1108339994 Q * DuckMaster Quit: Client exiting
1108340134 Q * duckx Remote host closed the connection
1108340183 J * duckx ~Duck@dyn-83-157-159-137.ppp.tiscali.fr
1108340586 N * Bertl_oO Bertl
1108340594 M * Bertl evening folks!
1108340628 M * Doener evening Bertl!
1108340667 M * Bertl hey Doener! how are you?
1108340706 M * Doener didn't sleep too well the last few days, but i guess a night without sleep will fix that ;)
1108340719 M * Bertl good starting point! ;)
1108340771 Q * duckx Remote host closed the connection
1108340774 M * Bertl so what are your plans for this night?
1108340876 M * Doener a) get some php mail reminder done, b) use a) to test my php framework (which itself is wip), c) take a look at the extended (at least the wiki ml says so ;) kernel to-do list (which is what i'm doing right now)
1108341206 J * duckx ~Duck@dyn-83-157-159-137.ppp.tiscali.fr
1108341309 M * Bertl and, do you like to address anything of the listed stuff?
1108341402 Q * yarihm Quit: Leaving
1108341584 M * Doener are you already working on something?
1108341598 M * Bertl not at the moment ...
1108341633 M * Doener guess i'll take a look at the copy_to_user stuff
1108341685 M * Bertl okay, good choice I guess ...
1108341700 J * yarihm ~yarihm@80-218-0-127.dclient.hispeed.ch
1108341709 M * Bertl wb yarihm!
1108341758 M * virtuoso Hey guys.
1108341768 Q * yarihm Quit: 
1108342138 M * Bertl hey virtuoso!
1108342677 M * Bertl I guess I'm off to bed now .. to tired for anything ...
1108342692 M * Bertl have a nice whatever everyone! cya tomorrow!
1108342697 M * Doener night Bertl!
1108342704 N * Bertl Bertl_zZ
1108356378 J * nox- ~nox@c207253.adsl.hansenet.de
1108356724 Q * nox Ping timeout: 480 seconds
1108356739 N * nox- nox
1108362437 Q * virtuoso Read error: Operation timed out
1108362865 J * virtuoso ~s0t0na@tranq.dorms.spbu.ru
1108363423 J * DaPhreak ~DaPhreak@pc-outside.uni-greifswald.de
1108365535 N * BobR_oO BobR
1108367063 J * lars||| ~ledj@213-156-52-105.fastres.net
1108368138 J * mhepp ~mhepp@r72s22p13.home.nbox.cz
1108368264 Q * lars||| Quit: using sirc version 2.211+KSIRC/1.3.11
1108369536 Q * berni Quit: Reconnecting
1108369537 J * berni ~berni@eth0.svr01.mucip.net
1108369837 J * DuckMaster ~duckx@195.75.27.158
1108370566 J * Val ~val@gj403.loria.fr
1108370666 J * prae ~prae@ezoffice.mandrakesoft.com
1108370715 N * BobR BobR_afk
1108370837 J * DJ-Bus-Driver aaa@CWPP-p-203-54-241-187.prem.tmns.net.au
1108372169 J * rs rs@ice.aspic.com
1108372172 M * rs hi there
1108372649 M * DaPhreak morning rs :)
1108373018 Q * DaPhreak Quit: leaving
1108373638 J * DaPhreak ~DaPhreak@pc-outside.uni-greifswald.de
1108373998 N * Bertl_zZ Bertl
1108374013 M * Bertl morning folks!
1108374046 M * DaPhreak morning Bertl 
1108374167 M * rs hey bertl!
1108374190 M * Bertl morning DaPhreak, rs!
1108374206 M * DaPhreak hmmm Bertl, -rc4 works nice @ home :)
1108374247 M * DaPhreak and today herbert xu sent me a mail with a patch, which hopefully fixes my local_bh_enable thingy
1108374311 Q * DaPhreak Quit: switching nic
1108374351 J * DaPhreak ~DaPhreak@pc-outside.uni-greifswald.de
1108374364 N * BobR_afk BobR
1108374377 M * Bertl DaPhreak: yeah, saw that one ...
1108375444 M * Bertl rs: so did you spend some thoughts on IPC limitations/accounting?
1108375741 M * rs not really, I just though about it but didn't search for all kind of accouning we should do
1108375780 M * Bertl k, so what are your priorities there?
1108375819 M * rs about IPC or in general ?
1108375986 M * Bertl well, if you ask ;)
1108376211 J * Hollow ~Hollow@home.xnull.de
1108376238 M * Bertl welcome Hollow!
1108376244 M * Hollow heya Bertl 
1108376291 M * Hollow Bertl: utils 203 failed with vconfig not found, but i didn't find a ./configure switch to turn it off, is it needed now?
1108376425 M * Bertl seems like enrico added a dependacy for the vlan tools, but I guess it should not hurt, pretty tiny package ...
1108376434 M * Bertl +n
1108376443 M * Hollow ok, just wanted to know
1108376986 M * Hollow hm, compilation failed http://paste.phpfi.com/51029
1108377044 M * Bertl non x86?
1108377048 M * Hollow no, x86
1108377053 M * Hollow athlon xp
1108377060 M * Bertl you got no syscall anymore?
1108377076 N * BobR BobR_afk
1108377079 M * Hollow dunno, where can i find out?
1108377111 M * DaPhreak should be linux-headers ;) or in your case probably linux26-headers :)
1108377115 M * Hollow ah.. important: i have hardened gcc
1108377136 M * Bertl which means?
1108377155 M * Hollow SSP, PIE, stack protector etc
1108377202 A * Hollow looks if 196 compiles with hardened
1108377232 M * Bertl hmm, looks to me like you are missing vital library/include files ...
1108377254 M * DaPhreak Bertl: is that sys/syscall.h what you mean ?
1108377257 M * Hollow i'll try reinstalling them..
1108377281 M * Bertl it should be either /usr/include/unistd.h or /usr/include/syscall.h
1108377297 M * Hollow i have both
1108377318 M * Hollow but util-vserver is compiled with dietlibc.. no?
1108377334 M * Bertl if available, yes ... which version?
1108377360 M * DaPhreak 0.27
1108377362 M * Hollow dev-libs/dietlibc-0.27
1108377364 M * Hollow ;)
1108377411 Q * DaPhreak Quit: switching nics
1108377419 M * Bertl okay, does the 0.30.196 compile there?
1108377432 M * Hollow no
1108377437 M * Hollow just failed
1108377447 J * DaPhreak ~DaPhreak@pc-outside.uni-greifswald.de
1108377466 M * Bertl DaPhreak: what does 'nics' stand for in your quit message?
1108377477 M * DaPhreak nic = network interface :)
1108377504 M * Bertl and you switch your nic for what purpose?
1108377556 M * DaPhreak heh, i was on cable to upload something (into our server network) and now im back on wireless :)
1108377578 M * Hollow :)
1108377618 M * halorgium DaPhreak: you should have the dhcpd giving you the same ip addy for both mac addys
1108377655 M * Hollow Bertl: can it be pie or ssp? i can switch to a different gcc profile for testing..
1108377664 M * Bertl well, question is, why leave wireless when connecting to cable ...
1108377674 M * DaPhreak nope .. it's separated (has to do something with those nasty students at our university)
1108377688 M * Bertl Hollow: try that, but I would look for some library issues ...
1108377705 M * DaPhreak yeah Bertl, i could also set a new route over the cable :)
1108377763 M * Bertl Hollow: try to test compile some simple test code which uses syscall() ...
1108377824 M * Hollow i'm no c programmer sry
1108377951 M * Hollow vanilla gcc works
1108377970 M * DaPhreak so it's an issue with hardened
1108377992 M * Hollow yep
1108378186 M * Hollow seems to be a PIE issue
1108378191 M * Hollow hardened without pie worked too
1108378385 N * BobR_afk BobR
1108378555 M * DaPhreak you should bug solar / pebenito with this Hollow ;)
1108378640 M * Hollow i'll do
1108379428 Q * monrad Quit: Leaving
1108379670 Q * DJ-Bus-Driver Ping timeout: 480 seconds
1108380704 M * prae Bertl: I talked about vserver to Mandrake's sysadmins... and, good new, Mandrake's sysadmins should use vserver in production servers.
1108380710 M * prae They are happy to test it now.
1108380734 N * BobR BobR_afk
1108380749 M * Bertl prae: sounds good!
1108380947 M * prae I haven't any time to test and integrate vserver into corporate mdk now, but two machines are installed (in localnet, another in net)
1108381001 M * prae this servers should be used to test vserver
1108381038 M * prae (localnet for me, kernel mnt; net for sysadmins)
1108381081 M * Bertl keep me updated if you encounter any (kernel) issues ...
1108381207 M * prae no problem'
1108382154 M * aba Hm, I have issues running portmap inside the vserver, because "cannot find any active local network interfaces". Any hint?
1108382184 M * Bertl make lo visible?
1108382200 M * Bertl configure portmap to _not_ use local interfaces?
1108382248 M * aba (running user space nfs-server is possible inside?)
1108382309 M * Bertl should be ... IIRC somebody reported it working ...
1108382456 M * aba Hm, how do I make lo visible inside?
1108382488 M * Bertl which kernel version?
1108382509 M * aba 2.6.
1108382525 M * aba (but no NGNet)
1108382575 M * Bertl you can try to just remove the hide_netif flag
1108382585 M * Bertl (maybe that's sufficient for the portmapper)
1108382737 M * Bertl if that doesn't work, add 127.0.0.1 as existing ip
1108382792 M * aba Hm. I'll try. Thanks.
1108382837 M * Bertl you're welcome ... but basically you should fix/config portmap.
1108382943 M * aba I'll try where I succeed more. Portmap seems to use 127.0.0.1 also for authentication purposes ...
1108383221 M * Hollow Bertl: do you know if vserver works with pax/grsec/selinux?
1108383307 M * Bertl it seems so, as a) se-linux is in the kernel, and folks provide patches for grsec for some time now ...
1108383318 M * Bertl forget the b) ;)
1108383338 M * Bertl argl *forgot ...
1108383407 M * aba Bertl: the portmap-problems seems to be that pmap_check doesn't consider my other ip-address as local.
1108384130 M * Hollow Bertl: ok, i'll try it with selinux some day this week
1108384139 M * Hollow i'm just migrating to hardened with my machines...
1108388633 T * * http://linux-vserver.org/ | latest stable 1.2.10, devel 1.9.4, ng9.0   -- He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime 
1108388633 T * Bertl -
1108389365 Q * mhepp Quit: mhepp caught signal: Autobus error
1108394065 Q * DaPhreak Quit: going homw
1108396526 N * BobR_afk BobR
1108396590 N * BobR BobR_oO
1108398049 J * ntrs ntrs@Dardeene-68.188.50.87.charter-stl.com
1108398370 Q * ntrs_ Ping timeout: 480 seconds
1108398628 J * pusling_ ~pusling@195.215.29.124
1108398653 Q * pusling Read error: Connection reset by peer
1108398677 N * pusling_ pusling
1108398818 J * ntrs_ ntrs@Dardeene-68.188.50.87.charter-stl.com
1108399159 Q * ntrs Ping timeout: 480 seconds
1108399748 M * Bertl okay folks, off for now, maybe back later ...
1108399758 N * Bertl Bertl_oO
1108400714 Q * pusling Ping timeout: 480 seconds
1108400870 J * pusling ~pusling@195.215.29.124
1108400912 J * pusling_ ~pusling@195.215.29.124
1108401054 Q * prae Quit: Client exiting
1108401265 J * kalou ~kalou@AToulon-201-1-5-83.w81-53.abo.wanadoo.fr
1108401267 M * kalou Hi
1108401298 Q * pusling_ Quit: leaving
1108401418 M * kalou Just patched my 2.6.10 source tree with 1.9.4, but got mm/mmap.c.rej . Not sure how to fix this file.
1108402363 M * micah with the 1.9.4 tools, when I create a new vserver it does not set my hostname to the vserver name... it does with the old tools somehow
1108404196 Q * rs Quit: leaving
1108409438 Q * kalou Quit: Leaving
1108411188 J * monrad ~monrad@213083190130.sonofon.dk
1108415685 J * DuckKing ~Duck@dyn-83-157-204-110.ppp.tiscali.fr
1108416100 Q * duckx Ping timeout: 480 seconds
1108419036 J * yarihm ~yarihm@80-218-0-127.dclient.hispeed.ch
1108419305 Q * yarihm Quit: 
1108419864 J * yarihm ~yarihm@80-218-0-127.dclient.hispeed.ch
1108419867 Q * yarihm Quit: 
1108422239 N * Bertl_oO Bertl
1108422258 M * Bertl evening folks!
1108422316 M * mugwump hi there Bertl !
1108422324 M * mugwump nice one on the release btw
1108422467 M * micah evening Bertl !
1108422523 M * micah does vserver clean /var/run somehow on starting up a vserver? I know that debian does this in its initscript, but if I comment out (or even remove that initscript), it still happens
1108422547 M * mugwump that used to be in /usr/sbin/vserver, a long time ago iirc
1108422560 M * Bertl hey mugwump! thanks!
1108422569 M * micah the reason I ask is because I do a mount --rbind in all my vservers that need to access /var/run/mysqld/mysqld.sock, but if I stop and start any one of those vservers, it removes the mysql socket for everyone :(
1108422591 M * mugwump dang
1108422605 M * Bertl hi micah! I'd suspect the stop scripts to do that ...
1108422611 M * micah yes, I am eager to try 1.9.4 final out once 2.6.11 is finalized as well (I'm running -rc4)
1108422624 M * Zoiah micah: yes, the new (not-legacy) vserver scripts do that.
1108422625 M * Bertl (I mean the vserver stop part)
1108422626 M * micah Bertl: the vserver stop scripts?
1108422632 M * micah hmm
1108422654 M * micah ok, I was really starting to wonder what was happening
1108422662 M * Zoiah     case "$INITSTYLE" in
1108422662 M * Zoiah         sysv)
1108422662 M * Zoiah             { find var/run  ! -type d -print0; \
1108422662 M * Zoiah               find var/lock ! -type d -print0; } | xargs -0r $_CHROOT_RM
1108422662 M * Zoiah             ;;
1108422682 M * Bertl yes, makes sense ...
1108422693 M * Zoiah Bertl: why?
1108422701 M * mugwump heh, it used to be just rm -rf var/run/* :)
1108422713 M * mugwump even typing a command like that in irc made my fingers nervous
1108422720 M * Bertl because too many services refuse to start when not stopped ...
1108422739 M * micah Zoiah: where is that?
1108422747 M * Zoiah micah: /lib/util-vserver/vserver.functions
1108422750 M * Bertl s/stopped/removed from var/run/
1108422777 M * micah hmm
1108422794 M * micah what should I do then to get around this?
1108422822 M * mugwump alter vserver.functions to allow this to be config'd, then send a patch to enrico imho
1108422832 M * micah I suppose configuring mysql to use networking and to access over the network
1108422838 M * micah instead of through the socket
1108422961 M * mugwump yeah.  then you can do host-based access control too
1108422977 M * Bertl micah: do you really use _one_ socket for more than one vserver?
1108422997 M * micah Bertl: the idea was to have a mysql vserver and all the other vservers that require mysql would connect to that one
1108423052 M * Zoiah Bertl: isn't that something that should be figured out inside a vserver itself?
1108423053 M * Bertl okay, network is probably the safest way to go here, but you can also try to copy the socket on vserver startup (via the startup script)?
1108423054 M * Zoiah Bertl: the removal?
1108423080 M * Bertl the sysv version doesn't call the 0 and 6 runlevels
1108423088 M * Zoiah Or atleast make it somethig configurable.
1108423109 M * Bertl (because of all the hardware related stuff there)
1108423129 M * Bertl but yes, feel free to submit a patch to make it configurable (guess enrico will accept it)
1108423165 M * Zoiah I just changed initstyle to minit. I'm not using the sysv stuff anyhow. :)
1108423167 M * micah if I setup mysql to be accessed over the network between vservers, do I just put it on a private IP and then configure each vserver to also have that network available?
1108423224 M * Bertl or use the 'public' ip or whatever of the mysql ...
1108423241 M * Bertl (but yes, you can add a network for that)
1108423375 M * mugwump or use the 127.* network
1108423381 M * micah if I do not configure a vserver to have an IP, then it simply does not have one?
1108423394 M * micah mugwump: i thought 127.* was not accessible between vservers
1108423480 M * mugwump it's not routable, to normally configured routers, that's all.  I think only 127.0.0.1 is treated specially
1108423561 M * micah hmm
1108423571 M * micah bind-address    = 127.0.0.1
1108423581 M * micah in my.cnf doesn't seem to cut it (netstat shows it bound to 0.0.0.0)
1108423612 M * Bertl 127.0.0.1 is mapped to the ip of the vserver
1108423636 M * Bertl (without ngnet that is ;)
1108423689 M * micah and if the vserver has no IP?
1108423823 M * Bertl good question, then it probably isn't mapped to anything but 0.0.0.0 ;)
1108423907 M * micah should I configure an interface for my mysql vserver to have 127.0.0.1?
1108423912 M * micah or 10.0.0.1 or something?
1108423928 M * Bertl 127.0.0.1 is no good idea ... 10.0.0.1 would be fine ...
1108423948 M * micah so every vserver that is going to want to connect to mysql will also need a 10.0.0.x address
1108424047 M * Bertl yes, tht's correct ...
1108424073 M * Bertl you could try to use 127.0.0.2 (just for a test, no guarantee)
1108424081 M * micah it seems like there must be a way to do that without requiring an IP, otherwise I will need to manage IP allocations for vservers
1108424100 M * Bertl and just access that from inside a vserver ...
1108424148 M * micah I will try that to see what happens
1108424218 M * micah aha!
1108424228 M * micah ERROR 1130: Host '127.0.0.2' is not allowed to connect to this MySQL server
1108424236 M * micah which means that it is just mysql access list I need to fix
1108424680 M * micah ok, last question from me today... with the 1.2 tools my hostname is set to the name of the vserver when I create it, however with the 1.9.4 tools, it is set to the host's hostname, if I change it in the vserver it will be reset to the host's name when I reset the vserver
1108424743 M * Bertl on 1.9.x you have 2 options ...
1108424776 M * Bertl you can a) configure a host name from outside, or b) allow the guest to change it from inside
1108424830 M * micah it seems like when I do b) it does not stick across vserver restarts
1108424835 M * micah but... I'd be happy to do a)
1108424927 M * Bertl let's see what the flower-page says ...
1108424954 M * Bertl # /etc/vservers/vserver-name/uts 
1108424963 M * micah the flower page?
1108424969 M * Bertl # nodename
1108424969 M * Bertl The node-/hostname of the vserver
1108424974 M * Bertl http://www-user.tu-chemnitz.de/~ensc/util-vserver/doc/conf/configuration.html
1108425007 M * micah ah the "flower" page :)
1108425024 M * micah which I have made more readable here: http://red.eggplantmedia.com/~arthur/micah.html
1108425068 M * Bertl hmm, looks the same to me ;)
1108425079 M * micah yes, but without flowers
1108425088 M * micah that flower page I cannot read the text at all
1108425093 M * Bertl that's selectable in my browser (thank's css)
1108425096 M * micah ahh
1108425097 M * micah :)
1108425107 M * micah do you know the difference between nodename and sysname?
1108425112 M * Bertl but it would be a nice thing to 'update' the page ...
1108425134 M * Bertl (each util-vserver contains an xml stylesheet which should be more up-to-date)
1108425153 M * Bertl micah: check uname and the various components
1108425190 M * micah is there a way to specify uts/nodename on the newvserver command line?
1108425190 M * Bertl (uname -s vs. -n)
1108425210 M * Bertl AFAIK newvserver is not part of the new tools ;)
1108425236 M * micah sorry... vserver <vserver> build
1108425237 M * Bertl so you have to ask/check with whoever is maintaining it ...
1108425254 M * Bertl yes vserver build should know about that with the 'options'
1108425279 M * micah ahh, ok, vserver build build --help
1108425290 M * Bertl exactly, seems to be --hostname ;)
1108425319 M * micah ah, this is great, I can do most of this :)
1108425321 M * micah on the commandline
1108425396 M * micah hmm, --netdev vs. --interface
1108425545 M * micah I will have to experiment