14:49 < centrix> Pentode, I am "scanning" the code of tens of little files, but I suppose so. I can see cc command is "linking" against it. 14:50 < post-factum> Pentode: take your children away from screen, the hardcore begins in 3ā€¦ 2ā€¦ 1ā€¦ 14:51 < lilltiger_> If your children cant handle words they shouldent be using the interwebs 14:51 < pingfloyd> ^ 14:52 < solidfox> your kids are swearing all day long at school anyway 14:52 < solidfox> lol 14:52 < post-factum> lol 14:52 < pingfloyd> not their little angels! 14:52 < post-factum> you want to believe 14:53 < pingfloyd> you should be kicking your kids off the computer and throwing them outside anyway 14:53 < solidfox> yes or they'll turn into a recluse like me 14:53 < pingfloyd> boot them out the door and tell them not to come back until dark 14:53 < post-factum> you mean, defenestrate them? 14:53 < lilltiger_> Well how do you think angels procreate ;) 14:54 < pingfloyd> welcome to the human race 14:54 < post-factum> if he/she flies instead of falling down after defenestration, he/she is definitely an angel 14:54 < pingfloyd> any parent trying to stop nature is only fooling themselves 14:57 < pingfloyd> the window separates the wheat from the chaff 14:57 < centrix> function defenstrate (kids) { if [ ! -z ${kids} ] ; then return 19; fi } 14:57 < solidfox> pingfloyd, I'm actually thinking of making myself stay out of my apartment until dark 14:57 < pingfloyd> go electronic free, except phone calls, until dark too. 14:57 < pingfloyd> no sms either 14:58 < pingfloyd> that will get you out of your hermit mode 14:59 < pingfloyd> I'm trying to remember how I survived without mobile phones growing up 15:00 < pingfloyd> closest thing then was payphones 15:00 < pingfloyd> seems like those are extinct and attract addicts 15:00 < AndroidKitKat> So I think I just royally screwed my path, is there anyway to get it back? 15:00 < pingfloyd> AndroidKitKat: depends. How did you royally screw it? 15:01 < AndroidKitKat> I forgot to append :$PATH onto an `export PATH=blah` then sourced my zshrc 15:02 < pingfloyd> fix the .zshrc and exit/start a new shell/term 15:02 < AndroidKitKat> that'll fix it? 15:02 < AndroidKitKat> it did 15:02 < pingfloyd> you've got to append the $PATH in it (fix) 15:02 < AndroidKitKat> haha 15:02 < pingfloyd> or prepend, depending which you want 15:03 < AndroidKitKat> i'll just fix my zshrc 15:03 < pingfloyd> I was going to say, "oh no! Now you have to reinstall!" 15:03 < AndroidKitKat> I think I would get fired for that 15:04 < pingfloyd> sounds like a harsh place to work 15:04 < AndroidKitKat> its not my box 15:04 < toffe> git push 15:04 < AndroidKitKat> and these are special computers or so im told 15:04 < toffe> .. 15:04 < toffe> dangit 15:04 < pingfloyd> magical ones? 15:04 < pingfloyd> sounds like you're working for evil overlords 15:05 < AndroidKitKat> i mea 15:05 < pingfloyd> need to stage a coup 15:05 < pingfloyd> that's the real fix 15:05 < AndroidKitKat> that would be treason 15:05 < pingfloyd> just don't fail 15:05 < pingfloyd> then they're the traitors! 15:06 < AndroidKitKat> lol 15:06 < pingfloyd> when you win, making me your second in command will be gratitude enough. 15:07 < AndroidKitKat> as you wish, m'lord 15:12 < centrix> pingfloyd, ldd /bin/mybin say that it needs "/usr/lib/libcrypt.a(shr_64.o)". So how come the binary lists this one instead of having functions copied into itself? 15:13 < post-factum> centrix: and what `file '/usr/lib/libcrypt.a(shr_64.o)'` says? 15:15 < centrix> /usr/lib/libcrypt.a: archive (big format) 15:16 < post-factum> centrix: are you sure about the name? i mean that (shr_64.o) part (wtf is thisā€¦) 15:16 < TRS-80> post-factum: Aha! Results: f__k: RIFF (little-endian) data, WAVE audio, Microsoft PCM, 16 bit, mono 8000 Hz Thank you, sir! 15:16 < post-factum> TRS-80: now you can play it 15:18 < Psi-Jack> AndroidKitKat: Heh. 15:18 < AndroidKitKat> lol 15:18 < TRS-80> post-factum: it's just someone (me?) saying "hello"... I still have no recollection of this whatsoever. lol Thanks though for helping solve the mystery. :) 15:19 < post-factum> TRS-80: i think this is *the* first contact 15:19 < eagle> hello, how can I install linux on my GNU operating system? 15:19 < post-factum> TRS-80: stay home, we are heading towards you already 15:19 < post-factum> eagle: wut 15:19 < Psi-Jack> eagle: Any major Linux distribution. :p 15:19 < eagle> post-factum: I cam kinda sick of hurd 15:20 < post-factum> eagle: install plan9 15:20 < eagle> Psi-Jack: I am still anoob 15:20 < eagle> post-factum: I am not from outer space 15:20 < post-factum> TRS-80: we've found the source of your voice ^^ 15:20 < Psi-Jack> eagle: Solus, Fedora, CentOS, Debian. There you go. 15:20 < eagle> debilian 15:20 < TRS-80> eagle bait 15:20 < post-factum> eagle: russian? 15:21 < eagle> side note, does anyone know where I can order 1995 vintage rms foot cheese? 15:21 < Psi-Jack> I smell... troll. 15:21 < post-factum> yeah 15:21 < TRS-80> eagles care about freeedom 15:21 < eagle> murica 15:21 < post-factum> eagle: yankee go home 15:22 < TRS-80> eagle hail rms our lord and savior 15:22 < eagle> hail 15:22 < eagle> saint ignucius 15:23 < jim> what did you start?! it's hailing foot cheese! 15:23 * AndroidKitKat praises RMS 15:23 < eagle> he isn't dead yet, jim 15:23 < dieggsy> is there a way to prevent bluetooth soft blocks entirely? (assuming no driver/card faultiness) 15:23 < jim> ... but not as we know it? 15:24 < TRS-80> OK so I'm installing Beancount via pip3 and it's pulling in all this Google crap, me no likey: Collecting google-api-python-client, google-auth, google-auth-httplib2... Oh maybe this is for the built in web server? there is also rsa... 15:25 < TRS-80> I suppose I need to look up what those packages are 15:25 < registeringIsBad> Does anyone here have any experience with xrandr? I'm trying to set up my tv as a second monitor, but it has overscan, with no --prop to fix it 15:25 < registeringIsBad> I have tried to use the transform option, but the tv for some reason also changes its resolution to match the transformation, even when I use the --size command. Part of the reason why it is overscanning is because its resolution is 1366x768, yet it does not list that as a resolution option, and any --addmode passed to it makes the screen just go blank. 15:25 < azarus> there's a reason why linux isn't GPLv3, so I don't think the kernel devs really like the FSF: https://lwn.net/Articles/200422/ 15:25 < TRS-80> registeringIsBad: Did you try and let it auto detect? 15:25 < azarus> registeringIsBad: likely settings on the TV itself 15:26 < registeringIsBad> TRS-80 by just doing xrandr --auto? yes 15:26 < TRS-80> azarus: What are you doing man! Don't open that can of worms! 15:26 < registeringIsBad> azarus there are no tv settings for overscan 15:26 < registeringIsBad> i have checked 15:26 < TRS-80> registeringIsBad: No by just plugging it in and booting up machine. 15:26 < registeringIsBad> TRS-80 yes 15:26 < TRS-80> how are you connecting? 15:26 < registeringIsBad> HDMI 15:26 < azarus> TRS-80: i'm igniting the flames, and nobody can stop me! 15:27 < registeringIsBad> Further information: using the --pos command to try to change the screen location does not change where the tv sources its output 15:27 * TRS-80 Checks expiration date and charge level of fire extinguisher 15:27 < registeringIsBad> only using the --transform command can I actually move the output on the screen 15:28 < registeringIsBad> Is there any way to create an output that has black edges(a poor mans overscan correction?) 15:29 < registeringIsBad> like by using a virtual output and then creating another larger virtual output thats a copy of the other virtual output, and then having the tv copy that second virtual output? 15:29 < BCMM> azarus: the other reason the kernel isn't gplv3 is that it basically can't change its licensing 15:29 < registeringIsBad> Problem is I have no idea how to create a virtual output 15:29 < TRS-80> registeringIsBad: Hmm. Well sounds like you are on the right track. If I can offer any help it's that my experience searching internet trying to get multiple monitors set up led me to a whole lot of outdated information about randr, etc. so be aware of that 15:30 < jprjr> Yeah I think you'd have to get sign-off from every single dev that's ever contributed to the kernel 15:30 < BCMM> it's not under "or later" licensing, and also copyright is usually retained by individual contributors 15:30 < azarus> correct 15:30 < registeringIsBad> TRS-80 what did you eventually do to set up your other monitors? 15:30 < BCMM> so you'd have to track down an impractical number of people and get their permission to relicence the kernel 15:30 < registeringIsBad> TRS-80 or did you have an entirely different issue 15:30 < jprjr> You could probably do an audit and figure out who's code is no longer there but that's really expensive to do 15:31 < BCMM> i mean, if Linus could change it to GPLv3, he clearly wouldn't. but as it happens, he can't anyway. 15:31 < jprjr> registeringIsBad: if it's an HDTV (using HDMI/DVI, not composite), I usually just adjust the overscan settings on the TV itself 15:31 < TRS-80> registeringIsBad: basically I stopped trying to set things manually in files and in randr and let the thing configure itself. With one exception. Some command about sinks or something. Standby lemme see if I can find that. 15:31 < jprjr> registeringIsBad: if you're using a CRT TV I don't have a lot of advice 15:32 < registeringIsBad> jprjr I am using an HDTV, and I am completely sure that there are no overscan settings 15:32 < TRS-80> registeringIsBad: yes that was the key, now if I can only find that bit of info... :/ 15:32 < jprjr> registeringIsBad: it's usually not explicitly labeled that - have you tried looking through any zoom settings? It's usually in there 15:32 < BCMM> registeringIsBad: it typically won't be a separate setting. it'll be under "picture mode" or "aspect ratio" or something 15:32 < registeringIsBad> BCMM I did check those locations 15:33 < BCMM> usually called something pointlessly confusing like dot-for-dot 15:33 < jprjr> registeringIsBad: you'll usually have a few different zoom levels, a widescreen, etc, and the mode with no overscan will be called "native", "dot-for-dot", etc 15:33 < azarus> or "cinematic zoom" 15:33 < jprjr> The long-term fix is to just use a computer monitor instead of a TV 15:34 < registeringIsBad> jprjr I do have an actual primary monitor, this is a secondary monitor to watch tv streams and stuff like that 15:34 < jprjr> Like my TV at home has "natural" (which means tons of overscan) and "native" (which means none) 15:34 < TRS-80> but isn't HDMI basically like DVI? Or am I mixing up something else? 15:34 < BCMM> i don't get why digital displays have an overscan option in the first place... 15:34 < jprjr> Yeah I figured it's a secondary monitor, I'm just saying using TVs as monitors always kinds sucks. The long-term fix is to not use a TV as a secondary monitor, get a proper monitor 15:35 < oneko> hi guys 15:35 < oneko> hello pbek 15:35 < registeringIsBad> jprjr doesnt have the same feel when you want to watch a movie or twitch stream though 15:35 < registeringIsBad> I might get a second monitor though 15:36 < registeringIsBad> once i replace my laptop 15:36 < jprjr> TRS-80: HDMI can carry a DVI signal, I believe. But of course it has a lot more options for carry sound data, I think there's some way to do ethernet over HDMI, etc etc 15:36 < registeringIsBad> 4 GB of RAM, cost $1k, apparently half the hard drives failed within 1 year 15:36 < pbek> hi there oneko 15:36 < registeringIsBad> its a pretty bad laptop 15:36 < registeringIsBad> I would not reccommend Toshiba 15:36 < jprjr> If the monitor is big enough it's about the same feel as a TV 15:37 < jprjr> Better, even 15:37 < erebel55> lo 15:38 < Psi-Jack> Toshiba is one of the few laptop brands I do recommend. 15:38 < TRS-80> registeringIsBad: OK I found the command, but not sure it will help you. IIRC my issue was setting up 20 30 20 PLP with 2 monitors running off main discreet GPU via DVI and then third monitor running off motherboard/chip (i5) built in GPU 15:38 < registeringIsBad> Psi-Jack are their other laptops better? 15:38 < toffe> dangit 15:38 < registeringIsBad> Psi-Jack do they actually work after 1 year 15:38 < TRS-80> registeringIsBad: anyway in my case it was "xrandr --setprovideroutputsource 1 0" 15:39 < Psi-Jack> I've got two Toshiba Satellites that are still running, and they're 2~5 years old. 15:39 < erebel55> I'm trying to create a systemd service that can be instantiated multiple times with multiple config options. I need to pass in a port and also a queryport. https://hastebin.com/mamuvocaju.service 15:39 < TRS-80> which I put in a script and run at every startup 15:39 < erebel55> How can I pass multiple config variables? 15:39 < jprjr> Every brand is going to have a line of trash laptops. My wife bought a pretty garbage Toshiba. It wasn't bad because it was Toshiba though, it was bad because it was like, $200 - new. 15:39 < registeringIsBad> Psi-Jack I have a satellite 15:39 < TRS-80> Original IBM ThinkPad FTW 15:40 < Psi-Jack> When I was choosing a laptop for work, on the other hand, I looked at Toshiba first, and decided to get an HP instead. 15:40 < Psi-Jack> I'm not unhappy with my decision on that either. :) 15:41 < jprjr> If you're buying laptops with 4GB of RAM you're gonna have a bad time. Not because 4GB of RAM isn't enough (depending on your use-case, obvi), but usually the ones that come with 4GB of RAM nowdays tend to have other garbage components 15:41 < jprjr> Exception being Chromebooks, I love my trashy Chromebook. But I also bought it fully knowing it was a weaker, kinda trash machine 15:41 < registeringIsBad> jprjr well this was about 5 years ago, as a Christma present 15:41 < registeringIsBad> jprjr i was not the one who picked out this laptop 15:42 < jprjr> Even then 4GB of RAM wasn't great 15:42 < registeringIsBad> i know 15:42 < jprjr> I'm just getting at like, it's probably not bad because it's a Toshiba, it's probably bad because it was just bad from the get-go 15:43 < registeringIsBad> jprjr so youre saying that although I bought a bad computer from a company, and thus know that the company is capable of making overpriced bad computers, I should definately consider them instead of other brands that I have heard make good computers? 15:44 < jprjr> Well yeah because every company is capable and totally makes overpriced computers 15:45 < jprjr> I'm saying don't write off Toshiba entirely, I'm saying go read some reviews of a particular model you're considering. There's garbage Toshiba Satellites and there's decent Toshiba Satellites 15:45 < jprjr> That'd be like saying "Yeah Toyota makes some really overpriced cars so I shouldn't ever consider buying a Toyota again" 15:46 < erebel55> anyone know of a more sys admin linux channel that my systemd question would fit better in? 15:49 < jprjr> I remember way back when I was in college, like every Dell Laptop was called in Inspiron. There was garbage ones, there were "high-end" overpriced ones, there were decent ones. It really varies from model-to-model and year-to-year. Whenever you're in the market for a laptop, you should find a few models in your price range, read some reviews, see if there's a store to try them out, etc etc. Nobody should ever 15:49 < jprjr> say "oh I love Toshiba, so I'll just buy the Toshiba without a second thought" 15:50 < twainwek> oh I love Toshiba, so I'll just buy the Toshiba without a second thought 15:50 < jprjr> and similarly, "oh one time I had a bad Toshiba, I should never buy Toshiba again." 5 years ago, Windows 10 didn't exist, Ubuntu was still using Unity, I mean things change drastically in a short period of time. 15:50 < registeringIsBad> jprjr I am more of a brand loyalty person. If I have a Dell laptop that lasts forever, I will almost certainly buy a Dell for my next computer, unless it has bad reviews 15:51 < pingfloyd> I'll buy a Toshiba without second thought 15:51 < pingfloyd> they last 15:51 < twainwek> 5 years ago windows 7 existed though 15:51 < pingfloyd> I have one that 9 years old and still running 15:52 < pingfloyd> they should've just kept upgrading win2k the whole time instead 15:52 < BCMM> registeringIsBad: thing is, brand loyalty doesn't really work in this day and age, for very large companies 15:53 < BCMM> if their brand name acquires some respect, they'll notice the value that has, and sell it out 15:53 < nekoseam> God I love zsh 15:53 < nekoseam> Makes me want to use the terminal 24/7 15:53 < jprjr> I have brands that like, I'll look at them first, sure. But with tech the only brand I'm loyal to is Linux 15:53 < BCMM> there is little to no long-term thinking, and brand image is just another asset to them 15:54 < registeringIsBad> nekoseam is zsh that much different than bash 15:54 < nekoseam> a gamechanger 15:54 < registeringIsBad> BCMM I know, there are lot of short-term thinkers in management 15:54 < pingfloyd> when you regularly see good stuff from a brand. You tend to stick with it. 15:54 < registeringIsBad> BCMM but a few companies still have the idea of a lasting brand 15:54 < pingfloyd> maybe other people ran into duds with the brand, but that's their problem 15:55 < BCMM> i'm not aware of such a company that sells laptops 15:55 < jprjr> Most things are looked at on a case-by-case basis. Big companies know brand-loyal people will just buy stuff. I mean look at Apple, they openly shit on their customers and people still gobble up iPhones 15:55 < registeringIsBad> BCMM not in the laptop buisness no 15:55 < registeringIsBad> jprjr apple did use to have a good brand though 15:55 < registeringIsBad> jprjr rip Steve Jobs 15:55 < BCMM> jobs turned apple from a technology company in to a fashion company. 15:56 < jprjr> Yeah and that's what I'm getting at, is there's people that still see it as a good brand because they don't even look at other brands anymore, because they're loyal to Apple 15:56 < djph> well, he started them as a tech company, then made them into a fashion company ... 15:56 < pingfloyd> like with HDDs. Seagate is consistently garbage and WD is consistently good. Then there's the all the brand in between. 15:56 < BCMM> his tenure as ceo did wonders for shareholders, not so great for the tinkers who comprised their original customers 15:56 < pingfloyd> s/brand/brands/ 15:57 < BCMM> ^tinkerers 15:57 < jfe> can someone explain to me the purpose of the -F arch=... option in kernel audit rules? 15:58 < winsoff> After I log in (lightdm?) on parrotlinux, it takes a few moments (and a few clicks, it seems--perhaps I'm superstitious) for awesomewm to start. How should I go about diagnosing and fixing this? 15:58 < post-factum> pingfloyd: wd? my 1tb passport is almost dead after 1 year of just laying in the wardrobe 15:58 < jprjr> So yeah my point is, in most cases brands are virtually meaningless, and being brand-loyal to laptops is a good way to get burned on a laptop purchase at some point. 15:59 < pingfloyd> Apple II lines was totally for tinkerers 15:59 < jprjr> Go out, read the reviews, try out what you can, you should absoultely do a ton of research when buying a laptop 15:59 < pingfloyd> then came Mac and he pitted Apple II against Mac 15:59 < pingfloyd> my WD elements has been working great 16:00 < BCMM> jfe: on a typical amd64 machine, there are often 32 and 64-bit version of a syscall (for compatibility with 32-bit binaries and so on) 16:00 < BCMM> jfe: that lets you specific why version you're talking about 16:00 < pingfloyd> all the WDs, that I've bought, have obsoleted instead of failed on me. 16:00 < pingfloyd> that's what you want technology to do 16:01 < pingfloyd> end its life on your terms 16:01 < jprjr> I treat hard drives like they could fail at any time, have solid backups, etc. I've noticed my WDs last longer in my NAS, but the whole idea of my NAS is that I plan on hard drives failing eventually 16:01 < post-factum> lucky you 16:01 < pingfloyd> always keep backups 16:01 < post-factum> this is a backup hdd, so i'm fine ;) 16:01 < pingfloyd> only sane discussion for backups is "always have them". 16:01 < BCMM> winsoff: i think you can use bootchart2 to profile the login process as well as the boot process 16:02 < jprjr> Also test your backups, I can't tell you how many backups I've seen that have never been tested 16:02 < lilltiger_> backups are like girlfriends, evry nerd knows we should have at least one, but most of the time we fail to set them up properly... 16:02 < jprjr> If it isn't tested it's not a backup. It's a hunk of data that may or may not function as a backup 16:03 < winsoff> BCMM, I think this version of debian uses systemd. Does that change anything? 16:03 < post-factum> "at least one" 16:03 < lilltiger_> ;) 16:03 < post-factum> i like that in the context of girlfriends 16:03 < post-factum> that implies i can have 2, which is a woohoo 16:04 < lilltiger_> post-factum: ofcourse you can, if you configure them corectly 16:04 < BCMM> winsoff: to profile login, i think you'd probably just want to start and stop it manually rather than starting on boot and stopping at the login prompt. but i've not actually used it myself, so you'll probably have to look it up yourself 16:04 < jprjr> I'd highly recommend against maintaining multiple girlfriends 16:05 < registeringIsBad> I finally got my TV to work as a second monitor 16:06 < registeringIsBad> It is currently using the --same-as command to copy a 680x440 virtual monitor, so the overscan is fixed 16:06 < pingfloyd> have fwb 16:06 < registeringIsBad> Now all I have to do is try to make my TV recognize its actual resolution :/ 16:06 < BCMM> while we're talking about backups, any recommendations for hardware? i want a NAS that is stable, uses a *lot* less power than an x86 box, runs linux and at least lets me have root (preferably lets me run a normal-ish distro) 16:06 < registeringIsBad> I have a TV that is both smart and dumb 16:06 < BCMM> was thinking of a rock64 for low power and cost, but not sure whether it's really ideal in terms of stability 16:07 < TRS-80> BCMM: you hit the nail righ on the head 16:07 < pingfloyd> make one 16:07 < BCMM> TRS-80: with rock64? or with the concerns for stability? 16:07 < post-factum> jprjr: but backups? 16:07 < TRS-80> BCMM: it's a tradeoff 16:07 < post-factum> jprjr: or, redundancy even in this case 16:08 < lilltiger_> jprjr: To me a raid0 array of girlfriends sounds pretty neat. 16:08 < TRS-80> BCMM: I urgently need bio but will brb to discuss 16:08 < TRS-80> lilltiger_: deegenerate lol 16:08 < BCMM> TRS-80: what i don't really know how much the lack of error-correct memory actually matters for an always-on device 16:08 < pingfloyd> couldn't you use an SBC for that? 16:08 < BCMM> pingfloyd: that's what the rock64 is 16:09 < BCMM> pingfloyd: a lot of SBCs kind of suck for NAS because of bandwidth issues, but the rock64 has independent USB3 and GbE 16:10 < pingfloyd> GbE would be good for using iscsi 16:11 < BCMM> i also haven't really looked at, like, commercial NAS boxes. i know there's a lot of them out there but i don't really know how to tell which ones are reasonably open and which are tivoized trash that are obsolete at release 16:13 < pingfloyd> I'm sure most are tivoized one way or another 16:14 < GNU\colossus> if you don't insist on (physically) hot-swapable drives, there's VERY little those pre-assembled NAS boxes can offer that a custom-built PC can't (do better) 16:15 < BCMM> GNU\colossus: the problem with a custom-built PC, imho, is that it uses like an order of magnitude more power than a nas needs to 16:15 < GNU\colossus> I built https://geizhals.eu/?cat=WL-760094 in 2017, it's running Debian and can do everything I want/need, and more 16:15 < GNU\colossus> BCMM, that's not true at all. the box I have has 2x3TB 2.5HDDs and clock in at 8W idle 16:16 < GNU\colossus> clocks* 16:16 < GNU\colossus> that's not nothing, but it's _almost_ nothing. :) 16:16 < BCMM> GNU\colossus: huh, didn't know x86 could do that. that's worth of consideration actually 16:16 < pingfloyd> do you use low power governor? 16:16 < GNU\colossus> modern intel CPUs are using virtually no power when they are in low power states 16:17 < GNU\colossus> pingfloyd, no, the default (ondemand, or whatever it's called these days). so it can clock up to more than 2GHz per core. 16:17 < GNU\colossus> I do use powertop to force low(er) power states, though 16:17 < BCMM> pingfloyd: do you mean the in-tree powersave governer, or somethign else? 16:17 < BCMM> because the powersave governor is often counterproductive 16:17 < strixdio> woo 16:18 < GNU\colossus> I have a systemd service that runs 16:18 < GNU\colossus> [Service] 16:18 < GNU\colossus> ExecStart=/usr/sbin/powertop --auto-tune --quiet 16:18 < pingfloyd> on newer intels you often only have performance or powersave 16:18 < BCMM> on modern hardware, for a given workload, it is almost always advantageous to run as higher frequency as possible in order to complete the work and return to an idle state sooner 16:18 < GNU\colossus> that pushes idle powr consumption from ~11W to a ~8W on that hardware I linked to 16:18 < BCMM> ^as high a 16:19 < strixdio> My job is going to get me a laptop, and I am going to put linux on it. Any suggestions? Something inexpensive and from a supplier that a school can put in a PO for? 16:19 < BCMM> now if you have a workload that will *constantly* max out the cpu until it's done and you want to avoid overheating, powersave might be the answer 16:19 < BCMM> GNU\colossus: thanks for that. what are you using to measure power consumption? 16:20 < GNU\colossus> BCMM, a watt-meter for the wall plug 16:20 < GNU\colossus> I have two different models, both agree on the measurement 16:20 < GNU\colossus> (the HDDs are kept in spin-down mode during idle via hdparm, and I boot from USB) 16:21 < drzacek> Hello there 16:21 < GNU\colossus> the case fits neatly into the IKEA KALLAX series of shelves, btw 16:21 < TRS-80> BCMM: Also remember how much power the HDDs will be pulling and all of a sudden what the CPU pulls may become irrelevant, depending on how many HDD(s) you are thinking about of course 16:22 < drzacek> I'm using my special linux pc along with LCD panel with native 800x600 resolution, but due to certain software I need to run, I run it at 640x480 resolution (use xrandr to set it in autostart script) 16:23 < registeringIsBad> drzacek okay 16:23 < drzacek> Today, instead of connecting it via hdmi I used a hdmi-to-vga adapter, and suddenly the 640x480 resolution is no longer available (same pc, same lcd). Can I FORCE it to use 640x480? 16:23 < Reventlov> I'm wondering about the difference between MVM and DVM (for iwlwifi) 16:23 < Reventlov> I know some drivers are for certain cards only, but why this difference? 16:23 < registeringIsBad> drzacek does it list the 640x480 mode when you just type out "xrandr" 16:23 < drzacek> no 16:23 < Reventlov> (also: what do MVM and DVM stand for?) 16:24 < TRS-80> BCMM I can also relate my experience: been running self hosted NextCloud + home media storage on a Cubietruck the last year, it's been flawless and low power but worked so well that we are loading it up with more ans more old photos off old SD cards, ripping CDs, scanning in old photos, etc. that I am now starting to worry about reliability of all that important data... 16:24 < registeringIsBad> drzacek then you will probably have to add a new mode 16:24 < drzacek> it starts with 720x480, and it echoes "cannot find mode 640x480 16:24 < registeringIsBad> drzacek do gtf 648 480 60 16:24 < registeringIsBad> that will give you a long list of stuff 16:24 < registeringIsBad> now do xrandr --newmode [all that stuff that was just printed out] 16:25 < drzacek> it gave me two lines of stuff 16:25 < registeringIsBad> start typing right after it says modeline(do not include the word modeline) 16:25 < TRS-80> ...so I have been looking into ZFS as a more robust solution, plus I have privacy/freedom concerns so I have purchased KGPE-D16 because of ECC support for ZFS and also libreboot support (https://libreboot.org/docs/hardware/kgpe-d16.html) 16:26 < registeringIsBad> drzacek have you done that? 16:27 < TRS-80> I suppose I have just accepted that it will use a little more power but I plan on ordering high quality, efficient PSU as I always do in my builds as well as following you and GNU\colossus power discussion with interest 16:27 < drzacek> registeringIsBad, yes. It added a new mode...for different output (HDMI-2 instead of HDMI-1) 16:27 < drzacek> (could it be because of the vnc connection?) 16:27 < registeringIsBad> drzacek weird. doing --newmode should not change anything in the xrandr output 16:28 < TRS-80> but honestly sometimes I wonder if I should just keep it simple and attach another HDD and do simple rsync or such backups nightly... 16:28 < registeringIsBad> try doing xrandr --addmode [your inputname] 640x480_60.00 16:28 < registeringIsBad> and then type xrandr 16:28 < GNU\colossus> TRS-80, getting something that's 80+ certified will help. it's a good idea to make do with the PSU that offers the _least_ capacity rating that still ticks your other boxes 16:29 < GNU\colossus> any PSU's efficiency is relatively terrible when it has to operate at low power usage levels 16:29 < drzacek> registeringIsBad, it added a new mode for correct output, setting it gave me "xrandr: Configure crtc 0 failed" 16:30 < GNU\colossus> the one I use in my build is cheap, highly efficient, and inaudible at a little distance from your ears 16:30 < GNU\colossus> I'd happily buy it again 16:30 < registeringIsBad> drzacek no idea what that means. 16:30 < registeringIsBad> does anyone else here have experience w/ xrandr? 16:30 < Sitri> Jut ask 16:37 < TRS-80> GNU\colossus: yes I am aware of power curve and efficiency, etc. 16:37 < drzacek> used "xrandr -s 640x480" and it worked. I put this inside my autostart script, lets see if this work after reboot 16:38 < registeringIsBad> drzacek good idea. I didnt suggest that because that never works on my setup 16:41 < dimm> What a minimum for get availability for shared internet? 16:41 < dimm> eth0 is internet intrface 16:41 < dimm> wlan0 is a shared wifi interface 16:42 < dimm> we must up hostapd, dnsmasq and put something into iptables 16:42 < dimm> right? 16:45 < ksk> dimm: some echo to proc-net-ipv4-forwarding 16:45 < ksk> you should be able to google it easiy though. 16:45 < ksk> I dont know what hostapd is. 16:45 < dimm> yes, forgot about ip_forwarding=1 16:46 < dimm> i'm just in situation now where yesterday it was working and today is not :-) 16:46 < dimm> ok, i will trying step-by-step from start 16:46 < oneko> Yes, dimm, ip_forwarding=1 is necessary to set up a VPN 16:48 < oneko> And, by the way, dimm, what were you setting up ? 16:49 < dimm> oneko: i'm trying share internet via wifi 16:55 < BCMM> TRS-80: thanks! i'd missed the cubietruck somehow; didn't realise it had SATA 16:55 < pingfloyd> dimm: which dist? 16:56 < dimm> pingfloyd: debian 16:58 < pingfloyd> dimm: https://wiki.debian.org/BridgeNetworkConnections#Bridging_with_a_wireless_NIC 17:00 < pingfloyd> this looks like an easier way https://serverfault.com/questions/152363/bridging-wlan0-to-eth0 17:01 < ksk> I did not read the wiki, but having done something similar a few days past I came to the conclusion you cannot bridge ethernet and wifi devices 17:01 < ksk> (proably because wifi is a shared medium) 17:02 < revel> I'm stingy with my WiFi. 17:03 < ayecee> that's kind of personal 17:03 < revel> Yes, my WiFi is very personal to me. 17:03 < pingfloyd> revel: marry it 17:03 < g1itch> is there a way to get full emoji support in the terminal? 17:03 < ayecee> why don't you and your wifi get a room 17:04 < revel> I would if I could ;_; 17:04 < revel> g1itch: Get one with full UTF-8 support. 17:04 * pingfloyd writing a note to never go near revel's WAP wife 17:04 < g1itch> revel a terminal you mean? 17:05 < revel> A terminal emulator, yes. 17:05 < g1itch> gotcha - using guake now, any suggestions? 17:05 < pingfloyd> emoji support in a TE?! 17:05 < pingfloyd> what's this world coming to? 17:05 < revel> Well, xfce4-terminal works. So does konsole. 17:06 < pingfloyd> you'll get ASCII and like it! 17:06 < revel> CJK support is good, at the very least. 17:07 < revel> And if it supports emoji, then it probably supports other non-ASCII UTF-8 stuff as well. 17:07 < pingfloyd> revel: is this you? https://www.youtube.com/watch?v=FuNJq_wI1ns 17:08 < revel> No. 17:09 < BCMM> ksk: you *can* bridge wifi and ethernet, in general. but a lot of wifi devices or drivers don't support it 17:10 < ksk> BCMM: thanks for clearing that up. 17:11 < pingfloyd> revel: I'm this guy https://www.youtube.com/watch?v=MsO8f_uybU8 17:12 < jhodrien> BCMM: arp_proxying I think I've found to work more widely. 17:13 < BCMM> ksk: something to do with at what level packets addressed to other devices are ignored, i think (since bridging involves accepting and routing packets not addressed to you, and normal wifi usage involves receiving pretty much every packet on the network and ignoring most of it) 17:14 < dimm> pingfloyd: will try, thank you! 17:16 < jhodrien> bridging means exposing multiple MAC addresses on the single authenticated connection, whereas arp proxying only shows a single MAC. 17:21 < dimm> it seems like access point working normal, but isp blocking my DNS request 17:21 < dimm> ping working fine 17:23 < kurahaupo> dimm: have you tried Ā« dig +trace host.name Ā» 17:31 < dimm> kurahaupo: will try it now 17:40 < dimm> kurahaupo: cool command, thanks a lot! 17:41 < diverdude> in my dmesg i have this: parport0: PC-style at 0xd010 (0xd000), irq 18 [PCSPP,TRISTATE] <---- What does that mean? 17:51 < widp> How do pipes work in linux? 17:51 < Psi-Jack> Like flowing water. 17:54 < widp> I know them as an abstraction, but not really how they are implemented... 17:55 < jim> widp, if you do something like... cmdA | cmdB, what the | does is connect the stdout of cmdA to the stdin of cmdB, so that anything cmdA does to print (like echo and printf) goes to the pipe (and therefore the stdin of cmdB, and... 17:56 < jim> widp, whatever cmdB does to read from stdin, will come from cmdA 17:57 < toothe> odd - I created a new user, but it cannot execute /bin/sh 17:57 < toothe> or for that matter, any binary. 17:58 < toothe> it says permission denied for /bin/bash, or /usr/bin/vi 17:58 < dimm> yes, it was blocking from isp 17:58 < toothe> isp? 17:59 < widp> jim: What about programs which are already running which also take commands form a fifo file? 18:00 < widp> do they just read the fifo data in another thread or something? 18:01 < GNU\colossus> processes usually interact with data sources via file descriptors; it doesn't matter (much - there's exceptions, for things like seek(), for instance) what is the underlying concept/machinery that's repesented by the kernel as a file 18:02 < GNU\colossus> you don't have to read data in a separate thread for that kind of I/O activity 18:02 < mirya> Hello Guys can anyone help me with iscsi multipathing?? https://paste.fedoraproject.org/paste/Otj-qMa~Vj0MRqkBlp5e5w 18:02 < GNU\colossus> (read() is a blocking syscall; it will only return data once there is any to return) 18:02 < widp> GNU\colossus: I see, makes sense, thank you GNU\colossus and jim 18:10 < High_Priest> hi guys, anyone here willing to help me eradicate malware that is probably mining (keeps cpu at 100%) on my debian/raspberry pi3 box? probably got inside via rogue streaming kodi addon 18:11 < toothe> err 18:11 < toothe> I just created a new user 18:11 < toothe> but, that user cannot execute /bin/bash 18:11 < toothe> or any binary 18:11 < toothe> I get immediately logged off - what gives? 18:13 < ayecee> to the batlogs! 18:13 < ayecee> what did you set the new user's shell to? 18:13 < GNU\colossus> toothe, what's the user's login shell, and what is that you're doing? 18:13 < toothe> its /bin/bash 18:14 < GNU\colossus> like in, which commands are you executing, nopasted verbatim? 18:14 < toothe> ayecee: but, sudo -u user vi also says permission denied. 18:14 < ayecee> something is different about this system 18:14 < ayecee> which distribution? 18:14 < toothe> its a fcuked up Kali machine. Got I hate Kali. 18:15 < ayecee> mind the language please 18:15 < toothe> I'm sorry... 18:15 < nekoseam> how do I disable this 18:16 < nekoseam> https://i.redditmedia.com/46pQB-5HYGR4LPPsDzTnYsLGoAhTr450dj8DWRaUv5A.png?s=5010469e9c9d6fef88cf62f201c0da06 18:16 < ayecee> wild guess is that something has changed the permission on / 18:16 < BCMM> toothe: is installing a proper distro an option here? 18:16 < toothe> BCMM: sadly, no. 18:16 < toothe> it must be Kali. 18:16 < ayecee> nekoseam: disable what 18:16 < arora> kali is a proper distribution imho 18:16 < BCMM> there should never be such a thing as a "kali machine" 18:16 < nekoseam> ayecee: the lgbt pride month thing 18:16 < nekoseam> distracting 18:16 < BCMM> arora: sure, as a live system. installing it is a mistake. 18:16 < toothe> BCMM: exactly. Kali running in a VM is cool. Kali running on the betal is not cool. 18:16 < arora> it's just debian with stuff thrown with it 18:16 < ayecee> )`: 18:17 < BCMM> arora: the thing is, it's not... it's *broken* debian with stuff thrown in 18:17 < toothe> yes, that's it. 18:17 < toothe> Its broken Debian. 18:17 < arora> it's good as long as the user knows what he's doing 18:17 < BCMM> arora: and if you're going to install to a hard disk, *you will have the ability to install packages* 18:18 < BCMM> so what's in the default install ceases to matter, and the user can just install his pentesting tools on real debian. (if he knows what he's doing, of course) 18:18 < arora> toothe: what's the problem here? 18:19 < BCMM> toothe: it would be interesting to see the permissions on /bin/bash 18:20 < toothe> arora: When I try to run sudo -u username /bin/bash, I get "Permission denied" 18:20 < BCMM> i wonder if it's somehow got no "others" execute bit, so you need to be in a group to use bash 18:20 < toothe> logging in at the tty console gets me the same error and it logs out 18:20 < toothe> ssh'ing does the same thing. 18:20 < BCMM> toothe: can you do `stat /bin/bash`? 18:20 < triceratux> toothe: there must be a reason why kali doesnt ship with an ordinary user enabled. theyre not prepared fully to support user account creation 18:21 < BCMM> i mean, the main reason is that kali is *supposed* to be a single-purpose distro, and that single purpose involves being root (since so many of the included tools need to send weird network packets) 18:21 < arora> but technically it 18:21 < arora> 's possible to create user 18:22 < triceratux> but the process wont be seamless as on distros designed for the purpose 18:23 < BCMM> it's like the classic "my sound doesn't work on kali" - the developers quite reasonably don't really care about sound 18:23 < revel> A distro designed for creating users? :D 18:23 < arora> toothe: are you able to log in? 18:24 < triceratux> kali really is a root account with enough linux to get the hackzor apps to work. its not a proper linux with an account that happens to be root 18:25 < nekoseam> https://i.redditmedia.com/46pQB-5HYGR4LPPsDzTnYsLGoAhTr450dj8DWRaUv5A.png?s=5010469e9c9d6fef88cf62f201c0da06 is there a way to fix (remove) this from appearing? 18:26 < arora> triceratux: define "proper linux" 18:27 < ayecee> itt: zealots for and against kali 18:27 < revel> ayecee: Who's for it, exactly? 18:27 < triceratux> arora: generalpurpose distros that prioritise catering to the gnu implementation of a unixlike os. debian, fedora, opensuse, ubuntu & their generalpurpose derivatives 18:28 * triceratux is solidly behind the kali wallpaper 18:28 < ayecee> revel: just zealots against then? 18:28 < ayecee> maybe apologists for. 18:28 < revel> lol 18:28 < triceratux> if youre going to use kali that way issues are to be expected. its a value free thing 18:29 < revel> Do you really have to be a zealot to say "maybe using kali as a general-purpose distro isn't a great idea"? 18:29 < toothe> I just switched terminals via alt-F2, and this kid at work was like "....how did you do that?" 18:29 < ayecee> revel: zealots don't use the word "maybe" 18:29 < toothe> well, I'm only about 6 years older, but he was amazed 18:29 < revel> toothe: You can also just use the super keys. 18:30 < toothe> revel: super key is what exactly? pardon. 18:30 < revel> Or whatever they're called. The ones that usually have the Windoze logo on them/. 18:30 < BCMM> is toothe even still here? 18:30 < revel> That and the menu key to swap between "adjacent" ones and the right super key for switching to the last one you were on. 18:31 < ayecee> BCMM: turn off your ignore and see 18:31 < BCMM> ayecee: apparently i can't read... 18:32 < BCMM> toothe: are you able to check permissions on the bash binary? 18:32 < BCMM> when did Windows Key start mapping to Super instead of Meta? 18:33 < ayecee> meta is alt, no? 18:35 < BCMM> ayecee: only sort of 18:35 < revel> Well, I'm not 100% sure what those keys are called :P 18:36 < nekoseam> yay my terminal colorscheme isnt garbage anymore \o/ 18:36 < revel> They don't have nice little labels on any of the keyboards I have, just icons. 18:36 < BCMM> revel: ... icons for ctrl and alt? 18:36 < revel> "those" keys. 18:36 < revel> Not "all of them" 18:37 < BCMM> oh right, makes sense 18:37 < BCMM> i don't think i've ever seen that key labelled 18:37 < BCMM> i went from KBs with a gap there to KBs with a windows flag 18:38 < toothe> Hm...why is this login failure occurring... 18:38 < toothe> BCMM: yes, they are set correctly. 18:38 < ayecee> toothe: what are the permissions on / 18:40 < toothe> rwr-xr-x 18:40 < toothe> I just changed it to 755. 18:40 < ayecee> what was it before? 18:40 < toothe> eek, I should have checked before changing it. I got frustrated. 18:40 < ayecee> doh 18:41 < BCMM> toothe: can you try `su user -c /bin`, out of curiosity? 18:42 < BCMM> obviously won't work, but the manner of failure will tell us something 18:44 < toothe> fixed it! 18:44 < toothe> somehow i had /bin's other permissions as 0 18:44 < toothe> I don't know how that happened - but I wrote a STIG script, so somewhere ni there it probably did that mistake. 18:45 < triceratux> might be a kali thing. maybe thats why they need the root account 18:46 < BCMM> that carries the extremely worrying implication that this kali install is supposed to be somewhat secure... 18:46 < revel> What's the point of 700 permissions if basically everything is going to be run as root anyway? lol 18:47 < revel> And do all daemons run as root to get around that then? 18:48 < BCMM> what daemons does it even need? 18:48 < revel> Well, dhcpcd/NetworkManager/something like that, for one. 18:49 < revel> Maybe dbus? 18:51 < jim> revel, no, not all. for example, the backend daemon for postgres will refuse to run as root 18:51 < revel> Could've patched them :P 18:52 < revel> Does kali even have postgres though? Also, a bunch of daemons probably don't even need access to /bin after being started. 18:54 < toothe> yeah, I wrote a fairly large CIS Benchmark script, and possibly it sets the /bin directory to 750 18:54 < toothe> not sure how/why, but perhaps it was some mistake in a find command. 18:54 < toothe> who knows. 18:54 < plexigras> sorry i just lost my conection 18:55 < plexigras> did someone respond to my question? 18:55 < revel> plexigras: What question...? 18:55 < plexigras> oh maybe the internet was already down and i just didn't realize :) 18:56 < plexigras> i was asking if there is a way to have gnu diff show the content of files that only exist in one directory but not the other? 18:56 < triceratux> toothe: good reason to get user accounts working or you wont notice that stuff ;) 19:04 < BCMM> plexigras: -N 19:04 < BCMM> --new-file treat absent files as empty 19:05 < plexigras> thanks 19:06 < BCMM> np 19:12 < phinxy> I found a kernel config option that sounds interesting, how do I go about finding if it can be compiled as a module? (and where is the .c) 19:15 < BCMM> phinxy: how did you "find" it? are you in menuconfig? 19:18 < toothe> triceratux: right! 19:18 < toothe> triceratux: well, they will be running as root most of the time. 19:19 < toothe> s/most/all/ 19:19 < _AxS_> Hey all -- I was wondering if anyone's had any success setting up multipath NFS for file sharing with a recent (ideally 4.4) kernel series on server and client with in-kernel NFS ? 19:20 < TRS-80> BCMM yeah they actually make an add on power board for the Cubietruck as well, which makes it super easy to power the HDD as well. 19:20 < TRS-80> Think about it... ;) 19:20 < S3ri0usMiah> Hi Folks! How can I grep "something" in irc logs? I found this suggestion of browsing irc logs from a google search but there "grep" is not described. 19:21 < _AxS_> (I've got a single large storage server that I'm trying to use as a disk image host for a few libvirt-VM hosts, with 4x1Gbit 802.3ad-bonded nic's each, and I'm trying to make file-io over NFS exceed the bandwidth of one nic.) 19:21 < TRS-80> I mean the power, you don't want to be running a full PC power supply, so what to do to power the HDD? I ordered a high quality high efficiency Mean Well brick power supply from Mouser I think, with a barrel connector, to go into the little board that powers Cubietruck and then HDD. Works a treat. 19:22 < TRS-80> S3ri0usMiah: man grep 19:22 < S3ri0usMiah> Specifically, I would love any clue of what "grep" is in terms of IRC logs. 19:22 < _AxS_> S3ri0usMiah: grep is a textfile search tool, so you would "grep 'thing to search' /path/to/your/irc/logs/*" 19:23 * TRS-80 heads back to Emacs tutorial (for the second time in his life) ;) 19:23 < BCMM> TRS-80: i don't know what part of the world you're from; do you know if they do that brick with a UK plug? 19:23 < S3ri0usMiah> As I understand, that command works with my local files. But what if it's hosted in remote server? 19:23 < TRS-80> BCMM: I'm in US but I'm pretty sure you can 19:24 < BCMM> S3ri0usMiah: grep is a command line tool that is practically certain to be already installed on your computer 19:24 < BCMM> S3ri0usMiah: it searches for text in files 19:24 < _AxS_> S3ri0usMiah: then you're SOL; there's no way to 'grep' files you don't have. That said, often people will use 'grep' as slang for search. 19:24 < TRS-80> BCMM: calculate your wattage needs (based on HDDs) if you need help or pull the trigger on CT just holler and I can give you some pointers 19:24 < BCMM> S3ri0usMiah: "grep" is also sometimes used as a verb meaning looking for or searching for things in general 19:25 < BCMM> TRS-80: thanks! 19:25 < TRS-80> or maybe dig up a link oor whatever 19:25 < BCMM> TRS-80: does it run normal-ish distros? 19:25 < S3ri0usMiah> I appreciate you. At least I've found an explanation of my question. 19:25 < TRS-80> BCMM: Mine is running Armbian, works a treat 19:26 < TRS-80> BCMM: good support of A20 that's why I picked it 19:26 < TRS-80> besides SATA 19:26 < sanroot> There any way to run pulseaudio in windowsbas daemon 19:27 < TRS-80> BCMM: see https://linux-sunxi.org/Buying_guide and also https://docs.armbian.com/ 19:28 < bls> sanroot: ask ##windows 19:29 < TRS-80> BCMM: Incidentally, I was looking at getting ROCK64 (I think that was the one) to play with as desktop replacement because they are so cheap but FWIW I've been running my CT based home server for like a year or 2 now and it's been rock solid! Can't speak highly enough about it! 19:34 < mavorsa> does anyone here use rofi? i'm trying it out, and trying to just open htop from it, but nothing happens when i select terminal based apps from it. what might i be missing? 19:38 < jim> what's a rofi? I notice it's in the string 'profile' 19:39 < msiism> jim: rofi is a program similar to dmenu. 19:39 < mavorsa> jim: it's a custom text-based pop-up application launcher 19:41 < msiism> mavorsa: you probably need to give it the termainl emulator to use as well. 19:42 < mavorsa> just found the answer after sorting through the manual. Shift-Enter, instead of just Enter 19:42 < msiism> mavorsa: have you tried 'xterm -e htop'? 19:43 < mavorsa> nope, but i figured it out 19:43 < msiism> mavorsa: so, what does shift do? 19:44 < mavorsa> msiism: when the launcher pops up, you can select entries with Enter. enteries that are command-line based are opened with Shift-Enter (holding shift while hitting enter) in order for it to open in the environment's default terminal emulator. 19:46 < msiism> mavorsa: ok, nice. i'm currently running dmenu with a wrapper script from suckless.org for history and auto-completion. i should really try rofi sometime. what's your distro (if i may ask)? 19:51 < msiism> i'm wokring on a small application (a CLI tool) that will make random note-taking easy for me. now, this program (currently just a bash script) will have to store two things: text files containing notes (done in a folder hierarchy based on the date) and one or more dotfiles, including a configuration file (used to set which editor to use, for example). files containing notes will be stored in ~/app_name/notes/YYYY/MM/DD/ by default. 19:51 < msiism> where would i best put the dotfiles? just /home/$user? or ~/app_name/? 19:52 < jiffe> are there multiple ways of writing to stdout? I have a program that prints a prompt without a newline that I'm trying to read from within perl using sysread (unbuffered) and it doesn't read this last line. If I write my own program to duplicate this (writes to stdout without buffer or newline) my perl script reads this as expected so I'm not sure what the difference is 19:53 < hexnewbie> msiism: https://freedesktop.org/wiki/Software/xdg-user-dirs/ 19:58 < hexnewbie> msiism: I'm sorry, I have you the wrong link. Config files should go to "${XDG_CONFIG_HOME:-$HOME/.config}/app_name"; data files should go in "${XDG_DATA_HOME:-$HOME/.local/share}/app_name"; cache and files that can be removed should go in "${XDG_CACHE_HOME:-$HOME/.cache}/app_name". Link: https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html 19:58 < msiism> hexnewbie: i've already had a look at that. it would simply be .config/app_name/ then. but this is not exactly a desktop (as in gui) program. 20:00 < jiffe> could there be a difference in interaction between this program and perl vs the shell such that the program wouldn't have written to stdout yet if run from perl but it would if run from the shell? 20:00 < hexnewbie> msiism: Yeah, but respecting XDG_CONFIG_HOME is good. 20:01 < msiism> hexnewbie: maybe you're right. it's one of the good points of the xdg user dirs definition after all. 20:02 < hendrix> it always irks me when an app dumps it's config file/directory to $HOME. grr. 20:05 < msiism> hendrix: if it's just a single file, i'm kind of ok with that. 20:06 < strixdio> ~. 20:10 < msiism> hexnewbie: ok, going with ~/.config/app_name now. 20:14 < hexnewbie> What I meant is that ~/.config/app_name is not "${XDG_CONFIG_HOME:-$HOME/.config}/app_name" 20:14 < Hejkki> hi. I have nokia 7 plus, how can i use the internet over usb cable on my linux pc? 20:16 < revel> msiism: He's saying that if the environment variable XDG_CONFIG_HOME is actually set, then it should be ${XDG_CONFIG_HOME}/app_name , not ~/.config/app_name 20:16 < Psi-Jack> Very important, that. 20:16 < dbe> Hejkki: I don't know about nokia 7 specifically; but what devices does the nokia expose over usb? (e.g. 'udevadm monitor' as the device is plugged in, or 'lsusb') 20:21 < Dagmar> USB tethering and bluetooth tethering basically "just work" if you're using a decent distro 20:21 < mirya> Can anyone help me with iscsi multipathing as I have successfully multipathed my node with 2 TGT servers, but now for the node to boot up it needs both the TGT servers to be up( which I dont want)I want the node to boot up with either of the TGT servers being up 20:22 < msiism> revel: oh, i missed that. thanks. 20:24 < Dagmar> mirya: You have to use active/standby. 20:25 < mirya> Dagmar where do i specify that ? This is what I have done https://paste.fedoraproject.org/paste/Otj-qMa~Vj0MRqkBlp5e5w 20:28 < dreadkopp> hey community. I run a custom xsession only displaying chromium in kiosk mode. however I found that mouse clicks aren't handled correctly unless I i.e. open chrome dev tools (f12)... Any ideas how to fix this ? 20:30 < lukey> dreadkopp: try running a minimal wm like evilwm 20:32 < Dagmar> "aren't handled correctly"? 20:32 < phogg> or any WM which allows borderless windows and full screen 20:32 < msiism> hexnewbie: this user dirs specification, i don't know, it seems kind of odd to me. is it really necessary to apply such a degree of management there? 20:32 < dreadkopp> lukey: hmmm... not sure I'd like that since now, when the user closes chrome for some reason it will automatically revert to lightdm, which wouldn't be the case with a wm, righty? 20:32 < phogg> msiism: so much yes 20:33 < phogg> dreadkopp: depends on how you set it up 20:33 < dreadkopp> Dagmar: right click not working, cannot click the menu button, clicks on dropdown selects are not recognised 20:35 < Dagmar> That would be normal afaik 20:35 < phogg> dreadkopp: if nothing else works you can at least run a background watchdog script which kills your WM if it doesn't seen any chrome process running. You could also launch chrome via a script of the form 'chrome ; killall your-wm-name' so it will issue a kill as soon as chrome returns 20:36 < dreadkopp> Dagmar: normal you say ? oh ... doesn't seem normal to me XD 20:36 < Dagmar> The general idea being they're at a _kiosk_ so it's probable that they shouldn't be allowed to open menus and do things that would let them escape the weaksauce sandbox 20:36 < msiism> phogg: ok. it's just that i have a rather different way of setting up dirs in my $HOME, which is why it feels a bit like being nannied to me. 20:37 < dreadkopp> Dagmar same behaviour when started in normal mode though 20:37 < phogg> msiism: Everyone does it his own way and can keep doing that. XDG user dirs are for telling applications where you want things put, not for applications to tell you where you must put things 20:38 < irgendwer4711> hi, how to show the un-effective (real) username? somehow id -un gives me the effective one 20:39 < phogg> msiism: it should always be up to the *end user* to decide what files and directories exist in $HOME, especially if they are not dotfiles. 20:39 < Dagmar> https://bugs.chromium.org/p/chromium/issues/detail?id=475506 20:44 < msiism> phogg: totally agree. and for the purpose you mentioned, these user dirs really make some sense. but i still don't fully get why somehting like this would have to exist. i'll have to think about that. 20:45 < phogg> msiism: Suppose you're writing an application and it needs to run as a regular user and write a file somewhere. Since regular users can only write to $HOME and /tmp you have one choice if it's not a temp file: $HOME. You now have figure out, somehow, where in $HOME the user wants his files. 20:46 < phogg> msiism: The user either has no opinion and doesn't care, or wants to specify where. 20:46 < dreadkopp> Dagmar: I guess I fixed it: simply running a 'xmodmap -e "pointer = 3 2 1" ' before starting chromium seems to have fixed it 20:46 < phogg> msiism: Some tools can just ask for a base directory, or even for each file. For some this is impractical. 20:46 < phogg> msiism: especially when it comes to storing config files 20:46 < dreadkopp> *1 2 3 20:47 < phogg> msiism: since every programmer thinks a little differently if we are all left to our own devices we all make slightly different decisions about where to put files by default and how to ask the user where he wants files. The result is (and was) very messy 20:48 < phogg> msiism: it's not necessary to have any elaborate infrastructure for *one* program, but even if YOU only write one program the *USER* never only has one program and he would generally prefer some kind of consistency. 20:48 < phogg> dreadkopp: are you running chrome in real kiosk mode or just full screen? 20:51 < dreadkopp> phogg first in kiosk mode but behaviour persists when in 'just full screen' or even totally normal start 20:51 < dreadkopp> anywho, it's fixed... now to the next task: remove url bar and menu button haha 20:52 < phogg> dreadkopp: use firefox, edit the chrome XUL file 20:52 < phogg> firefox makes it deadly easy to do this 20:53 < phogg> ... or it did before they decided to move away from XUL. I imagine it will become harder again in the future 20:58 < dreadkopp> phogg : hmm... I'd like to switch to Firefox but first I need to find a replacement in-browser screenshare extension like jit.si for chromium 20:59 < sstory> does a crontab have to be a root crontab to run all of the time--even when not logged in? 20:59 < bls> no 20:59 < ayecee> no 20:59 < msiism> phogg: ok, right, i see. that totally makes sense for config files and such (and downloads, too). my doubt is more about the structure prescribed by the things you can actually set there being a bit, well, preclusive. anyway, i'm getting a headache thinking about it, so i'll stop that for now. thanks for your explanations. 21:00 < bls> ..."but it works when I run it by hand" 21:00 < ayecee> classic 21:00 < triceratux> "use a systemd timer" 21:12 < phogg> sstory: cron jobs do not "run all the time"; they are started on the given schedule. This works so long as the cron daemon is running (which is normally "all the time"). 21:13 < phogg> sstory: the environment in which a cron job executes is not (by default) the same as your user's environment, so if you tried to run a job and aren't getting the expected result the first thing to do is to make sure you have set up all environment variables you depend on (even the ones you don't know you depend on) e.g. PATH inside your script. 21:13 < phogg> sstory: cron jobs also will not normally be able to interact with the X server unless you take extra steps, so if the program you are running needs a working X sessions you will need to do more work 21:14 < ayecee> maybe even reconsider your life choices 21:14 < Dagmar> ayecee++ 21:15 < phogg> ayecee: I know, let's use a cron job to change the slides in a slide show every two minutes! 21:15 < phogg> by killing and re-starting an image viewer, of course 21:16 < ayecee> too real man, too real 21:16 < icy`> any1 have that (sorta silly) ascii stream of world cup 21:16 < Sonolin> is displayport -> HDMI audio supported in linux? 21:16 < Sonolin> can't seem to find any HDMI devices via aplay, and lspci doesn't show it either 21:16 < ayecee> does displayport do audio? i didn't think it did. 21:16 < Sonolin> hmmm perhaps it doesn't 21:16 < Sonolin> althought the box seem to said it did 21:17 < ayecee> i may be mistaken 21:17 < Sonolin> s/said/implied/ 21:17 < ayecee> it's happened at least once before 21:17 < dreadkopp> irrc dp supports audio 21:17 < Lope> I'm busy trying out kde-plasma on ubuntu 18.04. It's very nice! But I can't get keyboard shortcuts working. Trying to launch gnome-calculator. I've tried just the program name and also the full path to it. I've tried it as a global shortcut and as an application shortcut. When I press the key-combo, nothing happens. I've setup the same keyboard shortcut on Mate desktop environment, and it works. I've set it to have 2 virtual desktops, now the 21:17 < Lope> hotkeys for switching virtual desktops is working... 21:17 < Sonolin> thanks dreadkopp for confirmation, wasn't sure either 21:17 < Sonolin> wondering if some kernel driver needs enabling, since I don't even see the audio device - I'm on Intel, and everything seems to be enabled for audio, though 21:18 < Sonolin> are there any DP (or HDMI) specific audio settings I should look into? 21:18 < phogg> Lope: you want it as a global shortcut. If you just can't get it going you can always resort to xbindkeys. That works on everything, even if it doesn't have a fancy config UI 21:18 < sstory> phogg: Thanks! I don't need any interaction. It will do some action and report to a Zabbix server. I just wanted to know if I could add it to a non root user cron and it still run. It isn't a script that needs root priv and don't want to do that. 21:19 < dreadkopp> Sonolin can you test with a dp-in monitor ? maybe your adapter is the problem 21:19 < Sonolin> perhaps, I don't have any DP monitors to test though, I suppose I should look for one 21:19 < Lope> phogg: thanks! I didn't know about xbindkeys! I actually prefer something I can script. Less GUI setup to do every time I reinstall!! 21:19 < dreadkopp> Sonolin no HDMI out on your machine ? 21:20 < phogg> Lope: plus you can take your favorite keybindings with you as you experiment with different DEs/WMs 21:20 < Lope> phogg: for shizzle 21:20 < phogg> Lope: of course it won't help if you plan to also switch to wayland, but that's tomorrow's problem 21:21 < Sonolin> dreadkopp nope this is an antique 21:21 < Sonolin> I suppose I can just stick to my aux out, just would be nice to listen through TV speakers 21:21 < Lope> I don't know what wayland is. I was using mint-mate. Then switched to ubuntu and ran mate on that, which was barely acceptable. Installed new ubuntu now, comes with gnome3. Gnome3 sucks so badly, it's like a dumbed down android. 21:22 < Sonolin> I'm just happy I get digital out now - I was using VGA (which hurts my eyes) before, much nicer with HDMI/DP 21:22 < Lope> Gnome3 is probably fine for a tablet for a desktop, they're outta their minds. 21:22 < dreadkopp> Sonolin well... maybe it's the gpus fault then since it does not support audio at all 21:22 < Sonolin> dreadkopp hmm posibly, its like gma 4500hd I think 21:22 < dreadkopp> check lspci 21:22 < phogg> Lope: Wayland is the prospective replacement for X. It's a new display server, new protocol, new everything. It is considerably different in detail. 21:22 < Sonolin> dreadkopp any pointers on what to look for? Both DP and HDMI not found there 21:23 < Lope> phogg: ah, when is it likely to hit? Probably not 18.04. 18.04 probably has feature freeze? 21:23 < phogg> personally I would have preferred if they called it Z instead of Wayland. 21:23 < phantomcircuit> i've got a usb-c flashdrive from sandisk that's presenting oddly 21:23 < Lope> They say X was very dated. X doesn't handle hot plugging monitors as gracefully as one might hope. 21:23 < phogg> Lope: I believe it is optionally available in the current ubuntu release. They keep talking about making it the default, but it's not clear when that might really happen. There are a lot of things still not quite done with it, in terms of user experience. 21:23 < phantomcircuit> it seems it's a pci thunderbolt device pretending to be a usbhub for some reason 21:24 < phantomcircuit> https://paste.linux.community/view/raw/550ff4b1 21:24 < phogg> Lope: the only major distribution to be heavily into wayland so far is Fedora, where the bleeding edge is the norm 21:24 < phantomcircuit> (works on windows and android (lol yes)) 21:24 < phantomcircuit> any ideas? 21:25 < dreadkopp> Sonolin: checking with lspci I see i.e my AMD GPU as 05.00.0 and the sound encoder on it as 05.00.1 ... not sure about the igps though 21:26 < Lope> phantomcircuit: see if it shows up in `lsusb` and `fdisk -l` 21:26 < Sonolin> thanks again dreadkopp 21:27 < phantomcircuit> Lope, it's not presenting as a disk 21:27 < phantomcircuit> but does add two root hub entries to lsusb 21:28 < Sonolin> I see Intel HD Audio controller, pretty sure that's just the AUX out/internal sound card 21:28 < Sonolin> oh well, not a biggie 21:28 < _AxS_> phantomcircuit: usb-storage module is loaded right? 21:29 < phogg> phantomcircuit: If it were me I'd guess the kernel is too old to figure out what it's doing. 21:29 < phogg> phantomcircuit: since it works on android I assume the device is probably not completely incorrect 21:30 < Lope> phantomcircuit: what kernel? 21:31 < Lope> phantomcircuit: trying the latest kernel available, trying other bleeding-edge linux live distros, good google-fu, or otherwise a USB-C to USB2 adapter is worth trying. 21:32 < dreadkopp> arg .... jets screen sharing in Firefox only works with window manager ... ugh ... :/ 21:32 < dreadkopp> *jitsi 21:32 < phogg> not surprising 21:33 < balletjebal> is there a way to get a cmdlist on linux via a terminal to get to know some of the major commands used in linux 21:33 < dreadkopp> in chromium it works without @ phogg 21:33 < ayecee> hit tab twice 21:33 < ayecee> shows all commands 21:33 < dreadkopp> ayecee .... seriously ? 21:34 < ayecee> hmm, maybe not. 21:34 < ayecee> it used to. 21:35 < balletjebal> ayecee, how ? on what screen 21:35 < ayecee> via a terminal 21:35 < dreadkopp> balletjebal start with something like this: https://files.fosswire.com/2007/08/fwunixref.pdf 21:35 < triceratux> Display all 1746 possibilities? (y or n) 21:35 < triceratux> still working for me 21:35 < ayecee> see, that's what i was expecting to see 21:35 < ayecee> maybe it's these stripped down vms i keep making. 21:36 < ayecee> always missing some little command or other i take for granted 21:36 < balletjebal> aint working overhere 21:36 < balletjebal> dreadkopp, Thanks 21:37 < ayecee> just as well. it's not a very helpful list. 21:37 < ayecee> but it is a very accurate list 21:37 < Kwicster> can run 'man (command)' to learn more about a command 21:37 < balletjebal> yeah i was rather looking for something to pop up from a terminal then switching to the web 21:38 < ayecee> -bash: syntax error near unexpected token `command' 21:38 < Kwicster> haha well replace (command) with the command you want to look at 21:38 < Kwicster> man ls 21:38 < ayecee> :D 21:41 < Siecje> I have a long running Python program, that uses subprocess for one task. After it has been running for a long time the subprocess fails because there is not enough memory. There is free RAM on the system, how can I determine if it is the overcommit setting? 21:42 < GNU\colossus> Siecje, how does it fail? is the OOM killer reaping it? 21:42 < GNU\colossus> or does memory allocation fail, and it quits voluntarily? 21:44 < Siecje> GNU\colossus: The subprocess fails, the original process is fine. 21:45 < GNU\colossus> Siecje, ok, but _how_ does it "fail" exactly? 21:48 < steamport> Question: does AMDGPU-pro work on Linux 4.16? 21:49 < Siecje> GNU\colossus: Python can't spawn the subprocess. OSError(12, 'Cannot allocate memory') 21:50 < GNU\colossus> ok, so it may be fork() that's failing. 21:50 < GNU\colossus> (or clone(), rather) 21:50 < longxia> Siecje: see how large the memory footprint of the parent is with: ps -eo vsz,args | grep the_prog_name # if this is approx. larger than free mem left, yes it has to do with the use of the clone() call by python. Better than to use overcommit, is to keep the parent's mem footprint small. 21:50 < GNU\colossus> that's gonna be tough to debug :/ 21:51 < GNU\colossus> you'll have to keep an eye on memory usage via /proc/meminfo, and yes, overcommit _can_ be a reason for stuff like this 21:52 < GNU\colossus> longxia, shouldn't the CoW semantics of clone/fork NOT require the process image's current allocation to be free? 21:52 < GNU\colossus> (I'm actually not sure about this) 21:53 < steamport> Question: does AMDGPU-pro work on Linux 4.16? 21:54 < longxia> GNU\colossus: you would think so, but apparently the initial claim on memory, even though it is immediately discarded by an exec() must be satisfied. I have to other explanation. I have seen this happen in Perl code too. Frankly, i think this whole clone() behaviour is Linux shooting itself in the foot. 21:54 < longxia> to=no* 21:55 < GNU\colossus> longxia, that's interesting, and would explain some rather odd allocation failures i've seen over the years 21:55 < Siecje> How does it fail if there is free RAM on the computer? 21:55 < Siecje> Can a single process only use so much RAM? 21:55 < ayecee> yes 21:56 < Siecje> How do I find that limit? 21:56 < longxia> Siecje: depends on how much free RAM 21:56 < ayecee> make larger and larger allocations until they fail 21:56 < GNU\colossus> depending on the overcommit setting, successfully allocated memory may come to bite you once you actually try to use it :) 21:58 < _AxS_> Siecje: it's not using a shared memory allocation or anything like that is it? those have their own limits independent of the OOM as well 21:59 < Siecje> It is a very small script, the reason I am using subprocess is because the script needs root permission and that user has permission to run that script as root. 22:00 < Siecje> So would this happen if I just ran `ls` in a subprocess? 22:00 < Siecje> I found this. https://stackoverflow.com/questions/1367373/python-subprocess-popen-oserror-errno-12-cannot-allocate-memory The link to the overcommit calculation is a 404 Not Found. Where can I find that. 22:01 < ayecee> to the wayback machine! 22:01 < hexnewbie> My experience is that it bites in regardless of the commit limits, in either directions, and under ā€˜reasonableā€™ limit for your workload you're threatened by both OOMs and commit failures (if you don't have swap, adding swap usually fills the gap, if you're willing to put up with extreme swapping instead of OOMs) 22:01 < GNU\colossus> https://www.kernel.org/doc/Documentation/vm/overcommit-accounting @ Siecje 22:01 < birdbolt1> need a quit bit of advice 22:02 < birdbolt1> I wanna set an env var in one of my projects which will contain a path 22:02 < birdbolt1> I know bash ignores double slashes, eg /home//Documents 22:02 < birdbolt1> and treats as a single slash 22:03 < birdbolt1> is it of concern if i set the path in the env var including the trailing slash 22:03 < Siecje> Here is my /proc/meminfo and the htop at the time of the failure https://dpaste.de/B910 22:03 < birdbolt1> and then when I reference it from anywhere, i still include the trailing slash for consistency and readability? 22:03 < GNU\colossus> birdbolt1, it's not bash that does that; it's the system's C library 22:04 < birdbolt1> oh nice, good to know :) 22:04 < hexnewbie> Siecje: Are there any limits for the virtual memory of the failing process? /proc/$pid/limits 22:04 < GNU\colossus> application log can still differ, depending on the presence of atrailing slash somewhere. rsync is a popular tool that does. 22:05 < Siecje> hexnewbie: Which process? The top level child of init or the bottom? 22:05 < hexnewbie> Siecje: The one that fails to fork()/clone() 22:05 < GNU\colossus> application logic* 22:05 < Siecje> hexnewbie: https://dpaste.de/q7s2 22:05 < birdbolt1> eg. MY_ENV_VAR = /repository/app/ then say in my project i reference it with /home/birdbolt1/$MY_ENV_VAR/tools 22:06 < birdbolt1> rather than /home/birdbolt1/$MY_ENV_VARtools 22:06 < GNU\colossus> birdbolt1, that should be fine. (that's not a _trailing_ slash in my book, btw ;)) 22:07 < birdbolt1> Im have extremely limited experience with linux but can that pose any sort of obvious/immediate security risks or weird behavior? I'm working with nginx, react, django, uwsgi 22:07 < birdbolt1> GNU\colossus, wait so.. what is a trailing slash then lol 22:07 < birdbolt1> at least in your book ;) 22:07 < GNU\colossus> /after/these/words/there/is/a/trailing/slash/ 22:08 < GNU\colossus> /after/these/words/there/is/NO/trailing/slash 22:08 < GNU\colossus> trailing := at the very end 22:08 < birdbolt1> aahh shouldve been more specific lol 22:08 < Siecje> A trailing slash doesn't make a difference. If the destination exists or not does though. 22:08 < GNU\colossus> regarding potential security implications: is the content of MY_ENV_VAR supplied by the user? 22:08 < birdbolt1> nope, I'm setting it in my config files 22:09 < GNU\colossus> then that should be fine 22:09 < birdbolt1> I only got into this so if i decide to change later i dont have to ctrl-f through the whole project lol 22:10 < hexnewbie> Siecje: So it's not that. 22:10 < Siecje> hexnewbie: What unit is Max locked memory 65536 ? 22:10 < microcolonel> Howdy folks 22:10 < Siecje> microcolonel: Nice nick! 22:11 < microcolonel> thanks :- ) 22:11 < hexnewbie> Siecje: Should be KiB, but I could be wrong 22:11 < lnnb> Siecje: kbytes, ulimit -a 22:11 < microcolonel> Ansyone know of a filesystem or filesystem configuration which can permit a file to be opened for write only once? 22:11 < microcolonel> *anyone 22:11 < mrig> Hi, how can i add a program for a non sudo user on linux? 22:11 < Psi-Jack> microcolonel: Not possible. 22:11 < Psi-Jack> Nor should it be. 22:12 < microcolonel> Psi-Jack: I mean, I know I can make it work, just wondering if there's a more straightforward way of configuring this 22:12 < hexnewbie> microcolonel: Use advisory locks in your program 22:12 < Psi-Jack> microcolonel: You're trying to work with a problem with a bad solution focus. 22:12 < microcolonel> you know nothing about my problem 22:12 < microcolonel> hexnewbie: I'll take a look, thanks. 22:13 < microcolonel> ahh 22:13 < Psi-Jack> Unless of course, you're looking to insure a file is only opened for write, in terms of a programatic manner. 22:13 < Psi-Jack> That differs a lot. ;) 22:13 < microcolonel> Psi-Jack: I meant that after the first time the file is written, it would no longer be writeable 22:13 < icy`> could make file just owned by a particular user and force others to su to that user every time they want to write 22:14 < microcolonel> otherwise I just muss with the permissions with inotify or some such similar 22:14 < Psi-Jack> microcolonel: Then no. Not possible. That wouldn't really be much of a filesystem. :p 22:14 < Psi-Jack> So again, I state, you're looking at the wrong solution. 22:14 < icy`> microcolonel, what's the use case for never wanting to write to it ever again? 22:15 < GNU\colossus> microcolonel, you can set the file immutable after you wrote to it 22:15 < GNU\colossus> but that's not absolute either 22:15 < microcolonel> GNU\colossus: sure, I mainly want to prevent accidental destruction or overwriting of data 22:16 < microcolonel> I have a separate log-based thing which I use for retention 22:16 < GNU\colossus> then the immutable extended attribute is for you :) 22:16 < microcolonel> cool, I'll take a look 22:16 < microcolonel> thank you GNU\colossus 22:17 < lnnb> microcolonel: it's possible but not trivial if you want to eliminate race conditions 22:17 < Siecje> lnnb: https://dpaste.de/HzWm ulimit -a 22:17 < microcolonel> lnnb: yeah, that was one main reason for wanting a filesystem-internal solution, rather than a bolt-on 22:18 < birdbolt1> anyone know if dockerfiles can inherit environment variables from docker-compose.yaml 22:20 < GNU\colossus> birdbolt1, I know nothing about docker, but that's probably possible. #docker-compose will know for sure. 22:20 < GNU\colossus> (or some other channel that's related to docker. check out /msg alis list *docker* ) 22:20 < microcolonel> GNU\colossus: it's a shame normal users can not set the immutable flag 22:20 < mrig> Ah ok, I compile and build into the users home .bin, seems like the best option 22:20 < microcolonel> unsetting is understandable 22:20 < birdbolt1> nice thanks mate 22:21 < GNU\colossus> microcolonel, for normal users, there's `chmod go-w` 22:21 < GNU\colossus> and that ought to be enough for anybody ;) 22:22 < noway96> So my hardware boots ubuntu on my lenovo desktop but not on my lenovo laptop. Any ideas why? There's a partition that's unknown according to gparted 22:22 < microcolonel> GNU\colossus: yeah, I think that'll be fine to at least hold off until I can set the attribute with a separate process 22:22 < noway96> harddrive* 22:24 < V7> Hey all 22:25 < rany> Hey! 22:25 < V7> Does anyone know how to make a new user be permanent in proteus ? 22:25 < V7> Is it just copy configs related for a user to rootcopy path ? 22:25 < rany> Did you install Proteus on a hard disk or is it running from a live disk? 22:26 < devslash> I want to set it up so within my own lan, pinging the host name git resolves to a local server in my LAN. So for example pinging git will resolve to 192.168.1.20 is that something that I can set up within my existing debian installation or is that something that I need to configure at the router level? 22:26 < Siecje> does every subshell run fork()? 22:26 < Lope> phogg: i tried xbindkeys. It's freakin amazing! 22:26 < rany> devslash: You can setup an avahi server that would do the job 22:27 < devslash> whats avahi 22:27 < devslash> is that a dns server 22:27 < rany> It's the open source bonjour implemention 22:27 < Lope> So much better to use a config file and the terminal instead of a GUI! 22:27 < Lope> Oh how linux has changed me. 22:27 < devslash> lope thats all i do for administration stuff in linux... 22:27 < V7> rany: A hard disk 22:27 < devslash> rany, is there an easier way to do that ? 22:28 < Lope> devslash: you mean you use xbindkeys? 22:28 < devslash> no 22:28 < devslash> I use the command shell without any GUI to administer and maintain my linux server 22:28 < rany> devslash: You can put the ip address in your /etc/hosts file 22:29 < V7> In some cases a terminal is a GUI, too 22:29 < devslash> rany but id have to do that for each client right / 22:29 < Lope> devslash: thanks for sharing 22:29 < devslash> V7 I was talking about a headless server 22:29 < V7> Comparing hexademics and binaries 22:29 < rany> devslash: Well, yes. but that's the way that I know of if you don't want to use Avahi 22:29 < devslash> hmm 22:30 < V7> devslash: Isn't a server no gui by default ? 22:30 < devslash> depends 22:30 < rany> but I don't think you have many servers anyway :P 22:30 < V7> gui just adds useless loads 22:30 < devslash> ubuntu server is headless but ubuntu desktop has a desktop window manager 22:30 < V7> What do you mean by word "headless" 22:31 < bls> devslash: check out unbound or dnsmasq, you could run one of those and then aim everything on the network at them via dhcp 22:31 < devslash> a headless server is a server that does not have XOrg installed and does not have a window manager installed 22:31 < devslash> everything is done at the command line level 22:31 < V7> devslash: I think you're not right 22:32 < birdbolt1> well well https://hastebin.com/cudokapidi.hs GNU\colossus 22:32 < V7> Headless: no monitor, no peripherals at all 22:32 < devslash> https://whatis.techtarget.com/definition/headless-server 22:32 < devslash> yes 22:32 < rany> V7: Turns out Proteus doesn't save changes you make to it by default. More info here: http://www.porteus.org/component/content/article/26-tutorials/general-info-tutorials/116-howto-create-a-savedat-container-and-why-you-need-one.html 22:32 < devslash> well you have to access it remotely 22:32 < devslash> but you can access it locally 22:33 < birdbolt1> That is on alpine, which I understand is without a lot unnecessary tools, etc 22:33 < V7> So, it can have GUI and Xorg, but without any local peripherals 22:33 < V7> So you can view it remotely 22:33 < V7> rany: Yes, of course, but how to save a user's data ? 22:34 < V7> I mean, it's not hard to persist data, but which one is needed for a user ? 22:35 < V7> 1) /etc/passwd; 2) /etc/group; are these enough for a user+password and group ? 22:35 < bls> and /etc/shadow 22:36 < V7> So these 3 fiels will be enough ? Thank you very much rany and bls, really. 22:39 < V7> rany: It's not only changes "cheatcode" which they provide, but "proteus/rootcopy" path to 22:39 < V7> rany: No, I don't have a GUI in proteus 22:40 < Siecje> GNU\colossus: lnnb hexnewbie longxia It is because overcommit was set to 2. 22:40 < Siecje> so there wasn't double the memory available and it fails. 22:40 < Siecje> Why does python subprocess need for fork()? And request double the current memory used by the current process. Just to run ls. 22:41 < triceratux> they assume youll just download more ram 22:42 < rany> V7: According to the docs, you should be able to enter the GUI by pressing TAB on startup 22:42 < hexnewbie> Siecje: You had 800 MiB free on those pastes, is your Python process using 800 MiB RAM? But yeah, I've had large Python processes fail to allocate memory due to overcommit set to 2. 22:42 < Siecje> hexnewbie: It was in the paste it was using 52% of 1.8 GB. 22:43 < rany> Maybe you should change the cheatcode and boot to the gui to generate a SAVE file 22:43 < azarus> Does Linux have a configurable malloc()? 22:43 < Siecje> 1 GB swap though so it must have just barely not been enough. 22:43 < bls> azarus: configurable how? 22:43 < V7> Sorry, but a gui is not needed currently 22:43 < azarus> (for example, to junk free()d memory) 22:43 < prussian> mallopt 22:43 < azarus> prussian: ah, thanks 22:44 < hexnewbie> Siecje: Ah yeah, I forget that overcommit_ratio is 50%. 22:44 < lnnb> Siecje: interesting, did not know about overcommit options 22:44 < Siecje> hexnewbie: We have it set to 100% 22:44 < V7> Thank you rany 22:44 < bls> azarus: not sure what you mean by "junk free()d memory" 22:44 < rany> Did you get anywhere, V7? 22:44 < azarus> bls: to basically put junk in freed memory 22:44 < azarus> (freed by the free() syscall) 22:44 < V7> Yup, it works now 22:45 < lnnb> azarus: you can override malloc symbol with your own implementation 22:45 < rany> Great :D 22:45 < lnnb> LD_PRELOAD 22:45 < azarus> lnnb: sure, but I was wondering if one could do it systemwide ;) 22:45 < bls> azarus: ah, check out the MALLOC_PERTURB environment variable 22:45 < lnnb> you have to modify mmap/brk i guess 22:46 < lnnb> in kernel 22:47 < bls> or MALLOC_PERTURB_ 22:49 < bls> azarus: there are also other tunables available: https://www.gnu.org/software/libc/manual/html_node/Memory-Allocation-Tunables.html 22:49 < azarus> Cool, thanks for the help! 23:11 < Lope> OMG KDE PLASMA IS $*##*(@#( AMAZING 23:12 < Psi-Jack> And sadly buggy as heck. :/ 23:13 < searedvandal> all this praise of Plasma, makes me wonder if I should give it another go at some point 23:14 < bb36e> searedvandal: I found it more familiar/easier to use than GNOME, however I keep running into minor bugs (keyboard shortcuts not working correctly for specific KDE windows, widgets looking buggy, text colour not being set properly etc). 23:15 < bb36e> I know they're legit bugs because I reported them and they were acknowledged as such, but it's just annoying sometimes 23:15 < searedvandal> that sounds like bugs that would irritate me a lot. 23:16 < triceratux> the dedoimedo guy agrees https://www.dedoimedo.com/computers/ubuntu-beaver-slow-boot.html i wont even go there 23:16 < bb36e> IMO many of those little bugs are a side effect of the customizability of KDE -- it comes with a lot of options out of the box and some configurations are just overlooked. but I'm on debian stable so maybe it's been fixed in the past year or so 23:16 < bb36e> dedoimedo was the guy who was reviewing distros using live CDs, right? 23:17 < Psi-Jack> Heh, my biggest bug was the taskbar flickering like crazy after being up a time period. Just goes crazy. 23:18 < bb36e> it's really disappointing, too. I want desktop linux to be great again but when the little annoyances pile up it's hard not to get frustrated 23:19 < bb36e> I wish I could go back to karmic koala, give me my desktop cube 23:19 < Psi-Jack> Well, I've been so for... Surprisingly happy with Budgie on Solus. 23:19 < electrosys> does anyone have anygood links for setting up hotmail/outlook with offlineimap and oauth2? 23:19 < triceratux> heh im running icewm because it runs so well on one of the xorg 1.20.0 distros. doesnt have to be a bugfest :) 23:20 < Psi-Jack> triceratux: Yes. But how much RAM is being used by what all you generally run? :0 23:21 < bb36e> it even got to the point where I considered installing fvwm, but then i realized that my beard wasn't grey enough 23:21 < bb36e> I'll spin up a VM and give budgie/solus a try though 23:21 < Siecje> Why does a fork exec request the amount of memory the parent process has? 23:21 < triceratux> Psi-Jack: MemTotal: 1964416 kB MemFree: 550440 kB 23:22 < Psi-Jack> What output is that even from? hehe 23:22 < bls> Siecje: for copy-on-write 23:22 < triceratux> Psi-Jack: modest hardware to begin with but im only running mc, hexchat, & opera 53 23:23 < Psi-Jack> Opera? 23:23 < bb36e> I thought all the opera folks moved to vivaldi 23:23 < Psi-Jack> triceratux: You use opera... We can't be friends anymore! LOL 23:23 < Siecje> bls: But if you don't have enough virtual memory it will fail. 23:23 < JimmyNeutron> All LVM documentations/videos that I've seen says you must change the partition type to "Linux LVM". However, I was able to create the LVM using the default linux partition type of 83(Linux). Is partition type "Linux LVM" really needed? 23:24 < triceratux> Psi-Jack: thats cat /proc/meminfo 23:24 < Psi-Jack> Ahhh 23:24 < bls> JimmyNeutron: not really, no. that's a legacy thing 23:24 < Psi-Jack> The Mem line of "free" is better. ;) 23:25 < JimmyNeutron> bls: Thanks! I'll do some more testing and change partition type to FAT32 and see what happens 23:25 < bls> I'm not sure what the issue is. how do you expect your system to behave when it's out of memory? 23:25 < bls> JimmyNeutron: just be wary if you dual boot. I think windows still uses the labels 23:25 < JimmyNeutron> bls: Thanks for that info! 23:25 < triceratux> total used free shared buffers cached 23:25 < triceratux> Mem: 1964416 1852904 111512 0 78048 565724 23:26 < Psi-Jack> So yeah.. Nearly 2GB of 2GB. 23:26 < Psi-Jack> And all you're doing is hexchat, mc, and opera? 23:27 < triceratux> but ive got 5G of swap. it kicks in at kvm time 23:27 < Lope> tricer 23:27 < Psi-Jack> Yikes. Too much swap. 23:28 < Siecje> bls: But if you are going to exec you will replace the fork anyways right? So why does it ask for the same amount of memory that the parent process has? 23:28 < Lope> I've not encountered any bugs with plasma yet. Only been setting up and testing stuff on it. 23:29 < Lope> It's amazing. Just fucking amazing. 23:29 < Lope> Makes all other DE's (incl w10 and OSX et al) look like toilet paper. 23:30 < Psi-Jack> Soft and cuddly? 23:30 < Lope> Not in a eye candy way (I've disabled animations) but functionality. 23:30 < Lope> night night 23:30 < Sonolin> Lope have you used plasma with touch yet? I'm wondering how it compares to latest GNOME for touchscreens 23:30 < Sonolin> oh nm he's gone 23:31 < Psi-Jack> NetworkManager? 23:31 < Psi-Jack> What about it? 23:31 < bls> for copy-on-write, if you don't want copy-on-write semantics, look into using clone() or vfork() 23:42 < bls> or tweak the overcommit settings as was mentioned above 23:44 < hexnewbie> Sonolin: My guess is that GNOME is far superior on touch. But I haven't tried the latest Plasma versions on touchscreens, but from using it on a desktop, I don't see how it could have improved. 23:44 < Happyhobo> Hi folks 23:44 < searedvandal> hi 23:45 < bb36e> I have to say though, KDE connect is awesome 23:45 < bb36e> if I try gnome I'm going to try and get it running on there 23:58 < nekoexmachina> hey there folks. I can't google this at all - what is proper word for panning in framebuffer? Like I want to have an area in tty where no text is printed ever period --- Log closed Thu Jun 21 00:00:06 2018