IRC Logs for #crux Monday, 2014-04-07

*** rta has quit IRC00:21
*** BitPuffin has joined #crux00:22
*** jdolan has quit IRC00:32
*** rta has joined #crux00:35
*** jdolan has joined #crux00:55
*** rta has quit IRC01:17
*** rta has joined #crux01:29
*** tilman has quit IRC02:04
*** tilman has joined #crux02:05
*** saptech has joined #crux02:13
*** BitPuffin has quit IRC02:18
*** jdolan has quit IRC02:19
*** jdolan has joined #crux02:20
saptechgreetings all02:23
*** mavrick61 has quit IRC02:25
saptechi'm planning on giving Crux a try sometimes soon and would like to know if i3 tiling window manager in the repositories?02:26
*** mavrick61 has joined #crux02:26
jaegersaptech: yes, it's in a couple repos02:26
saptechgreat, thanks02:27
jaegersaptech: if you don't find something in core, opt, xorg, contrib, or compat-32, take a look at http://crux.nu/portdb/02:27
jaegerlots of contributed stuff there02:27
saptechok, i'll bookmark it02:27
*** rta has quit IRC03:12
*** rta has joined #crux03:24
cruxbot[core.git/3.0]: man-pages: updated to version 3.6404:29
*** pitillo has quit IRC05:10
*** Feigr has joined #crux05:30
*** Workster has quit IRC07:23
*** Workster has joined #crux07:25
*** BitPuffin has joined #crux07:31
cybinsaptech: i'm using i3 and i don't remember to do something special to get it.07:41
*** pitillo has joined #crux07:53
Romsterlo dudes07:59
Romsterand dudeetts but i think we have none?07:59
BitPuffinhmm, maybe we should have pkgfile signing08:11
BitPuffinhowever that would make things a lot more complicated08:11
cruxbot[contrib.git/3.0]: zaptel: Dropping08:12
cruxbot[contrib.git/3.0]: asterisk: Updated to 1.8.26.108:12
cruxbot[contrib.git/3.0]: asterisk-core-sounds-en-alaw: Updated to 1.4.2508:12
cruxbot[contrib.git/3.0]: asterisk-core-sounds-en-ulaw: Updated to 1.4.2508:12
cruxbot[contrib.git/3.0]: asterisk-extra-sounds-en-alaw: Updated to 1.4.1408:12
cruxbot[contrib.git/3.0]: asterisk-extra-sounds-en-ulaw: Updated to 1.4.1408:12
cruxbot[contrib.git/3.0]: dahdi: 2.7.0.1 -> 2.9.1.108:12
cruxbot[contrib.git/3.0]: dahdi-tools: 2.7.0.1 -> 2.9.108:12
cruxbot[contrib.git/3.0]: libzip: 0.11.1 -> 0.11.208:12
cruxbot[contrib.git/3.0]: p5-gd: 2.46 -> 2.5308:12
cruxbot[contrib.git/3.0]: smplayer: 0.8.6 -> 14.3.008:12
cruxbot[contrib.git/3.0]: smtube: 1.8 -> 2.108:12
cruxbot[contrib.git/3.0]: exiv2: 0.23 -> 0.2408:12
nwehas someone try terminator (terminal emulator) ?08:13
*** jdolan has quit IRC08:17
*** Pingax has joined #crux08:58
prologicBitPuffin, what would that gain us?09:28
RomsterBitPuffin, the git trees are using keys already and you can see who is making the edits.09:30
Romsteronly a select few have access.09:31
diversethank goodness for git09:31
Romstermostly ssh actually09:31
Romsterfor auth09:32
prologicall ssh afaik09:32
diverseand that too, ssh rocks09:32
prologicgit:// is anon read-only09:32
prologicso pgp signing would buy us nothing ihmo09:32
BitPuffinRomster: Well I meant for ports09:33
BitPuffinprologic, Romster: it would mean that if I was a port author I would have a private key on my work computer with which I could sign a pkgfile after I am done with it and when someone adds my port they would add my public key. If someone would then take over my httpup thing on my server they couldn't tamper with the pkgfile without ruining the signature09:35
BitPuffinIt's basically saying "I trust this guy with writing pkg files, but I don't trust that he won't get his server compromised"09:51
Romsterhmm we dont have protection for that currently09:52
Romsterthat just adds far more overhead to our stuff though09:55
Romsterand we genearlly know security well09:55
Romsterbut the risk is there09:55
BitPuffinyeah exactly09:57
BitPuffinI mean in theory you *should* check the package files09:57
BitPuffinbut really who does :P09:57
BitPuffinand in theory you should read the source code of the software you are running as well09:57
BitPuffinRomster: the thing you'd have to solve very well though is still having the ability to locally customize the pkgfile after it is fetched09:59
BitPuffinso you'd do something like pkg-verify pkgfile10:00
BitPuffinand then like pkg-dontcare pkgfile10:00
BitPuffinlol10:00
Romsterlol10:00
BitPuffinor hmm, I guess the signature stuff could be separate from building10:00
BitPuffinso if you build something manually it won't care unless you run a verify command10:01
Romstercouldbe a part of ports command maybe10:01
BitPuffinyeah exactly10:01
diverseit probably won't be that much overhead either10:01
Romstertrusted gpg keys in the rsync/http files maybe10:01
BitPuffinis there any easy library for this kind of stuff? Like for having a keyring etc10:02
BitPuffinRomster: yeah I actually like that more10:02
BitPuffinjust files10:02
BitPuffinrather than a kind of database10:02
Romsteri havne't yet but looking at signature files when i get more advanced on my versionsort.com code.10:02
BitPuffinit's more like the crux way10:02
BitPuffinhmm actually it's pretty easy to sign with gpg I believe10:03
Romsteri mean the public key but then it could be in a keychain10:03
Romsteryeah simple gpg commands10:03
Romstermaking it robust and easy though will take a little thinking.10:03
BitPuffinso you could just have a simple wrapper thing that invokes a gpg command to verify the pkgfile10:04
BitPuffinRomster: to me it almost sounds like a couple of lines of bash only haha.10:05
*** diverse has quit IRC10:06
Romsteryeah it could be just that. but then each repo and Pkgfile needs to be verified. and what to do with those that don't pass.10:06
frinnstjaeger: have you ever had a linux vm change the time when you take/remove a snapshot?10:08
BitPuffinwe still won't be protected from the source remotes getting compromised I guess but neither is any linux from scratch and honestly no distro, if kernel.org gets compromised and say arch grabs the latest tarball and builds their kernel with it then everyone could get a malicious kernel haha10:09
BitPuffinRomster: I think as a transition thing it would probably just be a prompt to install anyway10:09
BitPuffinsince not everyone will be able to get the gpg stuff setup immediately10:10
BitPuffinit should also then encourage you to read the pkgfile to be sure10:10
BitPuffinor well10:10
BitPuffinI guess if there is a gpg key and it doesn't match then it should probably refuse10:10
Romsteri dunno this is a thing to add a feature request to flyspray10:10
BitPuffinbut if there is no gpg key, then prompt10:10
BitPuffinflyspray?10:11
Romsterand discuss on the mailing list with a possible patch10:11
Romsterour bug tracker10:11
BitPuffinah10:11
BitPuffinyeah10:11
BitPuffinit would probably at least just be a step before building in prt-get or something, doesn't seem like it would be too much10:12
Romsterwortha look at but i'm busy with other stuff atm10:12
BitPuffinyup same here10:13
BitPuffinbut it should be something to keep in mind for the future10:13
BitPuffincrux is not really a high priority target anyway :)10:13
BitPuffinbtw does crux use its own init system?10:14
Romsterit's styled after BSD but we are just using sysvinit10:16
BitPuffinah10:17
BitPuffinso it's kind of a wrapper then?10:17
Romsterthough i had in the past and someone else has made runit Pkgfiles10:17
Romsterthe rc files are just shell stuff.10:17
BitPuffinyeah10:17
BitPuffinwell, what's crux' view on systemd, one of the reasons I'm moving to crux is actually because I don't like systemd scripts that much, crux is much cleaner10:18
Romstersee core 3.1 eudev that's what we think of systemd10:19
Romsterwant to avoid it as much as possible.10:19
Romstergtk3 is another one i'm not so sure i wanna support but i fear i'll be forced too.10:20
BitPuffinlink to core 3.1 eudev? :)10:20
*** diverse has joined #crux10:20
BitPuffinyeah can't really keep stickling to gtk310:20
BitPuffin2*10:20
diversefucking lag, it was over 100010:21
Romsterhttp://crux.nu/gitweb/?p=ports/core.git;a=shortlog;h=3.110:21
Romsterover 9000?10:21
diverseyeah i wish /sarcasm10:21
Romster9000 nanoseconds10:22
diversethere you go10:22
Romsterwhat?10:22
Romsterid like tonot support gtk3 but really it'll be able to support the rendering stuff and less theming.10:23
BitPuffinRomster: ah I see10:23
BitPuffinyeah well good10:23
Romsterwhats the new xorg repalcment... geez i'm doing so much and now forgetting.10:24
BitPuffinseems like we won't left out on udev niceties at least then10:24
BitPuffinRomster: wayland?10:24
Romsterthat's it gtk3 has support for wayland10:24
BitPuffinyeah10:24
BitPuffinwayland is pretty important haha10:24
Romsterxorg's design has major latency issues on redrawing.10:25
BitPuffinwhy would you not wanna support gtk3 though?10:25
BitPuffinI mean we want to support as much as possible don't we?10:25
BitPuffinSince there will be applications and is applications running gtk310:25
Romsterevery gtk 3 release tens to break themeing.10:25
diverseand break API10:25
BitPuffinwell10:25
BitPuffinthat's a hitch10:25
BitPuffinThey'll get past it I'm sure10:25
BitPuffinbut for application compatability we need gtk310:26
BitPuffinand 4 whenever that comes10:26
BitPuffinapparently the gtk3 devs are unhappy with gtk3 and are betting on 410:26
BitPuffinlol10:26
Romsterhttps://igurublog.wordpress.com/2012/11/05/gnome-et-al-rotting-in-threes/10:26
diversewell a lot of apps are either going to migrate away gtk3 or stick with gtk210:27
RomsterTheme development is a tedious and difficult task, and for the GTK devs to be so careless in breaking their API at every turn disrespects the many hours people put into making themes for it. Yet as I read some of the GNOME developer comments below, I was given to believe that this breakage stems from a Microsoft-like climate of preventing users from customizing their systems, and deliberately breaking the work of others so that your ‘brand’ is th10:27
Romstere best. Anytime I hear the word ‘brand’ being used in Linux, I know something valuable is being poisoned. Just as a sample of what is to come10:27
diverseand an example of their current fucking with gtk3: https://igurublog.wordpress.com/2014/03/22/gtk-3-10-drops-menu-icons-and-mnemonics/10:28
BitPuffinhmm10:28
BitPuffinthat's a shame10:29
BitPuffinbecause I don't really like Qt either10:29
BitPuffinluckily I am just a game developer so I don't use those kind of toolkits that much10:30
BitPuffinalthough I do iOS dev at work10:30
BitPuffingah10:30
BitPuffinxcode10:30
BitPuffingtk is a lot better than xcode10:30
BitPuffin*app crashes* - xcode: It happened in main10:30
BitPuffinGEE THANKS10:31
Romsterthis stuff makes me mad10:31
BitPuffinRomster: xcode or toolkits10:32
Romstergtk systemd pulseaudio redhat in general.10:33
Romstergtk3*10:33
Romsteroh and firefox making there UI fugly.10:33
Romsterin the up coming 28 firefox it'll look like windows.10:33
Romsterwho needs menu items anyways?10:34
Romsterus power users that know how to navigate.10:34
Romsterno one cares about us its about making it simpler for the dumb ones.10:34
diversewe are already at firefox version 2810:35
Romsteror ws it 29...10:35
Romsterwas*10:35
Romsterthe up coming beta10:35
Romsterand these version numbers the hell.10:35
Romsterthey jut wanted to beat IE and chrome verion numbers10:36
BitPuffinlol10:37
BitPuffinthem version nambaz10:37
Romsteri'm getting rusty and out of touch with everything.10:38
BitPuffinit's actually just that they decided to work exponentially harder than before didn't you know10:38
Romsterlol10:38
Romsteri need to start focusing more on getting my code projects working than trying todo everything.10:38
diverseFirefox is doing what is called the "Rapid Release Cycle" http://en.wikipedia.org/wiki/History_of_Firefox_%28Rapid_release_development_cycle%29#Rapid_release10:39
Romsteryeah i've read it10:39
diverseso they push out more features as fast as possible10:40
Romsterthey will even include there own flash to html5 player soon too10:40
frinnstits 2910:44
frinnstand its completely unuseable10:44
frinnsti'll probably start using the lts version instead, and perhaps introduce it to opt or whatever10:44
diverseoh wow10:44
diverseso in 29 they started using gtk3?10:45
BitPuffindayum10:45
frinnstnot sure if its gtk3 or not, but it suuuuucks10:45
diverseLTS ftw10:45
BitPuffinseemed pretty good to me10:45
BitPuffinapparently a lot faster10:45
diverseMy nightmares are here to come...10:46
Romsterpale moon?10:48
Romsterhttp://sourceforge.net/projects/pm4linux/10:49
BitPuffinwoot10:50
BitPuffindoes it support the bookmark sync?10:50
Romsterdunno10:50
Romsterworth a try i haven't had time to try it yet10:51
BitPuffinfor windows operating systems10:51
BitPuffinlol10:51
BitPuffinhttp://www.palemoon.org/10:52
diverseHmm, what do we have here? This looks interesting.10:53
diverserofl10:53
diverse"Pale Moon is, and will always be, completely FREE to download and use! (Freeware)"10:56
diverseIt's not freeware, it's effing opensource, lol10:56
Romsteryeah i know the buggers.10:58
BitPuffinyeah saw that too lol10:59
diversefrinnst: so lts all the way?11:46
jaegerfrinnst: not sure, honestly, I've not checked the time after snapshots11:50
*** diverse has quit IRC12:28
*** BitPuffin has quit IRC13:05
*** BitPuffin has joined #crux13:15
frinnstjaeger: turns out the host had its time a bit offset. and even if you dont sync your guest with the host, it overrides this option and forces syncs when taking snapshots, vmotion etc13:15
frinnsthttp://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=118913:16
jaegerthat's annoying13:16
jaegerI do remember the forced sync, had a problem a couple years back where one host refused to sync properly with NTP so the VMs booted up way off13:17
frinnstyeah, i thought i was going crazy and some rouge cron job fucked it up :)13:18
frinnstturns out it was our backups :)13:18
jaegerwell, at least you found it :) would have been frustrating13:18
prologicBitPuffin, signing doesn't really achieve anything more than what git+ssh does13:19
frinnstmaybe it would be useful if we ran a lot of branches and more distributed13:20
frinnstas it is now, its all pushed to a central repo13:21
prologicfor example you can be reasonably assured that my ports aren't tampered with by comparing with the repository I maintain of them (https://bitbucket.org/prologic/ports)13:21
prologiceven so, those that have access to contrib say do so via ssk keys13:21
prologicso you in effect already have both auth and verification there13:21
prologicby way of ssh public key auth and git controled verification that user bob commited this change13:22
prologicsigning is great - don't get me wrong, but I don't think it buys you what you think it does13:22
*** jdolan has joined #crux14:06
*** jdolan has quit IRC14:25
*** jdolan has joined #crux14:30
BitPuffinprologic: well the point is thta you don't have to use git/hg+ssh to have a ports repo14:31
BitPuffinprologic: and with say a bitbucket repository somebody only needs your account to add their own public key to your bb account14:31
BitPuffinwheras with signing they would have to either break in to your pc and steal your private key, AND your server and tamper with the port. Or they would have to only break in to the server, convince everyone that you have a new private key and tell them to add a new public key14:32
BitPuffinso it does go further than that14:32
*** jdolan has quit IRC14:34
*** jdolan_ has joined #crux14:36
*** diverse has joined #crux15:01
diversefrinnst: they are staying with gtk2 for firefox 29. Other than changing their UI to look more like Chrome, what specifically sucks about it?15:02
*** AlexKraken has joined #crux15:02
*** pitillo has quit IRC15:11
*** rta has quit IRC15:12
frinnsttry it and judge yourself15:14
frinnsti've had the same firefox config for what,  10 years?15:14
frinnstim kinda used to it by now15:14
diversecan't you change the user interface back to what you are used to? From what I read, they still give you customizeablitiy, but I am hoping they are not pulling GNOME on us though.15:15
diverse*pulling a GNOME15:18
*** Pingax has quit IRC15:23
*** rta has joined #crux15:25
*** darfo has quit IRC15:32
*** pitillo has joined #crux15:45
*** pitillo has quit IRC16:02
*** darfo has joined #crux16:06
*** arjovr has quit IRC16:11
*** arjovr has joined #crux16:23
*** AlexKraken has quit IRC16:29
*** AlexKraken has joined #crux16:29
JackLFrostYay, my watchman wors on FreeBSD without any changes16:40
JackLFrostworks*16:40
JackLFrostOnly the scripts need to be different to accomodate the /{bin,sbin}, /usr/{bin,sbin} and /usr/local/{bin,sbin} separation16:41
JackLFrostBut that's easily fixed by mergin everything into /usr, lol16:41
cruxbot[core.git/3.0]: kmod: updated to version 1716:53
*** BitPuffin has quit IRC17:20
*** pitillo has joined #crux18:07
*** jdolan_ has quit IRC18:12
*** pitillo has quit IRC18:17
*** jdolan has joined #crux18:21
*** jdolan has quit IRC18:23
*** BitPuffin has joined #crux18:36
*** jdolan has joined #crux18:36
*** jdolan has quit IRC18:39
*** pitillo has joined #crux18:41
*** jdolan has joined #crux18:45
*** jdolan has quit IRC18:47
*** BitPuffin has quit IRC18:55
*** nlightnfotis has joined #crux18:59
*** pitillo has quit IRC19:01
*** jdolan has joined #crux19:03
*** pitillo has joined #crux19:05
*** BitPuffin has joined #crux19:46
*** BitPuffin has quit IRC19:51
*** pitillo has quit IRC20:00
*** pitillo has joined #crux20:00
*** jdolan has quit IRC20:00
*** diverse has quit IRC20:07
*** jdolan has joined #crux20:26
frinnstwow, usb i/o slows my system to a crawl with 3.14.020:39
jaeger:(20:39
*** jdolan has quit IRC20:39
frinnstanybody else have problems?20:39
jaegerI haven't tried 3.14 yet20:39
frinnsteven moving the mouse and typing is laggy20:39
*** jdolan has joined #crux20:39
jaegeryuck20:39
jaegerIs it usb 2, 3, or both?20:40
frinnstusb2 is the only thing i've tested20:40
frinnstjust regular usb sticks20:40
frinnstI boot one of my microservers off an usb-stick and it has just locked up a few times recently20:41
frinnstand it also runs 3.1420:41
frinnsti thought the usb-stick was a bit flakey, but maybe its something with 3.14?20:41
*** jdolan has quit IRC20:41
*** jdolan has joined #crux20:42
jaegerdoes it work fine with older kernels?20:44
frinnstyeah the microserver never had these issues20:52
frinnstnot sure about my desktop20:52
jaegerUSB support has been a pain in the ass for the ISO since 3.7.x20:54
frinnstyeah. i've run kernels as recent as 3.13 without issues20:54
jaegersince 3.7.x I've had to build USB controllers builtin instead of modules or booting from a USB drive fails :/20:55
jaegerShouldn't make any difference but it does20:55
frinnstload average: 4.40, 7.45, 8.5720:56
frinnstnot *that* bad20:56
jaegerheh20:56
frinnsti ran f2fs as the root fs on my microserver. switched it over to ext4 since fsck did detect some errors (but were unable to fix them), lets see if that solves my problems20:57
frinnstbut i doubt it :)20:57
*** BitPuffin has joined #crux21:25
cruxbot[opt.git/3.0]: zsh: fix compilation against current glibc21:30
*** jdolan has quit IRC21:32
*** AlexKraken has quit IRC21:50
*** jdolan has joined #crux22:46
*** nlightnfotis has quit IRC23:43
*** Sleepy_Coder has quit IRC23:46
*** Sleepy_Coder has joined #crux23:47
*** pips has joined #crux23:49
*** BitPuffin has quit IRC23:55

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!