--- Log opened Wed May 30 00:00:37 2018 00:00 < djph> they're devices 00:02 < Dagmar> They have a function. They're where your various drives and mounts go 00:02 < brutser> djph: the problem is, livecd-creator creates 3 volumes to the actual root of the iso, 3! then some small loop devices that the user cannot mount anyway 00:03 < brutser> in the livecd they have really no function 00:03 < brutser> for the user i mean 00:03 < brutser> and they are captured under "Places" 00:03 < Dagmar> So complain to the makers 00:03 < brutser> Dagmar: no way for me to "just" hide them? 00:03 < djph> that's not how it works 00:04 < Dagmar> No, and you'll wish you had them the moment you actually *do* something with mount 00:04 < brutser> djph: well if i include gnome instead of mate, they won't show at all under Places - i just wonder why mate threats them differently 00:04 < storge> where is Mate showing them? icons on the desktop? list items in a gui? 00:04 < brutser> treats* 00:05 < djph> because it's not Gnome 00:05 < brutser> storge: under Places and when you mount them or try to mount them, they appear on the desktop 00:05 < djph> storge: probably down the left-hand side (quick-navigation thing) of caja 00:05 < brutser> djph: menu on top, applications - places - system 00:05 < storge> maybe there's a setting to omit unmounted devices 00:06 < brutser> storge: yes that is what im trying to figure out 00:06 < brutser> or can i hide them with udev rules? but that's a bit silly i guess 00:24 < jnt> when I am in grub shell, is there a command to tell if it's booted via bios or uefi? 00:27 < Dagmar> Look at the variables 00:30 < jnt> Dagmar: thanks, echo $grub_platform did the trick 00:37 < ayjay_t> is there a way to undo a setfacl command? 00:37 < ayjay_t> like restore to "default" 00:37 < ayjay_t> i'm playing with this command and i have just effed myself 00:42 < xamithan> You mean like restorecon ? 00:42 < xamithan> oh nvm wrong one 00:42 < xamithan> -b flag 00:43 < xamithan> Might also need the -n 00:47 < supernov3h> is there a way to prevent a user accessing certain syscalls 00:47 < ayecee> could you be more specific 00:47 < supernov3h> well not syscall itself, but system calls provided by glibc, example: prevent use of setuid, select, write etc 00:47 < ayecee> for what purpose 00:48 < supernov3h> making a user account very restricted 00:48 < ayecee> nothing easy comes to mind 00:49 < ayecee> it would be better to provide them with a limited shell rather than trying to limit the library routines they can use 00:49 < supernov3h> sure but if they can access their shell remotely, they can write to their own root directory (given that they own it) and therefore provide whatever binary they like 00:49 < supernov3h> unless their fs is mounted as ro 00:49 < bls> hmm, could LD_PRELOAD something to stub them out and force them into rbash? 00:50 < ayecee> a limited shell would prevent them from running that binary 00:51 < zachary12> sounds like openbsd's promise() 00:52 < zachary12> *pledge() 00:52 < supernov3h> can you make a use load a shell that is outside of their root directory? 00:52 < bls> but you're trying to behead a hydra. if you don't trust someone, either put them in a VM they can destroy or don't give them an account to begin with 00:52 < supernov3h> it's more or less a practice in understanding linux better 00:52 < ayecee> supernov3h: no 01:03 < handturkey> what's that resource display thing for the background? 01:05 < aclaivi> supernov3h: Is this what you're essentially referring to https://cisofy.com/controls/HRDN-7222/ 01:08 < handturkey> anyone messed with deepin? 01:10 < ayecee> one way to find out is to ask a question about deepin 01:11 < aclaivi> Acquired that guaranteed 'maybe' 01:20 < ayecee> and he was never heard from again 01:20 < pfred1> I think I'm going that way over here 01:22 < supernov3h> aclaivi: they don't have to compile it on the system, they can just scp it onto the server later if they have write permission 01:23 < supernov3h> I have a question though, can two consecutive fstab entries operate on the same mount? eg, a mount --bind, then a mount -o remount,x,y,z 01:24 < ayecee> it wouldn't make sense to use remount in an fstab entry 01:24 < supernov3h> or is there potential for some isochronous problems 01:24 < supernov3h> ayecee: indeed it would: --binds can't be mounted with permissions set 01:25 < supernov3h> prior to some particular kernel, 26 I think 01:25 < ayecee> however the option would apply any time you were interacting with the device. 01:26 < supernov3h> ayecee: I don't get what you mean 01:27 < ayecee> imagine the first mount failed. the second mount will still be attempted. 01:28 < ayecee> err, you're thinking of doing the same directory here, right? 01:28 < ayecee> in that case, only the first would be processed, unless maybe if you were using mount -a 01:30 < supernov3h> it would be like mount --bind /path/to/safe/dir /path/to/exposed/dir, mount -o remount,bind,ro /path/to/exposed/dir 01:34 < ayecee> i don't think it would work like that, but i haven't tested it. 01:36 < Juesto> remount opts only seem to work on something like Android 01:36 < supernov3h> well the stuff I'm finding through google seems hazy at best 01:36 < Juesto> rooted Android* 01:36 < supernov3h> oh I'm pretty confident the read-only remount works when supplied asynchronously via manually typing those commands 01:36 < ayecee> seems like something you might have to test 01:37 < supernov3h> problem is doing them at boot with fstab could invite all sorts of problems I couldn't be aware of 01:37 < ayecee> maybe see if it works at all before worrying about unforeseen problems 01:37 < supernov3h> yeah, tias as they've told me many times before - also turns out the fstab on this server is cloud manage anyway kek, might opt for a startup script instead 01:38 < graff> when you use a seerver supplied to you by a coporation, like a paid server 01:38 < graff> how do you handle making iptable rules for it? 01:39 < ayecee> same way you do for one sitting on your desk 01:39 < graff> so i just runt he software and it will work the same way? 01:39 < ayecee> yes 01:39 < graff> I thought it required some interaction with the ISP 01:39 < graff> hmm. alright 01:39 < ayecee> i can't see why it would 01:40 < graff> yeah I think you are right. maybe I am just overthinking it 01:40 < aclaivi> Just don't lock yourself out remotely 01:40 < graff> oh jesus :p 01:40 < graff> that would be bad 01:40 < aclaivi> Your supplier/host should still give you ways to access it if you do manage to do that 01:40 < graff> on our main server I have an instance of gitlab and 4 shell accounts 01:40 < graff> and it seems to be running fairly slowly, the gitlab seems fine 01:41 < graff> but the shell accounts seem slow 01:41 < aclaivi> That's odd, but not likely to be due to iptables 01:41 < aclaivi> If you're not too sure about using iptables, ufw provides a friendlier layer 01:42 < aclaivi> If it's just git hosting, realistically all you would need is SSH and HTTP(s) enabled on the firewall 01:46 < Mead> whoever suggested the plop boot tool, thanks, it is pretty useful. 01:51 < oiaohm> aclaivi: it is possible to run git hosting without ssh. 01:51 < ayecee> it's more difficult to manage git hosting without ssh though 01:53 < aclaivi> I can imagine there would be use cases for hosting without ssh, but ssh is quite practical in most senses 01:54 < oiaohm> ayecee: yes and no. https://www.tecmint.com/shell-in-a-box-a-web-based-ssh-terminal-to-access-remote-linux-servers/ Just because you have not opened up ssh to open firewall does not mean you cannot have the ssh interface in webbrowser. 01:56 < oiaohm> aclaivi: sometimes you have to be able to git using https only due to network issues. Like some server some where blocking port 22 for some reason. 01:56 < aclaivi> I don't disagree with you 01:56 < oiaohm> Management wise well setup there is not much difference at all. 01:57 < oiaohm> Of course poorly set up pure https git is a nightmare. 01:57 < ayecee> aside from it being more difficult to upload material to the server 01:58 < oiaohm> ayecee: again that is how good are the https interfaces. 01:58 < ayecee> fair enough 01:58 < oiaohm> ayecee: ssh is a quite decent interface that is fairly hard to screw up. 01:59 < oiaohm> ayecee: http on the other hand is really easy to make something totally not useable. 01:59 < ayecee> nod 01:59 < aclaivi> I feel like we all agree on the same points here 01:59 < ayecee> violent agreement 02:00 < aclaivi> That is, most notable protocols have use cases where they have to be rerouted through different means 02:00 * aclaivi nods angrily 02:03 < ayecee> https://i.imgur.com/YAGpXPd.png 02:03 < oiaohm> https://stackoverflow.com/questions/43069655/gitlab-push-using-https-specifying-username-password git push as long as you enter everything right works over https. gitlab on https only is not insanely horrible. 02:04 < aclaivi> More angry nods 02:05 < aclaivi> id 02:05 < aclaivi> whoops 02:05 < ayecee> EGO 02:05 < ayecee> where'd those caps come from 02:06 < graff> so what is the basic premise with iptables? blocking ranges of potentially malicious ip? 02:06 < ayecee> graff: one normally uses it to whitelist good traffic, rather than blocking bad traffic 02:06 < aclaivi> It's a firewall 02:07 < graff> so if I make a generic rule that just allows https and ssh that will help? 02:07 < graff> hmm 02:07 < aclaivi> You need to determine what traffic goes through your system, then make rules according to that 02:07 < aclaivi> As ayecee said, firewalls are more whitelisting than blacklisting 02:07 < aclaivi> (usually) 02:08 < ayecee> more effective 02:08 < graff> what does that mean to whitelist something then? it gives it special precedence? 02:08 < oiaohm> graff: firewall designing is not always that simple of a process. Something remember is you normal go the route that keeps the firewall as small as possible. Remember it takes processing time that adds lag processing firewall rules. 02:08 < aclaivi> A whitelist is a set of things that are ALLOWED 02:08 < ayecee> whitelist means explicitly permitting something. blacklist means specifically forbidding something. 02:08 < aclaivi> a blacklist is a set of things that are DISALLOWED 02:09 < graff> like a utrusted user could repeatedly ping, but a non-trusted would get blocked from that? 02:09 < aclaivi> So from the perspective of security, you would start by disabling everything, and then enable the things that you need 02:09 < oiaohm> graff: whitelisting is a good start point to reduce the surface area before looking at blacklisting. Again this is about keeping the rules as compact in processing time as possible. 02:09 < lupine> don't disable ping, it's security theatre 02:09 < ayecee> graff: sure, though that could be accomplished by either approach 02:09 < lupine> and makes debugging much harder 02:10 < aclaivi> graff: It might be worthwhile for you to look into ufw, it just adds a layer on top of iptables which is a little more user friendly to configure 02:11 < lupine> (also, -j REJECT, not -j DROP) 02:11 < ayecee> keep in mind that you _will_ block yourself at some point, so you'd better have a workaround. 02:11 < aclaivi> If you block yourself, you prevent yourself from screwing your system up 02:12 < ayecee> win win 02:12 < aclaivi> Its a feature (tm) 02:13 < graff> I can't afford to make any mistakes. my users need to know that their resources are always available 02:13 < graff> I guess I need a dummy website to experiment with 02:13 < aclaivi> Wouldn't be a bad idea 02:14 < lupine> (if your host lacks an out-of-band shell access option, get a new host) 02:14 < aclaivi> Just setup a cheap vps with digitalocean or something to mess with 02:15 < oiaohm> lupine: https://serverfault.com/questions/157375/reject-vs-drop-when-using-iptables Reject and drop both have downsides. 02:17 < oiaohm> With ping blocking it completely can make diagnostics harder at times it can be important to put size limits on it. https://serverfault.com/questions/523965/blocking-udp-packets-with-a-length-of-4 And its not the only protocol you can consider putting size limits on. 02:17 < ayecee> sounds like increasing complexity for little gain 02:18 < lupine> mostly, people fiddle too much with this lot 02:18 < lupine> I find it hard to justify a firewall unless I have something complex to do based on source-address 02:18 < ayecee> with enough effort, you can make your configuration so fragile that you're scared to change it 02:19 < lupine> the kernel will already respond like -j REJECT for everything that isn't open, and I don't have anything open that shouldn't be open to the world 02:19 < ayecee> that you know of ;) 02:19 < lupine> I have a complete list of what is open 02:19 < oiaohm> ayecee: a lot of dos attacks have invalid sized packets. 02:19 < ayecee> would be a shame if someone went and changed that list while you weren't looking 02:19 < oiaohm> ayecee: and of course you want to straight up drop those. 02:20 < lupine> well, if the machine is compromised, it's obviously untrustworthy 02:20 < aclaivi> ayecee: You think someone would go on the internet with malicious purposes? Okay, Satan. 02:20 < lupine> in those situations an off-device firewall can be handy, but if that's compromised, it's also untrustworthy 02:20 < lupine> there's a point at which you have to shrug and accept some risk 02:20 < aclaivi> Security is about knowing what to trust 02:20 < aclaivi> and also some other things 02:20 < lupine> mine lies before several hours of pointless firewall rule tweaking 02:20 < ayecee> lupine: agreed 02:21 < lupine> most malicious code you're going to encounter is going to be downloaded and run autoamtically by your web browser anyway 02:21 < ayecee> diminishing returns 02:21 < lupine> the firewall is not going to be involved 02:21 < lupine> well, you could set it up to block all outgoing http(s) connections. that would improve quality of life generally 02:22 < aclaivi> Now it all comes down to what device you are using the firewall on? Client or server. 02:22 < aclaivi> Obviously a client is a bit more tedious than a server 02:22 < ayecee> maybe better to log outgoing instead of blocking it outright. 02:22 < ayecee> many a firewall has been set up to block things without knowing what was in use. 02:23 < aclaivi> If you had a server doing one thing and one thing only, it would make a lot of sense to just write up the firewall config and send it on it's merry way 02:23 < ayecee> q e d 02:23 < lupine> .:. 02:35 < graff> anyone here understand how to program these .gitlab-ci.yml files? 02:35 < graff> basically I woud like to not repeat the same code sections over and over 02:35 < graff> but to instead save them in a variable or something 02:42 < socomm> graff: there's a #gitlab channel on this network. 02:42 < graff> socomm: i had bad luck with it 02:43 < graff> considering hanging out there and answering things. but seriously got zero support every time i tried 02:43 < graff> I think they are really busy or working elsewhere 02:43 < socomm> graff: They've paid support. 02:43 < graff> yeah exactly 02:43 < socomm> Pretty sure free support is way down their list. 02:44 < aclaivi> Wecome to the ##linux free support help desk 02:44 < graff> bear in mind this is a foss project. I would have to get a donation to give to them 02:44 < aclaivi> (Quaility NOT guaranteed) 02:44 < Juesto> Graff: it's not that bad if you wait long enough 02:44 < graff> probably though if they knew this they would help 02:44 < Juesto> oh well 02:44 < graff> but the #gitlab seems super dead 02:44 < graff> Juesto: ah ok 02:44 < Juesto> don't they look at their issues list? 02:44 < graff> yeah I think they respond more to actual issue lists 02:44 < graff> like on their gitlab instance 02:45 < Juesto> and try #git if it's more general 02:45 < infinisil> Um, how can it be that a process doesn't get killed with a SIGKILL? 02:45 < graff> true true, they are very nice there 02:45 < Juesto> ##git 02:45 < Skunky> infinisil: they happen occasionally. 02:45 < Juesto> how's that 02:45 < infinisil> Soo what do, just restart? :/ 02:45 < Juesto> it's not supposed to happen 02:45 < Skunky> infinisil: usually when you take a filesystem out from under a running process. yes, just restart. 02:45 < Juesto> but depends on the hardware 02:46 < infinisil> Yeah it's filesystem related I think 02:46 < Juesto> sync, unmount and restart 02:46 < infinisil> The df command is the one hanging 02:46 < Juesto> but sadly sysrq is so locked out 02:46 < Juesto> .. nowadays 02:50 < Skunky> Juesto: if you didn't enable magic-sysrq, that's on you. 02:50 < Juesto> Skunky: it's not always fully enabled by default 02:50 < Juesto> for security reasons 02:51 < Juesto> it's compiled in through 02:51 < Skunky> and it's on you, if you didn't enable it. 02:52 < Juesto> remind me where one permanently controls sysrq allowance 02:52 < Skunky> bootloader 02:52 < Skunky> kernel config 02:52 < Juesto> Skunky: yes it's up to the user but also distros put a fairly secure sysrq allowance 02:52 < Skunky> it's not MY fault distros make idiot decisions. :) 02:53 < Juesto> Skunky: you didn't parse that the support is compiled, just secured 02:53 < Juesto> I'm asking where I'm supposed to change that because I don't remember, no i don't want to recompile the kernel 03:04 < ioerror88> Hello! is there a Linux PAM channel that is active? 03:05 < compdoc> might as well ask here 03:05 < ioerror88> I am considering using PAM to provide authentication of kerberos AND sql stored users (kerberos users can access additional 'stuff') but i'm concerned of cases that aren't obvious where such a configuration would endanger my machines 03:09 < ioerror88> systemd has made things confusing 03:18 < kusznir> hi all: is there a way to prevent I/O errors from one filesystem to slow down I/O on other mounted filesystems? 03:19 < kusznir> EG, if I have a network mount that may or may not be having issues, I want my local disk and OS to still work reliably even if IO is hung up on the network mount. Is this possible? 03:19 < kusznir> If so, how? 03:20 < aclaivi> What does the network mount deliver to your system? 03:21 < lupine> files :3 03:21 < lupine> kusznir: you want -oasync,intr and possibly a few other options too 03:21 < kusznir> In this case, its a VM server with 4 network mounts to different network filesystems. Currently, when one hangs, it takes the entire system (including local disk I/O) and hangs it for 30+ seconds at a time. 03:22 < kusznir> I want to figure out which mount is causing the problem by having I/O to only that mount hang and the others work. 03:22 < lupine> just browse the manpage for your filesystem and see what it supports 03:22 < kusznir> glusterfs 03:23 < lupine> ok, matrix is cancelled 03:26 < kusznir> Another question: I'm getting : blk_update_request: I/O error, dev dm-2, sector 0 (and random other sector numbers) messages on one of my fileservers. How do I figure out what dm-2 is actually? 03:28 < jim> maybe dm-2 is an lvm lv 03:28 < kusznir> quite likely...I have several lv's. How do I find out which lv it is? 03:28 < jim> I was just thinking of that, and I don't know how to map it... do you use lvm on that machine? 03:29 < kusznir> Yes, almost all disk space / partitions are lvms. 03:29 < kusznir> I have three pvs, and probably 5 LVs spread across them. 03:29 < kusznir> 3 vg's 03:29 < jim> it's possible that the node that the lv has, has that number in it 03:30 < jim> let me see if I can find out 03:31 < jim> ok, I think I have an answer... which vg is your lvs in? 03:33 < kusznir> it depends on the lv...2 are in one vg, antoher 3 in another vg, and 1 in the 3rd vg 03:33 < kusznir> I think I traced it down...ls -l /dev/mapper will show the dm mappings. 03:34 < kusznir> In my case, though, the base was a uuid, not immediately helpful lsblk showed me some more info, and the uuid was the parent of a vg. 03:34 < jim> kusznir, you should be able to see /dev/your-vg-name/*, and if you do ls -l /dev/your-vg-name then you might see the answer 03:35 < kusznir> It turns out that physical disk failed in a way that the OS just "ejected" or "removed" it....Strange, never seen that before. Looking further back in the logs, I see where an IO error occured on /dev/sdb1, and now there is no /dev/sdb* 03:36 < skizzy> what do i have to do to enable scp to be used . i want to copy from one linux machine in my network to another but am doing a bunch or stuff so figured i'd just ask 03:36 < skizzy> just enable ssh? 03:36 < kusznir> skizzy, Yep, enable ssh, and you should be good. 03:37 < jim> skizzy, what happens when you try to ssh to the machine in question? 03:37 < skizzy> connection refused. i will set it up just openssh-server right 03:37 < skizzy> y 03:38 < ioerror88> see if ssh'd is running -- sudo netstat -nlp --tcp|grep ssh 03:38 < jim> connection refused usually means no ssh server is there (possibly other things, and that's the first thing to check( 03:38 < jim> )) 03:39 < ioerror88> That'll return any ports sshd is listening on. (and what interface if any) they are bound to. You should see like tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 1234/sshd 03:40 < skizzy> ok got it running but permission denied. tried scp tim@192.168.0.121:/home/tim/file.cpp . 03:40 < skizzy> but when i run ssh on my local machine i can ssh intoo it 03:41 < skizzy> ok i can ssh into it not scp 03:42 < skizzy> I mean from machine 1 i can ssh into machine 2 but not scp into machine 2 i get permission denied 03:43 < aclaivi> Are you providing the credentials properly? 03:45 < skizzy> nevermind it was a local permission denied. i was not in my home directory 03:45 < ioerror88> Permission denied could mean you have a shell like rssh setup to restrict scp (see /etc/rssh.conf if present) or more likely the remote user cannot write where you are trying to write (are you calling sudo somewhere wwrong?) 03:45 < ioerror88> Cool. 03:45 < skizzy> it works! and taht was just an easy setup 03:48 < skizzy> woot! I got keys on one computer triggering macros on another computer for my everquest boxing lol. 03:48 < skizzy> I got everything i need in Linux now 03:48 < jim> everquest?! I heard about that thing years ago 03:48 < skizzy> yeah it's still around man! 03:49 < skizzy> They do like restarts from the beginning, that's what I'm on 03:49 < jim> paid per time? 03:49 < ioerror88> skizzy: Linux is the valhalla ;) 03:49 < skizzy> it doesn't cost me anything i make the currency in game to pay for my game time 03:49 < jim> halla at yo val! 04:01 < littlepyro> what is suid? 04:02 < littlepyro> does it allow us to be root for sometime? 04:03 < Happyhobo> Hi folks. 04:05 < Happyhobo> My cable company says I'm getting 10 Mb per second. Speedtest right next to the router says 70 but here through two three walls it says 10, is that the same as the 10 advertised and if it is why does it get 70 next to it. I have my new card in and the bluetooth is made to sit next to another computer and transfer files back and forth. 04:06 < djph> Happyhobo: two computers on the sane network will be significantly faster than "transfers over the internet" 04:06 < djph> s/sane/same/ 04:07 < Happyhobo> Oh no I don't care about transferring files. I wanted it to play pandora through my blue tooth stereo. It only finds the stereo if it's like 2 foot from it. 04:07 < djph> if you get 70 mbps to a remote server, then they *probably* allow some burst rate that exceeds your normal rate 04:07 < djph> which is normal for bluetooth 04:07 < djph> it's VERY shortrange 04:08 < djph> well, usually -- IIRC the spec is something like 2-3 meters 04:08 < Happyhobo> That and youtube. Now the 10 advertised, 70 speedtest.xinfity.com right next to it and 10 on it through the let's see 1 2 3 4 walls. My phone's bluetooth extends 30 foot 04:09 < Happyhobo> I need to get the dedicated Dell card and that has that kind of range. but is 10 through the walls the same as I should be getting? 04:09 < djph> huh? 04:10 < Happyhobo> Why did you say huh? 04:10 < djph> if you're only paying for 10 mbps, you technically "should" only be getting 10 mbps from anywhere. 04:10 < Happyhobo> It says Mb/s 04:11 < djph> however, ISPs can allow "burst" rates that exceed that limit 04:11 < kodrakan> Id be surprised if youre getting mucj more than advertised... Maybe 20% not 7x 04:11 < djph> mbps = Mb / s 04:12 < djph> kodrakan: yeah, I know some ISPs that allow bursts up to 100 on whatever plans; so .. 04:12 < supernov3h> is there some posix component that specifies user ID maps, like how 1000+ seems to be user-space, <1000 system, 65534 nobody etc? 04:12 < Happyhobo> I just got 16 through all the walls 04:12 < djph> so stop crying that you're getting *MORE* than you pay for 04:13 < Happyhobo> So I should be happy? 04:13 < djph> supernov3h: no, some distros use 500+ for users, for example 04:14 < aclaivi> This isn't the case of Mbps and MBs right 04:14 < supernov3h> djph: bluetooth what generation? ble 5 can far exceed 3 metres, within spec up to 30dBm (0dB) people have reached over a kilometre 04:14 < djph> speedtest does bits/sec 04:14 < aclaivi> Like your ISP tells you 10MB/s and speedtest tells you 75mbps 04:14 < aclaivi> Yeah thats what I'm saying 04:14 < djph> as do most ISPs 04:14 < aclaivi> is that the confusion 04:15 < Tahlwyn> @so@sk 04:15 < Tahlwyn> whoops 04:15 < Tahlwyn> ignore me 04:15 < supernov3h> aclaivi: that would have to be the first isp to say the data rate in bytes/sec not bits/sec lol 04:15 < Happyhobo> 16 Mbps on speedtest.xfinity.com isn't more than they advertise aclaivi 04:16 < djph> Happyhobo: what are you paying for, 10 Mb/sec right? 04:16 < Happyhobo> Yes.\ 04:16 < djph> 16 is more than `0 04:16 < djph> *10 04:16 < Happyhobo> Oh and it got 16 while I was playing Pandora. 04:17 < djph> you're doing better than you should. be happy and call it a day 04:18 < aclaivi> supernov3h: the only way these complaints made any sense to me was if there was a misunderstanding like that lol 04:19 < djph> or a typo that he's actually paying for 20, and he hasn't noticed it 04:19 < Happyhobo> now if my other laptop had blootooth they could sit side by side and talk like furbees. 04:20 * aclaivi increases level of confusion. 04:20 < djph> I think he's just stringing words together, without understanding anything he's saying 04:20 < djph> (or anything we're responding with) 04:22 < Happyhobo> aclaivi the bluetooth that is on this network mini card will not respond to my phone if it is across the room and it will interact with the stereo ten feet away. I read that this is a transfer blue tooth chipset. It has no range whatsoever. It's mean to do some kind of ftp over bluetooth with two units next to each other. Does that make sense? 04:23 < ioerror88> Happyhobo: Furbee do not 'talk'.. 04:23 < Happyhobo> If I set the phone on the table beside me just two foot away it will talk to the phone. It will hook up the phone through the netbook and scream at the caller on the other end with a distinct echo. 04:23 < Happyhobo> Yes, they do ioerror88. 04:24 < Happyhobo> I quickly realized making calls through the notebook(my other is a netbook 10.1) is not the best idea. 04:24 < aclaivi> Wasn't your problem related to your internet to begin with, not bluetooth? 04:26 < Happyhobo> It's all on one mini pci-e board. I had an older intel mini pci-e that had three antennas, in, out and Mimo. Using that card I was only getting 1 Mbps. I replaced it and now I'm trying to wrap my head around this. I can't believe two antennas is better than one. 04:26 < djph> funny how that wors 04:26 < djph> *works 04:27 < djph> 2x2 (2 antennas) tends to double the capacity 04:27 < strive> Is it possible to visit an internal web page on my server - displaying statistics about the server? 04:27 < djph> sure, if you set one up (or use something like webmin) 04:27 < Happyhobo> New technology I suppose. Now I need to get a bluetooth dongle with an extended range. Yes, but the other had 3 djph. 04:28 < djph> Happyhobo: then it was probably simply a damaged card 04:28 < djph> or you had the wrong driver 04:28 < djph> or any number of other problems that wifi can have 04:29 < Happyhobo> The 3 antenna one was very unstable. It liked to crash and says unable to activate network connection. 04:29 < Happyhobo> say* 04:29 < ioerror88> If you need range you need wifi not bluetooth.. 04:30 < Happyhobo> I use my phone 15 to 20 feet away from my car into the car stereo. I use my phone 10 feet from the house stereo ioerror88. 04:30 < Happyhobo> This card's bluetooth range isn't much longer than the average hoohoo length. 04:31 < Happyhobo> Why I need to have my callers listening to an obnoxiously loud echoing response is simply beyond me. 04:32 < djph> all semblance of sense is out the window 04:32 < aBound> Sense what sense. :P 04:33 < djph> fair enouhg 04:33 < k_sze[work]> Hmm, how should I feed the output of `grep -rZlI` to sed, as filenames to process? 04:33 < djph> you shouldn't 04:34 < littlepyro> i have the problem with suid. i did u+s file.py.. the owner of the file.py has -rwsrw-r-- and now if i do su - testuser and execute the command, it says permission denied 04:34 < littlepyro> what am i doing wrong 04:34 < Happyhobo> djph right now I could turn on my minibookshelf stereo on the other side of the room. It has a bluetooth connection available. I start my phone. I open the bluetooth dialog and click on the name and description of the phone. They link. I start pandora. Pandora starts through the minibookshelf stereo. Does that make sense? To get the notebook with this bluetooth card to link I have to sit the notebook on top of it. Does that make 04:34 < Happyhobo> sense? 04:35 < djph> Happyhobo: sure, especialy if you got the wiring wrong when going from 3 antenna connections to two 04:35 < ioerror88> Could be crap antenna design... could be crap chipset. 04:35 < Drakonan> so what is the problem? 04:35 < k_sze[work]> djph: why not? 04:35 < ioerror88> Ohh.. There's butchery involved. Am out! 04:36 < Drakonan> so one device can do it far away and another cant? 04:36 < djph> if you want to process files, a for loop is probably more sane 04:36 < Drakonan> you may've answered it yourself the other one just doesnt have the range of the other 04:36 < Happyhobo> There were no card with the MIMO other than the original on newegg. 04:36 < ioerror88> You could do while.. 04:36 < djph> ioerror88: yeah, he had an intel(?) card which was apparently faulty, now something else 04:36 < Happyhobo> This is a newer design Intel card with bluetooth. 04:37 < ioerror88> grep ... | sed ... | while read line; do stuff; done-- depending what that grep optionsdo 04:37 < djph> literally all (decent) 802.11a/b/g/n/ac cards have MIMO 04:37 < djph> 2x2 or if you want the expensive one, 3x3 04:38 < k_sze[work]> djph: sure, how do extract each filename from the output of `grep -rZlI` then? (They are null-separated because of the -Z option.) 04:39 < djph> there's something missing in that grep (like a pattern, for example) 04:39 < djph> ... although #bash would probably be a better place to ask 04:40 < k_sze[work]> djph: I omitted the pattern because it's not important to the question. 04:40 < ioerror88> Happyhobo: What antennas do you have connectors for? What does card have connection for? If 2x black & 1x grey - put a little polyolefin over the connector on the grey one so as to not have it Do Bad Things when the PC is moved... 04:40 < djph> k_sze[work]: then use a pipeline i guess 04:41 < Happyhobo> OK ioerror88 04:41 < ioerror88> If 2x in computer 3x connectors on card -- you're without much luck besides spending 13$ on amazon for a legit dell card (they're quite cheap nowadays with bluetooth even) 04:42 < ioerror88> im guessing old adaptor had no bluetooth? 04:42 < djph> ioerror88: most (even with BT) just have 2x antennas 04:43 < djph> unless they're 3x3 04:43 < ioerror88> djph: Yea, if card is 3x3 its not gonna work if he has 2 antennas.. if he's got 3x3 antennas, grey is the other array and should be safely insulated & stowed 04:43 < djph> defintiely 04:44 < pfred1> I could write a book, "How to Botch up Your PC in 80 million Hard Steps" 04:44 < djph> step 1, ask a question and don't understand the answers you've been given 04:44 < ioerror88> if antennas are such and properly attached & still not working? Crap chipset.. I keep finding chinese clone NICs :( 04:45 < Happyhobo> I'm digging through newegg looking for a 3x3 intel and all I can find is a refurbished like my original. 04:45 < djph> yeah, which is why I only buy from legit sellers -- even if it's $5 more than "mao's discount scamland" 04:45 < pfred1> ask stupid questions get stupid answers 04:45 < ioerror88> You dont have to do 3x3, just make sure the black cables are attached and grey is safely insulated (I like heat shrink) and stowed to not rattle or short things..... 04:45 < djph> Happyhobo: what MODEL are you looking for? 04:46 < Happyhobo> an intel chipset that is 3x3 and has bluetooth 04:46 < pfred1> djph $5? Do you know how many 6 packs of beer i can buy for $5? 04:46 < ioerror88> You dont want things shorted with Lithium chemistry batteries involved.. It's quite impressive show. Children do the worst things to computers... :\ 04:46 < ioerror88> pfred1: Beer or "American ale"? 04:47 < pfred1> ioerror88 I didn't say good beer 04:47 < ioerror88> in the proper bar/cafe on wing night, $5 will get you about a gallon of bud light, but im not sure if you can consider that beer 04:48 < ioerror88> I cant imagine what a 5$ 6 pack would be like :( 04:48 < pfred1> ioerror88 does drinking beer make you smarter? 04:48 < pfred1> ioerror88 it made Bud Weiser! 04:48 < ioerror88> pfred1: Unlikely. But it does keep you from choking people who are dumber, yes? 04:49 < djph> Happyhobo: quick search is intel doesn't even make 3x3 (anymore?) 04:49 < djph> what's the model number of the card you have (had?) 04:50 < ioerror88> pfred1: Ha. It's sad that we represent ourselves as a nation of civilized people, yet we still alllow the sale of budweiser and bud light. Worse yet, we allow them to claim to be "the American beer" :( 04:50 < djph> or rather, looks like they only did it in 802.11n 04:50 < djph> ioerror88: natty lite 04:51 < boingolov> natties and phatties 04:51 < dogbert_2> LOL - Intel is under investigation for potential age discrimination in its approach to layoffs initiated in 2016, according to a report. Engadget: 04:51 < dogbert_2> hey djph 04:51 < djph> yo 04:52 < ioerror88> djph: Ok well i can deal with that. It's still better than butt lice. Lol. 04:52 < ioerror88> Yall seem alright. 04:52 < pfred1> OK I got my video accelerated now I need to fix my audio 04:52 < sh1ro> ioerror88: they're not a poor representation 04:53 < pfred1> what's the command alsactl? 04:53 < ioerror88> sh1ro: We have some amazing breweries. But american style "beer" is not beer at all. 04:53 < pfred1> ioerror88 the best brewery in the US is 7 minutes up the road from me 04:53 < Happyhobo> I'd have to go downstairs to get it and the other is inside the notebook. 04:53 < djph> southern tier? 04:54 < ioerror88> pfred1: alsactl? That sounds like some sort of witchcraft.. Try poking it with a stick -- alsactl init 04:54 < djph> well, that makes it completely impossible to help you sort the problem out 04:54 < ioerror88> Tech support via telepathy isn't supported over IRC. Please try again in 2048. 04:54 < pfred1> ioerror88 heck at this point I don't evne know if i have any sound installed 04:55 < sh1ro> or try darpa 04:55 < pfred1> ioerror88 I installed a video card this afternoon and trying to get acclerated drivers going did not go well 04:56 < ioerror88> pfred1: also-info.sh 04:56 < ioerror88> sh1ro: darpa uses telepathy over avian carriers... 04:56 < ioerror88> Still faster than BT! 04:57 < pfred1> they use the H.A.A.R.P. array 04:57 < pfred1> adjust your tinfoil hats! 04:58 < ioerror88> HAARP is just a way to control the mirror for the particle weapons.... They hit the ionized gas mirror and are reflected to their target :O 04:58 < pfred1> I think Debian has it in for Nvidia binary drivers 04:58 < Happyhobo> bad f-ing ass it's easy to get all my photos from my phone to computer. It's not totally useless after all. 04:59 < ioerror88> pfred1: It is very complicated, but they use ionosphere as a parabolic reflector by heating up part of ionosphere (using the directed array) -- then various experiments are bounced off the mirror. I dont know anything about the experiments, however... 05:00 < ioerror88> it is for weather improvement... To stop global warming or something. 05:00 < pfred1> it was to burn a billion cubic meters of gas they couldn't get to a market 05:01 < nekoseam> I feel like I'm gonna vomit out instant ramen 05:01 < nekoseam> I feel like I'm gonna vomit out instant ramen 05:02 < ioerror88> LNG compression is viable for pipeline transport... 05:02 < ioerror88> nekoseam: User error: Replace user. 05:02 < pfred1> nekoseam it's tastier the second time around 05:02 < nekoseam> oh god 05:03 < ioerror88> That awkward moment when you reboot production during a maintenance window.. and 5 minutes later it's still not responding to pings :( 05:04 < pfred1> that awkward moment when you see your machine has 4 sound devices in it and you have no idea which one you're supposed to use 05:04 < ioerror88> pastebin it. 05:06 < ioerror88> And be patient -- i'm logged into a serial console in another country via dialup. Life is not good this tuesday night. 05:06 < ioerror88> Why is grub configuration handled so clumsily on every distro? :\ 05:11 < Happyhobo> OK next step usb dongle with a long range. Let my furbee talk with my stereo. 05:11 < ioerror88> There 'tis.. Long fscks always are a panic... 05:11 < nekoseam> I need to improve my health habits. The only thing I've done today is work on bash scripts, eat instant ramen and drink mtn dew 05:11 < Happyhobo> brb must go grocery shopping and find something to eat that I can eat without my teeth in. 05:12 < ioerror88> nekoseam: Sounds like you are well on your way to PFY (Jr Admin) position! 05:12 < nekoseam> ioerror88: oh boy 05:12 < nekoseam> Oh, I forgot I ate some icecream 05:12 < ioerror88> Repent ye, for there is still time to fix the error of thee ways. 05:13 < supernov3h> if you had say a thousand startup scripts that all remounted /proc, would there be any chance for trouble? 05:14 < ioerror88> supernov3h: That seems at worst very slow -- perhaps even likely to bork it - why not call mountpoint? What voodoo are you trying to do? 05:14 < nekoseam> Hey [R] 05:19 < pfred1> ah ha the old default master mute deal! 05:21 < supernov3h> ioerror88: I was just wondering, not trying to do anything 05:22 < socomm> supernov3h: I would never hand 1000 start-up scripts that all did the same thing. What are you /trying/ to do? 05:22 < socomm> s/hand/have/ 05:22 < supernov3h> socomm: it was just a question... I'm trying to drink some diet coke right now though :D 05:22 < supernov3h> s/$/g/g 05:23 < [R]> you know what they say 05:23 < [R]> you never see thin people drinking diet coke 05:23 < socomm> Heh. 05:23 < socomm> If its good enough for trump, its good enough for me. 05:23 < ioerror88> Tis not good for your microbiome, yo. 05:24 < [R]> socomm: rofl 05:24 < socomm> WTF "Trump reportedly drinks 12 cans of diet coke each day" 05:24 < [R]> thats a lot of coke 05:24 < socomm> I hope this is "fake news" 05:24 < [R]> i tell you what 05:24 < ioerror88> Praise be unto Odin! The git server has come back up and it seems noone even noticed it was down for a fsck... 05:24 < [R]> tahts slike 2.5 2 liter bottles 05:24 < socomm> I guess you can say ..... no one gave a fsck. 05:25 < supernov3h> I cycle 600k a week, a lot of people keep telling me to put weight on, I'll keep drinking it 05:25 < ioerror88> socomm: Oh one admin did. A very scared 8 1/2 minutes. 05:25 < littlepyro> how to find which user we are executing the program? 05:25 < supernov3h> I drink it for the caffeine, like most people drink coffee, but coffee tastes wrank 05:25 < ioerror88> littlepyro: whoami 05:26 < socomm> littlepyro: what program? also is this homework? We totally love homework questions here. 05:26 < [R]> socomm: you're homework 05:26 < ioerror88> If you mean in C you must use geteuid(); But i'll leave that to the manpages. 05:26 < littlepyro> ioerror88: im playing with suid.. i read from somewhere suid is used in Linux, for providing elevated privileges temporarily during execution 05:27 < ioerror88> Linux is user friendly, it's just picky about it's friends. 05:27 < supernov3h> does cat /proc/self/loginuid expose the original user who ran a sudo -Siuroot? 05:27 < littlepyro> socomm: ahaan.. this is nt the homework question.. so please refrain 05:27 < socomm> littlepyro: Are you sure? I've a teachers copy of the book. Trust me. I can help you. 05:27 < littlepyro> ioerror88: without c program can we not find it who is executing the program for a while? 05:27 < ioerror88> why do teachers come on IRC to troll? 05:28 < littlepyro> socomm: yes im 05:28 < littlepyro> socomm: dont waste my time please 05:28 < supernov3h> kek 05:28 < socomm> littlepyro: I appologize. 05:34 < ioerror88> littlepyro: WHich program? According to some quick estimation -- i have 3245 programs on my machine ( see also (echo "$PATH" | sed s/\:/\\n/g | while read i; do find $i -type f -executable 2>/dev/null; done) | wc -l 05:34 < e36freak> you're recursing 05:35 < ioerror88> True 05:35 < pfred1> I wish the old double tab trick still worked 05:35 < ioerror88> odn-noc:~> (echo "$PATH" | sed s/\:/\\n/g | while read i; do find $i -type f -maxdepth 1 -executable 2>/dev/null; done) | wc -l 05:35 < ioerror88> 2347 05:35 < littlepyro> usera has a file called filea.py and when i set suid, and do su - userb and then run /home/usera/filea.py... in this time how can i check that the filea was run by usera.. 05:36 < littlepyro> ioerror88: ^^ 05:36 < kajika> Hello everyone, I'm trying to find a way to check the parameter 'file=' of the udb_f_mass_storage that is already loaded in my embeded device. Does anyone know how to? 05:36 < [R]> kajika: /sys 05:37 < ioerror88> littlepyro: So you are using python -- that helps... https://docs.python.org/3/library/getpass.html go read this. 05:37 < nekoseam> Salem Oregon has a wanter contamination right now 05:37 < nekoseam> So I can't even drink some water FeelsBadMan 05:37 < littlepyro> ioerror88: there is nt any linux commands to check who is the user that gets executing that file? 05:38 < ioerror88> Yet CNN is still on about the first lady hasn't appeared in public in a few weeks (zomg!) 05:38 < ioerror88> littlepyro: You would need to check from inside that script (filea.py)... see that URL.. Add appropriate line or two of code.. All is well 05:39 < ioerror88> See also standard warranty: "If it breaks, you get to keep both pieces. If there are more than two pieces, please return them for further study!" or "Linux is user friendly, it's just picky about who it's friends are." 05:40 < pfred1> ioerror88 Linux wasn't too friendly to me today 05:41 < ioerror88> pfred1: That is error 37... 05:41 < pfred1> I just had to have a video card didn't I? 05:41 < ioerror88> EUSER (137): User Error: Replace User. [LIBMK] 05:41 < nekoseam> ioerror88: A FEW WEEKS? Unthinkable! 05:42 < ioerror88> nekoseam: Yea.Seems far more important than the water 05:42 < ioerror88> i mean people drink the first...wait...Yea..wtf CNN. 05:44 < socomm> pfred1: Technically its your vendors job to provide drivers not the OS. 05:46 < pfred1> socomm the vendor did. The distribution maintainers just did not make it easy to install 05:46 < socomm> Use a diff distro. You've 300002308430021 to choose from. 05:47 < pfred1> well the third time around it went smoothly enough 05:47 < pfred1> but the first two times were doozies 05:47 < socomm> Not sure if PEBKAC or actual problem. 05:47 < pfred1> I've been running Nvidia binary drivers since they were in beta too 05:48 < sauvin> socomm, don't exaggerate. You're high by roughly two orders of magnitude. 05:48 < pfred1> so it is not like I've never seen any of this stuff before 05:50 < ioerror88> dang phone's so fancy, im not quite sure why i even use the computer anymore besides the whole keyboard thing. and i could awkwardly plug one into the microusb...but i fear the model M would promptly deplete my battery 05:50 < socomm> IDK. I've never blamed the OS for vendors inability/general lack of interest in providing proper packages for their drivers. 05:51 < benjamin-l> nekoseam: ah, that's what that was 05:51 < socomm> ioerror88: Can you get real work done on a phone? If you're doing passive media consumption, phones are fine, but anything outside of that is a pain. 05:52 < benjamin-l> it appears that the website to check what the alert is has gone down 05:52 < pfred1> socomm you claim a lack of interest but tell me who has provided accelerated drivers for longer than Nvidia has? 05:53 < socomm> pfred1: Dont recall saying anything about nvidia, was talking vendors in general. 05:53 < pfred1> I see 05:54 < socomm> But yeah if you're not using one of the major distros, or bleeding edge version of a distro there is good chance your drivers may not work out of the box. 05:54 < nekoseam> Only tap water I've had today was filtered through a brita and put in instant ramen 05:54 < pfred1> socomm you don't think Debian is a major distro? 05:54 < nekoseam> Debian is technically the most popular distro when you take into account that Ubuntu and derrivatives are heavily based on it 05:55 < socomm> Well. Its mostly what the vendor thinks is major, thats what they invest money+time into. 05:55 < socomm> They may see Debian as a "server" distro and invest into distros like Ubuntu which are more end-user type. 05:55 < pfred1> nvidia knew all about Debian because whe n Itried to run their binary driver they named it by name 05:55 < sauvin> Debian is by *itself* plenty popular enough. 05:55 < nekoseam> I thought it was the 2nd most popular distro 05:55 < pfred1> it knew i was trying to install on Debian 05:56 < socomm> pfred1: What is the actual problem you had? 05:56 < nekoseam> Definitely in the top 5 05:56 < pfred1> socomm quite frankly it is too involved to describe here 05:56 < socomm> Hah. 05:56 < socomm> Paste bin it. 05:56 < pfred1> socomm if it was simple then I'd have dealt with it simply 05:57 < socomm> Post it on reddit. They like those kind of stories there. 05:57 < pfred1> one error alone was about 5 screens of text 05:58 < pfred1> it was actually the biggest error I've ever seen 06:01 < socomm> pfred1: hehe 06:01 < graff> any C programmers here looking to test their mettle on hlibc? 06:01 < lnnb> hay hay hay! 06:02 < lnnb> libc is for nerds, int 80 fo lyfe 06:02 < swift110> hmm 06:02 < graff> we have our CI up and can probably support more devs. we have all skill ranges from hobbyists to professional scientists. 06:02 < socomm> lnnb: I don't think they allow nerds on this channel, we should be safe here. 06:02 < graff> pretty good stuff yup. we have a lot of fun 06:03 < ioerror88> i hang out with some os nerds, but its my bedtime. 06:03 < lnnb> good point socomm, i can rest easy now 06:03 < graff> lnnb: :) 06:04 < ioerror88> Long past bed time but reboots in London happen at peculiar local times :) 06:06 < supernov3h> hey I'm trying to debug why a particular usb drive doesn't present itself and mount, lsusb shows identical output for all devices before and after I plug it in even though the usb drive works... what's up with that? 06:07 < [R]> supernov3h: what does dmesg say 06:09 < supernov3h> nothing by the looks of it 06:09 < supernov3h> THere was a VBUS_ERROR in a_wait_bcon earlier 06:10 < [R]> well tahts pretty bad 06:10 < [R]> i tell you what 06:10 < supernov3h> I know what vbus is I'm an electrical engineer lel 06:11 < supernov3h> I'll just put this down to a driver failure 06:12 < lnnb> drivers/usb/musb/musb_core.c:721(ish?) 06:12 < Anon-chan> Guys anyone has any idea why this appear at random in my console? 06:12 < ioerror88> Sure blame the software... 06:12 < Anon-chan> https://usercontent.irccloud-cdn.com/file/T3jPNewU/image.png 06:13 < [R]> Anon-chan: because you typed something weird 06:13 < Anon-chan> I mean, I'm literally not touching my PC and those appears 06:13 < Anon-chan> along tildes sometimes 06:13 < [R]> lets try not selleccting a tiny portion of the screen 06:13 < lnnb> man that function is like 6 screens, well over 80 columns 06:13 < [R]> what else is going on 06:13 < Anon-chan> or where can I get the values, thingies to see what button is "being pressed" 06:14 < benjamin-l> those look like escape sequences 06:14 < Anon-chan> [R]: it's SSH, and doesn't matters what I'm doing. Those just appear at random from time to time. 06:14 < [R]> lol 06:14 < [R]> well if it doesnt matter 06:14 < Anon-chan> when I leave it alone, somtimes I come to "~~~" 06:14 < [R]> then my help doesnt matter 06:14 < Anon-chan> I mean, they appear no matter what I'm doing 06:15 < Anon-chan> https://www.irccloud.com/pastebin/GaHAWgYu/ 06:15 < Anon-chan> they just appear :V 06:16 < benjamin-l> is it always the same number? 06:16 < Anon-chan> It's either that ^ 06:17 < Anon-chan> or just ~~~~ 06:17 < sanjibukai> Hi everybody 06:17 < sanjibukai> Does anyone used or know xkb? 06:18 < supernov3h> lnnb: don't bother poking at musb_core.c, there will be a problem with VBUS in this hardware, it's custom made, custom chip, custom cross-compile etc 06:18 < supernov3h> and besides our musb_core is modified 06:18 < lnnb> supernov3h: that dev_printk is zany 06:19 < benjamin-l> apparently `^[[28~` is the sequence for F15? 06:20 < Sleepygravel> How do I fix a high pitched/super fast microphone? It worked normal on a Bedian live install but after installing debian I got the problem back 06:20 < benjamin-l> Anon-chan: https://superuser.com/a/987505 06:20 < benjamin-l> this person seems to have a similar problem 06:20 < supernov3h> benjamin-l: approximately perhaps, it varies system to system 06:20 * Anon-chan blinks 06:20 < Anon-chan> there is f15? 06:21 < Anon-chan> and thanks 06:21 < ioerror88> all the way to f24.. 06:21 < ayecee> after enough drinks, anything is possible 06:22 < sanjibukai> Hi 06:23 < benjamin-l> hi 06:23 < supernov3h> the function keys are mapped in groups of four aren't they, they send some bytes that are like F1 = 4, f2 = 3 to F4 = then it skips for the next four, I don't see why perhaps the arbitrary nature of where that was chosen would have specified any limit 06:23 < ayecee> f all the keys 06:24 < supernov3h> Oh, groups of 5 06:24 < supernov3h> F15 should be 27 not 28... 06:25 < supernov3h> 28 should be unused 06:26 < sanjibukai> Hi everyone 06:26 < sanjibukai> Does anyone already tweaked his keyboard layout with xkb ? 06:27 < supernov3h> this is ludicrous, they chose 17, 18, 19, 20, 21, !!, 23, 24, 25, 26, !!, 28 06:27 < ayecee> inconceivable! 06:29 < benjamin-l> wow, that's awful 06:29 < sanjibukai> I wanted to know how can I add a third modifier in the symbol file mapping? 06:31 < oiaohm> supernov3h: some of that keyboard stuff is historic stupidity. 06:33 < oiaohm> supernov3h: missing numbers in F keys line up with keys that were on some keyboard but were removed because they were pointless. 06:34 < sauvin> Will the real pabed give the poseurs the heave, please? 06:34 < supernov3h> oic 06:35 < oiaohm> supernov3h: between scroll lock F14 and before Pause break F15 on some really old keyboards there was a release key. Yep does exactly what pressing scroll lock a second time does. 06:35 < supernov3h> interesting 06:35 < sauvin> pabed, answer me. 06:36 < oiaohm> supernov3h: of course pressing scroll lock second time on the systems with release key did nothing. 06:36 < oiaohm> supernov3h: think based of the ideas of old typewriters where caps lock applied the lock and then you pressed shift to release it. 06:37 < sauvin> .oO("Grandpa, what's a typewriter?") 06:37 < supernov3h> I didn't know what's what shift did but I guess it's obvious 06:37 < oiaohm> supernov3h: modern day shift does not have that feature. 06:38 < supernov3h> oiaohm: no I get it, you're talking about hitler times 06:38 < oiaohm> supernov3h: it explains some of the strange spaces. 06:40 < supernov3h> is there a command line wrapper for libconfig 06:40 < oiaohm> supernov3h: it is fun a times to put some of the old legacy computer keyboards in front of younger people and watch them have hell. 06:40 < oiaohm> supernov3h: with caps lock and other features not working how they are use to. 06:41 < supernov3h> About as fun as putting a modern high end PC in front of an old person and asking them to have fun playing PUBG or something 06:42 < oiaohm> supernov3h: that is why I do it. Gives the young person the feeling of truly being dropped of the cliff so are a little bit more understanding after it. 06:43 < oiaohm> supernov3h: particularly when it does not look that different. 06:48 < supernov3h> oiaohm: fair enough lol 06:49 < screwsss> im running linux of a live usb. theres no way to get it to save all my settings and all of that is there 06:56 < supernov3h> lol randomly reading a man page and it has female gender pronouns, how odd, most often its neutral, but I'd expect if there was a slip-up it would be for a male one 06:58 < lozy_86> I have a 2013 trash PC, powered by Intel pentium G2030 system. What linux distro will do fine there 06:59 < lozy_86> I would love to watch some videos, browse web and do some coding there 07:00 < [R]> lozy_86: whatever one you are confortable with 07:01 < Triffid_Hunter> lozy_86: distributions are just different ways of packaging the exact same software with basically the same performance. You'd be better off looking at lightweight DEs rather than trying to choose a specific distro given your criteria. 07:01 < benjamin-l> arch is clearly the only option :) 07:02 < benjamin-l> but, yeah, having a lightweight DE and programs means a lot more than what distro 07:02 < Triffid_Hunter> lozy_86: if you're new to linux, maybe linux Mint XFCE edition? XFCE is a fairly light desktop environment 07:02 < lozy_86> Triffid_Hunter: So can u suggest me one? I am not much into tweaking 07:03 < benjamin-l> xfce and lxde are pretty nice as far as floating wms go 07:03 < lozy_86> Triffid_Hunter: Ow, Linux Mint XFCE4, okey 07:05 < screwsss> lozy_86: mint linux 07:05 < screwsss> so running linux of a live usb wont save anything that i put on it or any settings i change? 07:07 < jnewt> how do i setup cutecom to look for ttyUSBx in addition to ttySx in /dev when getting list of available ports 07:07 < jnewt> i can't find it anywhere on the website or with man cutecom 07:07 < screwsss> jnewt: www.google.com 07:09 < jnewt> screwsss, omg, that's an amazing site. thank you so much. you're so helpful. 07:09 < screwsss> jnewt: np 07:09 < [R]> jnewt: it just asks qt for a list 07:09 < screwsss> glad to help 07:10 < [R]> jnewt: so if it's not finding them, thats a bug in qt 07:13 < jnewt> [R] it's not finding them. every CLI check for devices / ports I can think of picks it up as expected. i don't know much about qt (other than what it is on a very basic level). 07:14 < [R]> [10:10:07] <[R]> jnewt: so if it's not finding them, thats a bug in qt 07:16 < aclaivi> Suckless.org ssl cert expired rip 07:16 < aclaivi> Now how am I supposed to know what sucks and what doesn't 07:24 < Happyhobo> How would linux hook up two bluetooth cards?\ 07:24 < Triffid_Hunter> jnewt: mine reads ttyUSBx 07:25 < benjamin-l> aclaivi: that's bad. do you know if the project is doing okay? 07:25 < [R]> Happyhobo: "hook up"? 07:25 < Happyhobo> Great I have the song Hunter by Dokken running through my head now Triffid_Hunter 07:25 < Happyhobo> [R]: use both. 07:26 < [R]> Happyhobo: what? 07:26 < aclaivi> benjamin-l: Well it's only been out for a few hours so far so probs not a real issue 07:26 < aclaivi> **hopefully** 07:26 < Happyhobo> The internal card has no range. The dongle will have a long range, I want to use the dongle to hook up to my stereo.\ 07:27 < [R]> ok, so use th edongle 07:27 < Happyhobo> but how do I disable the one on the pci-e card:? 07:28 < [R]> why do you have to disable it? 07:28 < Triffid_Hunter> jnewt: https://i.imgur.com/STi0ylX.png 07:28 < Happyhobo> Because linux can't manage two. 07:28 < [R]> what? 07:29 < sauvin> Don tyou lov eho w[R] tpyes? 07:30 < jnewt> [R] Triffid_Hunter that's messed up. what's your /dev/ttyUSB0 show for perms and user / group 07:30 < Happyhobo> Linux only handles one [R] 07:30 < [R]> jnewt: ones that mean i have access ot it 07:30 < [R]> jnewt: do you have access to yours? 07:31 < jnewt> [R], rw, not x via group. owner is root, group is dialup which i belong to 07:31 < [R]> Happyhobo: says who? 07:31 < [R]> well then you have access to it 07:31 < Happyhobo> hmm 07:32 < Triffid_Hunter> jnewt: 660 root:plugdev 07:32 < Happyhobo> it doesn't need to receive a signal just send one to the stereo 07:32 < Triffid_Hunter> jnewt: dialout? you're on debian/ubuntu? 07:32 < jnewt> ubuntu 16.04 07:32 < Abbott> I am temporarily using my SWAP partition to house a small linux installation, but when I use fdisk to change the id, it stops at 82, so I can not set it to 83 for "Linux" 07:33 < Abbott> there are other options for "Linux root" or "Linux filesystem," does the id matter? 07:33 < [R]> Abbott: "it stops"? 07:33 < Adran> did you do swapoff? 07:33 < Abbott> Adran: yes, I did swapoff 07:33 < Happyhobo> sorry. 07:33 < Adran> also why not just use a vm 07:33 < Abbott> [R]: yeah like it says press 'L' to view all options and the last option is 82 07:34 < Abbott> Adran: I am following a guide to replace a current installation with another and it recommends using an interim installation on the SWAP partition 07:34 < [R]> Abbott: tahts gpt 07:34 < [R]> its not 83 for linux 07:34 < Triffid_Hunter> Abbott: can't just type 83 manually? maybe it's a scrolling list and your screen is tiny? have you tried cfdisk or parted or anything a little more modern? 07:34 < Abbott> Triffid_Hunter: if I type 83 manually, it tells me the partition type will remain unchanged and keeps it as "Linux Swap" 07:34 < jnewt> Triffid_Hunter, does that groupname seem odd (like there's a standard one that cutecom / qt looks for (plugdev?) 07:35 < Abbott> I don't have cfdisk installed, but I might have parted 07:35 < Triffid_Hunter> jnewt: plugdev is for random usb things in gentoo, dialout is normal for debian/ubuntu afaik 07:35 < Abbott> no, I don't have either installed 07:35 < Triffid_Hunter> jnewt: the software shouldn't care what the group is called, only that you're a member of it 07:35 < jnewt> got that covered. 07:36 < Abbott> [R]: what do you mean that's for GPT? 07:36 < [R]> Abbott: you're using gpt 07:36 < [R]> gpt doesnt use 82/83 for linux stuff 07:36 < Triffid_Hunter> jnewt: what happens if you ( cat <&3 & cat >&3 ; kill %%; ) 3<>/dev/ttyUSB0 ? (press ctrl+D to exit this serial terminal) 07:36 < Abbott> [R]: so does it matter what I set it to? or is it just an organizational thing? 07:36 < [R]> Abbott: set it to linux filesystem 07:36 < [R]> yo uknow 07:36 < [R]> common sense... 07:36 < Triffid_Hunter> jnewt: can use stty or anybaud to set terminal options first if you want 07:37 < Abbott> sounds good to me 07:39 < mikhael_k33hl> How do I bind mount the contents of two exported directories to a single directory without them overlapping each other? 07:40 < [R]> overlayfs 07:40 < [R]> unions 07:40 < [R]> unionfs* 07:40 < [R]> aufs 07:41 < jnewt> Triffid_Hunter, ran stty -F /dev/ttyUSB0 9600 cs8 -cstopb -parenb and then ( cat <&3 & cat >&3 ; kill %%; ) 3<>/dev/ttyUSB0 no output, just sits there 07:42 < Triffid_Hunter> jnewt: so your permissions are fine but there's no serial data coming in? 07:43 < Triffid_Hunter> jnewt: dunno what's up with cutecom.. point strace or gdb at it perhaps 07:43 < Mead> alright, I've got a sanity check, say I install a swap out a CPU (changing nothing else, continue using the same motherboard). Linux isn't going to freak out because of a slightly different cpu? 07:43 < jnewt> Triffid_Hunter, that sounds like a correct diagnosis. let me send some serial data in and see if it shows ( I can't work out exactly how those commands are chaining together to do whatever youre tryint to do) 07:44 < [R]> head: linux doesn't care about your cpu 07:44 < [R]> head: unless of course its so differnet its missing instructions, and your system is stupidly compiled to need those instructios 07:45 < Triffid_Hunter> jnewt: just a simple serial terminal.. background cat reads serial and prints to terminal, foreground cat reads from terminal (keyboard) and writes to serial, kill closes the background cat when foreground is finished, and they use bash filehandle redirects to both access the serial port through a single handle 07:45 < Triffid_Hunter> jnewt: if your serial data is mostly text, it's a very simple and quick way to get up and running without mucking around with the weird foibles of screen or gui serial terminals 07:46 < mikhael_k33hl> [R]: thanks mate 07:47 < jnewt> Triffid_Hunter, I'll never remember that. i just don't use that sort of stuff often enough for it to stick. it's easy to just fire up a gui, make my selections and hit open. 07:47 < Triffid_Hunter> jnewt: except when it doesn't work for no reason :P 07:47 < jnewt> Triffid_Hunter, thanks for the explanation though. maybe i'll get a sticky note. 07:47 < Dagmar> ...or you could just install minicom 07:47 < Triffid_Hunter> jnewt: in many of my microcontroller projects I stick that in the makefile as "console" target, so I can just 'make console' and talk to my chip 07:48 < Dagmar> Seeing as how it's the thing we used FOR NEARLY A DECADE to do modeming 07:48 < Dagmar> Forgotten knowledge, it seems 07:48 < Mead> [R]: who's head? 07:48 < [R]> Dagmar: well that's silly 07:48 < Dagmar> [R]: Yeah, why use a program extensively developed for communicating via serial port 07:48 < Triffid_Hunter> Dagmar: I've spent more time fighting with minicom than successfully using it, hence why I worked out how to do it more simply 07:49 < [R]> fighting 07:49 < [R]> rofl 07:49 < [R]> even the silly hardware guys at my work can use it 07:49 < Triffid_Hunter> Dagmar: minicom isn't designed for serial ports, it's designed for modem connections to BBSes.. using it for a plain serial port takes some digging deep in its settings 07:49 < Dagmar> Triffid_Hunter: Dude, I've watched guys too drunk to stand up compromise nationwide ISPs using minicom 07:49 < Dagmar> No one says you have to tell it to dial anything 07:50 < Dagmar> You point it at the serial port, you *maybe* set the baud rate to something other than 9600, and you just start typing 07:50 < Dagmar> If you had problems it's because you probably weren't using a nullmodem cable 07:50 < Mead> [R]: also thanks, CPU's are nearly identical except for clock speed and power, I'm moving from from a 125watt CPU to a 95watt cpu same architecture just clocked slower producing less heat 07:50 < Dagmar> Mead: Less heat with those is very good 07:51 < jcarder_> I feel like such and idiot for doing this. I though I was in another directory and ran 'rm -rf *' when in fact I was in my home dir. Is there any way I can recover all my data? thank you! 07:51 < Dagmar> jcarder_: If you have a second machine around, sure 07:52 < [R]> jcarder_: nope 07:52 < rcf> Dagmar: by default it's setup to dial using a modem. It's not a big deal if you use it regularly, though if you install it every now and again because you really need it *right now* it's an annoyance. 07:52 < Dagmar> jcarder_: Basically, every little write to the filesystem from this point forward stands a good chance of blowing away the things you want to recover 07:53 < jcarder_> ok 07:53 < [R]> rcf: step 1... minicom -D /dev/blah 07:53 < [R]> step 2... profit 07:53 < Dagmar> jcarder_: You can always try and download an ISO with photorec on it and boot it 07:53 < hexnewbie> jcarder_: Is /home a separate filesystem or part of the root filesystem? Like Dagmar says, avoid *any* writing before anything else. If /home is different filesystem, you may unmount it just in case. If it is the same as /, you may need another computer to analyse from. 07:53 < hexnewbie> jcarder_: Er, another OS, not computer. 07:54 < Dagmar> For somewhat obvious reasons, you're going to want to download that and burn it to a thumbdrive or optical disc using a second machine if at all possible 07:54 < jcarder_> hexnewbie: it is on / 07:54 < Dagmar> rcf: It might interest you to know that I've used minicom rather regularly _since_ modems became obsolete, because of all the bloody things with 9-pin DIN serial ports 07:55 < Dagmar> It's not as hard as you guys seem to think. 07:55 < jcarder_> why i accidently ran 'rm rf *' it only took a second which is weird as I have around 10 gigs on /home 07:56 < Dagmar> Heck, IIRC one can jsut tell it to dial anyway and the device on hte other end will hopefully ignore the ATDT command 07:56 < hexnewbie> jcarder_: So, shut it down as soon as possible and boot an OS from *another* media. Before shutting down, though, did you have any important files that you had open? Those may be recoverable and still on disk (even if deleted) before you shutdown. 07:56 < Dagmar> jcarder_: journaling filesystems are awesome like that 07:57 < Dagmar> You might want to consider standardizing on a prompt that has the cwd in it' 07:57 < jcarder_> hexnewbie: ok 07:57 < hexnewbie> jcarder_: In the other OS, you'd need recovery tools, here's a long list: Check here for packages useful for recovery https://packages.debian.org/stretch/forensics-all .. I'd start with photorec, and extundelete. Other than that, ext4magic, ext3grep, scalpel can also find files. 07:57 < Dagmar> This is literally the reason you see so many old hands using bash with user@host:/path$ 07:58 < Dagmar> We got tired of nuking the wrong directory 07:58 < Dagmar> I use a lot of shaded backgrounds for the same reason, actually 07:59 < Dagmar> dev boxes get slightly green backgrounds. test gets cyan. prod gets red. 08:00 < rcf> Dagmar: Oh, it's not bad, and I also use it regularly. I've just encountered a lot of people who don't, and only need it because they've completely hosed some important system and are just discovering that a serial console exists. 08:00 < sauvin> Or better even yet, use a TUI file manager like mc (midnight commander). 08:00 < hexnewbie> Also, using mc instead of rm is very useful. I began using mc again over fear what I might do with rm 08:00 < Dagmar> I made a mistake about six years ago and umounted an in-use filesystem (intermittent writing) because I'd just taken a new position, had too many people bouncing off my desk, and hadn't set up the color-coding on the new workstation yet. 08:00 < hexnewbie> Then I found mc extremely useful for 1001 other reasons. 08:01 < Dagmar> Actually, more like eight 08:04 < Dagmar> hexnewbie: Be careful with that kind of stuff, man. heh 08:04 < ziggylazer> The command ntpdate is not available ona RHEL server. Does anyone know what command should be used to set a NTP server? 08:04 < Dagmar> hexnewbie: I had a contract fall through last week because I was apparently "too cocky" during the interview 08:04 < hexnewbie> If you add obscure stuff like using C-x C-! as GIT frontend as a complement to tig, it really goes close to 1001. :p 08:04 < RayTracer> ziggylazer: configure chrony or ntpd 08:05 < ziggylazer> Yeah I'll go with ntpd 08:05 < ziggylazer> Thanks man 08:05 < rcf> ziggylazer: ntpd -q 08:05 < Dagmar> ziggylazer: Remember, use an *odd* number of sources 08:06 < ziggylazer> Dagmar, why? 08:06 < rcf> ziggylazer: if you want to emulate ntpdate with ntpd, that is. Though running the daemon is a lot nicer. 08:06 < Dagmar> ziggylazer: Because if ntpd sees two servers saying one thing, and two servers saying another (or one and one) it IGNORES the result of the polling 08:07 < ziggylazer> Oh! 08:07 < RayTracer> that pretty sounds like a bug 08:07 < Dagmar> I think the most common thing now is for people to include the local RTC as a source and cite four hosts, but that leads to people doing copypasta and getting just four hosts 08:08 < Dagmar> It's not a bug. The NTP protocol kind of takes obsessive correctness to disturbing levels 08:09 < RayTracer> I think it's a bug (logical). Ignoring information of one source because another source tells wrong doesn't make sense 08:09 < Dagmar> RayTracer: Which one do you believe? 08:10 < RayTracer> either one is better than just running on a local unreliable clock 08:10 < fearless_man> what are other ways to extend or use a different partition inside a different partition, bindmounts and symbolic links didn't work because the program that generates the folder and files is making it a hardlink 08:10 < RayTracer> if in doubt, keep it simple and choose the one with lower stratum 08:10 < Dagmar> ...and if your local clock is _reliable_ as is the case with a lot of machines? 08:10 < RayTracer> you'd never know 08:10 < Dagmar> I don't think I've seen anyone doing a lot of bothering with polling more than one stratum 08:10 < Dagmar> ...nor would that make much sense 08:11 < Dagmar> It's basically explained in the NTP RFCs that it's a deadlock condition 08:11 < sauvin> RayTracer, back in the before times, when sailors could navigate only by the stars and by their clocks, they took only one clock, or *three*. They explained: if they took only two clocks, and they disagreed, which do you believe? 08:11 < Triffid_Hunter> fearless_man: symlinked folders work perfectly for me.. maybe your program is stupid? 08:12 < RayTracer> sauvin: the one they would have taken with them if they only took one 08:12 < ziggylazer> So I configured the NTP server with chrony instead, 08:12 < fearless_man> Triffid_Hunter: yeah maybe, when I used fallalocate and monitor it with df it works 08:12 < sauvin> RayTracer, and which one would that be? 08:12 < RayTracer> sauvin: either one 08:13 < fearless_man> Triffid_Hunter: is there any other way beside bindmounts and symlinks? 08:13 < Dagmar> Oooo 08:13 < sauvin> So you're saying "flip a piece of eight". 08:13 < Dagmar> This reminds me to do something with the GPS dongle I dug out of a box 08:13 < Triffid_Hunter> fearless_man: well those are the ways.. you might be able to emulate a bindmount via fuse if you're feeling masochistic, but you'll be much better off making normal methods work 08:14 < fearless_man> Triffid_Hunter: then should I just create a cron tab and move those folders/files to a different partition? I'm out of ideas 08:14 < RayTracer> if you can't handle two dates, like getting the mean, throw one away and you have ended the confusion. Besides, you then strongly believe that the remaining one is correct.. 08:15 < fearless_man> Triffid_Hunter: I mean cron job 08:15 < Dagmar> If you take the mean then you are *knowingly* setting the clock incorrectly 08:15 < sauvin> No, you wind up pulling into port about three weeks after you starved to death. 08:15 < Triffid_Hunter> fearless_man: what exactly are you trying to achieve? 08:16 < sauvin> If you have ONE clock, it's either off or it's not, but you have to believe it. If you have THREE clocks, odds are that two of them will always agree. 08:16 < Dagmar> The expectation with NTP is that you are *going* to contact a majority of servers with The Absolutely Correct Time 08:16 < Dagmar> It's not about just setting the clock _close_ to the right time 08:16 < RayTracer> the kernel does that with adjtime anyways 08:16 < sauvin> For me, it's all about making sure my cell phones and my computer report the same time. 08:17 < fearless_man> Triffid_Hunter: well, A program keeps pilling up folders and log files and it's making the partition full, so I need a way to either create a bindmount or a symbolic link to a different partition 08:17 < Abbott> where will grub look for my grub.cfg first? I am added a second linux partition and want my machine to boot to that instead of what it currently boots to, but both partitions have /boot/grub/grub.cfg on them 08:17 < sauvin> "adjtime"? Forgive an old man for being a bit skeptical: I can remember when a PC could drift whole minutes within a 24 hour period. 08:17 < Dagmar> RayTracer: yeah, but it won't do anything with that without an ntpd passing it notes about how many cycles to skip 08:18 < Triffid_Hunter> fearless_man: that's literally what logrotate is for 08:18 < Dagmar> Triffid_Hunter++ 08:18 < fearless_man> Triffid_Hunter: even for folders? 08:18 < Dagmar> logrotate can do pretty much whatever you need done 08:18 < fearless_man> Triffid_Hunter: those log files are huge > 2 gig 08:18 < nekoseam> imagine a world without GIMP 08:19 < nekoseam> i see riots 08:19 < nekoseam> political corruption 08:19 < Dagmar> All looking like Ms Paint Adventures? 08:19 < sauvin> Without GIMP, my desktop background would not exist. 08:19 < fearless_man> Dagmar: Can I just use cron and move them to a different partition? 08:19 < Dagmar> fearless_man: *cough* That's what you use to trigger logrotate with 08:19 < Dagmar> It's fairly standard in every distribution 08:20 < fearless_man> Dagmar: can I just simply run mv -vrf source directory 08:20 < fearless_man> rather than using logratate? 08:20 < Dagmar> fearless_man: logrotate will be easier and more reliable 08:20 < fearless_man> I'll read the docs about logrotate 08:21 < nicholasBPM> I run ifconfig and got this for eth0 RX bytes:139880680082 (139.8 GB) TX bytes:1995626487 (1.9 GB)? I am just running a simple webserver with static pages.. only port 80,443 and ssh open any advice what could be wrong? 08:21 < ziggylazer> So how hard is it to configure rsyslog? 08:21 < Triffid_Hunter> fearless_man: thing is, unless you get the process generating the logs to close and reopen the log, simply moving them won't actually free any space 08:21 < Dagmar> Not from just that, other than to surmise no one wants to get the content 08:22 < Triffid_Hunter> fearless_man: most daemons that generate logs will accept a SIGHUP to reopen log files, and logrotate scripts basically always hup the daemon during log rotation 08:22 < Dagmar> nicholasBPM: Actually look at the webserver's logs 08:22 < ziggylazer> Got a Nagios logserver that I want my RHEL server to send logs to... But do I really want to send them unecrypted or should I create a VPN just for that? 08:22 < jim> nicholasBPM, about ifconfig... it's old, and the makers acknoledge that by deprecating it (which just means, it's here now, go ahead and use it, but it's in the process of going away) 08:22 < Triffid_Hunter> fearless_man: also, I've never encountered anything that had a problem with its log folder being a symlink to somewhere else 08:23 < nicholasBPM> Dagmar, thanks anything special to look for? 08:23 < ziggylazer> Sending unencrypted logs from AWS to another LAN seems dumb? 08:23 < jim> nicholasBPM, the more modern version is the program ip 08:23 < Dagmar> nicholasBPM: That perhaps you don't just have a zillion people doing a HEAD / and walking away 08:23 < Dagmar> ziggylazer: Well, ssltunnel is pretty easy, and there's always ssh as well 08:24 < nicholasBPM> jim, thank you will will look at ip 08:24 < ziggylazer> jim, checked out the HS we talked about. Nothing near me, witch is a shame. They seem to be doing some really cool stuff in that community 08:24 < nicholasBPM> Dagmar, thanks will analyze what my visitors are up to. 08:25 < Dagmar> nicholasBPM: Try not to look too closely at anything from Baidu 08:25 < Dagmar> It'll just make you angry 08:25 < nicholasBPM> Dagmar, evil search engine? 08:26 < Dagmar> In a word, "yes". 08:27 < Dagmar> Take a sample of URIs and sort them by frequency. Both the bottom and the top of the list will be interesting places to start 08:30 < nicholasBPM> Dagmar, thank you, i can see now what has been going on, multiple bruteforce connections to ssh and lots of connections to https from Asia 08:30 < Dagmar> lol 08:30 < alexandre9099__> hi, can sslh (a port multiplexer) pass the ip that connected to my server to the server applications like apache or ssh? 08:31 < alexandre9099__> sslh always says that i got a connection from ::1 :) 08:31 < Dagmar> nicholasBPM: Write this down: "okean.com" 08:31 < Dagmar> heh 08:31 < Dagmar> nicholasBPM: Simplest way to deal with the brute force attacks is to blacklist on that port with a broad brush 08:32 < nicholasBPM> Dagmar, thank you, I will do that. 08:33 < Triffid_Hunter> alexandre9099__: https://blog.cloudflare.com/how-we-built-spectrum/ may interest you 08:33 < Dagmar> Okean has a very nice and rather comprehensive list of netblocks that make it easy to just blacklist most of the planet by country. ;) 08:33 < jim> nicholasBPM, ip is kinda different, there's a lot to get used to 08:34 < nicholasBPM> Dagmar, over 100 GB, my server must be very interesting.. or maybe bandwidth is cheap ;) 08:34 < sauvin> I didn't do that. I just moved my SSH ports. 08:34 < Dagmar> There are, at *least* a dozen variants of worms that use brute force ssh attacks 08:35 < ziggylazer> Got my SSH on an obscure port as well 08:35 < Dagmar> Just having an IP address will catch around 2,000 attempts a day 08:35 < ziggylazer> And always use a more security then just a password 08:35 < sauvin> Funny, my logs are clear. 08:36 < nicholasBPM> jim, i agree but it looks neat 08:36 < nicholasBPM> Dagmar, i am happy i do not pay per GB 08:36 < ziggylazer> rsyslog anyone? Worth setting it up? Even if the logs are unecrypted? 08:36 < ziggylazer> nicholasBPM, 1GB/s Fiber 08:37 < Dagmar> ziggylazer: It doesn't *have* to be unencrypted 08:37 < nicholasBPM> ziggylazer, that would be awesome 08:37 < ziggylazer> Dagmar, Nah Guess I could send it over a tunnel 08:37 < Dagmar> It does tls 08:38 < ziggylazer> Might be an option then :) 08:38 < Dagmar> It's been an option for a decade now 08:39 < Dagmar> So... pretty well documented space, that. ;) 08:39 < ziggylazer> Hahaha 08:39 < ziggylazer> Sorry 08:39 < Dagmar> No worries 08:39 < ziggylazer> No coffee yet. Lazy... 08:40 < jim> nicholasBPM, the latest (but so far unreleased (I think)) version of iproute2 (which contains the ip program) has json output for ip addr and ip route at least 08:40 < graff> i'd lke some pre-rolled musl based distributions 08:41 < michael2> hi, Im at a syslinux prompt "boot: " does anyone know how to start a kernel with a bash shell from here? 08:41 < Dagmar> Someone should patch in protobuf support 08:41 < Dagmar> heh 08:41 < graff> I have alpine x86 and x86_64 installed 08:41 < graff> but I am having trouble tracking down much else that can be easily installed an booted up 08:41 < Dagmar> michael2: Maybe tab to get the list of images available 08:41 < nicholasBPM> jim, cool 08:41 < graff> I am not a gentoo user, so the idea of sitting down to try and make a gentoo-musl VM for 12 hours sounds terrible 08:41 < Dagmar> Most people have the good sense to at least leave a list on page f21 08:41 < Dagmar> or f2 08:42 < michael2> Dagmar: just reprints the "boot: _ " prompt 08:42 < Dagmar> No joy from f2? 08:42 < jim> I'm looking for a json command line tool, just a sec (I had it last week) 08:42 < michael2> Dagmar: did you mean me? 08:42 < jim> kq? jq? kj? 08:43 < Dagmar> michael2: yes 08:43 < michael2> F2 does nothing, what should it have done? 08:43 < Dagmar> michael2: Well, if someone had set that up to actually be used that way, it would have been where they'd have left some instructions 08:44 < jim> nicholasBPM, the json command line tool is jq, and essentially you pipe any json output into it, and using the arguments to jq, you specify what you want 08:44 < jim> no args to jq would cause it to prettyprint the input it gets 08:44 < michael2> Dagmar: ah, now Im with you, I set it up - and I didn't set up any greetings or help etc... 08:46 < Dagmar> You're kind of supposed to set up the labels _before_ this point 08:46 < nicholasBPM> jim, thats what i love about Linux so may options and variations and together with python you can do almost anything without being an expert 08:47 < michael2> yeah, I was hoping I could just start a kernel, get into a shell and create the labels etc... 08:47 < nekoseam> :-) 08:48 < Dagmar> Heheh. Whoops 08:53 < kuri0> how can I make bless hex editor load the whole file ? 09:00 < jnewt> i can't figure out how to get out of minicom. ctrl+a then z doesn't get me a menu or out. just going by ubuntu forum and man page for minicom. 09:03 < myxenovia> hi 09:04 < myxenovia> i installed realvnc server in centos 09:04 < myxenovia> but when im connecting to it it says connection was refused 09:04 < myxenovia> can u help me 09:05 < _abc_> Hello. I have multiple volumes which get mounted or not, and updatedb runs when all are mounted, then locate behaves as if I'd issue locate -e, not finding things on not (yet) mounted volumes. How do I make locate behave normally, showing db contents instead of -e ? 09:06 < well_laid_lawn> myxenovia: https://wiki.centos.org/HowTos/VNC-Server 09:09 < myxenovia> well i forgot the port number lol 09:09 < myxenovia> thanks well_laid_lawn 09:09 < well_laid_lawn> cheers 09:12 < myxenovia> i have question again 09:12 < myxenovia> how can i drag file from vnc to my desktop? 09:12 < myxenovia> and same the other way around 09:13 < _abc_> myxenovia: iirc there are add ons and plugins for that, but I may remember that from vbox 09:15 < myxenovia> _abc_ i see i will look for that. i just made realvnc work but it looks like being run on a virtual box 09:16 < _abc_> myxenovia: what server are you using? 09:16 < myxenovia> _abc_ centos 7 09:16 < myxenovia> _abc_ i mean the things that i do through vnc doesnt take effect in my server 09:17 < _abc_> myxenovia: thre are several vnc x-servers to use, at least one acts as an X11 window client. In theory, if all is well set up, copy and paste or drag and drop from such should put a usable network url file access path in the X11 clipboard. 09:17 < _abc_> Of course not myxenovia 09:17 < myxenovia> i see ill try to look at it. i will try it 09:18 < myxenovia> _abc_ why? the reason i installed vnc , is to do work remotely 09:18 < _abc_> myxenovia: only way you're going to see files in the vnc client in your / other machine(s) is by exporting network paths from the vnc box. 09:18 < _abc_> Maybe I missed on what you did? 09:19 < _abc_> You installed realvnc x-server on the remote centos machine, and access it with a vnc client from somewhere else, right? 09:19 < _abc_> Is this what you did? 09:19 < myxenovia> _abc_ yes 09:19 < _abc_> So what changes do you not see on the server? 09:21 < myxenovia> _abc_ i run a service in the server through realvnc but doesnt look like the service is visible to other pc 09:21 < myxenovia> this is what i did https://www.tecmint.com/install-and-configure-vnc-server-in-centos-7/ 09:21 < _abc_> myxenovia: 'visible to other pc' means, on the lan near the vnc server machine, or on the internet? 09:22 < myxenovia> well ill explain again i think u misunderstood 09:22 < myxenovia> im able to connect to my server through realvnc 09:22 < _abc_> The service you run is a 'server' in the internet sense, right? Like ircd or httpd? 09:22 < myxenovia> and after that, i run a 3rd application in the pc, and doesnt look like my pc server is affected 09:22 < myxenovia> yea 09:23 < _abc_> And the 'pc server' is your client, remote from the lan where the realvnc runs? 09:23 < tobylane> Is there a way to interrupt a multi command bash line at the end of the current command? 09:23 < _abc_> You need to forward ports in the firewall(s) 09:23 < myxenovia> yea 09:23 < _abc_> myxenovia: ^ 09:23 < myxenovia> _abc_ why? im already able to connect to my server using realvnc 09:24 < _abc_> myxenovia: your service needs it's own port forwarding. 09:24 < _abc_> For each service and port range. 09:24 < myxenovia> how can i do it 09:25 < _abc_> Read a port forwarding howto, access your rhel server and set up open ports as needed for the service, then each router between that server and the internet, set up the same way. 09:26 < myxenovia> _abc_ is there a way to figure out my port in centos? using realvnc 09:27 < _abc_> myxenovia: netstat -tua in a terminal 09:30 < prillian5> is there a cool tool which allows me a screencapture in timelaps? 09:30 < prillian5> I'd like to create a documentation about a full webpage installation-process. 09:30 < _abc_> prillian5: yes, x11 presentation tools, there are several 09:31 < prillian5> _abc_: can you name some, or point me to a list 09:31 < _abc_> google 09:32 < prillian5> I had, but find only default recorders like Kazam, SimpleScreenRecorder and other. 09:33 < prillian5> don't see how to get an timelaps screencapture with it 09:33 < prillian5> I would prefer to have one, who is lightweight with ressources 09:34 < prillian5> and the videofiles shouldn't be to big, because I wan't to record surveral hours 09:35 < prillian5> I really would apprecatet a ponited recommendation 09:38 < supernovah> hey with useradd, I specified a sha-512 string for --password, but I can't log in as the user, I also specified --system, does that prevent me logging into it? 09:39 < supernovah> format was $6$$ 09:40 < supernovah> the entry in shadow doesn't match the entry I gave 09:43 < supernovah> ah nvm, used double quotes and the $'s messed it up 09:52 < screwsss> hey 09:53 < megaxlr> hello 09:53 < screwsss> megaxlr: do you use kali linux 09:53 < megaxlr> No. I use Debian 09:55 < _abc_> https://itsfoss.com/best-linux-screen-recorders/ prillian5 seriously, go do your homework 09:56 < prillian5> I was exactly on this Page ^^ 09:56 < _abc_> Tough luck, then 09:56 < prillian5> At the moment, I try obs 09:57 < megaxlr> prillian5 from my experience, OBS has been the nicest. on Linux and Windows 09:59 < hehehe> https://www.ovh.co.uk/subscriptions-hubic-ended/ 09:59 < hehehe> anything similar? 10:20 < pankaj_> How to find the path of a module in linux? 10:25 < _0x40_> pankaj_: modinfo 10:35 < _abc_> So, any news about how locate/mlocate can be made to behave as-if not launched with option -e ? 10:36 < Pantsu> -e is not the default for me 10:37 < _abc_> Pantsu: try to index a mounted usb stick after editing /etc/updatedb.conf to permit indexing it, umount it, locate something on the stick. Then mount it, locate again. 1st time it won't show it, 2nd it will 10:37 < Pantsu> easier to just touch and rm to test :p 10:38 < _abc_> Pantsu: that does not work for locate 10:38 < _abc_> It has to be on another fs, for this bug. 10:38 < pankaj_> Their is a lot of resource on module handling on net but I want a standard one so that I can practice it practically and see the effect on my system and not just play with those install on my system. I have to also solve issues like about device driver for printer when I connect the printer to my laptop. 10:39 < Pantsu> https://bpaste.net/show/59810a598825 10:39 < Pantsu> ah 10:39 < Pantsu> report it then 10:41 < pankaj_> _0x40_: I just want to learn more about device drivers and module handling in linux. On internet their are variety of big and small as well as many differing tutorial but I want to practice it practically so that I can see the effects of it as well as load external ones. Please help if their is any guide or how to practice and learn about this stuff. 10:48 < Pantsu> pankaj_: kernelnewbies.org 10:48 < Pantsu> pankaj_: and the Documentation/ tree 10:49 < Pantsu> greg have a couple of good talks on yt too about writing your first kernel module 10:55 < catmando> that was odd 11:10 < Celmor> how do I copy a file between users' home directory without having to cache it or create a temporary file? 11:11 < Celmor> I'm using su to authenticate as the other user 11:11 < _abc_> Pantsu: I sort of did, on devuan... waiting for more discussions 11:11 < _abc_> Also I could strace it but not right now 11:14 < TJ-> Celmor: pipe it into su; something like "cat /home/$USER/somefile | su -lc 'cat >/home/someotheruser/somefile' someotheruser " 11:15 < _0x40_> TJ-: Why do you need to cat it? Can't you send the file to su directly with < ? 11:17 < screwsss> anyone here on linux 11:18 < Armand> No. 11:18 < _0x40_> screwsss: Doubt it. 11:18 < Celmor> TJ-, doesn't work, it seems su is reading the password from stdin 11:18 < Armand> screwsss: Welcome to ##sarcasm 11:18 < Armand> <3 11:19 < rukusza> I want to clean and redor on of my laptops, mint still the .. okay distro for general use ? (using mint with mate desktop) 11:19 < rukusza> redo* 11:19 < screwsss> empty channel Armand 11:20 < Armand> lol 11:20 < rukusza> screwsss: I'm representing a real person, I'm a bot, so no I'm not here. 11:27 < CustosLimen> hi 11:28 < CustosLimen> I'm looking for linux distro to put on my mom's laptop - I really don't like ubuntu/debian based though - and would like something with very reliable upgrades 11:30 < TJ-> Celmor: seems like it's easier with sudo :) cat somefile | sudo -H -u someotheruser sh -c 'cat > $HOME/test.txt' 11:30 < Armand> CustosLimen: Mint 11:30 < Celmor> TJ-, my current users isn't allowed to run sudo command 11:30 < Celmor> s 11:30 < TJ-> Celmor: you'll have to experiment then :) 11:30 < Celmor> CustosLimen, any issue with ubuntu derivatives? 11:30 < CustosLimen> Celmor, would prefer not 11:30 < Triffid_Hunter> Celmor: su - otheruser; cp -v /home/firstuser/file ./ ? 11:31 < Celmor> otheruser can't access /home/firstuser contents 11:31 < megaxlr> ArmandMint is basically Ubuntu. >I really dont like ubuntu. 11:31 < CustosLimen> Celmor, IMO ubuntu makes rather flimsy packages and I am afraid this will just translate to derivatives 11:31 < Armand> megaxlr: I know this 11:32 < Armand> But that conflicts with the latter remark. 11:33 < Triffid_Hunter> Celmor: why not? home dirs are usually 755 unless you change them 11:33 < Triffid_Hunter> Celmor: how about scp file otheruser@localhost:. ? 11:34 < Celmor> otheruser is not supposed to be accessible via ssh since that one can 'sudo' as root, firstuser is only supposed to be accessible via ssh 11:35 < Triffid_Hunter> Celmor: ok so you've gone to some effort to isolate the two users from each other, but now you've realised that it has worked and that's a problem? 11:36 < Triffid_Hunter> Celmor: user1: nc -l -p 1536 | tar -xv user2: tar -cv file | nc -p 1536 :P 11:36 < Celmor> Triffid_Hunter, they are not 11:36 < Celmor> 755 11:36 < TJ-> Celmor: this works for me: U=test1; su $U -c "cat /tmp/mce.log > /home/$U/test.txt" 11:37 < Triffid_Hunter> Celmor: then can copy 11:38 < Celmor> TJs, firstuser doesn't have a password, only ssh public keys for authentication 11:38 < Celmor> so I can't be seconduser and authenticate as firstuser 11:39 < Celmor> Triffid_Hunter, nc solution is worse than I came up with, using fifo's 11:40 < Celmor> thanks anyway 11:41 < TJ-> Celmor: if you've isolated them to prevent these things then you either need to relax the conditions or use some kind of mediator service 11:42 < Celmor> well, I wanted the "mediator service to be bash using pipes to directly stream data from one place to the other 11:45 < Celmor> as it is I'm stuck with something like `mkfifo /dev/shm/fifo && cat "$file" > /dev/shm/fifo & su otheruser -c 'cat < /dev/shm/fifo > "$file"' && rm /dev/shm/fifo` at best 11:48 < TJ-> Celmor: could you use an ftp server listening only on localhost? Then a CLI client command would do it 11:49 < Celmor> I'm trying to make it easier instead of harder 11:51 < vindicat0r_> hi 11:51 < Celmor> hi 11:56 < Celmor> CustosLimen, try https://elementary.io/ 12:07 < BluesKaj> Howdy folks 12:09 < djph> o/ 12:10 < BluesKaj> \o 12:13 < plexigras> when using `comm -3` is there a way to only print lines in file1 and not in file2 but not vice versa? 12:21 < plexigras> nvm 12:24 < zetheroo> is there any way to get the exitcode of a script printed at the very top of the script output? 12:25 < Celmor> why is the exit code printed? what kind of script? 12:26 < megaxlr> `$?` contains the last exit code if that's what you want 12:26 < Sitri> Except it isn't quite that 12:27 < Sitri> The only way to do it is to cache all the script's output, then print the exit code, then print the cached output. 12:31 < jim> zetheroo, usually the script would exit before you get the exit code, if it's a script you are writing, you would arrange to have the exit code tell you something... what they do in a lot of scripts, is have the exit code based on whether or not there was a problem running the script, and to some degree of detail, what that was 12:32 < zetheroo> jim: atm we have the exitcode echo'ed at the end of the script, but for monitoring purposes we need that code at the top of the script 12:34 < jim> zetheroo, so you need the exit code at the top of its own script? (note this could be a conflict of design: you're asking for the exit code before it's calculated 12:34 < jim> ) 12:34 < zetheroo> jim: no, just that on the output of the script - that it's printed at the top of the output 12:35 < zetheroo> I guess the output of the script would have to be parsed by something and then the exitcode sent to the top of the final output 12:35 < zetheroo> Just was wondering if anyone had experience with this 12:36 < jim> so you'd have a script that calls the script in question? 12:36 < jim> what language (alternatively, which interpreter) does the script use 12:36 < jim> ? 12:37 < Sitri> zetheroo: | tac | (read line; echo $line; tac) 12:38 < jim> Ch35TY, hi... are you having problems with your connection? 12:39 < jim> zetheroo, what language (alternatively, which interpreter) does the script use? 12:39 < zetheroo> jim: bash script afaik 12:42 < jim> ok... say the script in question (we'll call it the inner script) prints something, and that's how it returns data to your monitoring script, or outer script... the outer script can get what the inner script prints, like this: the_output=$(inner) 12:42 < jim> now, how to get the exit code, 12:43 < jim> I think maybe you can do this: 12:43 < jim> the_output=$(exitcode=inner) 12:44 < jim> but I'm not sure that will work 12:45 < jim> bash will put the exit code in a variable, I think it's $? 12:46 < jim> I actually have to leave at this point... getting cold :) 12:46 < jim> Ch35TY, hi... are you having problems with your connection? 12:46 < zetheroo> jim, ok thanks 12:47 < jim> you could try ##bash :) 12:47 < jim> but be careful, they;re notorious :) 12:50 < Xatenev> anybody using ripgrep? :p 12:50 < pankaj_> How do I find that which driver is loaded for each of my device and know its location? 12:57 < djph> modprobe, I think ... ? 12:59 < n3xuz> I need to remove my swap partition to resize my disk. Do I just turn it off with swapoff and remove it from fstab? 12:59 < Pantsu> define "location" 12:59 < Pantsu> n3xuz: sure 12:59 < n3xuz> Pantsu: thanks 13:09 < nevodka> hmm I wanted to add ssl to my transmission daemon that is internet facing 13:09 < nevodka> so I created an nginx config to listen on 8080 with ssl and created a new certificate with openssl and pointed towards it in the config 13:10 < nevodka> and under location did a proxy_pass to the transmission daemons original port 13:10 < nevodka> and restarted nginx and it does indeed let you connect via the 8080 port, but shows it as not using ssl still 13:10 < nevodka> ;-; 13:11 < nevodka> chrome says the certificate is invalid 13:11 < nevodka> does it take some time to propagate or something 13:13 < Pantsu> nevodka: self signed? 13:13 < revel> nevodka: Did you make a self-signed cert? 13:13 < Pantsu> if so you have to add it into your cert storage, and preferable also pin it 13:13 < nevodka> I just generated it as so: openssl req -x509 -nodes -new -keyout /etc/nginx/trans.key -out /etc/nginx/trans.crt 13:13 < Pantsu> right, see my last line then 13:13 < revel> Use Letsencrypt. 13:14 < Pantsu> eh, depends on the use case 13:14 < Pantsu> if this is going to be public, sure 13:14 < revel> I have a Letsencrypt cert on my bouncer. 13:14 < Pantsu> if this is just for your use then pinned self signed is fine 13:14 < nevodka> thanks, will look up what that means ;) 13:15 < revel> And I don't want to (or can't) add the cert to the local cert storage on every single device I want to connect from. 13:16 < nevodka> can you expand on what cert storage is Pantsu? do you mean i need to copy the .crt to /etc/ssl/certs or so? 13:16 < Pantsu> you distro should have documentation on how to correctly add the certificate to the storage 13:16 < Pantsu> and as I said, for self signed you want to pin them too 13:19 < wyoung> hi gang 13:24 < nevodka> hmm 13:24 < nevodka> i have somewhat of a strange setup actually 13:25 < nevodka> im running a shared vpn that i have a dynamic dns for, and my domain name is a CNAME to that 13:26 < djph> and ... ? 13:27 < nevodka> do i need to register to a certificate authority like CAcert? because they seem to require you have an email hosted at the domain 13:27 < djph> well, you don't *need* to use a CA-backed certificate ... but that's going to be trusted a lot more than self-signed. 13:29 < revel> nevodka: Letsencrypt doesn't require that. 13:29 < nevodka> oh really 13:30 < revel> Yes. 13:31 < JimBuntu> Let's Encrypt IS a Cert Authority 13:31 < revel> Did anyone claim otherwise...? 13:31 < djph> that they are 13:31 < Armand> Indeed, they are very much a CA 13:31 < djph> although, perhaps JimBuntu is speaking to my comment that one is not necessary ... ? 13:31 < Armand> lol 13:33 < JimBuntu> I was saying that to explain why a user of LetsEncrypt doesn't have to register elsewhere 13:33 < varliberate> lol I've been using mac for two years now, linux as a desktop since the 90s though, was just looking at my thunderbird settings (yeah I'm an old bofh) and it still had "gnome extensions" in the add on section 13:34 < varliberate> must have come in with my profile settings - was disabled (of course) 13:37 < novik> Can I see the data that is written to an open file descriptor of a process? Something to the effect of tail -f /proc/pid/fd/1? 13:39 < wyoung> hey gang 13:40 < varliberate> hey 13:40 < wyoung> varliberate: werd up? 13:40 < varliberate> ah just doing some digital housekeeping, you? 13:40 < wyoung> varliberate: watching Netflix. 13:41 < nevodka> so I failed the 'http challenge' during the letsencrypt setup 13:41 < wyoung> varliberate: Santa Clarita diet <3 13:41 < JimBuntu> nom nom, bone... nom nom, bone 13:41 < wyoung> JimBuntu: :) 13:41 < varliberate> :) 13:41 < nevodka> I don't have any page on port 80 13:42 < wyoung> nevodka: ok, I believe yuo 13:42 < varliberate> can I ask a quick n00b question (I haven't tried it myself yet) --- but you need the ACME client on your apache server for LE, is that correct? 13:42 < wyoung> varliberate: shoot 13:43 < varliberate> LE == Let's Encrypt btw 13:43 < varliberate> ^^ 13:43 < wyoung> varliberate: sudo apt-get install letsencrypt 13:43 < nevodka> can I not tell letsencrypt to check on a specific port? 13:43 < nevodka> it just asked for my domain name and then boop complained 13:44 < varliberate> that installs it yeah, but that is installing some version of what they call the ACME client, am I right in this? 13:44 < nevodka> the fact that my vpn is shared they don't allow you to use the conventional ports such as 80 13:44 < wyoung> varliberate: letsencrypt certonly --nginx -d yhg1s.com -d asshat.yhg1s.com 13:44 < JimBuntu> lol 13:45 < wyoung> JimBuntu: yhg1s is the worst. 13:45 < edd_lc> novik: A little offensive of an approach to the problem but might be helpful (not mine, btw). http://www.korznikov.com/ 13:45 < nevodka> I tried with the port included, no luck 13:45 < nevodka> guess i'll do a self signed certificate..? 13:45 < varliberate> or get out of the vpn long enough to do a LE 13:46 < varliberate> once the cert is signed you can go back into your tunnel 13:46 < varliberate> if that's an option 13:46 < BluesKaj> nevodka, vpns don't normally use "conventional ports" , mine uses one in 4 figures 13:47 < varliberate> wyoung, my question was on architecture, not usage ;) 13:47 < nevodka> hmm i suppose i could do that 13:48 < nevodka> but might it be easier to just do a self signed cert? 13:48 < wyoung> varliberate: k 13:48 < nevodka> this page is really only in use by me anyways 13:48 < varliberate> nevodka, if it's only for you and people you can share your private CA cert with totally 13:48 < JimBuntu> nevodka, but... can it be found by others... that's the question. 13:48 < varliberate> AWS do that for free BTW 13:48 < varliberate> no 13:48 < BluesKaj> nev depends on your vpn client's settings setup by the vpn service, usually 13:49 < BluesKaj> nevodka,^ 13:50 < altendky> I am trying to setup a bitnami phabricator container to accept ssh connections to the Phabricator Git repositories. It's a mess but I've gotten it to authenticate and now the client gets' `/usr/bin/env: php: No such file or directory`. Where is this coming from and how should I fix the PATH to find PHP? https://gist.github.com/altendky/a31cc669d7c4d9948caa83636a2bc817 13:50 < nevodka> so is the consequence of a self signed cert that any computer i'm on that does not have the private cert not use ssl? 13:50 < gunix> hey tdjb_ 13:50 < nevodka> i don't really understand the mechanics 13:50 < gunix> wrong nick, sorry 13:50 < brutser> for some reason when i build live cd with mate, caja is showing /dev/loop* and /dev/mapper/* all under Places from the top menu. if you try to mount them, they all appear on desktop, like normal hdd or usb - this does not happen when building with gnome and nautilus - how can i hide those when building with mate/caja? 13:50 < JimBuntu> altendky, do you have PHP installed? 13:50 < gunix> hey tds ... remember our chat about "ip rule" last week ? 13:50 < brutser> i tried udev rules, but no success yet 13:51 < gunix> i found another info: it seems some ASIC have a limitation to how many firewall rules they can get and ip rule can bypass that 13:51 < nevodka> or the self signed cert just means my server is doing the encryption and ssl will work from any client 13:52 < altendky> JimBuntu: yes, it was part of the bitnami phabricator container. it is at /opt/bitnami/php/bin/php. 13:52 < Pantsu> self signed just means that the one who created it signed it themself 13:53 < Pantsu> instead of having someone else sign it 13:53 < JimBuntu> altendky, then I would add that location (less the php at the end) to the path... 13:53 < Pantsu> you are basically saying "this is totally legit, I signed it myself" vs having someone (maybe) more trustworthy say "we checked the dude out and he is legit" 13:53 < nekoseam> Fun fact: xfce's window manager (xfwm) has roots from twm. (twm -> fvwm -> xfwm) 13:54 < altendky> JimBuntu: sure, but where do i configure the path being used by sshd at that point? 13:56 < JimBuntu> altendky, you are doing the 'git clone' locally, right? 13:56 < altendky> JimBuntu: from outside the container i am running git clone and connecting to the container. the container has my port 22 13:56 < altendky> my 22 is forwarded to the container 13:56 < brutser> can someone perhaps help me with udev rules? 13:57 < altendky> JimBuntu: if i open a docker console directly as the `git` user `which php` works fine 13:57 < varliberate> nevodka, you could tunnel port 80 through ssh to some outside machine too - this is all starting to get a bit crazy though ;) 13:57 < wyoung> I have all of this built up anger talking to Yhg1s. Any one want to help me? 13:57 < JimBuntu> altendky, I am unsure, this output looks odd. Is the error actually happening while you are cloning the repo? If so, then that's the system that needs to know where to find PHP 13:57 < wyoung> JimBuntu: use python 13:58 < wyoung> altendky: ^ 13:58 < altendky> wyoung: sure, i'll run phabricator with python. :] 13:58 < varliberate> nevodka, but if you plan it out right and not just hack away at it it's possible for sure 13:58 < nevodka> varliberate my domain points to a shared vpn that prohibits assigning anything to ports under 2000 or so 13:58 < nevodka> i'm just going to self sign the certificate 13:59 < altendky> JimBuntu: hmm... i'll install php locally and see. why would a git clone in this case trigger a need for php? hooks defined in the repo? phab is reserving a github repo that i can clone directly just fine. 13:59 < wyoung> altendky: nah, port your code bro / miss / bi-curitous 13:59 < varliberate> nevodka, this is essential reading imho https://www.davidpashley.com/articles/becoming-a-x-509-certificate-authority/ 13:59 < altendky> wyoung: phabricator is a large forge server. not gonna port it regardless of how much i would like it to be python. 14:00 < wyoung> altendky: :( you make me sad 14:00 < JimBuntu> altendky, That is the part I didn't understand. I have not had a repo trigger PHP before, or if it did... I didn't run into an issue as it was already installed... fairly certain I have not had one trigger before though. 14:00 < wyoung> altendky: what about java or scala? Both actual languaes. 14:00 < wyoung> languages* 14:01 < wyoung> PHP is what you give to homeless people 14:01 < altendky> wyoung: you don't seem to understand. phabricator is not a little utility you port really quick. have you heard of it? 14:01 < wyoung> altendky: It sounds terrible. 14:01 < altendky> wyoung: i'm just trying to enable ssh cloning from it (which it supports, just not configured in the bitnami docker container) 14:01 < wyoung> altendky: just do it 14:01 < nevodka> thanks for the link varliberate 14:01 < varliberate> nevodka, seriously, that's my jedi master tip for the night, once you read that you'll have the real skills hey 14:02 < varliberate> np my pleasure 14:02 < varliberate> thank the guy that wrote it hey, perfect article 14:03 < altendky> JimBuntu: i can repaste but at the no such file or directory is still there. running which php after the failed git clone command prints /usr/bin/php now (i did not have php before) 14:05 < JimBuntu> interesting. I imagine phabricator has info about this error. I'm not sure why it's happening. 14:07 < altendky> JimBuntu: ideas about how to control the PATH used by sshd for this stuff? in an effort to hack on and find other issues i made a wrapper for the previous script that yielded this error by adding php to the path in the wrapper. i'm not sure what script is getting called now though. 14:08 < tds> gunix: ah, that makes sense 14:08 < wyoung> indeed 14:08 < JimBuntu> altendky, I would expect that you can add the PATH in the SSH command line. This process is odd overall though. 14:09 < tds> gunix: I seem to remember some of the cumulus stuff I'd seen had support for hardware accelerated netfilter rules with some custom things added to iptables, I don't remember the exact details though 14:09 < altendky> JimBuntu: yes, my attempt to make it work is horribly hacky. i would fully expect to tidy it up once i have a working system. 14:10 < tds> ah, apparently only certain chains are accelerated 14:14 < bitSt0rm> Can anyone tell me why I'm unable to run: i8kctl fan x x , mapped to a key in Openbox? xterm mapped likewise works fine.. Screenshot: https://ibb.co/bH45Xd 14:14 < evanesoteric> Anyone use password-store - pass-tomb particularly? I can't seem to get it to tomb setup properly. Always "Error: Tomb is not present." 14:15 < bitSt0rm> (and i should say, Im able to run the i8k.. command from terminal fine) 14:21 < brutser> for some reason when i build live cd with mate, caja is showing /dev/loop* and /dev/mapper/* all under Places from the top menu. if you try to mount them, they all appear on desktop, like normal hdd or usb - this does not happen when building with gnome and nautilus - how can i hide those when building with mate/caja? 14:21 < brutser> i tried udev rules, but no success 14:23 < altendky> JimBuntu: thanks. got one step further. `echo 'PATH=/opt/bitnami/php/bin:$PATH' >> ~git/.ssh/environment` and `echo 'PermitUserEnvironment yes' >> /etc/ssh/sshd_config.phabricator` move me on to it not finding sudo 14:25 < altendky> JimBuntu: and got it cloned... thanks for talking through this with me. now to figure out how to actually set this up properly. 14:27 < ankk> hi. 14:27 < ankk> is there a way to log accessed files on ftp server? 14:27 < jack_rip_vim> hi ankk 14:27 < ankk> i want to see which ip downloaded which file 14:28 < jack_rip_vim> annk, check your ftp conf file 14:28 < jack_rip_vim> annk, you can enable it in there 14:28 < ankk> hmm 14:28 < JimBuntu> altendky, Sorry, I was AFK. Glad to hear you got it working, hacky or not. 14:28 < ankk> jack_rip_vim: which line is it? i don't have access to the server right now 14:28 < ankk> can you give me the keyword if you remember 14:29 < jack_rip_vim> ankk: when you open it, check this keyword --> log 14:29 < ankk> jack_rip_vim: :) 14:29 < jack_rip_vim> ankk, the conf file write clear 14:30 * Pantsu suggests replacing ftp while you are at it 14:31 < ankk> jack_rip_vim: thank you :) see you 14:31 < jack_rip_vim> see you! 14:33 < rcp> any is way of importing lines from text file and writing them to bottom of .htaccess file? 14:33 < rcp> from ssh 14:33 < rcp> I need to bulk add 301 redirects 14:34 < ananke> rcp: sounds like you should be investing time in some kind of config management tool 14:35 < silv3r_m00n> hi there 14:35 < silv3r_m00n> i dont know bash, can someone explain to me the lines here, https://pastebin.ubuntu.com/p/786PncRgpp/ 14:35 < silv3r_m00n> i need to modify that code, but cant figure out the syntax 14:35 < ananke> silv3r_m00n: explanation is in the comments. 14:36 < silv3r_m00n> ananke: ok, what exactly is happening here, request="${request#*?url=}" 14:36 < JimBuntu> a bit of 'find . -iname="*.htaccess" -exec cat >>{} \;` rcp 14:36 < Pantsu> silv3r_m00n: http://mywiki.wooledge.org/BashGuide 14:36 < silv3r_m00n> what does the $ sign indicate, and how is request put in there 14:36 < ananke> silv3r_m00n: for more details you can read about 'bash substitution'. eg: https://www.tldp.org/LDP/abs/html/parameter-substitution.html 14:36 < Pantsu> the tldp one is not very good 14:36 < djph> the '$' indicates the beginning of a variable... 14:37 < ananke> Pantsu: quite likely. it just happened to be first hit on google :) 14:37 < djph> the wooledge.org bash guide is very good... 14:37 * Pantsu would not write this in bash though, parsing urls in bash is not a thing you want to do 14:37 < djph> ^ 14:37 < JimBuntu> ^^ Python would be an easy way 14:37 < silv3r_m00n> actually i cant understand the replacement part in that code, how is the replacement working 14:38 < jack_rip_vim> hi Pantsu, haven't seen you for a long time! 14:38 < ananke> silv3r_m00n: you're going to have to read more 14:38 < silv3r_m00n> like this request="${1#*://}" The "" is for making a string, the $ is for variable value ? 1 is the first command line parameter, what is #* ? and where is the replace by blank ? 14:39 < ananke> silv3r_m00n: have you read anything that we gave you? 14:39 < silv3r_m00n> going through the tldp link right now 14:39 < jack_rip_vim> Pantsu: actully, I just saw you few days ago, but I haven't seen you speak like this way for a long time. 14:39 < ananke> silv3r_m00n: then finish doing so 14:39 < Pantsu> silv3r_m00n: read the wooleg one instead 14:40 < Pantsu> http://mywiki.wooledge.org/BashGuide/Parameters#Parameter_Expansion 14:40 < jack_rip_vim> :P 14:41 < date_night> Is anyone here using openbox and firefox 60? Currently having trouble with removing firefox's titlebar 14:41 < silv3r_m00n> got it 14:41 < silv3r_m00n> # means remove the pattern following 14:42 < date_night> The closest thing is the `Un/decorate` but that removes the maximise / minimise / close buttons 14:44 < Pantsu> date_night: click the three -, click customize, untick the title bar at the botton right 14:44 < Pantsu> bottom* 14:45 < date_night> I've done that and also set `browser.tab.drawInTitlebar` to True 14:46 < date_night> I can hide title bar in KDE 14:46 < date_night> Just switched to openbox, and now firefox cant hide titlebar 14:46 < Pentode> what version of openbox is it? maybe its not fully EWMH compliant 14:47 < jack_rip_vim> Ajohn 's quit message is amazing. 14:47 < date_night> Pentode: version 3.6.1 14:47 < silv3r_m00n> Pantsu: that did it all 14:47 < silv3r_m00n> thanks 14:47 < date_night> Pentode: Im on Debian Unstable 14:47 < silv3r_m00n> is there a way to extract a certain number from a string 14:48 < Pantsu> date_night: post screenshot 14:48 < Pantsu> silv3r_m00n: depends on what the string is 14:48 < Pantsu> silv3r_m00n: generally: don't use bash 14:48 < silv3r_m00n> great 14:48 < Pentode> date_night, did older versions of firefox work fine with it? 14:52 < date_night> I think the hide titlebar feature is only for firefox 60 14:52 < RevanOne> anyone using vpn libreswan ? 14:55 < Pentode> date_night, i've tried it on my other laptop and indeed it doesn't work. 14:55 < Pentode> best bet would be to submit a bug report or something 14:56 < Pentode> unless you want to get your hands dirty and try and fix it yourself 14:56 < RevanOne> I have installed VPN libreswan, L2TPD over IPSEC but I can only have 1 user connected at a time. 14:58 < date_night> Pentode: My guess would be that openbox is more of a "minimalist" wm and has less features than DE, thus unable to allow software to hide titlebars 14:58 < Pantsu> RevanOne: I would chose a vpn provider that uses openvpn instead 14:58 < date_night> I will look up how to file a big report now :) 14:58 < Pantsu> if you are setting up your own vpn, use openvpn or wireguard 14:59 < JimBuntu> Digital Ocean has some great walk-through guides for setting up OpenVPN, it really couldn't be much simpler while still being a proper install/configure 15:00 < RevanOne> I didn't want to use OpenVPN because I have to download additional software. This ones looked easier and the only problem I have right now is this client connection problem 15:01 < RevanOne> I've setup OpenVpn in the past 15:01 < RevanOne> I would rather find out why I have this problem and fix it than just ignore it 15:04 < Pantsu> silv3r_m00n: what is it that you are parsing? 15:04 < silv3r_m00n> Pantsu: https://pastebin.ubuntu.com/p/Z76FXwfwXT/ 15:06 < Pantsu> don't use bash 15:07 < Armand> Do use bash 15:11 < oneko> RevanOne: Wait, what are you trying to say ? Installing libreswan, L2TPD over IPSEC is easier than just downloading openvpn ? 15:14 < ayecee> oneko: operating systems like windows and osx have that built in. they don't have openvpn built in. 15:14 < ayecee> also android and ios have that built in. 15:19 < oneko> ayecee: I am a cross-platform human being and I'd rather people around here don't go as much as *trying* to shame me about it ;-) 15:20 < pocketmon> hello how can i unzip 7z? 15:21 < solidfox> pocketmon, install 7-zip 15:22 < revel> Engrampa works with 7z archives if you've got p7zip. 15:22 < zetheroo> I can wget a domain on one system instantly, but on another it times out always - both systems are Ubuntu 16.04 - the one that works in Server edition, while the one which doesn't work is Desktop edition. How to troubleshoot this? 15:23 < revel> zetheroo: It's not on ipv6, is it? 15:23 < revel> Also, does curl work? 15:23 < zetheroo> revel: no, and both systems are on the same network 15:24 < zetheroo> yes, curl https://domain.com works 15:25 < revel> Are the `wget --version | head -n 6` thingies too different? 15:26 < zetheroo> revel: sorry, I don't know about that 15:26 < zetheroo> I am just doing wget domain.com on both systems 15:28 < zetheroo> oh, sorry ... comparing them now .. 15:28 < zetheroo> both are identical 15:29 < rypervenche> ayecee: Technically though, you still need to install software to get IPsec working (libreswan), no? So I don't know that "I don't want to install software to get this working" is much of an argument. Although I may be wrong. I do understand the benefits of IPsec. 15:30 < revel> Not sure then. 15:32 < jhodrien> rypervenche: Not on Windows/OSX though. 15:32 < phre4k> my USB sound card is way louder under Windows than in Linux, I think I found the solution with some ALSA or sysctl settings but I can't seem to find them now :/ 15:32 < phre4k> how do I increase the volume of the sound card? 15:33 < rypervenche> jhodrien: Ah, fair enough then. 15:34 < phre4k> ok, what the heck, I just increased the volume in alsamixer and it worked – somehow it seems to reset after reboot though. Is there a "default volume" set somewhere? 15:40 < oneko> rypervenche: Wiregaurd is almost a drop in replacement for IPSec and, it's easy to configure 15:40 < oneko> And, it's been until now, bug free 15:43 < jhodrien> "WireGuard is not yet complete. You should not rely on this code." 15:43 < jhodrien> That's a fairly sound reason not to use it. 15:43 < jhodrien> Yet. 15:44 < oneko> But, you can always use it at your own risk :-P 15:44 < rypervenche> oneko: Looks very cool. Looks like something useful in the future. 15:52 < drzacek> Hello 15:52 < drzacek> Is there a way to force "emergency umount & shutdown" on linux? And how long would it take (if possible)? 15:53 < Desu> drzacek: read up on sysrq 15:54 < Desu> and why do you care about the speed? 15:54 < Desu> what is the actual use case 15:55 < Taoki> Hi, I have a question. Is the video card Sapphire Radeon RX 550 Pulse considered RadeonSI by the amdgpu / mesa drivers? 15:55 < phre4k> drzacek: what is an "emergency"? If you need your device to work in unreliable situations you should probably use a USV or some overlay FS where you commit changes 15:56 < drzacek> Desu, my boss told me that on his old system (dos) he was able to get information from PSU that the power was gone, and he got ~20ms to stop any disk operations, thus the system wouldn't fail and could safely start again 15:56 < phre4k> Taoki: the RX550 is Polaris architecture and supported by the amdgpu/mesa drivers. 15:56 < drzacek> Just want to implement some extra protection from sudden powerfails 15:56 < phre4k> Taoki: as the system is fairly new you should use a recent distribution/Kernel/mesa though 15:56 < Taoki> phre4k: Awesome. Will it be using the RadeonSI bridge though? The video drivers has two options: R600 and RadeonSI 15:57 < Taoki> I have that, sure 15:57 < Desu> drzacek: s/psu/ups/ you mean 15:57 < phre4k> Desu: ah, yes 15:57 < drzacek> no 15:57 < Desu> drzacek: and any ups system would give you more than enough time for a normal shutdown 15:57 < drzacek> the idea is not to use ups 15:57 < Desu> the psu itself does NOT store enough energy for a proper shutdown in any way 15:58 < Desu> you are lucky if it gives you 100ms or runtime 15:58 < Desu> of* 15:58 < Desu> if you want protection form powerfails, get a ups 15:58 < phre4k> drzacek: you can protect from power failures by using an UPS 15:58 < phre4k> drzacek: there's literally no other way 15:58 < Desu> there are combined ups/psu units for some servers 15:59 < phre4k> drzacek: you should use a journaling file system, too 15:59 < Desu> doubt your boss had one of those though 15:59 < JimBuntu> UPS/PSU combo, with comms to the OS 15:59 < Desu> probably just comfuses psu and ups 15:59 < JimBuntu> UPS have been telling the OS of power failure for well over a decade 15:59 < Desu> more like 50 years 15:59 < phre4k> Taoki: you shouldn't need to specify any driver/kernel module yourself, it just works™ (or it doesn't, if your Kernel is too old) 16:00 < Desu> hmm, more than 50 even 16:00 < JimBuntu> Desu, yeah, I don't remember how long exactly 16:00 < Taoki> I know. I'm only asking so I know for debugging, development, etc urposes. 16:00 < Desu> even back in the days of mechanical computing there were devices with a ups like system 16:00 < Taoki> Just wanted to know if it's technically treated as RadeonSI 16:00 < Desu> to avoid having the machine break/end up in a horrible state if the power dropped 16:00 < phre4k> Taoki: IIRC, yes 16:00 < Desu> electromechanical* 16:01 < Taoki> Awesome. Thanks :) 16:01 < Desu> were often flywheel based 16:01 < drzacek> well, wouldn't 100ms be enough to at least protect filesystems? 16:01 < Desu> no 16:01 < ayecee> probably a little short for that 16:01 < JimBuntu> drzacek, 100ms... doubtful 16:01 < Desu> try 5+ seconds 16:01 < JimBuntu> large file syncs can take seconds 16:02 < Desu> more realisticly 10-30 sec 16:02 < drzacek> the filesystems are mounted with sync option 16:02 < JimBuntu> don't forget, this shutdown period is going to use considerably more power than when the machine is in idle 16:02 < Desu> anyway, ups are cheap these days 16:02 < Desu> get a line-interactive one (offline are useless) 16:02 < drzacek> Desu, didn't found any under 200 16:02 < drzacek> euro 16:03 < Desu> drzacek: not long ago they were 10-50x that 16:03 < drzacek> doesn't mean they are cheap 16:03 < Desu> drzacek: also, you should be able to find some for 100ish eur even new 16:03 < screwsss> any kali users in here atm 16:03 < Desu> 200€ < 500€ damage to your computer due to power spikes and the like 16:03 < Desu> screwsss: they have their own channel 16:04 < screwsss> hook me up 16:04 < Desu> https://www.amazon.de/APC-Back-UPS-Unterbrechungsfreie-Stromversorgung-BX700U-GR/dp/B00T7BYRCK/ref=sr_1_1?ie=UTF8&qid=1527689045&sr=8-1&keywords=apc+ups 16:04 < Desu> 70€ 16:04 < Desu> there are some in the 50€ range too 16:05 < Desu> also can often find used ups systems really cheap, specially if you are willing to swap the battery yourself 16:06 < drzacek> any other steps one can do before using ups? 16:06 < Desu> people will trash or sell dirt cheap a 500€ ups just because the battery is worn out, when you can get a new battery for 30-100€ 16:06 < phre4k> drzacek: 70€: https://geizhals.eu/apc-back-ups-700va-bx700ui-a1236942.html 16:06 < Desu> drzacek: not really, no 16:07 < drzacek> ok thanks 16:07 < JimBuntu> drzacek, other option is to use a laptop instead... then it has a battery 16:07 < Desu> doesn't give you the protection against spikes etc that a ups can though 16:08 < Desu> spikes/brownout/frequency drift/ 16:08 < JimBuntu> Desu, only light protection... basically sacrifice the laptop's external power supply 16:09 < Desu> JimBuntu: spike will happily fry the mobo 16:09 < Desu> a spike* 16:09 < Desu> some line-interactive ups systems have some decent protection, the best ofcourse is online systems 16:09 < JimBuntu> Totally possible. 16:10 < Desu> with online systems you have the converters and between you and the outside 16:10 < Desu> can get them with galvanic seperation too 16:11 < JimBuntu> Air gapped? lol 16:11 < Desu> isolation transformer, flywheel, whatever 16:11 < Desu> online flywheel systems are still used as protection devices in some use cases 16:12 < Desu> big spinning chunk of metal with a motor at one end and a generator at the other :) 16:12 < Desu> not much that can go wrong 16:14 < Desu> the modern compact designs can more easily spark over though 16:16 < Desu> I've had fun at a place that had a 50+ tonn flywheel spinning in the open as the buffer. Built in the 30's, a work of art 16:17 < JimBuntu> I am familiar with these flywheel designs... more familiar with multiple balls, that rise as the spin rate increases. 16:27 < sruli> got a strange problem with inotifywait "inotifywait -m -r -e create /home/user/DIR1/" i get an error "Couldn't watch /home/user/DIR1/: No such file or directory! dir exists, if i execute "inotifywait -m -r -e create /home/user/DIR1/subDir" it doesnt complain 16:29 < fofalee> hi 16:30 < fofalee> if [[ cd / ]];then...;fi fails ? 16:30 < prussian> if cd /; then ... ; fi 16:30 < fofalee> what's the other way to execute a bunch of cmd ,and check it it's true; then do something? 16:30 < post-factum> fofalee: and what are you trying to solve by this? 16:31 < post-factum> if { true && false }; then yes; fi 16:31 < fofalee> I wrote a script if there is some dir, then run that program, else do nothing. 16:33 < post-factum> if [ -d "dir" ]; then… 16:36 < hollusion> how do you make an apache2 virtual host publicly accessible 16:36 < Oxyz> any 'sed' ninjas around ?! =) 16:37 < hollusion> do i need to edit /etc/apache2/sites-available/nextcloud.conf ? 16:41 < golden_rule> sruli: ending / vs. non-ending in / ? 16:41 < Oxyz> trying to extract the value "12" out of this example.. {}}i:12 but just can figure out how... 16:41 < Oxyz> echo ""matches";a:0:{}}i:12;O:10:"rRSSFilter":17" | sed -e 's/}i:\(.*\);0:10/\1/' 16:41 < Oxyz> any ideas ? 16:41 < sruli> golden_rule: final slash makes no difference, all dirs that work it doesnt care about the slash 16:41 < golden_rule> arright 16:53 < RebelCoder> Guys, why would the following happen: files are in a folder, you run the .sh script and you get an error that files is not found? 16:53 < RebelCoder> Like this: https://pastebin.com/4WNKvY1a 16:54 < ananke> RebelCoder: what's the output of 'head -1 olly-start.sh' ? 16:55 < dzove855> 2/ 16:55 < dzove855> heu sorry guys :) 16:56 < RebelCoder> ananke, 16:56 < RebelCoder> https://pastebin.com/CzYMTmP4 16:56 < RebelCoder> ananke, any idea ? Never had that problem before. Trying to run my scripts on a new linux image... 16:57 < twainwek> what could be the reason for sed not outputting the two files with similar contents 16:57 < twainwek> oops disregard 16:59 < golden_rule> you are trying to match a paren where is none 16:59 < golden_rule> \( <- literal 17:04 < twainwek> how does sed deal with eol characters? 17:05 < twainwek> would wanting to match 'key$' behave differently on files created on nix/windows? 17:06 < NewbProgrammer10> twainwek: I don't know, but if it does it unsatisfactorily, you can use gawk for that job. 17:30 < strixdio> are there foss/floss virtual sans for linux? 17:31 < phre4k> strixdio: is that a font? 17:31 < strixdio> lol 17:31 < strixdio> specifically storage SAN 17:31 < phre4k> ah, a SAN, not sans 17:32 < strixdio> ;) 17:32 < phre4k> well you could use cephFS or glusterFS or Greyhole, depending on your needs 17:33 < strixdio> cool, thanks for the info 17:33 < phre4k> strixdio: aaand look at https://github.com/n1trux/awesome-sysadmin, maybe there's more you didn't consider 17:34 < strixdio> ohboy. 17:34 < strixdio> Thanks! 17:36 < JimmyNeutron> Is there a way to pause a program in the middle of execution? The program takes 1 second to run, but I would like to run it and pause it before it finishes. CTRL+z doesn't work because program ends to quickly. 17:36 < phre4k> er, sorry strixdio, meant https://github.com/n1trux/awesome-sysadmin#distributed-filesystems 17:37 < phre4k> forgot to link the correct heading 17:37 < phre4k> JimmyNeutron: a debugger? 17:37 < phre4k> why do you want to pause it? 17:38 < JimmyNeutron> phre4k: Yea,h, I was thinking of that too.. I want to take a look at what files it's creating a lock on and simulate a lock 17:39 < mawk> ail 17:41 < phre4k> JimmyNeutron: maybe use strace? 17:45 < JimmyNeutron> phre4k: THanks!, but already did that and it did help to see what files are being lock by the program 17:46 < JimmyNeutron> but now I want to simulate a lock in where the users may actually, get lucky, and hit ctrl+z in time before the program exit. 17:46 < JimmyNeutron> on my test server, program takes 1 second to run but maybe on their server, it may take a few seconds. 17:52 < oo_miguel> I wonder if it is possible to symlink WITH command appplication.. create a file that symlinks to an existing files but applies a sed or diff to this file? Does this make sense? 17:52 < alexandre9099> hi, i'm using linux with openvpn, but i'm having trouble with dns, it changes from time to time on resolv.conf file, which leads to failing in solving dns requests (it changes to the local network that i'm connected) 17:55 < zax1> i have an oracle server that crashed a couple of days ago.. i am a linux newb - i cant find the syslog file 17:56 < zax1> any ideas or tips on diagnosing why it crashed ? 17:56 < phre4k> zax1: which distro? if it has systemd, the "syslog file" is journalctl ;) 17:59 < zax1> phre4k, running lsb_release give me this : EnterpriseEnterpriseServer 18:00 < zax1> code name carthage - looks like oracle 5.11 18:01 < uberwag> i get this selinux error all the time, how can i fix this? https://pastebin.com/GsEvUczZ 18:09 < pankaj> I was having issues with kernel compilation that I was unable to connect to net (and other problem) but now I know it was related to dual boot. 18:21 < BluesKaj> alexandre9099, which OS? 18:22 < alexandre9099> arch 18:22 < alexandre9099> (well, antergos, tbh) 18:23 < varliberate> yo 18:24 < alexandre9099> BluesKaj, i'll have to go in a few minutes, then i'll be back in like 40m :) 18:24 < plexigras> I'm trying to mount an iso and whatever i try i always get `mount: /media/iso: mount failed: Operation not permitted.` 18:25 < BluesKaj> alexandre9099, ok 18:25 < varliberate> some one needs to invent "git lite" 18:25 < varliberate> git for the node generation lol that doesn't have a billion flags and rules 18:25 < plexigras> is this the right command? `sudo mount -o loop my.iso /media/iso` 18:26 < varliberate> plexigras, looks good to me 18:26 < varliberate> never understood why -o is the flag there 18:26 < plexigras> well it's throwing the error i posted above 18:26 < lnnb> -o loop is needed ? 18:27 < varliberate> yeah usually..... hmmm 18:27 < plexigras> i get the same error when i add/remove `-o loop` 18:27 < prussian> is this some docker image or something? 18:27 < prussian> you mgiht not have a the capability 18:27 < plexigras> what do you mean? 18:27 < varliberate> what are you doing it on? 18:28 < prussian> create the loop device nodes with losetup 18:28 < prussian> if mount fails 18:28 < NewbProgrammer10> plexigras try running the command as root... 18:28 < varliberate> plexigras, what are you doing it on? (version of linux etc) 18:28 < plexigras> i already tried that and it did not work 18:28 < NewbProgrammer10> Hmph 18:29 < Desu> selinux? 18:29 < plexigras> arch 18:29 < varliberate> Desu, yeah could be 18:29 < NewbProgrammer10> plexigras, what Desu means is do you have selinux? 18:29 < NewbProgrammer10> Installed? 18:29 < NewbProgrammer10> And enabled? 18:30 < varliberate> look in /etc for selinux stuff 18:30 < n-iCe> No idea where to ask, I bought a repeater for my wifi, signal is great, but internet is slow, any idea what's happening? 18:30 < varliberate> n-iCe, wow that's "nice" and off topic lol 18:31 < plexigras> i dont see anything called selinux in /etc 18:31 < varliberate> find /etc | grep sel* 18:32 < useless-eater> http://paste.debian.net/1027290/ Is this related to buggy driver, or should I start looking for another card? using non-free driver packaged for debian stretch 18:32 < varliberate> plexigras, you have tried different locations I guess? I tend to use /1 and /2 because I'm slack (and I like solaris) 18:32 < twainwek> n-iCe: which signal 18:32 < useless-eater> some strange purple lines are appearing on my screen too :| 18:32 < qrvpzvb> is there a way to check my memory use 18:33 < qrvpzvb> other than top and free 18:33 < prussian> i doubt it's selinux. if you want to check plexigras ausearch --message avc --start recent 18:33 < plexigras> varliberate: there is nothing called selinux to be found 18:33 < varliberate> so strange 18:33 < prussian> i strongly doubt it though 18:33 < qrvpzvb> because 4gbs of ram are occupied and only samba is running 18:33 < varliberate> qrvpzvb, scanning electron microscopy? :D 18:34 < varliberate> take a look hehe 18:34 < plexigras> so what does this mean for me do i need selinux or something 18:34 < qrvpzvb> and it's not buffers either 18:34 < varliberate> plexigras, nah the opposite selinux can confuse things 18:34 < plexigras> oh ok good 18:35 < plexigras> so what else could it be 18:35 < prussian> what is the exact error message? 18:35 < prussian> have you tried creating the loop device with losetup instead? 18:35 < plexigras> i can post it again if you like 18:35 < lnnb> i don't think you need a loop for iso9660 18:35 < plexigras> prussian: no 18:36 < varliberate> like it really is straight forward https://www.cyberciti.biz/tips/how-to-mount-iso-image-under-linux.html 18:36 < triceratux> plexigras: i use -t iso9660 for that in fact 18:36 < varliberate> lnnb, yeah ya do 18:37 < varliberate> ok try :) don't let me stop you, I not know arch 18:37 < prussian> plexigras: so use it then 18:37 < lnnb> i just did it without a loop device 18:37 < prussian> losetup --show -f your.iso 18:37 < prussian> lnnb: losetup -a 18:38 < lnnb> varliberate: run `file your.iso` does it say iso 9660 ? 18:38 < varliberate> lnnb, it's plexigras that's got the problem 18:38 < varliberate> it's got my attention because it's normally a pretty day to day thing that works 18:39 < plexigras> well i think i just got it to work by using file-roller to extract the iso's contents 18:39 < varliberate> plexigras, yeah, whatever gets the job done, of course, but it's a strange one :) 18:39 < lnnb> oooh it does all of that automatically without telling you eh, thats kind of annoying i have a script that uses loop0 guess i have to fix it now 18:40 < varliberate> might you need fuse installed? 18:40 < varliberate> try install fuse-common or whatever and give it a go then 18:41 < sgautam_> hello 18:41 < sgautam_> how do i write a Windows ISO to a USB stick under Linux 18:42 < varliberate> plexigras, you aren't putting the iso and mount directory around the wrong way are you? 18:42 < varliberate> sorry to ask but we all do it 18:42 < plexigras> i also tried that :) 18:43 < varliberate> :) 18:43 < plexigras> but for now i will just use the extracted files 18:43 < varliberate> yep 18:43 < Pentode> sgautam_, WinUSB or poweriso. or something other utility that supports it. 18:43 < triceratux> sgautam_: http://multibootusb.org/ ? 18:43 < varliberate> well thanks for the distraction, I'm off to bang my head against git's billion and one settings 18:44 < triceratux> https://itsfoss.com/bootable-windows-usb-linux/ 18:44 < prussian> sgautam_: you copy the contents of the windows iso onto a fat32 formatted stick. 18:44 < varliberate> CI+CD is so nice when it's all set up, devs have no idea what a chore it is to do that though :) 18:44 < prussian> unless it's some windows server 20xx thing it should "work" 18:45 < sgautam_> prussian: yeah i did that but how do i write the bootloader? 18:45 < prussian> since win 2012 and up have massive archives too large for fat32 18:45 < prussian> no 18:45 < prussian> you don't it should have an \\EFI\ in the root of the windows image 18:45 < prussian> with a BOOTX64.EFI or whatever 18:47 < phre4k> sgautam_: https://github.com/slacka/WoeUSB 18:47 < prussian> your UEFI bootloader should be able to find it if at the root of the drive, there exists an \EFI\BOOT\BOOT.EFI 18:48 < phre4k> ms-sys -7 even 18:48 < sgautam_> okay thanks a lot phre4k and prussian i will try WoeUSB and report my findings 18:49 < prussian> note woeusb will not work with Windows server 2012 and up. you need a convoluted grub bootloader for that 18:54 < lexicon65> Do MPV have any equalizer? 18:56 < Trel> Using the dialog program, is there any way to set the colors for a one off dialog (without using a .dialogrc file)? 18:57 < Pentode> lexicon65, your best bet is to use the pulse audio equalizer or the alsamixer equalizer 18:57 < Pentode> bah 18:58 < stain_363> Pentode: Ok, thank you :) 18:58 < Pentode> omg this evil usb port 18:58 < Pentode> port / hub 18:59 < Pentode> it wasnt even cheap. thank god it didnt fry my usb interface on one of my machines 18:59 < Pentode> i plug it in and the machine resets, i'm like wtf. it's letting the 5v from the power supply backfeed back into the usb connector to the machine 19:00 < Pentode> couldnt afford a friggin diode? 19:00 < sadasaulna> wow 19:02 < pankaj> I have dual boot on my system. How can I chroot on that partition so that I can run command without restarting and logging in to that system. 19:03 < mawk> what's the problem with that Pentode ? the tension will be 0 since the usb connector also supplies +5V itself no ? 19:04 < Pentode> _most_ system boards should be able to cope with it 19:04 < Pentode> but the power supply can deliver 2 amps of current 19:04 < sadasaulna> mawk, presumably not quite zero 19:05 < mawk> ah if the computer is powered down then current will flow 19:05 < sadasaulna> they're not going to exactly match 19:05 < mawk> since the usb host controller won't supply +5V anymore 19:05 < mawk> I see sadasaulna 19:07 < sadasaulna> plus maybe ground return path is messed up. Its ages since I remember this stuff, but ground loops create all kinds of weird shit 19:08 < pankaj> Hello everyone. Is thier any command to get chrooted into the partition (as I am using dual boot). 19:09 < Pentode> ive just plugged it into several different ancient / half broken laptops and they all seem to cope with it aside from the one machine rebooting (newer) 19:09 < Pentode> however, it KILLED the macbook pro 19:09 < sadasaulna> pankaj, dual linux? Or are you talking bout Linux/Windoze? 19:10 < sadasaulna> Pentode: what a POS :( 19:11 < Pentode> honestly machines should have pretty good protection given the kinds of things people plug into their computers now-a-days 19:11 < Pentode> but stay away from 7port hubs made by "terminus inc", lol 19:11 < mawk> pankaj: yes: chroot, but you can also boot inside your second OS using systemd-nspawn for instance 19:12 < sadasaulna> should be called "terminal inc" lol 19:13 < Pentode> lol 19:13 < pankaj> sadasaulna: I have both or all three on dual boot. 19:13 < pankaj> mawk: What is this 'systemd-nspawn' 19:13 < mawk> a commandr 19:13 < sadasaulna> pankaj: you could just mount one of the partitions somewhere under /mnt or wherever you like and chroot to it 19:13 < dgurney> what is this search engine/man page 19:14 < mawk> it's the container command of systemd 19:14 < pankaj> sadasaulna: That is cool 19:14 < dgurney> systemd-nspawn is really neat 19:14 < Netham45> Is there any way to cancel a tab complete? It's trying to iterate through like 400k files on a network filesystem. 19:14 < mawk> try ^C a couple times Netham45 19:14 < Pentode> control+c? 19:15 < Netham45> Oh, it finally finished. 19:15 < domhnall> \o/ 19:17 < dsnvhlm> Anyone using arch linux here? 19:17 < bls> no, no one here uses arch linux, ever 19:17 < phatcat_> heya, I'm having a problem with Kali Linux - my touchpad freezes randomly and "modprobe -r psmouse" doesn't help. can someone help please? 19:18 < domhnall> phatcat_: well, check for synaptic drivers in the repo. also #kali-linux. 19:18 < pankaj> sadasaulna: Is their anything to care about using this command or in comparision to chroot how it is different? 19:19 < phatcat_> domhnall: thank you 19:19 < de-facto> i am searching a software to display a website in fullscreen mode: just the one website and no right click or menu interaction, no tabs, no new windows, no closing it. just the website, a mouse cursor, keyboard input and support for cookies. which is the most minimal solution for something like this? 19:19 < prussian> chromium in kiosk mode 19:20 < sadasaulna> ask mawk about systemd-nspawn, I don't know much about it as don't run systemd but I assume its more like running your other environment in a lightweight virtual machine like docker 19:20 < de-facto> prussian, i tried that, but it wont let me disable Ctrl+N or new tabs and such 19:20 < sadasaulna> ie, it uses linux containers 19:21 < phatcat_> domhnall: thing is, wifi doesn't work either) can I get the synaptic drivers in thru an usb? 19:21 < de-facto> plus fullscreen does not work properly 19:21 < sadasaulna> chroot is just chroot, its not very fancy 19:21 < domhnall> phatcat_: not sure what you mean by through usb. as in a wifi-dongle? 19:22 < de-facto> maybe something webkit based? 19:22 < sadasaulna> domhnall: i think he means a usb pen drive 19:22 < pankaj> mawk: Is their any downside of using 'systemd-nspawn' and in comparision to chroot how it is different. 19:22 < phatcat_> domhnall: nah, as in usb stick 19:23 < domhnall> sadasaulna: ok. 19:23 < sadasaulna> phatcat_: you not got wired ethernet on that? 19:24 < phatcat_> sadasaulna: no cable atm 19:25 < domhnall> phatcat_: also, which DE are you on? the 'freeze' may just be from a setting to disable touchpad-while-typing. 19:25 < domhnall> iirc, xfce does that. 19:26 < phatcat_> domhnall: gnome? 19:26 < bls> some of them also do palm detection if they detect lots of contact on both sides of the pad 19:26 < domhnall> hmm, havent used gnome in a while, but you can check. 19:26 < LordRyan> I can use `nc localhost 6600` and I can ssh into the machine I'm working with, but I can't connect to it from my computer. iptables rules are empty. 19:27 < LordRyan> Anyone know what might be up? 19:27 < TheWild> hello 19:30 < phatcat_> domhnall: it was true, set it to false 19:30 < bls> LordRyan: you'll need to rephrase that, because it reads as you saying: I simultaneously can and can't connect with nc and ssh 19:30 < phatcat_> let's see if it helps :) 19:30 < phatcat_> domhnall: oh no, froze agaibb 19:31 < TheWild> not really Linux question, but... I have some problems with the SSD in my office computer. It generally works on normal load. However, sh** happens when I have to read (*read*, not write) a lot of data (say 40 GB) in a short amount of time. It starts from about 100 or 200 MB/s, then goes down badly, sometimes even to 2 MB/s. Randomly stops responding, then continues. In rare cases, it stops responding completely, so Windows goes B 19:31 < LordRyan> bls: I can connect to port 22 using SSH, and can use nc 6600 on that machine. I can't, from *my* machine, nc 6600 19:31 < LordRyan> There's no iptables rules at all. 19:31 < TheWild> Am I abusing the disk? Is this a standard reliability of SSDs? 19:32 < dgurney> TheWild, how old is this SSD? 19:32 < koala_man> TheWild: are you sure that's a SSD? 19:32 < TheWild> pssh... about 9 months? 19:32 < michaelrose> is it a hybrid ssd hd arrangement? 19:33 < TheWild> it's not. Boss, despite my protests (SSD is 240 GB, HDD was 500 GB), removed it from the computer. 19:34 < michaelrose> which filesystem? 19:34 < TheWild> NTFS 19:34 < dgurney> either you're transferring a load of small files (which would explain the decreasing speeds), but excessively long thinking breaks? sounds like something could be wrong with the FS 19:35 < TheWild> hmmm... I haven't performed fs check for months. 19:35 < dgurney> well maybe you should then 19:36 < TheWild> once I get back to this machine and will not forget... 19:36 < domhnall> phatcat_: hmm, idk, not sure if requires reboot but I don't believe it's hw related. How about that wifi though? 19:37 < alexandre9099> BluesKaj, i'm back :) 19:37 < phatcat_> domhnall: tried a reboot as well, still freezing.. 19:37 < domhnall> sorry I can't help with anything gnome...haven't used it in 2yrs. 19:37 < alexandre9099> i can't reproduce right now the problem, only on work 19:38 < domhnall> phatcat_: okay, so is the freeze an hinder on your productivity or just annoying? 19:38 < phatcat_> domhnall: annoying + would help in solving the wlan0 prob if it worked 19:40 < domhnall> cool, so with the wifi, have you tried anything? being on kali, you can't expect spoon-fed. 19:41 < domhnall> and please...lets move it to #kali-linux 19:41 < BluesKaj> alexandre9099 set your DNS IPs in /etc/systemd/resolved.conf, and remove the # in front of the DNS line where the IP is entered. This worked for me til I dropped network manager 19:42 < domhnall> unless...others are willing to help, despite knowing you're on a distro not considered 'sane' here. 19:43 < alexandre9099> BluesKaj, which networking manager do you use? 19:43 < de-facto> how can i completely disable Ctrl key in Xorg? 19:43 < alexandre9099> would setting the dns there make it permanent? 19:45 < bls> de-facto: xmodmap can do that 19:45 < phatcat_> domhnall: thing is, I'd love to be on any other distro 19:45 < phatcat_> but I could only EFI boot kali.... 19:46 * domhnall likes kali... 19:46 < phatcat_> it wouldn't see anything else 19:46 < solidfox> phatcat_, ubuntu 19:47 < solidfox> (kali is based on ubuntu) 19:47 < domhnall> no...debian 19:47 < domhnall> backtrack was 19:48 < domhnall> but anyway... 19:48 < phatcat_> solidfox: tried booting ubuntu from a stick, didn't work :\ 19:48 < BluesKaj> alexandre9099, I just use ifupdown and set my DNS in the router (home network) the use the line: iface ethernetname inet dhcp in /etc/network interfaces with gateway IP , pc IP (address) and netmask 255.255.255.0 19:48 < phatcat_> just kali for some odd reason 19:48 < solidfox> domhnall, backtrack had a cooler name anyway 19:49 < BluesKaj> alexandre9099, not sure this works in arch, but it does in Kubuntu 19:49 < phatcat_> is there another distro I could try? 19:49 < BluesKaj> alexandre9099, so don't really use a network manager 19:50 < BluesKaj> I don't 19:50 < dsnvhlm> phatcat_: try linux lite 19:50 < alexandre9099> BluesKaj, :D 19:50 < phatcat_> dsnvhlm: doing it as we speak, but why wouldn't normal ubuntu work 19:50 < alexandre9099> having to set every network on wpa_supplicant is no easy task 19:50 < phatcat_> tried both 86 and 64 19:53 < BluesKaj> alexandre9099, on a laptop, oops I should have asked first...then we're in a different situation, but my first suggestion still stands using network-manager and /etc/systemd/resolved.conf 19:54 < alexandre9099> thanks (btw, on the desktop/server i have Network-manager, but i'm not really sure why :D i could just have dhcpc as it is wired) 19:57 < spear2> what is the difference between 'printenv PS1' and 'echo "$PS1"'? in the latter case the PS1 value i set in .bashrc is shown, in the former it is not shown 19:57 < spear2> other variables that show up with echo but not printenv are COLUMNS and LINES for example 19:57 < koala_man> spear2: printenv is external and can therefore only access exported variables 19:58 < bls> spear2: is PS1 exported or not? 19:59 < spear2> bls: i guess it is not in some cases, i just tried on another machine and it was 20:00 < koala_man> it's not exported by default 20:00 < bls> spear2: do you understand the difference between shell variables and environment variables? 20:00 < kubast2> Is it possible to check which app is utilizing gpu(nvidia prioprietary) ? 20:00 < phatcat_> can someone pls explain why I am only able to boot Kali from USB 20:00 < phatcat_> and any other linux distro doesn't work 20:00 < koala_man> phatcat_: what happens for the others? 20:01 < kubast2> got 100% gpu usage 81*C on a gpu in my laptop ,no intensive 3D apps running 20:01 < koala_man> coin mining malware 20:01 < bls> does kali use the microsoft secure boot shim and none of the others do? 20:01 < hexnewbie> kubast2: What desktop environment? 20:01 < kubast2> kde ,but it usually doesn't do that 20:01 < kubast2> it usually stays at 0-5% 20:02 < kubast2> my guess is that it's the spinning cricle of notification of plasmashell 20:02 < kubast2> cause it spins at the speed of light 20:02 < hexnewbie> kubast2: KDE (Plasma rendering) does that twice a week for me regarding CPU usage, I wouldn't be surprised if it also causes spikes in GPU usage. 20:02 < kubast2> yeah but the spike is VERY long 20:02 < hexnewbie> kubast2: Yes, anything moving in the Plasma panels (notification or tray icon) causes 100% CPU usage every time, even at low frame rates 20:03 < kubast2> that's a feature I guess lol 20:04 < phatcat_> koala_man: it just doesn't see the USB stick 20:04 < hexnewbie> kubast2: Again, haven't checked the GPU, but I suspect I would see usage there as well. Try killing the plasmashell, and see if it gets reduces. You can restart it with plasmashell5 or something. Check the start command before stopping it (there was also a way to quit it with kquitapp5 or something) 20:04 < spear2> koala_man: are COLUMNS and LINES environment variables at all or are they just "shell variables"? could it be possible to run printenv in some way to see them? 20:04 < hexnewbie> kubast2: Have only KDE 4 on the computer I'm on at present, so can't check the exact commands for Plasma 5 20:05 < bls> spear2: examine `printenv` vs `set` 20:05 < kubast2> yeah I had this command in krunner: it hanged pretty hard once killall plasmashell; kstart plasmashell 20:05 < kubast2> yeah 20:05 < kubast2> 3% 20:05 < spear2> bls: ah, ty =) 20:08 < gronke> Getting this error when configuring an install. Is there a package I can install via yum or? https://paste.pound-python.org/show/wd6oen2SDxwSGpLiBIDN/ 20:08 < bls> read the INSTALL instructions in whatever it is you're trying to compile 20:14 < gronke> bls hmm, okay, I'll take a look, thanks 20:18 < domhnall> phatcat: how are you creating the bootable usb drives? 20:19 < hexnewbie> phatcat: Which distros? 20:19 < domhnall> ^that too 20:22 < S_Gautam> I've tried writing the Windows ISO to USB for about 3 hours now and I can't seem to find any resources on how to do it. 20:23 < S_Gautam> I used UNetBootin, failed. I used MultiUsbBoot, failed (doesn't even recognize), I then formatted the disk with GPT (on GParted) and copy-pasted the contents of the ISO and still the computer doesn't read the ISO 20:23 < mawk> unless the official program to do that does magic (and then you might try Wine) you can dd it no ? 20:23 < illkitten> use dd 20:23 < koala_man> no, you can't use dd 20:23 < S_Gautam> I don't know how to use dd. 20:24 < koala_man> to boot an iso from usb it needs to be a hybrid image, and the Windows ISOs are not 20:24 < hexnewbie> S_Gautam: Have you tried WoeUSB? (Popular recommendation around here, is reported to work, haven't used it myself) 20:24 < S_Gautam> I used to use Rufus to write ISOs under Windows but there's no similar program to that 20:24 < S_Gautam> hexnewbie: I tried, apparently their github repo doesn't even build. 20:25 < mawk> let's see that S_Gautam then 20:25 < mawk> you tried to build master I guess ? 20:25 < mawk> try with the latest tag instead: https://github.com/slacka/WoeUSB/archive/v3.2.1.zip 20:25 < S_Gautam> alright thanks i'll do that 20:26 < BluesKaj> S_Gautam, https://www.ostechnix.com/how-to-create-bootable-usb-drive-using-dd-command/ 20:26 < mawk> if you're on debian you can install this dummy package to get all the build dependencies: https://github.com/slacka/WoeUSB/releases/download/v3.2.1/woeusb-build-deps_3.2.1_all.deb 20:26 < mawk> debian or derivated distro 20:27 < mawk> you do: sudo dpkg -i woeusb-build-deps_3.2.1_all.deb; sudo apt-get install -f 20:27 < hexnewbie> BluesKaj: I don't believe that has chances to work for Windows, or anything that isn't already a bootable USB (the formatting with vfat first is weird too) 20:28 < BluesKaj> Rufus usually works on windows 20:29 < BluesKaj> worked for me a couple of times 20:29 < mawk> on modified ISOs maybe 20:29 < kryptynasium> I wasn 20:30 < BluesKaj> not modded, ordinary kubuntu iso images 20:30 < BluesKaj> gotta use far 32 on the target usb iirc 20:30 < BluesKaj> fat 20:30 < mawk> ah, we're talkng about windows ISOs here 20:31 < kryptynasium> Ooops! I wasn't able to pull resource on Linux deployment - specifically related to ownership. What is the standard process. I know mysql creates a user and group and grants permission with the installation script/RPM. 20:31 < phogg> kryptynasium: It's not clear what you're asking. 20:31 < kryptynasium> Is there a resource someone can point out? Oracle does something similar - creates primary and secondary group (oinstall, dba) and then creates oracle user 20:32 < BluesKaj> assumed it was a linux iso, sinc ethis is linux support 20:32 < kryptynasium> phogg: Oracle installer adds user to group and gives appropriate ownership rights to install modules 20:32 < phogg> kryptynasium: So? 20:33 < phogg> kryptynasium: Let's start again. What Linux distribution are you using? What are you trying to do with it? 20:33 < kryptynasium> phogg: Is that the standard procedure? Or more appropriate way is adding current user to the group. 20:34 < kryptynasium> phogg: Linux distribution: RHEL 7 using RPM 20:34 < kryptynasium> phogg: I don't want user to explicitly run "chown" 20:34 < phogg> kryptynasium: No. That's what you're *asking* for. What are you *trying to do*? 20:35 < phogg> kryptynasium: Are you installing some software? What software? 20:35 < kryptynasium> phogg: I am creating an RPM to deploy my software 20:36 < phogg> kryptynasium: Okay, now we're getting somewhere! 20:36 < kryptynasium> Currently - post installation user has to explicitly run "chown -R localuser" 20:37 < phogg> kryptynasium: You should definitely make the RPM do that. Forcing the user to take manual steps should only be done when the package really cannot automate it. 20:37 < hexnewbie> kryptynasium: Is "localuser" an actual user account for a person (i.e. not a system account)? 20:37 < kryptynasium> Now - seeing Oracle and MySQL - they usually create ad hoc user for installation purposes either through script or they ask you to do it prior to installation - https://dev.mysql.com/doc/refman/8.0/en/binary-installation.html 20:37 < phogg> that said I am not familiar sufficiently with distribution policy on RHEL to be able to say what the standard procedure is 20:38 < kryptynasium> localuser is the currently logged user 20:38 < phogg> kryptynasium: mysql may ask for that, but RHEL-provided mysql packages do not. They create a user and group for you. 20:38 < hexnewbie> kryptynasium: So, not a system account like "mysql" or "postgres"? 20:38 < kryptynasium> I knew only IRC could help with this and not google 20:38 < prussian> distribute it as a tarball, let someone else bundle it 20:38 < phogg> kryptynasium: you should not need to or attempt to chown any files installed by an RPM to a non-system account 20:39 < phogg> kryptynasium: is your software a daemon or is it going to be run by a normal (non-root) user as a utility? 20:41 < kryptynasium> phogg: my software is a server - it can be run as daemon 20:41 < phogg> kryptynasium: if it expects to run as a daemon (and e.g. get started during system boot) then you should create a service account (UID <1000) and group to run it and own its files. 20:42 < hexnewbie> kryptynasium: What is this directory, part of your server, that you need to chown to give the localuser access and why? Why can't it use the user's home directory? Why does the user need to access it at all? What does the server do? 20:43 < kryptynasium> hexnewbie: default path is under /opt 20:43 < phogg> eww, opt 20:43 < phogg> /opt is a dumping ground for people who don't understand the FHS 20:45 < kryptynasium> hexnewbie: those are all important questions - MS ODBC drivers get installed under opt - so does sqldeveloper and perl5 20:45 < hexnewbie> And then get chowned to the local user account? Ouch 20:46 < prussian> you should be able to add a step to the rpm to chown 20:46 < phogg> kryptynasium: perl5 does not get installed under /opt/ in any sane distribution 20:46 < kryptynasium> hexnewbie: for the local user to run the exe's - they need to have permission 20:46 < prussian> it does when it might be a toolchain 20:46 < prussian> toolchain component 20:46 < phogg> kryptynasium: that doesn't mean they need to *own the files* 20:46 < phogg> kryptynasium: you just need to chmod o+rx, right? 20:48 < kryptynasium> phogg: correct - so you say standard procedure is to create group and add current user to group? 20:48 < phogg> kryptynasium: No. That's only for some things. Mostly you rely on world permissions. 20:49 < phogg> kryptynasium: Example: mysql DB files in /var/ are owned by user mysql group mysql. The mysql executable /usr/bin/mysql is owned by root group root, but permissions are 775. This way any user on the system has 5 (read+execute) access to the mysql client, and the client talks to the daemon which is running as user mysql and has write access to its DB files in /var/ 20:49 < hexnewbie> I've only seen such group permission for executing programs for /usr/games, and it was ages ago. 20:50 < phogg> kryptynasium: you normally only use a group to grant permissions when users need to share files between each other, not when users need access to programs 20:51 < phogg> kryptynasium: standard procedure is for each daemon to be run by a separate, dedicated user (and a group that no other users are in). This is good for security. 20:53 < kryptynasium> Yes my application in order to run requires write permission 20:53 < phogg> kryptynasium: write permission to what? 20:55 < kryptynasium> phogg: write permission to /dev/shm 20:55 < phogg> kryptynasium: your daemon requires that? 20:55 < kryptynasium> I am making use of shared memory 20:56 < phogg> that's neither here nor there 20:56 < kryptynasium> Yes - it is a requirement 20:56 < phogg> kryptynasium: do you have files somewhere in /opt which your daemon needs to be able to write? 20:57 < phogg> kryptynasium: I mean it's *irrelevant* to this discussion. 20:57 < kryptynasium> If shared memory gets accidentally created by "root" - it gets messy - it needs to be deleted 20:57 < phogg> kryptynasium: you should not be running your daemon as root 20:58 < kryptynasium> phogg: What I am saying is if someone accidentally started my exe as root - it would create it. I need to check for root user. bug 20:58 < phogg> kryptynasium: your daemon should detect root and drop privileges or abort with a fatal error. This is important for security but has nothing to do with packaging or standard file permissions practicers 20:59 < phogg> s/practicers/practices/ 20:59 < kryptynasium> phogg: Yes /opt/MySoftware/Logs - log directory 21:00 < phogg> kryptynasium: is there any reason a normal user would need to write to those logs? 21:00 < phogg> kryptynasium: and, by the way, you should put logs in /var/log/ like everybody else 21:00 < phogg> tl;dr follow the FHS or suffer later 21:01 < kryptynasium> The issue a lot of folks get desperate if something does not run - and then end up running stuff as root even though warned several times 21:02 < hexnewbie> kryptynasium: If your daemon runs as localuser, started by localuser, it can create /dev/shm/daemonname-localuser/ with *initial* permissions of 700 (KDE and X tend to do that in /tmp and /var/tmp), or better yet, /dev/shm/daemonname-localuser-XXXXXXX with mkdtemp() and store that somewhere in ~/.local/share/daemonname/. Logs should not be in /opt if they are created by something running as a local user account (in most cases) 21:02 < phogg> kryptynasium: that's why your daemon should upon start switch to its correct user and drop privileges. 21:02 < phogg> logs should not be in /opt *at all* 21:02 < phogg> nor should anything, really 21:02 < de-facto> when printing with lp from cmdline: how can i set a custom banner page for start and end of a job? 21:03 < hexnewbie> FHS mandates //var/opt/Vendor/MySoftware/logs ;p 21:03 < de-facto> lp option job-sheets= seems to only accept predefioned banners? 21:04 < kryptynasium> phogg: Sure - I will fix that - I am a newbie on Linux etiquette. Ok no logs in /opt 21:04 < djph> you have to create said banner page(s) 21:04 < phogg> kryptynasium: the reason is that /opt could be on a read-only filesystem. You should only assume that /tmp and /var are writeable, other places may not be. 21:05 < phogg> hexnewbie: it also mandates /etc/opt/ for opt-installed config files, but very few people respect that 21:05 < phogg> hexnewbie: most likely because anybody using /opt is unlikely to be familiar with the FHS in the first place! 21:05 < djph> /opt annoys me 21:05 < de-facto> djph, do you know if i can create those dynamically, lets say "JOB XY for John Doe begin" and "JOB XY for John Doe end"? 21:06 < kryptynasium> I think this clears up a lot of my confusion. I really don't need a chown on root software dir at all. Except for "log" but if that is moved out - then no point in chown-ing 21:06 < djph> de-facto: would probably require you re-writing 'lp' to be a customized script that generates the file(s), then calls 'real-lp' 21:06 < djph> de-facto: it's honestly been a stupid long time since I've used LP 21:06 < djph> *LPR 21:07 < phogg> kryptynasium: you will need to set the owner of a log dir or file in /var. 21:07 < de-facto> djph, so its not as simple as droping a dynamically generated file under a custom name in a dir and give that as job-sheets= in the lp command? 21:08 < kryptynasium> djph: if not /opt then? 21:08 < Ryvius> I have an idiot problem, I can't seem to figure out how to get this .desktop shortcut to work. The .desktop file has "Exec=wine "/home/ryvius/.wine/drive_c/Program Files (x86)/Diablo II/Game.exe" -3dfx" but no matter how I formulate it, it says either "exec format error" or "file not found" 21:08 < djph> kryptynasium: putting things in the right place 21:09 < phogg> kryptynasium: you can place logs in a file starting with the prefix /var/log/ or in a dir /var/log//log, but it is so uncommon I think you will be forgiven for not doing that. 21:09 < TJ-> de-facto: could you use FD redirection, as in "lp -o job-sheets=<( echo "JOB XY for $USER begin"),<(echo "Jo XY for $USER end") ... " 21:09 < djph> de-facto: Offhand, I don't think so (I mean, you have to create the file first) ... but I'd have to dust off the lpr manuals I ahve 21:09 < phogg> kryptynasium: http://www.pathname.com/fhs/pub/fhs-2.3.html 21:09 < fattredd> If I'm looking to migrate my install to a new hdd, I can just dd the partition over, and it will boot, right? 21:09 < djph> fattredd: "it should" 21:10 < TJ-> fattredd: not just the partition, but the boot files/sectors, too 21:10 < ayecee> mostly 21:10 < dgurney> fattredd, pretty much, though do keep in mind that partitions will not resize automatically 21:10 < phogg> kryptynasium: this is the Filesystem Hierarchy Standard. It describes dirs found in a standard Unix filesystem, what each of them are for and what you should put where. It also mentions /opt but only as a way of saying "use /opt if you can't be bothered to follow these rules." 21:10 < fattredd> So dd if=/dev/sda of=/dev/sdb? 21:10 < dgurney> sure 21:10 < kryptynasium> phogg: Thanks for the resource - I knew someone here will take me to the "promised land" 21:10 < fattredd> Will that carry over the partition data? 21:10 < TJ-> fattredd: so if UEFI then you'd need an EFI-SP and use the simple boot path (/EFI/BOOT/BOOTx64.EFI) 21:10 < fattredd> and the uuid? 21:11 < phogg> kryptynasium: all distributions follow this standard to some degree, mostly pretty closely. Importantly: Admins expect it. If you follow it you will not surprise them. 21:11 < jim> you -could- do that, and it would work... it will literally copy the filesystem 21:11 < TJ-> fattredd: whole disk copy should work if the new firmware uses the same boot protocol (UEFI or BIOS) 21:11 < jim> so you don't even need to format the destination first 21:11 < fattredd> Cool. jim, you seem like you don't like the idea of dd. Any other suggestions? 21:12 < jim> well see, I have to back up, because I don't know what you want to do... and expressing my preference is a waste of windbaggery 21:13 < fattredd> That's fair lol. I have an old hdd that is on ther verge of failing. I'd like to migrate to a brand new ssd without reinstalling everything 21:14 < jim> if I did, then I could say, well you should image like you're doing, or you should do file-by-file copying 21:14 < dgurney> fattredd, partition and format the new drive, copy over all files from the old one (making sure to preserve all permissions and attributes!), then fix fstab and bootloader (which is necessary if you use UUID) 21:14 < dgurney> that's a better approach than just DD, unless the disk size is 100% identical 21:14 < kryptynasium> phogg: hexnewbie djph Thanks! 21:14 < fattredd> Yeah okay 21:14 * justsomeguy likes fsarchiver or CloneZilla for this task. 21:14 < fattredd> That makes sense 21:15 < TJ-> or copy partitions to LVM LVs for flexibility 21:15 < jim> if you image it (i.e., with dd) you don't have to worry about the perms and attribs, because those are in the filesystem too 21:16 < fattredd> If I were to migrate to LVM, would I need to change much other that fstab? 21:16 < fattredd> *than 21:16 < jim> maybe that and how you boot? 21:16 < TJ-> fattredd: re-generate the initrd.img to ensure it includes the lvm tools as well 21:16 < fattredd> Gotcha. Thanks guys 21:16 < jim> what TJ- said :) 21:17 < TJ-> fattredd: the thing is, you mentioned the HDD is failing. If that includes corruption then you could copy already corrupted sectors over to the new device 21:17 < TJ-> fattredd: in which case a clean install and copying over settings/docs would be preferable 21:18 < jim> the thing about regening the initrd, it should look at how / is mounted, and establish the conditions to allow that at boot 21:18 < fattredd> TJ-: Any way to copy those settings w/o hating my life? 21:18 < jim> well if you haven't done lvm before, it takes some getting used to 21:19 < TJ-> fattredd: depends in which distro for the specifics, but generally it's all of the $HOME directory contents, plus all of /etc/ and possible some from /var/lib/ 21:20 < justsomeguy> Just thought I'd mention that fsarchiver lets you do file-level copies of filesystems, but retain xattrs, acl's and other stuff by default. Since it's file based, it doesn't rely on the partition you're restoring to being a certain size. 21:20 < jim> in the initrd, you would need... the module for the disk controller, the module for the filesystem that / is formatted as, and the module for lvm 21:20 < bls> and you don't have to worry about dealing with empty space, disk UUIDs, etc 21:21 < TJ-> fattredd: When I need to do this, on Debian/Ubuntu, I first do a "debfoster --show-keepers" to get the list of top-level installed packages. On the new target I copy over the /etc/ then install all those packages so the package manager asks what to do if there are any custom conf file entries 21:21 < jim> can I ask what the goal here is? 21:21 < fattredd> Yeah I'm thinking reinstall would be easiest 21:21 < fattredd> Goal is a running system with all my old services/files 21:22 < jim> ok, and what system are you running now? 21:22 < fattredd> Ubuntu 18.04 Server 21:23 < jim> is it that you want to change to a different linux dist? or keep the same? 21:23 < TJ-> jim: the issue is a failing HDD so migrate to new SSD I think 21:23 < fattredd> That's right 21:23 < jim> ugh... 21:24 < jim> so you're looking to save whatever you can 21:24 < TJ-> fattredd: I'd suggest to keep it simple do a while-disk 'dd' to begin with, keep the original HDD safe, if DD results in problems then try a different method 21:24 < fattredd> I've got the majority of my files from ddrescue 21:24 < TJ-> fattredd: you've already had to use ddresuce? 21:24 < Minnebo> is there like a apache channel with people in it? 21:25 < Desu> #httpd 21:25 < TJ-> fattredd: so doing a 'dd' isn't going to be the best solution since there may be uncorrectable sectors with random/bad data in 21:25 * graff read that as "normal people" 21:25 < jim> Minnebo, yep, that. and, there is a bot, alis, that can assist you in looking for channels on the Freenode irc net. To start, /msg alis help 21:25 < fattredd> Yeah, I was crossing my fingers that the bad sector wasn't important 21:26 < fattredd> But that's not good practice, so I guess a fresh install is the best option 21:26 < TJ-> fattredd: I'd say so; long-term it may well save you random problems 21:26 < fattredd> Looks like ddrescue got all my config though, so I guess it shouldn't be too bad to move over 21:26 < TJ-> fattredd: you can use 'debfoster' as I described to quickly generate a list of the packages you need to install on the target 21:26 < Minnebo> jim which one is it? :D 21:26 < fattredd> That's a solid tool. Thanks 21:27 < Minnebo> I find two apache channels with 3 people in it ^^ 21:27 < jim> ok, here's what I'd do (like right now): use file copy to copy your /home dir to a partition (or lvm lv) 21:27 < Taoki> https://www.ebay.com/p/AMD-Radeon-R9-390-8192-MB-R9390GAMING8G/220498527?iid=401535223159&chn=ps Could anyone please tell me if this video card is compatible with the latest Linux Kernel as well as version of MESA 3D? I can't find its exact GCN version and GPU family. 21:27 < jim> Minnebo, this is for apache2? 21:27 < Minnebo> ah 21:27 < TJ-> fattredd: if you're working from a ddrescue-ed image, maybe in a chroot?, then you'd need to install debfoster first then do "sudo debfoster -q" then you can get the list with "debfoster --show-keepers > /tmp/installed.list" 21:27 < Minnebo> yes 21:28 < graff> I would try to duplicate the entire partition with dd instead of using cp 21:28 < graff> and then run the rescue on that. 21:28 < Minnebo> I just needs some help configuring httpd.conf to secure more : ) 21:28 < jim> then (I dunno why they named it that but) the channel you want is #httpd 21:28 < graff> cp has to traverse the filesystem so it might causes accesses and updates 21:29 < jim> so Mr, More is feeling insecure/. 21:29 < jim> ? 21:29 < graff> I think dding the whole partition has the least chance of causing an inode update 21:29 < fattredd> graff: I' 21:29 < fattredd> ve already ddresue-ed the whole thing 21:29 < graff> fattredd: sharing my opinions in general 21:30 < fattredd> TJ-: Yeah I'm going to do this. Thank you 21:30 < jim> cool... if you did if=/dev/sea of=/dev/sdb, what you've done is copy the -drive- 21:30 < wonderer> hi all, am getting this repeating in my auth.log all the time https://pastebin.com/vvpEAfcR 21:30 < wonderer> can i reduce this? 21:30 < fattredd> Thank you all 21:31 < jim> wonderer, do you know who this user named pi is? 21:32 < TJ-> jim: the channel is named after the package (httpd) - Apache is the foundation that manages it and other projects :) 21:32 < wonderer> its on my raspberry pi 21:32 < Desu> don't constantly log in/out as user pi? 21:32 < jim> ok... yeah, it's just keeping track of what user logged into what, and what IPs failed to log into what 21:33 < wonderer> Desu does look like it 21:33 < wonderer> but ofc i am not doing this 21:33 < Desu> someone is 21:33 < wonderer> at best i log in a few times a day 21:33 < jim> ohh, really 21:33 < Desu> hint, the cron job 21:33 < TJ-> wonderer: it's a "cron" job running as the user 21:34 < wonderer> Desu yes, and here it is https://pastebin.com/94iF70xk 21:34 < jim> oh ok for a second there I thought he had a serious problem 21:35 < wonderer> i thought my times are at worse a few times an hour with all cron jobs added 21:36 < Desu> there are better ways to do keepalive 21:36 < Desu> which distro is this? 21:36 < jim> wonderer, and you didn't know this, so not your fault... and, pastebin.com does some questionable stuff, like has ads, capchas and lots of javascript (enough to crash a browser running on a machine with half gig of ram... like every time... so, 21:37 < jim> wanted to offer you a couple other choices, one, is in the topic of this channel, the other: you can pastebin the output of an arbitrary command by running "anArbitraryCommand | nc termbin.com 9999", and to include error messages, "anArbitraryCommand 2>&1 | nc termbin.com 9999" 21:37 < wonderer> i think i got an answe in rpi channel 21:38 < wonderer> jim 21:39 < wonderer> thats quite a specific/advance way of doing t 21:39 < wonderer> not sure i know how 21:40 < jim> stick around, I'll show you (it's easy) 21:40 < wonderer> well thx 21:41 < wonderer> and any other time i would but limitted on time 21:41 < Minnebo> Would you guys like disable SSLv2/v3 on webservers? 21:41 < wonderer> but i add these cmds to my wiki 21:42 < alipoor90> virt-manager pflash segments cannot be mapped under 21:42 < alipoor90> Hi, when i want to make a new UEFI virtual machine using virt-manager i get this error: https://paste.fedoraproject.org/paste/woHIID6y8~-t13rxSJg7Pw 21:43 < toothe> Anyone have this "Started Update UTMP about System Runlevel Changes" lockup? 21:43 < toothe> I am on Kali (um...Debian) 21:43 < toothe> I believe its a video graphics driver issue. 21:44 < toothe> Must be from my recent upgrade. 21:44 < Desu> read logs 21:44 < revel> toothe: Is Kali well-known for its stability? 21:44 < toothe> revel: ...no comment 21:44 < toothe> but, this is a Debian/Ubuntu issue. 21:45 < Desu> no, it is a kali issue 21:45 < toothe> what makes you say that? 21:45 < Desu> kali is not debian nor ubuntu 21:45 < domhnall> hah, somafm scared the mess out me just now, froze for like 1/2 hr, forgot it was even streaming. 21:45 < revel> Since it's related to Kali packages, most likely. 21:46 < toothe> hm...this is also a Debian issue per Google. 21:46 < Desu> no, since you are running kali it is a kali issue 21:46 < Desu> it's like going to citroen and ask them to fix your pegaut for free 21:46 < toothe> Desu: Kali is based off Debian. 21:47 < Desu> s/ask/demand/ 21:47 < Desu> and then argue when they tell you to go to the pegaut dealer 21:47 < domhnall> toothe: Debian-testing 21:49 < jim> wonderer, (aside, please spell thx out as thanks, helps with understanding by observers) ok, do you mean the pastebin method, is advanced? 21:52 < Desu> Minnebo: I would not allow anything but tls >=1.2 21:52 < Desu> Minnebo: there is no valid reason to accept anything else 21:53 < revel> How old is TLS 1.2 now anyway? About 10 years old? 21:53 < Desu> something like that 21:53 < Desu> the rfc came out in 2008 21:54 < ayecee> barely broken in 21:55 < wonderer> jim thanks :) - better? 21:55 < wonderer> yes/no pastebin method using output from cli 21:56 < wonderer> cli utput you suggested never seeen before 22:01 < arooni> anyway to modify this command so that i can get some space between the output of the two elements being combined? it seems to remove all spaces echo -n $(timew get dom.active.tag.1 | cut -c 1-35 | sed -e 's/$/:: /'); echo $mytime | sed " s .. " 22:01 < Desu> use awk 22:02 < Desu> just what format does timew spit out anyway? 22:02 < domhnall> toothe: also, https://bugs.kali.org/view.php?id=4536, tl;dr, one solution was reinstallation. 22:05 * domhnall inclined with it being specific to distro 22:06 < Desu> actual logs would help debug it 22:06 < domhnall> true 22:06 < Desu> but posting logs would be useful 22:07 < revel> lol 22:09 < domhnall> hm, second kali mention afaik, today, Mr.Robocop back on or something? 22:09 < dgurney> must be 22:10 < revel> I heard it derailed into a bunch of time travel stuff. 22:10 < bls> heh "how do I install kali to my hard disk and make it look pretty and use it as a daily driver for things other than pentesting so pentesting just kind of rubs off on me" 22:11 < revel> lol 22:12 < dgurney> well, the first step is to discard the advice of the kali developers/maintainers themselves ;P 22:12 < Desu> "I'm too lazy to ever read any documentation but I heard Kali magically makes you a 500k€/year pen tester and makes your e-panis 30cm longer" 22:13 < Desu> sums up my experience with most kali users 22:13 < domhnall> nice 22:13 < dgurney> Kali is such a meme that it even gives you a knockoff e-penis? damn 22:13 < justsomeguy> Pfft, arch linux gives you the best e-penis. 22:14 < domhnall> uuh okay. 22:14 < domhnall> So...how about that UbuntuBSD 22:15 < domhnall> :) 22:15 < revel> justsomeguy: So if you want to get some e-penis, you use Arch...? 22:17 < justsomeguy> Arch makes you install grub by hand, whiceans your a real sysadmin. 22:17 < justsomeguy> *which makes 22:17 < revel> lol 22:18 < Desu> when I installed RH the first time you lireally had to mknod pretty much everything yourself 22:18 * justsomeguy is irrational worried that he pissed off some archers with his bad joke just now. 22:18 < justsomeguy> That sounds crappy, Desu 22:19 < domhnall> justsomeguy: nah, arch is a tough community. 22:19 < justsomeguy> Good, good. 22:19 < ayecee> it was so bad, even the non-archers are getting the pitchforks 22:19 < Minnebo> Desu, thanks, I did and I have an A now on SSLLABS! 22:19 < Desu> the good old days 22:19 < Minnebo> yay : ) 22:19 < cyphex> I'm not an avid supporter of arch, but installing it for the first time made me learn how partitions work 22:20 < Minnebo> no A+ but it needs HSTS and i'm afraid to put it on ^^ 22:21 < justsomeguy> cyphex: I learned a ton when installing gentoo, tbh. I think there is a value in those kinds of distros, but people advocating their use for completely the wrong reason. 22:21 < dgurney> cyphex, yeah, that's a strong point of non-automatic installers 22:21 < justsomeguy> Wow, I'm making so many typos today. 22:22 < useless-eater> http://paste.debian.net/1027290/ should I start looking for another nvidia card, or could this be a buggy driver? Using the nvidia-common packaged for debian stretch 22:22 < cyphex> gentoo taught me quite a bit too, but I was much more proficient with linux when I installed it 22:22 < pankaj> justsomeguy: How is gentoo different from arch? 22:22 < cyphex> it's a lot more hands on 22:23 < Desu> useless-eater: most likely a driver/packaging issue 22:23 < Desu> what card is this though? 22:23 < Dagmar> If you're actually _still_ running an 8350 card tho', maybe if you play games it's time to upgrade 22:24 < justsomeguy> pankaj: The package manager is completely different, and it uses openrc - a different init system. I mentioned it in relation to arch because they both have very manual install processes. 22:24 < Desu> Dagmar: not to mention that nvidia doesn't support them anymore 22:24 < dgurney> however, it should be noted that while Gentoo's primary init system is openrc, systemd is fully supported and works well 22:24 < Desu> fully supported can be argued 22:25 < Desu> (in gentoo) 22:25 < useless-eater> Desu: thanks 22:25 < useless-eater> Desu: "NVIDIA Corporation GM206 [GeForce GTX 960] (rev a1)" 22:25 < justsomeguy> I found that systemd worked as well as it does on any other distro, from my limited experience with it as a gentoo user. 22:25 < pankaj> justsomeguy: I have installed arch now 3 times. Should I go to gentoo or will it be the same experience or learn more in gentoo? 22:25 < dgurney> pankaj, gentoo is worth trying 22:26 < cyphex> you will learn more in gentoo, but you might not like it as a daily driver depending on what you do on your computer 22:26 < dgurney> but if you're happy with arch, there's no need to hurry replacing arch 22:26 < justsomeguy> pankaj: I think if you are installing distros in order to learn things, you might get more value out of spending that time on the free linux system administration course by the linux foundation on edx. 22:26 < dgurney> justsomeguy, can confirm. I use systemd on most of my Gentoo systems, and it works without a hitch 22:27 < Desu> exherbo is imo much better, it is more like what gentoo used to be before the politics took over 22:28 < justsomeguy> pankaj: But you will at least get a grasp of how to compile a kernel (and other boot process stuff) if you follow the installation instructions for gentoo, which I have found useful. I guess it's just a matter of priorities. I can't say I didn't have fun tinkering with it. 22:28 < Desu> if you want something source based that is 22:28 < justsomeguy> Fyi, there is a binary distro called calculate linux that allows you to install gentoo without the manual install process, and you can still use the package manager with gentoo repos. 22:29 < pankaj> justsomeguy: OK. I knew that course on edx. There were (i think) 2 of them which were free that I completed. And I will definitely try gentoo. May be now but I am installing something. 22:29 < jim> wonderer, still here? (I'm a bit cramped for time atm) 22:29 < Taoki> https://www.msi.com/Graphics-card/R9-390X-GAMING-8G Can anyone please let me know if this card will use the amgdpu driver by default? 22:29 < justsomeguy> Have fun, pankaj! 22:29 < cyphex> make sure you read the handbook though pankaj 22:29 < jim> and, yes, better :) 22:29 < revel> Desu: Have you used it? 22:30 < pankaj> justsomeguy: Can you suggest some good resouce to get started with kernel modules (writing) and configuring? The resources on the internet that I google; much of them are based on old linux kernel version. 22:30 < Desu> revel: I've used both exherbo and gentoo 22:30 < Bashing-om> useless-eater: What version driver is intalled: http://www.nvidia.com/download/driverResults.aspx/134262/en-us . 22:30 < Desu> revel: used gentoo back in the days before all the good devs left due to the politics too 22:31 < revel> Last I heard, Exherbo its user and developer base are the same. 22:31 < Desu> revel: missing some s's there 22:31 < revel> I think I some words there. 22:31 < justsomeguy> pankaj: That's a completely different ball game. Hopefully some other people can chime in with advice. I guess I'd start by googling the kernel-newbies wiki. 22:31 < TJ-> pankaj: have you checked out the driver API implementors guide, and other sections, of https://www.kernel.org/doc/html/latest/index.html 22:31 < revel> Desu: Which good devs? 22:32 < pankaj> TJ-: Going to it. 22:33 < jim> wonderer, if you happen to be around, try this: echo "should I pastebin this" | nc termbin.com 9999 22:33 < triceratux> https://dev.gentoo.org/~mgorny/articles/the-story-of-gentoo-management.html 22:34 < jim> wonderer, the nc command will give you back a url... look at it in your web browser 22:34 < mattfly> hey 22:34 < mattfly> does anyone know how can i get more languages support with gocr? 22:34 < jim> wonderer, pretty much that simple :) 22:41 < Taoki> Any thoughts on my question too please? 22:42 < ayecee> what do you think of my question? 22:42 < Happyhobo> Hi folks 22:43 < Happyhobo> Hi ayecee 22:44 < Happyhobo> I'll try and not be so confusing today but I switched to Peppermint which is nothing more than glorified Ubuntu which is nothing more than screwed Debian. Now the HDMI outputs aren't in my list of sound options and my sound isn't even working with my built ins. 22:44 < Bashing-om> Taoki: Yes, see: https://support.amd.com/en-us/kb-articles/Pages/AMD-Radeon-GPU-PRO-Linux-Beta-Driver%E2%80%93Release-Notes.aspx . 22:45 < pankaj> exit 22:46 < Happyhobo> I've never had as many problems as I've had in the past two weeks. Shoot dependency Hell with Mepis was fun in comparison to this bulmalarkey. 22:46 < Happyhobo> brb let me try rebooting I'll windows this witch. 22:47 < triceratux> HappyHobo: np. thats why we are here 22:50 < jim> Taoki, I wonder if ##hardware has some additional info? 22:51 < Taoki> Oh... could ask there as well. Though this is a more Linux specific question 22:51 < Taoki> Bashing-om: I was thinking of the buildin amdgpu driver, not the pro one. Though I'd assume card support is similar between them 22:51 < jim> maybe it is... I've had them handle linux driver questions before 22:51 < Happyhobo> Well now the notebook's sound works but the volume control gives one choice hDMI display port unplugged. It's plugged into the TV I'm watching videos. 22:52 < jim> Happyhobo, could you see if this gives output: lspci -nn | grep -i audio 22:52 < Happyhobo> yes sir 22:52 < Taoki> I am seeing it listed there though 22:52 < TJ-> Happyhobo: I've seen such issues caused by ACPI bugs in the PC firmware, with more recent Linux kernels exposing the issues 22:52 < jim> my dad always used to say "don call me sir, I'm just as good as you are" 22:53 < Happyhobo> lspci -nn | grep -i audio 22:53 < Happyhobo> 00:1b.0 Audio device [0403]: Intel Corporation 6 Series/C200 Series Chipset Family High Definition Audio Controller [8086:1c20] (rev 04) 22:53 < Happyhobo> 01:00.1 Audio device [0403]: NVIDIA Corporation GF119 HDMI Audio Controller [10de:0e08] (rev a1) 22:53 < Bashing-om> Taoki: -pro is the proprietary overlay on top of amdgpu . If the card supports the overlay ..in great shape .. and your prospective card does :) 22:53 < Zharf> "sir" has that sort of connotations? (never used the word myself..) 22:54 < Happyhobo> I have concluded that this notebook just sucks with wireless regardless of the card. 22:54 < jim> Zharf, my dad was of course making a joke, partly (I would imagine) to get the other person to relax a bit 22:55 < Taoki> Bashing-om: So card support is never exclusive to the pro driver, yes? Just making sure 22:55 < TJ-> Happyhobo: there's an ACPI workaround can often solve issues like this, we see them alot on Ubuntu. I've an article explaining how to figure out the workaround at http://iam.tj/prototype/enhancements/Windows-acpi_osi.html 22:55 < jim> Happyhobo, then I wonder if it' 22:55 < jim> is about the antennas 22:55 < Happyhobo> so now I've moved on to other issues. Be nice if I could watch youtube on the TV and have the sound come from the stereo that the TV is hooked to for sound because the factory speakers are unpleasant. 22:56 < Happyhobo> I have the black on the black, white on the white and gray unplugged. 22:56 < Bashing-om> Taoki: AMD gave us what we wantsed .. open source driver .. and the effort to backport to older cards/drivers is an on-going effort that will not go back very far . 22:57 < Happyhobo> ACPI TJ- 22:57 < jim> Happyhobo, because there are only two connectors on the card? 22:57 < Happyhobo> Yep 22:58 < Bashing-om> Taoki: The AMDGPU-Pro Driver is only campatible with the later generation cards . 22:59 < jim> Happyhobo, maybe try different ones in different sockets.. maybe black on white or gray on white, see if it gets better (and be careful about plugging them in) 22:59 < Taoki> Bashing-om: Is it the case the other way around though? Just need to be sure that this card works under the amdgpu driver that comes with Linux 22:59 < Taoki> Not planning on getting the pro one 22:59 < dgurney> from what I know regular amdgpu should work with pretty much everything, provided the kernel is up-to-date 23:00 < Happyhobo> They're too hard to plug and unplug 23:00 < dgurney> well, pretty much everything modern anyway 23:00 < Happyhobo> My friend had to do it and had to gently use pliers to get them to click on. 23:02 < Happyhobo> Right next to the router I get 70 but here in my room I get .5 to 16 and the better performance is only for a minute or two. 23:03 < Happyhobo> is my kernel the issue Jim 23:03 < Happyhobo> shane@Paxton-City:~$ uname -r 23:03 < Happyhobo> 4.15.0-1006-oem 23:04 < jim> Happyhobo. are you saying it works better with other OSes? 23:05 < Desu> Taoki: there is no reason to use amdgpu-pro over amdgpu unless you are mining 23:05 < Taoki> Ok. So it will support my card then 23:05 < Desu> Taoki: amdgpu-pro is catalyst all over again 23:05 < Desu> it is horrible 23:05 < Happyhobo> The other card the OS made a difference, this one no. 23:06 < Happyhobo> We tried to roll a kernel on this unit before jim and it didn't work out too well. 23:06 < SpeakerToMeat> Any idea why lftp would be able to download individual files, but get "stuck" doing mirror or mget *.mxf for example? 23:07 < jim> Happyhobo, ok, on the card that made a difference, what was the other os, and what difference did it make? 23:08 < jim> SpeakerToMeat, would rsync work better? 23:09 < SpeakerToMeat> It would, if remote offered rsync :( 23:09 < Happyhobo> The other OS was Antergos, it connected but it was very unstable, other OSes with that card wouldn't connect at all not Mint, not Manjaro, not any of them. This card is the same with everything booom booom fast fast fast stay connected drop to nothing then fast again. 23:09 < Happyhobo> Sometimes the nothing lasts a while. 23:11 < JamesHarden> hi, I have some text files where the sentences are broken at completely the wrong places. I want to break it like a sentence per line, ended at the period. can awk/sed do this? 23:12 < pankaj> I need to download bios and some drivers from my device manufacturers site and it is demanding information. Unfortunately, the sticker at the back that was required for it has been blurred. Is their any command in linux that gives manufacturer related information? 23:12 < pankaj> I googled but most of them are windows specific ways. 23:13 < justsomeguy> JamesHarden: You might want to look at using the `fmt` command for that, instead. 23:13 < Bashing-om> Taoki: That R9 390 card is well supported. the driver - amdgpu - is in the kernel. 23:13 < m4estr001> Hey guys, i guess this question may be inappropriate. But i was using linux for quiet a bit, now caught some time to learn about it and, now i got confused. Here is my confusion: Hardi disks contain partitions, regions that contains information, information like files, directories etc..... Where i got completly confused is mounting partitions to some directiories, how if partitions cotain data can it be mounted to different partiti 23:14 < m4estr001> Just in different directory? 23:14 < djph> m4estr001: the directories are an abstraction of what's on the drive. 23:14 < pankaj> Hey, I also have a problem. 23:14 < JamesHarden> justsomeguy: will check it out, thanks! 23:15 < m4estr001> djph: so informations, that are on HDD - one partition lets say /dev/sda1 will be in directory /mnt and if moutned /mnt/123 23:15 < djph> m4estr001: sure 23:15 < Happyhobo> I hate this computer but I love this computer because it was a gift and it's the only 15.4 or .6 I own, 23:16 < Taoki> Bashing-om: Thanks. My current card only works on radeon by default, I need kernel parameters to use amdgpu. That's because it's a GCN 1.0. This one is a GCN 2.0... I was told that it too is not supported just yet. 23:16 < Happyhobo> Wireless issues, sound issues, bought the wrong damn keyboard off of Newegg so the keys are shaped funny. 23:17 < justsomeguy> JamesHarden: Glad I could help. Here's a url to the documentation, btw. https://www.gnu.org/software/coreutils/manual/html_node/fmt-invocation.html#fmt-invocation 23:17 < m4estr001> djph: so if i make partiton /dev/sda1 and mount it to /boot/efi , if i want t to be my efi partition, other os i mount it to lets say just /boot will recognize those information just set in /boot directory? 23:17 < djph> yes 23:17 < m4estr001> djph: in that case why i cant have two /boot directories on one partition for 2 oss? 23:17 < djph> what? 23:17 < bls> nothing's stopping you, but they'd have to stay in sync 23:18 < bls> or the bootloader config would at least 23:18 < djph> you can't have two directories on a single physical partition with the same name 23:18 < m4estr001> I read that it is problematic to have 2 linux kernels on one partition in same directory 23:18 < m4estr001> djph: what do you mean "physical partition" 23:19 < m4estr001> bls: so it is possible? 23:19 < bls> you're interchanging unrelated words 23:19 < dgurney> huh? 23:19 < djph> that is if you have partition /dev/sda1 (a physical partition), mounted to /mnt/sda1 - there can only ever be one /mnt/sda1/boot directory 23:19 < m4estr001> If partitions are just abstraction of what is on HDD how can one be physical? sorryi know those are dumb quest 23:20 < m4estr001> djph: much clearer now 23:20 < m4estr001> thanks all guys 23:20 < djph> the DIRECTORIES you see are a bit of an abstraction 23:20 < bls> if I understand what you're asking, yes, it's possible to have distro A's /boot on partition sdX1 and distro B's /boot on partition sdX2 23:20 < m4estr001> bls: i just wanted them on same partition 23:20 < m4estr001> bls: EFI, partition 23:20 < justsomeguy> T 23:20 < djph> to UNIX (and UNIX-like) systems, everything starts from the root directory ("/") 23:21 < bls> then no, a partition with a filesystem can only have a single root/mount point 23:21 < justsomeguy> Thats the default behavour for most installers, m4estr001. All OS's on a single system will discover the EFI partition and put their boot related files on it. 23:21 < djph> ... so you end up having a bit of an abstraction where "/" itself may be /dev/sda1; but "/usr" is actually the partition /dev/sdb1 ... 23:22 < djph> (or /usr is /dev/sda3, or similar) 23:22 < bls> you could try to share /boot between two distros, but there's not much point, and dual booting multiple distros is generally lots of annoying for little to no value 23:22 < m4estr001> Mhm, so what is simplest case scenario, for having 2 OSs , i want to keep it simple for now while learning 23:22 < bls> simplest case? don't dual boot. 23:22 < m4estr001> bls: i want one distro for testing stuff, and one to keep school files safe 23:23 < m4estr001> bls: i have ended alot time in kernel panic page 23:23 < justsomeguy> Maybe you should use a virtual machine for testing, instead. 23:23 < Happyhobo> jim I want a new notebook 23:23 < graff> dual booting has nothing to do with sharing a /boot dir 23:23 < m4estr001> justsomeguy: not a great laptop for that 23:23 < graff> in fact that seems like a really bad idea, for obvious reasons 23:23 < jim> Happyhobo, didn't you say you rent net from your neighbor? 23:23 < m4estr001> graff: maybe obvious for You, not for me at the moment of asking 23:23 < graff> m4estr001: ? 23:24 < bls> graff: it seems to go hand in hand with people that want dual boot linux distros...sharing as many partitions as possible 23:24 < graff> m4estr001: i am not talking to you, please don't try to troll 23:24 < m4estr001> graff: sorry, that was not my intention at all 23:24 < graff> bls: no, sharing a /boot has nothing to do with dual booting 23:24 < justsomeguy> m4estr001: Ah, well, you can use a container instead. It is less resource intensive, but still gives you the ability to run a different linux system on top of your existing one using the same kernel. 23:24 < graff> it's unrelated 23:24 < Happyhobo> Yes, I went and sat in his room with both of my notebooks, the one got 35 Mbps according to speetest.xfinity.com and this one got 70 Mbps jim 23:24 < bls> graff: yes, that's what we've been trying to explain 23:25 < m4estr001> Okay, gonna go try and see how linux "reacts" to all those changes. Thanks for patience and answering. You are very kind i really appricate that! 23:25 < graff> m4estr001: the obvious reason being that both distros need to share the file namespace. 23:26 < graff> in dual booting scenarios the bootloader has a concept of where the bootables kernels are on the actual filesystems 23:26 < graff> meaning it has the ability to mount filesystems itself 23:26 < jim> Happyhobo, so plenty of performance while sitting in his room? 23:26 < Happyhobo> Today in my house in my bedroom with his notebook he got 25 and I got 11, my netbook got 16. This one is only good if you're sitting on top of the router with it almost up your posterior jim 23:27 < m4estr001> graff: i guess that has to do something with boot process 23:27 < graff> so when you tell grub or syslinux that your kernel is /boot/mykernel and that is on /dev/sda1 then it actually opens up /dev/sda1 in order to boot it 23:28 < graff> so in a dual booting scenaro you might have one distro on /dev/sda1 and another distro on /dev/sda2 23:28 < m4estr001> graff: right 23:28 < m4estr001> graff: thats what i was trying to do 23:28 < bls> the common query we get is that can you have /boot/debiankernel and /boot/ubuntukernel and /boot/centoskernel all on the same partition 23:28 < Happyhobo> 9.3 up 10 down just now jim 23:29 < alexandre9099> hi, not sure if i may ask stuff about licenses here, so there is an android app that is licensed under lgpl-3, that app is made to work with one service, can i make a fork so that the app works for other service? (with the same codebase) 23:29 < graff> bls: that seems like a really useless booting arrangement our of all of the hundreds of different possibilities 23:29 < graff> out of all * 23:29 < bls> alexandre9099: should probably try #gnu 23:29 < alexandre9099> yep, heading to there, thanks 23:29 < bls> graff: and it is, but it's tough to talk people out of it 23:29 < m4estr001> So how i imagined my partitions is next: /dev/sda1 - EFI partition, /dev/sda2 - SWAP (sharred or not doesnt matter now), /dev/sda3 for distro A and /dev/sda4 for distro B 23:29 < jim> Happyhobo, from where? 23:29 * graff feels that he did a good job 23:30 < graff> in that sense 23:30 < Happyhobo> xfinity speedtest out of Richmond, VA 23:30 < jim> but I mean, where were you when you ran the test? did you run it from your laptop? 23:31 < Happyhobo> The last one was right here as we spoke from the Dell E6520. 23:32 < m4estr001> So now, i have one more question, do i want my bootloaders kernel to be installed to EFI partition 23:32 < Happyhobo> The Dell 2120 Netbook has a G card and is just as fast as this one and a lot of times faster; this is supposed to be an Ultimate-N. 23:32 < jim> Happyhobo, you know I can't see you :) I can't tell where is "here" 23:33 < jim> does your neighbor have an N router? 23:33 < jim> err access point 23:34 < Happyhobo> In my bedroom 45 foot away and through 2 walls, a corner of a closet across a stairwell and through the wall into my room. He gets 25 through that set of obstacles. 23:34 < Happyhobo> It's an N access point.\ 23:34 < jim> so... that's a good result? 23:35 < jim> remember too, that your results will be limited by his networking inside his house 23:35 < Happyhobo> 25 is amazing through all of those obstacles. I get 9 and lots of slows to nothing but remains connected mess. 23:35 < justsomeguy> Sounds good, at least for a home router. 23:35 < jim> and his connection speed 23:36 < Happyhobo> We're on the same network. He was almost sitting in my chair. 23:36 < jim> does he have a wired net? 23:36 < Happyhobo> Nope. 23:37 < Happyhobo> I just tested with another I got .14 Mbps down and .15 Mbps up, I have just enough of a connection to communicate on IRC. LOL 23:37 < jim> ok... so what kind of connection does he have? 23:37 < Happyhobo> regular N card from his N access point 23:38 < jim> yes, but how does he connect to the net? 23:38 < djph> you're still crying about getting exactly what you should be, given what you're paying for from teh ISP? 23:38 < djph> (or rather, *MORE* than what you should be, as I recall) 23:39 < Happyhobo> Through internet Explorer. He runs Windows 10. 23:39 < jim> does he have a dsl? a cable modem? eth over power? 23:40 < Happyhobo> A cable modem 23:40 < jim> ok, what does that cable modem connect to first? 23:41 < Happyhobo> No djph if I'm lucky I get that for a few minutes, it doesn't last for more than a few minutes, now I have just enough signal to IRC and it's not scrolling very fast. 23:41 < Happyhobo> The cable goes into it then it goes out from the cable into his TV. 23:41 < jim> yeah not much goin on here right now :) 23:42 < jim> the cable goes into... it? what's the it? 23:42 < Happyhobo> I would love to get this resolved then the HDMI and be able to watch videos and stuff on the TV 23:42 < Happyhobo> The router 23:43 < jim> so the cable modem goes directly into the router, and you get your connection from that router 23:44 < Happyhobo> the router and modem are one unit 23:44 < jim> there's not a hub or switch inbetween the router and the cable modem? 23:44 < djph> Happyhobo: and you're still on wifi, through two walls, right. Probably on 2.4 GHz on a stupid channel too, yes? 23:45 < jim> ok, so the cable modem/router has wireless, which is how you get your connection? 23:45 < Happyhobo> cable comes out of the wall right into this box that is about 9 inches tall, 3 inches wide with a 5 inch wide base, that box has two or three ethernet connections on it and puts out the signal from there. I have no idea if I'm on 2.4 but this card is supposed to be dual band. 23:45 < Happyhobo> Yes. 23:46 < djph> Happyhobo: through 2 walls, I'm telling you it's probably 2.4 GHz. 5 is usually lucky to get through 1 wall 23:46 < jim> how close are your buildings together? 23:46 < jim> close enough to get an eth cable across? 23:46 < Sonolin> yea +1 for eth 23:47 < Sonolin> cat5 ftw 23:47 < zaccanasta> maybe OT: readelf: Gap in build notes detected from 0x... to 0x... 23:47 < jim> well my cat5 had kitten6s 23:47 < zaccanasta> google is not my friend with this one 23:47 < Happyhobo> Heh there is a brick firewall between the townhouse apartments. 23:47 < Sonolin> haha nice one jim 23:47 < Sonolin> if real, congrats ;) 23:48 < jim> well that was long ago :) 23:48 < Happyhobo> I would have to wrap around the outside of the building about 75 foot 23:48 < jim> doesn't sound practical 23:49 < Happyhobo> It won't open a webpage at the moment. LOL 23:50 < Happyhobo> It's now saying Latency test error 23:51 < jim> I wonder if something could be done about the antenna in his house 23:51 < jim> I'm pretty sure this is an antenna issue 23:51 < Happyhobo> could gray be the problem? 23:52 < Happyhobo> I thought gray was MIMO and now cards have MIMO built in on 2x2 cards. 23:55 < jim> on the card, you have two antenna connectors (1 and 2)? 23:55 < jim> actually... unfortunately I've run out of time 23:56 < Happyhobo> yes 23:56 < djph> Happyhobo: they've always been ... actually, nevermind, not going down this rabbit hole again 23:57 < jim> but here would be my suggestion,,, you have two antenna connectors and three antenna cables... starting with antenna connector 1, try each of the three antenna cables on connector 1, then do the same on 2... see what's the best 23:57 < Happyhobo> that might work 23:58 < jim> (being careful,) make sure the antenna connectors are on the card pretty good 23:58 < jim> that's important :) 23:58 < jim> ok, gotta go... good luck --- Log closed Thu May 31 00:00:38 2018