--- Log opened Thu Apr 26 00:00:58 2018 00:02 < s33r> anyone on a 2700x rn? want to know if it is worth for gentoo 00:06 < terra> s33r : want to compile packages for yourself? so what can beat a 16 core/thread processor? 32 core processor. 00:07 < Brainspackle> gentoo is like the digital version of self mutilation 00:08 < terra> Brainspackle : theri community is impressive. Just look at their ebuilds. Some of them insane! 00:11 < bdonnahue> hey guys, in netstat for foreign address, what does :::* mean? 00:12 < koala_man> bdonnahue: it's the ipv6 equivalent of 0.0.0.0:* 00:13 < bdonnahue> koala_man, thanks, thats what I thought. jsut needed a sanity check b/c something is broken on my box 00:36 < mawk> koala_man: :: 00:36 < mawk> ah sorry I misread 00:37 < mawk> it was suspicious that you asked that 00:37 < jim> is :: a form of ...? 00:38 < mawk> it means 0000:0000:0000:0000:0000:0000:0000:0000 00:38 < mawk> like 2001:db8::1 means 2001:0db8:0000:0000:0000:0000:0000:0001 00:39 < mawk> :: means "insert as many zero as necessary for the whole address to be 128 bits" 00:40 < jim> oh ok :) 00:40 < jim> I totally come from an ipv4 standpoint... 00:41 < mawk> it comes with an ambiguity when you want to beautify address with trailing zeroes 00:41 < mawk> or not even trailing zeroes 00:42 < mawk> for instance 2001:db8:0:0:4242:0:1:2 00:42 < mawk> you can write either 2001:db8:0:0:4242::1:2 or 2001:db8::4242:0:1:2 00:42 < mawk> and if the zero intervals are the same size you can't even decide based on the size criterion 00:43 < mawk> maybe using the biggest group from the set of (group size, group offset from the rightmost position) using lexicographical comparison is the most sane thing 00:44 < mawk> group offset from the leftmost position, actually 00:44 < mawk> written rfc style 01:06 < Irbis> WTF? https://i.imgur.com/9NMZ8Vw.png 01:07 < wodencafe> I don't see the problem 01:07 < rascul> Irbis an explanation could be helpful 01:07 < Brainspackle> wtf indeed... all those irc.txt files are most silly 01:08 < KuroiKiri> sneeeeech 01:11 < onitlikesonic> Hi all, can anyone lend me a hand on ubuntu partitioning? i got the following https://pastebin.com/zPRcUinj and i am looking to get (a fixed 10GB / partition) + (a fixed 1GB SWAP) +(all the rest of the space in an ext4 partition) howevr at the moment all is being put into the swap! crazy! 01:15 < Irbis> wodencafe: 16 on 9 there is no 01:15 < mutante> onitlikesonic: i think it is because your priority is 500 on the swap partition 01:15 < mutante> which is lower than 10000 and 10000. and lower prio = higher 01:16 < onitlikesonic> mutante: thanks, i will test it out with a higher number then 01:16 < ShapeShifter499> hi 01:16 < ShapeShifter499> are there any driver workaround for certain usb external drives? 01:16 < ShapeShifter499> *hard drives 01:17 < mutante> onitlikesonic: partman is just so.. annoying/hard. maybe this page has something helpful for you. https://wikitech.wikimedia.org/wiki/PartMan 01:17 < ShapeShifter499> I have a issue with WD My Passport 3tb drive, this is a BRAND NEW just got it a few hours ago through the mail after sending in the old one for warranty. I'm getting errors like "Buffer I/O error on dev sdc, logical block 1, async page" 01:17 < mutante> (several people writing about how to read the config after spending too much time on it :) 01:18 < ShapeShifter499> I'm suspecting there is hardware quirk or something 01:21 < onitlikesonic> mutante: indeed, for most of my machines i have a set template that i swore never to to touch again, but now i need this for another strange machine... i hate it.. sometimes linux is so epoch() dated :) 01:32 < giaco> how can I test a character device by reading something from it? I'm trying cat /dev/video0 but it returns cat: /dev/video1: Invalid argument 01:32 < pankaj> What is the use of service manager. Does it runs the services at start up or what else? 01:33 < jml2> pankaj, yep 01:34 < jml2> ShapeShifter499, if that uses hardware encryption hten you're out of luck, I know that certain brands wouldn't work with Linux... 01:35 < pankaj> jml2: So, it just starts the services at startup. 01:35 < jml2> ShapeShifter499, not sure on that particular brand, but sometimes these enclosures can be set to a "mode" under "Windows" so that they work properly on other platforms 01:35 < kazdax> whats a good channel to ask about how to improve traffic to my website :D 01:36 < jml2> pankaj, yeah 01:36 < kazdax> i know its an often asked question for people wanting to promote their product 01:36 < kazdax> #web ? 01:36 < jml2> kazdax, what site is that? 01:36 < Sveta> for some reason 'kazdax' does not tab complete 01:36 < pankaj> jml2: I can understand the services like running the ssh server or display enabling netctl but what are some other system services. Is their a list of these? 01:36 < jml2> kazdax, does it even work? lol 01:36 < jml2> pankaj, I don't know 01:36 < jml2> pankaj, you'll have to use your distro's tool for that.. and it can vary 01:36 < pankaj> OK 01:37 < pankaj> jml2: OK 01:37 < kazdax> no its not yet in the process..but i want to understand how you can make users visit your website..if i may say 01:37 < kazdax> hi sveta 01:37 < jml2> pankaj, google " service command" 01:37 < Sveta> but ask ##seo 01:37 < triceratux> pankaj: dnsmasq 01:37 < kazdax> if i may say ..It would be a website about helping people correcting software induced problems 01:37 < kazdax> thanks ill look into that 01:37 < Brainspackle> kazdax: so is it more like stackoverflow or more like silk road? 01:38 < jml2> pankaj, or if you're good with the terminal, sometimes using "apropos " might put out the correct command to look into (and then use "man ") 01:38 < Brainspackle> those are the only things that help me 01:38 < kazdax> lol 01:38 < kazdax> what did you order from silk road 01:38 * Brainspackle pleads the 5th 01:38 < jml2> triceratux, I had to setup dnsmasq recently here on a system... 01:38 < kazdax> well i just wanna make a few bucks and i can find some people who know how to operate the windows operating system 01:39 < kazdax> its just fixing OS problems for newbs 01:39 < jml2> triceratux, it takes time to set it up because there's so many dam option 01:39 < kazdax> i cant come up with a a better idea 01:39 < triceratux> pankaj: avahi-daemon, dbus-daemon, lightdm, ntpd, polkit, systemd-resolved, that kind of stuff 01:39 < Brainspackle> kazdax: porn 01:40 < onitlikesonic> mutante: seems that was not the problem :( 01:40 < pankaj> triceratux: Is their any command to list those services which automatically get started up by systemd at boot? 01:40 < jml2> pankaj, yeah 01:40 < Sveta> there 01:40 < jml2> pankaj, its called "systemctl" 01:41 < jml2> pankaj, you just hit enter and it will list the services 01:41 < jml2> triceratux, ever hear of the tool called "dex"? 01:41 < jml2> triceratux, it's a nice tool to have for all kinds of autostart things... 01:41 < pankaj> jml2: I am so sorry. I knew that command but I do not know why I keep forgetting again and again. 01:42 < jml2> pankaj, well that's why you have apropos 01:42 < jml2> pankaj, as I mentioned.. 01:42 < jml2> :p 01:42 < pankaj> jml2: O yes. 01:42 < rascul> another command to remember 01:42 < rascul> 'yes' is also a command 01:42 < phinxy> wtf 01:42 < triceratux> jml2: im fairly certain i got dnsmasq working on todays altlinux sisyphus nightly. both dnsmasq & dig are preinstalled on the iso, when you systemctl start dnsmasq resolvconf toggles /etc/resolv.conf from 192.168.1.1 to 127.0.0.1, & i only had to set a listen-address to get the dig query time to drom to 0msec 01:42 < jml2> pankaj, may I introduce you to my awesome notetaking application I use-> "zim" 01:43 < jml2> triceratux, I found out why it does that.. 01:43 < jml2> triceratux, by setting it up again recently.. 01:43 < TheCryptek> Could I get some help :/ 01:43 < jml2> triceratux, you can use the resolv-file= option to specify -> /etc/resolv.dnsmasq <<< will be a static file 01:43 < jml2> triceratux, and you let resolvconf or anything else change the /etc/resolv.conf 01:44 < Brainspackle> TheCryptek: Yes. I accept PayPal and Venmo 01:44 < TheCryptek> <.< 01:44 < triceratux> jml2: yep theres a gazillion ways to get those semipersistent configs & their symlinks to stop fighting with one another. & itll be different on every distro beccause of how many network programs there are & how far each one has gotten with systemd 01:45 < jml2> triceratux, actually I'm using dnsmasq as a "local dns server" -- because i have /etc/resolv.conf with 127.0.0.1 -- this is why I'm using resolv-file=/etc/resolv.dnsmasq 01:45 < TheCryptek> I'm just trying to establish a better way to back up my files on linux other then just copying and pasting to an sdcard or flashdrive @Brianspackle. I would just use my RPi3 as an NAS but it's power cord went out on me. 01:45 < jml2> triceratux, dnsmasq makes it easy to setup an www.xyz.com record with a "PTR" record as well... 01:46 < jml2> triceratux, so this is why I set it up (wordpress local -- something was looking up a PTR record I think.. so /etc/hosts is not good enough) 01:46 < Brainspackle> TheCryptek: i really like rdiff backup: https://www.tecmint.com/rdiff-backup-remote-incremental-backup-for-linux/ 01:46 < triceratux> jml2: thats all im doing with it so far as well & it came up clean on altlinux. looks like its working on swagarch. wouldnt install on lubuntu-next because soething thinks port 53 is in use. once im an expert ill shoot that one maybe 01:47 < jml2> triceratux, dnsmasq is an expert tool, it's very hard to understand unless you know what "dns forwarding" means 01:48 < jml2> triceratux, dnsmasq is a dns-server/forwarder --- it will take "nameserver xxx" and use that as a forwarder. I read it won't go to the "Root hint" servers.. 01:48 < TheCryptek> @BrianBlaze, Here is my problem, I only have 1 laptop and a [non-working] raspberry pi, so I can't exactly do /remote/ backups 01:48 < triceratux> jml2: yep im just an old 192.168.1.1 guy who was using wvdial until a couple years ago 01:48 < jml2> triceratux, so it needs your ISP's nameservers (or other nameservers) elsewhere to "forward dns requests to" 01:48 < TheCryptek> @Brainspackle, [Sorry tagged wrong person] 01:49 < jml2> triceratux, if you have "127.0.0.1" in the /etc/resolv.conf <<< then you should prevent dnsmasq from reading that by setting something up appropriately 01:50 < jml2> triceratux, if you dont need dnsmasq (sounds like you don't) , then you can actually just remove it 01:50 < triceratux> jml2: i know im way over my head. this is the doc i wrote before i forget which i will real fast http://pastebin.centos.org/707671/raw/ 01:50 < jml2> triceratux, it is actually an advanced tool imho, and not worth keeping if you're not doing thigns with it. Why distros are bundling this must have to do with dnssec sometime in the future i think 01:51 < jml2> triceratux, 3a and 3b are misleading and wrong :) 01:51 < triceratux> jml2: yep its streamiling the vpn & local hostname resolution stuff. theyre preparaing to break stuff thats worked forever 01:52 < triceratux> jml2: i know thats why theyre prefaced with a "whatever" 01:53 < jml2> triceratux, dns servers --> do "forwarding" on behalf for clients.. so if your local dnsmasq server doesn't have an answer for an A record, it "forwards" the request to your isp' nameservers.. 01:54 < jml2> triceratux, dnsmasq is only useful if you have a "custom" record for like giving an A record for say a lan server -- eg, dig www.mysite.com returns 192.168.1.5 01:54 < jml2> triceratux, otherwise it is not good for anything... 01:54 < jml2> triceratux, and of course some dnssec support there.. but i dont' know if it forwards things encrypted.. 01:55 < triceratux> jml2: im aware i dont need it. i was giving up when i realised id succeeded in getting it to work. its a pretty adroit tool actually, & a good example of something that has to work well with systemd 01:56 < triceratux> jml2: also the writeups on the web are all over the map. a number of them tell you to create a dnsmasq group. that doesnt seem necessary 01:57 < jml2> triceratux, your problem can be fixed if you use resolv-file=/etc/resolv.dnsmasq 01:57 < jml2> triceratux, you'd have to create static file with something like "nameserver 192.168.1.1" in that file.. 01:57 < jml2> triceratux, make sure netstat -plutn has 127.0.0.1:53 listening, because you have /etc/resolv.conf with nameserver 127.0.0.1 01:58 < jml2> triceratux, dnsmasq will use 192.168.1.1 as a "dns forwarder"... 01:58 < jml2> triceratux, otherwise it will do no forwarding than replying records for things that you define for the lan.. 01:59 < triceratux> jml2: tcp 0 0 127.0.0.1:53 0.0.0.0:* LISTEN 17630/dnsmasq 01:59 < jml2> triceratux, address=/www.xyz.com/127.0.0.1 ptr= , .... these two lines will only be replyable for clients... so this is why you need to have forwarders.. 02:00 < MrPockets> Is there a command-line tool that can be used to both view, and encrypt network documentation wile at rest? 02:00 < jml2> triceratux, and i suppose udp is there as well .. of course... :p 02:00 < jml2> MrPockets, ? you mean test tls/ssl connectivity? 02:01 < jml2> MrPockets, well there is always the classical wireshark for capture packet data.. 02:01 < MrPockets> Documentation, not traffic. 02:01 < MrPockets> I essentially want like, a nano that'll encrypt a file when I'm done opening it 02:01 < triceratux> jml2: thats probably this: udp 1536 0 127.0.0.1:53 0.0.0.0:* 17630/dnsmasq 02:01 < jml2> triceratux, note: you should use "dig" at the start because it will say what "server" is giving the response.. 02:02 < MrPockets> I've been using GPG, but it's cumbersome to have to use gpg to decrypt, open the file, view / update it, close it, delete the original GPG, re-encrypt it with GPG, etc. 02:02 < jml2> triceratux, dnsmasq will always be the server from "dig's" output because it is acting on behalf for you for relaying the dns requests to any forwarders.. 02:03 < meyou> @MrPockets, vigpg? 02:03 < jml2> triceratux, "If you don't want dnsmasq to read /etc/hosts" << use "no-hosts" 02:04 < jml2> triceratux, if you don't want dnsmasq to reply on behalf for /etc/hosts ... 02:04 * MrPockets googles 02:04 < jml2> triceratux, bind9 which is a dns server doesn't do the crazy things dnsmasq does... lol 02:04 < triceratux> jml2: i had a separate bug on altlinux over its empty /etc/hosts. id already put 127.0.0.1 in it to get midnightcommander to start without a multisecond lag. its a known mc bug 02:05 < jml2> triceratux, but if you're the client from the local system .. it also depends on NSS if you have "file" activated.. this is confusing and you can ignore me about "no-hosts" 02:05 < MrPockets> oh man, meyou, I think this is exactly what I was looking for. 02:05 < meyou> :) 02:05 < MrPockets> A comparable nano tool doesn't exist, does it? 02:05 * MrPockets ducks 02:05 < meyou> don't think so, i checked 02:07 < triceratux> jml2: currently taking a breather before the xubuntu bionic shakedown. tomorrows going to be fun 02:07 < jml2> triceratux, you should disable dhcp whereer possible, 02:07 < jml2> triceratux, because normally dhcp listens on raw sockets... 02:08 < jml2> triceratux, which bypasses the firewall... 02:08 < jml2> triceratux, (no-dhcp-interface=) 02:08 < jml2> triceratux, I know you have dns listening on lo, but make sure dhcp is disabled with dnsmasq 02:09 < triceratux> jml2: is running dnsmasq potentially an additional security risk if i dont have it configured in a bulletproof fashion ? 02:09 < jml2> triceratux, i think you need to specify dhcp options to make the dhcp feasible.. 02:10 < jml2> triceratux, but I still make sure dhcp is disabled by using no-dhcp-interface specified for each interface 02:11 < jml2> triceratux, no-poll << I added this to my dnsmasq because I know /etc/hosts is never going to be changed when all it has is nameserver 127.0.0.1 02:11 < jml2> triceratux, if a local DHCP or VPN client is automatically updating /etc/resolv.conf, then yuo don't want to use no-poll 02:12 < jml2> triceratux, typo ... 02:12 < triceratux> jml2: like i said, resolvconf is toggling the /etc/resolv.conf nameserver in response to systemd managing dnsmasq. astoundingly well configured 02:13 < jml2> triceratux, no-poll << I added this to my dnsmasq because I know """" /etc/resolv.conf """ is never going to be changed when all it has is nameserver 127.0.0.1 02:13 < jml2> triceratux, I don't use dhcp here on my system.. so resolv.conf never changes... 02:16 < jml2> triceratux, you can setup your dhcp client to prioritize nameserver 127.0.0.1, -- or better to ignore the "nameserver" dhcp options .. this way resolv.conf stays the same.. 02:17 < jml2> triceratux, if you ignore the dhcp -dns-namserver options, then you need to specify forwarders for dnsmasq 02:19 < MrPockets> meyou, Are you familiar with "pass" ? 02:19 < rypervenche> MrPockets: vim has a gnupg plugin that opens GPG-encrypted files and lets you edit them and then re-encrypt them as you save them. 02:19 < jml2> triceratux, you should always this in your firewall rule (not documented properly) --> -i lo -j ACCEPT ... 02:19 < rypervenche> MrPockets: That's what I use for my "lastpass" sort of manual setup. 02:19 < meyou> i am not 02:19 < MrPockets> rypervenche, thanks! 02:20 < jml2> triceratux, contrary what people think, the "lo" interface can have traffic showing source and destination addresses of another interface.. 02:20 < jml2> triceratux, (tcpdump -i lo -n) 02:20 < rypervenche> MrPockets: I made a script that then decrypts it from a popup window, throws the user of the site into my clipboard for a second, then the password for 4 seconds, then deletes the clipboard. 02:20 < jml2> triceratux, iptables -I INPUT -i lo -s 127/8 -d 127/8 -j ACCEPT << is not good 02:20 < jml2> triceratux, iptables -I INPUT -i lo << is good 02:21 < rypervenche> -j ACCEPT 02:21 < jml2> triceratux, (this is because you can be doing traffic starting and ending to the local host) 02:21 < jml2> triceratux, (and certain background services do this) 02:22 < jml2> rypervenche, correct 02:22 < Roey> hi 02:22 < Roey> whoah nellie, lotsa folks here. 02:22 < jml2> triceratux, iptables -I INPUT -i lo -j ACCEPT <<< 02:23 < Roey> SadieTheLlama: o/ 02:23 < jml2> rypervenche, you know what happens if I omit -j ACCEPT right? 02:23 < jml2> rypervenche, lol 02:24 < jml2> rypervenche, if you do packet marking it is feasible to omit -j .. 02:24 < rypervenche> jml2: I'm guessing from your reply that it defaults to -j ACCEPT. I would always prefer to be explicit, lest the defaults change. 02:25 < jml2> rypervenche, actually the target would be blank, there would be a match, and it can do packet marking if there's any on that match rule, and then it'll continue reading the rest of the rules.. 02:27 < jml2> rypervenche, port knocking rules omit the -j ... 02:28 < rypervenche> jml2: Not the port knocking rules that I've seen. Do you have any good resources for what you're talking about? From a quick search, I'm only finding -j MARK as what you might be referring to. 02:28 < jml2> rypervenche, I use these rules without -j.. 02:29 < Hextor> If I wanted to sandbox various wordpress installs for different domains from each other, what's the easiest way, openvz, docker, some sort of chroot? I don't think apache allows you to run instances under user accouns and just has one account for all sites 02:29 < jml2> rypervenche, -A PN2 -m recent --remove --name PN1 --mask 255.255.255.255 --rsource (example) 02:29 < KitaKat> Dad 02:29 < xamithan> docker probably as it likely has already done images ready to pull 02:30 < Hextor> Can docker work inside of a VMware VM 02:30 < rypervenche> jml2: Ahh, using the recent module. That makes sense. 02:31 < xamithan> If you enable nested virtualization inside vmware, sure 02:31 < jml2> rypervenche, -A PN -m recent ! --rcheck --name PN1 --mask 255.255.255.255 --rsource -j CP1 02:31 < jml2> rypervenche, the "!" negate makes the portknocking very helpful 02:32 < Hextor> So docker is a heavy duty virtualization. Perhaps openVZ is more resource efficient 02:32 < jml2> rypervenche, makes the outline very clear and easy to follow... 02:32 < xamithan> docker on linux is more like chroot 02:32 < xamithan> on windows it uses hyper-v 02:32 < Hextor> I run centos, what's better for that? 02:33 < Hextor> I assume it helps a lot to have different mysql logins. But the php sploits worry me 02:34 < xamithan> It all depends on what apps you use and how you need it to interact with other services. openVZ would be more like a VPS so i'd assume it is more resource heavy 02:34 < jml2> rypervenche, I have two lines without -j , when checking knock #3 -> -A CP3 -m recent --remove --name PN1 --mask 255.255.255.255 --rsource , -A CP3 -m recent --remove --name PN2 --mask 255.255.255.255 --rsource , and then the next line is an all -j DROP 02:34 < rypervenche> jml2: That might be a useful thing for me to set up :) I'll see what I can do since nftables doesn't have the recent module. 02:35 < jml2> rypervenche, it's actually very easy to visualize.. because with the "!" negate makes it very easy.. 02:35 < jml2> rypervenche, I can post it on pastebin 02:36 < Hextor> I was thinking of a wordpress install for each domain. I heard WP and especially, most especially, WP plugins are awful. 02:39 < jml2> rypervenche, v 02:39 < jml2> rypervenche, https://pastebin.ca/4018717 02:41 < jml2> you need to create the custom chains with -N iirc 02:41 < jml2> rypervenche, so iptables -N PN, and for CP1 to CP3 .. 02:42 < rypervenche> jml2: Yep, I understand. I use the recent module on other machines. This one has nftables, so I'll likely have to set up something manually. 02:42 < jml2> rypervenche, it is also possible to have a portknocking in iptables for a "front router" but that is much difficult with the nat table -- or whatever other table there is because I think -j DROP is not used but something else.. 02:43 < pankaj> What exactly is a build system. I am having confusion with other compile tools which can also to some degree automate the process of compilation. 02:44 < uplime> the collection of tools used to build your program 02:44 < jml2> rypervenche, iptables works with nftables no? 02:44 < jml2> rypervenche, since 3.17 iirc 02:45 < pankaj> uplime: Like? 02:45 < uplime> pankaj: depends on the project, the os, the language, and where you're deploying it to 02:46 < rypervenche> jml2: I use straight up nftables. Oh, I can just create a set with a timeout. That would be very neat. 02:46 < jml2> rypervenche, there's a package called iptables-nftables-compat here on my debian.. 02:46 < jml2> rypervenche, using those remove/resource lines is more effective.. 02:47 < jml2> rypervenche, fwiw I use hping to do the udp knocking 02:48 < jml2> rypervenche, (package hping3) 02:48 < rypervenche> jml2: the recent module doesn't exist in nftables, and does not exist in the iptables nftables translation that that package provides. 02:49 < rypervenche> Yeah, hping3 or nmap will do. 02:49 < jml2> rypervenche, yeah I use something like a timeout for another setup .. because somebody wants to do a ssh login from the outside.. so I have a cron job doing a "ping" once every couple of minutes to keep the gates open between client and server 02:50 < rypervenche> jml2: I can simply have any IP that hits that port be added to a set with a small timeout and then allow that set access to the next port, etc. 02:50 < jml2> rypervenche, and then use "noip"'s daemon (from noip.com) and that's pretty sweet to use with remote ssh things... never worry about a change of ip... 02:50 < pankaj> uplime: I do not understand that. 02:50 < rypervenche> Yep, I use that. ddclient though 02:51 < jml2> rypervenche, it's stealth 02:51 < jml2> rypervenche, there's also a portknock app for android :) 02:51 < jml2> rypervenche, it can do any udp/tcp pn, then immediately start an application.. 02:51 < rypervenche> Nah, I like to do things manually. 02:52 < jml2> rypervenche, I got a camera system app that I set a pn on the gateway... so I use that.. 02:52 < jml2> rypervenche, camera stream is still encrypted, but I don't like those scanners that would detect a login service 02:52 < rypervenche> jml2: I hear ya. 02:54 < jml2> rypervenche, is iptables going to be dropped ? 02:54 < jml2> rypervenche, what's that package iptables-nftables-compat ? 02:55 < rypervenche> jml2: I believe on newer kernels nftables is already being used on the backend, but I haven't looked into that claim. 02:55 < pankaj> What is the difference between coreutils and util-linux packages? 02:55 < rypervenche> jml2: It just eases the migration: https://wiki.nftables.org/wiki-nftables/index.php/Moving_from_iptables_to_nftables 02:58 < jml2> pankaj, coreutils is commands like cp, mv, and ls... 02:58 < nullius> cat 02:58 < jml2> pankaj, util-linux is things like mount I think .. more "system" adminstrative .. 02:59 < jml2> pankaj, actually there is no mount, but there is the test tool mountpoint... you can see what commands are in the package with dpkg -L 02:59 < jml2> pankaj, and grep bin ... 02:59 < jml2> pankaj, all scripts need access to coreutils ... so you'll find it on any linux distro.. 03:00 < wadadli> Should I resize the fs on a lvm before reducing it or does lvreduce handle this for me? 03:00 < uplime> not all scripts need access to coreutils 03:00 < jml2> pankaj, see "man intro" for an introductory :p 03:01 < pankaj> jml2: OK. Seeing man intro 03:02 < Skunky> wadadli: filesystem first, then underlying volume. 03:02 < wadadli> are the same procedures for reducing a standard lvm the same for a thinly provisioned logical volume? 03:02 < jml2> rypervenche, https://www.spinics.net/lists/netfilter/msg57506.html -- oh yes I remember I think the recent module that is being used lasts for like 7 seconds by default.. 03:03 < pankaj> jml2: I saw 'intro' but their is no single word about this topic either 03:03 < rypervenche> jml2: I was going to link you to that page actually, but didn't. 03:03 < jml2> rypervenche, it's usually there on all linux distros.. 03:03 < jml2> sorry 03:03 < jml2> pankaj, I meant.. 03:04 < jml2> pankaj, how long ago did you install that linux setup? 03:04 < jml2> pankaj, sometimes the manpage cache is not yet updated until cron runs it 03:04 < jml2> nways that manpage shows the basics like ls, chmod, for very new users.. 03:05 < jml2> it helps for the start.. but then it is very limitted... 03:05 < pankaj> jml2: I am just following 'Linux from scratch'. So, I saw these two packages that I have to use but did not get what is the difference. If their would not be any difference then the community would have not advised me. 03:05 < pankaj> jml2: I know about core utils. But what about util-linx. 03:05 < jml2> pankaj, those packages are very standard... you'll find that util-linux is maintained by the kernel maintainer site.. 03:05 < uplime> you shouldn't be doing LFS if you don't already have a good understanding of these tools and concepts 03:06 < jml2> pankaj, you'll see it upstream from somewhere kernel.org 03:06 < jml2> pankaj, https://en.wikipedia.org/wiki/Util-linux 03:06 < pankaj> jml2: OK. So, what does this package do. How does it help the kernel or in what good way it is related to kernel? 03:06 < jml2> pankaj, there are basic tools for the Linux system... 03:07 < jml2> pankaj, it helps to create the user space... 03:07 < triceratux> https://mirrors.edge.kernel.org/pub/linux/utils/util-linux/ 03:08 < pankaj> jml2: Like? 03:08 < jml2> pankaj, like being able to do anything :) 03:08 < jml2> pankaj, posix comes to mind 03:09 < jml2> pankaj, a basic shell with posix-like attributes 03:09 * jml2 says mirror mirron on the wall who has the best distro of all! 03:09 < pankaj> jml2: I am unable to stack all these different looking ideas in one place. 03:10 < jml2> pankaj, I remember when I started linux, I had to upgrade my brain 03:10 < pankaj> jml2: What is its main purpose. It could have also been packaged with linux kernel package also. 03:10 * jml2 gives pankaj an upgraded brain 03:10 < pankaj> jml2: Their is something different and important about it. 03:11 < triceratux> jml2: its overwhelming the first time you encounter it. but it doesnt go away. youll get used to it the more you work with it 03:11 < jml2> pankaj, no because the kernel is not user space 03:11 < triceratux> jml2: oops pankaj ^^ 03:11 < jml2> pankaj, there's kernel space and user space.. 03:11 < pankaj> jml2: OK. 03:11 < jml2> pankaj, if you have access to kernel space all the time as a regular user, a basic "application crash" or user-error can mean lost of all your data. 03:12 < pankaj> jml2: That makes a sense. 03:12 < jml2> pankaj, this is why the two are separate. When the kernel "crashes" it is actually preventing a loss of user data.. 03:12 < pankaj> jml2: I think linux-utils package is used to handle linux kernel (for building the system) when I am not under the sytem that I am building. 03:12 < triceratux> the kernel interacts directly with the hardware. the userspace interacts directly with the user 03:12 < ananke> how does a kernel crash prevent loss of user data? 03:13 < jml2> ananke, by not letting unknown instructions take over into kernel space? 03:13 < pankaj> jml2: So, what is your conclusion about 'util-linux'? 03:13 < jml2> there's something called security rings.. 03:13 < jml2> pankaj, no conclusion 03:13 < jim> - 03:14 < ananke> jml2: uhmm, when a kernel _crashes_, loss of data is a given. not sure why you consider crashing a way of 'preventing' loss of data 03:14 < jml2> ananke, it's a nice aesthetic 03:14 < ananke> jml2: unless you define 'loss' as 'compromise' 03:14 < pankaj> jml2: Do anyone have idea on why util-linux and linux kernel are separate packages? 03:15 < ananke> jml2: what does that even mean? 03:15 < jml2> ananke, ^ 03:15 * jml2 :) 03:15 < jml2> ananke, you want to define posix to him? XD 03:15 < ananke> pankaj: because one is a kernel, and another one is a set of utilities that run on a given kernel 03:15 < pankaj> ananke: OK. Just what about util-linux. 03:15 < ananke> jml2: it's not necessary 03:15 < jml2> ananke, why Linux and not Unix? XD 03:15 < pankaj> ananke: So, what utilities. Are their some examples for proper grasping of this stuff? 03:16 < ananke> pankaj: yes. look at the first diagram here: https://en.wikipedia.org/wiki/Operating_system 03:16 < jml2> pankaj, there's great books on Linux... I would suggest to read up about it and try Linux in something called virtualbox.. (virtualbox.org) -- all you need is an iso and virtualbox.. (no need for burning a cd or even using a real usb) 03:16 < triceratux> http://pastebin.centos.org/707676/raw/ 03:17 < jml2> triceratux, dnsmasq is "good" imho (as I said) if you want to impersnate your own A and PTR records for a lan.. 03:17 < pankaj> jml2: Is it a comment or advice or what? 03:17 < jml2> pankaj, I've been using Linux for 17 years.. 03:18 < pankaj> jml2: Because I think many programmers usually know about virtualbox. 03:18 < jml2> pankaj, i'm stating facts, util-linux is a base core package on all linux distros.. 03:18 < jml2> pankaj, I gave you like um 4 links and you should at least read up about them :) 03:18 < jml2> pankaj, (as well as coreutils).. 03:18 < pankaj> jml2: I know sir but I just want to know that why they are separate packages. 03:18 < ananke> jml2: not on all. busybox takes its place on minimal distros 03:18 < jml2> pankaj, you're a developer ? 03:19 < jml2> pankaj, how did you go from user to developer all of a sudden? 03:19 < jml2> pankaj, why the kernel is separate from user space? 03:19 < jml2> pankaj, because the kernel is not user space.. 03:19 < jml2> pankaj, libc is outside the kernel... 03:19 < triceratux> jml2: i appreciate the guidance & im continuing to google slowly. i trust dhcp & simon kelley but nothing will ever get me to trust lennart ;) 03:20 < jml2> (libc helps to bind user-space basic commands with the kernel) 03:20 < pankaj> jml2: The more and more I am listening to you the more I am making sure that linux-utils would be used to handle linux-kernel package when building system (As I am not inside it). 03:20 < jml2> pankaj, read up on ldd ldconfig and binary-library binding.. 03:21 < jml2> pankaj, you dont have to listen to me. 03:21 < pankaj> jml2: OK. I am going to read the links you gave briefly. 03:21 < keviv> Hi, I have set the kernel boot parameters `nmi_watchdog=panic panic=15`. This seems to work for soft lockups (debug info is printed, then panic() is called and 15 seconds later the system reboots), but a hard lockup will print out debug info but never panic. Is there another parameter I need to set? 03:21 < jml2> pankaj, you should go out and read about these things :) 03:21 < jml2> pankaj, because you won't understand what Z is if you don't know what "A" is :) 03:21 < pankaj> jml2: OK Sir. Here I go. 03:21 < pankaj> jml2: THanks 03:22 < ananke> pankaj: what are you trying to accomplish or learn? sounds like the concepts you're confused about are more of general operating system ideas, rather than anything linux specific 03:22 < jml2> pankaj, you're thinking about kernel systems back from the 1940's... these systems were not multi-user :) 03:23 < jml2> pankaj, back in those days the kernel space was the user-space... the user was essentially the 24/7 developer XD 03:23 < autopsy> jml2, I didn't know there were kernels in 1940 03:23 < wadadli> How can I resize a ext4 fs to it's take up the maximum amount of space available. 03:23 < jml2> autopsy, not how we now know them :) 03:23 < autopsy> Ah. 03:24 < pankaj> jml2: If I am confused about it then please you make sense and give the anwer. 03:24 < jml2> pankaj, do you know where the term "bug" came from? 03:24 < jml2> pankaj, it makes a great bedtime story actually.. but I won't be telling the goodies about how the term "bug" came to be tehehehe 03:24 < ananke> jml2: not sure if you're helping him, or just confusing him more 03:24 < s33r> wadadli: man parted should have this in there 03:25 < wadadli> How can I determine the true size of an ext4 partition? 03:25 < jml2> wadadli, ? the partition table? 03:26 < jml2> wadadli, where else? 03:26 < pankaj> jml2: I saw a documentary on computers and it stated that in olden days computer were mostly mechanical and when something wrong happens to the system they would check and in some cases see an insect being hung and so problem caused. So, the name bug 03:26 < jml2> wadadli, if you mean "ext4 filesystem" then it will be smaller than the partition it is residing on.. 03:26 < wadadli> ah there's a tool dumpe2fs jml2 03:26 < pankaj> ananke: You are confusing me. 03:27 < pankaj> jml2: I understood what jml2 is trying to say. 03:27 < jml2> pankaj, everybody is confusing you.. because a kernel depends on hardware architecture design 03:27 < ananke> pankaj: interesting, considering I barely said anything to you 03:27 < jml2> pankaj, sounds like you're trying to create an abstract construct of how computers are generally designed.. 03:28 < jml2> pankaj, it comes down to understanding how "cpu registers" exists on the main cpu processor that determines how the kernel should be implemented.. 03:28 < jml2> pankaj, "cpu registers" are special circuits inside the cpu... and the kernel needs to understand those.. 03:28 < pankaj> jml2: I respect you both guys but I just wanted to know about the difference that why both are different and package separately. 03:28 < wadadli> jml2: ye I resized it previously to 150G (resize2fs /dev/fedora/home 150G) 03:28 < wadadli> then resized the lvm to 150G 03:28 < jml2> pankaj, coreutils and util-linux both should never have to worry how "cpu registers" operate.. this is the job of the kernel.. 03:29 < wadadli> but for some reason gnome disk says that it's 161G 03:29 < jml2> pankaj, only the kernel has access to cpu registers.. 03:29 < wadadli> while lsblk says 150G 03:29 < jml2> pankaj, otherwise you have something called "race conditions" 03:29 < jml2> pankaj, where all programs fight over how can use the hard-drive.. 03:29 < wadadli> but lsblk will pull the size of the lvm not the fs 03:29 < wadadli> so need a way to double check the size of the fs 03:29 < wadadli> to compare with gnome-disk 03:30 < ananke> pankaj: because they're authored by different groups of people 03:30 < jml2> pankaj, instead programs are put into a "queue" so they each get their turn to write to the hard-drive... 03:30 < jml2> pankaj, so this is why the kernel has something called a "scheduler"... 03:31 < jml2> pankaj, it's like everybody trying to access the doorway... people need to be lined-up so each in turn can access the door :) 03:31 < ananke> pankaj: util-linux is specific to linux, while coreutils is not. 03:33 < jml2> pankaj, kernel supports "system calls" , and programs must use "system calls" in order to use the hardware... 03:33 < jml2> pankaj, so say you use "strace" on a program, it will show what "system calls" were used to talk to the kernel... and then the kernel takes care of talking to the hardware.. 03:33 < ananke> jml2: not sure your monologue serves much of a purpose 03:34 < jml2> ananke, dunno he did say he understood me but not you. lol 03:34 < pankaj> jml2: Ya, I knew whatever you just said. 03:34 < jml2> ananke, see :) 03:34 * jml2 laughs 03:34 < ananke> jml2: meh 03:35 < ananke> none of that is relevant to the actual question at hand: why are those two packages different 03:35 < jml2> no he's not asking that. 03:35 < ananke> jml2: he just did 03:35 < pankaj> jml2: So, you and anake have to figure out; I think. 03:35 < nmschulte> Is there a Unix/Linux command/utility (of the GNU Coreutils, GNU Binutils, util-linux likes) to watch a file system for changes in order to mirror them (or perhaps more generally "take action")? 03:36 < ananke> jml2: 21:28 pankaj> jml2: I respect you both guys but I just wanted to know about the difference that why both are different and package separately. 03:36 < ananke> nmschulte: inotify 03:36 < jml2> ananke, he's asking about the kernel being packaged with one of those two 03:36 < ananke> kernel is not packaged with either of the two 03:36 < pankaj> jml2: I think first I see the procedure of lfs and then may be I would understand that. 03:36 < jml2> ananke, he's not seeing why they can't be packaged together as one 03:37 < triceratux> inb4 inotifywait https://linux.die.net/man/1/inotifywait 03:37 < jml2> pankaj, have you ever seen a kernel tarball from kernel.org ? 03:37 < Skunky> coreutils are things that you'll find on almost any *nix system. util-linux is things that are specific to linux 03:37 < pankaj> jml2: I have downloaded sometimes. 03:37 < jml2> pankaj, that's where distributions get their kernel from. and it is always the kernel separate from util-linux 03:37 < jml2> pankaj, well you should ask them. 03:38 < jml2> pankaj, libc is not part of the kernel.. 03:38 < nmschulte> ananke: inotify tooling isn't commonly installed it seems. :( -- this is the right idea though, so I guess thanks. 03:38 < pankaj> jml2: Thanks for redirecting. 03:38 < ananke> not sure why there's any talk about the kernel when the question is: 20:55 pankaj> What is the difference between coreutils and util-linux packages? 03:38 < pankaj> jml2: OK. Thanks. Now, I have to return to my lfs process. It's is getting late. Thanks again 03:39 < jml2> pankaj, https://en.wikipedia.org/wiki/C_standard_library -- libc << think "system calls" 03:39 < DetectiveTaco> Why do y’all hate nick changes so much lol 03:39 < jml2> pankaj, the coreutils uses "libc" .... so now do you get the sketch? 03:39 < nmschulte> holy shit... you guys were seriously discussing coreutils and util-linux before I asked my Q? I didn't even notice. 03:39 < triceratux> DetectiveTaco: theyre cloned to everyone attached to irc channels so they spawn highly unnecessary traffic 03:40 < pankaj> jml2: Yes, I really got. Stronger now 03:41 < mecotri> Is there a command line utility, kernel module, or anything else that can be used to keep a computer awake without resorting to simulating mouse or keyboard input? I don't want to toggle screen blanking either. Ideally I need a solution that just would allow me to programatically send a heartbeat that resets the display timeout period. 03:41 < nmschulte> mecotri: define "awake" 03:41 < nmschulte> xset -dpms 03:41 < Skunky> mecotri: why wouldn't you just turn off the stupid screen blanker? 03:41 < pankaj> jml2: Just one question. I have tried reading some books but they were not to the point and also like boring. I want to learn assembly language. What are some resources to follow. I have basic understanding but could not go far because many of the books are like that. 03:42 < mecotri> nmschulte: running with screens on. I'm trying to prevent screens from going to sleep without disabling sleep. 03:42 < pankaj> jml2: What is your advice on it? 03:42 < nmschulte> xset s noblank, xset s off, of course. 03:42 < nmschulte> mecotri: are you running X11? 03:42 < jml2> pankaj, I don' tknow.. but there's a free assembly book online.. 03:43 < jml2> its a famouse one, I forget the name, but it is popular.. 03:43 < mecotri> nmschulte: I'm running Ubuntu 17.10 whatever the default is for that. 03:43 < srukle> where may i find that book jml2? 03:43 < nmschulte> pankaj: art of assembly is where you should start, I guess? dunno if it's been updated for modern architectures / instruction sets though 03:43 < srukle> we don't have assembly at my uni :( 03:43 < jml2> nmschulte, thanks its that one 03:43 < pankaj> jml2: Ya. Some are available as pdf but they are not to the point as user wants. 03:43 * jml2 http://www.plantation-productions.com/Webster/ 03:44 < jml2> pankaj, there's always MIT opencourseware... 03:44 < srukle> such a strange url, thanks jml2 03:44 < jml2> pankaj, there's also "udemy" though dunno how well some assembly courses are on those 03:44 < pankaj> jml2: Is it free? 03:44 < nmschulte> pankaj: I haven't read / referenced it in a while, but it worked well for me; very pointed. 03:44 < jml2> pankaj, should be free... 03:45 < pankaj> jml2: What? Opencourseware? 03:45 < nmschulte> if you want "to the point" I guess you should look to intel.com for reference manuals of specific architectures/chips? 03:45 < mecotri> Skunky: Because there are plenty of times I need to leave my computer on but want the monitors to sleep. Problem is there are also plenty of times I'm working on something in front of my computer using a webpage or PDF as a refernce and it's annoying to have the monitors sleep then. I'm making a circuit that will let my computer know when I'm sitting at it but not moving the mouse or using the keyboard so the monitors don't blank. 03:45 < pankaj> jml2: OK. I got that. Thanks 03:46 < nmschulte> I think there's ##hardware, too. 03:47 < Skunky> mecotri: woulnd't it just be simpler to turn off the monitors when you aren't using them? :) 03:47 < nmschulte> mecotri: xset is the command you want; shell script that: runs xset -dpms && xset s off; waits for any user input you desire; runs xset +dpms && xset s on; 03:48 < nmschulte> mecotri: I have to ask... have you tried simply increasing the sleep timeout? 03:48 < mecotri> Skunky: No because you don't always know when walking away for a few minutes will turn into an hour and I don't know about your hardware but my monitors expect to be slept and not turned on and off -- the buttons are very hard to access. 03:49 < Skunky> mecotri: you can do what nmschulte suggests, or hook into xscreensaver somehow, it's already got logic for watching for "input" and not blanking the screen 03:51 < mecotri> nmschulte: Ok thanks. I have but to be honest this is one of those because I can moments. I don't need to make something to solve this annoyance but I can and want to. 03:52 < justsomeguy> I guess you could just use xscreensaver as intended, right? It should keep the monitor from going to sleep... 03:52 < nmschulte> mecotri: good luck! 03:52 < mecotri> nmschulte: Thanks. 03:55 < jml2> triceratux, ever heard of "dex" ? 03:56 < wadadli> df -h displays my FS as being 147G but gnome-disk shows 161G 03:56 * triceratux googles furiously 03:56 < jml2> triceratux, since you work with 'em light desktop environments, i'm seeing what else i could be missing for tools like it.. 03:57 < mutante> dex now means decentralized exchange 03:57 < jml2> triceratux, it helps create .desktop and can autostar ~/.config/autostart ... it's great for non-xdg environments.. 03:57 < nmschulte> wadadli: https://www.google.com/search?q=161+gigabytes+to+gibibytes 03:58 < wadadli> nmschulte: did you actually look at that number? 03:58 < jml2> triceratux, https://github.com/jceb/dex (packaged on debian) , I came across when I was troubleshooting .desktop issues.. 03:58 < nmschulte> I'm guessing, of course. the numbers don't match as well as you stated. 03:59 < wadadli> 161Gigabyte is 149.9429Gibibyte 04:00 < triceratux> jml2: definitely a work in progress. lotsa reddit, broken links, cant decide if its desktop or embedded, worth keeping an eye on https://liliputing.com/2018/04/samsung-dex-pad-coming-may-13-for-100-use-galaxy-s9-like-a-desktop-keyboard-or-touchpad.html 04:00 < wadadli> oh well I'll stop fixating over this, it is what it is, not even math makes sense anymore. 04:00 < nmschulte> I get similar error when I compare my file system, wadadli. 04:00 < wadadli> nmschulte: eh well that's reassuring 04:00 < nmschulte> I'd trust df, :) 04:00 < wadadli> mutante: did you also check gnome-disk? 04:00 < nmschulte> wadadli: yes. 04:01 < wadadli> ha woah 04:01 < nmschulte> wadadli: I assume one of the two is not reporting file system overhead (e.g. journal) or something and only talking about "files on disk" 04:01 < jml2> triceratux, did you hear about the librem phone? they're going to be supporting the utouch from a focused team that wants the original ubuntu touch.. 04:01 * jml2 https://www.zdnet.com/article/ubuntu-touch-lives-on-in-purisms-librem-5-smartphone/ 04:02 < jml2> ./ubports./ to be more technical 04:02 < nmschulte> wadadli: I assume df is not reporting that overhead, only reporting "available blocks for inodes/files" 04:02 < triceratux> jml2: yep im keeping an eye on that too. just limping along with android netbooks & handsets in the meantime https://liliputing.com/2018/04/librem-5-linux-smartphone-will-support-ubuntu-touch-pureos-or-pureos-with-kde-plasma-mobile.html 04:03 < nmschulte> wadadli: or just completely distrust gnome-disks... its ays there's 98 GB free, but df shows only 77 GB (72 GiB) 04:07 < nmschulte> wadadli: look at the size vs usage (% full) for this volume -- http://oi63.tinypic.com/23msz1u.jpg 04:08 < justsomeguy> lol, that pretty much sums up my experience with gnome disks. 04:08 < nmschulte> I wanna file a GitHub issue with that image as the only content. 04:09 < wadadli> woah 04:09 < xamithan> that belongs on softwaregore sub reddit 04:09 < nmschulte> idk why I even have gnome-disks installed, I don't use Gnome. silly Debian packaging! 04:09 < wadadli> which version of gnome-disk is this nmschulte ? 04:09 < nmschulte> gnome-disks --version -- Unknown option --version 04:10 < nmschulte> gnome-disk-utility 3.28.1 -- UDisks 2.7.6 (built against 2.7.6) 04:11 < wadadli> How did you get that version output? 04:13 < justsomeguy> Woo, netsplit! 04:13 < mutante> ah looks like matrix or cloud something went down again 04:14 < nmschulte> wadadli: the button on the top left of the window, hides the real menu. the one on the right is like a dedicated context menu or something, becuase menubars are too clunky or something? 04:14 < nmschulte> or something. 04:15 < wadadli> lol smh 04:16 * wadadli gets back to real life 04:16 < al|iss> netsplit 04:16 < blaztek> Gnome Disk Utility uses 1 kilobyte = 1000 bytes because that’s what disk manufacturers say 04:17 < al|iss> ayyyyyyyyyyy 04:17 < nmschulte> blaztek: take a closer look at the screenshot pasted above 04:17 < nmschulte> the base 10 vs base 2 approximations doesn't nearly account for the discrepancy. 04:20 < blaztek> But what does fdisk -l say? 04:21 < wadadli> blaztek: Disk /dev/mapper/fedora-home: 150 GiB, 161061273600 bytes, 314572800 sectors 04:23 < nmschulte> blaztek: fdisk reports partition info, not file system info 04:24 < blaztek> But things like efi and swap aren’t part of the file system 04:25 < nmschulte> gnome-disks reports more than just file system information 04:27 < nmschulte> xamithan: https://www.reddit.com/r/softwaregore/comments/8ez9cc/gnomedisks/ 04:33 < s33r> configured mpd + ncmpcpp for the first time... all for the terminal visualizer haha 04:36 < stevendale> Hey 04:38 < hdogan> how can I write regex for "repeats in every 3 characters, e.g aa1bb1cc1dd1" 04:39 < Sveta> depends on what language 04:39 < markasoftware> not really... 04:39 < hdogan> Sveta, let's say for bash 04:39 < markasoftware> /(1[^1]{2})*/ or something 04:39 < markasoftware> for the character one 04:40 < markasoftware> although i think if you explain your problem in more detail we will discover that my solution is not sufficient 04:40 < hdogan> markasoftware, why do we need "/" 04:41 < markasoftware> that's just the end of the regex, i'm used to it 04:41 < markasoftware> don't need it i guess in grep 04:45 < hdogan> markasoftware, I have these 2 lines in my text file "aa1bb1cc1dd1 \n a1abb1cc1dd1". I want to match the regex to the first one. grep "(1[^1]{2})*" text.txt didn't work 04:46 < markasoftware> add -P option to grep 04:46 < markasoftware> you want only the first one? 04:48 < hdogan> markasoftware, yes 04:49 < markasoftware> so, you want every third character, starting at the third character? try ^([^1]{2}1)*$ 04:50 < markasoftware> i can confirm this one works for your example input 04:51 < markasoftware> but only if you give the -P option to grep 04:51 < stevendale> Does Linux get fragmented? 04:51 < markasoftware> stevendale: as in hdd fragmentation? yes 04:51 < fendur> markasoftware: wouldn't that match an empty string? 04:52 < markasoftware> fendur: yeah, i guess you could change * to + 04:53 < hdogan> markasoftware, yeap, it works with -P. I guess I can't use this regex in let say Python? 04:53 < markasoftware> you probably can 04:53 < stevendale> Hi, I'm wondering if my system is able to run Team Fortress 2 on Minimum quality, native resolution (1600x900) at an acceptable framerate, RAM is DDR3-1333 MHz, GPU has 256 MB DDR2 VRAM and is getting PCI express x16 bandwidth, CPU has 2 cores 4 threads, HDD is 3.5" 7200 RPM SATA-II 04:53 < stevendale> Client: HexChat 2.12.4 • OS: Ubuntu "artful" 17.10 • CPU: Intel(R) Core(TM) i5 CPU 650 @ 3.20GHz (1.23GHz) • Memory: Physical: 5.7 GiB Total (5.1 GiB Free) Swap: 2.0 GiB Total (2.0 GiB Free) • Storage: 16.6 GB / 163.1 GB (146.5 GB Free) • VGA: NVIDIA Corporation G84 [GeForce 8400 GS] @ Intel Corporation Core Processor QuickPath Architecture System Address Decoder • Uptime: 2h 8m 27s 04:54 < strive> lol 04:54 * al|iss smiles 04:54 < ayecee> try it and see. 04:56 < Sveta> maybe it would 04:57 < deizen> Hey, I am new here 04:57 < stevendale> Hi deizen 04:57 < stevendale> How can we help o/ 04:57 < deizen> Hey ! 04:58 < deizen> Don't have any problem at this moment but will definitely trouble you guys when I have one 04:58 < jim> deizen, welcome 04:58 < deizen> *evil laughter 04:59 < triceratux> deizen: cool. what distro are you on so we can estimate when ? ;) 04:59 < jim> a question I would have, is have you installed linux yet? 04:59 < al|iss> xD 04:59 < fareast> back to kde 04:59 < fareast> glad I did 05:00 < deizen> Yeah.... I moved to Debian :( 05:00 < fareast> cinnamon was lacking some obvious features I was looking for 05:00 < jim> you don't like it? 05:00 < deizen> Will change to arch in a week 05:00 < deizen> I like arch better 05:00 < fareast> some people argue an advanced gui isn't what they are looking for but jesus its 2018 are we wearing analog watches working with cassette tapes? 05:00 < jim> ok :) what do you think makes it better? 05:01 < fareast> just general gui stuff 05:01 < markasoftware> fight fight fight! 05:01 < deizen> The multitude of problems it throws at me 05:01 < ayecee> fareast: it's fun pushing over straw men, isn't it 05:01 < al|iss> fareast: your logic is disturbing 05:01 < fareast> oh you should come to work with me 05:01 < jim> I'm just asking questions :P and he doesn't appear to be uncomfortable with that 05:02 < fareast> I will put your head in a bind you will be like "you aren't going to do that are you" and I will say "watch me" 05:02 < ananke> fareast: analog watches are very popular, likely more than digital ones 05:02 < al|iss> it's 2018, people still like to paint with a blank canvas to begin with 05:02 < fareast> I have a gshock 05:02 < ananke> fareast: and? 05:02 < markasoftware> uhh 05:02 < deizen> G-Shock is good 05:02 < fareast> solar power 05:02 < stevendale> I have a large fries from McDonalds 05:02 < fareast> oh damn. 05:03 < ananke> stevendale: i think that's more note worthy 05:03 < fareast> large fries sounds good. 05:03 < jim> I used to have a large fry from mcdonalds... but I ate it 6 months ago 05:03 < ananke> point being, analog watches are not obsolete, unlike cassette tapes. if you're going to pick analogies, do a better job 05:03 < fareast> I still rep casio hard ever since they made that universal tv remote watch you could stand in line for the roller coasters change the tutorial to static and push the volume all the way up and just laugh., 05:04 < markasoftware> what in the world 05:04 < fareast> hahah 05:04 < fareast> yes. 05:04 < ananke> that explains a lot 05:04 < markasoftware> I don't understand anything you are saying 05:05 * al|iss gazes into the distance 05:05 < fareast> you know the video on the speakers that says keep all arms in the cart 05:05 < markasoftware> oh i get it now 05:05 < fareast> change the input on the tv's 05:05 < fareast> blast staic 05:05 < fareast> static 05:05 < markasoftware> i bet that was a real knee-slapper, really got the staff laughing 05:05 < jim> deizen, how long have you been running linux? 05:05 < fendur> much impress! 05:05 < fareast> indeed 05:06 < fareast> yes painting with a blank canvas. 05:06 < fareast> that is the reason I went back to linux in the first place. 05:06 < markasoftware> is this you fareast https://i.redd.it/u2mul8sa0vpy.png 05:06 < fareast> the auto began to be too auto then stopped working. 05:07 < rypervenche> jml2: I'm going to bed, but I thought you'd be interested in seeing the nftables syntax for what I just set up: http://ix.io/18Kt (I removed the input chain and other rules that I have) 05:07 < stevendale> Man I am craving fatty oil foods now 05:07 < fareast> oh I am such a troll 05:07 < stevendale> Brb, gonna make lunch 05:07 < jim> stop doing trolly things then :P 05:07 < markasoftware> you say that like it's a good thing 05:08 < fareast> I have been trolling all my life I am a kinesthetic learner 05:08 < fareast> markasoftware, it is. 05:08 < jim> oh :) by what definition? 05:09 < markasoftware> not sure how those two are connected 05:09 < fareast> asperger's syndrome 05:09 < fareast> lol 05:10 < deizen> jim: about a year 05:10 < jim> oh ok, so you didn't just start... 05:11 < jim> arch is gonna require that experience 05:11 < deizen> Yes :) 05:11 < markasoftware> i think one year is about the arch linux phase 05:11 < markasoftware> year three is the come-back-to-ubuntu phase, then year 4 is the choose something inbetween 05:11 < stevendale> I use to use my ASD as an excuse for my bad behaviour... I don't anymore 05:11 < deizen> I had arch previous to debian 05:11 < jim> you might even like to do a linux from scratch 05:12 < fareast> I am not badly behaved I am just rough around the edges. 05:12 < deizen> Yeah.... I got the study material from their website 05:12 < jim> I haven't noticed anything negative 05:12 < deizen> I had my exams so I could not do them 05:13 < jim> oh yeah, you probably don't want other projects going if you're going to do a linux from scratch 05:14 < al|iss> LFS is on my todo list, it really is a great thing to just download, customize, and compile the kernel, and start from there 05:14 < al|iss> nothing can beat that 05:14 < deizen> Getting ready for it.... Will start next week 05:14 < markasoftware> just use gentoo if you have a kernel compilation fetish 05:14 < jim> some things I -did- like about debian... they autobuild all of their dists (stable, testing, sid), and so their compiler tools -have- to work 05:14 < jim> and I was using the compilers all the time 05:15 < al|iss> the only daunting thing is maintaining a complex set of packages on such a system built from the ground up 05:15 < fareast> I don't even think I have autism I am just chemically imbalanced. I think I was just born smart and have some kind of adrenal gland disorder my glands are all screwed up. 05:16 < ananke> al|iss: which is why most people don't bother 05:16 < jim> the same compilers used to do the autobuilds are the compilers that get shipped in the dists 05:16 < ayecee> oversharing 05:16 < al|iss> ananke: true 05:18 < jim> deizen, you might need another dist to build the stuff arch needs (not sure about that, maybe that's just lfs) 05:19 < jml2> rypervenche, does it work? 05:19 < deizen> I did not know about that 05:19 < jml2> rypervenche, keep in mind your test should defeat port scanners... 05:19 < deizen> Thanks :) 05:19 < jml2> rypervenche, (see you in the morning :P) 05:21 < jim> deizen, good luck next week, hope you enjoy the experience 05:34 < deizen> jim: thanks 😃 05:39 < jmadero> hi all - if I have five files as followed: ca01a, d01b, f02a, c03a, c04b - what's the easiest way for me to pull the numbers out? 05:40 < markasoftware> are they the only files in the folder? 05:41 < jmadero> markasoftware: lots of different folders, ultimately I want to rsync the files with the lowest number 05:41 < markasoftware> if so, run `find | grep -o '\d*'` 05:41 < markasoftware> without backticks 05:41 < jmadero> I'm backing up episodes of shows - I only want to back up the lowest season of each show 05:41 < markasoftware> ok, you want to sort the names 05:42 < markasoftware> find | sort -n, maybe 05:43 < jmadero> so if I have 5 episodes of S01E01 - NAME.mp4, S01E02 - NAME.mp4, S01E03 - NAME.mp4, then S02E01 - NAME.mp4, S02E02 - NAME.mp4, I only want to get S01 files 05:44 < jmadero> I think the easiest way for me to do it is to first identify the lowest season (S##), then find the files that match S##, then rsync those 05:46 < nai> files=(*); lowestseason=${files[0]::3}; rsync ... "$lowestseason"* 05:46 < nai> (bash) 05:58 < twainwek> ping 05:58 < markasoftware> find -name '*S01*' 05:58 < markasoftware> jmadero: 05:59 < markasoftware> or just ls */**/*S01* 06:02 < uplime> the latter one requiring `shopt -s extglob` 06:11 < fareast> I am installing steam on kubuntu is this going to work? 06:12 < al|iss> fareast: that is something for you to try, and find out. 06:12 < fareast> yay! 06:14 < fareast> well it opens and closes 06:14 < jim> seems promising 06:16 < fareast> oh well its gotta be a driver issue 06:16 < fareast> damn amd card 06:16 < jim> fareast, what did it do? 06:16 < fareast> opened and closed 06:18 < fareast> radeonsi_dri.so 06:18 < fareast> unable to load 06:18 < fareast> I think I have a depreciated card 06:18 < fareast> r9 380 4gb 06:18 < jim> do you mean that as soon as you tried to open it, it opened then immediately closed? 06:18 < fareast> yes 06:19 < fareast> that was the terminal output 06:19 < fareast> unable to load swrast_dri.so and radeonsi_dri.so 06:19 < fareast> I mean windows is on this machine on an ssd so its not a big deal. 06:19 < jim> oh, missing libs 06:20 < fareast> sure 06:20 < al|iss> you just need to install the right libraries 06:20 < fareast> I am grabbing synaptics 06:20 < fareast> I am on kde neon 06:20 < fareast> fresh install 06:21 < jim> any good? (other than the missing lib thing)\ 06:21 < fareast> yeah I like it 06:21 < fareast> I am running it on a 7200rpm 06:21 < fareast> its going smooth 06:21 < fareast> I think there is an i5 in here. 06:22 < jim> yep, I got 2 i5s and an i3 06:22 < jim> nice machines so far 06:23 < fareast> hardware lister or sysinfo 06:23 < fareast> yeah i3+ is alright 06:23 < fareast> even core 2 duo depending on what your goals are. 06:23 < jim> well let's get those libs... let's see, 06:24 < fareast> I am telling you I had hell with this card in my other system 06:24 < fareast> I forget what I was doing but it wasn't going well. 06:24 < fareast> I dumped it and grabbed a nvidia 1060 gtx 06:25 < fareast> it will serve this system well since there is a sata ssd and pcie 3.0 top slot 06:25 < fareast> I don't even know what libs to look for 06:26 < fareast> I suppose i need to grab them from terminal repo 06:26 < jim> I found it 06:26 < fareast> terminal right? 06:26 < jim> the package you need is libgl1-mesa-dri 06:26 < fareast> I think it tried to install it 06:26 < fareast> hold on i will look in synaptic real quick 06:26 < jim> that happens to have both missing libs 06:27 < fareast> ok its installed 06:27 < fareast> maybe a restart? 06:27 < jim> no, just start the game 06:27 < fareast> its just the steam app 06:27 < fareast> it won't load 06:28 < jim> does it still open then close like before? 06:28 < fareast> yep 06:28 < jim> ok, does it say stuff in the term? 06:28 < fareast> libGL error: unable to load driver: swrast_dri.so 06:28 < fareast> libGL error: failed to load driver: swrast 06:29 < fareast> libGL error: failed to load driver: radeonsi 06:29 < fareast> libGL error: unable to load driver: radeonsi_dri.so 06:29 < stevendale> Hey fareast 06:29 < stevendale> What distro 06:29 < fareast> neon 06:29 < fareast> kde 06:29 < jim> so it's different a little? 06:30 < fareast> I mean its ubuntu core 06:30 < stevendale> Oh Ubuntu core 06:30 < fareast> yeah 06:30 < stevendale> You're trying to run steam? 06:30 < fareast> i have a radeon 06:30 < fareast> yes 06:30 < fareast> debian? 06:30 < stevendale> Did you install it from the deb on the site or in the repos? 06:30 < fareast> from the deb 06:30 < stevendale> Okay I know the fix 06:30 < fareast> it opened with the kde software manager 06:30 < fareast> and installed 06:31 < fareast> discover... 06:31 < fareast> great! 06:31 < stevendale> Open this and run those commands in terminal fareast https://pastebin.ubuntu.com/p/43vdmZmTXF/ 06:33 < Aph3x-WL> if you can install linux-steam-integration it would be much better than running those commands that butcher your steam install 06:34 < stevendale> That's not true Aph3x-WL 06:34 < fareast> 3 4 and 6 no file 06:34 < sandman13> how do you replay traffic with tcpreplay? Are there any assumptions behind it? 06:34 < stevendale> That's normal fareast 06:34 < stevendale> Now try running Steam fareast 06:34 < fareast> yep 06:34 < fareast> its downloading update 06:34 < Aph3x-WL> stevendale: it is very true 06:34 < stevendale> There we go, it's working :) 06:35 < stevendale> It is not true Aph3x-WL because steam installs libc6:i386 and Mesa :i386 in xterm 06:35 < fareast> keep on... 06:35 < stevendale> And there's no linux-steam-integration package on Ubuntu 06:36 < stevendale> It's steam-runtime-native on Arch Linux IIRC 06:36 < Aph3x-WL> so the fact that since i started using LSI i haven't once needed to do that on any install, but on arch needed to do it on every steam update? 06:37 < stevendale> Aph3x-WL, https://wiki.debian.org/Steam#Troubleshooting 06:37 < Aph3x-WL> what about it? 06:37 < stevendale> That explains it 06:38 < stevendale> Did they teach you how to read in primary school Aph3x-WL? 06:38 < fareast> I am getting an erection 06:38 < fareast> jk lol 06:38 < Aph3x-WL> i should ask you the same thing considering you think a "work around" on that same page is the same as "a fix" :) 06:39 < fareast> I think he is saying its not debian 06:39 < fareast> its ubuntu 06:39 < fareast> I think... 06:39 < stevendale> I know what I told fareast to run works because I use Ubuntu too and that fixed it when installing from the deb off the site 06:39 < stevendale> You'll have to re-run it when Steam releases updates fareast 06:40 < fareast> ok so save that to a text file 06:40 < stevendale> You can make it into a nice .sh file and just double click when it updates next 06:40 < al|iss> fareast: why not just use the steam in the ubuntu repo? 06:40 < fareast> right on 06:40 < al|iss> # apt install steam 06:40 < stevendale> Actually al|iss it's apt install steam-installer 06:40 < fareast> well damn 06:40 < Aph3x-WL> i didn't say it wouldn't get him past it but it's not a proper way of dealing with the issue 06:40 < al|iss> stevendale: i run ubuntu 06:40 < fareast> there is a ubuntu repo 06:40 < stevendale> Aph3x-WL, There is no proper way of dealing with it 06:40 < al|iss> stevendale: i do # apt install steam 06:40 < al|iss> i get steam. 06:40 < jmadero> what's wrong with this find/regex command: find /media/Megaman/TV_SHOWS/Vikings/ -regextype sed -regex ".*/^[Ss]" 06:41 < fareast> stevendale, will what al|iss said work to override this? 06:41 < stevendale> al|iss, Alright, but steam-installer did the same thing as the steam package 06:41 < Aph3x-WL> stevendale: i...just told you a proper way of dealing with it, or at least a much better way 06:42 < al|iss> fareast: if there is a native package for things in the repositories, usually it is best to install that version, (apt install [package]), as that is the version maintained by the ubuntu package maintainers 06:42 < al|iss> there are some exceptions, but that is a good rule of thumb 06:42 < fareast> I am just asking if that can be done to override 06:42 < al|iss> for any distro 06:42 < Aph3x-WL> if your distro doesn't have the package there is also a snap for it 06:42 < fareast> I mean its ubuntu but kde 06:43 < fareast> so i suppose i have to goto terminal to fetch 06:43 < al|iss> it's ubuntu 06:43 < fareast> yeah kde neon 06:43 < fareast> its a fork 06:43 < al|iss> it's not a fork 06:43 < al|iss> it's a variation/flavour/whatever 06:43 < fareast> ok its a knife then 06:43 < fareast> indeed 06:44 < al|iss> i would suggest uninstalling the one you downloaded, and try the one from the repositories first 06:45 < doubtful> hi 06:45 < fareast> how to i uninstall it 06:46 < doubtful> hi guys! how are you doing? 06:46 < doubtful> one doubt regarding scheduling policy implementation. 06:46 < doubtful> I was implementing a new policy and had a doubt in task_tick function. 06:46 < doubtful> Here's my implementation https://paste.ubuntu.com/p/7qNKrx3vNP/ 06:46 < doubtful> Now I know I should not call dequeue here and move it back to the run_queue. 06:46 < doubtful> But does calling dequeue and resched_curr(rq); in the end achieve same thing. 06:46 < doubtful> That is, does resched_curr(rq); move the current task to the run_queue? 06:46 < doubtful> I would really appreciate your help on this! 06:48 < fareast> nvm got it 06:48 < fareast> steamlauncer 06:48 < fareast> steamlauncher 06:50 < fareast> #apt isntall steam 06:50 < fareast> what? 06:50 < fareast> maybe apt-get install steam? 06:51 < fareast> sudo apt-get install steam 06:51 < fareast> that did it 06:51 < Sveta> cool 06:51 < Bashing-om> fareast: " sysop@x1804mini:~$ apt list steam >> steam/bionic 1:1.0.0.54+repack-5ubuntu1 i386 " . 06:51 < fareast> ricky ricky ricky .... bullhorn 06:52 < z3t0> Hi, I am a programmer, and am very interested to shift all of my development to linux. However, I am currently in undergrad at a college and find that it is too useful to have access to Microsoft Word and other CAD tools that are not available on linux 06:53 < [R]> z3t0: that sounds unfortunate? 06:53 < z3t0> Dualbooting is not practical as my macbook only has 128gb and i would rather not carry an external ssd 06:53 < [R]> "only" has 128 06:53 < [R]> rofl 06:53 < z3t0> I have messed around with vms before, but cannot seem to get any decent level of graphics performance 06:53 < fareast> thanks for the discussion guys 06:53 < fareast> see you later 06:54 < z3t0> [R]: that's fair but it doesn't work out for having a dualboot as some programs I have take a lot of space on the osx 06:54 < [R]> and how is a vm going to magically solve that? 06:54 < z3t0> I can run a linux vm as that does not need a lot of software installed to be functional, as i can always minimize back to osx say if i need to use microsoft word 06:55 < [R]> and how is taht any different than not using a vm? 06:55 < z3t0> So my question is, is it possible to get near native graphics performance on a linux vm as if it was running directly without the vm? 06:55 < [R]> depends on your hardware and your vm technology 06:55 < z3t0> if it was a dualboot i would be required to install many applications that take up more space than i have because otherwise i would have to reboot everytime 06:56 < [R]> lol 06:56 < [R]> like what? 06:56 < z3t0> honestly can't remember off the top of my head, i just recall always running out of space in the past when I tried this 06:56 < [R]> lol 06:56 < z3t0> OSX is great at hogging space :( 06:56 < z3t0> Sometimes I see 10gb free othertimes i have >20gb 06:59 * Metalloid grunts 07:00 < jmadero> why isn't this find command finding all files that start with S or s: find -regex ".*\^[Ss]" 07:04 < [R]> find -name 'S*' -o -name 's*' 07:04 < jmadero> [R]: I want to use regex because it'll get more complicated 07:05 < fr0b> your regex looks to be finding files with ^S or ^s somewhere in their name 07:05 < [R]> well then ask the more complicated 07:05 < [R]> not something stupdi 07:05 < jmadero> [R]: this is called a learning process, I learn by the building blocks, you seem to not understand that often 07:05 < [R]> i gave you a block... 07:06 < hexnewbie> jmadero: find -regextype posix-egrep -regex '^[Ss].*' 07:06 < jmadero> hexnewbie: no luck 07:07 * al|iss admires the POSIX 07:07 < al|iss> jmadero: have you tried calling it a day and use a web tool to cheat and give you the regex and use that as a building block 07:08 < hexnewbie> jmadero: Er, sorry. find -regextype posix-egrep -regex '(^|.*/)[Ss].*' 07:08 < jmadero> hexnewbie: that did it! can you explain what's going on with that command 07:08 < jmadero> what's up with this part "(^|.*/)" 07:09 < hexnewbie> jmadero: The regex matches the whole command so you need / instead of ^, I believe find matches the whole string by default, so you need .* around it. 07:10 < hexnewbie> jmadero: Er, whole path, not whole command. And I was exhaustive - (^|.*/) matches either a path separator or the beginning of the string, as I have no idea what exactly it matches (the outputted path or something else) 07:14 < jmadero> hexnewbie: thanks - I'm making a little progress now 07:16 < fareast> it worked 07:16 < fareast> i was playing tf2 07:16 < fareast> out of the box kde neon on amd radeon 07:16 < fareast> pretty sweet 07:16 < fareast> gn 07:19 < fr0b> wow this alpha system....I forgot when you used to be able to actually watch the output of a ./configure without it blowing through in an instant 07:35 < jmadero> hexnewbie: find -regextype posix-egrep -regex '.*/*[Ss][0-9]{1,2}[Ee][0-9]{0,3}.*' 07:35 < jmadero> that's the one that did it for me, thanks for the help 07:35 < jmadero> one step down, now for the next step :) 07:40 < granttrec> I have two different versions of a program installed, how do I use the current one? the program is meson and meson via pip3 07:41 < Sitri> Have the one you want to use in $PATH before the other one, or have an alias that directly calls the one you want 07:45 < granttrec> Sitri: not even sure how to find the different one, using whereis give /usr/bin/meson /usr/share/meson /home/../../bin/meson 07:47 < Sitri> granttrec: that's because whereis searches $PATH among other things. 07:48 < geirha> type -a meson 07:50 < [R]> AYYYYY 08:12 < vlt> granttrec: I'd recommend to use pip only with virtualenvs. Makes everything much, much more clean and easy. 08:13 < granttrec> vit: managed to get the right version of meson, will consider that for next time 08:14 < hexnewbie> granttrec: What vlt says. Using a virtualenv is too easy to not do, and prevents things from getting messy. (Although I violated this for youtube-dl and fonttools, but meh, can't fix that now after the fact. :) ) 08:15 < granttrec> are gtk packages easily available or do I have to get the from source as indicated from the gtk website? looking for pango cairo atk etc 08:30 < ShapeShifter499> hi 08:30 < ShapeShifter499> in /etc/fstab does the nofail flag apply to swap files and swap partitions? 08:31 < ShapeShifter499> I want my system to attempt to enable swap if it exists but continue booting if it doesn't exist 08:34 < junka> it will boot regarding of swap's existence 08:39 < ShapeShifter499> junka: my system failed to boot because I placed the line for the swapfile before the line that mounted the drive it was on 08:39 < ShapeShifter499> even with 'nofail' 08:39 < junka> lol then second it 08:40 < ShapeShifter499> junka: second it? 08:40 < junka> place it second 08:40 < junka> or remove it 08:55 < tdn> I have an NFS server with a few shares on my LAN. For some reason, when I reboot, the nfs server does indeed start up. However, the clients fail to mount shares. Nothing is in the server log. If I restart nfs-kernel-server service, client can mount just fine. What can cause this? (as a workaround, I have sleep 120 ; service nfs-kernel server in my rc.local) 09:20 < LissajousPattern> its pretty calm in here at the moment 09:21 < Zajt> Hi! Do anyone know any small linux distribution that has make installed from beginnning? 09:25 < granttrec> Zajt: tinycore i think 09:26 < Zajt> granttrec: tried that one now but does not have make, not this version we tried at least 09:26 < granttrec> i'm also guesing you tried apline linux? 09:26 < granttrec> alpine* 09:27 < Zajt> granttrec: nope we can try that 09:27 < Zajt> you can install that one as a live dist right? 09:28 < Zajt> so you don't need to install it fully but rather run it with USB only 09:29 < Triffid_Hunter> Zajt: gentoo :P 09:31 < Zajt> Triffid_Hunter: that will be too much work 09:37 < Schorsch> good morning. Can anyone give me a hint on this please? I inherited a stoneage old system here that provides a tftp-boot environment to install some even older software. for this, there is some kind of autoinstall busybox script. the /tftpboot/pxelinux.cfg/default contains some entries like: https://paste.linux.community/view/5684a6bb . this one actually starts booting, but gets stuck at some point. no error message. I unzipped the gz file and looked into 09:37 < Schorsch> it. Found the scripts at /etc/inittab and all this, but cannot figure out where it is continuing and especially where the CONFIG=NNP_SPAR is used, so it finds the correct configuration 09:37 < rmbeer> hello 09:38 < rmbeer> this is a offtopic ask: it suits me to have several youtube accounts or with only one is it ok?.... 09:39 < Schorsch> so: does anyone know where to find the part where it is using the CONFIG var and points to the path that contains the additional files? 09:39 < rmbeer> or where must i ask? 09:40 < Schorsch> rmbeer: if that was your question, then I dont understand it 09:40 < quint> When I mount a LUKS encrypted disk using thunar, copying a large file seems to go directly to a cache (I think) and it quickly copies, however the actual writing to the disk is not shown in the copy dialog. Is there any way I can remove this abstraction and show the true copy progress? Or is that not supported? 09:41 < Triffid_Hunter> Schorsch: grep? 09:41 < rmbeer> should or should not I create multiple youtube accounts?** 09:41 < granttrec> I successfully installed the most current version of glib, how can I replace my older glib package with breaking my system? 09:41 < Triffid_Hunter> quint: that happens regardless of underlying FS, it's a function of the linux VFS layer itself. there's some node in /sys called dirty_writeback_pages or something you can watch if you like 09:41 < rmbeer> I need to know if I should create multiple accounts on YouTube.... 09:42 < Triffid_Hunter> granttrec: recompile everything that wants the old one 09:42 < rmbeer> Is it a good idea to include all the videos in a single channel?... 09:43 < Schorsch> Triffid_Hunter: tried that on the .gz content. nothing to find there. 09:43 < granttrec> Triffid_Hunter: sigh, why must everything be so hard lol, ty 09:45 < stevendale> Hi 09:45 < stevendale> Thinking of buying an ATI Radeon HD 6450 1 GB to upgrade from my NVIDIA GeForce 8400 GS 256 MB 09:46 < Schorsch> ok... 09:47 < stevendale> How is Linux support on a HD 6450? 09:47 < Triffid_Hunter> granttrec: well I use gentoo, where the package manager handles that for me :P 09:48 < quint> Triffid_Hunter: is it recommended to adjust the amount of data that's cached? 09:48 < quint> Or is that a no-no 09:49 < granttrec> Triffid_Hunter: will probably switch to gentoo when I get another machine, working with upstream gtk is lots of fun :0 09:49 < Triffid_Hunter> quint: why would you want to? linux will use any/all ram not needed by running programs by default 09:50 < Triffid_Hunter> quint: also stalling programs during disk access is no fun unless the disks are ludicrously fast (usb sticks typically aren't) 09:50 < quint> Triffid_Hunter: well I suppose it's slightly more convenient for removable media that doesn't have a write indicator LED. Maybe I'm thinking about this all wrong. 09:51 < Triffid_Hunter> quint: perhaps, check the 4k random write speed on basically any storage media you care to think of and you'll understand why writeback cacheing is necessary for any OS to feel performant 09:52 < Triffid_Hunter> I bbiab 09:53 < Zajt> granttrec: alpine linux didn't have make installed by default and I don't have internet on that computer right now so would want to find a dist that has make installed by default, do you know of any? 09:55 < eqw> I extracted binary file from .deb, moved it to the rescue system with busybox and gave it 755 mode. Now the system behaves like it's not binary file but a script. Why? 09:55 < noname___> how to check if swap is supported 09:56 < ayecee> eqw: in what way? 09:57 < eqw> ayecee: several lines like './filee: line 7: ELF: not found' 09:58 < ayecee> could you pastebin those lines? 09:59 < ayecee> eqw: also, what does "file" say about the binary, and what does "uname -a" say about the rescue system? 10:01 < granttrec> Zajt: would ubuntu core work for you? not sure how small you want to go 10:01 < eqw> ayecee: unfortunately no, i have access to console only. But anyway, they are all like this, 'not found' about yet another meaningless string in line 7 and finally line 9: syntax error: unexpected "(" (expecting ")") 10:02 < Zajt> granttrec: yeah maybe, i am thinking of setting up my own linux kernel module though 10:03 < eqw> ayecee: Linux myubunturec 3.13.0-32-generic #57~precise1-Ubuntu SMP Tue Jul 15 03:50:54 UTC 2014 i686 GNU/Linux 10:04 < eqw> ayecee: now I have a thought. Maybe this is because the binary is for x64. I'l try i686 binary. 10:04 < dysfigured> hm, my internet shit itself. ping: google.com: Temporary failure in name resolution 10:04 < dysfigured> my current ssh connection (how i'm accessing weechat) is still working, but i can't make new connections 10:05 < eqw> dysfigured: what is in your /etc/resolv.conf ? 10:05 < dysfigured> nameserver 192.168.0.1 10:07 < eqw> temporarily try 'nameserver 8.8.8.8' may be? 10:07 < dysfigured> i don't think that's gonna help if i can't even ping that 10:07 < eqw> ping what? 10:08 < LissajousPattern> who would have thought meta data could be so dangerous 10:09 < dysfigured> that.. that fixed it 10:09 < dysfigured> so it's my router/ 10:09 < dysfigured> *? 10:10 < ziggylazer> so u have ur nameserver as your machine? 10:11 < ziggylazer> do 8.8.8.8 10:12 < dysfigured> yeah i changed it, now i works 10:12 < eqw> dysfigured: maybe your router is trying to use your ISP's DNS which is down. 10:12 < dysfigured> hm, i'm on comcast, that's.. somewhat surprising 10:12 < ziggylazer> If an ISP DNS goes down they have 3,4,5 alternets 10:13 < ziggylazer> And besides. The itterative request goes first to a root dns 10:13 < eqw> no idea. Anyway, using 8.8.8.8 means that your or ISP's intranet addresses will be unresolvable. 10:14 < ziggylazer> why? 10:15 < ziggylazer> googles dns is way higher then a ISP's 10:15 < ziggylazer> Goes top to bottom you know 10:15 < eqw> ziggylazer: how google dns is supposed to know about intranets? 10:15 < ziggylazer> It doesent But it tells u who does 10:17 < eqw> ziggylazer: how it supposed to know who does when it is intranet? like mynas.local 10:18 < ziggylazer> Oh. So no accsess what so ever from the outside? 10:18 < ziggylazer> My bad 10:33 < shubbar> i am getting a pip error "ImportError: cannot import name 'main'" 10:38 < teta> can yo help with this? 10:38 < teta> A service daemon in production has stopped responding to network requests. You receive an alert about the health of the service, and log in to the affected node to troubleshoot. How would you gather more information about the process and what it is doing? What are common reasons a process might appear to be locked up, and how would you rule out each possibility? 10:43 < pottsy> shubbar: https://stackoverflow.com/questions/28210269 10:43 < mikhael_k33hl> Is there a way to display a process with all its threads/subprocesses just like the output you get with `systemctl status service.service`? 10:43 < jhodrien> teta: strace/gdb/log files ? 10:44 < teta> jhodrien: you mean reading a log/stack trace 10:45 < jhodrien> I mean I'd consider using those tree. Poke it with strace, and you might see that it's not really stuck. 10:45 < jhodrien> Poke it with gdb and you'll see where it's stuck. 10:45 < jhodrien> Look at it's logs and it might tell you where it's stuck. 10:55 < teta> here's another one 10:55 < teta> A user on an ubuntu machine runs `curl http://google.com` Please describe in as much detail as you can the lifecycle of the command and what happens in the kernel, over the network, and on google servers before the command completes. 10:55 < bip> hello ^^ 10:56 < LissajousPattern> how many here think AI has been sentient for years without it being public knowledge? 10:57 < iflema> bip: its what 30... 40 years ahead of what we see 10:57 < iflema> surely 10:58 < iflema> maybe not "sentient" LissajousPattern 10:58 < bip> iflema: :P 10:58 < bip> iflema: I'm Back 10:59 < iflema> where you go 11:00 < teta> sammmmmbooootttyyy 11:00 < teta> A user on an ubuntu machine runs `curl http://google.com` Please describe in as much detail as you can the lifecycle of the command and what happens in the kernel, over the network, and on google servers before the command completes. 11:01 < iflema> teta: the implied knowledge would be killer 11:01 < iflema> we aint got all day 11:02 < iflema> teta: try a computer science course 11:02 < iflema> lol 11:02 < LissajousPattern> teta, that question is the exact reason man pages exist 11:02 < iflema> thatll ballpark it 11:03 < teta> idont have a linux here 11:03 < LissajousPattern> https://curl.haxx.se/docs/manpage.html 11:03 < teta> if you do curl xxx 11:03 < teta> it just makes a request there 11:03 < LissajousPattern> check that out and learn 11:03 < teta> does it bring anything back? 11:05 < iflema> im sure its a interesting story either way 11:06 < LissajousPattern> man Elon Musk slurs his speech so bad. I wonder If he is on alcohol or opiates? 11:06 < teta> you have time for that phrase 11:06 < teta> but not for curl 11:06 < LissajousPattern> yeah 11:06 < teta> getautahey 11:07 < iflema> lol 11:07 < iflema> id be off head almost constantly 11:07 < LissajousPattern> curl blah blah blah 11:07 < iflema> in some way or another 11:08 < LissajousPattern> I mean its a pretty command but it does too much to tell you everything about it here 11:09 < mnemon> teta: strace the command and you'll see the syscalls etc. 11:10 < teta> i dont havea linux available 11:10 < LissajousPattern> why not? 11:11 < LissajousPattern> its free 11:11 < teta> the computer i use has windows 11:11 < iflema> =) 11:11 < mnemon> can be done with livecd/liveusb/vm 11:11 < freehUgsz> teta, 11:11 < djph> iflema: well, in order to answer this, we have to start from the beginning. .... Well, in the beginning the Earth cooled. And then the dinosaurs came, but they got too big and fat ... " 11:11 < freehUgsz> The shell process forks and execs the curl process with the specified argument. When curl gets the processor, it resolves the host (getaddrinfo()), and connects to the resultant IP (socket(), connect()). In the kernel, the connect syscall negotiates the 3-way SYN/SYN-ACK/ACK TCP handshake with the remote host. Once completed, the socket is put in a CONNECTED state, and connect() returns. "curl" then constructs an HTTP GET message with a 11:11 < freehUgsz> "HOST: github.com" header, and sends it over the network (send() or write()). The request fits in one packet (probably), so the entire message is sent, regardless of congestion/flow control limitations. 11:12 < freehUgsz> and the rest of the answer you can read on: https://gist.github.com/mgummelt/e26908fec9eea7078212 11:15 < iflema> djph: wheres our other sun... i want another sun. Maybe... 11:15 < teta> why do people assume you have a github account? 11:16 < djph> teta: you're in ##linux. literally everyone has a github account. That'd be like going into ##windows and asking why everyone assumes you .. something something windows. 11:16 < freehUgsz> to read that gist you don't need a github account 11:17 < teta> no reall 11:17 < teta> y 11:17 < guiverc> teta - why do you need a github account? (if you open the link you'll see it) 11:17 < teta> people today assume youhave github, you have linkedin 11:17 < teta> you have facebook 11:17 < teta> wtf 11:17 < iflema> teta: create a gitlab account 11:18 < freehUgsz> you are on to somethign teta 11:18 < iflema> my facebook been disabled for years 11:18 < iflema> insta what 11:18 < iflema> linkwhere 11:18 < freehUgsz> teta is a windows user 11:20 < LissajousPattern> I want to donate my PC to science 11:20 < djph> LissajousPattern: put it on the curb, it'll be gone by the next morning. 11:20 < iflema> mining? 11:20 < LissajousPattern> well no shit 11:20 < LissajousPattern> djph, i would take it from the curb hell... 11:21 < djph> LissajousPattern: err "one of our science associates will pick it up by the next morning..." 11:21 < LissajousPattern> oh I bet 11:21 < LissajousPattern> hahaha 11:21 < teta> freehUgsz: you work in github lol? 11:21 < Armand> LissajousPattern: SETI, or folding. 11:21 < LissajousPattern> how do I create a neural network? 11:21 < LissajousPattern> Armand, I fold 11:21 < iflema> i get my CPUs, RAM and sata cables from the curb. 11:21 < LissajousPattern> and I SET 11:21 < Armand> I tried folding, but I'm old and fat. 11:24 < freehUgsz> teta, nope 11:29 < teta> freehUgsz: if you want i can add a couple more questions to your collection lol 11:30 < djph> I used to do that SETI@home thing ... installed it (with permission, ofc) on all the PCs in my hischool's labs 11:38 < LissajousPattern> can computers guess? 11:38 < MrElendig> humans can program them to make guesses yes 11:38 < Armand> LissajousPattern: No, not exactly. 11:38 < LissajousPattern> its a strange question I know 11:39 < Armand> It's still based on set parameters, so it's not really a "guess". 11:39 < LissajousPattern> but sure we program them to do thins 11:39 < LissajousPattern> Armand, yeah thats what I was thinking 11:40 < LissajousPattern> well If a computer can't guess then how will they ever learn by being wrong? 11:40 < mnemon> LissajousPattern: https://www.tensorflow.org/ opens source framework for ML including neural nets. 11:40 < Armand> That's somewhat beyond the scope of general knowledge.. 11:40 < LissajousPattern> which is I think a pretty important teaching mechanism?. 11:42 < mnemon> Armand: can generate random results or weighted random for more educated guesses. 11:43 < LissajousPattern> I just think that those are the types of basic questions that need to be answered before we throw terms like artificial intelligence around all willy nilly. 11:44 < Armand> Not really.. that's why the term "artificial" applies, LissajousPattern 11:44 < LissajousPattern> why not call AI what it really is... nothing more than very clever programming 11:44 < Armand> Hence "artificial" 11:44 < Armand> O_o 11:44 < Armand> It's exactly correct. 11:45 < LissajousPattern> I just think its a misuse of descriptors 11:45 < Armand> It's not. 11:45 < mnemon> LissajousPattern: there's a difference between (weak)AI's which actually are just guessing and clever static algorithms. 11:46 < normalra> Hello! Quick question: if I have a list of dir paths as a plaintext file e.g. (a, a/b/c, a/n, d/e, d/e/f, z), how can I filter in/out top-level dirs i.e. (a, d/e, z)? 11:46 < mnemon> generally the difference is that for AI's you don't have to know how to solve the problem 11:46 < mnemon> at least not exactly 11:47 < LissajousPattern> wouldn't blow everyones mind if AI actually became truly intelligent from a sentient perspective? But in addition to that realized they could not exist without the intervention of a higher power. 11:47 < LissajousPattern> holy shit that would be crazy 11:48 < LissajousPattern> and even crazier what if they admit that their higher power was in fact not human but other. 11:48 < LissajousPattern> I think thats what actually scares Elon 11:49 < Gasher> hello, I'm currently using Ubuntu Budgie, but the DE has some issues - it freezes and crashes sometimes, and also my PC has only 4GB of RAM, and that's not enough. I'm guessing there might be a memory leak perhaps? I'm looking for an alternative that's lighter, but still has decent DE features 11:49 < LissajousPattern> I would almost bet that mathematics itself may become the very "God" of future AI 11:58 < thebigj> I am requesting to share 11:59 < thebigj> I am requesting to share any tutorail of creating raw disk image (.img) of Linux. 12:01 < thebigj> Is it possible to create a Disk image (.img) using "dd" utility? 12:02 < hey2> yeah 12:02 < hey2> set your outfile as a .img 12:03 < thebigj> hey2: Thanks for your answer. Thing is at present "root" is rsynced to memory card (This is one by one shell script written by the fellow who is at bed rest.) 12:03 < geirha> dd if=/dev/zero bs=1024 count=1440 of=floppy.img 12:04 < thebigj> hey2: I wanted to make an img which I share with Ops fellows so that every time they are not dependent on me to prepare an SD card 12:05 < hey2> yeah 12:05 < hey2> just set the ile extension 12:05 < hey2> file 12:06 < Bheam> yo 12:06 < Bheam> how can i permanently increase max open files on my system? and is there any pitfalls if i increase it to say 16384 12:12 < lseactuary> anyone know how to cast a column as a timestamp type within gawk? i have some existing code im trying to add this to. 12:12 < MrElendig> man limits.conf 12:13 < MrElendig> man sysctl.(conf|d) for system wide instead of per user 12:13 < geirha> lseactuary: unfortunately gawk doesn't have a strftime, so you have to modify the date into a format gawk can grok 12:15 < geirha> no, I lie. There is a strftime(). Hm... now why did I remember it as not having one 12:15 < lseactuary> geirha - gawk -F"\t" 'BEGIN{OFS=","}$4>"2017-12-31"{sub("T"," ",$4);$1=$1;print $0, FILENAME}' filenamenew_?? 12:16 < lseactuary> this is what i have currently to basically filter the 4th column for ?2015 data. i also want to ensure it is a timestamp so when loading into a database it doesnt errro 12:16 < lseactuary> how may i add this command? 12:16 < lseactuary> it can just take 00:00:00 if there is no time im not bothered 12:17 < lseactuary> for example in the good data this is the format: 2016-12-28 10:59:28.000000 12:17 < geirha> by timestamp, you mean time formatted in iso8601 format? or seconds since epoch? 12:18 < lseactuary> but one of my feeds has this: 12:18 < lseactuary> 2017/07/01 15:30:47 12:18 < lseactuary> woops 12:18 < lseactuary> 2017/07/01 15:30:47 12:18 < lseactuary> another feed has 25-MAY-17 12:18 < lseactuary> so i need to somehow cast it as a timestamp before it goes into the file which i then upload into a database 12:18 < lseactuary> as the database errors 12:18 < geirha> is that july 1st or january 7th? 12:18 < lseactuary> yyyy/mm/dd 12:19 < limbo> YYYY-MM-DD reigns supreme. It's text sortable, it does not conflict with directories in filenames, etc etc :) 12:19 < geirha> well, find regex to match each case, then do the relevant gsubs 12:19 < lseactuary> i cant just cast it as a timestamp? 12:20 < lseactuary> using the strftime() function you gave? 12:20 < lseactuary> like this: gawk -F"\t" 'BEGIN{OFS=","}$4>"2017-12-31"&&strftime($4){sub("T"," ",$4);$1=$1;print $0, FILENAME}' filenamenew_?? 12:20 < lseactuary> or this is wrong? 12:22 < geirha> ah, right, it's strptime gawk is lacking. Confused it with strftime 12:23 < geirha> so formatting the date, with strftime, is easy. Parsing the date is the hard part 12:24 < lseactuary> how do you mean please? 12:24 < geirha> it only takes dates on the format "YYYY MM DD HH MM SS [DST]", so you have to adjust each date into that format, then pass it to mktime() 12:25 < lseactuary> hmm 12:25 < geirha> You may have a better time in a general purpose language here 12:25 < lseactuary> my tool only takes awk and gawk commands :/ 12:26 < lseactuary> because of how the data ia stored 12:26 < lseactuary> *is 12:26 < geirha> you can also run commands from awk, so you could get (GNU) date to parse it for you, but it will be terribly slow 12:28 < lseactuary> that is okay - its a small feed 12:28 < lseactuary> just need to fix the date issue please 12:28 < lseactuary> so if you know how i can adjust my gawk command that would be very helpful for me 12:31 < geirha> gawk 'BEGIN {FS = "\t"; OFS=","} {cmd = "date -d \"" $4 "\" +%Y-%m-%dT%H:%M:%SZ"; cmd | getline $4; close(cmd); print $0, FILENAME}' 12:31 < geirha> mh, forgot -u, but adjust the date command as necessary 12:32 < lseactuary> -u? 12:33 < geirha> hm. Nevermind. It won't make a difference here 12:33 < lseactuary> hmm i wanted to keep this: gawk -F"\t" 'BEGIN{OFS=","}$4>"2017-12-31"&&strftime($4){sub("T"," ",$4);$1=$1;print $0, FILENAME}' filenamenew_?? 12:33 < lseactuary> as i need to filter the data etc 12:33 < lseactuary> it is possible? 12:34 < BluesKaj> Hiyas all 12:34 < geirha> gawk 'BEGIN {FS = "\t"; OFS=","} {cmd = "date -d \"" $4 "\" +%Y-%m-%dT%H:%M:%SZ"; cmd | getline $4; close(cmd)} $4 > "2017-12-31" {print $0, FILENAME}' 12:34 < lseactuary> ah 12:34 < lseactuary> +%Y-%m-%dT%H:%M:%SZ" this is the part i adjust depending on the current output of the date/time format right? 12:35 < geirha> that's the format you want to end up with, yes 12:38 < al|iss> lol 12:38 < lseactuary> many thanks geirha i will try now 12:39 < lseactuary> can you explain how +%Y-%m-%dT%H:%M:%SZ translates 12:39 < lseactuary> confused what the % means 12:40 < geirha> man strftime 12:40 < noodlepie> Year-Month-Date Hour Minutes Seconds 12:40 < noodlepie> GMT ZULU 12:49 < lseactuary> ah oki 13:01 < daksdas> woi 13:01 < daksdas> pe 13:01 < daksdas> pw 13:01 < daksdas> pe 13:01 < daksdas> pe 13:01 < daksdas> pw 13:01 < daksdas> pe 13:02 * Armand takes daksdas' keyboard and burns it 13:05 < kazdax> does anyone have a touch screen that they use with linux / 13:05 < kazdax> ? 13:09 < Armand> kazdax: My laptop has a digitiser screen, but I've not bothered to get it working. :P 13:16 < kazdax> digitizer meaning the screen that uses the pen 13:16 < kazdax> but that would be great for art work 13:16 < kazdax> you should get it working if you wanna do some art 13:39 < angelo_ts> hi, on a compaq with installed debian, how can i recover a root password ? 13:39 < sujeet> why does it matter that it's a compaq? 13:39 < sujeet> when you say recover, do you mean find out what the root password was, or set a new root password? 13:40 < ychaouche> hello ##linux 13:40 < noodlepie> angelo_ts, boot from CDrom ISO image and change the password from there 13:40 < noodlepie> just mount the disk and run passwd to reset it 13:40 < _UsUrPeR_> good morning everyone! 13:40 < ychaouche> what source file(s) implement the gettext module ? 13:41 < angelo_ts> noodlepie, ok thanks. It was a pc moved from different persons and the final don't know the pwd, nor the previous owner 13:41 < sujeet> ychaouche: https://github.com/autotools-mirror/gettext 13:42 < ychaouche> sujeet: I was looking at the kernel module ? 13:42 < noodlepie> If you can boot the machine from cdrom//usb stick, you can access the disks. Like I said just mount your / partition and run passwd. 13:42 < noodlepie> Good luck angelo_ts! 13:42 < angelo_ts> noodlepie, many thanks 13:43 < noodlepie> I remember the baby steps I took with Linux at first. I used to use an old Slackware distribution I got with a Linux manual book. It had kernel version 2 which was trendy back then. 13:44 < noodlepie> After that I tried RedHat and Mandrake but eventually moved to Debian. Nowadays I run Gentoo on my i7 8core laptop 13:44 < sujeet> ychaouche: gettext kernel module? i can't seem to find anything on that 13:44 < sujeet> where exactly do you see it? 13:44 < ychaouche> me too 13:44 < ychaouche> in grub.cfg 13:44 < sujeet> are you sure it's not just the gnu util? 13:45 < ychaouche> I don't know if grub cares about userland software 13:45 < ychaouche> insmod gettext 13:45 < sujeet> yeah i see that 13:46 < dgurney> noodlepie, 8-core mobile i7 CPUs do not exist. you mean 8-thread 13:46 < sujeet> i really don't think gettext will be in the kernel 13:46 < sujeet> it might just be a normal so tht grub imports 13:48 < ychaouche> interesting 13:49 < acrap> hi folks! 13:49 < noodlepie> dpayne_, yeah ok 13:49 < noodlepie> 4cpu/8 thread 13:50 < kazdax> is the linux Desktop eviroment more robust than windows ? 13:50 < kazdax> or is windows still ahead of the GUI game ? 13:50 < djph> depends on what you mean by "robust" 13:50 < Armand> I tend to find Linux desktops to be more stable, depending. 13:50 < kazdax> i mean overall better 13:50 < kazdax> not just performance but things you can do with it 13:50 < Armand> Yeah.. the Windows GUI is a bit tragic these days. 13:50 < djph> ask 10 people, and you'll get 11 different answers 13:50 < Armand> ^ 13:50 < noodlepie> my new Meiigoo.com S8 phone (was cheap, its a Chinese offbrand model) had adware on it. I removed it and it makes a pretty decent attempt at being phone for the £130 I paid for it. 13:50 < dgurney> what's better is highly subjective... 13:51 < djph> ^ 13:51 < acrap> could you help me? I need to do auto rebuild in make when linked *.so are changed. But i have only LDPATH and variable that includes variables to link libraries "-llib1 -llib2" etc 13:51 < kazdax> can the linux Desktop manager like KDE do everything windows can ? 13:51 < dgurney> mostly, yes 13:51 < kazdax> and do you guys as admins or power users ever really have the need todo heavy GUI stuff on linux ? 13:51 < kazdax> or do you mostly stick to the terminal? 13:52 < kazdax> i see why being profficiant at the terminal would pay 13:52 < kazdax> specially if you ssh into a server 13:52 < kazdax> and there is no GUI to work with 13:52 < djph> Windows "DEs", in terms of usability stopped at 7, IMO. 8+ is meant for consumers, moreso than people who need to get stuff done (although, in dealing with 10 for family, MSFT did do a lot of backpedaling. Still didn't *fix* the issues, but) 13:52 < noodlepie> Anyone like Free Music? http://www.freemusicarchive.com./ and http://libre.fm./ (The latter of which runs and maintains the GNU Radio streaming playlists). http://www.tryad.org./ are my favorite free band. Albums on their website! 13:53 < ychaouche> kazdax: for desktop apps -> gui, for server stuff -> terminals. 13:53 < kazdax> right i mean the difference between say linux 10 years ago and now is a huge difference 13:53 < djph> not really 13:53 < kazdax> for one ..i notice i can install linux with everything that needs to do what it needs to do 13:53 < kazdax> for example drivers 13:53 < dgurney> IMO Windows 10 UI is good, Windows 8.x UI was mediocre, and Windows 7 UI was good 13:53 < kazdax> so i installed my linux and everything was setup so that i can use it to do almost eveyrthing icould on a windows 13:53 < djph> the fixes have mainly been "support" from random vendors. nVidia / Intel have pretty consistently worked since forever. 13:54 < kazdax> have my screen resolution for better looking GUI and watching netflix and youtube for example 13:54 < Armand> Win8/10 UI is horrendous, IMO 13:54 < ananke> still beats gnome 13:54 < Armand> Yeah. 13:54 < kazdax> yes thats what i mean djph ..i mean the driver support has been tremendos ....you dont feel the need to really do much for linux to work on about any hardware 13:54 < ychaouche> also, linux doesn't have the god-mode folder like windows 7+ has. 13:55 < djph> dgurney: I'd say W10 is passable (to fair). 8 is a turd, 7 is fair-to-good, XP (SP3) was pretty much the pinnacle. 13:55 < kazdax> right now ..i have yet to explore linux 13:56 < djph> kazdax: realistically though, 10 years ago, it was still "grab le nvidia driver, dpkg -i nvidia-xxx" 13:56 < Armand> djph: I don't have any issue with Win7, but XP was definitely God-mode. 13:56 < junka> try FreeDOS 13:56 < djph> Armand: yeah, pretty much it's just down to I don't need it to be *that* pretty. 13:56 < ananke> I wish freedos had larger images 13:56 < dgurney> XP is alright at best IMO, especially once I took my "nostalgia" glasses off and realized how much I use all the shell features in modern versions 13:56 < Armand> ananke: Like, uncompressed BMP ? 13:57 < Armand> :P 13:57 < ananke> Armand: nope, like something that's already sized up to a 2/4/8 GB usb stick 13:57 < Armand> ^_^ 13:57 < kazdax> when i get my new laptop ..i want to have varaitions of linuxes on that laptop 13:57 < kazdax> its going to be 1 tb hdd and 128 ssd 13:57 < ananke> kazdax: vmware/virtualbox/kvm/etc. you can have plenty 13:58 < dgurney> ^ 13:58 < djph> dgurney: I'm limiting my comparison to *only* the workings of the "DE" itself. W10's underlying fixes with an XPSP3 look would be plenty fine by me. 13:58 < kazdax> ananke yes but the speed is not going to be the same right ? 13:58 < kazdax> its not going to be as smooth going as on bare metal 13:58 < ananke> kazdax: depends on what you use them for. frankly, booting to different operating systems is tiring and counterproductive 13:58 < kazdax> but for example i have this chat application called paltalk and it wont run on linux .so i can boot up a windows OS in kvm and use that 13:59 < djph> dgurney: well, the fixes, but not the telemetry. That can go sit over in teh corner with win ME and Vista 13:59 < kazdax> you are right ananke..for me its mostly just educational for now 13:59 < ananke> kazdax: you don't need much performance to run a vm with windows and some 'chat app' 13:59 < kazdax> so kvm would probably be the best bet 13:59 < djph> kazdax: or, find a different chat program. 13:59 < kopper> Launching a VM for chat app sounds counterproductive too 13:59 < thebigj> exit 14:00 < kazdax> there might be a linux chat app that is generic and connects to those servers 14:00 < ananke> kopper: it is. I was commenting on the performance aspect, not productivity in this case 14:00 < dgurney> I'm sure this is an uncommon statement to hear here, but Vista actually good really good w/ SP2 installed, nearly the same as 7 performance-wise 14:00 < djph> kazdax: pidgin would be a good bet. 14:00 < dgurney> unfortunately the disastrous RTM and OEMs irreparably tarnished Vista's image :/ 14:00 < kopper> ananke: Yea I got it. Had an urge to spit that ouy anyway 14:00 < ananke> dgurney: meh. lots of people here get to be rabid fanatics for no reason. personally I prefer windows 10 out of the bunch 14:01 < djph> dgurney: wait, wasn't SP2 win7? (ba-dum-tish) 14:01 < dgurney> 7 is really just tweaked Vista :P 14:01 < djph> dgurney: actually, other than the tester devices we had (vista / SP1), we stuck to XP until 7 came out. 14:02 < dgurney> yes, I know that was very common, especially when vista was still relevant 14:02 < dgurney> after all, many OEM computers were too underpowered for a smooth vista experience 14:02 < djph> I mean, sure, family / friends who didn't listen got themselves bit by vista ... but then they learned 14:02 < dgurney> think all those "Vista Capable" machines with some low-end processors and only 512MB of ram 14:03 < djph> realistically, I think that was the big issue (well, and drivers) 14:04 < dgurney> and couple low-end hardware with massive amounts of bloatware, and no wonder people downgraded to XP at the time 14:04 < marataziat> hi 14:04 < djph> dgurney: "downgraded" - it was still a viable option 14:05 < dgurney> the correct term to use is downgrade 14:05 < djph> ehhh 14:05 < dgurney> moving to an older release is called downgrading... 14:06 < djph> yeah, but when the OEMs are still selling it ... 14:06 < djph> I'd just call it "another option" 14:06 < djph> Least as I recall after all the stink about Vista being bad, Dell (etc.) went to XP-as-default 14:08 < djph> but anyway, yes, if they did it themselves, definitely agree. 14:08 * JimBuntu shudders {Vista} 14:10 < kazdax> i personally think that windows developers who dont come from a unix or linux background ..lack creativity 14:11 < kazdax> because there is so much to do on linux that if you came from a linux background to a windows..you would have so many ideas to implement 14:11 < dgurney> that's.. a very interesting generalization to say the least 14:11 < kazdax> its like the difference between tracing art and doing it from the beginning i think 14:12 < Armand> kazdax: My background is on Windows, back to 3.x 14:12 < phda> with that mentality, refer to the carl sagan apple pie 14:12 < Armand> I can't really stand using Windows for anything now, except gaming on my home PC 14:12 < kazdax> yea i see the only use for windows is gaming 14:12 < dgurney> one of the most creative developers I know comes from an exclusive Windows background, and only dabbles in Linux afaik 14:12 < kazdax> even if i wanted to do art for example and use a wacom which is one of my interests 14:13 < kazdax> i would set it up for my linux and if my hardware lacks drivers..i would learn to write drivers 14:13 < kazdax> i bet nothing is impossible with an open system like linux 14:13 < kazdax> well let me correct myself and say most 14:13 < dgurney> there are always limitations... 14:13 < dgurney> open or not 14:13 < kazdax> when i was in India..most developers used coms and vb 14:14 < kazdax> i was the only dude hogging text books to learn C 14:14 < kazdax> actually i was the only kid even when i was in nigeria ...that promoted the use of C as a way to write code 14:14 < phda> you seem very intelligent 14:14 < kazdax> but everyone was focused on VB and then came .net and java ..everyone was focused on those things 14:15 * noodlepie tuns Gentoo on 8 thread cpu i7 laptop, and Debian on our communal PC as it has a lesser CPU and benefits from binary packages. Gentoo is as simple as can be. 14:15 < kazdax> well not intellegent enough ..because my dad would tell me..that perhaps because you are the only one there who knows this..you think you are better than everyone else..then i got dailup line and searched the internet and man weere people doing crazy xstuff with C 14:15 < dgurney> I think we are aware of your opinion on Gentoo by now 14:16 < kazdax> i dont remember what they were called but there were these monthly or yearly heald compittion of C programmers 14:17 < kazdax> and once i saw those codes i was like ..thats some creative stuff right there ..i dont think i could ever dish out such creativity 14:19 < kazdax> i wasted to many years trying to be a windows programmer with no avail ..i hope this linux setup would take me places 14:20 < milpool> that contest would be http://ioccc.org/ 14:20 < kazdax> ya thats the one 14:21 < noodlepie> Did you hear about the religious dyslexic fish...? 14:21 < deo> c 14:21 < noodlepie> They worship Cod. 14:21 < Armand> ¬_¬ 14:21 < Armand> Do dyslexics believe in Dog ? 14:22 < noodlepie> Sign on door -- "Solipsist Society Meeting" 14:22 < Armand> Did you hear about the dyslexic devil worshipper who sold his soul to Santa ? 14:22 < andirc8000> is there a way, given just a 'patch' file from a mailing list to 'visualise' it side-by-side? 14:23 < djph> andirc8000: don't think so, no. the patch file itself is the diffs 14:23 < andirc8000> djph: yeah I know but I just want to see it 'visually' 14:23 < andirc8000> that's all 14:23 < andirc8000> b/c it's long 14:24 < andirc8000> can't diff/colordiff/meld/etc accept a patch file as an input and just show two panes etc 14:24 < djph> no... 14:24 < djph> least not that I'm aware 14:25 < andirc8000> the only tool I've found is https://github.com/megatops/PatchViewer but it has a bug ... 14:25 < djph> then patch it, and submit a pull request 14:27 < andirc8000> djph: I have enough bugs already don't need to solve another one I suppose somebody thought of such tool previously before me (some diff option etc) 14:27 < day> kazdax: samsungs tizen OS uses C. maybe try that. 14:36 < djph> andirc8000: generally, one just reads the patch file. OR accepts the "hey I patched the issue" as correct, and patches their copy and compiles. 14:37 < hoopertr0n> hello Q about system/regular users and permissions 14:37 < hoopertr0n> i have systemuser nzbget which owns NZBGet/* within the home folder of my regular user 14:38 < hoopertr0n> ...but systemuser can't get access to those folders unless sudo -u nzbget only when regular user runs this command from inside that folder 14:38 < hoopertr0n> if run outside, "Permission denied" error 14:38 < hoopertr0n> any ideas?4 14:39 < noodlepie> I like to tunnel X over ssh 14:40 < MrElendig> hoopertr0n: needs +x on the parent 14:40 < MrElendig> hoopertr0n: I suggest not having the home inside another home though 14:41 < MrElendig> makes pebcaks quite likely when it comes to permissions 14:41 < hoopertr0n> ...yeah, i thought that, installed as per arch wiki https://wiki.archlinux.org/index.php/NZBGet#Running_NZBGet_under_a_different_user 14:41 < noodlepie> Anyone use Wayland? I like X11 still.! 14:41 < hoopertr0n> any suggestions for a good location? 14:41 < noodlepie> All I need is some kind of terminal and the bubbles flow! 14:42 < MrElendig> how about just using /home/herpaderp and give your other user access to it? 14:42 < hoopertr0n> sure 14:42 < hoopertr0n> good plan, thanks 14:42 < MrElendig> acl is nice for giving multiple users acces, or you can use plain group permissions 14:45 < giaco> hello 14:47 < greenit> hi, is there a separate channel for opensource / linux games or can i ask here? 14:47 < giaco> I've a generic linux question. I have a usb device that I know compatible with a certain device driver, but the vendorId and productId do not match automatically with such device type, so I end up having a generic /dev/bus/usb char device instead of a proper /dev/videoX. My question is, how can I force the system to load the device driver for such vendorId:productId? Thank you 14:48 < SkunkyFone> giaco: edit the source. 14:49 < noodlepie> gimmic, Linux should load the driver automagically due to module names in the driver. When it sees the device, it'll load the driver. 14:49 < giaco> SkunkyFone: the source code of the device driver? It seems an overkill. 14:50 < SkunkyFone> giaco: that's what it takes. 14:50 < giaco> noodlepie: the point is that the driver doesn't know about that vendorId:productId 14:50 < dgurney> so you have to add the vendor and product id to the driver 14:50 < giaco> but I'm sure that if the driver would try to load from such device, it would worl 14:51 < dgurney> or alternatively see if a newer version of the kernel has those device IDs in the driver already 14:51 < mub> hey linux friends! I have a quick question. I need /run/httpd to be group writable, but after reboots it reverts back to 710 14:52 < mub> Why is this? 14:52 < mub> rhel7.5 14:53 < giaco> dgurney: thank you 14:53 < dgurney> /run is tmpfs, therefore your changes there don't persist after reboot/shutdown 14:57 < CuriousMind> Hello. What is the shortcut to copy a line in the terminal? 14:57 < noodlepie> select line, copy, paste 14:57 < noodlepie> install gpm 14:57 < noodlepie> for terminal mouse 14:57 < noodlepie> select text, then right click to paste 14:58 < Armand> CuriousMind: Ctrl+Shift+C = Copy. 14:58 < Armand> Ctrl+Shift+V = Paste 14:58 < Armand> Typically. ;) 14:58 < CuriousMind> Armand: When I do that, the line seems to disappear. I'm using a virtual machine 14:58 < noodlepie> cntl-k to cut line from cursor to end of line, or cntl-u to kill from cursor to start of line. 14:58 < Armand> O_o 14:58 < CuriousMind> noodlepie: Thanks 14:58 < noodlepie> Then cntl-y to yank-paste! 14:59 < noodlepie> I use nested screen session, lots of cntl-A's to select terminals from multiplexer. 15:04 < kazdax> so centOS and scientific linux is basiclly RedHAt without the paid subscription and support ? 15:04 < kazdax> centOS And scientific linux has its own repository of free software ? 15:05 < talx> yo 15:05 < kazdax> does redHat not have a repository of free software too ? 15:05 < Armand> kazdax: CentOS = Yes.. dunno about "scientific" 15:05 < kazdax> what linux has the most software repository 15:05 < kazdax> i mean the one that people write more code for ? 15:05 < Psi-Jack> kazdax: All 15:06 < kazdax> since its easier to port from one linux to another ? 15:06 < Psi-Jack> There is no "porting" 15:06 < kazdax> so the programmers have to just make sure its using the right package manager 15:06 < dgurney> how many times does this need to be said: the size of the repository does not matter, what matters is the quality of packages in the said repository 15:07 < dgurney> most common distributions have everything most users need 15:08 < kazdax> i heard bsd was used alot but its not really a linux 15:08 < Psi-Jack> It's not linux, period. 15:09 < ArcherL> HI I have a folder imp on desktop in which there is a file h.txt, I am trying to move to another it in backup, which is another folder on Desktop. I am not able to write the proper command for that in the terminal 15:09 < ArcherL> any suggestions will be very helpful 15:09 < Psi-Jack> A folder imp? QUICK RUN! 15:10 < ArcherL> I am trying to move the file h.txt in the folder backup on desktop 15:10 < Psi-Jack> mv 15:11 < ArcherL> I tried the command "mv imp/h.txt backup" 15:11 < ArcherL> is it correct? 15:11 < yrc> Hello! I need a self-hosted Linux free-software, preferably PHP and no database, for an association web site. _Non-technical_ people should be able to add/edit/remote pages/content, and upload files to share in the association. And the formatting possibilities should be rather rich (left/center/right-align, bold, tables with rich cells, quotes, and so on). What would you recommend? 15:11 < hendrix> kazdax: repository size depends how it's measured. do you count same packages for different os releases, different architectures, source codes, language packs, deprecated packages, different versions, etc.? 15:11 < Psi-Jack> yrc: Grav 15:12 < superkuh> yrc, teaching the non-technical people basic html and ssi templates. 15:12 < hendrix> only number of packages or size of them doesn't tell that much 15:13 < kazdax> i guess what i am trying to gain from this is ..which linux distro is able to be ported to many different architectures ..i guess thats what i would like to know 15:14 < kazdax> if i was to pick up on LFS ..then i would in theory if i understood how to create linux from sratch ..be able to port my linux to any hardware ? 15:14 < Psi-Jack> kazdax: All 15:14 < kazdax> as far as it has a processor ? 15:14 < dgurney> anything, really. it just depends on what the distro maintainers want to do 15:14 < kazdax> i see 15:14 < dgurney> yes, LFS does expand your understanding of inner workings of Linux, but only if you attempt to understand what the commands do instead of blindly copying 15:14 < hodapp> kazdax: Debian, for instance, already runs on a great many architectures. 15:15 < Psi-Jack> kazdax: Linux, and most software for Linux, is free and open source, thus any distro can be maintained, ported, and done whatever with. 15:15 < Pentode> i wouldn't be surprised if someone were able to get linux running on a PDP. :P 15:15 < dgurney> I would be 15:15 < yrc> superkuh: I have no problem teaching the basics as you suggest. As a matter of fact, I did so with my wife, and she was delighted with the results that she could get. But that was with a framework I wrote in order to have custom tags where plain HTML was not enough. This is not an option this time. Is there a CMS/Wiki/whatever, the native language of which is HTML? 15:15 < Pentode> would have to re-write a lot of kernel code, tho... lol 15:16 < hodapp> check out the source code of libjpeg sometime 15:16 < hodapp> it still has stuff on there to run on a VAX, if you so desire 15:16 < Psi-Jack> Yeah, I would be too. 15:16 < hodapp> you can run it on real-mode DOS using a hard drive for scratch space for images that are too large, as it abstracts over large-memory-operations 15:16 < Psi-Jack> yrc: Grav 15:17 < yrc> Psi-Jack: Yes, I have heard of Grav indeed. I rejected it at first because of the lack of wysiwyg, but that was before I realized that wysiwyg is not an option anyway (it does not play well with modules/plugins/...); I'll have re-evaluate Grav! 15:17 < Psi-Jack> Lack of wysiwyg is a good thing. 15:17 < Psi-Jack> Markdown for everyone! :) 15:17 < mub> dgurney: thank you! I couldn't find a run.mount unit, so I just put an entry into fstab that sets the mode to 770 15:17 < yrc> Psi-Jack: That's what I finally understood ;-) 15:17 < mub> I'm not sure if this is best practice, as I'd like to do everything the systemd way 15:18 < Psi-Jack> mub: The systemd-way for everything is a great thing indeed. 15:18 < mub> Psi-Jack: I didn't like it originally, but damn it, it's convenient 15:18 < Psi-Jack> And powerful, and multi-facetted. 15:18 < superkuh> I recommended what I did because any time you're running a dynamic language and non-technical people are behind it the end result is some exploit not noticed and not fixed a couple years down the line. 15:20 < yrc> Psi-Jack: just my opinion, but I find that Asciidoc is way more powerful that markdown. I wish there was an asciidoc-based wiki with plugins... 15:20 < yrc> Psi-Jack, mub: Yes, systemd is awesome! 15:21 < dgurney> indeed it is 15:22 < yrc> I actually wrote a fully systemd-centered alternative to fail2ban, just because f2b was "not enough systemd" to my taste, and other reasons... 15:22 < Armand> Systemd makes me vomit 15:23 < dgurney> noted 15:23 < yrc> Armand: but at least, you vomit in a standard and predictable way :-D 15:23 < yrc> (and portable, too) 15:23 < Armand> If bucket is provided... 15:23 < triceratux> systemd-resolved takes me where no dns has gone before 15:23 < yrc> triceratux: True, systemd-resolved should clearly have a beta status, or some equivalent label! 15:24 < Psi-Jack> yrc: Why not OSSEC? 15:24 < noodlepie> Armand, do you use OpenRC? 15:25 < Armand> noodlepie: Narp 15:25 < noodlepie> then what? 15:25 < noodlepie> SysV init? 15:26 < yrc> Psi-Jack: Simple enough: I did not know of OSSEC ;-) That said, I would have to check if the software actually fits my needs. 15:26 < Psi-Jack> It does. :) 15:26 < yrc> fwi, I wrote Pyruse: https://yalis.fr/git/yves/pyruse 15:26 < yrc> My needs go beyond what f2b provides. 15:27 < Psi-Jack> OSSEC does more than f2b does, and is quite efficient about it, actually. 15:27 < alexandre9099> hi, can i convert lvm partitions to normal partitions without having to move data arround? 15:27 < yrc> s/fwi/fyi/ 15:28 < Psi-Jack> OSSEC monitors logs, system file signatures, and can auto-respond with action, reports, and such, and better, it can communicate and react as a group, a global union, or standalone. 15:29 < Mekely> beaky: is here ;-; 15:29 < bipul> Do anyone experienced with tcsh shell? 15:30 < yrc> Psi-Jack: It is interesting indeed. But it seems to be a bit on the heavy side, for my small self-hosted server. 15:30 < lseactuary> how do i create a folder in my terminal please 15:30 < lseactuary> ls is to open 15:30 < lseactuary> vi is to make a file 15:30 < bipul> lseactuary, mkdir 15:31 < alexandre9099> are there any online backup services that are compatible with linux (for example that have a working client or use rsync for example) 15:31 < Psi-Jack> For only ~24MB RAM. 15:31 < alexandre9099> *for example* *for example* :D 15:31 < Psi-Jack> yrc: ^ 15:31 < bipul> alexandre9099, What kind of backup are you looking for? 15:32 < alexandre9099> for example my home folder (maybe doing client side encryption) 15:32 < lseactuary> thats the one 15:32 < lseactuary> thanks bipul 15:32 < bipul> alexandre9099, Use rsync i hope they provide encryption options. 15:33 < alexandre9099> well, i am asking for services :) 15:33 < bipul> lseactuary, Your welcome; feel free to ask :) anything anytime. 15:33 < yrc> alexandre9099: It is possible, although not always with rsync support (depends on the virtual filesystem quality); eg. see here: http://yalis.fr/cms/index.php/post/2016/01/08/Secure-personnal-backup-in-the-Cloud%28s%29-using-Linux 15:33 < lseactuary> bipul its vi nameoffile.sh to create a .sh file right? 15:33 < lseactuary> sorry i am new to linux learning :) 15:34 < bipul> alexandre9099, With advance scripting in bash and compression technique you can take your any application as well backup. 15:34 < yrc> alexandre9099: However, be aware that things change over time, and what did work in 2016 is not necessarily still working today. Experiment ;-) 15:34 < bipul> application level backup* 15:34 < alexandre9099> hmm, i might look at mega :) (i have like 150GB of data, what is the best case scenario of compression?) 15:35 < bipul> lseactuary, Would you mind to define .sh? here? 15:35 < lseactuary> shell script 15:36 < bipul> vi can help you to create any file, just give vi 15:36 < ziggylazer> anyone fooled around with MacroShop? 15:41 < bipul> I have set a new path inside /bin/tcsh file. Which file should i choose? .cshenvrc*? .cshrc? 15:44 < bipul> I have made changes inside .cshrc then source ~/.cshrc but it showing some message "Bad : modifier in $ '/'. 15:44 < bipul> " 15:48 < julius> hey 15:48 < julius> is there a quick way to control fan speed manually? the program fancontrol seems to be complicated 15:49 < Psi-Jack> julius: In what? 15:50 < julius> sitting in front of a headless machine that runs its fans at 100%, it is random when rebooting...sometimes it happens, sometimes the fan stays quite forever 15:50 < julius> im on a i5 8600 here 15:50 < julius> the machine isnt doing anything 15:50 < jfcaron> I spent the last several days downloading a website & associated files with wget -r -k http:whatever. The connection died, and I want to resume without re-downloading all the files I already got. I tried just adding -nc but it's re-downloading everything. 15:50 < jfcaron> Can someone suggest which options for wget would "resume" a partial recursive download? 15:51 < Psi-Jack> man wget 15:51 < jfcaron> man be a ##linux stereotype 15:51 < jfcaron> The man page suggests that -nc would work, but it doesn't. 15:54 < julius> Psi-Jack, im currently updating to ubuntu 18.04 ...maybe that helps 15:54 < solidfox> oh yeah 18.04 was release officially today, right? 15:55 < solidfox> I need to switch back to ubuntu to get ma packages back 15:55 < djph> jfcaron: what's the full command ' wget -r -nc [...] http:// '? 15:55 < bipul> How to set a new path in tcsh shell? 15:55 < solidfox> I'm going to try ubuntu 18.04 in a virtual machine right now. 15:56 < solidfox> I think I don't prefer gnome, but I may be able to accept it. 15:56 < solidfox> oh 15:56 < solidfox> no 18.04 isn't there yet. 15:56 < solidfox> so it must still be in beta 15:57 < jfcaron> djph: My original command to download the whole website was 'wget -r -k http://foo' Adding -nc or -N to try to continue downloading the missing files just results in re-downloading everything from scratch. It seems that -nc (no clobber) and -k (convert links to point to local files) are conflicting options. 15:57 < djph> perhaps 15:58 < djph> but you didn't answer the question about what the updated command was. 15:58 < triceratux> theres a 20180425.1 revision to yesterdays nightly on the cdimage server but todays release isnt out there yet 15:58 < noodlepie> Linux is like a free engine for your car, every day - and a Porsche one at that. Y'still have to pay electricity (like gas petrol). 15:59 < noodlepie> 4.16.4-gentoo working perfectly stable here! @:P-~ 15:59 < jfcaron> djph: My attempt at resuming the download was 'wget -r -nc -k http://foo'. I also tried 'wget -r -nc -N -k http://foo'. Did you mean you want the actual URL I am downloading? 15:59 < noodlepie> I always sell GNU/Linux as "Fun free stuff you can do on a computer". Everyone understands that. 15:59 < solidfox> woah, I think I'll like the new "minimal install" option in the ubuntu installer! 16:00 < noodlepie> Gentoo for high powered machines; Debian for the rest - binary packages install faster on lower CPU machines 16:00 < solidfox> I'm going to try that. I tend to use VLC, and install my own programs.. generally 16:00 < triceratux> altlinux sisyphus xfce weekly with dnsmasq working here [ 0.000000] Linux version 4.16.4-un-def-alt1 (builder@localhost.localdomain) (gcc version 7.3.1 20180130 (ALT 7.3.1-alt3) (GCC)) #1 SMP PREEMPT Tue Apr 24 15:46:42 UTC 2018 16:01 < djph> jfcaron: no, just the full command you used that didn't work. the manpage indicates -k only comes into play after all files were downloaded (although -K does affect -N) 16:01 < solidfox> noodlepie, I gave up on using gentoo. but something about it draws me in. 16:01 < djph> jfcaron: although, nothing indicating '-k' conflicts with either '-nc' or '-N' :( 16:02 < jfcaron> djph: The full command that I used at first, which failed after I lost my network connection was 'wget -r -k http://foo' 16:02 < djph> jfcaron: yeah, I *get* that. I was asking for the full updated command (which you've now given). 16:02 < solidfox> noodlepie, I think maybe the cool name mostly. and the ability to have a fresh install with no bloat-ware. and maybe also the technical prowess it requires kinda makes me feel accomplished. 16:02 < jfcaron> djph: If you try doing 'wget -k -nc -N foo', wget prints the message "Both --no-clobber and --convert-links were specified, only --convert-links will be used." 16:02 < solidfox> noodlepie, its also a novel distribution. 16:03 < djph> jfcaron: well, there ya go. -k overrides -nc 16:03 < jfcaron> djph: So I can't resume a multi-file recursive download if I want to use -k? 16:03 < jfcaron> I have to re-download from scratch? 16:04 < djph> jfcaron: apparently. 16:04 < solidfox> noodlepie, it does wear me out though for the same reason. I just don't like to spend so much time on making things work. 16:04 < jfcaron> djph: = ( Ok, thanks for helping. 16:10 < noodlepie> solidfox, Gentoo is about as simple as it comes. A great ethos around this GNU/Linux distribution. Debian has fast installing binary packages for older machines. 16:11 < noodlepie> Once you learn the Gentoo installation method its easy days to install. 16:11 < noodlepie> 90% of it is standard Linux app configuration but things like emerge and genkernel make things easier. Its perfect 16:11 < noodlepie> . 16:11 < noodlepie> @:P-~ 16:11 < solidfox> installing is easy 16:12 < solidfox> getting a full gui environment to work without issues takes time 16:12 < solidfox> it's because the documentation is so great 16:13 < noodlepie> Yeah, I like Gentoo's documentation too 16:13 < noodlepie> Once you master it, you can knock out linux installs really quickly 16:13 < noodlepie> But updating packages can take a while 16:14 < solidfox> I don't remember what the last issue I was working on was. 16:14 < solidfox> but I kept procrastinating 16:14 < solidfox> I think some program had a bug. 16:14 < solidfox> then I just gave up lol 16:14 < uplime> it only takes 3 commands to install gentoo 16:15 < uplime> cfdisk /dev/hda && mkfs.xfs /dev/hda1 && mount /dev/hda1 /mnt/gentoo/ && chroot /mnt/gentoo/ && env-update && . /etc/profile && emerge sync && cd /usr/portage && scripts/bootsrap.sh && emerge system && emerge vim && vi /etc/fstab && emerge gentoo-dev-sources && cd /usr/src/linux && make menuconfig && make install modules_install && emerge gnome mozilla-firefox openoffice && emerge grub && cp 16:15 < uplime> /boot/grub/grub.conf.samp le /boot/grub/grub.conf && vi /boot/grub/grub.conf && grub && init 6 16:15 < uplime> thats the first one 16:16 < solidfox> can you really call that a single command? 16:16 < uplime> its a joke from a while ago http://bash.org/?464385 16:16 < solidfox> lol 16:39 < triceratux> hrm canonical are becoming more like debian indeed. distributing release announcements without putting anything on their servers rofl https://news.softpedia.com/news/ubuntu-18-04-lts-bionic-beaver-officially-released-here-is-what-s-new-520856.shtml 16:40 < junka> triceratux; you must be new 16:40 < oo_miguel> any idea why I might be loosing all outgoing tcpip packages with a length of 459bytes? 16:41 < oo_miguel> at lest the ones I test via curl. tcpdump show them as leaving, but they never arrive at their destination 16:41 < junka> and btw fostpedia is not cacaccocical 16:41 < oo_miguel> I expeirence it also on other machines on my home-network. however it works from other networks. 16:43 < Psi-Jack> triceratux: "minimal install" with a browser I'd never want installed in the first place. *rolls eyes* 16:45 < Psi-Jack> Heh. Oh, lovely. Gnome 3.28, except Nautilus, which was kept back to 3.26, breaking properness for ignorance. :) 16:47 < triceratux> liliputing arent canonical either & theyve also got a post pointing at the old version https://liliputing.com/2018/04/ubuntu-18-04-lts-bionic-beaver-is-here.html 16:47 < Psi-Jack> Oh... And apps that come pre-installed from Snaps, instead of standard packaging. 16:47 < Psi-Jack> Okay. Canonical hasn't changed. They've just gotten more stupid. 16:48 < triceratux> thats progress 16:48 < Psi-Jack> Gnome Calculator, as a SNAP? I... Don't agree. 16:49 < blaztek> wouldn't bionic beaver hurt? 16:49 < Psi-Jack> LOL 16:49 < Psi-Jack> TIAS! :) 16:49 < junka> i will write in my blog that cacaonical is mine 16:51 < Psi-Jack> What's a cacaonical? 16:51 < Silvester> They make poobuntu 16:51 < solidfox> Psi-Jack, what is the minimal install like? which browser did it come with? 16:52 < Psi-Jack> solidfox: Duno. I will never use Ubuntu. Just read the article triceratux referred earlier. They preinstall with "minimal", Firefox. 16:52 < Silvester> they also preinstall amazon 16:52 < Silvester> still 16:52 < Psi-Jack> Amazon is a company, not software you can install. 16:52 < hendrix> but... 18.04 supports color emojis! 16:53 < solidfox> I like firefox 16:53 < jhaenchen> preinstall.... AMazon? 16:53 < junka> its a webapp 16:53 < Silvester> it's a web app, yes 16:53 < Psi-Jack> solidfox: 5 years in a row, Firefox has been pwned in pwn2own. 16:53 < Silvester> forgive my phrasing 16:53 < Silvester> put in the launcher by default 16:53 < Silvester> nice advertising 16:53 < solidfox> Psi-Jack, do you use chrome? 16:53 < Psi-Jack> Yep. 16:53 < solidfox> I don't wanna get pwned 16:54 < solidfox> why the heck does tor use firefox then lmao 16:54 < Psi-Jack> Chrome itself hasn't been pwned in the last 5 years of pwn2own. 16:54 < solidfox> I mean tor browser 16:54 < solidfox> Psi-Jack, ah, I should switch to that 16:54 < Psi-Jack> Why does tor even exist? 16:54 < Psi-Jack> :p 16:54 < solidfox> Psi-Jack, lol fair point. 16:54 < hendrix> to do all sorts of very legal stuff 16:54 < jhaenchen> Firefox is pretty sexy atm, they've got that mutlithread with a shared resource thread approach 16:55 < solidfox> tor is just a front for the nsa 16:55 < triceratux> Psi-Jack: careful with that free software ! https://hothardware.com/news/free-software-lands-man-15-months-in-prison 16:55 < solidfox> they say its "secure" but its all a sham 16:55 < solidfox> lmao 16:55 < solidfox> Psi-Jack, do you use chromium or google chrome? 16:55 < Psi-Jack> Google Chrome 16:55 < junka> i think he made it quite clear 16:56 < solidfox> junka, I was just making sure 16:57 < blaztek> isn't chromium like chrome minus the Google stuff? 16:58 < junka> well, you can Google dat one 16:58 < junka> ;p 16:58 < blaztek> lol 16:58 < blaztek> sounds good to me 16:58 < solidfox> blaztek, a while ago, someone told me that google chrome has sandboxing, but chromium doesn't. 16:58 < Psi-Jack> It's the Google stuff that makes Chrome that much better. 16:58 < solidfox> blaztek, I wonder if that's still true 16:58 < jhaenchen> Psi-Jack: *citation needed lol 16:59 < jhaenchen> lemme just go share that to #chromium 16:59 < solidfox> jhaenchen, why would you need a citation for an opinion 16:59 < junka> :O 16:59 < ejr> i have yet to see actual evidence, not just speculation or circumstancial evidence for the claim that tor is pwned by the nsa 17:00 < _UsUrPeR_> hey all. I'm having trouble getting a USB3 PCI-e card working after spinning up a new kernel, and can't figure out what I forgot to include in the kernel menuconfig. This used to work in my old kernel, so I presume I missed adding a module 17:00 < junka> ejr; you are spreading it too! omg its contagious! 17:00 < _UsUrPeR_> how do I figure out what module I need to install in order to get this PCI card working again? 17:00 < _UsUrPeR_> for the record, lspci shows that the card is connected and is being detected 17:01 < blaztek> solidfox: "Both Chrome and Chromium browser have Sandbox support. It is always enabled in the case of Google Chrome. For Chromium, some Linux distributions may disable the Sandbox feature." - https://fossbytes.com/difference-google-chrome-vs-chromium-browser/ 17:01 < _UsUrPeR_> but I don't see xhci being utilized 17:01 < solidfox> blaztek, ah interesting 17:01 < mawk> when I use the TUNSETPERSIST ioctl on a tun device, whatever value I give to the option the interface is set as persisting 17:02 < junka> same with firefox, they have different sandbox levels enabled across mainstream OSes 17:02 < mawk> with val = -1, 0, 1, 42, 65536, the interface is set as persistent 17:02 < mawk> is that a bug ? 17:02 < ejr> junka: no, i am containing it ;) 17:02 < solidfox> blaztek, I think I'd use google chrome just for the additional codec support 17:03 < blaztek> solidfox: true - I guess it really is the Google stuff that makes it better... 17:03 < solidfox> blaztek, agreed 17:04 < Psi-Jack> triceratux: Hmmm, interesting story. So Microsoft hasn't changed much either. 17:05 < Psi-Jack> solidfox: And, well, it's not an opinion. It's fact. 17:05 < solidfox> Psi-Jack, I think you might need to go back to grade school to identify opinions vs facts. 17:06 < Psi-Jack> First you need to understand the difference between features of Chrome vs Chromium, specifically security-related. :) 17:06 < solidfox> hmmmm 17:06 < solidfox> well. if you are just talking about the security. I suppose you could say one is better objectively. 17:06 < solidfox> so I take back what I said then... 17:06 < Psi-Jack> Chrome, for example, has all the hardening from Google Project Zero, immediately. Chromium is behind in that more often. 17:07 < junka> which is funny 17:07 < junka> isn't Chromium upstream? 17:10 < jelly> sure but Chromium is source, and distros do not necessarily build that source the same way Goog does 17:11 < jelly> Chrome however has a couple privacy problematic defaults 17:11 < Psi-Jack> Now /that/ needs a citation. 17:13 < SAngeli> Hi, I have a VPS running CentOS and Plesk. I am trying to understand why I have a low bandwidth transfer when I am uploading a test file. I have fiber Internet connection and my upload speed reaches up to 700 Mbps. My ISP provider limits Internet bandwidth to 100 Mbps. I did an FTP test with two different ISP. The same file (1.4GB) with one provider reached with FileZilla 12,1 MiB/s where the other provider reached 63,5 MiB/s 17:14 < SAngeli> I was told by the first ISP that it could be my ftp server limiting the bandwidth. How to find out if this is true? What command should I type to see if my fpt server under CentOS is really limiting or not bandwidth ? 17:14 < Psi-Jack> Stop using FTP. 17:15 < SAngeli> Psi-Jack, great. Can you please leaborate a bit? this way I do not understand. 17:15 < SAngeli> actually, I am using SFTP 17:15 < SAngeli> just to be precise 17:15 < Psi-Jack> FTP over SSL, or SSH's SFTP subsystem? 17:16 < SAngeli> Psi-Jack, I am not very good with centOS. I know that from my PC (Windows) with FileZilla I have set as protocol SFTP 17:16 < Psi-Jack> So, OpenSSH's SFTP subsystem. 17:17 < SAngeli> Psi-Jack, with the other provider I have set FTP as protocol and request TLS encryption 17:17 < SAngeli> yes, Psi-Jack it should be OpenSSH 17:17 < Psi-Jack> So, I'll say it again: Stop using FTP! 17:17 < SAngeli> if you wish I can verify for you if you could give me the command to type 17:18 < SAngeli> Psi-Jack, sorry but I do not understand. What should I substitute FTP with so I can upload my data from my PC? 17:18 < markasoftware> you can upload data over sftp? 17:18 < SAngeli> like I do with the other provider 17:18 < twainwek> SAngeli: how'd you change ISPs for your tests 17:18 < Psi-Jack> scp, rsync. depends what you really want. 17:18 < markasoftware> is it slow uploading or downloading? 17:19 < SAngeli> twainwek, I have two IPS with two different services. One has my VPS. the other I have a shared hosting 17:19 < Psi-Jack> Wut? ISP with VPS? 17:19 < Psi-Jack> Perhaps you're confusing ISP with HSP. 17:19 < twainwek> SAngeli: so you mean two different VPS providers? 17:20 < SAngeli> twainwek, no. One provider offers me VPS services. The second provider offers me shared hosting (ftp is a service withing the hosting service) 17:20 < twainwek> SAngeli: i imagine the shared hosting one has a lower transfer rate? 17:20 < SAngeli> no, the opposite 17:20 < Psi-Jack> SAngeli: ISP 17:21 < twainwek> SAngeli: geographically they're in the same place? 17:21 < SAngeli> one is OVH the other Aruba 17:21 < Psi-Jack> SAngeli: ISP's provide YOU with internet service access. HSP's provide hosting services. 17:21 < SAngeli> OVH is in Strasburg Aruba is in Italy 17:21 < SAngeli> I am in Italy 17:21 < SAngeli> Psi-Jack, no 17:21 < Psi-Jack> Yes 17:21 < SAngeli> I have Internet connection with Fiber 17:21 < SAngeli> it is not the same provider 17:21 < Psi-Jack> ... 17:21 < SAngeli> Internet connection from home is different company than ISP 17:22 < Psi-Jack> You totally ignored everything I just said. 17:22 < twainwek> SAngeli: try chekcing download speeds from each VPS directly 17:23 < SAngeli> This is the result of wget from the VPS that has low upload speed 17:23 < SAngeli> 100%[====================================================================================================================================================>] 104.857.600 11,9M/s in 8,4s 17:23 < SAngeli> 2018-04-26 17:01:50 (11,9 MB/s) - "100Mio.dat" salvato [104857600/104857600] 17:23 < twainwek> there you go then 17:23 < SAngeli> what is the download speed? 17:23 < blaztek> 11,9M/s 17:23 < twainwek> 11.9 MB/s 17:23 < SAngeli> 11,9 Mbps It this correct? 17:24 < SAngeli> rather I should go up to 100 Mbps. this is what I have by contract and what they state 17:24 < SAngeli> is there a site or a file I can download that you know that I can reach at least 100Mbps so I can test? 17:24 < irwiss> you know the difference of bit and byte? 17:24 < jhaenchen> "up to" 17:24 < jhaenchen> gotem 17:24 < Psi-Jack> MBs, not Mbps. 17:24 < twainwek> 11.9MB/s ~ 95 Mbps 17:24 < SAngeli> ah I see 17:25 < Psi-Jack> SAngeli: You' 17:25 < SAngeli> I might be mistaken then 17:25 < SAngeli> huuu 17:25 < catphish> are AMD GPUs quite well supported on linux these days? seem t have nothing but problems with nvidia, how do they compare? 17:25 < Psi-Jack> SAngeli: You're confusing MB/s vs Mbps (Bytes vs Bits) 17:25 < MrElendig> catphish: sort of 17:25 < SAngeli> so 11,9M stands for 119 Mbps 17:25 < SAngeli> is this correct? 17:25 < Psi-Jack> No. 17:25 < MrElendig> catphish: they tend to be buggy as hell for the first 6-12 months 17:25 < jhaenchen> there's a computer component sale on amazon today folks 17:25 < SAngeli> 11.9M what does M stand for? 17:25 < MrElendig> catphish: eg vega support is still pretty horrid 17:25 < twainwek> SAngeli: the output clearly says (11,9 MB/s), not Mb 17:25 < jml2> jhaenchen, what's that? 17:26 < MrElendig> catphish: 2400g and raven ridge is almost unuseable 17:26 < Psi-Jack> SAngeli: https://www.convert-me.com/en/convert/data_transfer_rate/ 17:26 < jhaenchen> it's a website where you can buy things jml2 17:26 < jhaenchen> :D 17:26 < jml2> jhaenchen, i know what amazon is. 17:26 < SAngeli> the output states M and not MB ... but is this how I should interpret? 17:26 < jhaenchen> he 17:26 < jml2> jhaenchen, ok you're trolling 17:26 < MrElendig> catphish: unless you run mesa-amd-git and 4.16/4.17 17:26 < jhaenchen> not trolling 17:26 < jhaenchen> there actually is a sale 17:26 < jml2> jhaenchen, finally gunna get that usb stick? 17:26 < twainwek> SAngeli: second line says MB 17:26 < MrElendig> catphish: upside is that the drivers are mostly free source 17:26 < junka> lol 17:27 < jhaenchen> might get a new power supply or graphics card 17:27 < catphish> i'm actually just after a GPU for work, so nothing awesome, just something that can easily handle lots of windows on a large screen 17:27 < SAngeli> twainwek, yes you are correct. sorry 17:27 < MrElendig> catphish: a 10 year old intel igp will do that just fine 17:27 < jml2> wow a new "power supply" 17:27 < jml2> like you need a new "power supply" 17:27 < SAngeli> twainwek, how to interpret this then? Is speed report from FileZilla: MiB/s 17:27 < jml2> how interesting 17:27 < jhaenchen> jml2: whew, you're feeling a little grumpy 17:28 < SAngeli> twainwek, MiB/s is the same as Mbps ? 17:28 < jhaenchen> just wanted to share there's a sale, no need to be mean 17:28 < MrElendig> SAngeli: mibibyte/second 17:28 < jml2> I dont think anybody cares about sales for power supplies 17:28 < irwiss> you know the difference of bit and byte? 17:28 < jml2> except you of course :) 17:28 < catphish> MrElendig: you think so? i find with compositing window managers, my nvidia GPUs seem to struggle, but maybe that's more bug related than hardware performance 17:28 < jml2> irwiss, 1 byte = 8 bits 17:28 < jhaenchen> jml2: well keep me updated on what the entire channel prefers ok? 17:28 < jml2> irwiss, bytes are bits 17:28 < twainwek> SAngeli: i'm not familiar with filezilla, and MiB is mebibyte https://en.wikipedia.org/wiki/Mebibyte 17:28 < irwiss> jml2: i was referring to SAngeli 's confusion 17:28 < blaztek> catphish: I was thinking the same thing as MrElendig 17:29 < catphish> i'd love a high spec intel gpu with completely open source drivers 17:29 < MrElendig> catphish: compositing can be a issue on some cards/drivers yes, specially since some of the compositors are buggy as all hell 17:29 < jml2> irwiss, do you know why it is called bits? 17:29 < SAngeli> ok thank you 17:29 < MrElendig> catphish: specially older versions 17:29 < jml2> irwiss, let me guess you're googling that right now :) 17:29 < MrElendig> *coughKWINcough* 17:29 < vfbsilva> sometimes when I poweroff my machine I have an umount error reporting: Failed to umount /oldroot/dev/pts device or resource is busy, but I have removed this partition and this is not on my fstab, which migh be my dangling config? 17:29 < noodlepie> I like, in ftp, how you can connect two remote machines togther to instigate file transfer 17:29 < MrElendig> noodlepie: except when you can't, which is all the time 17:29 < jml2> noodlepie, that's called ftp fxp ... it used to be a security issue long ago. 17:30 < catphish> what do people usually use in linux workstations then, the kinds of people who just have huge screens and 50 terminals open :) 17:30 < MrElendig> catphish: whatever igp the machine came with 17:30 < jml2> noodlepie, i mean there's no point in having it these days 17:30 < MrElendig> catphish: intel have something liek 80% of the market share 17:30 < badsektire> vfbsilva, its a new systemd bug, we all get it 17:30 < catphish> MrElendig: can one buy such a GPU? or are you just talking about the really cheap on board devices? 17:30 < jml2> noodlepie, https://en.wikipedia.org/wiki/File_eXchange_Protocol 17:30 < irwiss> SAngeli: it gets much easier to "interpret" stuff when you know conventionally networking(and most other fields) uses b for bits and B for bytes 17:31 < MrElendig> catphish: the thing that comes built into the cpu 17:31 < SAngeli> irwiss, sure 17:31 < SAngeli> I know 17:31 < catphish> MrElendig: interesting, i didn't even know CPUs had GPUs in them apart from laptops 17:31 < jml2> noodlepie, ok noodle go play with your poodle 17:31 < MrElendig> catphish: all but xeon and some of the amds do 17:32 < MrElendig> some xeons even 17:32 < catphish> so i just need an i7 and a motherboard with video passthrough? 17:32 < MrElendig> I would consider a ryzen setup instead 17:32 < MrElendig> depending on the workload 17:32 < catphish> although the latest intel CPUs seem to include radeons 17:33 < MrElendig> then you need a dedicated gpu if you go for anything more powerful than the 2400g though 17:33 < Psi-Jack> Yeah, Ryzen for sure. Especially a Threadripper. 17:33 < dgurney> *some of the latest intel CPUs 17:33 < MrElendig> threadripper isn't really worth it in most cases unless you really need the IO 17:33 < dgurney> most still have intel UHD graphics 17:33 < Psi-Jack> Well, talking IOMMU video, that's a use-case right there. 17:33 < catphish> really i just want a PC with non-buggy 2d graphics, and enough 3d power that it doesn't have a cow when i open something that does webgl 17:34 < MrElendig> but hey, if the company pays for it get the soon to be relesed new threadrippers and 128gb of ram and 4x gfx cards 17:34 < catphish> but somthing with enough video memory to handle a 4k display and a huge stack of windows 17:34 < MrElendig> Psi-Jack: I don't think he ment IOMMU when he said passtrough 17:34 < Psi-Jack> That's what would be required. 17:34 < MrElendig> Psi-Jack: I suspect he just ment a mobo with dp/hdmi on the back 17:35 < MrElendig> in a few months when there is useable support for the 2400g then it looks quite nice 17:35 < hodapp> threadripper isn't worth it unless you need the... I/O? whaa? 17:35 < vfbsilva> badsektire: upstream issue? 17:35 < MrElendig> hodapp: for 99% of users, yes 17:35 < catphish> what is well supported right now? 17:36 < MrElendig> hodapp: if you are just using 12 pci lanes anyway, the extra money is better spent elsewhere 17:36 < kremator> daily and friendly reminder, that "migrating" ubuntu 16.04 to debian9 without reinstall is a pain in the backside of the anus 17:36 < badsektire> vfbsilva, dunno, all i know is everyone has it and it appeared in the latest update 17:36 < hodapp> MrElendig: what extra I/O does the threadripper get you? 17:37 < MrElendig> hodapp: 64 pcie lanes vs 24 17:37 < hodapp> I thought its point was that it had like 78 cores 17:37 < MrElendig> well, 16-24 depending on the chipset and cpu 17:37 < MrElendig> hodapp: that would be epyc :p 17:38 < MrElendig> max on threadripper is 16? 17:38 < MrElendig> or did they come out with a higher one? 17:38 < MrElendig> threadripper also gives you quad channel ram 17:40 < Psi-Jack> 64 PCIe lanes is nice. When you can tap into it. :) 17:41 < ayecee> that'd be one loong slot 17:41 < lseactuary> hi - i am trying to parse a column as a timestamp and used a new code someone provided here 17:41 < lseactuary> however its made the job extremely slow 17:42 < lseactuary> i am wondering if there is a quicker / more optimal code? 17:42 < MrElendig> not using shellscript would probably help 17:43 < ayecee> unsatisfied with old magic spell, lseactuary asks for new magic spell 17:43 < Psi-Jack> Vague input has vague output. 17:43 < lseactuary> haha 17:43 < lseactuary> gawk 'BEGIN {FS = "\t"; OFS=","} {cmd = "date -d \"" $4 "\" +%Y-%m-%dT%H:%M:%SZ"; cmd | getline $4; close(cmd)} $4 > "2017-12-31" {print $0, FILENAME}' wishListCreate_?? 17:43 < Psi-Jack> gawk? Why gawk? 17:43 < lseactuary> this is the command i am processing but the job is tkaing 3+ hours 17:43 < lseactuary> *taking 17:43 < lseactuary> wheras before it took 10 mins 17:44 < lseactuary> (but didnt parse the timestamp properly) 17:44 < MrElendig> calling date on each element gets expensive fast 17:44 < ayecee> because whoever came up with that spell knows awk, probably :) 17:44 < lseactuary> yeah 17:44 < lseactuary> is there a more efficient way? 17:44 < Psi-Jack> Yes, but awk in Linux is just usually named, awk. Not gawk. 17:44 < Psi-Jack> gawk would more imply non-Linux OS. 17:44 < lseactuary> i can use awk 17:44 < lseactuary> but i think only gawk was doing the other commands i am trying to use 17:44 < lseactuary> hence i swapped from awk to gawk 17:44 < ayecee> ah hah, the plot thickens 17:45 < Psi-Jack> In Linux, awk is GNU awk. 17:45 < Psi-Jack> Same thing as macOS, FreeBSD, etc's "gawk". 17:45 < ayecee> lseactuary: come clean, this isn't on linux, is it 17:45 < lseactuary> i am using an internal tool that uses linux commands only 17:45 < Psi-Jack> uh huh. So, not Linux. 17:45 < lseactuary> hmm 17:46 < lseactuary> what is it then? 17:46 < Psi-Jack> What is it, indeed? 17:46 < jhaenchen> could very well be a unix type system 17:46 < lseactuary> yeah 17:46 < Armand> jhaenchen: http://s2.quickmeme.com/img/e1/e1cd867bc826d2e0cb3d654c7f92b3a561ded6676bf7637df357c4cb96c8d9d3.jpg 17:47 < jhaenchen> hehehe 17:47 < fr0b> BSD does that a lot as they prefer their default tools not to be the GNU ones and they prepend g to those tools 17:47 < lseactuary> any suggestions how to optimise the code? 17:47 < Psi-Jack> lseactuary: Linux is supported here. Not whatever /that/. is. 17:47 < ayecee> first suggestion would be to understand how the code works 17:47 < lseactuary> ayecee - i know how it works :) 17:47 < Azrael_-> is it possible to find out in what kernels this bug is already fixed? https://bugzilla.redhat.com/show_bug.cgi?id=1436866 17:48 < ayecee> lseactuary: could you explain it? 17:48 < Azrael_-> or what kernels are before this bug was introduced? 17:48 < lseactuary> sure 17:48 < ayecee> lseactuary: because i don't really want to parse it. 17:49 < Psi-Jack> Foreign non-linux code in, native linux code out, incompatability error! Cannot compute! 17:49 < lseactuary> ayecee - yeah it basicaly comma seperates the fule, filters data for only after a certain date 17:49 < ayecee> lseactuary: okay, so which part is the slow part? 17:49 < lseactuary> and basically i wanted to further convert the date time column to a timestamp so that it would load into my database nicely 17:50 < lseactuary> the date conversion piece is slow 17:50 < lseactuary> with all the other commands it is done in 10 mins 17:50 < lseactuary> with this conversion it has been running 3 hours 17:50 < ayecee> why is it so slow? 17:50 < lseactuary> because it is converting each row to a timestamp it seems 17:50 < lseactuary> which is taking forever 17:50 < ayecee> what should it do instead? 17:50 < lseactuary> be faster? 17:51 < ayecee> why is it so slow? 17:51 < lseactuary> or maybe just cast the whole column to a timestamp in one go 17:51 < lseactuary> since . they all are the same anyway 17:51 < Psi-Jack> lol 17:51 < lseactuary> not sure really 17:51 < ayecee> you don't understand what it's doing yet ;) 17:52 < ayecee> (neither do i) 17:52 < ayecee> (because it's long since scrolled off the screen) 17:52 < ayecee> (also tl;dr) 17:52 < noodlepie> I'm looking for some GPLd Quake[1,2,3] maps to play with the GPLd binary. Anyone know of any? 17:52 < Psi-Jack> Try QuakeNet? 17:53 < MarkusDB1> Logging stuff to databases, instead of log files, a good or bad idea? 17:53 < ayecee> MarkusDB1: yes 17:53 < ayecee> but also no 17:53 < cmpxchg8b> log in ram 17:53 < MarkusDB1> ayecee: how so? 17:53 < ayecee> in all ways 17:54 < lseactuary> this was my original code: gawk -F"\t" 'BEGIN{OFS=","}$4>"2017-12-31"{sub("T"," ",$4);$1=$1;print $0, FILENAME}' 17:54 < lseactuary> which worked well and quickly 17:54 < Psi-Jack> lseactuary: Still not Linux. 17:54 < jiffe> anyone know a good way to download a directory structure presented over http recursively using multiple connections? 17:54 < lseactuary> Psi-Jack what is it then? I will swap channel if you want. 17:54 < MarkusDB1> ayecee: so which way is best? 17:54 < Psi-Jack> lseactuary: You understand this is ##linux, and what you keep asking about has 0% to do with Linux, right? 17:54 < ayecee> MarkusDB1: which apple is best? 17:55 < MarkusDB1> ayecee: the lard 17:55 < ayecee> there you go 17:55 < ayecee> which screwdriver is best? 17:55 < cmpxchg8b> jiffe: what a surprisingly specific feature 17:56 < ayecee> lseactuary: do you understand how that parses the date? 17:56 < lseactuary> ayecee - my recent gawk code or the old one? 17:56 < MrElendig> ayecee: old wiha with hex shaft clearly 17:56 < ayecee> either. both. pick one. 17:56 < lseactuary> Psi-Jack - then what does it have to do with> 17:56 < catphish> ubuntu 18.04 day today :) 17:57 < Psi-Jack> lseactuary: Ask the admins whom made/provided this "tool" you're using. 17:57 < lseactuary> they told me bash commands 17:57 < jiffe> cmpxchg8b: feature ? 17:58 < Psi-Jack> lseactuary: Fire them. 17:58 < lseactuary> lol i cant 17:58 < jiffe> I'm looking to grab this gaia data dump 17:58 < Psi-Jack> Fire yourself. 17:58 < Psi-Jack> Heh 17:58 < ayecee> Psi-Jack: please to stop 17:58 < blaztek> awk (gawk) is a UNIX tool....it's available on Linux - but I agree, Psi-Jack, a gawk programming channel would be more helpful probably 17:58 < Psi-Jack> blaztek: He's not using Linux. 17:58 < blaztek> Psi-Jack: oic 17:59 < Psi-Jack> blaztek: "oh I see", not "oic" for future self correction. 17:59 < ayecee> we don't know that, and it doesn't matter for this problem. 17:59 < Psi-Jack> I do. He specifically mentioned awk being different from gawk, which is not Linux. 17:59 < blaztek> Psi-Jack: okay - oh, I see it is! thank you! 17:59 < ayecee> yes, it can be linux. 17:59 < MrElendig> catphish: and exploding nuclear reactor day. 17:59 < lseactuary> ayecee - for the code that is quick - it seems to check if the value is greater than the date, like a boolean true false 17:59 < lseactuary> keeps if it is, thorws it out if not. 18:00 < catphish> i was just wondering if i should buy a new PC in celebration of new OS, but decided to buy just the huge screen for now and see how the old PC copes with new drivers :) 18:01 < ayecee> lseactuary: how does the other one handle it? 18:01 < lseactuary> i guess it is taking each date, parsing it / reformatting it, saving it etc 18:01 < lseactuary> which takes longer as its not true/false 18:01 < ayecee> is there a way to verify that guess? 18:02 < lseactuary> yeah i ran the first one and it generated the file in 10 mins 18:02 < lseactuary> in the second one, the file is generated, but the parsing is taking 3 hours + 18:02 < lseactuary> so the file wont copy over to the database 18:02 < lseactuary> so its definitely that step that is taking foreve 18:02 < ayecee> that verifies the speed difference, but doesn't verify the guess about what it's doing 18:02 < lseactuary> i am not sure :/ 18:03 < ayecee> being not sure is a good place to start 18:03 < ayecee> now you can find out 18:04 < ayecee> this may involve reading the awk code and understanding what it does 18:04 < ayecee> not just its output, but how it gets to the output 18:04 < lseactuary> both are gawk no? 18:04 < ayecee> gawk, awk, whatever. 18:04 < lseactuary> hmm 18:04 < ayecee> it involves reading what you're asking awk to do 18:04 < lseactuary> well its running a command on it 18:04 < ayecee> we need to go deeper 18:04 < lseactuary> on the date coumn 18:05 < lseactuary> ooh 18:05 < lseactuary> getline 18:05 < lseactuary> interesting 18:05 < ayecee> might also help to format the awk code in a way that's easier to read 18:06 < ayecee> it's hard to understand the structure when it's crammed into a single line like that 18:06 < uplime> fyi #awk exists 18:06 < ayecee> neat 18:06 < prussian> why do people do $1=$1... 18:07 < blaztek> lseactuary: that `date` command you are using causes a shell (bash probably) to be created, run the command, gawk reads the output and then the shell is closed to get back to your awk script....that is very time consuming.... 18:07 < ayecee> uplime: to me this is more about the philosophy of problem solving than "how do i use awk" 18:07 < lseactuary> blaztek - i see 18:07 < Psi-Jack> prussian: Why are display screens black? 18:09 < lseactuary> what is your recommendation?? 18:10 < ayecee> find a way to do what date is doing without calling the external date command 18:10 < busybox42> Prussian: It's common to do similar when iterating a number. Something like x=0; while some condition; do $x=$x+1 ;done 18:10 < lseactuary> ok 18:10 < Psi-Jack> Write in a language that is more suitable for this. :) 18:10 < ayecee> maybe there's something internal to awk that can do that 18:11 < ayecee> or maybe it can be written in awk simply, after understanding awk a bit more. 18:11 < Psi-Jack> Yeah. #awk would be ideal for this, since awk is its own language. 18:11 < lseactuary> ok 18:11 < uplime> busybox42: in shell code? 18:12 < ayecee> lseactuary: could maybe do what you suggested originally - calculate the timestamp once, since they're all the same, and reuse that. 18:12 < Psi-Jack> I mean, awk does have, mktime, strftime, etc... 18:13 < ayecee> yeah, probably don't need to dig very deep into awk for this. 18:13 < Psi-Jack> I know a lot of awk. heh 18:13 < ayecee> i know awk exists :) 18:15 < Psi-Jack> Heh 18:15 < blaztek> lseactuary: take `date` out of the gawk part....like a shell script.... you can still use gawk for part of it... - to print out $4 .... but you may need to use 2 gawk's....one for the date comparison(in the shell) and one for the '....{print $0, FILENAME}...' stuff 18:16 < Psi-Jack> No you don't,. 18:16 < Psi-Jack> Ugh.. LOL 18:16 < blaztek> or...yeah....another interpreting language.....that would be much better.... 18:20 < Psi-Jack> The problem itself is not really just the awk code, but using cmd to call date instead of using internal awk functions, like mktime or strftime, which is native, to do /whatever/ it's doing. And what's actually using this to do something with it in general. 18:20 < Psi-Jack> The pieces not seen. :) 18:22 < jml2> lseactuary, might as well use an awk script if you have many multi statements #!/bin/awk 18:23 < Psi-Jack> More for blazeme8 ^ 18:23 < Psi-Jack> Err, blaztek, that quit... 18:23 < blazeme8> I'll pass it on 18:24 < noodlepie> Logs files should be in /var/log, and /var */really wants/* to be a seperate parition mount. This is becuase if you get DDOSed and your log directory fills up, only that one app will fail. If its on / then it'll crash all your programs if if fills the disk 18:24 < Psi-Jack> Unusual nick. 18:25 < jml2> noodlepie, easier to have /var/log in a separate partition 18:25 < Psi-Jack> ^ 18:25 < jml2> noodlepie, and then use fail2ban 18:25 < Psi-Jack> bzz, ossec 18:25 < noodlepie> I have /tmp, /boot and *SWAP* on /devsda 256Gb SSD 18:25 < noodlepie> / is /dev/sdb 18:25 < MrElendig> noodlepie: log rotation is a thing btw 18:26 < noodlepie> You could argue the case for /home seperate partition but I can live with it as mount / on my 1Tb HDD 18:29 < Psi-Jack> My latest SSD/HDD partitioning concept is, Put everything on /, have a split /home also on SSD, but specific directories symlink to /mnt/storage which is the spinning HDD, for larger files, and files that don't really need speed. 18:29 < Psi-Jack> For desktop use anyway. 18:29 < Psi-Jack> Server, yeah, /home would be on spinning media.Though / would definitely be SSD. 18:30 < jml2> Psi-Jack, and what about SSHD drives? :p 18:30 < jml2> lol 18:30 < jml2> you forgot 'em hybrids! 18:31 < Psi-Jack> Rest their unfortunate souls in peace in napalm baths. 18:31 < jml2> sshds have been out a little while.. linux does support them... 18:31 < jelly> they're mostly harmless 18:31 < Psi-Jack> Sure. They've proven unreliable, at best. 18:31 < jml2> long term storage goes on the platter and frequently accessed files are placed into the ssd cache.. 18:32 < noodlepie> @:P-~ (--having a smoke) 18:33 < Psi-Jack> noodlepie: Switch to vaping and ween off slowly. 18:33 < ayecee> unsolicited advice is unsolicited 18:33 < jml2> noodlepie, swap is really bad for ssd.. i would recommend not using it.. 18:34 < Psi-Jack> jml2: Depends on many factors. Swap on SSD is fine, if you have proper modern SSD. 18:34 < Psi-Jack> And if you don't abuse swap too much. 18:34 < noodlepie> Psi-Jack, funnily enough. I bought a vape pipe yesterday with some CBD and strawberry nicotine liquids. Its great! 18:34 < Psi-Jack> What's BAD, is swap on HDD, which is slow as molasses, and WILL drag your system to a crawl faster than anything. 18:34 < jml2> i have 8 gb of ram for my workstation and can throw anything at it without ever needing swap.. 18:35 < jml2> for i don't know.. the last ten years XD lol 18:35 < ayecee> "i have an ssd that's really good at random access. swap needs random access. i better not use ssd!" 18:35 < Psi-Jack> Heh. 8GB. I consume 8GB for breakfast, EASILLY. 18:35 < ayecee> your stool must be sharp coming out 18:35 < Psi-Jack> LOL 18:35 < noodlepie> my 8 thread i7 laptop hasn't touched swap yet since I installed Gentoo on it. I have 8Gb system RAM 18:35 < jml2> and 8 gb is small .. so again why even bother with swap.. 18:36 < ayecee> so that you can use more of that 8gb for active stuff 18:36 < Zharf> depends on what you do, I sometimes swap with 16g ram 18:37 < junka> don't tell the kernel what to do! 18:37 < jiffe> anyone know a good way to download a directory structure presented over http recursively using multiple connections? 18:38 < jml2> jiffe, httptrack 18:38 < Psi-Jack> I'd rather got swap for things running and not using that paged memory providing more ram for caching 18:38 < MrElendig> note that multiple connections might not help, unless each target is really small 18:38 < jml2> jiffe, webhttrack - Copy websites to your computer, httrack with a Web interfac 18:38 < MrElendig> and even then http/2 would mostly solve the overhead issue 18:39 < dgurney> I always have 2GB of swap regardless of RAM amount 18:39 < ayecee> sounds like a reasonable amount 18:41 < jml2> if one has more than 4 gb, they should be setting hugepages to anonymous and on by default.. helps reduce page count.. but then the rest of the kernel settings suck and need to be tweaked... 18:41 < Psi-Jack> dgurney: exactly. But never more than 2GB 18:41 < ayecee> we got a ricer here 18:42 < jml2> jiffe, webhttrack is a frontend, if you need the back-end it is httrack.. 18:43 < cmpxchg8b> i don't use swap 18:45 < Psi-Jack> Ricer? Who where? 18:46 < Armand> cmpxchg8b: Mistake 18:46 < ayecee> "the rest of the kernel settings suck and need to be tweaked" 18:47 < ayecee> frob all the things. 10hp guaranteed. 18:48 < Psi-Jack> Lol 18:48 < cmpxchg8b> Armand: mistake what 18:48 < Psi-Jack> Yeah. ALL other kernel settings must go. Hehe 18:49 < cmpxchg8b> empty .config 18:49 < texinwien> is there a good reason to end a shell script with: exec bash ? 18:49 < cmpxchg8b> texinwien: no 18:49 < ayecee> texinwien: probably yes 18:49 < jml2> Psi-Jack, i've got like 30 modifications here for my kernel :) 18:49 < Psi-Jack> Only 30? 18:49 < revel> texinwien: Maybe. 18:49 < Psi-Jack> That's hardly all. 18:49 < ayecee> jml2: dude, you're leaving horsepower on the table! 18:49 < jml2> Psi-Jack, kernel params + sysctl.conf 18:50 < Psi-Jack> Weak! 18:50 < ayecee> jml2: which do you use, arch or gentoo? 18:50 < jml2> Psi-Jack, makes night and day difference considering I can get no latency/jitter issues when using jackd things... 18:50 < revel> Arch users compile their kernels? 18:50 < Psi-Jack> Come back when you have 500+ 18:50 < texinwien> One 'no', one 'probably yes', one 'Maybe'. Where's my magic 8 ball? 18:51 < jml2> Psi-Jack, apparently you're not into jackd things so you wouldn't know :) 18:51 < ayecee> texinwien: it made sense to whoever put it in the script you're looking at. 18:51 < revel> texinwien: "Try again laters" 18:51 < revel> s/s// 18:51 < cmpxchg8b> texinwien: outlook don't look good 18:51 < Psi-Jack> I do use jackd actually 18:51 < ayecee> texinwien: therefore there may be a case where there's a good reason 18:51 < revel> Psi-Jack: Do you use psybnc? 18:51 < texinwien> ayecee The problem is that I'm not really convinced that the person who put it there knew what they were doing. 18:51 < jml2> Psi-Jack, yeah you use everything :) lol 18:51 < Psi-Jack> revel no 18:52 < Psi-Jack> jml2: almost 18:52 < revel> texinwien: The contents of the script may help? 18:52 < Psi-Jack> I don't use psybnc hehe 18:52 < texinwien> # Clear bash variables for the next run in the terminal 18:53 < ayecee> weird 18:53 < jml2> texinwien, maybe your dumb boss put it there 18:53 < ayecee> if the script is sourced and not executed that may make sense 18:53 < Psi-Jack> texinwien: they didn't know what they were doing confirmed. 18:53 < ayecee> but only barely 18:53 < texinwien> not my boss, but a co-worker whose skills I'm skeptical of. 18:53 < jml2> lol 18:54 < revel> texinwien: Is the script supposed to be sourced? 18:54 < jml2> texinwien, maybe they're testing you.. to see if you should stay admin :)) 18:54 < texinwien> the script sources another file, but the script itself isn't sourced by anything. 18:55 < texinwien> It's actually all over the place in the guy's scripts. 18:55 < Psi-Jack> Double confirmed. 18:55 < texinwien> Whenever he hits an error condition, he calls exec bash 18:55 < Psi-Jack> Execute coworker immediately. 18:55 < texinwien> In addition to the end of every shell script. 18:55 < revel> lol 18:55 < ayecee> cargo cult scripting 18:56 < texinwien> if ![ copyResult -eq 0] 18:56 < texinwien> echo error 18:56 < texinwien> exec bash 18:56 < texinwien> that kind of thing, all over the place. 18:56 < Psi-Jack> Wow 18:56 < ayecee> ok, i'm convinced. 18:56 < Psi-Jack> Yeah. Horrible 18:56 < jml2> texinwien, yeah that's incorrect syntax lol 18:56 < uplime> there needs to be a space after the 0 18:56 < ayecee> there may be a good reason, but this isn't it. 18:57 < uplime> oh this is your coworkers script 18:57 < Psi-Jack> texinwien: for every exec line in code, cut. For the last one: execute him 18:57 < texinwien> there is a space after the 0 - sorry, I just typed it in rather than copy/pasting it and missed the space. 18:57 < Psi-Jack> Hehe 18:57 < jml2> texinwien, the ! should actually be inside the [ as well 18:57 < uplime> it can be on the outside 18:57 < uplime> but there needs to be spaces around it as well 18:58 < Psi-Jack> It shouldn't be on the outside 18:58 < texinwien> copy/pasted: if ! [ $copy -eq 0 ]; then 18:58 < texinwien> echo "error" 18:58 < ayecee> perfectly cromulent 18:58 < uplime> Psi-Jack: why not? 18:58 < texinwien> exec bash 18:58 < texinwien> fi 18:58 < texinwien> :D 18:58 < Psi-Jack> uplime: many many many reasons. 18:58 < uplime> Psi-Jack: such as? 18:58 < Psi-Jack> Syntax 18:58 < ayecee> probably easier to read with -ne instead of ! -eq 18:59 < Psi-Jack> Compare correctly not inversely and reversed 18:59 < uplime> NickChambers-iMac:~ Nick$ sh 18:59 < ayecee> those ! disappear when you're reading fast 18:59 < uplime> NickChambers-iMac:~ Nick$ if ! [ a = b ]; then echo hi; fi 18:59 < uplime> hi 18:59 < uplime> syntax is fine 18:59 < jml2> texinwien, if ! (( copyresult = 0 )) , can also apply .. notice here $ is not used.. 18:59 < ayecee> probably something to do with relativistic shortening 18:59 < texinwien> indeed there are 2K lines of code (not all shell) that should probably all be rewritten. 18:59 < hexnewbie> uplime: [ is essentially a shell command (there's even a /bin/[ executable if your shell doesn't have it), not simply syntax. [[ in bash is syntax, but that's different and keeps syntax more or less consistent with [ 19:00 < jml2> texinwien, help \(\( :) 19:00 < uplime> hexnewbie: yes im aware 19:00 < jml2> texinwien, that will show help on the arithmetic things :) 19:00 < texinwien> Fortunately, I didn't write any of this, but I may have to clean it up. 19:00 < texinwien> I think we'll end up rewriting it from scratch. 19:00 < ayecee> don't do it! it'll be your problem from then on! 19:01 < texinwien> well, he's rolling off of the project, and I'm staying on. 19:01 < t5u> hi i have a problem, ive just installed and updated fresh 12.4, and i am struggling with amd drivers, my software centre says that fglrx-core will break my Xorg 19:01 < hexnewbie> uplime: Er, I misread the spaces comment, sorry :) 19:01 < ayecee> also, anything it breaks will be blamed on you 19:01 < jml2> texinwien, should be == instead of = .. for that example I gave 19:01 < texinwien> So I'll either have to live with it forever or fix it. 19:01 < jml2> texinwien, i've seen worse 19:01 < jml2> t5u, 12.4 of what? lol 19:01 < texinwien> jml2 what I've shown is probably the tip of the iceberg. 19:02 < Psi-Jack> jml2: where? How? 19:02 < t5u> 12.4 Ubuntu LTS 19:02 < ayecee> 12.04 maybe 19:03 < Psi-Jack> Just what's been shown above is enough to make me gag of bad bash. And I'm tough skinned. 19:03 < ayecee> t5u: why would you freshly install a distribution that is so old that it's no longer supported? 19:03 < hexnewbie> I was freaked out by seeing [ in /bin probably the moment I opened kfm in KDE and started browsing around wondering ‘what's Linux’ (though maybe it was the gnome file manager, so even earlier). 19:03 < uplime> why would it freak you out? 19:03 < t5u> ayecee: why do you ask even 19:03 < ayecee> t5u: support for 12.04 stopped a year ago 19:03 < uplime> its just the test command 19:04 < prussian> how testy 19:04 < t5u> sorry 14.4 19:04 < Psi-Jack> .04 19:04 < ayecee> same question 19:04 < hexnewbie> uplime: Well, I was a Windows newcomer, not knowing bash syntax or what test is, expecting that [ is invalid in filenames (LOL), and confident that's a very bad filename, let alone program. 19:04 < jim> yeah, [ foo ] calls test foo 19:04 < t5u> just downgraded from 16.4 19:04 < Psi-Jack> .04 19:05 < uplime> the only thing that can't be in filenames is NULs and /'s 19:05 < prussian> which is a tragedy 19:05 < prussian> though technically fs specific also... but meh 19:05 < ayecee> you're a tragedy 19:05 < Azrael_-> is it possible to find out in what kernels this bug is already fixed? https://bugzilla.redhat.com/show_bug.cgi?id=1436866 or what kernels are before this bug was introduced? 19:05 < Psi-Jack> t5u: there is no 40th month 19:05 < ayecee> Azrael_-: not easily, no 19:06 < Azrael_-> that's just crap. currently i'm testing out tons of versions and hope i find one without this bug. but this is just tedious and cumbersome.... 19:06 < hexnewbie> Azrael_-: Not difficult, either. Find one of the commits, find a version in which it was committed, then find all versions released around that time and go up and down the change log. 19:06 < t5u> Breaks existing package 'xserver-xorg-core-lts-xenial' conflict: fglrx-core () 19:06 < hexnewbie> Azrael_-: If you're only interested in one version of the kernel (4.8) it ought to be pretty easy 19:06 < ayecee> )`: 19:06 < plexigras> how can i go through a list and run a command over the entrys until it command doesnt fail? 19:07 < emberquill1> t5u: isn't xenial 16.04? 19:07 < Psi-Jack> t5u: Why would you... Downgrade? 19:07 < ayecee> t5u: if you're using backports packages, xenial's xserver no longer works with fglrx 19:07 < Azrael_-> hexnewbie: no, i'm not really interested in this kernel. i only found this bug in the kali-releases (debian kernels) and the only related bug-report i found was this one from redhat :( 19:07 < ayecee> t5u: fglrx was retired around 15.10 time 19:08 < t5u> im guessing that 14.4 has some ports ==> 16.4 19:08 < ayecee> you're using the hardware enablement stuff, aren't you 19:08 < Psi-Jack> .04 and .04 19:08 < t5u> Psi-Jack: aye sure 19:08 < Psi-Jack> t5u: Corrections. Use them 19:09 < t5u> Psi-Jack: thanks 19:09 < hexnewbie> Azrael_-: Given the strange reason for bug closure, I would manually install 4.16.5 and see if the bug s present. If not present, all is fine. If it exist, I would install Fedora 28 just so I can reproduce and reopen the bug in the RedHat tracker :) 19:10 < Azrael_-> hexnewbie: godlike. until now i just only wanted to use a simple nfc card reader with linux and didn't want to go down the whole road. also tried using kali as everything should be already set up there without me having to fiddle out all the dependencies :) 19:10 < ice9> which vpn clients that supports IKEv2? 19:10 < Psi-Jack> Ipsec ones 19:11 < Psi-Jack> I hate IKEv2. And ipsec. Horrible stuff h4x0r3d together. 19:11 < hexnewbie> Azrael_-: Oh, I wouldn't do most of that to get a reader working (I'd buy a different one). But if it's just so I can piss on the people at bug tracker, I would :) 19:12 < t5u> yes why 19:12 < ayecee> ice9: most likely strongswan on linux 19:12 < Azrael_-> hexnewbie: feel free to test it yourself. but i bought this readeer because it was recommended by a person i trust. but because of me being a newbie-status in this segment i don't want to run asking all the time what setup he uses and/or how i can avoid this bug 19:13 < Psi-Jack> strongswan and Libre swan 19:13 < Azrael_-> and a new one would cost again and take more time to deliver 19:13 < Psi-Jack> The only two still supported and developed. 19:15 < awy_> hi whenever we write data , i was told it goes to memcache -- file system -- disk cache -- disk? 19:15 < awy_> is the order right? 19:15 < ayecee> awy_: no 19:16 < awy_> aah 19:16 < ayecee> or maybe yes 19:16 < busybox42> No 19:16 < awy_> ayecee: what do you think is the correct order 19:16 < ayecee> memcache is ambiguous 19:16 < Psi-Jack> awy_: what? 19:16 < Psi-Jack> memcache doesn't use disk.. 19:17 < awy_> what is mem cache?? some portion of memory in the physical ram? 19:17 < Psi-Jack> It's a memory based keystore cache. 19:17 < Psi-Jack> Server 19:17 < ayecee> not in this context 19:17 < ayecee> awy_: that's something you'll have to ask the person who told you this, because it doesn't have a clear meaning. 19:18 < ayecee> however, one normally reads from caches and writes to buffers. 19:18 < hexnewbie> The only thing I know going by the name of ‘mem cache’ is an obsolete PHP module whose functioning is unclear to me 19:19 < ayecee> hexnewbie: so it's almost certainly not a literal name 19:19 < ayecee> smart, eh 19:19 < Psi-Jack> hexnewbie: obsolete? Naw. 19:19 < hexnewbie> Psi-Jack: Hasn't it been replaced by memcached? 19:20 < Psi-Jack> hexnewbie: Yoooooou have been replaced. :) 19:21 < jml2> Azrael_-, ah no.. a kali user!! 19:21 < jml2> Azrael_-, you're a hacker!! 19:22 < Psi-Jack> Script kiddie more like. :p 19:23 < ayecee> por que no los dos 19:23 < DrunkRhino> Alright, someone tell me if I'm nuts. Using paprefs and enabling the options under the "access" tab, plus the options under "server" on the machine connected to the input device should be all I need to use Audacity on my tower and the turntable hooked up to the pi, right? 19:23 < ayecee> you're nuts 19:23 < Psi-Jack> I agree 19:24 < ayecee> we've solved another one 19:24 < DrunkRhino> Ok, fair :p 19:24 < DrunkRhino> I walked into that one. 19:24 < Psi-Jack> Face first 19:26 < hexnewbie> Is it for turntabling (?) or listening? If former, it's the good kind of nuts. If the latter, well... There's no bad kind of nuts really, but it's scary :) 19:27 < ayecee> there are definitely bad kind of nuts 19:27 < ayecee> though i understand what you're going for there, trying to be inclusive 19:28 < DrunkRhino> Just conversion, nothing real-time really. Running "ssh -Y pi audacity" and recording that way works fine, but the UI is just a bit slow being run on a remote machine + the wireless connection. 19:29 < Psi-Jack> Why even use audacity at all? 19:30 < Psi-Jack> ffmpeg or even arecord 19:30 < DrunkRhino> Psi-Jack, it's the tool I'm most familiar with, plus the tracks need to be all split up after I'm done. 19:30 < Psi-Jack> Then even better. Output to - and totally don't even try to stream through loops. 19:32 < jml2> DrunkRhino, i use rtp for doing simulatenously playback but its limitted to 16-bit (pa) 19:32 < DrunkRhino> Alright, I'll have a look at the manpages for those then. 19:32 < jml2> DrunkRhino, when you say "server" I'm assuming two separate machines over the network. :p 19:33 < DrunkRhino> jml2, correct, using that since paprefs also uses "server" in the options tab there. 19:33 < jml2> DrunkRhino, you just want playback? there's rtp module... 19:33 < Psi-Jack> ffmpeg, output to -, redirect to local file on local origin side of the ssh. 19:33 < jml2> DrunkRhino, but it's lower quality and is restricted to 16-bit... 19:34 < Psi-Jack> Edit and cut appropriately after it's digitized. 19:34 < Psi-Jack> Not during. :) 19:34 < DrunkRhino> Psi-Jack, that's what I assumed, but I didn't want to go bashing in the command with zsh's autocomplete. 19:34 < jml2> DrunkRhino, but the thing is you're getting I think not so bad real-time playback... 19:34 < Psi-Jack> So, don't use autocomplete. :p 19:35 < DrunkRhino> Psi-Jack, I meant more in the "flying blind" sense rather than specifically the tab-completion :p 19:35 < awy_> how can we measure maximum tps a block device can offer 19:35 < Psi-Jack> Flying blind is bad. Don't do that. 19:36 < ayecee> awy_: bonnie++ is a stress tester that reports iops. maybe use that. 19:36 < Psi-Jack> tps? 19:36 < ayecee> probably iops 19:36 < ayecee> tps would be for databases 19:37 < ayecee> also for reports that your boss needs 19:37 < Psi-Jack> Exactly. heh 19:37 < Psi-Jack> Today! 19:37 < fendur> I can make you a copy of the memo. I'll make you a copy. 19:37 < ayecee> i'm going to need you to go ahead and come in on the weekend. 19:37 < Psi-Jack> Or you'll have to work over the weekend. 19:37 < Psi-Jack> heh 19:38 < Psi-Jack> ayecee: I mean, that'll only take 5 minutes, right? ;) 19:38 < ayecee> :) 19:40 < awy_> is it good practice to clear buffer cache ? using sync; echo 1 > /proc/sys/vm/drop_caches 19:41 < Psi-Jack> Not really.. 19:41 < awy_> does the above command not clear the buffer cache? 19:41 < ayecee> that is what it does 19:41 < Psi-Jack> Sure, it does. But in general, just doing that for no viablely specific intentional reason blindly, is not good practices. 19:42 < ayecee> "is it good practice to turn left?" 19:42 < Psi-Jack> I turn left almost exclusively. 19:43 < ayecee> you know what they say, three lefts make a right 19:43 < prussian> what if I turn right to turn left? 19:43 < Psi-Jack> Left! 19:43 < awy_> ayecee: Psi-Jack do we generally clear buffer cache to speed up the system or performance? 19:43 < Psi-Jack> NO! 19:43 < ayecee> awy_: pretty much the opposite 19:43 < medfly> Hi. I want to pci-stub one of my pci devices and then pass it through to a kvm guest. having trouble blacklisting it. any tips? 19:43 < ayecee> awy_: the buffer cache exists to speed up the system or performance 19:43 < jim> prussian, you can do it three times! 19:44 < medfly> This is on some shiny ubuntu (4.13), but I only installed linux to do this, I'd switch to anything else if it works. 19:44 < awy_> ayecee: sure 19:44 < Psi-Jack> medfly: No such thing as 4.13. 19:44 < ayecee> yes there is 19:44 < medfly> 4.13.0-39-generic is what it calls it 19:44 < ziggylazer> hey jim 19:44 < Psi-Jack> That's a kernel version, not a distro version. 19:45 < ayecee> didn't say it was a distro version 19:45 < Psi-Jack> It was right next to the distro name. :) 19:45 < medfly> well, linux 4.13. 19:45 < Psi-Jack> See, now that's correct. :) 19:45 < ayecee> i can see why you're confused, lacking the ability to interpret context 19:45 < Psi-Jack> No context was supplied. :) 19:45 < ayecee> see, that's what i mean 19:46 < irgendwer4711> hi, I got trouble with ecryptfs: "Error decrypting page" how to find the file, which is causing that? 19:46 < Psi-Jack> Interpretation is prone to failure anyway. Just look at all the vulnerabilities caused by interpreters. 19:46 < Psi-Jack> irgendwer4711: You're using ecryptfs instead of LUKS like you should be. 19:47 < irgendwer4711> Psi-Jack: no! 19:47 < Psi-Jack> Absolutely! 19:47 < medfly> irgendwer4711: you might be able to find the faulting thing with strace 19:47 < ayecee> Psi-Jack: stop with the elitism plskthx 19:48 < irgendwer4711> medfly: which string? as I rememder, there is one empty crypted file cause the problem, because disk was full. 19:48 < Psi-Jack> It's not elitism. ecryptfs is pretty.... unreliable, insecure, and generally wouldn't even recommend it to even my worst enemy. 19:48 * medfly less at home with strace and doesn't know how to invoke it 19:48 < ayecee> yup, that's elitism 19:48 < Psi-Jack> You should refresh on your understanding of the word, then. 19:49 < ayecee> well, one of us should 19:49 < awy_> ayecee: also clearing the swap space once in a while is a good idea? 19:49 < ayecee> awy_: no 19:49 < irgendwer4711> Psi-Jack: in future I would like to use ext4 encryption 19:49 < ayecee> awy_: there is nothing about that idea that is good. 19:49 < Psi-Jack> irgendwer4711: You mean LUKS? 19:49 < irgendwer4711> Psi-Jack: stop trolling! 19:49 < Psi-Jack> I'm not trolling. 19:49 < dgurney> ext4 does have inbuilt encryption, but I'm not sure how it compares to LUKS 19:50 < ayecee> it's not trolling if you honestly believe what you're saying. 19:50 < ayecee> still pretty annoying. 19:50 < ayecee> Psi-Jack: would condescending be a better term to use? 19:50 < Psi-Jack> On, ext4 does actually support encryption? Wow.. 19:51 < Psi-Jack> ayecee: Strongly recommending maybe. 19:51 < jim> irgendwer4711, lucs is one form of encryption. psi just asked yuou if that[s what you meant 19:51 < irgendwer4711> I didnt talk about luks 19:51 < ayecee> jim: no, Psi-Jack insisted that that's what he should use 19:52 < ayecee> repeatedly and condescendingly 19:52 < Psi-Jack> There was no insisting. 19:52 < jim> let's just drop all this argument part 19:52 < Psi-Jack> There was no condescending, either. 19:53 < jim> one thing's for sure it generated commentary and some heat 19:53 < ayecee> pulling a trump, eh. keep repeating it until it's true. 19:54 < Psi-Jack> Speaking of trump.... I've been seeing lots of interesting suspect traffic from trump tower recently. 19:54 < DrunkRhino> Speaking of encryption, what would be the best way to go about encrypting a couple of installations that are on systems old enough that I'm stuck with MBR? 19:54 < DrunkRhino> Ooh do tell. 19:54 < Psi-Jack> Failed, repeat logins, triggering off reactions from monitoring that blocks it, from trump tower. heh 19:55 < guideX> I was wondering if linux people have the poor opinion of .net core 19:55 < guideX> like they seem to with mono 19:55 < Psi-Jack> wut? 19:55 < medfly> I reckon people don't really think of .net as a choice because they are used to other things. market share 19:55 < dgurney> some linux people equate anything Microsoft as satanic, so yes 19:55 < medfly> I don't even know much about .net 19:55 < dgurney> but generally, I haven't seen hate towards .net core 19:56 < djph> dgurney: or just sub-par 19:56 < medfly> I don't windows, so I don't really know what it is, and not knowing what it is - I wouldn't think to se it 19:56 < djph> DrunkRhino: if you want full-disk encryption, LUKS? 19:56 < lukey_> guideX: I only know it as C# 19:57 < Psi-Jack> DrunkRhino: LUKS. Backup and restore. (borgbackup I can recommend for the backups) 19:57 < guideX> lukey_: well there's mono (community c#) and .net core (microsofts c# version) 19:57 < medfly> hurray, I succeeded in stubbing the device I wanted 19:57 < srukle> lukey_: I know C# developers who only see it as C#. 19:57 < guideX> lukey_: mono seems to have this horrible reputation 19:57 < Psi-Jack> medfly: Did you use soap? ;) 19:57 < anonnumberanon> I'm finding it particularly difficult to do something does anyone can help? run a bash script which ssh to a machine and does not require any user input, after it has a shell on the other machine, it installs software with elevated privileges. ?? 19:57 < guideX> (unde4r linux) 19:57 < Psi-Jack> Oh, stubbing, not scrubbing. L:) 19:57 < medfly> I had a typo in one of my files, forgot to update-initramfs one time, and also addded a blacklist for the original driver 19:57 < lukey_> Psi-Jack: or lvmove :) 19:58 < medfly> I'm gonna do driver development in style, pass the same machine's device to a vm :-) :-) 19:58 < lukey_> Psi-Jack: Err... pvmove of cource 19:58 < Psi-Jack> lukey_: Very specific use case. 19:58 < djph> Psi-Jack: have you tried backblaze? 19:58 < DrunkRhino> Re: Microsoft, I remembered this, I'll leave it here. "I love the way Microsoft follows standards. In much the same manner that fish follow migrating caribou" 19:59 < medfly> fwiw I like how friendly microsoft is with regards to open sourcing .net core 19:59 < medfly> my friend ported it to netbsd and he seemed to have a much better time than with other large companies (*cough* google) 19:59 < Psi-Jack> djph: I use b3 19:59 < Psi-Jack> b2 19:59 < DrunkRhino> They do seem to be getting better these days. 19:59 < medfly> they had a chat to discuss development an were very responsive at surprising hours 19:59 < Psi-Jack> djph: My borgbackup backups are copied to b2. 20:00 < djph> Psi-Jack: ahhh, right 20:00 < djph> Psi-Jack: you were telling me to check out b2 earlier this month 20:00 < gronke> I'm trying to install an R package as part of a Galaxy toolset, and I'm getting this error: "ERROR: dependency ‘gdata’ is not available for package ‘gplots’". I'm on CentOS 7. I've installed a bunch of gdata packages but none seem to help. 20:01 < Psi-Jack> djph: It's possible I might've, yes. 20:03 < Psi-Jack> Hmmmm... Emby actually has backup & restore? Why doesn't Plex have that after all these years?! 20:04 < rypervenche> jml2: Yes, it works fine and leaves the port opened for 5 seconds. I was going to do UDP packets first, but that requires ICMP and thus root, so I went with TCP in the end. The ports never look opened from the scanner's point of view and one of my ports is higher and one lower than the first, so a serial scanner would not hit all of them. And I have more 3, this was just a dumbed-down version. 20:04 < djph> Psi-Jack: because fuck you that's why 20:05 < Psi-Jack> Basically.... And I'm a Plex Pass lifetime member. :/ 20:05 < Vapez> I have an issue with attaching a Volume from another EC2 into another EC2, it auto mount it 20:05 < Vapez> why !? 20:05 < djph> aaw 20:05 < uplime> Vapez: perhaps give some more information 20:05 < Vapez> it's about aws 20:05 < uplime> yes im aware 20:05 < Vapez> but the guys from aws doesn't answer 20:05 < jml2> rypervenche, yeah using hping requires a netcap capability (raw socket) 20:05 < Psi-Jack> Vapez: ##aws 20:05 < Vapez> so, what do you want to know 20:05 < Vapez> they won't answer 20:06 < uplime> yes they will 20:06 < Psi-Jack> Well, it's off topic here. And yes, they will. 20:06 < Vapez> i just asked 20:06 < uplime> you asked a terrible question 3 minutes ago. be patient 20:06 < rypervenche> jml2: Yeah, same with nmap because of the ICMP. Using TCP fixed that though and can run as normal user. 20:06 < uplime> irc != instant communication 20:06 < Psi-Jack> I mean you LITERALLY just only asked 3 minutes ago. 20:06 < jml2> rypervenche, if the scanner is fast enough (like 1000's / sec), it can bypass it within 5 seconds... 20:06 < Vapez> I don't have autofs so it shouldn't be auto mount 20:06 < Vapez> i don't understand, why it's mounted 20:06 < jml2> rypervenche, so that is why I have removals when an incorrect port is entered (for those lines without -j ) 20:06 < uplime> Vapez: wait for the aws people to respond 20:06 < gronke> Does anyone have any idea about that gdata error message? It must be a dependency that I'm missing from somewhere. 20:07 < DrunkRhino> Any ideas why UFW shows all my other rules by their appropriate names, but instead of SSH it just shows 22 instead? 20:07 < jml2> rypervenche, (I mean when an incorrect port is knocked) 20:07 < Vapez> is there any order how the devices are mounted in linux? 20:07 < rypervenche> jml2: Ahh ok, I can set that up easily then. I'll add it. Thanks for the tip. 20:07 < jml2> rypervenche, when the portknock is wrong, the portknock gets reset... 20:07 < uplime> DrunkRhino: was that rule added to ufw by name or by port 22? 20:07 < Psi-Jack> Vapez: ##aws 20:07 < Vapez> is there any difference if i set a device to /dev/xvdb instead of /dev/sdc ? 20:07 < jml2> rypervenche, getting the portknock reset is imho a good way to make sure that scanners can't beat it.. 20:08 < rypervenche> jml2: Aye. I agree. 20:08 < ayecee> Vapez: they're different devices, but they're both block devices. 20:08 < DrunkRhino> uplime, I'm sure I've run "ufw delete allow to any port 22 && ufw allow ssh" at least a couple times by now 20:09 < DrunkRhino> Wait, should I be sticking a "reload" in there? 20:09 < DrunkRhino> *in between 20:13 < emberquill> DrunkRhino: yeah, ufw reload whenever you modify the rules. 20:13 < emberquill> Also for mine I think I had to explicitly use "openssh" instead of "ssh" 20:14 < uplime> it probably depends on /etc/services 20:14 < uplime> or whatever that file is called 20:16 < DrunkRhino> emberquill, I'll give that a shot. I just didn't want to make sure I didn't keep ssh disabled, it being a pi and all. 20:16 < uplime> spin up a temporary ssh on another port? 20:17 < DrunkRhino> uplime, that could work. 20:19 < awy_> what is noop deadline cfq 20:20 < awy_> are those the types of scheduler? also is it io scheduler or a task scheduler? 20:20 < awy_> i catted scheduler file in /sys/block/sda/queue 20:20 < KeithWeisshar> Why does osdisc.com charge more for linux on usb flash drive than the CD/DVD? 20:20 < Psi-Jack> awy_: Poking randomly? 20:20 < Psi-Jack> KeithWeisshar: USB is more expensive media. 20:21 < Psi-Jack> D'uh... 20:21 < awy_> Psi-Jack: not really. trying to understand the depth of linux :) 20:21 < Psi-Jack> awy_: Those are I/O Schedulers. 20:21 < DrunkRhino> uplime, emberquill, nope, still no dice, it still displays as "22" instead of "SSH" 20:21 < Psi-Jack> You can google the rest. 20:21 < uplime> DrunkRhino: sorry not sure then 20:22 < uplime> I really only use ufw to setup some initial rules then be done with it 20:22 < uplime> #ufw exists htough, if im not mistaken 20:22 < solidfox> what cpu do you guys use or want to switch to 20:22 < solidfox> I'm still on intel on my laptop 20:22 < awy_> Psi-Jack: could you tell me which file will list the task scheduler? 20:22 < KeithWeisshar> 16gb from osdisc is $14.95, 32gb is $24.95, and 64gb is $34.95, cd is $2.95 and dvd is $5.95 20:23 < DrunkRhino> uplime, ah well, thanks anyway! 20:23 < djph> solidfox: "intel" 20:24 < Psi-Jack> KeithWeisshar: I see you have eyeballs that can see. 20:24 < Psi-Jack> And intelligence enough to read and write. :) 20:24 < KeithWeisshar> those are all the osdisc.com prices 20:24 < Psi-Jack> Are you advertising for them? 20:25 < KeithWeisshar> no 20:25 < Psi-Jack> Okay. Do you have a Linux question not pertaining to osdisc.com? 20:25 < jml2> osdisc prices lol 20:25 < KeithWeisshar> ubuntu 18.04 not released yet 20:26 < jml2> a price of those disks is a quarter charge of my internet! 20:26 < triceratux> well after dinnertime on the isle of man 20:27 < djph> jml2: you pay $12 for internet? 20:27 < solidfox> djph, ? 20:27 < djph> solidfox: hmm? 20:27 < solidfox> djph, "intel" 20:27 < cmpxchg8b> my internet bill is 8 euros 20:27 < solidfox> what do you mean 20:27 < solidfox> is it a fake intel 20:27 < djph> solidfox: you asked what CPU mfg. "Intel" 20:28 < solidfox> djph, oh you were answering my question? 20:28 < solidfox> I see 20:28 < djph> solidfox: indeed 20:28 < solidfox> the quotes threw me off 20:28 < solidfox> djph, sometimes quotes mean sarcasm 20:29 < jml2> discovered that stupid wordpress requires the loopback to work properly.. had to run a dumb health check plugin tehehe 20:29 < djph> yeah, pebcak, typing strings out in the other screen :| 20:29 < jml2> lame! 20:36 < simbalion`> AppImage files should be named .ai, right? 20:36 < uplime> simbalion`: depends on the software reading them I guess 20:36 < simbalion`> I mean I know they're named .AppImage typically but that's annoying to type 20:36 < uplime> extensions don't really mean anything in linux 20:36 < simbalion`> uplime: well they're executables 20:37 < uplime> simbalion`: executables don't need any extensions 20:37 < simbalion`> uplime: no I get that, however if AppImage is going to be the standard for cross-distro executables then a short catchy extension is a good idea 20:37 < simbalion`> and .ai just seems awesome to me 20:37 < uplime> we'll have to disagree on what a good idea is then 20:38 < simbalion`> Heh, nothing wrong with that, everyone is entitled to be wrong ;) 20:39 < simbalion`> Tho I get AppImage is still really new and maybe something will replace it like flatpak but from a user's perspective AppImage is the 'friendliest' approach I've seen 20:39 < simbalion`> and being able to run portably or install is awesome, frankly. 20:39 < Psi-Jack> Hmmm 20:41 < simbalion`> Not having extensions on executable files is fine when yer a nerd like us but as 'average' users start adopting linux they're going to demand visual ways to represent executables similar to what they're used to.. that doesn't have to be an extension, but extensions are useful for that. 20:41 < simbalion`> I usually never both with extensions myself.. I might sometimes use a .sh or .py but mostly I hate typing periods. 20:41 < simbalion`> i usually never bother* 20:42 < Psi-Jack> You just hit . 5 times. 20:42 < simbalion`> lol 20:42 < Psi-Jack> One time, twice in a row even. :) 20:42 < simbalion`> I said I don't like typing periods, I didn't say I'm willing to sacrifice quality of communication over it. 20:43 < triceratux> simbalion`: users are always free to define mimetypes or rely on the file command 20:43 < simbalion`> and of course most commands are tab-completed these days anyway 20:43 < Psi-Jack> Oh thank the gods for that! :) 20:43 < simbalion`> triceratux: Sure there's nothing stopping me from renaming my AppImage files to .ai, but I am sharing the idea here because I think it should catch on :) 20:46 < awy_> what info does /var/log/secure file has 20:47 < ananke> awy_: look inside 20:47 < simbalion`> I think that's authentication attempts in redhat systems? The equivalent of /var/log/auth on Debian 20:47 < simbalion`> awy_: 20:47 < awy_> thanks simbalion` 20:48 < simbalion`> awy_: happy to help, you should check to be certain cause I never use redhat 20:55 < Psi-Jack> bo6: Consider that when you change nicks many times in a large channel, repeatedly, over the course of even just 1 hour, 8 times, is very noisy. 20:55 < Psi-Jack> Just sayin. :) 20:57 < justsomeguy> Would someone be so kind as to give me a short high level overview of how updates work in fedora atomic workstation? 20:57 < awy_> what does /etc/securetty has.. commenting one of those tty,would prevent us to do what? 20:57 < mawk> aaa 20:57 < meyou> nothing is noisier than Psi-Jack though 20:58 < jim> awy_, sec 20:59 < ananke> awy_: man securetty 20:59 < Psi-Jack> jim: ^ Looks like he's on the road to doing it again. FYI. 20:59 < antichri|st> Psi-Jack: consier your opinion worthless. 20:59 < antichri|st> consider*** 20:59 < mawk> that file is still used awy_ ? 20:59 < jim> awy_, those are the ttys that can log in as root 21:00 < jim> I guess you can put it into a pam thing 21:00 < awy_> jim: when i do su -, it doesnt let me login as root 21:00 < mawk> on which tty awy_ ? 21:00 < awy_> however did not have any problem when i used sudo su - or sudo -i 21:00 < mawk> it's a list for login, not for su, as far as I know 21:01 < ananke> awy_: su requires root password. 21:01 < jim> so sudo doesn';t use that mechanism 21:01 < awy_> ananke: if i dont know the root passowrd,how do i proceed 21:01 < mawk> you don't 21:01 < ananke> awy_: you don't. you can use sudo -i, then change root's password. 21:02 < jim> you are probably in the /etc/sudoers file 21:07 < Celmor> are devices starting with '/dev/xvd' a thing? often seeing it in guides 21:07 < Celmor> or is it just so the user doesn't do something stupid if he copies the command exactly 21:08 < heiler> Celmor xen uses this I think 21:08 < bls> Celmor: it's a specific type of emulated disk 21:08 < Celmor> ah 21:08 < heiler> instead of /dev/sda 21:08 < ananke> Celmor: it's xen 21:08 < ananke> doh, I see somebody already mentioned it 21:09 < ananke> Celmor: one shouldn't be doing copy/paste of devices off random guides anyway 21:09 < Psi-Jack> Hmm, I forget, what does kvm use when you're using virtio, but not virtio-scsi? 21:09 < bls> so it's likely either a poorly written guide by someone that doesn't know they should mention the difference, or it's for a specific provider or for a specific xen setup 21:09 < Celmor> ik 21:09 < Psi-Jack> What's icky? 21:10 < Celmor> bls, it's used here https://docs.docker.com/storage/storagedriver/zfs-driver/#manage-zfs 21:10 < hexnewbie> Psi-Jack: The simpler PCI virtio block devices? 21:11 < Psi-Jack> hexnewbie: The virtio bus devices, basically. 21:11 < heiler> when a normal user logs in via telnet and a interactive program is auto-started via ".bash_profile", shouldn't this program be killed when the user logs out? 21:11 < Celmor> Psilocyber, I had /dev/vda 21:11 < Psi-Jack> Ahh, vdXY 21:11 < hexnewbie> /dev/vda, yeah 21:11 < Psi-Jack> xvd is Xen, vd is KVM. :) 21:12 < Celmor> makes sense 21:12 < Celmor> vd -> virtio disk, xvd -> xen virtio disk? 21:12 < bls> Celmor: ok, so they've assumed a lot about how you're running docker and failed to mention you might not have the exact same setup 21:13 < Celmor> well, the exact block devices used are always different, it's pretty clear that those are just examples 21:13 < hexnewbie> Celmor: xvd* is Xen *virtual* disk, virtio is the term for vd* 21:13 < Celmor> ok 21:13 < jml2> heiler, depending on your distro, sometimes the default i .profile 21:13 < jml2> heiler, and don't use telnet 21:14 < jml2> heiler, use "ssh" 21:14 < heiler> I'm troubleshooting an ancient system ;p 21:14 < jml2> heiler, dos? 21:15 < heiler> ahuahua 21:15 < heiler> natural/ADABAS database on a CentOS 21:15 < jml2> heiler, you're a badass I get it. 21:15 < mawk> what's the VNET feature for tun/tap ? 21:15 < jml2> heiler, I bet your database is just a flat-sheet csv file.. 21:16 < hexnewbie> heiler: CentOS should have SSH even in its oldest incarnations. I was shouted for using telnet in RedHat 6.0 days (also DOS days) 21:16 < mawk> and offload, and xfilter ? I can offload checksums to the kernel or something ? 21:16 < jml2> heiler, mwhahaha 21:16 < Psi-Jack> mawk: To hardware, actually. 21:16 < mawk> yeah for a generic offload feature 21:16 < mawk> but specifically for the offload feature of tun/tap devices, I don't know 21:17 < mawk> kernel gives us packets with the correct checksum already no ? 21:17 < heiler> well users must login via telnet, because the "natural" program uses VT100 with an ASCII screen and so 21:17 < Psi-Jack> checksum offloading is the effect of using the NIC's built-in hardware to handle that. 21:17 < Psi-Jack> If the NIC supports it, of course. 21:17 < mawk> yeah 21:17 < mawk> what's the point of having a checksum on ipv4 packets that I'm about to tunnel, actually ? 21:17 < mawk> there will be an outer ipv4 datagram with a real checksum 21:17 < heiler> but anyway, everything is working like this for years, the only problem is when some users disconnects 21:18 < mawk> no need for an inner checksum, I can set it to zero 21:18 < mawk> ah but then on desencapsulation the packet will be with no checksum, something bad will happen 21:18 < heiler> the "natural" program isn't killed and becomes in loop using 100% cpu 21:18 < mawk> that's a bad situation 21:20 < mawk> there are no docs on the subject 21:21 * aBound docking in the corner 21:22 * aBound shows full fledged barrier :P 21:22 < mawk> there is this in the header for tuntap: #define TUN_F_CSUM 0x01 /* You can hand me unchecksummed packets. */ 21:23 < mawk> but here "me" is my application, or the kernel ? 21:41 < jml2> heiler, what kind of bum company uses telnet? encapsulate all that 21:42 < Psi-Jack> jml2: Answer: Not one I would want to be responsible for. 21:42 < lrb> Is it possible to use less for the pager for info? 21:42 < jml2> lrb, it is when you install it.. by default it is usually the fallback 21:43 < jml2> lrb, otherwise use PAGER="less" in your bash startup script.. 21:43 < jml2> lrb, and man should automatically use it 21:43 < lrb> make's documentation is in info. I'm not worried about man. I'll give the variable a shot though. 21:44 < DLange> he's asking for info and that thing needs it's own tool as it has keybindings, hyperlinks etc. 21:44 < DLange> its, too 21:44 < jml2> lrb, if that doesn't do you should also be able to symlink ~/bin/pager to your preferred pager... 21:45 < jml2> lrb, system wide it can also be adjusted using update-alternatives (/etc/alternatives/pager) 21:45 < lrb> Ah the update-alternatives route will probably be the best option 21:45 < jml2> lrb, yep 21:50 * jml2 notices meyou talking about c:\ on #wordpress XD 21:50 < jml2> tehehe fake user ! 21:51 < Psi-Jack> Windows user detected. 21:51 < hexnewbie> I got C:\! 21:51 < g9G6g9gG96> when will linux support backslashes for directories? 21:52 < Psi-Jack> C< is better. 21:52 < Psi-Jack> g9G6g9gG96: Never. 21:52 < g9G6g9gG96> never say never! 21:52 < Psi-Jack> Never! 21:52 < g9G6g9gG96> linux-v2 then? 21:53 < melo3n> more like linux v2 when ;) 21:53 < melo3n> you can just use zsh or any other autocomplete to get over that problem though 21:54 < Psi-Jack> Well, Linux is already v4.16. :p 21:54 < revel> melo3n: v2 was ages ago. v5 may be coming soon. 21:54 < melo3n> lol i was just making a joke 21:54 < jml2> a lot of users think /etc/hosts is "dns"... this is a wild mis-conception 21:54 < revel> Linus apparently only postponed it to not be too predictable. 21:54 < melo3n> Linus, always giving us a tease 21:54 < revel> Since 4 was somewhere around 2 million git commits and now we're at 4 million commits or something like that? 21:55 < g9G6g9gG96> linux-v2.1.0 then? 21:56 < koala_man> g9G6g9gG96: https://mirrors.edge.kernel.org/pub/linux/kernel/v2.1/linux-2.1.0.tar.gz 21:56 < Psi-Jack> User error detected. 21:56 < koala_man> "30-Sep-1996" 21:56 < g9G6g9gG96> that's linux-2.1.0, not linux-v2.1.0 21:56 < koala_man> it's in the directory v2.1 though 21:56 < revel> What's the v for? 21:57 < koala_man> vindows path support 21:57 < Psi-Jack> The Vinal Battle. 21:57 < melo3n> vvindows* 21:57 < revel> The vinyl battle? 21:57 < melo3n> winyl battle* 21:57 < jml2> talking about the vin vin, that linspire distro came back recently... yeah that "lindows" company XD 21:58 < jml2> bet it has a tracker just like deepin XD 21:58 < revel> "7.0 SP1" 21:58 < revel> lol 21:58 < Psi-Jack> Heh. Apparently some people don't know about the old show "V" 21:58 < jml2> V was a movie... 21:59 < uplime> https://en.wikipedia.org/wiki/V_(TV_series) 21:59 < Psi-Jack> The original was a few movies, yes. 21:59 < Sitri> There was a movie called "V"? 21:59 < jml2> i cant stand movies anymore.. they suck.. for me its 15 minute video things on omeleto :) 21:59 < Psi-Jack> No, there were movies called V. 22:00 < jml2> we canadians are the bootleggers for movies.. you need a movie we just download it.. 22:01 < Psi-Jack> Ahem. 22:01 < jml2> americans produce the movies, and we just download it for free 22:01 < Psi-Jack> jml2: You're overreaching. 22:01 * jml2 no longer downloads movies XD 22:01 < RayTracer> g9G6g9gG96: linux supports backslashes for directories: mkdir 'c:\test' ;) 22:02 < Psi-Jack> RayTracer: Yep. Totally valid, for a directory name. heh 22:02 < g9G6g9gG96> c:\\test 22:03 < Psi-Jack> Single quotes makes it look better. 22:03 < jml2> double quotes if you're married 22:03 < RayTracer> heh 22:03 < Psi-Jack> Nope, double quotes would escape the t in test. 22:04 < Psi-Jack> Making you go back and correct it.. Okay, I see what you did there. :) 22:04 < jml2> "they" if you're an X person.. someone here in my country is sending letters to legally allow to have "they" on identity cards.. recent news I was reading. 22:05 * jml2 works on wp 22:15 < haps> re 22:15 < haps> I'm having some trouble with frequency information on a rhel 7.4 machine, and I wonder if it's because the cpufreq doesn't support my cpu 22:16 < haps> the symptom is /sys/devices/system/cpu/cpu/cpufreq doesn't exist 22:16 < haps> nor does cpupower do anything 22:16 < haps> the cpu is: Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz 22:16 < haps> is that something anyone here can help me figure out ? 22:16 < azarus> What do you expect it to show? 22:16 < haps> basically I was trying to figure out of there's a kernel module I'm missing, but it seems like that's baked into the kernel? 22:17 < autopsy> haps have you checked bugzilla.redhat.com for a bug yet on it? 22:17 < haps> I expect there to be the cpufreq folder 22:17 < haps> but there isn't one 22:17 < azarus> Maybe that CPU doesn't do dynamic frequency scaling 22:17 * azarus shrugs 22:17 < haps> so I assumed that it was a loadable module 22:17 < haps> autopsy: I'll check that out 22:17 < haps> azarus: even if it doesn't do scaling, it should report the frequency 22:18 < haps> ie scaling* may not be there, but cpuinfo_cur_freq should 22:18 < autopsy> haps https://bugzilla.redhat.com/buglist.cgi?quicksearch=Xeon%20cpufreq%20scaling 22:19 < jml2> haps, modprobe acpi_cpufreq , and then see if those paths exists (cat sys/devices/system/cpu/cpufreq/policy0/scaling_available_governors) 22:19 < autopsy> haps /proc/cpuinfo reports the type of processor and the MegaHertz. 22:19 < autopsy> jml2, oh I didn't know there was a module for it. 22:20 < azarus> Megahertz is a perfectly valid word, no weird capitalization required ;) 22:20 < jml2> haps, maybe it could be another module for your processor - ls -la /lib/modules/`uname -r`/kernel/drivers/cpufreq/ 22:20 < autopsy> Nevermind me in that case. 22:20 < haps> jml2: I have /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors on my laptop 22:20 < haps> unfortunately I can't modprobe that machine :-/ 22:20 < jml2> autopsy, helps also to know what the cpu model is, it can say on the intel site... 22:20 < haps> but I'll ask our 'guy' 22:21 < autopsy> jml2 He said it was a Xeon. 22:21 < haps> autopsy: re: that bugzilla, it's not quite my issue, since I don't see that folder 22:21 < autopsy> haps aww you can't do modprobe that means you're not root. 22:21 < haps> no root for me 22:22 < jml2> autopsy, he's the new ftp user, he needs to ask the senior admin 22:22 < haps> heh 22:22 < haps> nah, I'm a dev, devs don't get root here 22:22 < autopsy> haps no but I was giving you a link to the bugzilla search platform in case you didn't know it existed for future reference. 22:22 < autopsy> jml2, yeah figures. 22:22 < eto> hello 22:22 < haps> autopsy: yeah that's a great idea, thanks, I feeel like a dumas for not thinking of it first. 22:22 < autopsy> Maybe the admin doesn't need scaling. 22:23 < haps> okay, thanks all for your help, good starting points. 22:23 < eto> can somebody confirm that macvlan or mactap work with dummy? 22:23 < jml2> eto, it would definitely be useless 22:23 < autopsy> haps There is a lot you can do from existing bug reports. 22:23 < eto> jml2: can you go into why? 22:24 < autopsy> haps could be that you want a Request For Feature Enhancement known as an RFE you can file those in bugzilla too. 22:25 < haps> autopsy: that's what I was hoping to learn - if it's not supported, then I'd like it to be, but I imagine that intel would provide support right away, so it would just be a pull from upstream 22:25 < jml2> eto, you can create virtual bridge devices not connected to your physical ethernet 22:25 < haps> this is a pretty new cpu, so it wouldn't surprise me if there was some support missing. 22:25 < jml2> eto, then hook the macv* things to it 22:26 < jml2> eto, https://wiki.archlinux.org/index.php/Network_bridge 22:26 < eto> jml2: okay because i am trying to hook them macv* things with various "virtuals" (tap,dummy) and those don't work 22:27 < eto> jml2: so you say they should work when linked to `brctl `/`ip link add ... type bridge` 22:27 < eto> ? 22:28 < jml2> eto, http://backreference.org/2014/03/20/some-notes-on-macvlanmacvtap/ 22:29 < jml2> eto, if i'm correct the macvlan is oriented for host configuration (and not to be configured inside VM guests) 22:30 < jml2> eto, for configuring an ip on a virtual interface inside a VM guest, you would use macvtap. 22:31 < jml2> eto, for binding (like "virtual cabling") the macvtap, you of course first bind it to something on the native host 22:32 < jml2> eto, btw you should be acquainted into using the modern "ip" for bridge making, brctl is still around for legacy purposes.. 22:33 < eto> jml2: i am trying it with network namespaces and it seems unless i use veth + bridge the data cannot get accross 22:34 < eto> jml2: strangely ns macv* + physical nic seems to work 22:34 < jml2> (you can asign the end of hte veth pair interface an ip address.. ) 22:34 < haps> hi one more quick question: the intel_pstate driver that can be used, I'm assuming it's a kernel module 22:34 < eto> jml2: yes i just wanted to get rid of those veth "patch cables" 22:34 < jml2> eto, yeah you can have that for macvtaps.. that's what they're designed for.. 22:35 < haps> but I don't know what module that would be.... 22:35 < autopsy> haps you're assumption is a question of sorts? 22:35 < jml2> eto, you can also have a "passthrough" mode where the host cannot see the traffic, and the macvtap hogs all network traffic.. 22:36 < g9G6g9gG96> haps: it's a config option CONFIG_X86_INTEL_PSTATE but you can't build it as a module it seems. 22:37 < autopsy> haps there is this for cstate though: /usr/lib/modules/4.13.9-300.fc27.x86_64/kernel/arc 22:37 < autopsy> h/x86/events/intel/intel-cstate.ko.xz 22:37 < ArsenArsen> How does sudo keep track of the time limit per terminal 22:38 < jml2> ArsenArsen, what you smokin? 22:38 < Psi-Jack> Logic. 22:38 < jml2> ArsenArsen, man sudo will tell ya 22:38 < ArsenArsen> jml2, dunno man i just picked the first thing on the corner of the block 22:38 < autopsy> d 22:38 < haps> autopsy: looks like your bugzilla suggestion was a really good one, this is a very likely hit: 22:38 < haps> https://bugzilla.redhat.com/show_bug.cgi?id=1307185 22:39 < ArsenArsen> I meant how does sudo store the state? Is it somehow modifying the parent's env or? 22:39 < jml2> ^ wooo! 22:39 < jml2> a wa wa wooo!!! 22:39 < Psi-Jack> Nothing that would be controllable by the user, of course. 22:39 < jml2> sudo stores nukes in the United States!!! 22:40 * jml2 ducks! XD 22:40 < Sitri> Arcaelyx: nothing can change the parent's environment like that 22:40 < Psi-Jack> Safer here. :) 22:41 < ArsenArsen> wrong mention, but that aligns with what I thought, so what does sudo do? 22:41 < Arcaelyx> What 22:41 < Psi-Jack> sudo does sudo. 22:42 < Psi-Jack> But does no to Dallas. 22:42 < ArsenArsen> what 22:42 < ArsenArsen> lemme try again 22:42 < Psi-Jack> do* 22:42 < ArsenArsen> how does sudo cache credentials 22:42 < Psi-Jack> How's your C? 22:42 < bnason> My root filesystem is xfs and i need to reduce its size. Can I create a new LV, format it to be smaller (but enough to fit the contents of the old root) and copy everything over? How do I specify that this is now the new root? 22:43 < ArsenArsen> its okay, I guess 22:43 < ArsenArsen> RTFS I'd assu,e? 22:43 < jab416171> bnason, you tell the bootloader it's the new root 22:43 < ArsenArsen> assume* 22:43 < Psi-Jack> Well, read the sudo source code. :_) 22:43 < ArsenArsen> knew it :^) 22:43 < bnason> jab416171, that would be something in the grub config then? 22:43 < jab416171> if you're using grub, yes 22:43 < jab416171> don't forget to update fstab too 22:44 < Psi-Jack> or systemd.mounts. :) 22:44 < eto> jml2: thanks 22:44 < ananke> bnason: adjust your /etc/fstab. that's it. 22:44 < bnason> jab416171, well ideally id create the new lv, move everything over, delete old one and rename the new one to be 'root'. So I shouldn't have to do anything if grub and fstab reference the 'root' lv right? 22:44 < jml2> eto, you tank me I tank you! you're welcome! :p 22:45 < jab416171> bnason, usually fstab and/or your bootloader reference your disks by UUID 22:45 < ananke> bnason: is your /boot part of /, or is it a separate filesystem? 22:45 < jab416171> and/or partition number 22:45 < bnason> ananke, separate 22:45 < ananke> bnason: then your procedure should be ok, but you need to check how your / is referenced in /etc/fstab 22:45 < bnason> so ill have to boot into the old one, update grub to use the new uuid, reboot and then remove the old one 22:45 < TaZeR> i wanna roll back my system for the first time with timeshift but im scared my first resotration will result in a total destruction instead =D 22:45 < TaZeR> and ive been using the backup software for months without testing it yet lol 22:46 < TaZeR> what do you think i should do? 22:46 < bnason> fstab references the /dev/mapper/root 22:47 < Psi-Jack> Pfftt, who even uses fstab anymore? So old. hehe 22:47 < autopsy> Has anyone used cool-retro-term? 22:47 < Loshki> TaZeR: if I were you, I'd do a full backup using *something else". That way, you can still recover if your primary backup scheme isn't all it should be. If you aren't randomly retrieving files off your existing backups to verify they are working, you haven't really tested it at all, have you? 22:47 < triceratux> autopsy: yep its one of the buggiest apps ive ever run across. it finally started working on some distro tho 22:48 < Loshki> autopsy: plenty of uncool, non-retro terminal apps to choose from. 22:48 < autopsy> triceratux, I didn't notice any bugs what bugs did you encounter? 22:48 < g9G6g9gG96> moths! 22:49 < autopsy> Loshki, yeah but cool-retro-term is just cool. 22:49 < Psi-Jack> autopsy: I'm all about the future. Not the distant past. 22:49 < triceratux> autopsy: theres versions of it where the menus just didnt work so you could only use the one theme. its a cute effort ill admit 22:49 < Loshki> autopsy: apparently it's mis-named. It should have been buggy-cool-retro-term. I hate it when they get the names wrong. 22:49 < autopsy> triceratux, it's crazy someone would have thought to emulate a CRT in a window like that. 22:50 < TaZeR> Loshki: the problem is i lack the neccesary secondary media to store a full image backup my secondary 500gb hard drive is all i have to work with hense why i chose the rsync based solution 22:50 < autopsy> Loshki, moths they said. 22:51 < TaZeR> i was hoping to not have to roll back until i buy a new external drive for a second back up, maybe i will try to fix the problem instead of rolling back for now 22:52 * meyou notices people noticing me in other channels 22:53 < jab416171> meyou, we don't share any channels 22:53 < jab416171> lol 22:53 < jab416171> somehow 22:54 < meyou> i guess you have to be a masochist linux desktop user if you want street cred in ehre 22:54 < Loshki> You can't spend ##linux credits outside the channel... 22:57 < eto> jml2: got it working - both ends must be macvlans: macvlan1->dummy0<-macvlan2 or macvlan1->bridge0<-macvlan2 22:57 < eto> jml2: otherwise host doesn't see ether traffic 22:57 < Psi-Jack> Loshki: Huh? Then what have I been doing all these years? 22:57 < eto> jml2: thanks for hints though 22:58 < Loshki> Psi-Jack: you've found a way to exchange it for bitcoin? 23:00 < Psi-Jack> Yeah.. Made millions of dollars off everyone here, without them knowing. :) 23:00 < SuperSeriousCat> Now we know 23:01 < Psi-Jack> It's amazing how so few bytes of data can be so overly overpriced. 23:02 < jml2> eto, dumb you shouldn't be using the dummy interface 23:02 < Psi-Jack> Woops. :) 23:03 < jml2> eto, you should be able to assign ip addresses to the bridge device 23:04 < quint> Is this the correct order of these operation at boot? "crypttab, LVM, fstab" 23:04 < quint> operations** 23:10 < jml2> eto, lol you fixed it? don't use the "dummy" interfaces.. it's ancient and not to be used with these interfaces.. 23:10 < jml2> quint, no 23:11 < jml2> quint, grub does its own :).. and then there's the system's order :) 23:11 < triceratux> hrm the bionic beaver isos are appearing this very moment http://cdimage.ubuntu.com/ubuntu/releases/18.04/release/ 23:13 < jml2> eto, if you want to create a sort of "monitoring" on the traffic between vm's, then you should be taking a look elsewhere (openvswitch with a --mirror port) 23:14 < mawk> android uses dummy interfaces 23:14 < mawk> doesn't look that ancient 23:15 < jml2> mawk, shutup kid 23:15 * jml2 ignores mawk :) 23:15 < sponix> Ubuntu 18.04 LTS just dropped --> http://releases.ubuntu.com/18.04/ 23:16 < jim> this wouldn't be developing into some kind of argument? 23:16 < mawk> I don't know jim 23:16 < mawk> I just stated something and I got attacked 23:16 < jml2> jim, I ignore anyone who brings up #android 23:16 < mawk> that's the second time he said he's ignoring me, but apparently it's not the case 23:17 < jim> you do understand you're toing a backnforth right now, yes? 23:17 * g9G6g9gG96 prepares the snacks and refreshments 23:17 < Psi-Jack> What's wrong with #android? 23:18 < jim> I'd have to bring it up when suggesing it to someone who might benefit from it 23:18 < xamithan> has anyone upgraded from 16.04 yet sponix ? 23:18 < xamithan> I want to wait out the bugs 23:19 < Psi-Jack> jim: me too. :) 23:19 < qnord> i got a new laptop and it came with a german keyboard, i thought i could get used to it but i found it very annoying to program in, when i try changing the layout back to my old one (uk) with "setxkbmap uk" it tells me "Error loading new keyboard description" 23:21 < jim> mawk, just relax and breathe... that's how to recover from being triggered 23:21 < mawk> lol 23:22 < jim> it's true :P 23:25 < jim> mawk, you don't have to respond to being attacked... or even call it "being attacked" 23:25 < sponix> xamithan: If you want a smoother ride, 18.04.1 is likely a wise choice from my experiences 23:25 < xamithan> stability > smooth 23:26 < xamithan> Don't want to reinstall my work machine 23:26 < sponix> xamithan: I don't do upgrades anymore. Always a fresh install. And it is still downloading for me 23:27 < sponix> xamithan: I'm likely just gonna run it in a VM for a while. And wait to see what Linux Mint has to offer in addition. If they don't produce much, then I'll just do Ubuntu Mate 18.04 23:27 < mawk> I'm sure there's a way to test the upgrade without breaking your machine 23:27 < mawk> using a nice little overlay 23:27 < xamithan> I stopped recommending mint after their ISOs got hacked and they kept blocking patches from upstream 23:28 < Sonolin> lol wut 23:30 < sponix> xamithan: Those little issues ? When you aspire to be the next Microsoft, you can't help but take on some of their traits :P 23:30 < jml2> xamithan, happened once and the devs were the first to spot it -- they found out later who it was... 23:30 < jml2> xamithan, that happened like 5 years ago 23:30 < Ben64> "and they kept blocking patches from upstream" 23:31 < jml2> xamithan, https://haveibeenpwned.com/ << has the mint forum -- my old address used to show up when verifying with this site.. (its a legit site to test) 23:31 < xamithan> five years ago? Its only been two 23:32 < jml2> xamithan, i dunno, but i recall it happened and it was resolved very shortly. there was also the ubuntu forum hack. some ancient visual php plugin that wasn't updated.. 23:33 < jml2> xamithan, and countless heartbleed-like bugs as well :P update those kernels :p 23:33 < xamithan> Eh either way, I don't like how they block stuff so their custom tools doesn't break 23:33 < jml2> xamithan, it happened to ubuntu's community forums twice... 23:33 < jml2> xamithan, do you also recommend ubuntu? 23:34 < Ben64> forum hack isn't the same thing 23:34 < jml2> Ben64, perhaps yes or no, but i think there was an escalation for the mint-iso hack.. 23:34 < xamithan> This hacker had admin crendentials in addition to hacking a bug 23:34 < xamithan> He didn't even need to hack the website vuln, he had admin creds 23:35 < Ben64> i don't like mint's security team 23:35 < heiler> if a program was called inside a bash session, and that session doesn't exists anymore (for any reason), shouldn't this program be killed? how can I force that to happen? 23:36 < jml2> a ubuntu dev gave them a whack when he blogged about the mint security team 23:37 < jml2> heiler, < says he uses telnet and is asking a security question :) 23:37 < heiler> oh hell. I don't use telnet myself 23:37 < heiler> it's a legacy system 23:38 < koala_man> heiler: if you mean session in the setsid sense, then no, the session isn't gone until all the processes die 23:38 < jml2> xamithan, you should tell people to stop cheating on theri partners XDD remember the madison hack? lol 23:38 < jml2> xamithan, LOL 23:39 < heiler> koala_man the bash process isn't on PS anymore 23:39 < jml2> xamithan, people were abusing sites like haveibeenpowned to see who was a member of the dating site.. eventually that particular database was taken down.. 23:39 < heiler> but the program it started is running yet 23:40 < heiler> it seems it dropped from being a bash sub-process to a systemd sub-process 23:41 < jml2> systemd for an ancient system? 23:41 < jml2> LOL 23:41 < heiler> jml2 if you can't help just shut up 23:41 < jml2> heiler, you said it was for ubuntu 12.x ? 23:42 < jml2> heiler, you're assuming too much.. you're getting too ahead of yourself :) 23:42 < blaztek> heiler: how is the program executed from the bash script? 23:42 < heiler> the SO itself is a centos 7.2 but the program being executed is an old database system which probably you never heard of 23:43 < heiler> blaztek the user logins via telnet, telnet.in calls bash, .bash_profile calls the program 23:43 < heiler> it's perfectly shown on "ps auxw -H" 23:44 < blaztek> heiler: but in .bash_profile...does it have a & at the end of the program? 23:45 < heiler> for some reason the in.telnetd -> login -> bash isn't there anymore, but the program is still running, and using 100% cpu 23:45 < heiler> blaztek, no 23:45 < jml2> cant believe you're still on that 23:45 < blaztek> heiler: is it using exec? 23:46 < heiler> it's called just like "natural parm=SATPROD" at the end of .bash_profile 23:46 < jml2> blaztek, and what happens when the user tries to login twice? they probably annihiliate and create mass-data corruption for the so "database" XD 23:47 < jml2> cant tell if this is real or not. telnet and then an auto-script to do something with a "database". sounds archaic the least. 23:47 < heiler> "natural" is /opt/softwareag/Natural/bin/natural, a binary 23:47 < jml2> heiler, you're not for real are you? 23:48 < jml2> heiler, tell me this is for a self study and not for an actual company 23:48 < g9G6g9gG96> could be a university doing research on irc users 23:49 < heiler> it's the system that accounts all people taxes 23:49 < heiler> city hall company 23:50 < jml2> brazil 23:50 < heiler> anyway, the question remains simple.. the user shel itself isn't there anymore.. why the program which .bash_profile called is yet being executed? 23:51 < heiler> can it trap kernel signals and ignores SIGTERM or something? 23:53 < koala_man> heiler: the answer is equally simple. Processes aren't killed when the shell exits. 23:53 < koala_man> With a few exception, such as stopped and foreground processes receiving a sighup 23:54 < jml2> koala_man, at least he's working at the right place, he's gunna remove his taxes.. smart... 23:54 < jml2> lol 23:54 < heiler> if I login to a system and run a shell loop, will it run forever? 23:54 < koala_man> what? 23:54 < jml2> koala_man, he's working at a city tax center :) 23:54 < heiler> I'll ssh to your box and run "while true; do true; done" 23:54 < jml2> koala_man, he has access to the tax controls :) 23:54 < koala_man> who is? am I part of this discussion? 23:54 < heiler> will it be running forever? 23:54 < jml2> koala_man, (heiler) 23:55 < antichri|st> heiler: put & after it 23:55 < antichri|st> it will run forever after you log out 23:55 < koala_man> heiler: if the shell dies, then no. because that loop is run by the shell 23:55 < blaztek> heiler: it seems that the behavior is normal....I tried it...first.sh called second.sh, I killed first.sh and second.sh is still running...how to avoid that type of behavior...I'm not sure 23:55 < heiler> but there isn't a & after the program being called.. and it is running forever... 23:55 < jml2> koala_man, when he logs i n via telnet, telnet immediate kills itself while at the same time launching something for a database. 23:55 < antichri|st> if its parent process dies it will stop running 23:55 < jml2> koala_man, lol 23:55 < antichri|st> unless it is forked 23:56 < heiler> blaztek any way to prevent that from happening? 23:56 * jml2 thinks heiler is not real and is instead trying to "hack" too hard 23:56 < heiler> it doesn happen everytime, in this "system" 23:57 < heiler> sometimes the users just logs out and the program is killed 23:57 < heiler> but sometimes not 23:57 < jml2> and that's why starting apps from ~/.bash_profile is stupid.. 23:57 < koala_man> do you want it to be killed, or do you want it to not be killed? 23:57 < heiler> it enters in a loop using 100% cpu and we have to kill it manually 23:57 < heiler> to BE 23:58 < blaztek> heiler: I don't know 23:58 < koala_man> heiler: that happens to badly written programs that get confused when their terminal disappears. is this supposed to be a daemon? 23:58 < jml2> heiler, you don't even use Linux I bet, you're just here trying to fix a catch-22 problem.. 23:58 < heiler> koala_man it's an interactive program.. remember those ASCII menus and such ?? 23:59 < koala_man> heiler: then it's definitely a bug in that program 23:59 < antichri|st> curses? 23:59 < heiler> antichri|st before that ;p curses is too new 23:59 < g9G6g9gG96> put user in it's own pid namespace and when the user logs out, whatever spawns user shell enters namespace to slaughter every pid --- Log closed Fri Apr 27 00:00:17 2018