--- Log opened Mon Apr 30 00:00:03 2018 00:06 < phinxy> Is this a correct bash if-branch? It executes startx even tough $(tty) is tty2 http://termbin.com/q7fw 00:07 < phinxy> I'm sure it worked before the linux updat 00:07 < uplime> if [[ $(tty) = '/dev/tty1' ]]; then startx ...; fi 00:07 < bls> phinxy: == only works in bash's [[, not [ 00:08 < uplime> it works in bash's == as well 00:08 < uplime> erm, in bash's [ as well 00:08 < bls> phinxy: also, shellcheck.net 00:08 < uplime> its just not standard 00:08 < uplime> (so as soon as you go to POSIX sh, it breaks) 00:09 < bls> right, so be explicit #!/bin/bash [[ and == 00:09 < mawk> "" aren't mandatory in [[ ]] 00:09 < mawk> it's prettier without 00:10 < uplime> it is for the right hand side 00:10 < uplime> for expansions at least 00:10 < mawk> around $() I mean 00:10 < mawk> even around a variable containing spaces 00:10 < uplime> just for the left side 00:10 < mawk> yeah 00:10 < uplime> expansions on the right side still need to be quoted 00:16 < stevendale> https://www.ebay.com.au/itm/DELL-INSPIRON-6000-1-73GHZ-512MB-RAM-60GB-HARD-DRIVE/401528637091 00:16 < stevendale> I bought it :D 00:16 < stevendale> Gonna get to revive it with Linux! 00:17 < ococooccoo> c00l 00:18 < triceratux> stevendale: http://antix.mepis.org/index.php?title=Main_Page 00:18 < stevendale> triceratux, I was thinking off Debian netinstall :) 00:18 < Psi-Jack> ococooccoo: Another new nick, dannylee? 00:19 < phinxy> I changed the if branch to double brackets and when switching to tty3 its still starting x 00:19 < phinxy> http://termbin.com/yvs6 00:19 < triceratux> stevendale: thatll work too. be sure to start with the nonfree unofficial community netinst 00:19 < stevendale> triceratux, Why's that? OwO 00:20 < triceratux> stevendale: itll have the network drivers whose absence can otherwise thwart an install 00:21 < bls> phinxy: then why not try running it with `bash -x` or spitting out some debug to a file? 00:24 < Psi-John> "it'll", please use the queens approved apostrophe for future correctness 00:25 < triceratux> Psi-John: "queen's" 00:25 < Psi-John> triceratux: "Queen's" ! 00:26 < jim> umm wut 00:26 < pr3c0g> also have no clue what's going on, but I just arrived so 00:27 < AOL_> cwene f (nominative plural cwenan) 00:27 < AOL_> a woman 00:27 < jim> oh :) what brings you? 00:27 < apb1963> phinxy, Figure it out yet? 00:27 < AOL_> From Proto-Germanic *kwenǭ, from Proto-Indo-European *gʷḗn. Cognates with Balto-Slavic *genāˀ, Old Saxon cwena, Dutch kween ‘barren cow’ 00:28 < stevendale> https://www.ebay.com.au/itm/ASUS-EEEPC-4G-UNTESTED/401529302477 Is it worth the risk, those things were $200.00 back in the day 00:28 < phinxy> apb1963• its both tty1 and tty5 at the same time 00:28 < stevendale> I have chargers for it 00:28 < stevendale> And spare batteries 00:28 < apb1963> phinxy, ?? 00:28 < phinxy> apb1963• my computer is broken after updating. cant even grep 00:28 < apb1963> phinxy, What is $(tty) supposed to be? A variable? 00:30 < apb1963> phinxy, is that a script you wrote? Or did you get it from somewhere? 00:31 < jim> stevendale, maybe, what do you want to do with it? 00:31 < phinxy> apb1963• the .profile? It's just a branch to start xorg in tty1 when logging in 00:31 < phinxy> and some keyboard settings 00:31 < apb1963> phinxy, I ask, because typically one uses ${} not $() for a variable. 00:32 < bls> $() is the syntax for command output substitution 00:32 < apb1963> i.e. braces {} instead of parentheses (). $() starts a subshell. 00:33 < bls> which is exactly what he wants to happen 00:33 < mawk> if I swap two bytes in an ipv4 packet, it doesn't change the checksum right ? 00:33 < apb1963> doesn't look like it 00:34 < stevendale> jim, Web server probably 00:35 < apb1963> phinxy, are you trying to execute the tty command? 00:35 < [R]> mawk: try it and see 00:37 < aaro> mawk, if those two bytes are in the ip header, and the cheksum is the ip header checksum then yes 00:42 < jim> stevendale, first off, is the guy close enough to get to personally? 00:42 < stevendale> jim, No 00:44 < bls> you don't want to use garbage old desktops/laptops for a server 00:44 < mawk> alright, what is the TUN_F_CSUM flag for then ? for tun devices 00:45 < mawk> when I enable it the kernel refuses my packets with no checksum, and the kernel hands me checksummed packets 00:45 < [R]> mawk: well you hae the source code... 00:45 < mawk> doesn't look at all like checksum offload 00:45 < mawk> yeah I read it 00:45 < mawk> it enables checksum offloading 00:45 < mawk> but it seem to have no effect 00:45 < mawk> it's a shortcut for the ethtool checksum offloading flag 00:46 < bls> do you have a NIC with checksumming hardware? 00:47 < mawk> no, but why ? 00:47 < mawk> the NIC is the tun device here 00:47 < mawk> which clearly has checksumming feature (if it's not unimplemented for now, that is) 00:47 < phinxy> bash -x was very helpful, it shows that the comparison is correct, if tty1 = tty2 00:48 < phinxy> the login bash is not the same as /bin/bash 00:48 < phinxy> I'll just nuke the machine. 00:49 < bls> it's the same command, it just behaves different based on whether or not it's a login shell, an interactive shell, or a neither/a script interpreter 00:52 < mawk> looks like unimplemented, actually 00:52 < mawk> that's so lame 00:52 < Psi-Jack> No YOUUUU! 00:52 < mawk> :( 00:52 < mawk> case TUNSETNOCSUM: /* Disable/Enable checksum */ /* [unimplemented] */ 00:53 < Psi-Jack> vwvvwvwwvw: Would you mind not changing nicks so frequently? This is a large channel, and nick changes are noisy. 00:53 < vwvvwvwwvw> ok i'll go back to my old nick 00:53 < mawk> you can trick the system, Psi-John 00:53 * Psi-Jack rolls his eyes. 00:53 < mawk> /part, /nick, /join 00:57 < Psi-John> also, if you can address me as, sire or "my liege", thanks for future correctness 00:57 < Psi-Jack> Not happening. 01:00 < instantp10neer> Anyone have an opinion on the safest VM'd distro for financial transactions including cryptocurrency that stores data (no livecds)? I was using Ubuntu and am now evaluating CentOS, RedHat or Ubuntu Server. 01:01 < [R]> "safest"? 01:01 < [R]> its all the same crap 01:01 < instantp10neer> secure* 01:01 < bls> it's all the same software, so none is going to be safer than the other 01:01 < Psi-Jack> Well, security comes from knowledge and experience. CentOS comes with SELinux, provided you use it properly, it can be secured heavil. 01:01 < instantp10neer> most secure 01:01 < Psi-Jack> "most" anything is opinionated. 01:01 < Psi-Jack> In such a context anyway. 01:01 < [R]> do a console install and removea ntyhing y udont need 01:02 < bls> and if you break out the "safest out to the box because I'm not going to do anything to secure it" I'd ask wtf you're handling financials 01:02 < instantp10neer> OK. If a console was done with the three, which would you choose? Compatibility should weigh in. 01:02 < [R]> its all the saem crap 01:02 < [R]> so... close your eyes and pick 01:02 < Psi-Jack> instantp10neer: Stop polling. 01:02 < Psi-Jack> We're not a voting booth. 01:02 < instantp10neer> hostile much? 01:03 < Psi-Jack> Only towards people that ask the same question 5 times over. 01:04 < rotb> I have two servers with CentOS 6, one is running 4.9.63-29.el6.x86_64 and the other is running 3.18.34-20.el6.x86_64. Both running Xen 4.6. First server suffers a NIC lockup every two weeks requiring a full power cycle. Second server (running 3.18) never locks up. I'm debating downgrading 4.9 server down to 3.18 via yum. Anyone ever encounter problems doing a downgrade like that? 01:04 < bls> I'd replace the NIC before I regressed OSs 01:05 < sauvin> There's no difference between Ubuntu and Ubuntu Server. Linux is client; Linux is server. This is true even of budget models. Also: "security" is a word many folks use to justify fat paychecks. 01:05 < ananke> that 4.9.x kernel is not what comes with centos 6, which makes me wonder why you'd pick it 01:06 < Psi-Jack> elrepo sounds like. 01:08 < phogg> instantp10neer: how secure something is has everything to do with you and little to do with the OS, basic functionality taken as a given. Even Windows can be secure if you put in enough effort. 01:08 < ananke> rotb: downgrade should be safe, and probably the easiest way to determine if this was the culprit 01:09 < sauvin> I put in all KINDS of work into securing my Windows machine. I ripped out its ethernet. 01:09 < ananke> hah. I personally disagree with all of the definitions of security presented so far, because each one is incomplete and very biased 01:09 < phogg> sauvin: that's effective but has a negative impact on the the number of connections per second that you can process with it. 01:10 < bls> then you give us yours so we can all disagree with it as well 01:10 < phogg> ananke: you're WRONG! 01:10 < instantp10neer> I don't argue a lock is safer latched than not. Some locks are difficult to close. Some are difficult to use. Does anyone have experience with Cent, RedHat or something else they feel is fairly compatible and easier than others to secure, without being cryptic? 01:10 < phogg> ... 01:10 < phogg> did I jump the gun a little? 01:10 < AOL_> sauvin: i did that too, i don't run a windows machine but once in a blue moon i plug in a HDD with windows 7 on it to flash VBIOS of GPUs to mine crypto, that is the only thing i ever need it for, and it has no ethernet connection 01:10 < Aph3x-WL> instantp10neer: alpine might be what you want 01:10 < phogg> instantp10neer: it's all exactly the same 01:11 < AOL_> on that subject, does anyone here have experience of flashing VBIOS of GPU within linux? i am sick of using win7 just for this purpose with atiflash, it's only reason i still have w7 on a dusty old HDD 01:11 < ananke> security is the net result of procedures, techniques and products used for the purpose of protecting a given computing system and its assets 01:11 < bls> instantp10neer: most of have used them all. once you know what you're doing, none is easier than another. if you don't know what you're doing or are listening to people that don't, they all look like different magic black boxes 01:12 < instantp10neer> AOL_, you could always Windows To Go it, as a step in the right direction. 01:12 < phogg> instantp10neer: do you know what's different between CentOS and RHEL apart from the branding? It comes down to a tiny handful of packages. Quick! Which is more secure? You might as well ask whether a chocolate bar in the wrapper or out of the wrapper has the greater amount of chocolate. 01:12 < ananke> dismissing any of the components and placing emphasis only on the product, or technique, or procedure misses the mark 01:12 < [R]> ananke: you put the wrong emphasis on the wrong cilabil 01:13 < sauvin> Very interesting spelling, [R]. I'm guessing the dictionary I keep telling you to get didn't happen to be an English one. 01:13 < AOL_> instantp10neer: i don't want to use windows at all, i use linux for desktop and server 100%, the only time i use the old dusty w7 hard drive is to flash VBIOS of GPU for friends or myself when we get new GPU for mining 01:13 < AOL_> i know there is an atiflash for ethos, linux, but i don't use ethos 01:13 < ananke> and 'experience and knowledge' are contributing factors in achieving security, but they don't make much difference if they're not used. knowing something versus actually doing something are two very different things 01:13 < [R]> sauvin: lol 01:14 < phogg> [R] is just a fan of bad movies 01:14 < [R]> AOL_: linux is linux 01:14 < [R]> AOL_: is a program runs on linux... it runs on linux 01:14 < [R]> phogg: great movie 01:14 < AOL_> [R]: ethos is a paid solution, and atiflash is not open source, they bundle it with ethos 01:14 < phogg> [R]: interestingly that's not always true. Some programs depend on kernel features that are only patched into specific distributions. 01:14 < AOL_> there are versions of atiflash uploaded various places, i wont trust them 01:15 < sauvin> Let me guess... the movie was "Cybill Does Acid". 01:15 < phogg> sauvin: I honestly don't remember the title. 01:15 < [R]> phogg: so whats stopping you from appling those patches? 01:15 < [R]> sauvin: no, something with mike myers 01:15 < phogg> [R]: Nothing, but it sort of contradicts the "Linux is Linux" angle. Linux isn't just one thing, so they don't compare equal. 01:16 < sauvin> Not one of those stupid Austin Powers things, was it? 01:16 < AOL_> i also don't agree with the ethics of ethos and other mining distros which are making money from open source free software 01:16 < phogg> sauvin: no 01:16 < AOL_> i wont support them with money 01:16 < [R]> sauvin: a) austin powers is great b) no 01:16 < phogg> sauvin: also, only the sequels were stupid. The first Austin Powers was brilliant satire. 01:16 < sauvin> No, it wasn't "great". It was just... um... shagadelic. 01:16 < [R]> AOL_: as long as they are abidiing by the licesnes... ain't nothign wrong with taht 01:16 < [R]> sauvin: baby 01:17 < nrg> not a good way to lose ones head 01:17 < [R]> you could say he wasn't ery head strong 01:17 < AOL_> [R]: i just don't agree with their ethics, so i choose not to support them, plus i configure everything manually anyway, it's just atiflash i am missing ... maybe i will try the random uploads of it someday and possibly brick a GPU 01:18 < phogg> [R]: I feel like we need to hang out some time. Have a movie marathon or something. 01:18 < instantp10neer> Aph3x-WL I will check it out. I appreciate it. Asking a simple question such as any opinions on which distro is most compatible and secure brings about ten minutes of butt hurt for the intelligence of the admin. I miss people remembering Linux is a community open source, free, welcoming movement for software. 01:18 < [R]> phogg: and then have you murder me? i think not 01:18 < phogg> instantp10neer: you asked an impossible question so you're going to get unsatisfactory answers 01:18 < [R]> instantp10neer: what does "most compatible" even mean? rofl 01:18 < AOL_> [R]: also, i use KVM with PCI passthrough for the GPU mining, and i notice there is an option to load a BIOS file for each PCI device, each GPU, but it doesn't seem to stick with the file and resorts back to the VBIOS of the GPU, any insight into that? 01:19 < phogg> instantp10neer: if you want to ask a *specific, answerable* question you'll find nothing but help. Except maybe from Psi-Jack who may also give you snark, 01:19 < [R]> AOL_: all i heard was "blah blah blah" 01:19 < AOL_> i think it would be cool to be able to just specify the VBIOS file for the PCI devices of the KVM guests 01:19 < AOL_> ROM/VBIOS 01:19 < [R]> i think it would be cool if all the aholes that did mining... stopped 01:19 < [R]> lol 01:19 < bls> instantp10neer: and alpine isn't a very good recommendation as it uses a different libc and userland than normal distros. if you don't know what you're getting into, it could bite you badly 01:19 < instantp10neer> phogg because thinking outside of the box to answer with an opinion is so far from expectations 01:20 < phogg> instantp10neer: if what you want is advice on which Linux distribution to choose in the name of security you have your answer: The distro choice doesn't matter that much. 01:20 < AOL_> [R]: why are you so salty? 01:20 < phogg> instantp10neer: I'm an engineer. I don't guess. 01:20 < [R]> phogg: CHOO CHOO 01:20 < AOL_> i have been mining since way back when 01:20 < phogg> [R]: Not *that* kind. The kind which pays better. 01:21 < [R]> phogg: i'm pretty sure train engineers get paid handsomly? 01:21 < phogg> [R]: I've honestly no idea but I figure it's a blue collar gig. 01:21 < [R]> haha 01:21 < pnbeast> instantp10neer, your question basically implies that have no idea what to do with any answer you get, here. "most compatible"? With what? "most secure"? They are, basically, all the same, and if you don't know it already, it means you don't understand their use, and so you're screwed on security a priori. 01:21 < [R]> lots of blue collar jobs rake in the bank 01:21 < [R]> i hear garbage men make a ton 01:21 < phogg> [R]: good unions will do that 01:21 < [R]> lol 01:22 < instantp10neer> elitist nonsense 01:22 < sauvin> phogg, what kind of engineer? 01:22 < phogg> sauvin: software, things related to software 01:22 < phogg> instantp10neer: you are entitled to your opinion, but don't go blaming a failure of the community. Ask a better question. 01:23 < pnbeast> instantp10neer, is that why the whole history of the internet is rife with the perfect security that people got by randomly choosing a Linux distro recommended in an IRC channel? 01:23 < instantp10neer> pnbeast, the question was in regard to cryptocurrencies, as stated 01:24 < phogg> instantp10neer: Is your question "Which distribution is best to run a miner?"? 01:24 < bls> "what's the best vehicle?" "the world is in full agreement. the best vehicle in the world is a 1967 Ford Ranchero" 01:24 < AOL_> instantp10neer: security requires you to actually make an effort, not just pick a distro and expect it to be secure 01:24 < instantp10neer> this is getting nowhere. thanks anyhow. with this kind of hostility and yes 'snark' it is a small wonder nobody is willing to share 01:24 < jml2> instantp10neer, thanks anyhoo? 01:24 < pnbeast> Oh, wait, now that we know it's for cryptocurrency, so the answer is clearly OS X. You don't want Linux, instantp10neer. OS X is *your* choice. 01:25 < jml2> cryptocurrency? you mean those morons who put $1000 into things and then get .0001 cents out ? 01:25 < phogg> instantp10neer: We're getting nowhere, yes, but it's not *hostility*. I just can't answer your quesiton because there is no answer. There is no best, you're going to need to ask a more specific question. 01:25 < jml2> LOL 01:25 < bls> this isn't some "we know the answer but we're not willing to tell you because you're not in the in crowd" this is a situation where you don't like the correct answer 01:25 < [R]> phogg: what soda is the best? 01:25 < irwiss> clearly pepsi 01:25 < phogg> [R]: Coke. No, Mountain dew! 01:25 < jml2> whp cryptocurts ? 01:25 < AOL_> jml2: hey don't compare all miners and node runners to people who bought in at $19,000 because CNBC told them to 01:26 < AOL_> :) 01:26 < [R]> phogg: coke, no pepsi? 01:26 < Psi-Jack> Okay... These https://www.homedepot.com/p/Ideal-RJ45-Cat6-Modular-Plugs-25-Pack-85-366/202276269 suck... Never buy them. 01:26 < sauvin> Java Monster Mean Beans. 01:26 < phogg> [R]: too sweet 01:26 < AOL_> some of us have been in this from the early days jml2 01:26 < [R]> phogg: thats why its better... 01:26 < phogg> Psi-Jack: I sense a story 01:26 < DennisA> Root beer! 01:26 < phogg> [R]: for a treat like candy yes, but if you're sipping all day long to keep the buzz going you don't want that 01:26 < [R]> when i was a kid, i would get sick from root beer 01:27 < [R]> phogg: lol, i drink 1 soda at lunch time, and then if i go out to dinner i'll get soda then, otherwise i dont drink too much anymore 01:27 < Psi-Jack> phogg: I'm trying to make a few more cables to plug into my secondary NIC ports to setup an openvswitch internal network, and using these RJ45 heads. The tip modules aren't through-hole so it makes it 10 times harder than even traditional RJ45 heads. 01:27 < [R]> phogg: in college, i used to drink a liter o cola a day 01:27 < phogg> [R]: so for you Pepsi is fine. For the long haul drinking I prefer something milder. 01:27 < [R]> haha 01:27 < phogg> [R]: Super Troopers 2 was good, by the way. 01:27 < DennisA> am in college, am drinking a liter a day 01:28 < bls> mmm, fresca 01:28 < [R]> phogg: yeah, i saw it 01:28 < jml2> Psi-Jack, you have to trust online reviews even for the simplest things :)) 01:28 < phogg> DennisA: Hi, this is the future calling. Don't keep doing that. Your teeth cannot take it. 01:29 < Psi-Jack> Yeah, I seriously should've just bought the through-hole plugs and used my flush cutters or razor to trim them down. 01:29 < DennisA> My teeth thank you. Is it the year of the Linux desktop yet? 01:29 < phogg> Psi-Jack: you live, you learn 01:29 < Psi-Jack> Thankfully I only have... 25. 01:29 < phogg> DennisA: for me that year was over a decade ago 01:29 < Psi-Jack> And an odd number, no less. 01:30 < phogg> no wait... 1.5 decades now 01:30 < [R]> phogg: i dont buy soda at home anymore... sometimes it gets deprsssing 01:30 < DennisA> @phogg I'm new to irc. how are you directing your message to me? 01:30 < DennisA> What command are you using 01:30 < phogg> [R]: I just get them from the vending machines at work. Sixteen ounces per day is my limit. 01:30 < Psi-Jack> DennisA: den 01:30 < jml2> I like guanabana and the clementina https://www.beachbasketbelize.com/671-large_default/san-pellegrino-sparkling-clementine-beverage-clementina.jpg -- if you want something very sweet go for either of them, I guarantee you will love it 01:30 < jml2> XDXDD 01:30 < bls> DennisA: the IRC convention is to use nickname: not @nickname 01:31 < DennisA> Psi-Jack: thank you 01:31 < phogg> DennisA: a direct message is the /msg command, but to send a private message unsolicited can be taken as very rude. 01:31 < DennisA> Got it! Thanks everyone! 01:31 < Psi-Jack> DennisA: @nick doesn't generally tab complete, nor is it traditional IRC. 01:31 < lupine> or nickname, 01:31 < instantp10neer> "Anyone have an opinion on which distro to use in a VM for storing cryptocurrency wallets? Compatibility with mainline (e. g. Ubuntu) and greater ability for security (see: less bloat, easier U. I. for non to semi power user) are pluses." 01:31 < phogg> lupine: some people do that, but it's like 1-2% 01:31 < lupine> I think it's original xchat andm um 01:31 < bls> instantp10neer: ubuntu 01:32 < lupine> nobody 01:32 < lupine> ubuntu is not mainline 01:32 < instantp10neer> bls ok, ill pretend. which flavor? 01:32 < phogg> instantp10neer: No. Use what you're familiar with already. If you want low bloat start with Debian's netinstall disc. 01:32 < lupine> better: abandon the project 01:32 < bls> instantp10neer: if it's for a vm/server, flavor doesn't matter 01:32 < lupine> cryptocurrency is evil and you are going to hell for being involved 01:32 < phogg> instantp10neer: pro tip: *ALL UBUNTU FLAVORS ARE IDENTICAL*. It's just different default packages. 01:32 < lupine> well, hell or gulag 01:32 < bls> phogg: oh no, don't start that again :P 01:33 < Psi-Jack> instantp10neer: You've already been answered. Stop repeating. 01:33 < phogg> crypocurrency isn't evil, but speculation bubbles are 01:33 < jml2> phogg, you can use any major distro minimal iso installer these days.. most have one... 01:33 < [R]> phogg: LIES 01:33 < phogg> jml2: But most distros aren't Debian, so I recommend the one that is. 01:33 < jml2> [R], lies naked with you 01:33 < [R]> phogg: more than milk bubbles? 01:33 < DennisA> what is the difference between Ubuntu flavors and Ubuntu-based distros like Linux Mint and Elementary? 01:33 < bls> DennisA: default package lists 01:34 < Psi-Jack> DennisA: Ubuntu is Ubuntu is Ubuntu. 01:34 < jml2> DennisA, just the cake icing.. they're mostly ubuntu packages 01:34 < phogg> DennisA: ubuntu-based distributions change a variety of things and always include additional packages Ubuntu does not. 01:34 < Psi-Jack> There is no "flavors", just distributions of. 01:34 < phogg> if there are no additional packages then it's not really a new distribution 01:34 < DennisA> So if Canonical forks Ubuntu, it's a flavor. Anyone else is fork 01:34 < bls> DennisA: although mint has a variant that uses debian packages instead of ubuntu (which are from debian to begin with) 01:34 < phogg> DennisA: No. 01:36 < phogg> DennisA: Ubuntu has a package repository. It also has an install disc which will install Ubuntu and some packages from the repo. If they provide two install discs with different sets of packages it's still the same distribution; these are different "flavors". If I make my own install disc with my own default set of packages *it's still Ubuntu*, just not an official flavor. 01:36 < jml2> DennisA, most of the software is pulled from upstream<->bugreported among distro communities, so it's not that it is like 100% fork black and white.. 01:36 < catphish> DennisA: distros like mint add new packages, and change which are installed by default, i believe official ubuntu flavors only do the latter, and of course have the difference of being ubuntu branded 01:36 < phogg> DennisA: If I add packages, alter the way the system is configured, add additional repositories and then make a new install disc now it's an Ubuntu-based distribution. 01:36 < jml2> DennisA, (upstream, meaning the internet oss projects -- outside distro projects) 01:37 < DennisA> So it's the addition of addtitional packages that makes Linux Mint a different distro 01:37 < phogg> DennisA: additional packages, removal of packages, or replacement of packages with altered versions 01:37 < Psi-Jack> DennisA: Mint uses Ubuntu repos, plus their own repositories. 01:38 < DennisA> Thanks 01:38 < AOL_> also creates a repository of fanboys 01:38 < jml2> DennisA, most of *buntu distros are derivatives with "cake icing" -- where repositories on those derivatives is 99% exactly the same repotories for the official Ubuntu.. 01:38 < Psi-Jack> The ones in their own repos overrides what's in Ubuntu's. 01:38 < phogg> some of them are little more than different wallpaper and default GNOME theme 01:38 < bls> DennisA: I'd say it's more about organization structure than anything else. you can run the "mint extras" like MATE or Cinnamon on ubuntu if you wanted 01:39 < jml2> DennisA, a lot of such derivative projects die out after time, because no communities build around them.. or its some lame "Hannah Montana" thing jut for the gigs... 01:39 < bls> or elementary extras like pantheon 01:39 < phogg> the irony is that mint has sort of achieved Debian's original mission of creating a universal base distro 01:39 < DennisA> phogg: How so? 01:39 < phogg> DennisA: How what? How is it ironic? 01:40 < catphish> didn't debian achieve that quite nicely 01:40 < [R]> isn't it ironic 01:40 < [R]> don't ya think 01:40 < jml2> phogg, the GNU still has you 01:40 < DennisA> phogg: No, what's a "universal base disto" mean? 01:40 < phogg> catphish: Not as intended, otherwise Ubuntu would use Debian repos. 01:41 < catphish> phogg: good point, it's interesting that they don't 01:41 < phogg> DennisA: rather than require every new distribution to re-invent everything from scratch the idea was to provide all the boring plumbing stuff and let new distributions merely add the parts that differ, then share the work of maintaining the lower level parts. 01:42 < catphish> while other distros are happy to use ubuntu's repos 01:42 < phogg> catphish: that's just mind share. 01:42 < jml2> DennisA, you know Linux is here to stay when even the evil empire MS has just released their own Linux distribution 2-3 weeks ago :)) tehehehe 01:42 < jml2> DennisA, and true too 01:42 < phogg> catphish: most "distros" could be and should be based on a common base like that, very few are run by people who want to re-invent the lower layers 01:43 < DennisA> jml2: ha ha 01:43 < bls> this is also why there's no good answer to "what's the most X distro?", because 99% of anything you'd want to actually use are based on 2-3 base distros 01:43 < triceratux> DennisA: https://www.maketecheasier.com/how-ubuntu-based-distros-differ-from-ubuntu/ 01:43 < phogg> jml2: that's an exaggeration. The MS "distro" isn't what you would call a Linux distribution at all. A kernel and some minimal stuff around it strictly for use with one embedded device. 01:44 < jml2> phogg, the Debian's universal OS is also an exxagerration :)) even though I use debian, I find it kind of lame they use this title in their documents... 01:44 < jml2> phogg, lol 01:45 < jml2> phogg, but again the GNU has you :P 01:45 < jml2> the GNU will ALWAYS have you no matter what! 01:45 < phogg> jml2: It's not an exaggeration; it was a founding goal of the project. Whether it has been achieved is another matter. 01:45 < jml2> phogg, GNU man! 01:45 < phogg> jml2: GNU was never about making a universal OS, just a Free one. 01:45 < catphish> does gnu have its own full working os release? 01:45 < lupine> just about 01:46 < lupine> and Debian's universality is about hardware, not users 01:46 < phogg> catphish: depending on how you define full and working (and how you define "its own") 01:46 < [R]> catphish: depends on your definition of "working" 01:46 < f91w> Does Debian still have other kernels besides linux? 01:46 < lupine> turns out microkernels are *really hard* 01:46 < jml2> catphish, they're working on it XD lol "guix os" or something like that.. it will be using its "own package manger" with the similar naming of guix*** .. 01:46 < lupine> debian/kfreebsd,etc? 01:46 < [R]> lupine: that's what she said 01:46 < lupine> think so 01:46 < f91w> i thought they discontinued that 01:46 < lupine> [R]: ITYM moaned 01:46 < catphish> i was thinking of an operating system called "gnu" that one can install and use 01:46 < jml2> catphish, (and i've tried it out as well back some time ago, it's horrid to install) 01:46 < phogg> catphish: The FSF supports a small handful of Linux distributions which are sufficiently Free for its liking. Their own OS the HURD also exists and can be run, but it's not as complete as you might want. 01:46 < revel> Guix SD. 01:47 < christianbundy> (SD stands for system distribution) 01:47 < revel> Yep. 01:47 < christianbundy> (and "Guix" is pronounced "geeks") 01:47 < jml2> phogg, they're making their development so they can bind hurd to it later.. for now they are using a fork of the linux kernel 01:47 < phogg> revel: fun story about guix... I eventually got around to installing it. It was refreshingly non-hand-holding. And then it choked trying to start X with my laptop's graphics card. 01:47 < lupine> I recall getting an email from stallman asking us to switch from debian to gnewsense 01:48 < revel> Hehe. 01:48 < lupine> obviously, we said no 01:48 < revel> lol 01:48 < revel> lupine: What? Why? 01:48 < phogg> lupine: Which "we" is this? 01:48 < revel> How'd you get in that situation? 01:48 < bls> phogg: is it too soon to start claiming it has execellent documentation? :P 01:48 < lupine> some shitty ISP I used to woirk for 01:48 < catphish> i pictured gnu as an os like bsd, a unified piece, wasn't sure if it existed 01:48 < lupine> gnewsense didn't support all the hardware in the dedis we rented 01:48 < phogg> bls: The documentation definitely exists and the parts that exist are good. 01:49 < lupine> (we published a downstream of debian to aid in self-hosting various common things) 01:49 < catphish> their opening line is "GNU is an operating system" 01:49 < lupine> but now I work elsewhere and don't use that downstream distro on principle, so 01:49 < christianbundy> Are there design/marketing folks working on GNU/FSF? I love the ideas behind the project but it doesn't seem like it's really marketed as an actual product for non-technical users. 01:49 < phogg> catphish: In this case they mean "an entire operating system" and not just a kernel. GNU was complete-ish apart from the kernel by around 1990. 01:49 < bls> I've played with nix and guix (just the package managers) but using them felt too much like the 90s and/or gentoo and I don't have time for such a thing 01:50 < christianbundy> lupine: which principle? 01:50 < lupine> general principle 01:50 < lupine> clue's in the name 01:50 < revel> bls: I don't know, Gentoo feels a lot simpler to me than {N,Gu}ix. 01:50 < phogg> bls: it's hard to unlearn everything you know and then learn guile and *then* learn their configuration system. 01:50 < revel> Though maybe that's because I only use the former full-time. 01:50 < jml2> catphish, RMS calls GNU to 100% alias the term Linux 01:50 < pnbeast> christianbundy, did you ever hear of Bill Hicks? He had an interesting take on marketing people. 01:50 < catphish> phogg: that's exactly what i thought they meant, but they don't seem to actually provide the operating system as a whole 01:50 < revel> They use their own scripting language for *everything*, don't they? 01:51 < lupine> stallman is the premier intellectual of our generation 01:51 < phogg> catphish: again, they do: In the form of several Linux distributions they have the complete OS. 01:51 < lupine> well, of the generation before my generation, but he's still going strong 01:51 < phogg> catphish: and with the HURD as well, although as I said it's less complete 01:51 < bls> revel: I meant more in the sense of having to compile things myself, fix broken builds, etc 01:51 < christianbundy> lupine: not a GPL fan? 01:51 < lupine> ? 01:51 < lupine> I am strongly in favour of the GPL 01:51 < lupine> although copyfarleft is intriguing 01:51 < phogg> "intellectual" is not a pejorative. 01:51 < revel> bls: Doesn't nix let you use binary packages? 01:51 * jml2 https://www.youtube.com/watch?v=B2vcvZt8JdA -- recent interview with RMS 01:51 < lupine> ohhhh, of course 01:52 < lupine> americans think intellectualism is bad 01:52 < lupine> yeah, I was using it as a compliment 01:52 < bls> revel: it does now; didn't at the time I trialled it 01:52 < lupine> what a world this is 01:52 < catphish> phogg: i think i understand, so gnu consider gnu with a linux kernel to be "gnu"? 01:52 < revel> Ah, okay. 01:52 < phogg> catphish: yes 01:52 < revel> catphish: GNU/Linux. 01:52 < revel> :D 01:52 < lupine> more precisely, they consider it to be gnu "plus" linux 01:52 < phogg> that is indeed why they call it that 01:53 < lupine> it's almost like they use words with intent or somethinf 01:53 < phogg> catphish: from Stallman's POV the actual software you run (the applications such as e.g. bash) are mostly GNU, so the Linux part is incidental. 01:53 < catphish> i thought it would be more interesting for gnu (without the plus anything) to be a downloadable os in its own right (including hurd) but i guess that's not an immediate goal 01:54 < phogg> catphish: It was and still is a goal and it exists, it's just not called GNU. Try Debian GNU/HURD on for size. 01:54 < bls> catphish: it is and has been a goal, but hurd developement stalled, so this is a compromise on that goal 01:54 < lupine> I recall downloading and running gnu/hurd a while ago and being reasonably unimpressed 01:54 < catphish> i see 01:54 < jml2> catphish, according to RMS yes that's correct --- see their gnu.org site XD 01:54 < jml2> catphish, "We recommend installable versions of GNU (more precisely, GNU/Linux distributions) which are entirely free software. " 01:55 < phogg> lupine: there's nothing impressive about it unless you are really into settrans and the things it lets you do in terms of system architecture 01:55 < phogg> it is really, *really* elegant 01:55 < lupine> I'm mostly into freedom from capitalists 01:55 < lupine> linux is, unfortunately, being captured 01:55 < lupine> it's a slow process, but they are definitely making progress 01:55 < bls> lupine: the problem is that the GPL didn't include anti-commercialization clauses 01:55 < lupine> right, hence copyfarleft 01:56 < jml2> RMS' opinion on MS' sphereos can be seen right here -> https://www.youtube.com/watch?v=Kqp9LIl5MH4 interesting.. 01:56 < catphish> on the contrary, gpl is designed to allow commercialization afaik 01:56 < lupine> yeah, it's a serious failing 01:56 < lupine> it makes the commons available to exploitation 01:56 < catphish> is it? without that allowance, i can't see many people using it 01:56 < bls> so people that want software freedom *and* oppose capitalism throw around terms like theft 01:57 < lupine> catphish: success or failure is measured against goals that are implicit to our worldviews 01:57 < catphish> for me, if i couldn't sell free software, i wouldn't contribute to it 01:57 < catphish> but i'm sure there's a lot of views on the subject 01:57 < joel1nux> So, I have a Logitech f310 gamepad that shows up as /dev/input/event22 but I can't get higan to accept it as a key setting, despite claiming it accepts gamepads...any help? 01:57 < lupine> when I say failure, I mean that it doesn't help to free people from capitalist exploitation 01:58 < phogg> lupine: I don't really see how you could do a copyfarleft with software very well. 01:58 < triceratux> lupine: it allows genuinely free software to benefit from the development activities & contributions of capitalists. the gpl allows the commons to exploit the corporatists 01:58 < lupine> when other people say success, they mean it enables capitalist exploitation 01:58 < bls> there are people that want software freedom, but only for individuals and other open source projects, not commercial entities 01:58 < lupine> phogg: it's pretty easy 01:58 < jim> catphish, (the usual line is) you don't have to sell it, but you could offer services around it 01:58 < lupine> certain CC licenses meet all the requirements 01:58 * jsgrant remembers being 15 as-well. 01:58 < jsgrant> Le Edgy Evil-Capitalism. 01:59 < phogg> jsgrant: I don't think any of the participants in this conversation are at or below that age 01:59 < lupine> nothing edgy abou tit 01:59 < lupine> it's simply the lived experience of billiona 01:59 < catphish> jim: well really those things are the same, at its absolute simplest, you provide the service of making a physical copy of it for someone 01:59 < phogg> jsgrant: It is doubly insulting to *imply* that someone's point of view is naive and unworkable. It's better to say it directly and allow honest disagreement. 01:59 < lupine> another name for copyfarleft is copyfair 02:00 < jml2> lupine, I invented copylow... it's a hybrid of opensource and free concepts.. 02:00 < lupine> what's the USP? 02:00 < phogg> lupine: it's a good idea, but I'm dubious about the mechanics. Maybe for non-software things it will work out. 02:01 < lupine> software is the main battleground, so one has to hope it will work there 02:01 < jml2> gotta get back to my wordpress excuse me.. 02:01 < jsgrant> phogg: Yeah, I don't think my second/follow-up statement was implying anything. 02:01 < lupine> "run away! run away!!!!!" 02:01 < phogg> lupine: if it can provide a better solution than AGPL that would be something 02:01 < lupine> I don't see that the two spheres overlap much 02:02 < phogg> jsgrant: I would certainly say it was more implication than anything. 02:02 < lupine> I'm not familiar with an agpl-alike copyfarleft, but it would be interesting to see 02:02 < nohop> Hey guys. This is probably a dumb (and not spcifically linux-related) question. I'm using this modbus tcp server, and I'm logging all connections and actions to the syslogger. Now, I want to log when a host disconnects, too. Is there a way to retrieve the remote address from a simple int fd; ? (Like how getsockname() would give me the local address?) 02:03 < phogg> lupine: I don't think anything different would be required; since AGPL is designed to thwart people who run services that sidestep the GPL's distribution clause anything that requires a license for mere *use* neatly solves the problem. 02:04 < lupine> nohop: see getpeername() 02:05 < jsgrant> phogg: Eh, if it was implying it was so self-evident what my view based on that -- that it's more of technicality. 02:05 < lupine> phogg: AGPL permits commercial entities to run services as long as they make the code available 02:05 < jsgrant> if it was more implying* 02:05 < nohop> Oh! Pfff. Thanks, lupine :) 02:05 < lupine> a copyfarleft AGPL would forbid that 02:05 < nohop> Looks like I can s/getsockname/getpeername/ then :) 02:05 < lupine> yup 02:06 < bls> that's still not going to solve the monetization problem 02:06 < lupine> you say it like it's a problem 02:06 < bls> to a lot of people, it is 02:06 < lupine> but copyfarleft is actually designed to solve the monetization problem 02:07 < lupine> it allows you to bleed capitalist entities (i.e., where the capitalist scrip is) for money in return for using stuff 02:07 < nohop> lupine: Funny. I suddenly realised that there must be a function for this. In the past I've actually 'manually' kept track of things like that. 02:07 < nohop> *idiot* 02:07 < lupine> it's rare I do C-level sockets programming, but I'm familiar with being able to get the remote address via some call or another 02:09 < nohop> I'm using a ModBus tcp server that is very very minimal, so I have to :) 02:09 < jsgrant> lupine: You're referencing http://wiki.p2pfoundation.net/Copyfarleft ? 02:09 < lupine> actually the venture communism manifesto, but I think that page works off it pretty hard 02:09 < sauvin> Like this? https://stackoverflow.com/questions/3060950/how-to-get-ip-address-from-sock-structure-in-c 02:10 < bls> all the larger companies I've worked for have been accused of theft from the open source community. being in compliance with licenses isn't enough to the accusers 02:10 < lupine> reasonably so 02:11 < jsgrant> bls: Well to be fair, it's reasonable to question -- but hard to verify; So comes into conflict often. 02:12 < jsgrant> And how many real-big GPL violation lawsuits have there been to scare away bad behavior to say it's a "big deal" to make explicit. 02:12 < jsgrant> ?* 02:12 < lupine> there's been a couple 02:12 < nohop> lupine: Apr 29 20:12:02 greenian UniPLC[4496]: Connection closed: 192.168.10.30 02:12 < nohop> Thanks again, works :) 02:12 < lupine> <3 02:12 < lupine> go forth and multiply 02:13 < nohop> As always. 02:13 < jsgrant> lupine: Any with a HUGE payoff from offender? 02:13 < lupine> perhaps the USB stick one 02:13 < bls> jsgrant: this wasn't a "I think they're violating OSS licenses", this was "they're following the licenses to the T. but they're making tons of money while doing it. that money should be going back into the OSS projects/community" 02:14 < lupine> http://gpl-violations.org/news/20130626-fantec_judgement/ 02:14 < jsgrant> bls: Ah. 02:14 < lupine> bls: right, this is what copyfarleft is for 02:14 < jsgrant> Yeah, that's a bit different. 02:14 < lupine> if you're not happy with your software being used for great profit that does not benefit you, you should copyfarleft it 02:15 < jsgrant> So that's not '4 freedom' compliant then, I guess. 02:15 < jsgrant> ?* 02:16 < triceratux> lupine: folks are free to do that, & then their software cant & wont be incorporated into gnu/linux/x11. look at kutvonen microemacs for example with its anti-profit license 02:16 < bls> lupine: fully agree. and the flip side is also true. if you don't care and CC/PD your stuff, no one should be able criticize thatA 02:18 < lupine> bls: one can certainly criticise it 02:18 < jsgrant> Wasn't Linux originally not-for-profit before GPL adoption? 02:18 < lupine> but it's certainly within peoples' individual rights to be exploited 02:19 < lupine> one could almost say the whole system is set up for that 02:19 < jsgrant> lupine: Depends how you define exploited -- pretty sure you know that though. 02:19 < lupine> I'd use it in the normal sense 02:19 < lupine> i.e., not provided with fair recompense for your efforts 02:20 < jsgrant> What/who defines fair? 02:21 < triceratux> lupine: there just arent that many options its true http://www.newsweek.com/senior-citizens-japan-are-committing-crimes-because-they-want-go-jail-854842 02:21 < Zeioth> The witcher 3 running on linux at 140fps -> https://www.youtube.com/watch?v=KMZLUlCDaw8 02:21 < jsgrant> I'm sure I'm sounding difficult, but it's actually really hard to come with a unified agreed upon mean between active seperate parties. 02:22 < lupine> it's certainly a problem when you have mutually antagonistic groups making up a society 02:22 < lupine> one can generally argue for the eradication of all but one of the groups as a means to solve the problem 02:22 < jsgrant> Also, find it a bit odd when there's no doubt more people in the history of FOSS getting actively payed for their work; And only a few years of such technologies actively making such a thing "practical". 02:22 < bls> and how much should I charge someone that looks at, but doesn't use, my code and gains some insight? 02:22 < lupine> ideally, you wouldn't charge at all 02:23 < lupine> you want to develop the commons to the point where all your material needs are met 02:23 < jsgrant> Needs vs wants. 02:24 < jsgrant> There's certainly enough in the commons to meet my needs to survive. 02:24 < lupine> funny, I'm not aware of a surplus of commons-accessible food and housing 02:25 < lupine> it's easy to lose sight of the basics when you're immersed in software for a living 02:25 < triceratux> lupine: go to berkeley & stand in the food not bombs line for a few years 02:25 < lupine> unfortunately, I can never travel to the US 02:25 < lupine> well, I say unfortunately, I don't really mind 02:28 < sauvin> Why? 02:28 < lupine> it has the death penalty 02:29 < [R]> plan on commiting crimes? 02:29 < lupine> most crimes are of passion 02:29 < [R]> rofl 02:29 < lupine> death-penalty-garnering crimes, I mean 02:29 < lupine> it's simply not sensible to subject yourself to such a jurisdiction willingly 02:29 < lupine> I'd like to visit japan, but I won't unless they drop it too 02:30 < sauvin> Some states have the death penalty and others don't. 02:30 < lupine> by the by 02:31 < lupine> and there is a federal death penalty anyway 02:32 < lupine> there is no room for this in a civilised society 02:37 < AOL_> i think sometimes life in prison until death could be harsher punishment than execution 02:38 < AOL_> lupine: Japan is a nice place to visit, why would you let their law on capital punishment stop you? 02:39 < [R]> AOL_: because apparently he plans on commiting horrific crimes 02:39 < AOL_> they also had a eugenics program from the 40s up until the 90s i believe 02:40 < autopsy> LOL 02:40 < AOL_> heh 02:40 < AOL_> why doesn't he just join the US military then and ask to be stationed in Okinawa? 02:42 < AOL_> https://www.stripes.com/former-kadena-worker-reveals-gruesome-details-of-okinawan-woman-s-death-1.453763 02:43 < AOL_> sickening 02:43 < lupine> I'd have to go to the US first for that 02:43 < lupine> transitive liability is a hell of a drug 02:43 < lupine> anyway, i've made it clear that planning is no part of this 02:44 < [R]> i dont sit in fear everyday 02:44 < [R]> "omg, im accidently going to murder someone" 02:44 < lupine> neither do I 02:45 < lupine> principally because I'm nowhere where such a mistake would lead to my death 02:45 < sauvin> I have a very dangerous temper, *have* had it for six decades. Only time I ever saw the inside of a police station was when bailing out a buddy for DUI. 02:45 < sauvin> lupine, you can get killed just crossing a street. By your line of reasoning, it's unreasonable to live anywhere there's motorised or horse-drawn vehicle. 02:46 < [R]> lol 02:46 < lupine> sauvin: accidental and premediated killings are rather different 02:47 < lupine> but congratulations on being a smartarse, I guess 02:47 < sauvin> Chances of getting killed crossing the street are probably far higher than the chances of being murdered unless you live in a hot spot. I'd avoid the Middle East and southeastern Europe, maybe. 02:48 < ayecee> or if you roll with gangs 02:48 < [R]> sauvin: no, hes worried about him murdering someone 02:48 < lupine> I've been to morocco, which officially has a death penalty, but hasn't used it for decades 02:48 < Sveta> in one book on psychology of driving, the author alleged that a yawning driver is more dangerous than a driver who is loading a handgun 02:48 < lupine> (despite having > 100 people officially on death row) 02:48 < Sveta> because yawning drivers just kill so many more people than drivers with guns 02:49 < ayecee> omg.. i yawned not 15 minutes ago! 02:49 < Sveta> I found that reasoning flawed... 02:49 < sauvin> Point is, you're probably afraid of the wrong things. 02:49 < lupine> anyway, accidental deaths happen everywhere. state-sponsored premeditated deaths only happen in a subset of the world 02:49 < Sveta> ayecee, were you driving? 02:49 < ayecee> thankfully no! 02:49 < sauvin> Sveta, it's not flawed, but it may be just a tad distorted. 02:49 < lupine> avoiding that subset is sensible 02:49 < ayecee> that was a close one 02:49 < Sveta> ayecee, good 02:49 < sauvin> lupine, do you know what's meant by "mathematical expectation"? 02:49 < lupine> funny how people normalize this idea that they may be sentenced to death at any time 02:49 < lupine> sure 02:49 < sauvin> Apply it. 02:50 < lupine> I already did 02:50 < lupine> it's a low-probability event I can avoid with no negative impacts on myself 02:50 < sauvin> No, you didn't. You're predicating your odds on unusual conditions. 02:51 < lupine> I'm explicitly acknowledging that the odds are only slightly north of 0 02:52 < pnbeast> When you look across the observable universe, it seems pretty likely that life is an unusual condition. 02:52 < lupine> well, drake has some opinions on that 02:52 < ayecee> a terminal condition too 02:52 < pnbeast> Yeah, that. :( 02:52 < sauvin> Terminal for life as we understand it, yes. :D 02:52 < lupine> it has very little bearing on my day-to-day decisions though 02:52 * pnbeast yawns and presses down on the throttle a little harder. 02:54 < triceratux> lupine: no negative impact ? youre missing out on a great deal of community farmed polenta & kale served by credentialed anarchists. & theres plenty of folks youll be standing in line with wholl be executed before you 02:55 < AOL_> xD 02:55 < lupine> I have negative interest in meeting anarcists 02:55 < lupine> bunch of nutcases, the lot of htem 02:55 < AOL_> > credentialed anarchists 02:55 < lupine> any kind of anarchists 02:56 < lupine> "yeah in the absence of explicit power structures, implicit power structures totally won't form, because, um, magic" 02:56 < lupine> get out 02:56 < Sveta> hey pnbeast 02:56 < sauvin> Yup, magic. Uh huh. That's certainly consistent with human nature., 02:57 < lupine> I'm not 100% certain anarchists even qualify, but that's probably one for another channel 02:57 < lupine> ##gulag or whatever 02:57 < [R]> pnbeast: are you saying we're alone in the universe? 02:57 < sauvin> "We are unable to confirm or deny such an allegation at this time." 02:59 < pnbeast> [R], no, I've seen Space Ghost Coast to Coast so I know Zorak and Moltar are around. 03:00 < Psi-Jack> YES! Finally got my sick dog to eat, even a little bit. 03:01 < [R]> pnbeast: lol 03:02 < sauvin> Probably just needed to pee out some liver enzymes :D 03:02 < sauvin> What did the vet say was wrong with the dog? 03:02 < Psi-Jack> REAAALY high liver enzymes, and GI. Likely cause is something he ate, and likely what he ate, might've been a mushroom. 03:03 < uplime> ouch 03:03 < uplime> i just got out of the hospital for GI stuff 03:03 < uplime> not fun 03:03 < Psi-Jack> They've been popping up in my backyard like crazy suddenly. 03:03 < triceratux> being an omnivore has its downsides 03:03 < sauvin> You might want to have a mycologist identify the mushrooms. 03:03 < Psi-Jack> I'm not exactly sure what "GI stuff" actually is. heh 03:03 < sauvin> Psi-Jack, "gastro-intestinal". 03:04 < uplime> your stomach area kind of 03:04 < Psi-Jack> sauvin: Well, thankfully, it wasn't one of the 7 most deadly mushrooms. 03:04 < Sveta> what was it 03:04 < uplime> 7 deadly mushrooms was a good anime 03:04 < Psi-Jack> Not entirely sure. 03:04 < sauvin> Even the "not so deadly" mushrooms are just as deadly. 03:04 < Psi-Jack> Heh yeah. 03:04 < Sveta> i like the idea of bringing a mushroom to a mycologist 03:05 < Psi-Jack> Yeah. WHen I see another one pop up, I might just. They're poping up right at the edges of raked up piles of leaves,. 03:06 < uplime> does anyone happen to know if webroot is supported for wildcard letsencrypt certificates? 03:06 < sauvin> If you do nothing else, get pics, but if you think your dog is eating them, you probably DO want to know what they are. 03:06 < Psi-Jack> I know one of my other dogs definitely has eaten a mushroom or two, and he's fine. 03:06 < sauvin> Be aware that many times you can't tell one mushroom from another just by looking at it. Sometimes, the slice'n'dice chromatograph is the only reasonably sure way. 03:07 < Psi-Jack> Yep. 03:07 < Sveta> and dogs react to them in different ways 03:07 < Psi-Jack> Just that the 7 most deadly ones, have a "skirt" around the stem near the top. 03:07 < sauvin> Don't depend on that. 03:08 < AOL_> nibble a very small amount of it and see what happens 03:08 < Psi-Jack> Yeah, I know. I'm definitely no mushroom expert. :) 03:08 < AOL_> see if you go into elf land 03:09 < Psi-Jack> Naaah, Mario Bro's. Where the most deadly mushrooms exist that you walk on, jump on, spit fire at. heh 03:10 < uplime> apparently not. only DNS validation 03:10 < sauvin> "DNS"? 03:11 < uplime> for getting LE wildcard certificates 03:11 < sauvin> Oh. I got contexts mixed up. 03:11 < Psi-Jack> Heh 03:11 < elf> AOL_: get off my lawn 03:11 < uplime> my bad 03:11 < Psi-Jack> Yeah, only DNS validation can get wildcard certs with LE. 03:11 < uplime> just wanted to give the answer in case someone looked it up in their logs 03:11 < uplime> thanks Psi-Jack 03:11 < sauvin> Of course, of COURSE, clusters of mushrooms will handle their own DNS. 03:12 < uplime> depends on the hypervisor 03:12 < Psi-Jack> Gotta find each other, right? ;) 03:12 * sauvin gonna start naming his VMs after mushrooms 03:12 < uplime> hm 03:12 < uplime> mushrooms as a server name scheme isn't a bad idea 03:13 * AOL_ uproots the mushrooms before getting off the elf lawn 03:13 < sauvin> Hrm... wonder how much of a deal it would be to hook up Chantarelle with Portobello... 03:13 < AOL_> i'm going to dance with the forest nymphs 03:13 < elf> -_- 03:13 < uplime> im a little jealous of your nick 03:14 < sauvin> Oh, and can't forget the Windows machine... call it Psilocybin... 03:14 < Psi-Jack> lol 03:14 < Psi-John> don't insult psilocybin like that 03:14 < stevendale> OwO 03:14 * sauvin snerks 03:14 < uplime> oh god 03:14 < uplime> theres two of them 03:14 < Psi-Jack> uplime: Yes, my son? 03:15 < uplime> lol 03:15 < Psi-John> Psi-Jack: what linux distro does your dog run? 03:15 < sauvin> o.O 03:15 < stevendale> Puppy Linux 03:15 < Psi-Jack> Psi-John: None you would know about, dannylee. 03:15 < uplime> ah 03:19 < stevendale> I am going to go hit a hard drive with a sledge hammer 03:19 < Psi-Jack> That might damage it. 03:19 < stevendale> Good 03:19 < Psi-Jack> It might decide to open up and hit you back with its spindles, like a baws. 03:23 < jml2> triceratux, so how is the new redstar? LOL 03:23 < jml2> mwahahaha 03:23 < jml2> muscovite 03:24 < jml2> this wordpress i'm using is nut, best plugins, sexy themes.. all i need is content.. 03:24 < jml2> yep, this is gunna be great 03:27 < slondr> Anyone have opinions on systemdbus vs kdbus? 03:27 < slondr> I think distros are starting to switch to kdbus lately 03:30 < jml2> dbus is used by systemd... 03:30 < jml2> so its only 1 dbus.. 03:30 < triceratux> jml2: i got the resolved thing fleshed out a bit more. i have no fear. its quite a schism, tho. theres 16.04, 18.04, & all the rest of the distros https://paste.opensuse.org/3011521 no reason why it cant work 03:31 < slondr> yeah but kdbus won't rely on systemd anymore 03:31 < slondr> systemdbus still will for obvious reasons, but idk some people are pushing back against kdbus for some reason 03:32 < jml2> "Since version 221, systemd will automatically load the kdbus module (if available), mount kdbusfs and create the system and user bus via kdbus. A dbus1 compatibility daemon, called systemd-bus-proxyd, is spawned for each bus." 03:32 < jml2> systemd uses dbus... whether it is to use either one.. 03:38 < slondr> hm 03:39 < triceratux> systemd-resolved is all about the dbus. & just because a distro is systemd doesnt mean its using systemd-resolved 03:39 < slondr> ^ 03:46 < jml2> triceratux, not all its 3 modes though... it redundantly supports legacy, so it also supports nss-resolv via tcp/udp queries 03:46 < jml2> triceratux, ./nss-dns./ rather for that case.. 03:46 < jml2> triceratux, it also parallel-supports nss-resolv for the 2nd mode :p 03:47 < jml2> triceratux, NM talks to systemd-resolved if /etc/resolv.conf is a symlink to systemd-resolved.. 03:47 < jml2> triceratux, so those "extra name servers" to my understanding (you know the name server 192.168.1.1) do not have to be written anywhere afaict because they're told to systemd via dbus.. 03:48 < slondr> Just run `diff /bin/kdbus /bin/systemdbus` and you'll see what I mean. 03:48 < jml2> triceratux, so this is why you get 127.0.0.53 ... 03:49 < jml2> triceratux, a solution for you is if you edit your script to force a symlink to NM's resolv.conf -- NM alway writes a sane copy of the nameservers because it is the dhcp client.. 03:49 < jml2> triceratux, (/var/lib/NM*/resolv.conf -- somewhere) 03:49 < jml2> triceratux, I dunno why you don't do thi, it's basically just 1 symlink ... 03:49 < jml2> triceratux, so that you can keep the good old legacy ways... 03:50 < jml2> i won't diverge more into it XD 03:52 < triceratux> jml2: again, i dont have to do anything. it works everywhere except where it doesnt & then its easy to fix. the only thing that can go wrong is systemd-resolved crashes which is prettttty unlikely 03:52 < jml2> triceratux, funny i'm finding wordpress to be a lot harder than systemd-resolved XD 03:53 < jml2> triceratux, what about those (R) columns.. 03:53 < jml2> triceratux, you got 2 distros totally bRokened... 03:53 < jml2> hahaha 03:53 < jml2> triceratux, remember symlink /etc/resolv.conf -> to the /var/lib/NM*/resolv.conf is the easiest fix... 03:54 < jml2> triceratux, it will never fiddle between 192.1.68.1.1 and 127.0.0.53 when you use this symlink.. 03:54 < triceratux> jml2: thats just the dnsmasq install. it wont start until you systemctl stop systemd-resolved. after that youre golden 03:56 < jml2> triceratux, you're forgetting another wonderful distribution on that list.. it's quite a shame. 03:56 < triceratux> jml2: btw nobody does that. on this distro i dont have a /var/lib/NetworkManager/resolv.conf 03:57 < jml2> triceratux, it's documented.. 03:57 < jml2> triceratux, in the NetworkManager.conf page ... if you mean "stock-ready", nah you're right, no distro does this, but it is a sane configuration that works properly and predictably.. 03:57 < triceratux> jml2: what am i missing ? siduction ? 03:57 < jml2> triceratux, redstar.. 03:57 < jml2> triceratux, LOL 03:58 < triceratux> astra ftw 03:58 < jml2> triceratux, you've been punked! 03:58 < dell00> Heeheheheee 03:58 < jml2> dell00, he a muscovite 03:58 < dell00> O-O 03:59 < jml2> dell00, a person from moscow and loves the russian distributions (hint altlinux :p) 03:59 < jml2> which is not a bad distro I have to admit 04:00 < jml2> interesting openindiana is actually still maintained... 04:00 < jml2> its been such a long time since i tried opensolaris related things.. 04:00 < jml2> maybe i should try that .. time me.. XD 04:01 < jml2> tehehe 16megs/second :P 04:02 < jml2> its less than half a meg, gotta use my special jump server :P 04:02 < triceratux> jml2: i got openindiana running on a latitude from a flashdrive a couple years ago. its definitely not like linux but it can be done 04:03 < jml2> better at 1.5meg/sec .. its still so slow.. 04:03 < jml2> triceratux, oh there's usb image support here.. 04:03 < jml2> 5%.. i should of been done already.. 04:04 < jml2> 13 minutes more.. 04:04 < jml2> i usually can get it within couple minutes.. 04:04 < jml2> these clowns are not using cdn mirrors.. 04:05 < jml2> siduction looks interesting.. 04:06 < jml2> i recall you mentioning it at least 2 times XD, it must be good therefore 04:06 < triceratux> the swagarch github servers are so slow http downloads via the browsers never complete. its the first time i was forced to script a wget just to get the iso 04:06 < jml2> you muscovite 04:06 < triceratux> in my own mind ;) 04:07 * Psi-Jack pictures triceratux's mind, wondering why all the truffles and poppies and fairies. 04:07 < triceratux> jml2: siduction is just a tidy way for me to get a liveiso of sid running xfce. & yeah its surprisingly solid & they seem to be releasing more often 04:07 < dell00> Anyone here using ERC like me? 04:07 < Psi-Jack> dell00: Nope. Only people using vim. 04:08 < dell00> Dang. I'm using Spacemacs, which is Emacs and Vim combined. 04:08 < dell00> I love it. 04:08 < jml2> triceratux, it's actually the first time i've been on sid for like 5 months... despite me being on debian for over a decade.. the break-ing and fixing is pretty minimum to my surprise.. 04:08 < Roserin> nvim is nice 04:08 < jml2> triceratux, you might be able to use stock vanilla debian on testing... 04:08 < Psi-Jack> dell00: You mean EmacsOS finally got a decent text editor by incorporating vim? 04:08 < jml2> triceratux, sorry 04:09 < dell00> Psi-Jack: basically. 04:09 < jml2> triceratux, I mean "testing" for 5 months.. never "sid" -- which is more for developer/maintainers.. 04:09 < jml2> triceratux, not sid not sid!! 04:09 < triceratux> jml2: sidophobe ! 04:14 < jml2> my dedicated machine is limitted to 3mb... but still I'm saving 7 minutes mwhahaha 04:15 < jml2> its funny how some of these *bsd projects come to be, -- i recalled trying out ubuntubsd back a couple years ago.. it was rock stable when I tested it (although just in a vm) 04:16 < jml2> very interesting and good to have alternatives.. more and more is better 04:18 < lonepurplesmurf> Anyone know how ssh-keygen generates the sha256 fingerprint from /etc/ssh/ssh_host_ed25519_key.pub 04:19 < jml2> lonepurplesmurf, finger print? lol 04:19 < lonepurplesmurf> err not sure what's funny 04:19 * jml2 puts a toe print on his keyboard ....tap tap 04:20 < lonepurplesmurf> it's a hash of the public key but it's not base64 decoded first 04:20 < lonepurplesmurf> like the old md5sum one was 04:20 < jml2> ^ lol 04:20 < jml2> like the old md5sum one.. 04:21 < lonepurplesmurf> jml2 this is above your head...hush please 04:21 < jml2> lonepurplesmurf, i'm sure it is 04:21 < jml2> lonepurplesmurf, except that I don't differentiate keys by a checksum 04:21 < lonepurplesmurf> yeah..it's above your head 04:22 < jml2> lol 04:22 < lonepurplesmurf> do you understand what a fingerprint is? 04:22 < jml2> you're trying to use vocabulary out of their intended semantics XD 04:22 < lonepurplesmurf> no, that's literally what it is 04:23 < ananke> jml2: ffs, your 'lols' are obnoxious. it's not a punctuation mark, don't use it as such 04:23 < lonepurplesmurf> plus you don't know what you're talking about 04:23 < jml2> ananke, excuse me who are you? his question doesn't even make sense 04:23 < ananke> jml2: who I am has nothing to do with it. neither his question. 04:24 < lonepurplesmurf> jml2 you don't know what you're talking about...sorry 04:24 < jml2> ananke, shakespearean.. 04:24 < jml2> lonepurplesmurf, i didn't say anything.. 04:24 < jml2> lonepurplesmurf, i'm letting someone else try to help you oout on your little warrior adventure 04:24 < lonepurplesmurf> the question doesn't make sense to you because you don't understand pki 04:24 < jml2> pki? lol 04:24 < jml2> LOL 04:24 < jml2> omg.. 04:24 < lonepurplesmurf> derp 04:24 * jml2 hides 04:24 < ananke> jml2: are you retarded? 04:25 < jml2> uh I know his question and setup has nothing to do with pki 04:25 < jml2> LOL 04:25 < jml2> wow XDXDXD 04:25 < lonepurplesmurf> jesus 04:25 < jml2> the package installer genereates those keys. 04:25 < lonepurplesmurf> derp 04:25 < jml2> that's why they are "unique". 04:26 < lonepurplesmurf> old fingerprint was echo "public_key" | base64 -d | md5sum 04:26 * jml2 says go check your postinst scripts from dpkg 04:26 < lonepurplesmurf> that would do nothing 04:26 < lonepurplesmurf> now if you told me to read the source code of ssh-keygen then you'd have something 04:26 < jml2> I just did. 04:27 < lonepurplesmurf> no, no you didn't 04:27 < lonepurplesmurf> go away fly 04:27 < jml2> lonepurplesmurf, I can't, i'm stuck on debian for another 10 years 04:27 < jml2> thanks for reminding me i'm not drunk 04:27 < jml2> like you! 04:36 < ayecee> just think, you're like this without any drinks 04:39 < jml2> triceratux, this is sweet I can choose a location on antartica with this openindiana setup https://imgur.com/a/mVsXLHr XDXD 04:39 < jml2> triceratux, I always wanted to do that! 04:40 < rascul> that doesn't seem unique to openindiana 04:40 < jml2> rascul, are you a troll? 04:40 < jml2> rascul, XD --> "troll" city -> https://imgur.com/a/zMOplGR 04:40 < jml2> lol 04:41 < jml2> rascul, those devies have some sense of humour... 04:41 < rascul> oh 04:41 < jml2> currently installing it.. 04:41 < jml2> (vm) 04:41 < rascul> i'm still trying to determine if i'm a troll 04:41 < jml2> ;-) 04:42 < rascul> i don't seem to quite match this https://en.wikipedia.org/wiki/Troll 04:43 < jml2> oh it must be this- https://en.wikipedia.org/wiki/Troll_(research_station) 04:43 < jml2> kewl XD 04:43 < jml2> didn't know hahahah 04:44 < jml2> "Troll Station" -- I bet those Norgewians use irc on a frequent basis! 04:49 < weswes887> Hello 04:50 < sauvin> Zz. 04:53 < ananke> jml2: 3.36% of all your lines in this channel include 'lol', while 2.34% uses 'xd'. you have a real problem 04:54 < jml2> ananke, LOL 04:54 < jml2> ananke, 99% of your lines have % in them... 04:54 < jml2> ananke, count me more on this ! :p 04:55 < neopub> any good epub reader that doesn't like from 70s? 04:55 < jml2> neopub, calibre? 04:55 < neopub> ideally something that uses gnome/gtk proper. 04:55 < neopub> not that gnome is nice, but is the okay. 04:55 < neopub> jml2: doesn't it use qt? can't change the theme, the icons on the reader takes 20% of my screen. lol 04:56 < jml2> neopub, dunno, it does use python bindings 04:56 < jml2> neopub, yeah it does use pyqt 04:56 < pnbeast> zomg, jml2 is a loller! 04:57 < jml2> neopub, there's a "fast" mode with calibre iirc so that you can do fast page moving 04:57 < jml2> neopub, the scroll-look alike icon can be toggled... 04:58 < neopub> jml2: where? don't see anything under look and feel. 04:58 < jml2> neopub, dis!! -> https://imgur.com/a/8y9YFlf 04:58 < neopub> and to be honest, calibre just generally feels like overkill and too complicated to me. 04:59 < neopub> oh, I know that one 04:59 < jml2> neopub, that's the toggler if you have the rendering too slow 04:59 < neopub> i actually like to have separate pages 04:59 < Qatz> But Calibre jst generally works 04:59 < neopub> anyways 05:00 < neopub> anything else? tried hard to love Caliber, but it is probably just not to meant to be, it is not Caliber, it is me. 05:00 < jml2> neopub, maybe you're using an older version here I can do this with the one I have 05:00 < Qatz> It converts to what ever book format I'm using, Don't need to "love" it. It just does the job 05:01 < neopub> yeah 05:01 * jml2 is using calibre 3.15 and can do multi pages 05:01 < neopub> it is me. but it is not working mate. 05:01 < neopub> that is not my problem tough 05:01 < neopub> though* 05:01 < neopub> I just don't like Calibre. it is ugly and clunky. 05:01 < neopub> it is a very useful tool for converting books though 05:01 < neopub> and updating meta data 05:01 < jml2> neopub, https://imgur.com/a/YhnXP10 05:01 < neopub> certainly something to have around but not the best reader 05:02 < jml2> neopub, its ugly? you can blow the view fullscreen 05:02 < jml2> neopub, I find it rather pretty 05:02 < jml2> neopub, that's why I use it 05:02 < jml2> neopub, and it is imho one of the best readers out there... 05:02 < neopub> i just don't like eye candy 05:02 < neopub> i use apvlv to read pdfs 05:02 < neopub> use vim for text editing 05:02 < neopub> i3 for vm 05:02 < neopub> then calibre comes with all the blings and jlings filling up my precious screen real estate. 05:03 < neopub> you feel me? 05:03 < jml2> neopub, you were asking about .epubs.. where does pdfs come into this? XD .. i prefer something else than calibre for pdfs actually 05:03 < neopub> i was giving you a general idea of my taste for ui 05:03 < neopub> it is minimal :) 05:03 < jml2> i'm quite happy here with the latest calibre.. apparently you need to UPGRADE 05:03 < jml2> mwahah 05:04 < neopub> nvm 05:04 < neopub> found bookworm 05:04 < neopub> it is okay-ish 05:04 < neopub> needs more shortcuts but will do 05:04 < neopub> at least no bling dling all over the place ;) 05:05 < jml2> it's extremely customizable to meet your preferences... 05:05 < neopub> jml2: You see, it is not calibre, I am telling you, it is me. 05:05 < neopub> It is not a love made in heave, there is just no love. 05:05 < neopub> You feel me? 05:05 < weswes887> Hey. Anybody reccomend a distro for someone tired of arch but like to customize 05:06 < jml2> it's GUI and a few checkboxes to change... I don't see what is so hard to try it. jc... 05:06 < neopub> weswes887: archlinux? 05:06 < Disconsented> weswes887> gentoo 05:06 < jml2> if I had to tell you to type things in a ~/.setting -- then hey you can complain, but calibre makes it easy peasy to customize everything you're complaining about!!! 05:07 < neopub> jml2: How do I make the sidebar auto hide? 05:07 < Jay-Ros> sorry to bother, but would anyone be able to help with a network issue on opensuse? 05:07 < jml2> neopub, you mean the TOC? that's easy 05:07 < pnbeast> Jay-Ros, don't ask to ask. 05:07 < Jay-Ros> pnbeast - fair call 05:08 < jml2> neopub, click on the blue finger icon and it toggles it 05:08 < neopub> jml2: Also, how do I have a day/night-light/dark theme? 05:08 < Jay-Ros> I cloned an opensuse server and gave it a new eth, moved eth1 to eth0 so it's the primary and removed the secondary 05:08 < jml2> neopub, you can definitely change the background color 05:08 < Jay-Ros> but for some reason I'm now losing my route/gateway on reboot 05:08 < neopub> jml2: I managed to pull it out and close it. lol how do I bring it back? 05:09 < Jay-Ros> I've moded /etc/sysconfig/network/routes and /network/ifcfg-eth0, but it's still dropping 05:09 < jml2> neopub, when you open a .epub file, you can click on the screw driver icon, that's where you can configure those things you were talking about.. 05:10 < jml2> neopub, rather than the main window preferences dialog box.. i think that's why you couldn't spot these settings... 05:10 < pnbeast> Jay-Ros, hopefully someone familiar w/ your distro will chime in, but can you paste the ifcfg file? Did you set it to *not* be managed by networkmangler? Is systemd involved, here, someplace? 05:10 < neopub> jml2: so I dragged out the sidebar and top bar, then closed them. How do I get them back to get Screwing? lol 05:11 < neopub> never mind 05:11 < neopub> a fullscreen and out did the trick 05:11 < jml2> neopub, ctl-t 05:11 < neopub> ctl-t is for table of contents? 05:11 < jml2> neopub, y.. if there is one for that epub 05:12 < jml2> neopub, otherwise i dont think it would show one 05:12 < neopub> how do I make the icons small? 05:12 < neopub> this is my biggest problem 05:12 < jml2> just as not all pdfs have a bookmark outline 05:12 < neopub> they're massive 05:12 < jml2> only dumb lazy and incompetent authors dont include them 05:12 < neopub> even after enabling the optimize for high res they are still bigggg 05:12 < neopub> i got this book for free 05:12 < neopub> and it was written in 1970s 05:12 < jml2> well your epub book was written by a moron 05:12 < neopub> still has a TOC 05:13 < neopub> no. 05:13 < neopub> k. 05:13 < jml2> .epub wasn't around in the 1970's.. 05:13 < jml2> sounds like you opened something else entirely.. 05:13 < neopub> the book, I said, not the current file. 05:13 < neopub> anyways. 05:13 < neopub> jml2: good lolz 05:13 < neopub> but how do I make icons small? 05:14 < jml2> trivial! 05:14 < jml2> walk further away from monitor and it will look small!!!!!! 05:14 < jml2> easy !!!! 05:15 < Jay-Ros> ls 05:15 < Jay-Ros> oops :) 05:15 < neopub> jml2: dang! worked like a charm. 05:15 < neopub> You're a good and useful monkey. I am proud of you. :) 05:15 < jml2> neopub, and your books are all moronic from the 1970's.. :) 05:16 < neopub> it is an excellent read if you like good math. 05:16 < jml2> neopub, remember, .epubs weren't invented back in the 1970's while... 05:16 < neopub> as in, mathematics beyond application, and have a thing for graphs and charts. 05:16 < jml2> neopub, sgml was still too advanced for the little man 05:17 < jml2> neopub, and you may have stumbled on something way too difficult for your comprehension! 05:17 < neopub> jml2: Introduction To Graph Theory, Richard J. Trudeau, Rev. ed. of Dots and Lines, 1976. 05:17 < neopub> Still better than yours. 05:17 < ananke> jml2: ffs, you really need to stfu 05:17 < neopub> jml2: Possibly, so far it is seems pretty approachable. 05:17 < jml2> ananke, dont need to while I parse my distributions here :) 05:18 * jml2 hmm installing installing in the background ... hmmmm check check.. 05:18 < neopub> jml2: Do you have a high res screen or a CRT? 05:18 < jsgrant> lupine: Sorry about crank earlier. Btw actually skimmed a bit of the Peer-Production License & Idk man. I prefer copyleft to permissive, but think there is a "too limiting" factor and that seems to be straddling the edge. 05:19 < neopub> jsgrant: Why do you like copyleft to permissive? 05:19 < neopub> jsgrant: I like to get it free, and charge for giving. I gives you a good economic position. 05:19 < jsgrant> neopub: More changes go back upstream, from what I can tell. 05:19 < jml2> neopub, you must have long arms!! How far did you have to walk away in order to make your icons smaller XD 05:20 < jml2> neopub, it worked just like I told you it would XD congratulations chubby XD 05:20 < neopub> jml2: Sadly that almost never happens, most people just fork. Look at Google for example. 05:20 < jsgrant> Also seems less likely to fragmentation. 05:20 < jml2> neopub, wrong nick noob! 05:20 < jml2> neopub, LOL 05:20 < jsgrant> Ideologically I line a bit more there too, but think I've transcended that. 05:20 < neopub> jml2: 2 meters. I generally have to sit that far when I am feeling in appropriate content anyways. So nothing new. 05:21 < neopub> jml2: Yes, I can't stop thinking about you!! 05:21 < jml2> perhaps with a little more effort you'll figure out how to use calibre you little devil :)) 05:21 < neopub> jsgrant: Oh, I am sure, look at Blink, Webkit, and whatever Safari started. No fragmentation. ;) 05:21 < jsgrant> neopub: Forking is inheret to OSS; It's about limiting it to a maxim of usefulness & not just "becuase we can". 05:22 < neopub> jml2: I just wish it had an easy option to change themes. 05:22 < neopub> jsgrant: But how does copylift stops "just because"? merging things upstream is hard work, it is not just putting the code out there. 05:23 < neopub> jsgrant: I use lots of open source projects and getting fixes upstream can become a fulltime job, sometimes I just say screw it and maintain a fork :( 05:24 < jml2> neopub, well then you bugreport them. 05:24 < jml2> neopub, that's what most do.. 05:24 < neopub> jml2: Why no text layout for Continuous Mode? 05:24 < jml2> neopub, you must do the reportbug thing, otherwise it is anti-community-wise 05:25 < jsgrant> I'm really half-out-of-it; So hopefully somewhat clear -- but it's the aggregate effect & when you compare the features/quirks over time I'm seeing a lot more net-effect on those under 'less-permissive' licenses. 05:25 < jml2> neopub, if you gunna just do the freeloading without contributing then at least spare a 2-coin change for the rest of us :) 05:26 < jsgrant> Sometimes just being forced to as a entity to release your changes to the open-source'd depencency before releasing product to the public. 05:26 < jml2> neopub, and no, it's a misnomer that an "unmaintained" project cannot be on debian's alioth for example.. 05:26 < jml2> neopub, distro maintainers would host a copy of upstream. 05:26 < jsgrant> Maybe fragmentation is a bad modus to argue; Availability I think is more tennable. 05:26 < neopub> probs. 05:27 < sauvin> That would be "tenable", I think. 05:27 < neopub> But based on my limited observation, copyleft makes companies just run away. ;) 05:27 < jml2> that's a dumb statement, that "upstream is hard work". 05:27 < neopub> jml2: Spoke like a true sysadmin. ;) 05:27 < jsgrant> sauvin: Yeah I saw red squiggle, too 'woah' to care. 05:27 < jml2> if upstream is too much a problem to report and get things done, the upstream is merely cloned downstream.. 05:27 < jml2> neopub, pathetic 05:27 < jml2> lol 05:27 < neopub> come on. 05:28 < neopub> You know it. lol 05:28 < jml2> no that was a dumb statement actually 05:28 < neopub> but true nonetheless. 05:28 < jml2> I'll let you go back to your calibre-learning, I'll let someone else help ya on the separation of pages :)) 05:28 < neopub> In essence, it is truth that matters, and truth is a matter of opinion. 05:28 < neopub> So? 05:28 < christianbundy> I'd argue that it has less to do about efficiency and more about not allowing people to use your work to create Copyrighted Works© they can use to sue others. 05:28 * jml2 tries oindiana in vm 05:28 < jsgrant> neopub: x=x? 05:29 < neopub> christianbundy: That is a good argument. I think that is fair enough. 05:29 < jsgrant> jml2: Wat dat? 05:29 < neopub> jsgrant: JS Grant? 05:29 < jsgrant> J S Grant 05:29 < neopub> I was going to make a JS joke. 05:30 < neopub> But okay. 05:30 < neopub> x=x could be falsed given certain formalism. 05:30 < jsgrant> neopub: Should of made a Bach joke 05:30 < jsgrant> I'll be 05:30 < uplime> does anyone know how to get the gateway of a server in ubuntu 17.10? 05:31 < autopsy> uplime, can't you use the route command? 05:31 < jim> neopub, to play with someone's name, that's something I avoid 05:31 < jsgrant> neopub: 1x = 1x? 05:31 < uplime> autopsy: perhaps. im not familiar with it though 05:31 < revel> `ip route`, I think. 05:32 < revel> Or `ip r` if you're lazy. 05:32 < neopub> uplime: gateway depends on the NI, no? 05:32 < uplime> neopub: right 05:32 < uplime> but i don't see why that matters 05:32 < neopub> So you could have more then one! 05:32 < neopub> ifconfig? 05:32 < jml2> "than one" 05:32 < jml2> "then!=than" 05:32 < neopub> than one* 05:32 < neopub> fuck. 05:32 < jml2> good 05:33 < neopub> I hate when people get that wrong. 05:33 < neopub> Not sure what happened here. 05:33 < uplime> neopub: i know i can more than 1 05:33 < neopub> i can more than 2. ;) 05:33 < neopub> try ifconfig? 05:33 < uplime> can have* 05:33 < uplime> why? ip route shows it 05:33 < jsgrant> Lol, I think libinput doesn't support this touchpad since Fedora 28 Pre update -- keep trying to reach for it. 05:33 < neopub> cool. 05:34 < neopub> jsgrant: Use Arch. Don't be a Noob. 05:34 < jml2> neopub, you're too new to Linux to call other people noobs.. 05:34 < jim> neopub, please watch the f-bombs (and other language) while you're here 05:34 < jsgrant> neopub: I was for awhile on my dev-box, but back to NixOS 05:34 < neopub> uplime: ip is a good command, ifconfig is not. if ip does it, and ip exists, use ip. 05:34 < neopub> jml2: no. 05:34 < jml2> ye 05:34 < uplime> neopub: yes im aware 05:34 < neopub> jsgrant: NixOS sounds promising, at least one of my tools is packaged for it. 05:34 < jsgrant> Makes Ansible look banal/boring 05:34 < neopub> so i might consider it. 05:35 < jml2> ifconfig is as well being obsoleted... suggesting others "ifconfig" tells me you're not ready for using "ip" which everybody should be now using.. 05:35 < revel> neopub: What does using Arch have to do with not being a "noob"? 05:35 < jml2> neopub, and that's not a good thing.. 05:35 < jml2> distros are moving ifconfig out of their default installs.. 05:35 < jsgrant> revel: Memes 05:35 < neopub> revel: if you have used just about any other distro, you realize that at some point the upgrade is going to screw you over. 05:35 < jsgrant> New Gentoo 05:35 < neopub> It always happens. 05:35 < jml2> neopub, and that is particularly noobish. 05:35 < jsgrant> Newer I guess 05:35 < neopub> jml2: no. 05:36 < jim> well it's just that ifconfig is old, they found certain problems and limitations with it, so someone came up with ip, and soon after, they deprecated net-tools (the package ifconfig and route are in) 05:36 < jsgrant> Is Gentoo still a meme on 4chin? 05:36 < jml2> neopub, yep. try installing redhat, centos, centos, and other distros, you'll see it's deprecated.. 05:36 < revel> neopub: "the upgrade"? You mean `pacman -Syu`? 05:36 < jml2> heck even debianners are saying to use ip 05:36 < neopub> jml2: I know. which part of " ip is a good command, ifconfig is not. if ip does it, and ip exists, use ip." do you not understand? 05:36 < jsgrant> 'ip address' works fine; So used to ifconfig though 05:37 < jml2> neopub, you not reading. "ifconfig' is not available on distro installs.. 05:37 < jml2> neopub, so asking to try ifconfig doesn't make sense 05:37 < neopub> revel: I mean, when I did a major upgrade with Fedora, Ubuntu, and you name it, at some point they always broke my system. Arch never did. 05:37 < neopub> Rolling release is the tits. 05:37 < jml2> or holds bearing for the near future when all distros are likely to drop it.. 05:37 < neopub> Brace it. 05:37 < jsgrant> jml2: Well a lot of time you get it on a live-medium there's just no reason to 05:37 < revel> It's "embrace" 05:37 < neopub> NO. 05:37 < neopub> brace it. 05:37 < revel> Unless you meant" brace for impact" 05:37 < neopub> Brace yourself for tits. 05:37 * jsgrant wishes more distro installers/live-disks would ship tmux though 05:37 < neopub> You don't get it. 05:37 < flopwiki> iirc, my previous install of centos still used ifconfig 05:37 < jml2> funny is he's giving bad advice and calling other people noobs... 05:37 < revel> Yes, I don't. 05:38 < neopub> adios. Me gotta go. 05:38 < revel> Smooth. 05:38 < jsgrant> neopub: o/ 05:38 < jml2> neopub, adiot to you too and smarten up 'em ip skills XD 05:38 < jml2> ./*adios*/ rather 05:38 < blaztek> jsgrant: me too 05:38 < jsgrant> jml2: Lol - if that was a true accident, that's pretty neat 05:38 < jml2> oops my typies.. 05:38 * neopub walks into the sunset 05:38 < jml2> jsgrant, I hope that it was.. did it look bad? :/ 05:39 < jml2> neopub, sorry about that.. 05:39 < jml2> neopub, typo. 05:39 < jim> neopub, here's something that might help http://inai.de/2008/02/19 05:40 < jsgrant> a priest, a kohen, and a neopub walks into a sunset 05:41 < jsgrant> jml2: That was like a Dexter's Labism; neopub you aaaaadiot! 05:41 < jim> jsgrant, please don't do namecalling 05:42 < jsgrant> jim: I wasn't; I was saying what it looked like to me & wasn't sure if it was subtle jab or actual mispelling 05:42 < revel> I think he was just making a joke. 05:42 < jim> I was responding to this: neopub you aaaaadiot! 05:42 < revel> In reference to a cartoon boy getting mad at his big sister. 05:43 < revel> And yelling something like that. 05:43 < jml2> revel, I feel pretty bad when I mock retards... so let me have a little sniffles... that was a sincere typo.. 05:43 < jsgrant> revel: ++ 05:43 < revel> lol 05:43 < dell00> Anyone here an expert with CMake? 05:43 < dell00> Or is there an IRC channel for CMake? 05:43 < jsgrant> dell00: #cmake 05:44 < jim> I know I have it... 05:44 < jsgrant> No Idea how active they are 05:44 < jsgrant> probably safer bet regardless 05:44 < revel> 241 people, at least. 05:45 < revel> "checking whether we are using the GNU C Library >= 2.1 or uClibc... yes" 05:45 < jsgrant> jim, neopub: Well, I guess sorry it came off as 'namecalling' if it did. :^P 05:45 < revel> But which one, configure, you smartass? 05:45 < jim> and, you might as well ask here (question should be specific, and include lots of informative detail) 05:46 < jim> jsgrant, well, it would if they didn't get the context 05:46 < jim> revel, namecalling :P 05:47 < revel> I can't namecall software either? 05:47 < jsgrant> Oh, there's a tooling question! Is mdbook worth it writing a smallish text-book/instructional guide or should I just use a regular static-site generator with a theme; Right now using hugo + grav's learn theme and decent but feel like a lot of little extra work. :^P 05:47 < jim> revel, the thing is you do so at the risk of triggering the people who like it 05:48 < revel> Hey, I didn't say I didn't like it. 05:48 < revel> What was it again? Autoconfigure? 05:49 < jim> jsgrant, maybe you're best off writing it in a form that can be translated into many forms automatically 05:49 < jsgrant> jim: Well, that's why I kinda want to stick with markdown -- was going to use reST but then saw how little actually supported it 05:52 < littlepython> i have 2 website. one from a region and other from a CDN.. now i want to compare their timings 05:53 < littlepython> can i do a time curl -X GET http://website.com and time curl -X GET http://websitefromCDN.com 05:53 < littlepython> is that the best way to compare which one comes faster? 05:53 < jim> you -could- write a tool that translates markdown to hugo/grav 05:54 < c50a326> test 05:54 < jsgrant> jim: Well does that now automatically; Might do a reST to Grav -- might actually be a fun project 05:55 < c50a326> managing dotfiles in a git repo, how do you then push these into your home? best solution I can think of at the moment is bash script to create sym links 05:55 < c50a326> requiring special condition blocks for .local and .config kinda paths 05:56 < jim> littlepython, I would think that by making that comparison, your bandwidth to the two sites would figure into those measurements 05:56 < jsgrant> Idk, weird to me that this isn't a lot more competitive market. Guess Sphinx (though funnily not reST) has become little contested "king' with stuff like readthedocs 05:56 < littlepython> jim: does that mean i can find which website loads faster? 05:57 < jim> you can find which loads faster -to you- :) 05:59 < jsgrant> Eh, I guess I'll eye-up the Hugo api -- more webdevish stuff than I'd like but again eh. 05:59 < jsgrant> Certainly Grav Learn looks a LOT better than the default (maybe limitations) of what mdBook has shown 05:59 < c50a326> apparently gatsby.js is the thing now 06:00 < c50a326> with a remark plugin for markdown 06:00 < c50a326> I'm just looking at that at the moment actually 06:04 < jsgrant> c50a326: Yeah, idk, there's a ton of static-site stuff; Just find it weird there's not a real big focus on "digitally aware" text-books or even really plugin-suites and the like I've found; No trivial way to define an index, list of figures, list of example question, you'll see some auto ToC stuff but that's about it. 06:05 < jsgrant> Isn't Remark that Web Markdown driven slideshow thing? 06:05 < jsgrant> Ah, that's a different "remark"; Gotta love name reuse! 06:09 < jsgrant> Eh, I'll probably just get annoyed enough that thinks "aren't perfect" and do a NIH in 3 years or something. Been loosely considering it now, just trying not to give into that bikeshedding tick. :^P 06:09 < jsgrant> Do still think it's weird though. 06:12 < c50a326> jsgrant: a lot of people like asciidoc i think 06:12 < c50a326> guess you probably came across that already 06:13 < c50a326> pandoc seems good for converting between such formats (asciidoc, markdown, tex etc) 06:13 < c50a326> emacs org mode... yeah it's all a big mess :D 06:14 < jsgrant> c50a326: Yeah, and was even about to mention was looking into Pandoc; Which also lead me to question if I'm worrying too much limiting myself to markdown 06:14 < jsgrant> Paradox of choice stuff, yeah 06:14 < keith-man> Anyone got a good resource explaining how mmap works with the under lining file changing size? Do you need to remapp the file for each resize that happens or can you mmap much large than file? 06:15 < jsgrant> Lol, maybe I should worry about getting a few more chapters down too -- guess all of this is already bikeshedding. 06:19 < c50a326> well pandoc will out to asciidoc but not in 06:19 < c50a326> it can in gfm, commonmark and docbook 06:19 < c50a326> oh docbook isn't what i thought it is 06:19 < c50a326> I was thinking of gitbook 06:19 < c50a326> jsgrant: did you look at gitbook btw? 06:20 < c50a326> but yeah for whatever reason I feel like writing in GFM or CommonMark and then using pandoc if you need to out to asciidoc and then you can also use gitbook 06:20 < c50a326> and then you can use hugo or gatsby or what the hell ever too, or github pages and so on 06:21 < jsgrant> c50a326: Have only played with the web-portal (which has sadly been a pretty horrible experience) -- should probably try local, yeah, especially since mdbook is so heavily inspired by it. 06:21 < c50a326> that seems like the least limiting path anyways 06:23 < jsgrant> c50a326: Yeah, think that's a very fair, level-headed guidance; Ty. 06:23 < [R]> who remembers hugos house of horrors? 06:24 < jsgrant> Again, probably shouldn't even be worrying about it ... but eh, obsessive going to obsess 06:24 < jsgrant> [R]: Who remembers Hugo Drax? 06:24 < [R]> dunno who that is 06:24 < jsgrant> Moonraker 06:24 < jsgrant> Bond. 06:24 < jsgrant> James Bond. 06:25 < [R]> yeah, no clue 06:26 < jsgrant> [R]: Villain from the 11th James Bond movie; Where they ran out of ideas and sent him to space 06:27 < littlepython_> if i do curl -X GET http://d2qn5ag4ypxct0.cloudfront.net i get temporary redirect warning 06:27 < [R]> lol 06:27 < [R]> sounds old 06:27 < [R]> littlepython_: pastebin the full output 06:27 < jsgrant> [R]: 1979, I think? 06:28 < [R]> yeah... super old 06:28 < jsgrant> Think the 2nd Austin Powers parodys this a few times iirc. 06:28 < c50a326> man, github are awesome https://dotfiles.github.io 06:28 < c50a326> littlepython_: add -L to follow redirects 06:30 < quint> Can anyone explain what the advantages of gpg-agent are? Why does it keep recently used keys in memory? 06:30 < jsgrant> [R]: Just looked into House of Horrors; Before my time. We got a home-computer like late 90s. Floppies were starting to go out -- cdroms were in 06:31 < c50a326> quint: not having to re-enter your master password over and over 06:31 < [R]> jsgrant: it was so impossible to play 06:31 < [R]> c50a326: thats what she said 06:31 < c50a326> lol 06:32 < quint> hrm. I guess 06:32 < Triffid_Hunter> quint: reduces attack surface for shoulder-surfing attacks :P 06:32 < jsgrant> [R]: I heard Universal has the Rock staring in the 3-movie blockbuster next year. 06:32 < [R]> lol 06:33 < quint> Triffid_Hunter: good point there 06:34 < jsgrant> "The truth is that Hugo's House of Horrors was actually inspired after playing the original Sierra game Leisure Suit Larry in the Land of the Lounge Lizards" 06:34 < jsgrant> Would have never guessed 06:38 * sauvin finds an actual honest to $deity legitimate bug in perl6 and reports it 06:40 < CrazyTux> hello, how to remove all the traces of xfce on OpenSuse Leap? 06:41 < jim> I dunno... uninstall its packages? 06:42 < Triffid_Hunter> and purge configs, might have to clean ~ manually 06:43 < Psi-Jack> Ohhh yeah! Got my private networking up and running between multiple physical hosts. 06:43 < Psi-Jack> GRE tunnels over OpenVSwitch Bridges baby! 06:43 < CrazyTux> jim, I have uninstalled xfce through yast. When I try to upgrade Leap, it asks for installation of some xfce packages. 06:44 < Dominian> CrazyTux: there's probably some things in the upgrade that are linked to xfce packages 06:44 < jsgrant> I wish there was a standard way to remove dangling config & cache files. 06:44 < CrazyTux> probable it has not been purged of xfce entirely. 06:45 < jsgrant> Seems like such a simple thing; Guess Flatpak and the like could somwhat solve it or at least trivialize everything to one root dir 06:45 < gogeta> you must construct additional pylons 06:45 < [R]> jsgrant: what defines dangling 06:46 < Psi-Jack> Finally now I have real "public" vs "private" vm networking. 06:46 < jsgrant> [R]: Non-User data like logs, I guess. 'Nix' the config files for graphical apps you probably aren't going to git and have on multiple machines? 06:46 < [R]> what? 06:46 < jsgrant> Most cache 06:47 < littlepython_> [R]: this is the result of pastebin after including -L option https://pastebin.com/V8gnjsVb 06:47 < jsgrant> [R]: Non-user data that is relevant to the app that is removed? 06:47 < pnbeast> [R], do you need a visual? It won't be pretty. 06:47 < littlepython_> c50a326 [R]: the link im trying for is a cloud front url 06:48 < [R]> littlepython_: i dont see anythhing about any redirect in there... 06:48 < littlepython_> when i see the time of a normal url its faster 06:48 < littlepython_> [R]: this is the normal url http://www.etis.com.s3-website.ap-south-1.amazonaws.com/ 06:48 < [R]> littlepython_: ? 06:48 < littlepython_> [R]: I have included -L flag to avoid redirect 06:48 < c50a326> s/avoid/follow/ 06:49 < [R]> too much rambling... 06:49 < [R]> do you love the song rambling man? 06:50 < littlepython_> [R]: ah sorry to confuse you 06:50 < littlepython_> the question was always the CDN url loads faster than the any other url 06:50 < littlepython_> so I was thinking why cloudfront which is a CDN loads quite slow compared to that url 06:50 < pnbeast> Lawd, I was bone a ramblin' man, trying to make a livin' and doin' the best I can, then my abusive gym teacher, uh, did things. 06:50 < [R]> what you said was seomthing about redirects 06:51 < [R]> you should probably ask your real question to begin with... 06:53 < jsgrant> Okay guys, bedtime. o/ 06:54 < pnbeast> You can't make me sleep. You're not the boss of me. 06:58 < CrazyTux> I had installed xfce on Opensuse Leap. I have removed it now. I feel some xfce packages are still left. How can I remove them completely without creating any problem? 07:04 < [R]> you're not the boss of me now... you're not the boss of me now... 07:49 < c50a326> hey, this link http://dotfiles.github.io and also those "awesome" lists you get on github... it'd be great to be able to kinda scrape-grep them, you know? 07:50 < c50a326> like `some-cmd url` and it hits 1 level deep of links on the given url and greps all the text in content blocks... 07:53 < fareast> missing desktop icons kde after font dpi scaling anyone know how to fix this. 07:54 < c50a326> ah brb 07:56 < c50a326> back 07:56 < fareast> welcome back 07:56 < sauvin> Zz. 07:56 < fareast> well looks like noone can answer my question 07:57 < sauvin> Might want to check back in when the banks open in Boston. 07:59 < fareast> take me to toast get me a job. 08:45 < storge> i agree completely. 09:00 < adac> With smartctl I do get: 09:00 < adac> Please specify device type with the -d option. 09:00 < adac> how can I find out the device type actually? 09:03 < storge> -d auto 09:04 < Boyeto> is there a channel for the amd mesa drivers? 09:06 < adac> storge, smartctl -d auto /dev/vda 09:06 < adac> Please specify device type with the -d option. 09:06 < adac> /dev/vda: Unable to detect device type 09:07 < sauvin> .oO("vda"?) 09:07 < sauvin> adac, what is that device? 09:08 < Triffid_Hunter> adac: what is vda? do you mean sda? 09:09 < adac> sauvin, Triffid_Hunter df -h shows me that device name 09:09 < adac> https://pastebin.com/RkBQ4ffF 09:10 < sauvin> So, adac, what is that device? 09:11 < Triffid_Hunter> adac: vda is for virtual disks, what makes you think your virtualizer will expose hardware smart info through that interface? 09:11 < adac> sauvin, actually it is a KVM machine So I guess the name /dev/vda comes from this yes 09:11 < Triffid_Hunter> adac: https://unix.stackexchange.com/questions/145332/difference-between-sdx-and-vdx 09:11 < adac> OK I understand I need to check it on the host machine 09:12 < adac> thanks! 09:12 < sauvin> And this is why I ask stupid questions. :D 09:14 < okamis_> In docker, im running a basic rhel7:latest image, and the logname command returns "logname: no login name" How can that be resolved to show root or myuser? 09:15 < Triffid_Hunter> okamis_: try 'whoami' 09:17 < okamis_> Triffid_Hunter: whoami returns root. My problem is that some package I use uses logname so it would be nice for it to work. 09:24 < Triffid_Hunter> okamis_: log a bug with redhat then, and in the meantime maybe symlink logname to whoami 09:29 < jim> well if you can find a command that outputs the content blocks to stdout, then you just pipe that to grep 09:32 < dzaczek> hi all 09:33 < okamis_> Triffid_Hunter: Could you point me in the direction of making a bug, first timer for me. 09:35 < velix> Is there a shell tool, which shows the time expired while waiting running an app? 09:38 < MaximB> Hi, I've installed the new ubuntu and it shows browsers (firefox/chrome) processes per tab, so I see several chrome processess with 'top' , how can I combine those processes to see the overall cpu consumption of say chrome? 09:38 < janco> Hi guys. I'm on arch linux and i've a USB c type 3.1 port on my laptop. Trying to connect a dell usb C dock but it doesn't work at all. This is my dmesg output https://jancokock.me/f/75cd5/ . (I disconnected it at the end) 09:38 < janco> someone an idea how I could get this working? :) 09:39 < janco> I see some pci things happening but no screens detected and usb ports are also not working 09:39 < MaximB> interesting to know that there are type C laptops now, about time :) 09:40 < janco> Lol.. 09:40 < janco> Already for years 09:41 < janco> I've got a clevo laptop, so no dell 09:41 < janco> Would be fun if I could get it working. It looks like it,'s some driver thingy. Becuase it does detect some pci and usb things 09:42 < velix> Reformated my question: I want to display total time elapsed since the process began. 09:42 < geirha> ps can do that 09:43 < velix> geirha: I'd like to have it always on the screen. 09:43 < velix> geirha: oh gnomon can do this. 09:43 < velix> https://github.com/paypal/gnomon 09:43 < geirha> if it's part of a pipeline, you can put pv in between to get some stats on time and/or throughput 09:43 < velix> geirha: pv starts only, if there's content. Actually, I'm running psql, which queries much data from the DB. 09:44 < velix> geirha: so psql outputs nothing and pv doesn't show "waiting" time. 09:44 < velix> ah, maybe I should remove -W from pv. 09:45 < ELLIOTTCABLE> o7 all. anybody know how I can print exactly what my terminal app sends for a given key?' 09:45 < velix> geirha: pv -F %t did it. 09:46 < ELLIOTTCABLE> (tryin' to map an exotic key in Vim; but I'm not seeing any reaction to the keyboard input. tryna figure out which keys it *does* support sending.) 09:47 < hendrix> ELLIOTTCABLE: 'cat -v' and the key 09:48 < geirha> Ctrl+V then the key 09:53 < stevenm> Hey is it possible to have a poor-mans dedupe-ing filesystem? e.g. a ext4 filesystem where dir1 (e.g. 8gb) gets cp'd to dir2 but using cp -l (hard link creation)... then have something watch dir2 in case it's trying to change the existing files... and unlink them just before they write a new one? 09:55 < Sitri> stevenm: overlayfs 09:55 < stevenm> hmm i think i've heard of that before 10:00 < oiaohm> stevenm: https://lwn.net/Articles/747633/ really I know this is not for production now but it is interesting watching what xfs is up to. So subvolumes might come design possible on a lot of old file systems. 10:00 < stevenm> oiaohm, yeah I've read all about xfs getting the dedupe support 10:00 < V7> Hey all 10:01 < V7> Should I use sync command after cp or rsync commands ? 10:01 < stevenm> i'm just not sure if I ever want to use xfs tbh - I was already prepping for btrfs 10:01 < stevenm> V7, I only use it after low level stuff like a dd operation 10:01 < stevenm> not sure what proper procedure is though tbh 10:01 < V7> I've aksed this one because sometimes when cp command is over and disconnecting a usb flash drive I notice that stuuf hasn't been copied yet 10:02 < V7> So, how to check if it's really has been copied ? 10:02 < V7> moved, copied etc. 10:03 < High_Priest> V7, sync & umount should be sufficient 10:03 < well_laid_lawn> umount should let you know the usb is still bbusy 10:04 < V7> so, for example, cp 1GB /dev/sdc1 && sync && umount /dev/sdc1 whould be enough ? 10:04 < V7> would * 10:09 < oiaohm> stevenm: the way xfs is doing subvolume if it works this could be portable back to ext4. As it avoids altering file system disc structures. The method btrfs and zfs use depend on a full cow file system. B 10:19 < istevenmon> hi guys, hope someone could help me out a bit. I dd'ed a disk from an old HP server which used cciss raid controller. With the dd image I created a VM which when starts shows me grub screen but when booting it kernel panics. I assume the problem is the cciss driver, the VM shows the disks as /dev/sd* and not as /dev/cciss/c0d* 10:27 < gareppa> is 20000 a high number of damaged sectors on a hd? 10:29 < Minnebo> gareppa, i'd replace the disk anyway 10:29 < gareppa> ok 10:29 < kopper> gareppa: Yes 10:40 < pingfloyd> if damaged sectors are showing up it means the drive has run out of spare sectors 10:41 < pingfloyd> drives have damaged sectors out of the factory, but those get remapped out the gate when they LLF them 10:42 < MrElendig> when it comes to disks: if you intend to keep anything semi-important on it: if in doubt, replace it 10:46 < promach_> For https://paste.ubuntu.com/p/9WbHW9BfYM/ , is there a way to clone this 32GB SD card with only the usable space ? I mean I want sdd1 and sdd2, but not all 32GB 10:48 < MrElendig> promach_: it is called "cp" 10:48 < gareppa> to test the health of a hd, is badblocks ok? or should i run fsck? 10:48 < MrElendig> or "rsync" 10:48 < promach_> cp only copies file, it does not copy the partition information 10:48 < MrElendig> gareppa: https://en.wikipedia.org/wiki/S.M.A.R.T. 10:49 < gareppa> i'm talking external usb hds 10:49 < MrElendig> promach_: not that you really need that 10:49 < MrElendig> gareppa: smart works for sane external disks 10:49 < gareppa> smart is greyer out 10:49 < gareppa> oh, for this one was greyed out 10:49 < promach_> What I am trying to do is to clone this 32GB card to a 16GB card 10:49 < MrElendig> and you can always take the disk out of the case 10:49 < MrElendig> promach_: set up the partitions as you want on the new card then copy the data over, problem solved 10:50 < promach_> ok, let me try 10:50 < MrElendig> may have to copy the first 446 bytes over too if this is something that is supposed to boot 10:51 < MrElendig> though if it is for some sbc arm device, they don't usually rely on the mbr anyway 10:51 < promach_> you mean /dev/sdd1 2048 32767 30720 15M b W95 FAT32 ? 10:51 < dzaczek> exit 10:52 < MrElendig> gareppa: does the device have both usb and esata? 10:54 < cluelessperson> can you format a disk directly? 10:55 < cluelessperson> mkfs.ext4 /dev/sdb 10:57 < cluelessperson> it seems you can, but meh, it appears to be empty when I put stuff in it and mount it elsewhere 10:59 < gareppa> MrElendig, it's an external hd. i'm testing a number of external hds. they're just regular (sata, i think?) hds. i totally could remove them from their case, but i'm using a laptop ad it would be a loss of time 11:03 < MrElendig> get yourself a usb -> sata controller that supports smart passtrough :) 11:03 < MrElendig> handy to have in your toolbox 11:03 < MrElendig> badblocks alone is not a good way to judge drive health 11:03 < gareppa> ok 11:04 < MrElendig> badblock might be fine, but smart might show that it is on the limit of remaps and the load cycle count is 200k past old age 11:05 < jelly> aren't almost all usb sata bridges capable of that? it's just that smartctl didn't know how to deal with specifics 11:07 < gareppa> jelly, most of them do, but this one didn't 11:09 < jelly> gareppa: did you try using hdsentinel on it instead of smartctl 11:11 < jelly> it often works better with ssds, and usb-connected devices, than smartmontools https://hdsentinel.com/hard_disk_sentinel_linux.php 11:17 < wootwoot> I'm using android studio on linux and I want to carry out an update. I'm getting an error saying: "Studio does not have write access to /opt/android-studio. Please run it by a privileged user to update." I tried chmod 777 /opt/andorid-studio (since it's my personal computer), but I'm still getting the same error. How do I fix this? 11:17 < Skunky> do it as root? 11:17 < wootwoot> Skunky: yup, I'm root 11:17 < Skunky> wootwoot: does that dir exist? 11:18 < wootwoot> Skunky: ah never mind, there's a bug with android studio itself :) 11:18 < MrElendig> wootwoot: don't use the auto-update if you installed it system wide 11:18 < MrElendig> +using your package manager 11:19 < MrElendig> don't 777 11:20 < ac1d_r41n> good night everyone. Does anyone have a suggestion for a long range wifi adapter compatible with Wifi pineapple nano ? Thank you. 11:20 < wootwoot> MrElendig: the auto-update, as in the one on android studio gui? 11:22 < MrElendig> yes 11:23 < MrElendig> if you want to use that feature, then install android-studio to $HOME 11:23 < MrElendig> else rely on your package manager to update it 11:23 < cluelessperson> ac1d_r41n: what is wifi pineapple 11:23 < cluelessperson> ah 11:27 < pretodor> hi, in my terminal, when I write: 11:27 < pretodor> $ sleep 1; xdotool key a 11:27 < pretodor> and I click on the window with my text editor 11:27 < pretodor> the 'a' letter shows up after one second. 11:27 < pretodor> But when I go to 'System Settings' -> 'Keyboard' -> 'Shortcuts' 11:27 < pretodor> and I bind `xdotool key a` to Ctrl+E, nothing happens 11:27 < pretodor> Does anyone know what is causing this problem? 11:27 < pretodor> (I'm using Linux Mint 18 and xdotool v3.20150503.1) 11:28 < MrElendig> sounds like a xyproblem 11:30 < repys> I have to rsync from server1 to server3 passing throught a server2, what could it be the best approach to do that? 11:31 < MrElendig> openvpn 11:31 < repys> server1 (rsync) -> server2 -> server3(rsync server) 11:31 < MrElendig> or wireguard 11:31 < MrElendig> ssh if this is just a one-off and speed doesn't matter that much 11:32 < repys> I need to schedule this task via cron but the speed doesn't matter too much 11:32 < repys> ssh what do you mean exaclty? 11:32 < repys> do you have a quick example? 11:33 < MrElendig> if this is not a one-off then personally I would set up a vpn for it 11:34 < MrElendig> for ssh: ProxyCommand ssh -W ...... 11:34 < mAniAk-_-> repys: ssh proxycommand 11:34 < mAniAk-_-> stackexchange should have something 11:35 < mAniAk-_-> requires a new-ish ssh version though 11:35 < repys> and is this ssh proxycommand encrypted connection encrypted? 11:36 < MrElendig> ssh man page documents it 11:36 < repys> I mean, it has to pass throught the internet, is it secure? 11:36 < mAniAk-_-> yes 11:36 < MrElendig> if you use ssh as the proxy, and set up your keys sanely, yes 11:39 < ShapeShifter499> hi 11:39 < luna_> not every day you have to SSH into a Russian Debian installation 11:40 < ShapeShifter499> If I symlink to a file that has the following permissions "0 lrwxrwxrwx 1 root root" shouldn't it be fully readable by anyone? 11:40 < Triffid_Hunter> repys: rsync -e "ssh server1 ssh server2" .. 11:41 < Triffid_Hunter> repys: assuming you have appropriate private keys and maybe key forwarding 11:41 < djph> ShapeShifter499: what's the permission on the target though? 11:41 < repys> key forwarding? 11:42 < ShapeShifter499> djph: the same 11:42 < ShapeShifter499> djph: the original location's folder isn't 777 though, only the file is 11:42 < promach_> Does https://paste.ubuntu.com/p/jHc4krqXBn/ mean the death of my SD card ? 11:43 < djph> ShapeShifter499: then yes. I *think* (and it's before-coffee here) that a symlink's target permissions comes into play (so if they're -rw------- root root ; no one's gonna be opening it) 11:43 < pingfloyd> it's the target's perms that matter 11:43 < pingfloyd> the symlink's perms don't 11:44 < ShapeShifter499> djph: /somefolder/file > /original-folder/original-file If the original file in my example is 777 but the original folder isn't that matters? 11:44 < Triffid_Hunter> symlinks don't have perms, they all show up as 0777 (lrwxrwxrwx) 11:44 < pingfloyd> ShapeShifter499: of course the directory permissions matter 11:44 < ShapeShifter499> Triffid_Hunter: right but the original file is 777 11:45 < Triffid_Hunter> ShapeShifter499: then anyone can do anything with it 11:45 < pingfloyd> also makes sure the directory has execute permissions 11:46 < ShapeShifter499> pingfloyd: adding chmod +x to the folder where the original file was (not the symlinked one) worked. Why? 11:47 < Triffid_Hunter> ShapeShifter499: because you can't do stuff in directories unless you have execute permission 11:47 < ShapeShifter499> the file in question is just a let's encrypt ssl 11:48 < djph> ShapeShifter499: as long as the intended (non-root) user or group has read (and possibly execute) permissions, they should be okay. 11:49 < pingfloyd> ShapeShifter499: you need to be able to traverse the directory 11:49 < pingfloyd> that's what the execute bit does with directories 11:50 < Triffid_Hunter> ShapeShifter499: symlinks don't let you bypass regular permissions handling.. 11:50 < pingfloyd> symlinks are always 777 anyway 11:51 < pingfloyd> their perms don't matter 11:51 < pingfloyd> the target's is what does 11:51 < ShapeShifter499> so the file was ok, it was the folder it was in 11:51 < ShapeShifter499> I didn't know execute allowed you to traverse a folder 11:52 < ShapeShifter499> I thought it was just for scripts and executable files 11:52 < pingfloyd> perms have different meanings depending on whether you're dealing with a file or directory 11:53 < ShapeShifter499> more reason why I should be RTFM 11:53 < pingfloyd> think of it in this case, you need to be able to get to the file to be able to have its perms come into play 11:55 < pingfloyd> take write perm for example. on the file it determines if you can modify the file, on the dir, it determines if you can delete or rename the file. 11:57 < promach_> Does https://paste.ubuntu.com/p/jHc4krqXBn/ mean the death of my SD card ? 12:14 < c50a326> stow always starts symlinking in the dir above current dir? how to make stow always symlink starting at the home dir? 12:14 < c50a326> ah -t 12:24 < mawk> often my system freezes for like half a second, and then all windows are mixed up and I can't click anywhere anymore, I have to restart X with ctrl-alt-backspace 12:24 < mawk> how can I debug this 12:24 < mawk> I'm on KDE 12:24 < pingfloyd> promach_: what's the X problem? 12:24 < mawk> it happens randomly after such a freeze, or it happens like a third of the time when I run QEMU 12:25 < pingfloyd> mawk: how's memory usage looking? 12:25 < mawk> the current window loses its top decoration, and I can see the contents of the window behind in place of the top decoration 12:25 < pingfloyd> mawk: is the system thrashing when it happens? 12:25 < mawk> thrashing ? 12:25 < mawk> memory usage is 5.5G over 8, during the freeze I don't know 12:26 < Triffid_Hunter> mawk: sounds like kwin crashing 12:26 < mawk> yeah 12:26 < pingfloyd> probably 12:26 < mawk> I'll try to make a coredump or something next time 12:26 < mawk> to send them to the devs, I'm on debian testing 12:27 < pingfloyd> is it more like there' a problem with the windows or more of a graphical display issue? 12:27 < mawk> it doesn't look like a graphical display issue 12:27 < pingfloyd> is just window decoration transparent? 12:28 < mawk> the cursor can move, I can click, nothing looks frozen 12:28 < mawk> yeah 12:28 < mawk> and I can't click on the taskbar at the bottom either 12:28 < sveinse> I'm trying to decide using docker vs running the services on bare metal. Now, this is not a VM vs docker vs bare-metal kind of discussion. I have this small server that shall run a few services, such as a web server, and I'm trying to evaluate if I should run them through docker or not. E.g. for added security due to containment. Is the overhead of running a container in docker very large? 12:28 < pingfloyd> is there a shortcut you're accidentally invoking to change the window decoration transparency? 12:28 < pingfloyd> or gesture 12:28 < mawk> no, I don't think so 12:28 < Triffid_Hunter> sveinse: what about chroot jail? 12:28 < mawk> otherwise the window manager wouldn't be stuck like that 12:29 < mawk> last time it happened because I was doing alt-tab too fast 12:29 < pingfloyd> what do you mean by stuck? 12:29 < mawk> clicking has no effect 12:29 < mawk> the applications themselves still work, but not the desktop manager items like the taskbar 12:30 < sveinse> Triffid_Hunter: That is what docker essentially is, isn't it? Doing a manual jail requires some work, doesn't it. Especially if you're using your distro's packages for the service, right? 12:30 < mawk> the overhead is pretty much null sveinse 12:30 < mawk> namespaces live side by side, in the kernel 12:31 < mawk> overhead could come from extra security features used by docker such as apparmor or seccomp 12:31 < mawk> or could come from the NAT done by docker 12:31 < sveinse> mawk: yeah, great. I wanted to confirm that. 12:32 < sveinse> mawk: yes, NAT has overhead wherever NAT is used, so it's not dockers "fault" per se. Just a consequence of the chosen architecture 12:32 < mawk> yeah 12:35 < sveinse> Doesn't docker has some filesystem overhead? With aufs or whatever system being used to build layers and CoW? 12:35 < gareppa> bad blocks in hds are like 90% of the time a hardware èproblem, right? 12:36 < promach_> pingfloyd: it seems to me that my SD card is dead ? 12:37 < BCMM> sveinse: aufs is pretty efficient isn't it? 12:37 < pingfloyd> gareppa: pretty much 12:37 < pingfloyd> gareppa: but have tried changing drive cables? 12:37 < gareppa> no, it's a notebook 12:37 < BCMM> a bad sata cable would tend to manifest as bad blocks in unpredictable locations 12:38 < BCMM> if the bad blocks are always in the same place, blame the platters 12:38 < pingfloyd> in that case, swap that drive out 12:38 < pingfloyd> and move on 12:38 < gareppa> ok 12:38 < BCMM> gareppa: if you have data you want to recover from the drive, look at gnu ddrescue 12:42 < sveinse> Do you have any SMART data available? 12:42 < c50a326> hey in my dotfiles I have nvim/.local/share/nvim/site/autoload/plug.vim and then when I run nvim .local/share/nvim also gets filled with a bunch of other stuff 12:42 < gareppa> it's strange, smart test says the hd is ok, but i have like 20k bad sectors,mh... gonna swap that out anyway, but are there reliable ways to somewhat predict hds failures? 12:42 < sveinse> gareppa: You ran the extensive self check? 12:42 < gareppa> no 12:42 < c50a326> how do I specify which dir to symlink, like how leafy, you know 12:42 < c50a326> or the file even 12:43 < jhodrien> gareppa: pastebin the full "smartctl -a" output for the disk. 12:44 < ananke> gareppa: 'reliably predicting failure' is a pipe dream 12:44 < gareppa> eheheh 12:44 < jhodrien> Good indicators of some types of failures. 12:44 < jhodrien> That's the best you'll manage. 12:44 < ananke> gareppa: smart checks and results are as good as it gets 12:45 < jhodrien> Yep, paired with scheduled poking of the disk whether that's via SMART of not. If you don't read from a sector, SMART can't know it can't read from it. Just looking at the SMART values isn't enough on its own. 12:47 < gareppa> jhodrien, https://hastebin.com/qicaqaheyu.sql 12:47 < sveinse> yeah, sometimes the extensive self check is capable of discovering that some secors are bad, sometimes not. And the most extensive selfcheck takes time. 12:48 < jhodrien> Current_Pending_Sector looks a bit exciting. 12:49 < jhodrien> Current_Pending_Sector is a little bit exciting. 12:49 < jhodrien> Coo, I'm echoing myself. 12:49 < jhodrien> Does a long SMART test pass? 12:50 < sveinse> A value to keep a lookout for i 5 Reallocated_Sector_Ct. If the drive detects a bad sector it has some extra capacity to relocate into. In this case, it seems fine. So yes, the drive might not be selfaware of the bad sectors 12:50 < jhodrien> That figure looks implausible. 12:50 < jhodrien> sveinse: Again, reallocated can be deceptive. 12:50 < jhodrien> If you've got a big archive of data, you'll never reallocate a sector. 12:50 < the_drow> How do I give only one user (not root) permissions to read and write files to a folder without setfacl? 12:50 < sveinse> jhodrien: It can, but as an indication 12:51 < jhodrien> You need to write to a sector to reallocate it. 12:51 < jhodrien> the_drow: Via a group. 12:51 < the_drow> su myuser -c 'chmod u=rw /folder' doesn't do the trick 12:51 < sveinse> I've exchanged drives under warranty on SMART failures, so they're not completely useless. But not completely reliable either 12:51 < the_drow> jhodrien, can you please explain why what I did was wrong? 12:52 < jhodrien> the_drow: Who do you want to own the folder? 12:52 < the_drow> a normal user 12:52 < jhodrien> The same normal user? 12:52 < the_drow> yup 12:52 < jhodrien> mkdir /blah, chown someuser /blah;chmod 700 /blah. 12:52 < jhodrien> That not what you want? 12:52 < the_drow> but what's wrong with what I did 12:53 < the_drow> I can never remember the octal codes for permissions 12:53 < the_drow> It's like the tar cli... 12:53 < sveinse> It sais that a extended self-check on gareppa's drive takes 74 minutes 12:54 < gareppa> what are the problematic values in the smartctl output i posted? 12:55 < the_drow> jhodrien, why 700 and not u=rw,g=rw ? 12:55 < the_drow> why the first works and the second doesn't? 12:57 < jhodrien> I suspect you're only changing the user bits, and leaving group and other untouched. 12:57 < jhodrien> To be honest, other than a+rX I only ever use octal. 12:58 < sveinse> gareppa: Nothing that strikes the eyes as being immediately wrong, no 12:59 < jhodrien> You're not changing the permissions of other. 12:59 < jhodrien> So if you started of with broad permissions, it's going to leave them in place. 13:09 < BluesKaj> Hey folks 13:14 < jim> hi blues 13:14 < BluesKaj> Hi jim 13:18 < leshaste> I want to run unison and type < for each of the thousands of lines. But first I need to enter my password so I can't use yes. Is there an easy solution 13:20 < yakiza> HEllo guys i am trying to configure a startup script to check what interface is up eth1 or eth0 so i can use the proper rc.firewall file 13:21 < yakiza> but i cant see a way to do it in bash or at least i dont know 13:21 < SuperSeriousCat> Can you curl/ping or w/e where you specify interface? 13:22 < armin> yakiza: ip a | grep 'state UP' 13:22 < yakiza> armin: that worked 13:23 < yakiza> armin: is there a way to only get eth0 so i can do something like if a eq eth0 13:25 < armin> yakiza: what exactly do you want to do? check if eth0 is up or not? or get a listing of the devices that *are* up? 13:25 < yakiza> armin: i want to check if eth0 is up use rc.firewall0 if eth1 is up use rc.firewall1 13:26 < armin> yakiza: what if both are up? what if none of them is up? 13:26 < nazarewk> any idea if "automatic provisioning of desktop computers" is solved in any way? 13:26 < yakiza> you are right about that but due to my routine one of them is always up depense where i put the hard drive 13:26 < yakiza> i usually put it in 2 computers that 1 of em is eth0 and the other one is eth1 13:27 < nazarewk> i'm looking for tools i could use to configure machine i'm working from 13:27 < nazarewk> cba installing arch linux manually every time :P 13:28 < armin> yakiza: for dev in eth0 eth1; do if ip link show $dev 2>&1 | grep 'state UP' >/dev/null 2>&1; then echo "$dev is up"; fi; done 13:29 < yakiza> armin: i am now leraning bash thats why my question might seem a bit dummy 13:29 < armin> nazarewk: use a distro that cares about your brain-damage then. 13:30 < armin> nazarewk: there are arch derivates that come with an installer btw. 13:30 < yakiza> armin: thank you 13:30 < armin> nazarewk: if you really want the arch way, that is. 13:30 < SuperSeriousCat> Do you install a full DE system every time? Else Arch take 5-10 min tops to install. You can automate a lot with a few scripts if you want 13:30 < SuperSeriousCat> Other distros are not that much faster to install 13:31 < sagax> hi all! 13:31 < armin> i can even run a fully usable desktop session WHILE my operating system installs, including full-disk-encrypted setups, while listening to music and chatting on IRC, why the hell would i want to go through that trouble again? 13:31 < yakiza> SuperSeriousCat: use slackawre 13:31 < yakiza> slackware* 13:32 < armin> SuperSeriousCat: if you get satisfaction out of elitism go ahead, but i see no reason to hurt myself. 13:32 < SuperSeriousCat> Im not the one wanting Arch 13:32 < SuperSeriousCat> Too unstable in my experience 13:32 < sagax> how to i can see what colorscheme in pdf file? 13:33 < sagax> example rgb or cmyk 13:33 < sagax> ? 13:33 < armin> SuperSeriousCat: compared to what? 13:34 < armin> heck, even freebsd with zfs on root and full disk encryption via geli is less pain to setup than the arch clusterfuck. 13:34 < SuperSeriousCat> I use Gentoo as desktop and Debian as server. Compared to them I suppose. Everytime I tested Arch something broke in an update within a few months 13:34 < armin> SuperSeriousCat: as long as you don't propose gentoo as being less hassle to setup... 13:35 < junka> 1 broken piece every few months? thats a record! 13:36 < SuperSeriousCat> Its not that big a hazzle. Mostly copy-paste and with a few use flag changes. The boring part is lspci -k to get all the right kernel changes correct 13:38 < mnemon> SuperSeriousCat: My desktop arch installations haven't broken in the last 2-3 years apart from the screen brightness controls not working on the laptop(had to do some manual actions few times to fix the update process itself). before that it was a bit more prone to breaking, mostly with the nvidia blobs 13:38 < armin> SuperSeriousCat: oh, *that* simple even, wow. m( 13:39 < SuperSeriousCat> Guess I was unlucky then. Tested it with Xfce and it refused to boot into DE more than once after an update 13:39 < armin> installing nvidia drivers for me was the end of predictability of my computer. no matter the distribution, the behaviour of my machine was seriously weird after then. 13:40 < Neobenedict> https://i.imgur.com/5ScL1gL.png why are there so many disks? raid 5 with 4 disks 13:40 < Neobenedict> why are there six for each disk 13:40 < SuperSeriousCat> Never bother with nvidia drivers. I dont game on my Linux system and it often breaks in kernel updates 13:40 < armin> nazarewk: there are 4. 13:41 < mawk> since I updated KDE the perfectly good behavior of opening .png links in gwenview doesn't work anymore for distant content, I hope it comes back within a reboot 13:47 < plexigras> is using symlinked dotfiles a bad idea? 13:47 < dgurney> why would it be? 13:47 < paulcarroty> nvidia means you use old crap or wait updates like idiot 13:48 < hk238> how can I restart desktop without log out? :o 13:48 < hk238> on debian linux with kde? :o 13:49 < plexigras> so im good with doing `ln -s dotfiles/vim .vim`? 13:49 < hendrix> plexigras: sure 13:50 < Sitri> plexigras: I do that 13:50 < plexigras> ok then i will give that a try 13:53 < BluesKaj> hk238, check this out, https://www.makeuseof.com/tag/refresh-linux-desktop-without-rebooting/ 13:53 < hk238> anyone know about KDE desktop, is there a way to increase textfield size? Thanks Blueskaj I just foundout away to do it from google too 13:57 < hk238> https://pasteboard.co/Hj0kN5g.png here's the problem essentially 13:59 < BluesKaj> hk238, textfield size ? not sure waht you mean 14:00 < hk238> underneath the icons there's a label that contains the foldername, but after reducing the size of the grid, the textfield only displays 1 character or something like that 14:01 < hk238> where as I would like the text to wider so that the folder names can be read 14:01 < BluesKaj> increase the dpi perhaps 14:03 < lseactuary> i have 30 .sh scripts in my desktop folder, want to copy it to my terminal folder. is it possible? 14:03 < mawk> cp Desktop/*.sh Terminal 14:03 < BluesKaj> hk238, system settings>fonts>"force fonts dpi " 14:03 < mawk> without more precision 14:04 < hk238> while the font might be a little large the problem is the area that is being used for text display is too small relative to the icons / cellsize 14:04 < Peetz0r> hey! What would be the cheapest small linux board with wifi today? would it be the Orange Pi i96? or something else? 14:04 < hk238> but I'll try changing that dpi 14:04 < Peetz0r> I don't care much about specs because it's going to do almost nothing 14:05 < istevenmon> hi guys, I had an old HP server with cciss driver for controller, I made an image with dd and i am trying to connect it to a virtual machine with IDE controller. The problem is that booting from livecd the drive gets recognized as sda, when chrooting the / df shows as cciss/c0d0p1. How can i fix this so when booting from the disk wont kernel panic 14:05 < lseactuary> mawk i ssh'd into my Ec2 and want to copy the scripts from my local desktop to there 14:05 < lseactuary> hence stuck 14:05 < mawk> ah ok 14:05 < mawk> use scp then 14:05 < lseactuary> ah 14:05 < mawk> scp source user@host:destination 14:05 < SuperSeriousCat> Peetz0r, https://www.adafruit.com/product/3400 14:05 < SuperSeriousCat> Thats a tiny one who use almost no power and got WiFi build-in 14:06 < hk238> there's a thing called PlasmaComponents.TextArea in FolderView.qml , I wonder if changing something in that would result to larger text field? :o 14:06 < acresearch> people i am trying to use paperspace (paperspace.com) to preform some computation. but they only have ubuntu 14.04 and 16.04 (not 18.04 nor any other linux distro), anyone knows when will they provide 18.04 or is there a way to install another distro myself? 14:07 < Peetz0r> SuperSeriousCat: I don't really care about power usage, but the raspberry pi zero W is not cheaper than the Orange Pi Zero or i96 or others 14:10 < promach_> hi, for u-boot, what does the hypher mean in "bootm 0x3000000 - 0x2A00000\0" \ ? 14:11 < promach_> hyphen 14:15 < BCMM> promach_: no initrd, i think 14:16 < promach_> what does that exactly mean ? 14:16 < repys> I got a power failure and after that I noticed an error: ABRT in smartctl output from my disks. what is that exaclty? 14:16 < promach_> BCMM: Bypass the initial ramdisk ? 14:16 < repys> there are no bad sectors from smartctl but I see this Error: ABRT errors 14:16 < BCMM> promach_: the 2nd argument would be the address of an initrd image, if there was one to load 14:17 < acresearch> people what is better for cloud computing (number cruching not web hosting) paperspace or linode? 14:17 < BCMM> promach_: may i ask what you're running u-boot on? 14:17 < promach_> Zedboard 14:19 < promach_> BCMM: how should I modify https://github.com/Xilinx/u-boot-xlnx/blob/master/include/configs/zynq-common.h#L266-L273 to achieve the purpose of no initrd ? 14:19 < repys> any idea guys? 14:19 < repys> or can you tell me how to know? 14:22 < BCMM> promach_: i'm just going on this https://www.denx.de/wiki/view/DULG/UBootCmdGroupExec#Section_5.9.4.2. 14:22 < BCMM> promach_: i haven't actually played with u-boot myself 14:23 < BCMM> promach_: based on that document, it seems like ${ramdisk_load_address} should be set to "-" if you're not using an initramfs 14:25 < nothos> Hey all, I tried updating my fedora 26 install to 27 and it basically murdered everything. Managed to get the system booting by doing a distro-sync in a chroot 14:25 < nothos> But now a lot of stuff seems to be looking for glibc 2.26 which is from fedora 27 and not 26 14:25 < nothos> Including curl so I can't even use dnf to fix it :D 14:26 < promach_> BCMM: someone just told me to use setenv ramdisk_load_address - 14:26 < promach_> what do you think ? 14:26 < Pentode> nothos, sounds like a good time to do a fresh install. ;p 14:26 < nothos> I installed a copy of glibc from 27 (version 2.26) but changing the symlink of libc.so.6 to point there just breaks more stuff 14:26 < BCMM> promach_: that seems right 14:26 < nothos> Pentode Realistically yeah, but I'm at the point that I'm getting stubborn 14:27 < Pentode> it may just lead to insanity and bitterness, lol 14:27 < nothos> Pentode used to that ;) 14:28 < LambdaComplex> I'm trying to get a RAID array working after I replaced a failed drive. There should be two RAIDS (both mirrors), md0 and md1. md0 seems fine. md1 doesn't exist anymore, and mdadm --examine is saying that sda3 and sdb3 don't have md superblocks 14:28 < Pentode> carry on then ;D 14:28 < nothos> Mainly I really, really don't want to have to try and mess around getting xrdp working again >.< 14:28 < LambdaComplex> so, how do i recreate md1 from the existing partitions? 14:29 < nothos> I guess that's my punishment for not snapshotting the LV of the server first 14:32 < noodlepie> Hiya guys! All good today? 14:32 < LambdaComplex> no 14:32 < LambdaComplex> :) 14:33 < noodlepie> Twisted firestarter! 14:42 < rigid> ahoy 14:42 < SuperSeriousCat> Ahoy, matey 14:42 < rigid> is there a way to retrieve the backlog of a kernel panicing at boot? 14:42 < rigid> the resolution is so small, i just see the end of the stack trace and shift+pgup doesn't work 14:44 < jhodrien> rigid: Serial consoles are you friend. 14:45 < rigid> oh, ok... that could get hard on a box without a serial port :) 14:46 < jhodrien> LambdaComplex: If it's RAID1, then they're individually mountable, so you'd still have access to the data. 14:46 < noodlepie> Linux is fun and fast. 14:46 < noodlepie> I like it a lot. 14:47 < jhodrien> rigid: No virtual serial (like IPMI)? 14:47 < jollyjester> hello 14:47 < jollyjester> anyone wanna chat on mumble about linux? 14:47 < HyP3r> Short question, how can I search for a specific line in a file with sed and then replace the line after that line with some new text? 14:49 < jollyjester> hello 14:51 < z88> HyP3r: Insted of using sed to search for the line, use grep with the opcion "A", after. Then sed that line. 14:51 < plexigras> how can i link something like `ln -si a b` but ignore if a is already linked to b? 14:51 < z88> grep -A after, grep -B before. 14:52 < LambdaComplex> jhodrien: while true, that doesn't answer my question 14:53 < HyP3r> z88: ok but this line is not uniqe 14:54 < mawk> of course plexigras 14:54 < mawk> just use --force 14:54 < mawk> ah, I didn't see the -i 14:55 < mawk> well I don't think so but you can add a test right before 14:55 < plexigras> yeah thats the issue i cant just ignore it if the link already exists 14:56 < kazdax> whats the use of alisas 14:56 < kazdax> any practical use ? 14:56 < mawk> plexigras: [[ -L b && $(readlink b) == a ]] || ln -i a b 14:56 < jollyjester> hello anyone willing to chat on mumble? 14:56 < kazdax> does it just make it easier to search things ? 14:56 < jhodrien> LambdaComplex: Well, no, but it possibly gets you your only route out if you can't get md to recognise the partitions. You've presumably got one good and one empty, given you had a failed disk. 14:56 < mawk> subject to race conditions blah blah 14:57 < promach_> For https://github.com/Xilinx/u-boot-xlnx/blob/master/arch/arm/config.mk#L68-L73 , why am I getting this error https://paste.ubuntu.com/p/hz4NGwPN9G/ ? 14:57 < jhodrien> So why not just create md1 as a raid1 with a missing member on the new disk, mount the old filesystem, dump, restore, add the old partition in? 14:57 < plexigras> mawk thank i will try that 14:57 < jollyjester> anyone willing to chat with me on a mumble server? 14:58 < jollyjester> hello! 14:59 < mawk> hello jollyjester 14:59 < jollyjester> mawk, are you here for a specific reason e.g help? or are you here just to chat? 15:00 < mawk> I'm here to help and occasionally get helped 15:00 < jollyjester> Ok 15:00 < jollyjester> If you want you can chat with me on a mumble server if you are willing to 15:00 < mawk> I don't have a microphone 15:00 < mawk> and I don't like speaking 15:00 < rypervenche> Sounds like someone just found a new Mumble vulnerability. 15:00 < jollyjester> Understandable 15:00 < mawk> lol 15:01 < repys> does dd run in kernel space or user space? 15:01 < jollyjester> lol 15:01 < mawk> user space repys 15:01 < mawk> using syscalls 15:01 < mawk> but I'd say the great deal is in userspace, as a copy happens 15:01 < repys> mawk: I am running dd, it uses 90% of cpu and this 90% used is in system space reading from top 15:01 < repys> why then? 15:02 < jhodrien> Pissy small block size? 15:02 < jhodrien> set bs=4M and then try again. Is this a desktop or something embedded? 15:02 < repys> desktop 15:04 < LambdaComplex> jhodrien: do you know how to get the superblock version from an existing array? 15:04 < repys> system space are the syscall basically, right? 15:06 < jhodrien> Does mdadm --misc --detail /dev/md0 give you what you want? 15:06 < LambdaComplex> oh, yeah, it does. guess i'm illiterate 15:07 < LambdaComplex> thanks jhodrien 15:07 < jhodrien> Easy to be blind to it. No problem. 15:07 < LambdaComplex> someone else was working on this before, btw. looks like all the data is actually on both drives 15:08 < LambdaComplex> so i'm gonna try to --create it again (specifying the superblock version) 15:09 < noodlepie> Oooh, Linux kernel 4.16.6-gentoo available, installing it now... @:P-~ 15:17 < plexigras> hey mawk do you have an idea how i could get this to work when doing something like `ln -si * /example` ? 15:18 < promach_> Could anyone help with https://www.reddit.com/r/linuxquestions/comments/8fzc0a/gcc_version_mismatch/ ? 15:19 < acresearch> people what is better for cloud computing (number cruching not web hosting) paperspace or linode? 15:23 < tharkun> acresearch: depends on your software for ai and numbercrunching if you are using gpu software I believe paperspace is your place to got. There are some distributed services. You need to research those too. 15:24 < Roenie> GRUB2 question: I want to replace a very noisy HDD (/dev/sda) that contains root (/dev/sda2, sda1 contains Windows). I'll be replacing the entire harddisk with an SSD, and that SSD has been in use before and probably has GRUB1 or GRUB2 on it. I will partition the SSD with 2 partitions as well, and when it replaces the HDD it will become /dev/sda. Currently it's attached as /dev/sdf. What is the correct method here to make the system boot co 15:24 < Roenie> rrectly off the SSD? (Can booting off a live cd and chrooting be avoided?) 15:24 < acresearch> tharkun: well right not i am not using AI not GPU, just a CPU but i guess i need updated distros ,,, but i am not sure if linode is made for computaion or just web hosting 15:25 < Roenie> (note: the SSD has been in use in a differnt system) 15:25 < mawk> not really plexigras 15:25 < mawk> well you can use find to generalize my method 15:26 < mawk> or write a wrapper for ln that will do my test 15:26 < mawk> in bash it should be pretty easy 15:30 < LambdaComplex> Roenie: well, first, use UUIDs in fstab 15:30 < Roenie> there currently is a UUID in there for the HDD, and I see the same one in one of grub's config files 15:30 < LambdaComplex> I guess update the fstab to reflect the SSD's partitions' UUIDs, reboot, and hope for the best? 15:31 < Roenie> but but but :) I will have to grub-install to the SSD and it's at /dev/sdf. 15:31 < Roenie> will it then still use /dev/sda? 15:32 < noodlepie> I just use BIOS booting. Should learn UEFI at some point in the future though. 15:32 < noodlepie> legacy I mean 15:32 < Roenie> because once I remove the old HDD , the SSD becomes /dev/sda 15:32 < Roenie> but I have to install grub to the SSD while it's still /dev/sdf 15:33 < Roenie> I probably should mention it is a BIOS system, no UEFI mobo 15:33 < LambdaComplex> grub-install just writes the necessary executable stuff to the MBR of the drive you specify 15:33 < LambdaComplex> the fact that you're writing that data while the drive is /dev/sdf and will boot when it's /dev/sda is not important at all 15:33 < iflema> Roenie: run blkid 15:34 < Roenie> I guess update the fstab to reflect the SSD's partitions' UUIDs, reboot, and hope for the best? 15:34 < Roenie> would I not also have to do update-grub to get that change into grub's config? 15:34 < LambdaComplex> yeah, that too 15:35 < LambdaComplex> assuming this is ubuntu 15:35 < LambdaComplex> (or debian? does it use update-grub as well? can't remember) 15:35 < Roenie> debian 15:35 < Roenie> I think it does, yeah 15:35 < Roenie> iflema, thanks I am familiar with blkid 15:35 < LambdaComplex> well if it doesn't have update-grub then it's just `grub[2]-mkconfig -o /boot/grub/grub.cfg` 15:36 < Roenie> I see 15:37 < LambdaComplex> the [2] because it's either grub-mkconfig or grub2-mkconfig...because some distros have grub and grub2, so grub2 stuff is grub2-, but other distros have grub2 but use grub- for its commands 15:37 < Roenie> I miss the old lilo days when things were simple 15:38 < kuri0> how do I change my screen's white balance ? 15:38 < plexigras> mawk: i made this bash function any tips for improing it? 15:38 < plexigras> ln_i() { 15:38 < plexigras> for i in ${@:1:(${#@})-1}; do 15:38 < plexigras> local o=$([[ -d ${@: -1} ]] && echo "${@: -1}/$(basename $i)" || echo "${@: -1}") 15:39 < plexigras> [[ -L $o && $(readlink $o) == $i ]] || ln -si $i $o 15:39 < plexigras> done 15:39 < plexigras> }y 15:39 < Roenie> thanks LambdaComplex 15:39 < plexigras> it works but i dont know what im doing so there is probably a better way 15:39 < mawk> maybe add quotes here and there 15:39 < kuri0> on windows i can change the white balance by saying how much of each color to use in calibration settings 15:39 < triceratux> Roenie: youre always welcome to use lilo. its still in an archive somewhere 15:40 < Roenie> reason I didn't is it would feel like a cop out :P 15:40 < LambdaComplex> Roenie: oh, and be sure you have os-proper installed before running update-grub 15:40 < LambdaComplex> *os-prober 15:41 < mawk> what's the use of the first line inside the loop plexigras ? 15:42 < Roenie> I'll dd the entire HDD device to a USB backup disk just in case it fails, because there's no going back. 15:42 < plexigras> mawk: its for if the last argument is a dir 15:42 < iflema> cp -a all the dirs, not /sys /proc /dev, to the ssd (break it up across partitions even), get the UUIDS and edit fstab, and update bootloader 15:42 < Roenie> the act of update-grub on the running system after editing fstab does that. 15:42 < mawk> oh ok plexigras 15:43 < mawk> why do you use basename ? 15:43 < mawk> ah no sorry ok 15:43 < mawk> yeah it seems right 15:43 < mawk> now just a little error handling, checking that $# >= 2, etc 15:44 < plexigras> i will do that 15:44 < iflema> no /tmp or /mnt either 15:45 < nepho> Hello. I'm trying to poll (2) on a serial interface (/dev/ttyACM0), and have trouble with CPU usage. Opening the file with O_RDWR (because I need to both read from/write to it) means that my polling condition will (almost) always have POLLOUT true. Which means that the loop pretty much equals a "while (true)" loop in term of CPU usage. Could anyone help me? Thanks in advance. 15:45 < iflema> and no /run 15:46 < mawk> iflema: cp --one-file-system 15:47 < iflema> -a will get you there 15:47 < mawk> ah 15:48 < jelly> -a doesn't do -x 15:48 < jelly> likewise with rsync 15:48 < iflema> jelly: havnt we done this before? 15:49 < jelly> probably! :-D 15:50 < noodlepie> 4.16.6-gentoo stable here. 15:50 < triceratux> 4.16.4-1-ARCH swagarch here 15:51 < noodlepie> My kernel is perfectly optimised for my system with Gentoo. 15:51 < mawk> lol 15:51 < noodlepie> Its such a simple idea of how to do a distro. 15:52 < noodlepie> but you need the CPU power to compile everything. 15:52 < noodlepie> But it works.. so.. :P 15:53 < n13z> noodlepie: https://me.me/i/ill-protect-you-ill-keep-you-warm-until-youre-ready-13726455 15:55 < HyP3r> If I run 'su -c "echo $USER" -s /bin/bash foo' why is $USER then root instead of foo? 15:55 < ayecee> HyP3r: because the $USER is expanded before the su is run 15:56 < HyP3r> is there a way to expand those vars while execution? 15:57 < ayecee> use single quotes instead of double quotes 15:58 < HyP3r> ayecee: thanks 15:58 < LambdaComplex> nepho: try ##programming or maybe ##posix or even ##C (maybe) 15:59 < kuri0> does anyone know how to change the screen white balance (color sliders) on linux ? 15:59 < kuri0> with xorg 15:59 < LambdaComplex> nepho: or perhaps even #osdev because they've been happy to help with userspace programming every time i've asked (and they're good at it) 16:01 < Alexander-47u> i get 'not found' when trying to run an executable that is in local directory 16:01 < noodlepie> Anyone like fractals. I made some by myself from source here --> www.google.com/+PhillipHugo They're good and worth a look. 16:01 < Alexander-47u> how to fix path? 16:01 < mawk> nepho: you can solve that POLLOUT in several ways 16:02 < mawk> the dirty way is assuming that you can always write to the tty, and just losing bytes when you can't 16:02 < mawk> another way is to have the POLLOUT event in oneshot mode, and activate it only when you can't write to the tty (write() returns EAGAIN/EWOULDBLOCK) 16:03 < mawk> and in the meantime halt whatever operation that fills your buffer 16:03 < mawk> a third way is to use edge-triggered mode, nepho 16:04 < mawk> that's with epoll; epoll will send you only one notification per event, and no further events will be sent until you fully consume the event (fully read() from the file descriptor until you get EAGAIN/EWOULDBLOCK) 16:04 < mawk> so only one EPOLLOUT will be sent, you write all the things you want, and no new EPOLLOUT events will be sent until the next EAGAIN/EWOULDBLOCK from write() 16:04 < mawk> you got everything nepho ? 16:06 < promach_> is https://developer.arm.com/open-source/gnu-toolchain/gnu-rm/downloads cross compiler or native compiler ? 16:06 < melkor> If I have a python or java program that is reading stdin, it is limited to 4k characters at a time. I think this is a limitation of the shell and I wonder if there is an enviroment variable I can change. 16:06 < mawk> a limitation of the shell melkor ? 16:06 < mawk> where did you get that ? 16:07 < mawk> the behavior of python buffering input in 4k chunks with a pipe is a behavior of...python 16:07 < mawk> you can read whatever size you want from a pipe 16:07 < melkor> because it worked in a different environment. 16:07 < mawk> yeah it depends on the type of stdin 16:07 < mawk> pipe vs tty 16:07 < mawk> but overall python decides, not bash 16:08 < mawk> PIPE_BUF is hardcoded to 4k, you can't change that 16:08 < nepho> Yes, thanks a lot @mawk! :) 16:08 < geirha> you can always tell the other side of the pipe to use no buffering, or line buffer 16:08 < mawk> but with a size larger than 4k you can perfectly write/read, it's just that it won't be atomic anymore 16:08 < mawk> yes 16:08 < mawk> the application itself does the buffering 16:10 < mawk> but indeed there are environment variables melkor 16:10 < melkor> So all the program does it: i = input(); print(len(i)), if I paste 15k of text w/out any returns I get 4k. How would I get the rest of that. The reason I asked here is because this happens with both python and java, so I thought it could be my environment. 16:10 < mawk> you can unbuffer stdin using the -u flag for python 16:10 < mawk> or rather modify stdin inside python itself to make it line-buffered 16:11 < geirha> I thought that only unbuffered stdout 16:11 < noodlepie> God's computer is better than mine because he has Trees, Heap, Hashes, Man and Memory 16:11 < mawk> ah sorry geirha you're righty 16:11 < mawk> -y 16:11 < mawk> so just modify stdin from inside python 16:11 < geirha> melkor: it's the command on the left side of the pipe you need to coerce into not buffering 16:11 < geirha> If you're lucky, it has an option for it. If not, you can trick it with commands like unbuffer or stdbuf 16:14 < melkor> okay, so it seems that pasting has some buffering. eg. If I paste to a text file it is all there. but if I paste to the program it only gets 4k. 16:15 < geirha> yeah, pasting to a terminal emulator is problematic, and you often lose bytes along the way 16:16 < melkor> okay even pasting to cat or wc cause this problem. 16:18 < Psi-Jack> Eh? I have pasted quite huge buffers before, no problem. 16:19 < melkor> I don't really think I need to paste 4k+ to the command line, but if I do cat > test.txt then ctrl+shift+v 15k of text followed by ctrl+d I only get a 4k file. 16:20 < Psi-Jack> What terminal? 16:21 < melkor> gnome terminal. 16:21 < melkor> Pasting into vim gives me the full output. 16:22 < Psi-Jack> Yeah, I generally paste directly into vim. . Whare are YOU pasting into that only gives you 4k? 16:23 < melkor> The cat example above only gives 4k. I was looking at a java program that only recieved 4k, and I made a python program to test it. 16:25 < jollyjester> hello everyone 16:25 < jollyjester> anyone wanna talk with me on my mumble server 16:25 < jollyjester> i really want to talk to someone about linux 16:26 < n13z> melkor: when I use python I create a container to save every chunk of data to avoid this kind of issues. 16:26 < post-factum> jollyjester: call your mom 16:26 < Psi-Jack> Ask quesions, actual questions, not ask to ask. And avoid using enter as punctuation and/or pauses in thought, jollyjester. 16:26 < Psi-Jack> post-factum: Condescending much? 16:26 < n13z> "too much" 16:27 < post-factum> not enough 16:28 < Psi-Jack> Well then. 16:31 < texinwien> I'm getting a "bad variable name" error when I start a new tmux session. 16:32 < texinwien> Helpfully, the msg tells me on which line the error occurs. 16:32 < texinwien> Unhelpfully, it doesn't tell me in which file it occurs. 16:32 < post-factum> texinwien: ~/.tmux.conf ? 16:33 < texinwien> post-factum the error occurs on line 5 (of something). Line 5 of ~/.tmux.conf is part of the commented copyright notice. 16:33 < MrKeuner> Hi, how can use capturing groups in gedit regex search? () and $1 combination doesn't seem to work 16:33 < post-factum> texinwien: /etc/tmux.conf ? 16:34 < texinwien> post-factum /etc/tmux.conf doesn't exist. 16:34 < acresearch> tharkun: well right not i am not using AI not GPU, just a CPU but i guess i need updated distros ,,, but i am not sure if linode is made for computaion or just web hosting 16:35 < post-factum> texinwien: ok, post an actual error then 16:35 < jhodrien> strace will probably clarify, by showing you what file it opened before the error was output. 16:35 < cart_man> how can I write a command in one line that will execute a build and then shutdown the system? 16:35 < jhodrien> Do you care whether the build succeeds? 16:36 < Psi-Jack> acresearch: I cannot recommend Linode. At all. 16:36 < cart_man> jhodrien, It should yes 16:36 < uplime> if you care about the build succeeding: build command && sudo shutdown now 16:36 < ananke> cart_man: command1 && shutdown -h now 16:36 < jhodrien> ^^ that 16:36 < cart_man> ananke, Thanks 16:36 < acresearch> Psi-Jack: why? what is your angle? 16:37 < Psi-Jack> acresearch: They are horribly unskilled people running that service very badly. 16:37 < acresearch> Psi-Jack: oh 16:37 < acresearch> Psi-Jack: hmmmm i have 2 websites there,,,, maybe i should remove them? 16:37 < Psi-Jack> Yes 16:37 < Psi-Jack> Heh 16:38 < acresearch> Psi-Jack: where can i take them? i just need a cloud server, linode was very easy to setup 16:38 < Psi-Jack> You'd be far better off with Digital Ocean, Vultr, or AWS. 16:38 < Schrostfutz> Hi, on my system kthreadd seems to always take up one CPU core. I'm trying to debug the issue but have no idea where to start. Do you have any suggestions? 16:38 < acresearch> Psi-Jack: ok let me search them 16:39 < post-factum> Schrostfutz: cat /proc/2/stack 16:39 < post-factum> acresearch: i use vultr 16:39 < texinwien> post-factum: -sh: 5: export: {some paths}: bad variable name 16:39 < Psi-Jack> I use all three for differing purposes. 16:39 < acresearch> post-factum: ok let me check them out and their prices for linode i pay $5 / month for a server 16:39 < Schrostfutz> post-factum: http://ix.io/196M 16:39 < HyP3r> How can I execute a multiline bash script with su? 16:40 < post-factum> texinwien: to me it looks like it is your shell that produces the error. either from local config, or from global. what do you use, bash? 16:40 < Psi-Jack> acresearch: Yep. All three have similar options 16:40 < Schrostfutz> HyP3r: su -c 'script.sh'? 16:40 < acresearch> Psi-Jack: nice 16:41 < post-factum> Schrostfutz: what's in top? 16:41 < post-factum> 100%? 16:41 < texinwien> post-factum zsh, and I'm also poking around in zsh-related files. 16:41 < HyP3r> Schrostfutz: yes but I want to write this code in the same file as the main code 16:41 < HyP3r> Its a install script and I want to execute some lines of the script as another user 16:41 < texinwien> Interestingly(?) it only occurs when I open zsh inside tmux (not in screen, for instance) 16:41 < acresearch> post-factum: Psi-Jack what about GPU, you know for (my) future when i get a grip on AI and deep learning are these services good enough? 16:41 < Schrostfutz> post-factum: ~70% for kworker/0:1 and ~30% for ksoftirq/0 16:42 < Psi-Jack> acresearch: AWS might be your best bet. 16:42 < post-factum> Schrostfutz: that's different… check their stacks then 16:42 < acresearch> Psi-Jack: ok, got it :-) 16:42 < Psi-Jack> acresearch: I usually recommend AWS above all, but it's not as simple as the others. But it's a lot more flexible, powerful, and you have many MANY more options. 16:43 < acresearch> Psi-Jack: i understand. i prefear SSH and doing things more manually than on layers 16:43 < acresearch> Psi-Jack: it is just less confusing for me 16:43 < Psi-Jack> Including.. For examnple. AWS has Elastic GPUs. 16:43 < azarus> Psi-Jack: do they bounce? 16:44 < Psi-Jack> Maybe. :) 16:44 < azarus> *boing* 16:44 < Hdphn> anyone using nixOS? 16:44 < Hdphn> heard great things about it 16:44 < Hdphn> wanna switch from arch to nixOS. worth it? 16:44 < Psi-Jack> TIAS for yourself. 16:45 < ayecee> what is TIAS 16:45 < Psi-Jack> This is a great use of virtual machines for testing. 16:45 < Schrostfutz> post-factum: The stack for ksoftirq shows this: http://ix.io/196N. The kworker's PID seems to change, the stack of the busy one only contains 0xffffffffff 16:45 < Psi-Jack> TIAS is a very common acronym. 16:45 < azarus> try it and see 16:45 < ayecee> so what does it mean 16:45 < Hdphn> TIAS? 16:45 < ayecee> ah, i see 16:45 < Psi-Jack> azarus said. :) 16:46 < Hdphn> English please. Please obey topic rules. use full and proper english. Psi-Jack 16:46 < post-factum> Schrostfutz: fine. then, go for "pref top -Ug" 16:46 < ayecee> should probably expand it so it's clear to non-english users. 16:46 < kuri0> how do I load a windows icc ? 16:46 < azarus> Hdphn: let's use esperanto instead, make it universal 16:46 < Psi-Jack> TIAS has been around since the early BBS days, heh 16:46 < post-factum> ayecee: "Try It And See" 16:47 < ayecee> nick checks out 16:47 < jhodrien> kuri0: I'm not sure that's a linux question. 16:47 < Schrostfutz> post-factum: pref or perf? 16:47 < kuri0> jhodrien, in linux 16:47 < post-factum> Schrostfutz: meh. perf, of course 16:47 < Hdphn> Psi-Jack: I see you keep breaking the rules of ##Linux 16:47 < jhodrien> You want to run a windows icc compiler under linux? 16:47 < kuri0> jhodrien, icc color profile 16:47 < azarus> jhodrien: no, icc is a color profile 16:47 < azarus> kuri0: https://wiki.archlinux.org/index.php/ICC_profiles 16:47 < kuri0> i have one from windows but it doesn't work 16:47 < jhodrien> Ah sorry, wrong end of the stick. 16:47 < kuri0> ones from the internet work though 16:48 < azarus> (basically. kuri0, use xcalib) 16:48 < kuri0> ill try it 16:48 < kuri0> i used my distro's gui 16:48 < Schrostfutz> post-factum: That shows me 40-50% being used in xhci_hub_control 16:48 < post-factum> interesting 16:49 < Schrostfutz> most of which in ret_from_fork though 16:49 < post-factum> Schrostfutz: smth like https://forums.linuxmint.com/viewtopic.php?t=240569 ? 16:49 < kuri0> azarus, when i do xcalib with that file the screen colors stay the same 16:50 < kuri0> so its like not recognized by it or something 16:50 < kuri0> the icc works on windows 16:50 < Psi-Jack> Hdphn: SMS speak, aka shtspk, is the "abbreviating" of more simpler words usually. u vs you; ic vs I see; Actual acronyms like LOL, TIAS, BRB, AFK, etc are not SMS-speak. 16:50 < Hdphn> ok but ##topic doesnt allow you to speak SMS 16:50 < Hdphn> either 16:50 < Hdphn> so you better be careful next time 16:51 < Hdphn> before OPS ban you 16:51 < uplime> what? 16:51 < ayecee> Hdphn: not going to win any points by whipping this dead horse. 16:51 < Hdphn> true 16:51 < Hdphn> he is just a troll. leave him alone 16:51 < Psi-Jack> Also, hitting enter excessively as if punctuation or pauses in thought, is frowned upon. 16:52 < ayecee> ... 16:52 < ayecee> a little early in the morning for a t-bomb. 16:54 < Schrostfutz> post-factum: Yeah, that looks exactly like my issue. Restarting the USB hub as described in the post fixed the issue. I'm gonna try disabling usb autosuspend, too, s.t. it does not reoccur. Thanks for the help! 16:54 < post-factum> welcome 16:58 < noodlepie> Free Music available at http://www.freemusicarchive.com/ and http://libre.fm. Enjoy! 17:06 < newpy> I'm wondering if anyone here knows how to edit color scheme on gitbash? 17:06 < uplime> isn't gitbash a windows thing? 17:08 < acresearch> Psi-Jack: ok i made an account in vultr, i think it is the cheepest 17:11 < uplime> acresearch: i really like vultr 17:11 < uplime> its nice 17:11 < acresearch> uplime: :-) 17:11 < uplime> ive got 5/6 servers through them 17:11 < acresearch> uplime: what do you do with 5 servers? 17:12 < SuperSeriousCat> Server 2 check uptime of server 1 by pinging, 3 of 2 and so on 17:12 < uplime> 2 of them are dns, 1 is a web server for serving static files, 1 is a dev server for an irc project (will probably actually scrap that one though), 1 is for web/irc 17:12 < uplime> oh and one is an irc testnet 17:13 < acresearch> uplime: wow, can't you run all these from 1 server? i mean i have a website + IRC bot + twitter bot + some other bots like 5 in total all in 1 server 17:13 < uplime> yes 17:15 < Psi-Jack> acresearch: Eh, they're pretty decent at least. I just don't like not being in control of backups. 17:16 < Psi-Jack> uplime: webserver, just for static files? Man,.. S3 buddy. 17:16 < uplime> Psi-Jack: i know :c 17:16 < acresearch> Psi-Jack: i totally understand, i moved my website from hostgator to linode once i figured out how to setup a LAMP and apache (which is still confusing) 17:17 < uplime> i didn't know how cool aws was 17:17 < uplime> but im slowly migrating to it 17:17 < Psi-Jack> heh 17:17 < uplime> im very excited to move my dns stuff to route 53 and setup my first vpc 17:17 < plexigras> is there a better way to do `$(dirname $(realpath $0))`? 17:17 < Psi-Jack> plexigras: What are you using that for? 17:17 < plexigras> to get the dir the bash script is in 17:18 < uplime> plexigras: /msg greybot !faq location 17:18 < Psi-Jack> ^ 17:18 < Psi-Jack> I was about to suggest #bash :) 17:18 < uplime> \o/ 17:18 < plexigras> thanks guys 17:19 < uplime> #bash might be the second most pedantic channel, but there are definitely a lot of people there that know their stuff 17:19 < Psi-Jack> Including me. :) 17:19 < kazdax> whats a good linux webhosting servive "? 17:19 < Psi-Jack> And I'm very pedantic. 17:19 < Psi-Jack> kazdax: Got a Linux question? 17:20 < noodlepie> 1and1.com, rackspace.com 17:21 < Psi-Jack> Someone get a rope. 17:21 < noodlepie> dsvr.co.uk is dead now, they used to sell virtual services 17:21 < Psi-Jack> We're gonna need rope and shovels. :) 17:21 < uplime> we bury the rope 17:21 < uplime> and then hit noodlepie with the shovel 17:22 < ayecee> as is tradition 17:22 < Psi-Jack> heh 17:22 < uplime> the way they did in the motherland 17:24 < Psi-Jack> plexigras: Hopefully learning not to use $0 will do you good. : 17:26 < blurry_light> anyone here familiar with the Linux Foundation's certificates? 17:27 < blurry_light> i was thinking of preparing for the LFCS instead of the RHCSA since the former is vendor-neutral and distro agnostic 17:27 < plexigras> Psi-Jack: what else should i use? 17:27 < jml2> blurry_light, rhcsa is more advanced 17:28 < Psi-Jack> plexigras: That FAQ page shows you a lot. 17:28 < blurry_light> jml2: i see. how would you compare LF engineer then with RHCSA? same level? 17:28 < jml2> blurry_light, lfcs is for babies, rhcsa is asked in industry... 17:28 < kazdax> blurry i am preparing for my RHCSA 17:29 < Psi-Jack> plexigras: Why do you need the path the script is in? 17:29 < kazdax> what book are you using or resources to study for it ? 17:29 < jml2> blurry_light, ask kazdax -- he'll agree with me 17:29 < blurry_light> jml2: ok ... what about LF engineer then? their next certificate 17:29 < jml2> kazdax, book? 17:29 < blurry_light> kazdax: i see 17:29 < blurry_light> kazdax: a video course 17:29 < kazdax> which one blurry ? 17:29 < blurry_light> of course nothing beats actually sitting down and doing previous exam problems in an actual distro 17:30 < kazdax> rhcsa i think is better because its hands on 17:30 < blurry_light> kazdax: one by this Dutch guy 17:30 < kazdax> and not a written test like everyone else 17:30 < jml2> kazdax, get the safarionline subscription -- it has tons of video series for redhat 17:30 < blurry_light> kazdax that's the thing, so is LF stuff 17:30 < plexigras> this seams to work `echo $(dirname "$PWD/$BASH_SOURCE")` 17:30 < blurry_light> it's also hands on 17:30 < jml2> kazdax, (along with books iirc) 17:30 < blurry_light> you solve actual problems in a web-based virtual machine 17:30 < kazdax> ohh ill look it up jml thanks 17:30 < blurry_light> it's not like CompTIA's stuff 17:30 < blurry_light> like Linux+ 17:30 < jml2> kazdax, it's expensive though, but you have access to 1000s of other books and videos-- like ccna and much more.. 17:31 < jml2> kazdax, you pay a yearly subscription, but you get a 30-day free trial iirc 17:31 < jml2> kazdax, (i have the safarionline) 17:31 < jml2> or blurry_light 17:31 < plexigras> Psi-Jack: is echo `$(dirname "$PWD/$BASH_SOURCE")` ok? 17:31 < Psi-Jack> No 17:31 < jken> Does anyone know if it is possible to write a sparse image to a block device using DD? Doing so with the conv=sparse option right now causes a kernel panic when I try to boot it, but I am wondering if I just need to zero the entire disk first? 17:32 < jml2> jken, yeah 17:32 < blurry_light> jml2 kazdax this is what i have https://www.sandervanvugt.com/course/linux-foundation-certified-system-administrator-complete-video-course/ 17:32 < Psi-Jack> plexigras: Again, why do you need the script's path? 17:32 < blurry_light> this guy is good 17:32 < jml2> jken, but not to a DD -- "from a block device" 17:32 < jml2> jken, you can write to a sparse file using dd 17:32 < jml2> jken, one of the conv options 17:32 < jml2> blurry_light, who? 17:32 < plexigras> bc i am linking files in the same directory as the script 17:33 < Psi-Jack> plexigras: "because" not "bc" 17:33 < blurry_light> jml2 the website I linked, Sander Vanvugt 17:33 < jml2> blurry_light, the author of those redhat training books? XD. .. it's a belgium guy iirc 17:33 < jml2> blurry_light, yeah its him 17:33 < blurry_light> dutch 17:33 < seven-eleven> hi 17:33 < jml2> blurry_light, he also has a yt channel.. 17:33 < blurry_light> not Belgian 17:33 < blurry_light> right, yeah 17:33 < blurry_light> him 17:33 < jken> jml2, that's how I created the image. so writing it back to the block device with conv=sparse should work? 17:33 < blurry_light> he also has a bunch of RHEL stuff 17:33 < jml2> blurry_light, and puts other videos there as well.. you can find his books i think on sf-online as well 17:33 < seven-eleven> is there a way to tune a DHCP server to make it win a race situation if there are more than one DHCP servers in the same subnet 17:33 < Psi-Jack> plexigras: So, create a script to check for the existance of ~/.myscript.conf, or fallback to /etc/myscript.conf, and have that file provide a path to where the included scripts should be located. 17:34 < jml2> blurry_light, his whole series is available in safarionline -- you wouldn't find them on youtube... 17:34 < ayecee> seven-eleven: no 17:34 < jhodrien> seven-eleven: Many ways to win the race. Tuning to lose is easier. 17:34 < blurry_light> jml2: right of course, it's a paid course 17:34 < jml2> blurry_light, he writes redhat training books -- so he knows his stuff :) 17:34 < blurry_light> (which i paid for) 17:34 < blurry_light> jml2 indeed 17:34 < blurry_light> he's good 17:34 < jken> jml2, and its currently failing for me because I didn't zero the disk first? 17:34 < seven-eleven> ayecee, jhodrien okay sounds its not easy, then i dont try :-) 17:34 < ayecee> that's the spirit 17:35 < blurry_light> seven-eleven: then you will never learn 17:35 < jml2> blurry_light, you're paying a lot for a certificate less valuable than rhcsa imho 17:35 < kazdax> blurry i am using his 2014 RHCSA video tutorials 17:35 < blurry_light> jml2: i see 17:35 < kazdax> he is good but for some reason i prefer a book 17:35 < jhodrien> Winning a PXE contest is often easier than a broader DHCP battle. 17:35 < ayecee> blurry_light: never learn why people don't do that, maybe 17:35 < jml2> blurry_light, i'm betting evenutally the LF will drop their certification like Ubuntu did.. 17:35 < jml2> blurry_light, rh has been doing certifications/training for well more than i can remember XD 17:36 < blurry_light> jml2: hm thatd be a shame 17:36 < kazdax> blurry do you know other courses that require you to do hands on examination 17:36 < jml2> jken, ? 17:36 < kazdax> certifications i mean 17:36 < blurry_light> jml2: one of the things that bothers me is, with only a Red Hat cert if i go some company that doesnt use RHEL or CentOS or whatever they'll be like "go away" 17:36 < jml2> jken, you can save a stream of bytes from a disk ->> to a sparse file.. 17:37 < blurry_light> thats why the distro agnostic nature of LF certs appealed to me 17:37 < jml2> jken, but the filesystem needs to have "sparse support" 17:37 < jken> jml2, yes, I have done that. Now I want to go the other way. 17:37 < Psi-Jack> blurry_light: EVERYONE respects the RHCSA and more so, the RHCE. Because it's a real literal hands-on certification. 17:37 < jml2> jken, dd if=/dev/[a device] of=/path/to/sparsefile.raw conv=sparse iirc 17:38 < blurry_light> Psi-Jack: yes, but as i said, so are the LF certs. they have you do stuff in a virtual machine to solve real-life admin and engineer problems 17:38 < blurry_light> it's not like Linux+ 17:38 < jken> jml2, yep. now I want to `dd if=sparse.img of=/dev/sda conv=sparse` and am wondering it that's even possible. 17:38 < Psi-Jack> Never done the LF ones. Just Linux+ and RHSCA+RHCE. 17:38 < blurry_light> linux+ is pffft 17:38 < blurry_light> like, boring 17:39 < kazdax> LF seems new 17:39 < Psi-Jack> I know. 17:39 < blurry_light> memorize a bunch of crapola 17:39 < blurry_light> boom 17:39 < Psi-Jack> I even had to correct LPI on their questions. 17:39 < blurry_light> LOL 17:39 < blurry_light> seriously? 17:39 < jml2> jken, no it isn't because "sparse" is something that is only effective with the "underlying" filesystem. Not all filesystems can support "sparse files" 17:39 < jml2> jken, no it isn't because "sparse" is something that is only effective with the "underlying" filesystem. Not all filesystems can support "sparse files" 17:39 < jml2> jken, ^ 17:39 < Psi-Jack> Yes, Seriously. 17:39 < blurry_light> hahahahah 17:39 < blurry_light> thats awesome 17:39 < blurry_light> wow 17:39 < kazdax> yeblurry if you already bought the course..then go for it 17:39 < jml2> Psi-Jack, LPI with LF ? 17:39 < Psi-Jack> Only reason I even took the L+, Army used it. They probably stopped after me, because of the hell I caused. :) 17:40 < kazdax> hell to your own infantry :P 17:40 < jken> jml2, damn, I am trying to find a faster way to write a 30gb image to disk.. it was faster with conv=sparse but wasn't happy when I tried to boot it ¯\_(ツ)_/¯ 17:40 < blurry_light> wow now thats something 17:40 < jml2> jken, there is a fast way 17:40 < blurry_light> a candidate having to correct a thing in some cert 17:40 < jml2> jken, mount loop it with losetup or qemu-nbd --- then clone the files.. 17:41 < jml2> jken, but then you have to manually fix the boot-loader 17:41 < blurry_light> that proves how cruddy that cert is 17:41 < jml2> it seems Psi-Jack mixed LPI and LF together XDXD 17:42 < jml2> LOL 17:42 < Psi-Jack> jml2: You fool! :) 17:44 < jml2> LPI is more advanced than LF's cert... 17:44 < blurry_light> is it? isn't LPI just a written test? 17:44 < jml2> LPIC1, LPIC2 and has interesting LPIC3 options 17:44 < jml2> blurry_light, it's more advanced than LF's.. 17:45 < jml2> blurry_light, has more levels and has I think become quite popular -- I actually have two levels with them.. 17:45 < blurry_light> the thing here is its status in the industry ... you see as Psi-Jack said, RHCSA and RHCE are considered valuable because the actual test is done hands on an actual system 17:45 < blurry_light> so whether one is more "advanced" than the other isn't that relevant 17:45 < jml2> blurry_light, thats correct 17:45 < jml2> blurry_light, because RH doesn't trust multiple-choice exams.. they want to know you can actually do your shit 17:45 < blurry_light> right 17:46 < blurry_light> and apparently LF also thought that 17:46 < blurry_light> since their tests are also hands on 17:46 < jml2> blurry_light, so memorizing quiz banks is not necessarily a good thing 17:46 < blurry_light> exactly 17:46 < blurry_light> especially if the answer is wrong lol 17:46 < blurry_light> and they give you credit for it anyway 17:46 < Psi-Jack> 2+2=22, right? 17:46 < blurry_light> haha 17:46 < Psi-Jack> True: If concatenating strings. :) 17:47 < uplime> in shell code it is :| 17:47 < deo> true if using iphone calculator 17:47 < Psi-Jack> deo: Umm.. No. 17:47 < blurry_light> deo: what 17:47 < pepermuntjes> hi i have a question. Should i use bsd or linux, and use vim or emacs on it? 17:47 < uplime> actually he's right 17:47 < noodlepie> Linux/Emacs/Screen 17:47 < uplime> there was a bug where it would do that if you pressed buttons fast enough 17:48 < blurry_light> pepermuntjes: while this is more a Linux channel, my personal suggestion is to use all those combinations and stick with what you like 17:48 < blurry_light> as there's no one single objectively valid answer 17:48 < deo> Psi-Jack: it's an old one - https://9to5mac.com/2018/01/31/calculator-bug-ios-11-3-animation-returns/ 17:48 < Psi-Jack> uplime: No. 17:48 < jml2> my tde trinity setup so far is pretty... been trying to get it to work on me debian for the last 2 years.. finally this time it actually works! 17:48 < uplime> no wait im misremembering the bug 17:48 < uplime> https://9to5mac.com/2018/01/31/calculator-bug-ios-11-3-animation-returns/ i was thinking of this one 17:48 < Psi-Jack> heh 17:49 < uplime> ha mine still does it 17:50 < deo> yep lots of devices are still doing the broken math 17:50 < Psi-Jack> I use Calculator ∞ on my iPhone anyway. 17:50 < jml2> i just fire up my quake with "bc" when I need to do calculations 17:50 < uplime> https://xkcd.com/1537/ 17:50 < Psi-Jack> I needed more... 17:50 < jml2> nothing beats bc 17:50 < deo> Psi-Jack: the question would be why using iphone at all 17:50 < uplime> i rolled my own rpn that i used for a hwile 17:50 < Psi-Jack> jml2: I beat bc./ 17:50 < deo> :) 17:51 < Psi-Jack> With a dowel rod the diameter of my index finger. 17:51 < blurry_light> Psi-Jack: the last straw for me regarding Apple was when they started remotely downclocking iPhones to reduce battery usage 17:51 < jml2> short key with guake? and bc is right there? 17:51 < jml2> impossible! 17:51 < blurry_light> guake is awesome 17:51 < twainwek> hah. so all the 1+2 checks people do on their calculators before actually doing anything is legitimate 17:51 < blurry_light> i use f12 on mine, nice drop down terminal ;) 17:51 < Psi-Jack> guake? Awesome? No.. 17:51 < jml2> blurry_light, that's right guake is king of terminals! 17:51 < blurry_light> come on 17:51 < blurry_light> it's awesome. combine it with neofetch 17:51 < blurry_light> and you're banging 17:52 < Psi-Jack> Doesn't even come close to comparing to yakuake. 17:52 < jml2> faster than a ctl-alf-f1, I can be in terminal in a swift 17:52 < blurry_light> heard of yakuake never tried it 17:52 < Psi-Jack> But, now-a-days, I use i3. So, visor terminals just suck. 17:52 < blurry_light> yeah i like i3 for lower spec systems, really unburdens resources 17:52 < blurry_light> still workable 17:53 < Psi-Jack> I use i3 on my 16 core system with 64GB RAM. 17:53 < blurry_light> LOL 17:53 < blurry_light> you could put KDE on that 3 times and it wouldn't even notice 17:53 < Psi-Jack> Yet, I would notice. 17:53 * noodlepie has i7-8thread cpu 17:53 < Psi-Jack> KDE 5 is horribly buggy. 17:54 < Psi-Jack> I was a KDE user since pre 1.0.0. 5 was the tip of the iceburg to finally make me WANT something else. 17:54 < noodlepie> I like GNOME 17:54 < solidfox> current kde is pretty good 17:54 < solidfox> I remembered so many bugs in kde on kubuntu 16.04.0 17:54 < jken> jml2, how would I go about the loop method? 17:54 < Psi-Jack> I can't stand the flickering taskbar items after a few days of uptime, for absolutely no effing reason. heh 17:55 < jml2> jken, depends on how skilled you are XD 17:55 < jml2> jken, LOL 17:55 < j0seph> current KDE is aight. only gripe is how nvidia doesn't play with it so well. 17:55 < solidfox> Psi-Jack, lol, never noticed that one. its because I turn off my computer when I don't use it. 17:55 < blurry_light> solidfox: hopefully 18.04 LTS Kubuntu fixed that 17:55 < jml2> jken, what is is that you cloned? 17:55 < jken> jml2, willing to learn lol 17:55 < jken> jml2, its a debian image generated using mkosi 17:55 < jken> for a fleet of iot devices 17:55 < jml2> jken, mkiso? 17:55 < jken> mkosi* 17:55 < jml2> wtf is that 17:56 < jken> jml2, http://0pointer.net/blog/mkosi-a-tool-for-generating-os-images.html 17:56 < jml2> jken, ok, so basically you want to test an iso by dumping it to a disk? 17:56 < jml2> jken, why not just open it up in virtualbox? 17:56 < jken> exactly. mkosi gives me a sparse image. 17:56 < blurry_light> or compare its SHA1/md5 hashshum with the reported value where you got it from 17:56 < Alexander-47u> hi guys 17:56 < jml2> jken, I mean it is supposed to be a "Bootable iso" right? 17:56 < blurry_light> wouldnt that be easier to check integrity 17:56 < Alexander-47u> i just got a usb sound card 17:57 < jken> jml2, I know the image works, I am trying to find the quickest way to write it out to the devices (need to do ~1000 of them) 17:57 < Alexander-47u> its not being detect 17:57 < jml2> jken, wait 17:57 < Alexander-47u> however, i can hear sound when i run the following command 17:57 < Alexander-47u> speaker-test -Dhw:2,0 -c2 -twav 17:57 < jml2> jken, so you are NOT doing a "system restore" here 17:57 < blurry_light> jken: what devices are these? usb flash drives? 17:58 < jml2> jken, and you wouldn't want to create an iso with alot of "unused space" -- that would be stupid 17:58 < jken> jml2, blurry_light, intel compute sticks. I cant remove their disks, but I can boot anything via USB. Right now I have a 2nd image which contains the image I want and writes it out to the device's disk using d 17:58 < jken> dd* 17:58 < jml2> jken, sparse is only useful if you're backing up a system.. 17:58 < Alexander-47u> anyone? 17:58 < jml2> jken, not for creating isoz faster... 17:58 < Psi-Jack> Alexander-47u: Yes? 17:59 < jml2> jken, otherwise you're making isoz ineffectively in the first place 17:59 < jken> jml2, well mkosi gives me a sparse image with a / parition the size of the device disk (~30GB), file size is ~600mb 17:59 < Psi-Jack> Alexander-47u: I am anyone, speaking to you with Psi-Jack's keys. What did you need? 17:59 < Alexander-47u> Psi-Jack, got an usb sound card 17:59 < jken> jml2, I did that so when I write it to the device, it can use its entire disk for storage 17:59 < Alexander-47u> its not being detected in ubuntu 17:59 < Alexander-47u> speaker-test -Dhw:2,0 -c2 -twav 17:59 < jml2> jken, well then you have to go back to STAGE 1 and fix the size of your target iso 17:59 < Alexander-47u> running that command does produce sound 17:59 < Alexander-47u> how do i configure it to be detected 17:59 < Psi-Jack> Alexander-47u: Enter is not punctuation or pauses in thought. 18:00 < jken> jml2, I can make it as small as I want, but then Ill need to figure out how to grow the / partition after writing it to the device. 18:00 < Psi-Jack> uplime: You and uptime need to merge. 18:00 < jml2> jken, well then that has nothing to do with your original question 18:00 < jml2> jken, you're asking about how to create a distribution, and I'm not here for this :))) 18:00 < jken> jml2, Right you are.. 18:00 < Psi-Jack> uplime: That xkcd earlier was good. heh 18:00 < uplime> what about updime and uprime? 18:01 < uplime> Psi-Jack: yeah its one of my favorites :D 18:01 < Psi-Jack> uplime: heh, dunno about them. They're not here. :0 18:01 < uplime> :o you're right 18:01 < Psi-Jack> uplime: But uptime is here. 18:02 < T-Rog> So I heard Gnome3's backend is JavaScript 18:02 < Psi-Jack> Destroy him. Take his disc and merge it with yours. 18:02 < Psi-Jack> T-Rog: And CSS 18:02 < j0seph> Psi-Jack: and HTML 18:02 < T-Rog> So I heard KDE's backend is JavaScript 18:02 < j0seph> Gnome is Electron confirmed 18:02 < Psi-Jack> T-Rog: You heard wrong. 18:03 < jken> I guess my question needs to be, what is the best way to extend a partition and grow the ext4 filesystem automatically via a "live cd" (my other image used for writing the desired image) 18:03 < Psi-Jack> And you must be penalized for it... By fetching a shrubbery! 18:04 < Psi-Jack> Alexander-47u: Do you think people 18:04 < T-Rog> Psi-Jack: what is kde's backend then? 18:04 < Psi-Jack> T-Rog: Qt 18:04 < T-Rog> Uh 18:04 < Psi-Jack> Really. 18:04 < Alexander-47u> does anyone have experience with this ? 18:04 < T-Rog> I thought Qt was a way to make guis 18:04 < T-Rog> And qt integrated with c++ 18:04 < Psi-Jack> Alexander-47u: like to try to read 18:05 < Psi-Jack> Qt is a library 18:07 < jken> jml2, this might be the proper approach.. https://packages.debian.org/sid/cloud-initramfs-growroot 18:07 < jken> Thanks for the help 18:07 < T-Rog> Psi-Jack: Budgie is moving to Qt right? 18:07 < Psi-Jack> Alexander-47u: unfocused multi-line messages? 18:07 < Psi-Jack> T-Rog: I pay no attention to Budgie. 18:07 < jml2> you want to shrink a 30gb file for 600 megs for your linux distribution? 18:07 < Alexander-47u> Psi-Jack, :p.. 18:07 < jml2> pfff! nobody will want to wair 3 hours for it to complete! 18:08 < Psi-Jack> Alexander-47u: Not fun is it? 18:08 < jml2> bad bad designing 18:09 < Alexander-47u> I want my sound card to work lol, dont care for broken lines. 18:09 < Psi-Jack> Here's your cake. No you can't eat it. No you can't have the cherry on top. :) 18:10 < LambdaComplex> ...Can I have a chocolate milkshake? 18:10 < uplime> we only serve cake here, sorry 18:10 < ozymandias> cake, or death 18:10 < ozymandias> https://www.youtube.com/watch?v=rMMHUzm22oE 18:11 < T-Rog> LambdaComplex: AnomalousMaterials 18:11 < LambdaComplex> hey, someone gets the reference 18:11 < LambdaComplex> neat 18:11 < T-Rog> Prepare for unforseen consequences 18:12 < LambdaComplex> i want a milkshake now 18:12 < LambdaComplex> dammit 18:12 < T-Rog> Does Gnome Team have a plan to fix the memory leak? 18:13 < promach2> I have some syntax error at https://gist.github.com/promach/e5b5584a8b3a85bae14fdfcb6263750e#file-zynq-common-h-L272 Could anyone help ? 18:13 < LambdaComplex> T-Rog: i'd tell you it's a feature, not a bug, but the gnome team seems to like getting rid of features 18:14 < SuperSeriousCat> T-Rog, https://bugzilla.gnome.org/buglist.cgi?bug_status=__open__&content=memory%20leak&no_redirect=1&order=Importance&product=&query_format=specific 18:14 < T-Rog> I think they should just get rid of their desktop and tell everyone tonuse a different DE 18:15 < Psi-Jack> T-Rog: Did you get the TPS report done? 18:15 < T-Rog> SuperSeriousCat: jesus fucking christ 18:15 < Psi-Jack> Kindly mind the language, 18:15 < T-Rog> Are all of those related to memory leaks 18:15 < T-Rog> Sorry 18:15 < turkeyhand> with arch, if I have a similar computer 18:15 < SuperSeriousCat> I just searched "memory leak". Probably not all 18:16 < T-Rog> Psi-Jack: TPS report? 18:16 < turkeyhand> basically an identical computer, thinkpad x220 tablet and x220 (normal) and I switch the SSD from one to the other 18:16 < Psi-Jack> T-Rog: you didn't did you? Well, I'm going to need you to come in on the weekend and finish that report then! 18:16 < turkeyhand> will it boot arch linux? 18:16 < turkeyhand> does the motherboard have any special... shit 18:17 < T-Rog> Oh man I just wrote a report on Natural Language Processing 18:17 < Psi-Jack> turkeyhand: hi! did you know that hitting enter as 18:17 < turkeyhand> what 18:17 < LambdaComplex> turkeyhand: make sure the drive is specified via UUID in the fstab file and i'll go with "it'll probably work" 18:17 < Psi-Jack> turkeyhand: pauses in thought is really hard and 18:18 < jcjordyn120> what would cause virtual memory to be exhausted even if I have a large swap file and 90% ram free? 18:18 < LambdaComplex> Psi-Jack: do you offer lessons on complaining? i think i'm pretty good at it but i kinda wanna get to the next level 18:18 < turkeyhand> the drive is going to the same location 18:18 < Psi-Jack> turkeyhand: annoying for otehrs to read, and is 18:18 < turkeyhand> the msata slot 18:18 < turkeyhand> are you drunk Psi-Jack 18:19 < LambdaComplex> no, you're just annoying him 18:19 < LambdaComplex> well, he might be drunk. idk 18:19 < T-Rog> How can I install the arc theme on a system where I do not have administrative privileges 18:19 < LambdaComplex> turkeyhand: specifying via UUID is still way better than specifying via device file 18:19 < Psi-Jack> turkeyhand: very annoying. Don't use enter as if punctuation or pauses in thought. 18:19 < LambdaComplex> so i'm gonna say do that anyways 18:19 < pepermuntjes> lets all correct each other 18:19 < LambdaComplex> pepermuntjes: *let's 18:20 < turkeyhand> it's the same location, sdb 18:20 < turkeyhand> it's nearly an identical computer 18:20 < turkeyhand> I only say nearly because this is mising a touchscreen and a wacom pen input 18:21 < BCMM> what should i use to batch-convert FLAC music to Opus? i know this is easily shell-scriptable, but i figure it's gotta be a solved problem 18:21 < turkeyhand> I had to butcher both computers together just now so this one had an IPS display 18:21 < BCMM> i would particularly like a solution that will gracefully ignore existing Opus files, so i can efficiently re-run it after adding more FLAC 18:22 < LambdaComplex> BCMM: write a shell script wrapping ffmpeg, probably 18:22 < BCMM> LambdaComplex: i figured that might be the answer, but this has to be common problem... 18:22 < SuperSeriousCat> If you like to keep a nicly tagged music lib, you got beets who can convert for you 18:23 < SuperSeriousCat> Your whole music library gets saved in a database it maintain 18:23 < BCMM> SuperSeriousCat: i'm happy to assume the FLAC metadata is correct and just transfer it to the opus 18:24 < BCMM> SuperSeriousCat: ooh, beets does look interesting though 18:24 < SuperSeriousCat> The dev is active in #beets if you run into problems with it 18:36 < MrPockets> Well. This is strange. 18:37 < MrPockets> This ubuntu server seems to be accepting SSH connections on port 53. Netstat doesn't show 53 open, and I've configured iptables to reject traffic to 53 18:37 < LambdaComplex> neat 18:38 < LambdaComplex> MrPockets: try telnetting to port 53 and see if you get some openssh message 18:38 < MrPockets> I sure do. 18:40 < MrPockets> I sure do. 18:40 < oxagast> are you trying to connect to it from the outside, or to localhost? 18:40 < MrPockets> externally 18:40 < LambdaComplex> neat 18:41 < MrPockets> no NAT involved. 18:41 < oxagast> that was my next question, some NAT shit going on 18:41 < BCMM> SuperSeriousCat: this really is the nicest interface for cleaning up metadata I've ever used. thanks for the recommendation. 18:41 < MrPockets> This server's directly on the internet. I'm connecting to it ffrom over the internet. 18:41 < rypervenche> MrPockets: What netstat command are you running? 18:41 < BCMM> SuperSeriousCat: (i've forgotten all about transcoding stuff to fit on my phone, because i'm not going outside until all my metadata is fixed now) 18:41 < MrPockets> I can SSH and auth to it on port 53. Netstat on my locla workstation shows a connection to the server's WAN IP, on port 53. Local netstat on the esrver shows it conected to port 22 18:42 < SuperSeriousCat> BCMM, np :) 18:42 < MrPockets> Something, is redirecting port 53 to port 22? 18:43 < BCMM> SuperSeriousCat: is beets automatically arranging music in to a directory structure, or is it just taking the existing dir structure? 18:43 < rypervenche> MrPockets: ss -plunt to see what's listening, and "iptables -nvL -t nat" to see any silliness going on. (run both as sudo/root) 18:43 < oxagast> MrPockets: are you on a trusted network? 18:44 < oxagast> or did you get one of those Someone could be doing something nasty messages when you auth'd? 18:44 < BCMM> i think this thing gets extra points for knowing that music belongs in Artist/Album/01 Title.flac, which seems to be weirdly controversial these days 18:44 < SuperSeriousCat> You decide in the config file. "beet config -e" to open it. Read the documentation 18:44 < oxagast> it sounds to me like someone is messing with you 18:44 < MrPockets> oxagast, as far as I'm aware, I'm on a trusted machine on a trusted network 18:44 < MrPockets> i AM, however, connecting over the internet, 18:44 < BCMM> SuperSeriousCat: i *think* it's already rearranging it but the defaults are what i want anyway 18:45 < MrPockets> hey 18:45 < SuperSeriousCat> You can do folder name by tags of your choosing. Put them in different "buckets" by genre or other tags and basically anything you like by tweaking the config file 18:45 < MrPockets> rypervenche, -nvL 18:45 < MrPockets> shows some kinda REDIRECT 18:45 < rypervenche> MrPockets: Well, we can't help if we're left guessing what it is :) You can pastebin stuff if you'd like some help understanding it. 18:45 < MrPockets> understandable! Thanks! 18:45 < MrPockets> I'm digging into it atm 18:46 < SuperSeriousCat> It can also find corrupt files with an addon. Nice to check if you copy to a new HDD or similar 18:46 < rypervenche> MrPockets: The router could also be doing the magic. 18:47 < MrPockets> whoops. Just iptables -F 'd the server.. 18:47 < rypervenche> LOL 18:47 < MrPockets> Worth mentioning, though, that i can no longer connect on 53 o_0 18:47 < rypervenche> hahahahahha 18:48 < rypervenche> If INPUT was set to default DROP, then you're going to have to force a reboot or get someone with physical access to it to do it. 18:48 < rypervenche> Otherwise, you should be able to get in on the default SSH port. 18:48 < MrPockets> So I'm confused as to why my iptables script, that starts with -F and seemingly has no mention of 53, and only the single line permitting 22 from all, didn't undo this forward. 18:48 < MrPockets> rypervenche, indeed. But, FIRST! I'm going to re-run this vuln scan that's been giving me issue s;-) 18:49 < rypervenche> MrPockets: I don't know that you should be doing that first... 18:49 < MrPockets> sshhh 18:49 < rypervenche> MrPockets: Depending on your distro, you may have a firewall rules file being loaded by an init script elsewhere that was set. 18:50 < rypervenche> mkay, you do you 18:50 < Selfsigned> so now you have to learn emacs and vim uh 18:51 < Selfsigned> Oh man my weechat buffer is always scrolled way up 18:51 < Selfsigned> I should check if there's a way to have some kind of indicator 18:53 < mattfly> hi 18:55 < oxagast> i don't know why by default iptables would have a rule to defer 53 to 22 18:56 < oxagast> that seems weird 18:56 < uplime> DNS over SSH is the next big thing 18:56 < revel> Default according to what? 18:56 < rypervenche> They wouldn't. Someone else would have done it. But he hasn't given us much information to work with. So we'll have to assume that he's not the only one with access to the machine. 18:56 < oxagast> someone said a init script from a distro 18:57 < mawk> show us that rule 18:58 < jack3> Hello. Ubuntu 16. ran out of space. deleted some logs, space not reclaimed. Googled. Found lots. Most start with this: lsof +L1 to find files held open. But when I run that command... it finds nothing. Not yet found what to do when that happens. Help with how to reclaim space would be greatly appreciated. 19:00 < rigid> what could be the reason my raid1 on /dev/nvme0n1p1 and /dev/nvme1n1p1 isn't detected on boot? I have CONFIG_MD=y, CONFIG_MD_AUTODETECT=y and CONFIG_MD_RAID1=y also CONFIG_NVME_CORE=y and CONFIG_BLK_DEV_NVME=y... grub is also setup correctly 19:05 < MrKeuner> Hi, how can use capturing groups in gedit regex search? () and $1 combination doesn't seem to work 19:08 < lukey_> rigid: I never got Raid autodetect by the kernel to work properly. AFAIK you at least need the old metadata format i.e. 19:09 < rigid> lukey_: but strangely it gets detected by a live CD without problems 19:10 < seven-eleven> can you configure consumer routers to have two subnets and route all incoming traffic from subnet A through the default gateway of subnet B? in linux you would just add one default route like so `ip route add default gateway ip_of_subnet_B` and also do `# iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE` 19:11 < djph> seven-eleven: likely "no". 19:11 < SkunkyFone> seven-eleven: unlikely, unless it's a router that you can put wrt on 19:11 < seven-eleven> mhm ok, good to know thanks :-) 19:11 < djph> seven-eleven: but then it depends on "what" you consider a "consumer router" to be -- e.g. *wrt or something from Ubiquiti / Mikrotik ... 19:12 < seven-eleven> djph, yeah like $40 asus wlan routers :-) 19:13 < seven-eleven> for my purpose (just need one ethernet interface) i'll just use a raspberry pi then 19:13 < djph> probably not (although, *maybe* that plus whaterver-wrt). Definitely the (about) $50 UBNT ER-X can do that (but no wifi) 19:13 < snugger> Why, for example, might I switch from Zsh to something like Dash or fish? In my opinion zsh is already the go-to Bash alternative 19:13 < lukey_> rigid: Because it detects it by mdadm in the initrd probably not by the kernel 19:13 < snugger> replacement* 19:13 < uplime> snugger: because different people like different things 19:14 < djph> seven-eleven: although, if you're not concerned about it considering the other LAN as "WAN", cheapo consumer routers can route to upstream 19:14 < uplime> zsh, bash, fish, and dash are just different shells used to administer your system 19:14 < uplime> none are inherently better than the other 19:14 < revel> I don't think people use dash for its extensive feature set. 19:14 < uplime> people use dash because its POSIX 19:15 < revel> And small. 19:15 < uplime> yep 19:16 < revel> So, if you want fun features or something, then {ba,z,fi}sh is probably better. Pick whichever one floats your boat. 19:16 < rigid> lukey_: yeah, that's it i guess. I have no initrd 19:16 < seven-eleven> djph, i have an LTE router which is in another subnet than my LAN, so I would connect the LTE router to a linux box and use the linux box as gateway for the LAN 19:16 < rigid> lukey_: thanks 19:17 < seven-eleven> djph, i think using a cheap pi v3 will do the job and shouldn't bottleneck for a 150 MBit WAN, latency should be fine too 19:18 < snugger> It's funny how some people take this for something serious 19:18 < snugger> http://harmful.cat-v.org/software/ 19:19 < snugger> >GNU screen is considered harmful software 19:19 < snugger> >musl is considered a replacement for glibc 19:20 < lukey_> >head is considered harmful software 19:20 < uplime> you definitely want to get rid of head 19:20 < uplime> thats why there are so many headless servers out there 19:21 < mawk> "Harmful things that are so superfluous and useless that require no alternative: Dynamic linking, in particular in Unix systems." 19:21 < snugger> >vim 19:21 < snugger> >emacs 19:21 < snugger> >useless 19:21 < uplime> it would be great if they could explain why its useless 19:21 < uplime> or harmful 19:22 < snugger> These people think that if anything has less lines of code it means it's better 19:22 < lukey_> mawk: We now have Docker because devs are to lazy to build their software statically 19:23 < uplime> lol awk is not a replacement for perl or ruby 19:24 < mawk> yeah lukey_ 19:26 < mutante> Harmful: Jabber Less harmful: IRC What they are actually using now: SLACK .booo 19:26 < sauvin_> Yeah, that page is pretty comical. 19:27 < sauvin_> http://harmful.cat-v.org/software/OO_programming/ 19:27 < sauvin_> OOP is also apparently inherently harmful. 19:31 < snugger> “XML is like violence. Sure, it seems like a quick and easy solution at first, but then it spirals out of control into utter chaos.” – Sarkos in reddit 19:32 < snugger> "reddit" 19:32 < lukey_> "http" 19:33 < aw1> on macos the makefile http://codepad.org/a8BuqtdP produces no output ... as if foo was never defined ... while the same makefile works in linux ... could somebody help me understand that? 19:34 < uplime> because they're different make's? 19:34 < samort7> join /#vagrant 19:34 < uplime> and im pretty sure thats not a standard way of defining make variables 19:35 < Alexander-47u> hi all 19:35 < Alexander-47u> anyone know what the normal cpu load of pulseaudio should be? 19:35 < aw1> uplime, i see. thanks. 19:37 < Alexander-47u> brb 19:40 < no_gravity> Hello! I have a bunch of files where I want to change all occurances of lazyPlanetAnimal to lazySunAnimal. I can find the files with 'grep -r lazyPlanetAnimal'. What's a good way to do the replacements? 19:41 < ]|[[|]|[[]> sed 19:43 < no_gravity> ]|[[|]|[[]: How would you feed all the files to sed and make it write them back? 19:44 < revel> -i? 19:45 < ]|[[|]|[[]> find . -type f -exec {} \; 19:45 < ]|[[|]|[[]> only add -i once you verify it works as intended because it will modify the file 19:45 < rypervenche> sed -i --follow-symlinks is recommended. 19:45 < rypervenche> Otherwise it overwrites symlinks as real files. 19:53 < ayecee> TIL 19:54 < zack6849> ]|[[|]|[[]: did username verification regexes kill your family? 19:54 < Psi-Jack> Obviously. 19:55 < Psi-Jack> That, once again, is dannylee, aka stupid1, aka,... Too many to list. :p 19:55 < zack6849> am i supposed to know who that is tho 19:55 < MyLiege> just ignore the channel troll, zack6849 19:55 < zack6849> his whois looks fun though 19:56 < zack6849> ttps://www.zack6849.com/uploads/eEjKyXaxnBMX0.png 19:56 < Psi-Jack> zack6849: Ahh, not a regular denizen here? 19:56 < zack6849> nope, just lurk and often don't read 19:56 < LambdaComplex> hi zack6849 19:56 < zack6849> oh no 19:56 < zack6849> it's here 19:56 < zack6849> hi LambdaComplex 19:56 < twainwek> why isn't follow-symlinks default? 19:56 < zack6849> how are you today 19:57 < mawk> you forgot the h zack6849 19:57 < mawk> now because of you I must copy paste the link into chrome instead of just clicking on it 19:57 < mawk> you ruined my evening 19:57 < zack6849> that's impressive, i'm not even sure how i did that 19:57 < zack6849> my script posts it to my xclip 20:00 < mawk> you script was stripping the h all this time 20:01 < zack6849> no 20:01 < zack6849> lies and slander 20:02 < MyLiege> just like the other false accusations flying around here 8-) 20:03 < laggger164> I was just looking around on ebay for some RAM for my laptop and I found this: 20:03 < laggger164> https://www.ebay.com/itm/For-Kingston-16GB-2x-8GB-DDR3L-1600M-Hz-Laptop-Memory-PC3L-12800-SODIMM-Notebook/302720299413?hash=item467b893195:g:LY8AAOSwNEBa5GzS 20:04 < laggger164> Am I the only one calling BUUUUULSH**? 20:04 < Psi-Jack> laggger164: ##hardware ? 20:04 < ayecee> what does that even mean, "am i the only one" 20:04 < laggger164> Psi-Jack: OK, it's just that this is a bigger community :D 20:04 < MyLiege> it means only YOU can have an off topic convo in this channel 20:04 < Psi-Jack> laggger164: This is a Linux support channel, not a hardware support channel. 20:05 < ayecee> i mean, it's obviously a rhetorical device, but what are you trying to say with it? 20:05 < laggger164> Psi-Jack: OK, I'll stop, but I was asking for an opinion, not for technical support. 20:05 < MyLiege> about some mushrooms your dog eats because you don't feed it well enough 20:05 < laggger164> Psi-Jack: OK, I'll move on though. 20:05 < ayecee> what are you asking an opinion about? 20:05 < Psi-Jack> sauvin_: Still here? 20:06 < laggger164> ayecee: About what you think of the stuff in the link I posted 20:07 < sauvin_> Psi-Jack, what's up? 20:07 < ayecee> laggger164: why would i have an opinion on that? 20:07 < laggger164> ayecee: Is your opinion that it's a scam or not? 20:07 < ayecee> ah, that's a more sensible question 20:08 < ayecee> it seemed like you were pretty convinced that it was. 20:08 < laggger164> ayecee: I am, so I am asking you guys 20:08 < laggger164> I don't know if I am wrong... 20:08 < ayecee> i don't think we'd know either 20:13 < Trieste> I've got a bit-for-bit same binary that acts differently on two different machines, is there a way to trace which files the binary accesses, etc.? 20:14 < ayecee> strace, perhaps 20:14 < hexnewbie> Trieste: strace to trace system calls, ltrace to trace library calls (less useful for what you ask) 20:15 < toothe> I am running Kali on these test laptops, but they constantly change the IP with the network manager. 20:15 < toothe> Is there a way to disable that? 20:15 < toothe> disable that even at reboot. 20:15 < Psi-Jack> toothe: Easy. Don't use Kali. 20:15 < ayecee> that's your answer to every kali question :P 20:15 < Psi-Jack> It is. :) 20:15 < ayecee> should probably keep it to yourself. 20:15 < Trieste> I'll take a look, thanks 20:16 < SuperSeriousCat> What about Parrot? 20:16 < toothe> Psi-Jack: we have to. 20:16 < toothe> its not my choice. 20:16 < SuperSeriousCat> Never tested it or Kali. Just seen it mentioned as a better Kali alternative 20:16 < mutante> blames network-manager 20:16 < Psi-Jack> ParrotSec is pretty good so far in my minimal experience with it so far. 20:16 < toothe> I'm really annoyed at its Sandboxing. 20:16 < Psi-Jack> toothe: How is it not your choice? heh 20:17 < toothe> Psi-Jack: These aren't my laptops. I'm working for a company whose policy said use Kali. 20:17 < mutante> Hacking Academy? 20:17 < Psi-Jack> ... 20:17 < toothe> these are assessment laptops. 20:17 < Psi-Jack> Leave, Find a new job, heh 20:17 < triceratux> toothe: the MrRobot Cult Foundation ? 20:17 < ayecee> enough distro bashing plskthx 20:17 < toothe> ? 20:17 < mutante> but if he uses another distro and network-manager still sucks, what then 20:18 < toothe> I don't really wish to discuss why/how I have to use this specific distro, but it entirely makes sense in this context. 20:18 < toothe> These are assessment laptops. 20:18 < Psi-Jack> toothe: #kali-linux 20:18 < sauvin_> Kali has its place. My personal viewpoint is that said place isn't on the desktops or laptops of newbies. 20:18 < toothe> Psi-Jack: I believe this is a general MATE thing, no? 20:18 < ayecee> an assessment laptop would be a place for it though 20:18 < Psi-Jack> toothe: No. 20:18 < toothe> ahh yes! network-manager! 20:18 < Psi-Jack> It 20:18 < Psi-Jack> It is a uniquely specific Kali setup. 20:18 < toothe> mutante: yes ,that's it. I need to figure out how to kill that. I'll google it, I Just needed to know its name :). 20:19 < sauvin_> or even just google the whole problem. 20:19 < toothe> thanks everyone! 20:19 < triceratux> toothe: technically its NetworkManager. killing it can be good. manage your resolv.conf & all that stuff by yourself 20:19 < toothe> and Kali isn't that bad. I don't get the Kali-hate. 20:20 < mutante> toothe: https://miloserdov.org/?p=542 20:20 < sauvin_> toothe, you'd have to hang out *here* for a while to get it. 20:20 < Psi-Jack> heh 20:20 < toothe> I don't run Kali as my desktop OS. That might be odd. 20:20 < Psi-Jack> toothe: Others do. 20:20 < sauvin_> It'd be very odd. It's broken. 20:20 < ayecee> i don't 20:20 < toothe> But its a decent platform to use in airgrapped environments when you can just throw away the OS and start a'new. 20:20 < uplime> it comes with a lot of tools configured towards a specific environment, makes it "easy" to use but these aren't meant to be easy to use tools, and it breaks very easily 20:20 < ayecee> and i hang out here a lot 20:20 < toothe> which is what we routinely do. 20:20 < invisisith> kali...schuck thee day. - Indiana Jones Temple of Doom 20:21 < toothe> After almostevery assessment we throw away the entire install and re-image it. 20:21 < mutante> it's not even related to the distro, it's just the normal network-manager and "how to set a static IP" thing.. 20:21 < florianbAT> someone mentioned kali? 20:21 < toothe> If you want to be really really upset -- the client thinks that data is FOREVER on a hard-drive, s owe routinely destroy 500gb hard-drives that have only been used for 2-3 weeks. 20:21 < ayecee> florianbAT: yup. need a pitchfork? 20:21 < ayecee> --E 20:22 < sauvin_> toothe, yeah, that's a felony. 20:22 < florianbAT> ayecee: :p 20:22 < toothe> I mean that literally - we ordered a box of a few hundred 500 GB hard-drives. We install Kali once, go on assessment, then destroy the hard-drive after week 2-3. 20:23 < ayecee> toothe: i guess that takes care of a lot of potential carelessness 20:23 < triceratux> certainly sounds like yer serious about security 20:23 < toothe> ayecee: I know. When I learned that was the case, I was personally upset. 20:23 < ayecee> i can understand the approach, even though i probably wouldn't do it myself 20:23 < hexnewbie> How do you destroy the drives? 20:23 < Psi-Jack> Why even "install" kali? Just shove it on a USB stick like it was designed ror. 20:23 < sauvin_> That's throwing away an awful lot of money for an extreme solution to a perceived *potential* problem. 20:23 < Psi-Jack> for* 20:23 < ayecee> drives are cheap, data leaks are expensive 20:24 < toothe> I don't know ,honestly. 20:24 < plexigras> why does `echo a/*` not work? 20:24 < SuperSeriousCat> It dont matter if it cost a lot of money. Its cost gets pushed to the clients 20:24 < hexnewbie> plexigras: It works. What do you expect it to do? 20:24 < toothe> SuperSeriousCat: well, in this case, the tax payer. 20:24 < florianbAT> toothe: BAE? 20:24 < plexigras> to print all files that are inside the dir a 20:24 < hexnewbie> plexigras: Do the files that don't get printed start with a dot? 20:24 < florianbAT> anyways, wouldn't it be cheaper to just use USBs and then destroy them afterwards? 20:25 < ayecee> plexigras: what does it do instead? 20:25 < triceratux> toothe: but doesnt a practice like that contribute needlessly to global warming ? 20:25 < plexigras> hexnewbie: yes 20:25 < toothe> they bought really really expensive laptops. Guess what? the environment we work in has to be air-gapped. So they physically removed the wifi PCIe cards. Then destroyed them. 20:25 < toothe> I asked to keep one. 20:25 < ayecee> lol 20:25 < toothe> Because I want to write a driver for it on BSD. They said no and destroyed it. 20:25 < ayecee> brute force security 20:25 < toothe> want me to go on? We spent $1 million on super beefy machines that run and do nothing. 20:26 < hexnewbie> plexigras: Globbing doesn't expand to dot-files. In Bash, you could use ‘shopt -s dotglob’ to change that. 20:26 < ayecee> nah 20:26 < toothe> beacuse they were not properly security baselined. 20:26 < ayecee> sounds rage inducing, and we already have enough of that here. 20:26 < hexnewbie> plexigras: Or you could use find 20:26 < uplime> plexigras: it does work, files that start with a dot are "hidden". you can use shopt -s extglob, then do something like echo !(.|..)* 20:26 < ayecee> toothe: just be glad it's someone else's money :) 20:27 < Psi-Jack> toothe: That sounds almost as bad as the US Navy kicking millions of dollars of equipment into the ocean, just because next year budgetting purposes. 20:27 < SuperSeriousCat> If it is taxmoney it is partly his money too 20:27 < hexnewbie> * with extglob still shouldn't match . and .. 20:27 < uplime> hexnewbie: it doesn't, bncause * isn't an extglob 20:27 < plexigras> hexnewbie: uplime and when i want to have this in a shell script i dont need to disable it aferwards? 20:27 < uplime> !(.|..)* is though 20:27 < uplime> plexigras: correct 20:27 < uplime> extglob is only enabled for that instance of the script 20:27 < uplime> it doesn't affect your parent shell 20:27 < hexnewbie> uplime: Er, sorry, that's second time today I confuse dotglob with extglob 20:28 < uplime> s'ok 20:28 < plexigras> ok thanks 20:28 < toothe> oh, and Parrot OS is pretty decent, their terminal PS1 variable is really pretty. But I would prefer just stock Ubuntu. 20:28 < hexnewbie> toothe: And wallpaper is cool 20:29 < Psi-Jack> it's all about the wallpaper. 20:29 < Psi-Jack> Nothing else matters. 20:29 < hexnewbie> Well, PS1 is more important :p 20:29 < Toadisattva> I've been a debian kid for years and I'm looking to try out arch on a machine, is there a popular learning distro anyone would recommend along gthe lines of the buntus with debian? 20:30 < bls> Toadisattva: there's manjaro, but it's had mixed reviews 20:30 < rendar> if i have an user, let's say git, and i want that another user let's say rendar can write everything in /home/git, as if /home/git would have owner rendar, what is the best way to do this? 20:30 < Toadisattva> manjaro eh, thanks I'll check it out 20:31 < Psi-Jack> Toadisattva: I wouldn't. 20:31 < Toadisattva> no? you got an arch distro you like? 20:31 < Psi-Jack> It's a really bad insecure unstable fork of arch. 20:31 < Psi-Jack> Toadisattva: Sure. Arch. 20:31 < Toadisattva> mm okay 20:31 < Toadisattva> just standard arch okay 20:32 < Psi-Jack> At the most, Antergos, which doesn't overlap Arch too much. 20:32 < bls> Toadisattva: part of arch's popularity is its lack of user friendliness. user friendly variants of it are frowned upon 20:32 < Psi-Jack> Not entirely true. :) 20:33 < Psi-Jack> Arch is definitely not user friendly for everything that likely matters most, that's for sure. If you don't truly know Linux strongly, don't use Arch. it's not for you. 20:33 < Toadisattva> I'm okay with that I've put in enough time wtih debian I feel capable of learning to navigate arch, but a bit of hand holding int eh beginning would probably facilitate that 20:33 < hexnewbie> It's not the user friendliness, it's that they skip on security updates to make it ‘not rolling’. (And user friendliness is subjective) 20:33 < Psi-Jack> Toadisattva: There will be no hand holding. 20:34 < Psi-Jack> And what you learn with arch will mostly be arch-specific stuff. 20:34 < Psi-Jack> Heh, I wonder if Manjaro even got the retpoline patches yet. 20:35 < bls> eh, I've seen pretty much every user assistive tool scornfully labelled bloat by the arch community, but yes that's the community and not the distro itself 20:35 < Psi-Jack> They are horrendous regarding security. 20:35 < Psi-Jack> Yep. 20:35 < Psi-Jack> Oh, you use an AUR wrapper? You are worse than scum! heh 20:35 < Psi-Jack> That's Arch's "Community" 20:36 < zack6849> Psi-Jack: i dont see that often at all? 20:36 < zack6849> they aren't supported, depending on22:00 < knucker_> i made a program in go i execute the commands and start them in the background so my other code can run without block by waiting each command to finish . i want to check by pid of each program to see if finish or exit by eror each time the user type the pid number 22:01 < bls> knucker_: you can send the programs/pids the 0 signal to see if they're still running 22:01 < bls> knucker_: or you can set up a signal handler for sigchld 22:02 < knucker_> the singal handler will return any error if the program exit with error ? 22:04 < bls> no 22:04 < knucker_> some people tell me to use a rabbitmq or something familiar if i want to total control them without interput my other code 22:04 < knucker_> but i wanted to learn some more linux methods about this 22:04 < bls> that sounds like massive overkill in the absence of any other requirements 22:05 < knucker_> this is what i have thinking that is overkill 22:05 < knucker_> and study to find a better solution 22:07 < bls> since you're programming this, you can also just do a loop over your children using a non-blocking wait/waitpid call 22:07 < dgurney> well I guess that's one way of taking care of your children 22:08 < revel> "how to kill all children while leaving the parents alive?" 22:09 < revel> "What do you mean I'm on a watchlist now?" 22:11 < djph> revel: arsenic in the koolaid 22:12 < The_Schmidt> parents also drink koolaid 22:15 < djph> The_Schmidt: sshh 22:21 < The_Schmidt> is that a new shell? 22:21 < Psi-Jack> knucker_: Here's a question. Why are you writing something in go that forks external processes at all? Heh 22:22 < revel> No, it's an extension to ssh. 22:23 < uplime> super secure shell 22:23 < jim> maybe he likes the way that's expressed in go 22:23 < SmashingX> so if I create a file under /etc/sudorers.d/sampleuser 22:24 < SmashingX> and then inside of that file I put: sampleuser ALL=(root) NOEXEC: /usr/sbin/httpd 22:24 < SmashingX> can I make it not to ask for a password as well? 22:25 < revel> With NOPASSWD, I think? 22:25 < SmashingX> NOPASSWD: ALL 22:25 < SmashingX> sampleuser ALL=(root) NOEXEC: /usr/sbin/httpd NOPASSWD: ALL 22:25 < SmashingX> would that be correcT? 22:25 < Psi-Jack> ALL means anything. 22:26 < Psi-Jack> I don't think that's what you want, but you want NOPASSWD, possibly. 22:26 < revel> Haven't really edited the sudoers file... 22:26 < SmashingX> so NOPASSWD alone? 22:26 < SmashingX> revel: I’m not editing the sudoers file 22:26 < bls> there are reams of documentation and examples on this topic, including the sudoers man page, syntax checking by visudo, and diagnostics via sudo -l 22:26 < SmashingX> I’m editing the file inside /etc/sudoers.d/sampleuser 22:27 < revel> Sudoers file, file in /etc/sudoers.d . Potayto, potahto. 22:27 < revel> Same syntax. 22:27 < SmashingX> the problem is that I don’t have the password for that user 22:27 < SmashingX> but I have root access to that server 22:27 < SmashingX> I just wantr to test if the user would be able to change the directory 22:28 < uplime> well sudo cd; won't really do anything anyways 22:28 < SmashingX> I mean no change the directory but edit anything inside the directory 22:28 < Psi-Jack> SmashingX: NOEXEC: NOPASSWD: ... 22:28 < revel> Err. 22:28 < revel> SmashingX: If you have root access, then you could just `sudo -u sampleuser -i`, right? 22:29 < SmashingX> is that the same as su - sampleuser? 22:30 < revel> As far as I know, yes. "Login shell as the user sampleuser" for both. 22:31 < revel> Except sudo would ask for your password instead of for the password of sampleuser if you weren't root, I think? 22:31 < SmashingX> ok that’s the thing 22:31 < SmashingX> I don’t know the password for sampleuser 22:31 < revel> Yeah, but you don't need it if you're root. 22:31 < ozymandias> sudo uses your password 22:31 < Psi-Jack> Oh, sorry, NOEXEC:NOPASSWD: 22:31 < ozymandias> su uses the password of the target 22:31 < revel> Both su and sudo just give you access if you're root already. 22:31 < revel> ozymandias: Even if you're root? 22:32 < ozymandias> if yo