IRC Logs for #crux Sunday, 2015-02-22

cruxbot[contrib.git/3.1]: xscreensaver: removed dependency netpbm, it's not being used anymore and is not linked to netpbm so a rebuild is not necessary00:11
*** arduo has quit IRC00:43
*** tilman has quit IRC01:03
*** tilman has joined #crux01:05
cruxbot[opt.git/3.1]: wine: 1.7.36 -> 1.7.3701:26
*** orbea has quit IRC01:29
*** orbea has joined #crux01:29
*** BitPuffin has joined #crux01:48
*** JSchilli1 has joined #crux01:50
JSchilli1hey, anyone know what went wrong here? i didn't change anything as far as i know.
jaegerinstall start-stop-daemon, it's been recently added to core01:51
JSchilli1jaeger: ahh, thanks a lto01:51
Romsteryou may want to join the crux-lists mailing list01:55
Romsterbtw JSchilli1 you can do prt-get fsearch start-stop-daemon to see what package is missing.01:56
Romsterprovided you did a ports -u already01:56
*** cezar has joined #crux02:01
*** cezar has quit IRC02:03
*** pejman has quit IRC02:16
*** namra has joined #crux02:21
*** pejman has joined #crux02:24
*** BitPuffin has quit IRC02:35
cruxbot[compat-32.git/3.1]: util-linux-32: 2.25.2 -> 2.2602:35
cruxbot[compat-32.git/3.1]: mesa3d-32: 10.4.4 -> 10.4.502:35
cruxbot[compat-32.git/3.1]: gtk-32: 2.24.24 -> 2.24.2602:35
JSchilli1Romster: thanks, i had issues subscribing to the list before, it kept saying i had an invalid code or something, i just got re-subscribed this week02:45
JSchilli1that fsearch command is helpful, though02:45
Romsterah me too i think that might be fixed now02:45
Romsteryeah i use fsearch heavily when i'm looking for a missing file.02:45
*** nogagplz_ has quit IRC03:18
*** nogagplz has joined #crux03:19
*** mavrick61 has quit IRC03:33
*** mavrick61 has joined #crux03:34
*** namra has quit IRC03:36
*** greduan has quit IRC04:21
*** diverse has quit IRC04:27
*** pejman has quit IRC04:56
*** pejman has joined #crux05:05
Romsteri'm already looking at that. i tried to run chrome dockerfile it failed.06:42
Romsterdocker run -it --net host --cpuset 0 --memory 512mb -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY -v $HOME/Downloads:/root/Downloads -v $HOME/.config/google-chrome/:/data -v /dev/snd:/dev/snd --privileged --name chrome jess/chrome06:42
Romster[1:1:0222/] Running without the SUID sandbox! See for more information on developing with the sandbox on.06:42
RomsterNo protocol specified06:42
Romster[1:1:0222/] Gtk: cannot open display: :006:42
Romsterecho $DISPLAY06:43
prologicI'll have to try this as well06:49
Romsterprologic, you busy i'm toying with a CI setup. thats init's early stages. and packages.06:49
Romsterin it's*06:49
Romsterbeing that our packages can vary between systems as we are not strict on optional dependencies. i thought it was an idea to build each port in it's own container. hash the directory for the package.06:50
Romstermake a deptree of the port and assign a hash to it.06:51
Romstermake a new pkg-get to download the required hashed directory/package06:51
Romsteralso collect the package list off the users system and output of finddeps, and if no suitable match is found. trigger a new build with them packages installed in a container. and add that to the hashed package directory for pkg-get06:52
Romsterit'll over time end up with all the package variations of every optional dependency it can link too.06:53
prologicsounds complex06:53
Romsterwhy not? i'm part way there now06:53
Romstersee gist ^06:54
prologicyeah looking at it now06:54
Romsterit's a really rough idea.06:54
Romsterports with no hash are in core.06:55
Romsterso they can be skipped.06:55
Romsteri haven't implemented a way to remove old package versions yet. prt-orphan might do my needs.06:55
Romsternot looked into that yet.06:55
Romsteri could make a master html table of optional dependencies found for a package.06:57
Romsteri could code this better in python, but just hacking it up atm.06:59
Romsteralso once i have a master list of deps a port can use i can then test rebuilding all those on a port version update. and report back if an update broke anything.07:00
Romsterthe only thing it wont test is dynamic loaded, calling executives and compile time deps.07:01
*** deus_ex has joined #crux07:02
Romsteri just tried to run spotify07:03
RomsterStatus: Downloaded newer image for jess/spotify:latest07:03
RomsterNo protocol specified07:03
Romster(spotify:1): Gtk-WARNING **: cannot open display: unix:007:03
Romsterok then so i set DISPLAY=$DISPLAY and...07:03
RomsterNo protocol specified07:03
Romster(spotify:1): Gtk-WARNING **: cannot open display: :007:03
Romsterclearly it can't talk to my Xorg server.07:04
Romsterthe chrome one is probably broken in some way other than the failure to connect to Xorg07:04
Romsterdocker run -it --name lynx jess/lynx07:06
Romsteri expect this to work.07:06
Romsteryeah that works fine.07:08
prologicactually if those examples aren't working07:08
prologicI think they might be a bit wrng07:09
prologicI remember successfully running GUi apps as docker containers on my dekstop07:09
prologicbut it was a bit tricky07:09
prologic    $ docker run -i -t -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -v $HOME/.Xauthority:/$HOME/.Xauthority --net=host crux/xchat07:10
prologicsomething like this I think worked for me07:10
prologicI'll try it07:10
Romstermight be because crux is different07:12
Romsteri read somewhere to run X app in container and VNC connect to it.07:13
prologicno no07:13
prologicdont' do that07:13
prologicthis *does* work07:13
prologiclet me rebuild crux/gtk and crux/xchat images here07:13
prologicand test it out for myself07:13
prologicthen if that works, any other GUI app will will work the same07:13
Romsteri was thinking that be ok for testing somethign but... why can't we jsut use the hosts xorg-server to connect to it.07:13
prologicwe can07:14
prologicyou bind mount the socket07:14
Romsteri was using ssh in a container now i use -v for bind mounting.07:14
prologicdocker run -i -t -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -v $HOME/.Xauthority:/$HOME/.Xauthority --net=host crux/xchat07:14
Romsterthe weird part is -v when new files are added at that bind mount. they are left behind after docker rm container_name07:14
Romsteri'd of expected the bind mount to be unmounted after docker stop.07:15
Romsterwhich would also not show any files on host after.07:15
RomsterError: image crux/xchat:latest not found07:17
Romsterand prologic/xchat don't exist either07:18
*** arduo has joined #crux07:35
*** lnds has joined #crux07:42
prologicit exists here locally07:43
prologicI experimented with it all, got it woring07:43
prologichad some font issues07:43
prologicand left it :)07:43
prologicso it's unpublished works atm07:44
Romsterfont issues would be missing fonts.07:49
*** hhhhhhhh has joined #crux07:50
prologicI think last time I tried all this08:01
prologicI concluded that I needed to build a crux/xorg image08:01
prologicwith all of xorg (minus video driers)08:01
prologicand plus fonts08:01
prologicStep 5 : RUN localedef -i ${LOCALE} -f ISO-8859-1 ${LOCALE}08:02
prologic ---> Running in 3585f3f7f78b08:02
Romsterbut yeah a xorg image that has fonts, then you can make a Docker container use that for xchat etc.08:09
nwegood morning guys08:41
*** hhhhhhhh has quit IRC08:44
*** asie has joined #crux09:19
asieafter a sysup, my start-stop-daemon disappeared and sshd's rc script depends on it09:20
asiewhat went wrong?09:20
prologicwtf is start-stop-daemon?09:26
prologicSun Feb 22 19:29:3609:26
prologic$ prt-get fsearch start-stop-daemon | wc -l09:26
prologicstart-stop-daemon is not a CRUX thing AFAIK09:27
prologicwe generally don't wrap our rc stuff in black magic09:27
prologicRomster, ...09:28
prologic$ dki -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -v $HOME/.Xauthority:/$HOME/.Xauthority --net=host crux/xchat09:28
prologicNo protocol specified09:28
prologic(xchat:1): Gtk-WARNING **: cannot open display: :0.009:28
prologicI get this too09:28
prologicsomething has changed in Docker 1.5 and/or latest Xorg09:28
asieprologic: check core/sshd09:28
asieas in the sshd file in openssh09:28
prologicyeah ok09:29
prologicbut there is no start-stop-daemon09:29
prologicso it cannot have disappared because crux oes not have one :)09:29
asiefor me it is on line 609:29
prologic$ grin start-stop-daemon /etc/rc.d/sshd | wc -l09:30
prologicyou must be running something else?09:30
asiei am definitely running crux, what's going on09:30
prologicyou tell us :)09:30
prologic HexChat: 2.10.1 ** OS: Linux 3.12.17 x86_64 ** Distro: Unknown Distro ** CPU: 8 x Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz (GenuineIntel) @ 3.50GHz ** RAM: Physical: 15.5GiB, 52.8% free ** Disk: Total: 3.5TiB, 91.0% free ** VGA: 10de:11c0 ** Sound: HDA-Intel - HDA Intel PCH1: HDA-Intel - HDA NVidia ** Ethernet: 14e4:16b1 ** Uptime: 5d 21h 5m 28s **09:30
prologicCRUX version 3.109:30
prologicLinux daisy 3.12.17 #18 SMP PREEMPT Wed Feb 11 17:54:24 AEST 2015 x86_64 Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz GenuineIntel GNU/Linux09:30
prologicI *am* definately on CRUX :)09:30
asieso am i, i mean, i just ran a prt-get sysup yesterday09:30
asie and this is my /etc/rc.d/sshd09:30
asieright there just below the first comment09:31
prologicwth this is weird09:31
prologicgimme a sec09:31
asieand the issue only started happening after i ran sysups yesterday09:31
prologicI haven't updated in a while09:31
asieprior to that sshd worked fine09:31
prologicbut AFAIK we *do not* have a start-stop-daemon09:31
asiei'm an user09:31
prologicunless this is new in some port/package/software09:31
asiei don't care if you do or don't, your script does09:32
asiein core/openssh09:32
Romsterprologic, core/start-stop-daemon is a new depe3ndency09:32
prologicRomster, are you serious?09:32
asieRomster: why doesn't openssh depend on it, then?09:32
prologiccore ports don't normalyl list their deps09:32
Romsteri am dead serious.09:32
prologicit's assumed you *have all of core installed*09:32
asieyes but when someone's updating09:32
asiewith sysup09:32
prologicRomster, why?09:32
asieit doesn't install it09:32
Romsterports -u ; prt-get depinst start-stop-daemon09:32
prologicthis is nuts09:32
asieand openssh is kind of a bad thing to break in a sysup, you know09:32
Romsterit was sent to the crux list mailing list.09:33
prologicI don't get it09:33
prologicwhy are we using this?09:33
prologicwho introduced it?09:33
prologicI never read anything about it09:34
prologicthis is all new to me :/09:34
prologicjue, ping?09:34
prologicalso because why?09:34
Romsterwe had some issues with some dhcp pid file other crap wnt stop cleanly debians small program for services makes writing RC scripts cleaner and simpler.09:34
prologicdoes it do something that /sbin/sshd doesn't?09:34
prologici see09:35
prologicI did read about some dhcp issues that frinnst  was erporting09:35
prologicisn't this a problem with upstream dhcpd through?09:35
Romsteralso yes09:35
Romsterupstream bug reports have been made afaik09:35
prologicI'm not happy about this :) haha09:36
prologicthis adds a layer of complexity to rc scripts and a new dependency09:36
prologicwe may as well switch init systems09:36
prologicso no discussion on the ml09:37
prologicjust a core port notifyication09:37
prologichmm :)09:37
Romsterthere was discussion in #crux-devel09:37
prologicI'm regreetting not being a part of htat now :) haha09:37
Romsterfrankly the sshd rc script should not have been changed this soon... and i was abit mad it was added now and not when we are releasing crux 3.209:38
Romsterbut i'm just a packager what do i know.09:38
Romsteri'm not core team.09:39
prologicI'm a bit disappointed tbh09:39
prologicthis seems like an architectural change of core really09:39
Romsterit's a tiny ass package so i don't see the big deal now except update openssh and not install this new dependency and poof no remote access09:39
prologicAFAIK start-stop-daemon really is just a small C binary that wraps around what you'd do with things like kill, pidof, etc09:40
Romsterwhat if the box is several hindered kilometers away... and no on site admin?09:40
Romsteryeah to make rc scripts easier to write.09:41
prologicyeah this is what I mean09:41
prologicchanging core like this is bad09:41
prologicbecause we don't list core deps09:41
prologicso you end up breaking things09:41
prologicnot to mention a sysup or update will not install new deps09:41
prologicit's a manual step09:41
prologicI don't really care to be dragging in Debian tools into the CRUX ecosystem either really (but that's just a personal thing)09:42
Romsterno we do list core deps we link too except glibc and gcc09:42
Romsterkey word /link too/09:43
Romsterif it does not show up on finddeps foo then it's not listed.09:43
Romsterand because start-stop-daemon does not link to anything well.. it's a essential core port but nothing to inject it with after a ports -u09:44
Romsterother than the admin seeing it in the ports -u and the mailing list.09:44
prologic# prt-get depends openssh09:44
prologic-- dependencies ([i] = installed)09:44
prologic[i] openssl09:44
prologic[i] zlib09:44
prologic[i] openssh09:44
prologichmm I guess you're right09:44
prologicI keep forgetting we changed that policy09:44
prologichowever openssh does not list start-stop-daemon09:44
prologicIHMO we should undo this change09:45
Romsterand because we are experienced you'd think they would be watching the mailing list.09:45
prologicthere has to be better ways to solve dhcpd issues09:45
Romsteropenssh does not link to start-stop-daemon09:45
Romsterand even if it did, prt-get wont install it for you.09:45
prologicare you defending the introduction of start-stop-0daemon or what? :)09:46
prologicdoesn't matter if sshd doesn't link against it09:46
prologic/etc/rc.d/sshd now requires it09:46
Romsterno i am on my own side...09:46
prologicso it should list it as a dependency09:46
Romsterhow crux works and it's policy09:46
prologicshoe mw09:46
prologicthat's silly09:46
Romsterwhat should have happened is this new dep not be introduced until crux 3.2 was released.09:47
prologicindeed - it's a major change09:47
Romsterbuild dependencies provided by core are not listed in the dependency header09:48
Romsterrun-time dependencies from core which aren't dynamically linked in are not to be listed, either09:48
Romsterso start-stop-daemon can't be listed.09:48
Romsterits in core it's essential must be on all systems.09:48
prologicrun-time dependencies from core which aren't dynamically linked in are not to be listed, either09:48
prologicI guess you're referring to this09:49
Romsterso says in the mailing list.09:49
prologicwhich in this case is a stupid policy09:49
prologicbecause it totally breaks sshd09:49
prologicand any remote access09:49
Romsterprt-get depinst sshd... it's in core it's installed already it wont pull in start-stop-daemon09:49
prologicunless you were diligent, read the #crux-devel discussion, saw the email notification and handled it09:49
prologicwhich has already caught one poor user09:49
prologicthis is really bad form09:49
Romsterbecause we talked about that feature in prt-get and it's not gone anywhere.09:49
Romsteri do agree it's bad form to introduce it now.09:50
Romsteri'd say revert sshd rc script until crux 3.2 is released.09:50
prologicI think it's also bad form to bring in a new dependency to solve some other unrelated issue09:50
prologicbut this is me as a software engineer talking here09:50
Romstereh take that up with jue09:50
prologicI always argue against introduction of new dependencies in any software or software ecosystem09:51
Romsteri had no say in this matter09:51
prologicI will09:51
prologicI'm quite upset over this09:51
Romsterafaik it was frinnst and jue09:51
prologicjust lucky it didn't bite me int he ass09:51
prologicanyway famiyl time09:51
Romsteri'm not arguing i'm just pointing out policy and i don't agree bringing it in now.09:51
Romstertake care.09:52
prologicpolicy in this case is broken09:53
prologicand this change should not hafve been introduced09:53
Romsterthen that needs to be discussed09:53
prologicand this should have been discussed on the crux ml at large09:53
Romsterbut think about it it wont pull it in even if it is listed now.09:53
prologicthis is all really poor form09:53
prologicno I know that09:54
prologicwe have a missing feature of prt-get09:54
prologicwe've "put up" with that for a while09:54
Romsteri must note i even voted against that but now i'm reconsidering.09:57
Romsteri don't want a smart package manager that decides for me.09:57
Romsteroh you need this or lets remove that for you.09:57
Romsterbad form for a experienced distro.09:58
Romsterwe use crux because it give freedom you want remove glibc sure does it.09:58
Romsterit'll break the system but the sysadmin knows best09:58
Romsteri don't mind if its a notice or something. or an alert.09:59
Romsterssh will not run if this port is also installed. do you want me to install start-stop-daemon? [Y/n]09:59
Romstershould of been added for crux3.2 not now to catch everyone out.10:02
prologicI think a warning is sufficnent10:05
prologicas long as we list the dep10:05
prologicand an optional prt-get flag could be useful too10:05
prologicbut I suspect the reason we've not added to or improved some edge-cases of prt-get is because of devleopment time and effort required10:06
prologicwhich is fair enough10:06
nweis it any todo for crux3.2?10:18
Romster not made yet10:21
nweoh I see10:24
*** arduo has quit IRC10:33
*** zXter has joined #crux10:40
*** lnds has quit IRC10:41
zXterHi all, new cruxer here...10:49
Romsterhi zXter10:49
zXterirclogs @ shortcircuit are down?10:50
nwehello zXter10:50
Romsterno they load for me but not for someone i know in the usa10:50
Romsterso some network outage10:50
zXterAhh OK, so they will be back, cos I usually read them first thing in the morning :)10:52
Romsterlook  for a proxy i'm sure you'll find it then10:53
Romstersince the site itself is up10:54
*** namra has joined #crux10:59
Romstermaybe i need to paste this every hour for at least irc people.
*** leo-unglaub has joined #crux11:11
*** leo-unglaub1 has joined #crux11:56
*** leo-unglaub has quit IRC11:56
*** asie has quit IRC12:05
*** cezar has joined #crux12:56
*** cezar has quit IRC12:58
*** hhhhhhhh has joined #crux13:21
*** namra has quit IRC13:49
*** EmoSpice_Home has joined #crux13:58
EmoSpice_HomeGood morning.14:03
EmoSpice_HomeSo I'm trying to build a tool to scrape the portsdb (so I can write a tool to search it for me). Am I missing something, or is there no API for the portdb?14:04
*** greduan has joined #crux14:14
*** greduan has quit IRC14:14
*** greduan1 has joined #crux14:14
greduangood morning :)14:15
EmoSpice_HomeGood morning :)14:15
nrxtxEmoSpice_Home: there is no offical api but i started to parse and provide those information on a third-party site14:17
EmoSpice_Homenrxtx: Oh really? I'm currently parsing the html directly14:18
EmoSpice_Home(and I'm glad that went through - I was afraid I hadn't yet authed and was muted or something :P)14:19
EmoSpice_HomeI'll have a look over that. Thanks!14:19
nrxtxgood morning frinnts14:28
*** heroux has quit IRC14:33
*** heroux has joined #crux14:35
*** Feksclaus has joined #crux14:35
*** toriso has joined #crux14:40
jueprologic: you are not subscribed to the CRUX mailing list, shame on you ;)14:47
*** leo-unglaub has joined #crux14:50
*** leo-unglaub1 has quit IRC14:51
*** doomicide has joined #crux15:05
*** tvaalen has quit IRC15:23
*** tvaalen has joined #crux15:24
*** lowe has joined #crux15:24
*** irclogger_ has joined #crux15:25
*** Feksclaus has quit IRC15:27
*** BitPuffin has joined #crux15:38
*** BitPuffin has quit IRC15:38
*** BitPuffin has joined #crux15:40
*** BitPuffin has quit IRC15:52
*** hhhhhhhh_ has joined #crux15:54
*** dougl has joined #crux15:57
*** doomicide has quit IRC16:03
rmullEmoSpice_Home: Someone said there was actually an "API"16:12
*** lnds has joined #crux16:12
rmullSee port "portdbc"16:12
EmoSpice_Homeinteresting. I hadn't seen that one.16:13
EmoSpice_HomeI'm pushing my own script as we speak though :P16:13
EmoSpice_Home, if anyone is interested.16:16
*** mechaniputer has joined #crux16:18
*** kori has quit IRC16:25
*** kori has joined #crux16:25
*** kori has quit IRC16:25
*** kori has joined #crux16:25
*** kori has quit IRC16:26
*** kori has joined #crux16:27
*** kori has left #crux ()16:27
*** kori has joined #crux16:30
*** kori has quit IRC16:35
*** kori has joined #crux16:41
*** Pingax has quit IRC16:53
*** EmoSpice_Home has quit IRC17:01
*** EmoSpice_Home has joined #crux17:21
*** dougl has quit IRC17:29
EmoSpice_HomeWell, I guess a rainy, crappy Sunday is as good a day as any to work on this portdb tool. Back to work.17:30
greduanrainy doesn't need to mean crappy. :), depends on if you're in a city or more outside the city though, I guess17:37
EmoSpice_HomeIn the city(ish) and it is definitely pretty crappy :P17:41
EmoSpice_Homebut generally - you're right17:41
greduanif there are plants around, rain is so nice. although if it's all concrete and gray rain sucks ass17:45
greduanoh yeah, how explicit can I be in #crux?17:45
EmoSpice_HomeNo one's called me out yet :P17:45
EmoSpice_Homeand yeah - Atlanta is basically a city in a forest, which is nice. It's been rainy and really damn cold for the last week or so, so I'm getting tired of it17:47
greduanI see17:48
greduanall the other kids with the pumped kicks you better run better run17:49
EmoSpice_HomeThat song is entirely too catchy to be as sad as it is17:50
greduanI never pay attention to meaning in lyrics17:51
greduani will however notice sexual meanings and innuendos. I will pick up on that but if it's sad stuf chances are I won't pick it up17:52
EmoSpice_HomeFrom wikipedia: The lyrics to "Pumped Up Kicks" are written from the perspective of a troubled and delusional youth with homicidal thoughts.[4] The lines in the chorus warn potential victims to "outrun my gun" and that they "better run, better run, faster than my bullet."17:53
greduanwell then17:54
greduanwould not have guessed that17:54
*** EmoSpice_Home has quit IRC18:06
*** kori has quit IRC18:29
*** kori has joined #crux18:31
*** kori has quit IRC18:31
*** kori has joined #crux18:31
JSchilli1do you guys think it would be a good idea to put changes like the start-stop-daemon addition on the homepage under news?19:00
JSchilli1i know it's kind of expected (or, at least encouraged) to watch the mailing list but it's not realistic to expect everyone to do that19:00
JSchilli1arch linux seems to do something like that, with system-breaking changes being noted on the homepage, i liked that19:01
*** hhhhhhhh has quit IRC19:08
*** hhhhhhhh- has joined #crux19:08
*** hhhhhhhh- has quit IRC19:15
*** hhhhhhhh has joined #crux19:17
*** leo-unglaub has quit IRC19:17
*** Pingax has joined #crux19:19
*** hhhhhhhh_ has quit IRC19:21
*** hhhhhhhh has quit IRC19:24
*** BitPuffin has joined #crux19:26
*** hhhhhhhh has joined #crux19:28
*** hhhhhhhh has quit IRC19:35
*** hhhhhhhh has joined #crux19:38
*** mhe has joined #crux19:39
prologicjue, are you serious? I have *always* been subscribed!19:44
*** jdolan_ has quit IRC19:49
*** jdolan has joined #crux19:50
greduanI would like if the homepage showed news about changes to core, not just new releases. Also makes the distro look more active19:52
greduanwhich is very important for attracting new users19:52
*** mikel has quit IRC19:53
*** mikel has joined #crux19:56
frinnsti like the "repelling new users" feature :)20:00
frinnststart-stop-daemon is the best thing since sliced bread20:02
frinnstmost of our rc.d-scripts sucked butt20:02
frinnststuff used killall -q <path/to/bin> and that didnt work when the binary was overwritten20:02
frinnstand a bunch of other crap20:02
frinnsthaving to manually kill processes before restarting them when upgrading20:03
*** dougl has joined #crux20:03
*** dougl has quit IRC20:04
frinnststart-stop-daemon allows for simple scripts with added features. very small price to pay i think20:04
jaegerI think it's a nice change20:04
frinnstand is the default to really overwrite scripts in rc.d/ ? maybe its just me but i have that turned off on all my systems20:05
joacimfrinnst likes it. it must be  bad!20:05
frinnstits probably better to tweak pkgadd.conf20:05
jaegerUPGRADE     ^etc/rc\.d/net$     NO20:07
jaegerthat's default20:07
jaegeryeah. that's default. so just net20:07
joacimfrinnst: you use linux software raid, right? Do you think RAID6 is safe?20:10
frinnstI though that was the default for everything in rc.d :)20:10
joacimbeen thinking about getting another 3 drives for my home server and configure make a RAID6 array.20:10
frinnstmd is safe yes. If raidN is safe it depends on what you intend to use it for20:10
frinnstwith 4 drives raid10 is a better choice i think20:11
frinnstfaster too20:11
joacimfiles and backups of other computers20:11
frinnstavoid raid520:11
joacimthe plan was to build a raid5 array, then transfer data over, and then grow to raid620:11
frinnstyou could build a degraded raid10 array with 2 drives i think20:12
joacimand then grow to raid6 or raid10 (with 4 drives) from there?20:13
frinnstim unsure how you can change raidlevels with md20:13
joacimi think you add spares, and then grow20:13
joacimi think20:13
frinnstwith raid5 you would probably want to get a ups20:14
prologicjust been reading the backlog a bit20:14
prologichave to catch bus to work very shortly20:15
prologicjust one question20:15
prologicwhat's wrong with using kill, pkill, pidof anyway?20:15
joacimwhat about ecc ram? recommended?20:15
frinnstprologic: many of our initscripts are broken20:15
joacimwas planning on getting some cheap amd apu for the task, which doesnt support ecc ram20:16
prologicjoacim, only if you care about data integrity20:16
prologicfrinnst, can you explain why/how?20:16
frinnstecc is recommended yes20:16
prologicI personally haven't had issues with any of our init scripts myself20:16
frinnstprologic: do you use dovecot or nfsserver ?20:17
frinnsttwo examples that comes to mind20:17
frinnstor bind20:17
joacimthe data isnt exactly important, but it would be a pain in the ass to lose all those files20:17
prologichmm I have used them; not recently20:17
frinnstif you upgrade these ports and then run /etc/rc.d/<port> restart20:17
frinnstit wont kill the process20:17
prologicwhat happens? does stop) not kill the daemons proeprly?20:17
prologichmm really?20:18
prologicbecause the binary was overwritten?20:18
jaegerYou have backups anyway, right? right? :)20:18
prologicanyway have to dash -- will answer from work20:18
jaegerMy data is on a server with ZFS and ECC but I still back it up externally20:18
joacimi'm not doing that many writes, so i figured raid6 would be decent. I mostly read files.20:23
joacimit'll host media files that i access from my desktop pcs20:23
greduanok so with the start-stop-daemon thing, is there an example how this would be used in the rc scripts?20:24
joacimthe new openssh script i guess =)20:24
frinnstexim too20:24
frinnstnight night20:24
joacimthis early?20:25
joacimgood night old man20:25
*** cezar has joined #crux20:48
nwecezar: hello20:50
nwedid you get it to work with splix.. ?20:50
*** Feksclaus has joined #crux20:56
*** doomicide has joined #crux21:18
*** dougl has joined #crux21:18
*** dougl has quit IRC21:21
*** z3bra has joined #crux21:24
*** lnds has quit IRC21:30
cezarnwe:sorry for the delay and tks for your attentio and interest. unfortunately I did not. but it is not a splix problem but a compilation problem. I will show you just a moment...21:41
nwecezar: do you have Pkgfile for cezar I can get?21:51
*** diverse has joined #crux21:53
cezarnwe: :( no...21:58
nweokey I will check one thing..21:59
*** prodhe has joined #crux22:03
*** prodhe has quit IRC22:04
*** prodhe has joined #crux22:04
*** prodhe has quit IRC22:05
*** dougl has joined #crux22:15
nwecezar: there?22:20
nwego into you splix-2.0.0 directory22:21
nwethen cd src22:25
nwevim ppdfile.cpp22:25
nwego to row 28522:25
nweand change PPDFile::Value::Value &val _TO_ PPDFile::Value &val22:26
nweand then cd ..22:26
nweand build the source22:26
nwebut now time to for sleep22:27
cezarnwe:great will let you know the result tomorrow tks enormously for now22:27
*** Workster has joined #crux22:30
*** JSchilli1 has quit IRC22:33
*** dougl has quit IRC22:35
*** mhe has quit IRC22:36
*** zXter has quit IRC22:54
*** pidsley has joined #crux23:06
*** pidsley has left #crux ()23:07
*** dougl has joined #crux23:26

Generated by 2.11.0 by Marius Gedminas - find it at!