1142294443 Q * rs Quit: rs 1142294551 M * Aiken http://pastebin.com/600714 1142294622 M * Bertl sounds interesting for linux-vserver ... many substems seem to be affected 1142294784 M * Aiken the 2nd run was 18 seconds faster 1142295190 M * Bertl doener_: ping? 1142295197 M * Bertl phreak``: ping? 1142295418 Q * peter_ Remote host closed the connection 1142295596 M * Bertl okay, guess I will have a nap .. probably back later ... have fun and thanks for testing! 1142295620 N * Bertl Bertl_zZ 1142296107 Q * doener_ Quit: leaving 1142296958 J * matta ~matta@71.224.125.126 1142297940 J * Adi MC@212.106.211.219 1142298010 P * Adi 1142298444 Q * click Ping timeout: 480 seconds 1142299052 J * click click@ti511110a080-3151.bb.online.no 1142300375 J * rs ~rs@vol75-7-82-229-177-124.fbx.proxad.net 1142300858 Q * lost_eps Quit: Leaving 1142300862 Q * phycho Quit: 1142300993 J * phycho ~phycho@ext-gw.darktech.org.uk 1142300995 M * phycho hey guys 1142301000 M * phycho does the vservers not have a loopback address? 1142301051 M * mugwump no 1142301069 M * mugwump blame glibc for unnecessary assumptions 1142301084 M * phycho k 1142301091 M * phycho should i add one? 1142301096 M * phycho (on the vlan interface) 1142301097 M * mugwump vserver 2.1.x has a method to virtualise entire network stack, I think this is what openvz does 1142301099 M * phycho i need loopback 1142301111 M * mugwump what for? 1142301116 M * phycho its 2.1 im running 1142301121 M * phycho i want some services to listen on localhost only 1142301124 M * phycho (mysql for starters) 1142301135 M * Cru you may assign an IP on lo to the guest and add a localhost record to /etc/hosts, but not every programm will accept is as 127.0.0.1 is often hardcoded 1142301152 M * mugwump in particular the DNS resolver in glibc makes this assumption 1142301155 M * phycho yea thats what i was gonna do 1142301161 M * phycho ah k 1142301162 M * mugwump s/DNS // 1142301184 M * mugwump if you want entire virtualised network stack, you need to disable legacy network support 1142301192 M * phycho k 1142301210 M * phycho would people be able to sniff if i want virtual network stack? 1142301255 M * phycho (obviously thats not what i want) 1142301260 M * Cru I assume only when CAP_NET_RAW is permitted 1142301269 M * phycho k 1142301292 M * phycho i assume if i allow that and the users on vlans, they can only sniff their own vlan right? 1142301371 M * mugwump AIUI with virtualised network stack, sniffing won't see other stack's traffic 1142301391 M * phycho k 1142301393 M * phycho cool 1142301393 M * mugwump however normally tcpdump can't be used in a vserver because it can't be properly virt 1142301398 M * phycho hehe k 1142301404 M * phycho sounds like what i want 1142301411 M * phycho thanks 1142301517 M * phycho unable to connect back to your own external ips in vserver either? 1142301522 M * phycho root@www:/chroot/apache# telnet 192.168.2.3 22 1142301523 M * phycho Trying 192.168.2.3... 1142301525 M * phycho that does nothing 1142301534 M * phycho the ip is 192.168.2.3 on the box (im ssh'ed into it) 1142301579 M * phycho thats one thing that i need more than anything else.. for apache to interact with mysql (both are chrooted properly so they cant interact with /tmp/mysql.sock) 1142304203 J * Smutje_ ~Smutje@xdsl-84-44-247-110.netcologne.de 1142304319 Q * Smutje Ping timeout: 480 seconds 1142304319 N * Smutje_ Smutje 1142305510 Q * phycho Quit: 1142307232 J * _ruben_ ~probe@84.90.83.125 1142311351 Q * Hunger jupiter.oftc.net europa.oftc.net 1142311429 J * Hunger Hunger.hu@Hunger.hu 1142312088 Q * _ruben_ Quit: leaving 1142312560 J * Hmmmm ~Hmmmm@221.135.51.19 1142318149 J * coocoon ~coocoon@p54A0570A.dip.t-dialin.net 1142318157 M * coocoon morning 1142318463 J * _chris ~chris@210.213.216.88 1142319880 J * mnemoc ~amery@user4-2.tutopia-dialup.ifxnw.cl 1142320191 M * coocoon some slackware guru here 1142321818 M * bubulak what you need coccoon / 1142321819 M * bubulak ? 1142322060 M * _chris has anybody tried installing vserver on slack 10.2? 1142322352 M * bubulak yes ... i 1142322368 M * bubulak no problem whit lastest utils 1142322426 M * _chris got notes on the installation? 1142322454 J * DataCompBoy ~datacompb@217.8.236.1 1142322476 M * DataCompBoy Hi guys :) 1142322490 M * bubulak _chris: you dont need notes 1142322500 M * bubulak just ... patch youre kernel 1142322503 M * DataCompBoy Does someone knows how to make one SSH server for all vservers? Like virtozzo can? 1142322508 M * bubulak compile utils 1142322514 M * bubulak and use wiki page 1142322544 M * _chris how bout the init? works fine on default configurations? 1142322569 M * bubulak yes ... works fine 1142322577 M * _chris thanks 1142322579 M * bubulak just 1142322581 M * coocoon biubulak: sorry 1142322582 M * bubulak w8 1142322615 M * DataCompBoy is that possible at all? 1142322636 M * bubulak in /etc/vservers//apps/init/style must be word "plain" 1142322638 M * bubulak to work 1142322647 M * coocoon bubulak: wanted to remove the HOSTMNAME in the guests but then I got message that the hostname isn't there, but in other guests there is no hostname and it works 1142322664 M * coocoon bubulak: that the vserver given name appears 1142322696 M * bubulak hmm 1142322729 M * bubulak if you use vserver start ? 1142322753 M * coocoon vserver guest yes 1142322787 M * bubulak hmm 1142322806 M * coocoon but also when I removed HOSTNAME in the console appears the name root@darkstar: 1142322808 M * bubulak for initscript is that name of directory /etc/vserver/ 1142322833 M * bubulak an than file name in that directory 1142322839 M * coocoon what does that mean 1142322848 M * coocoon thats all there therefore I ask 1142322851 M * bubulak coocoon: :))) ach thaaat 1142322852 M * bubulak :) 1142322906 M * coocoon another one is that starting pkgtool in slackware 10.0 10.1 10.2 failled in 9.0 9.1 it works fine 1142322927 M * coocoon also I wanted to know that I can fix the slackware sources to the network folders 1142322929 M * bubulak edit /vserver//etc/HOSTNAME 1142322938 M * coocoon yes but how 1142322943 M * coocoon remove all in there 1142322956 M * coocoon the name also appear if there is no HOSTNAME 1142322966 M * bubulak hmm 1142322970 M * coocoon yes 1142322972 M * coocoon ;-) 1142323024 M * bubulak edit /etc/hosts to 1142323027 M * bubulak o 1142323027 M * coocoon I think it must be slackware specific, who has invent HOSTNAME 1142323169 M * coocoon that doesn't work 1142323183 M * coocoon mom 1142323186 M * bubulak hmm ... i just every time edit these two files and than that work 1142323227 M * coocoon how u edit HOSTNAME 1142323252 M * coocoon u give in the hostname right 1142323259 M * coocoon but this mustn't be 1142323270 M * coocoon I wanted to use this guest as template 1142323300 M * bubulak echo "hostname" > /vserver/guest/etc/HOSTNAME ? 1142323306 M * bubulak that you mean ? 1142323354 M * bubulak w8 i test that 1142323365 M * coocoon in my other guests there is no hostname file 1142323392 M * coocoon now I have removed the HOSTNAME file and restart the guest now it is there again 1142323402 M * bubulak yes 1142323427 M * coocoon also my mtab changed everytime 1142323429 M * bubulak thats is defined in /etc/rc.d/rc.M if doesn't exist create one 1142323459 M * coocoon how? 1142323468 M * coocoon wait 1142323553 M * bubulak hmm 1142323566 M * bubulak i wrote this script 1142323587 M * bubulak http://www.wnet.sk/~bubulak/ 1142323605 M * bubulak http://www.wnet.sk/~bubulak/build.slack.vserver 1142323620 M * bubulak just edit it for youre purpose 1142323627 M * bubulak and on end 1142323627 M * coocoon i have copied rc.d form another slack10.2 guest, could I use it also for other distros 1142323637 M * bubulak change /vserver/guest/etc/HOSTNAME 1142323809 M * DataCompBoy agh, people -- one also question: how to create separate loopback adapter for guests ? 1142323863 M * bubulak hmm 1142323956 M * DataCompBoy virtuozzo also able to do that :( 1142323966 M * bubulak ask bertl 1142323987 M * bubulak but now sleep 1142323987 M * DataCompBoy he is 'zZ' now :) 1142323990 M * bubulak :) 1142324006 M * DataCompBoy who is sleeping at 2pm ?! :] 1142324026 M * bubulak maybe have another timezone :) 1142324047 M * bubulak i have 9:00am 1142324049 M * coocoon thanx for the script 1142324060 M * bubulak np 1142324100 M * coocoon sorry but I have done it in another way, and it works with 9.0 and 9.1 but not with 10.* 1142324151 M * coocoon why does slack creat everytime new mtab and hostname file 1142324172 M * coocoon DataCompBoy: could u give more details for what u need this 1142324443 M * DataCompBoy coocoon: I have several services, that hardcoded to use 127.0.0.1 to work -- it creates socked and connect to itself. 1142324483 M * DataCompBoy i think, that best idea is create new loopback adapter, call it lo-vhost1, and rename it in that vserver to lo 1142324537 M * coocoon DataCompBoy: http://list.linux-vserver.org/archive/vserver/msg11528.html 1142324591 M * coocoon maybe this will help, u know mailinglist http://list.linux-vserver.org/archive/vserver/, I have searched loopback 1142324744 M * DataCompBoy oo, my case:) thanks 1142324753 M * coocoon uw 1142324754 M * DataCompBoy  1142324760 M * DataCompBoy ok, reading 1142324765 M * coocoon bubulak are u still there 1142324768 M * DataCompBoy but -- what to do with SSH ? 1142324779 M * coocoon nat 1142324791 M * coocoon dnat 1142324823 M * DataCompBoy I have two vservers 1142324827 M * coocoon maybe i do not know exactly what u want, but I have no official ips so i use that 1142324855 M * DataCompBoy and virtuozzo allow to have both servers share one ssh port 22 -- i just log as root@vserver1 or root@vserver2 1142324870 M * DataCompBoy on one IP. 1142324975 M * coocoon i do not know how virtuozzo works, but i know to assign another port for the guest to use ssh and so u must create a rule, to access the vservers ssh 1142325014 M * DataCompBoy yep, via different ports that easy, but ugly -- and also problem when software have hardcoded connect to 22 :( 1142325014 M * coocoon i am not a specialist u know i do it in this way, haven't use virtuozzo ever before 1142325033 M * coocoon but it works with ssh 1142325050 M * coocoon some cofniguration must be done 1142325053 M * coocoon ;-) 1142325062 M * coocoon handwork 1142325084 M * DataCompBoy I just have friends in swsoft:) virtuozzo use some proprietary auth module, that allows to auth on user in vserver, and after successfull auth it log into that vserver. 1142325118 M * Wonka sounds quite b0rked, that approach 1142325128 M * Wonka and incompatible 1142325165 M * Wonka how would it differentiate root@vs1 and root@vs2? username is both "root" 1142325181 M * DataCompBoy usernames are root@vs1 and root@vs2 1142325191 M * DataCompBoy root -- is root for host 1142325333 M * Wonka aah 1142325340 M * Wonka even more brokenness 1142325365 M * Wonka but then, the hosts just acts as a proxy 1142325391 M * Wonka and he can sniff anything going through the line *shudder* 1142325467 Q * shedi Quit: Leaving 1142325513 M * DataCompBoy no, hosts not a proxy. 1142325539 M * DataCompBoy first of all: host have access to any vservers. that is mainline. 1142325620 M * DataCompBoy so if host act as proxy -- that not lowering a security of vservers. 1142325837 M * DataCompBoy ooo, also good question: does vhosts patch iproute2, so I can match on xid ? 1142326178 M * daniel_hozac no. 1142326249 M * daniel_hozac primarily because all networking happens on the host. 1142326267 M * daniel_hozac ergo, there is no xid. 1142326316 M * coocoon daniel_hozac: do i need e2fsprogs in the guest 1142326349 M * daniel_hozac depends on the distro. e2fsprogs has a library that's quite common to use. 1142326367 M * coocoon daniel_hozac: slackware 1142326388 M * daniel_hozac i have never used slackware. 1142326403 M * daniel_hozac (the 5 minutes it took to realize it had no sane package management was enough) :) 1142326414 M * coocoon daniel_hozac: I have problem slackware always overrides my fsatb and mtab will be erased 1142326469 M * coocoon a virtual scanner hasn't be invented till yet 1142326505 M * daniel_hozac coocoon: you'll probably want to edit your initscripts. 1142326530 M * coocoon ok I must have a look at them 1142326599 M * coocoon but for what nust i have a look exactly 1142326622 M * coocoon i have no idea and plan for that 1142326684 M * coocoon ok I think I have found something 1142326722 M * daniel_hozac you could use http://debian.marlow.dk/vserver/guest/_slack_10.diff as a base. 1142326852 J * pagano ~pagano@lappagano.cnaf.infn.it 1142326853 M * coocoon ah ok set this instead of the other rcM.d 1142327149 M * coocoon thanx a lot 1142327863 J * gudi ~maurer@nm.it-mn.de 1142327864 M * gudi hi 1142327872 M * gudi is it possible to get traceroute wrok in a vserver ? 1142327897 M * eyck NET_RAW is probably needed 1142327911 M * gudi allready have that 1142327926 M * gudi web:/# traceroute web.de 1142327926 M * gudi traceroute: ifaddrlist: SIOCGIFADDR: eth1: Cannot assign requested address 1142327933 M * eyck then it should work 1142327947 M * eyck what address does it request? 1142327956 M * gudi what you mean ? 1142328043 M * eyck it looks like it's trying to use wrong source address 1142328059 M * gudi how to fix it ? 1142328095 M * eyck hmm, try traceroute -i YOUR_CORRECT_IFACE web.de 1142328117 M * eyck or even -i iface -s CORRECT_SOURCE_ADDRESS 1142328224 M * pzYsTorM (using nmap u have to do the same stuff with iface and source adress) 1142328258 M * eyck if it works, you should figure out why does it try using wrong ones 1142328273 M * gudi eyck, nope it not work 1142328287 M * gudi web:/# ip addr show 1142328287 M * gudi 2: eth1: mtu 1500 qdisc pfifo_fast qlen 1000 1142328287 M * gudi link/ether 00:14:4f:01:a1:77 brd ff:ff:ff:ff:ff:ff 1142328287 M * gudi inet 217.195.4.227/29 brd 217.195.4.231 scope global secondary eth1 1142328287 M * gudi web:/# 1142328298 M * gudi web:/# traceroute -i eth1 -s 217.195.4.227 -d 217.195.15.252 1142328496 M * eyck it should work... which version are you using? 1142328513 M * gudi vserver 2.0.1 1142328569 M * eyck I don't know... maybe Bertl will get up in few minutes 1142328589 M * gudi ok 1142328591 M * gudi thx 1142329419 Q * pagano Ping timeout: 480 seconds 1142329499 J * shedi ~siggi@tolvudeild-200.lhi.is 1142329584 Q * DataCompBoy Quit: Õðÿï 1142331512 M * coocoon what makes "touch noisy"???? 1142331634 M * SiD3WiNDR eh? :p 1142331702 M * coocoon as command ;-) 1142333238 Q * Aiken Quit: Leaving 1142336059 Q * pzYsTorM Ping timeout: 480 seconds 1142336640 J * terrorgrl ~terrorgrl@86.59.25.121 1142336651 P * terrorgrl 1142336729 J * lilalinux ~plasma@dslb-084-058-229-157.pools.arcor-ip.net 1142336898 J * pagano ~pagano@lappagano.cnaf.infn.it 1142337295 Q * shedi Quit: Leaving 1142337887 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1142337965 J * coocoon ~coocoon@p54A0570A.dip.t-dialin.net 1142338029 M * _chris i have downloaded kernel 2.4.30 and patched the stable version vserver 1142338051 M * _chris when i ran make menuconfig, i didnt see "Linux VServer" 1142338183 M * coocoon have u patched it again 1142338185 M * coocoon ;-) 1142338242 M * _chris nope, earlier, i only tried util-vserver 1142338257 M * _chris without the kernel patch. 1142338266 M * _chris now im ready to patch kernel and try util-vserver 1142338364 Q * Hmmmm Quit: Ex-Chat 1142338364 M * coocoon chris i would patch it again 1142338371 M * _chris oh yea, ive also tried patching the default slack kernel 2.4.31 didnt work either so i downloaded 2.4.30 and tried it, i cant locate the Linux Vserver category anywhere. 1142338523 M * _chris ok 1142338857 M * eyck _chris: why not 2.4.32 ? 1142338898 M * _chris weird still a no go 1142338910 M * _chris amma try downloading .32 1142339076 Q * Smutje Quit: leaving 1142339140 J * Smutje ~Smutje@xdsl-84-44-247-110.netcologne.de 1142339472 Q * FireEgl Ping timeout: 480 seconds 1142339496 M * coocoon how to use the dif from ther have no idea http://debian.marlow.dk/vserver/guest/_slack_10.diff 1142339521 M * coocoon shall i make a bash of it or patch something 1142339606 M * coocoon sorry i mean bin.sh# 1142339700 M * _chris i don't think vserver 2.4.30 is compatible with kernel 2.4.32 1142339715 M * _chris i had a few errors 1142339716 M * _chris Hunk #1 succeeded at 1509 (offset 6 lines). 1142339717 M * _chris Hunk #3 succeeded at 1539 (offset 6 lines). 1142339949 N * Bertl_zZ Bertl 1142339953 M * Bertl morning folks! 1142340002 M * Bertl _chris: did you try if it a) compiles and b) what the testme.sh says? 1142340052 M * _chris i havent compiled it yet, im looking for the Linux Vserver category, should i look for that or do i go ahed with the compilation? 1142340194 J * doener ~doener@i5387DAD9.versanet.de 1142340209 M * Bertl _chris: show me what commands you did to get you 'patched' kernel 1142340243 M * _chris bzip2 -dc ../patch-2.4.30-vs1.2.10.diff.bz2 | patch -p1 1142340328 M * Bertl after you unpacked and cd-ed into the kernel source, I presume 1142340361 M * _chris yes 1142340390 M * Bertl okay, that should be fine 1142340402 M * Bertl in 2.4 there is no 'special' vserver menu 1142340425 M * _chris oh okay 1142340459 M * Bertl the only 'things' you can select are: 1142340480 M * Bertl CONFIG_BLK_DEV_VROOT (and the same as module) 1142340494 M * _chris http://www.howtoforge.com/linux_vserver_debian < i was reading this and assumed 2.4 would be the same 1142340498 M * _chris ok 1142340519 M * Bertl you might also look at http://vserver.13thfloor.at/Experimental/patch-2.4.31-vs1.2.11-rc1.diff 1142340538 M * Bertl it contains a minor fix we did for vs1.2 1142340548 M * _chris okay 1142340551 M * _chris will try that 1142340566 M * Bertl why are you working with a 2.4 kernel if I may ask? 1142340608 M * doener morning :) 1142340618 M * Bertl good morning doener! 1142340656 M * _chris i believe 2.4 is much more stable than 2.6 1142340661 M * _chris :) 1142340692 M * Bertl okay, good argument ... 1142340765 M * coocoon bertl: having some problems with a slackware guest, all times it overrides my fstab and removed my mtab, this happens with 10.0 version, daniel has told me to have a look at the http://debian.marlow.dk/vserver/guest/_slack_10.diff, there also appears an error "chmod 1777 tmp", in the rcM script this also will be checked, with slackware 9.1 all works fine, i do not understand anything 1142340804 M * Bertl the guest overrides your fstab/mtab (inside the guest)? 1142341091 M * eyck Bertl: seee... :) 1142341107 M * eyck Bertl: so, how about 1.2.11 ? 1142341132 M * eyck ;) 1142341147 M * coocoon bertl: yes mtab has been overwritten "proc /proc proc rw 0 0" after reboot -f "none / none rw 0 0" , after starting the vserver df -h in the guest schows "none 977M 16K 927M 1% /" 1142341496 M * Bertl eyck: you know, we are working on it, right ... ? 1142341538 M * eyck we are? 1142341549 M * coocoon bertl: here si the rcM http://pastebin.com/601212 maybe u can help me 1142341574 M * Bertl coocoon: well, I assume it is started via init (plain init style) so you might scan the scripts (probably rc.*something*) run by init? 1142341592 M * Bertl coocoon: and clean them up to not do that? 1142341628 M * coocoon ah ok i will try 1142343615 J * pagano_ ~pagano@lappagano.cnaf.infn.it 1142343711 M * Bertl welcome pagano_! 1142343734 M * pagano_ :) 1142343829 M * Bertl phreak``: ping! 1142343862 M * Bertl doener: regarding the sendfile issues, are you sure that your patch doesn't break anything? 1142343899 Q * pagano Ping timeout: 480 seconds 1142343978 M * doener at least on 2.6.16-rcX i am pretty sure, as long as the breakage is not in vanilla as well 1142344005 M * Bertl okay, sounds good enough to me, I'll add it back for 2.6.16 for now 1142344100 Q * mountie Remote host closed the connection 1142344169 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com 1142344204 M * coocoon what could this be line 331: 20736 Segmentation fault 1142344220 M * coocoon bertl: I think I have found it thanx a lot 1142344221 M * Bertl bad code, in whatever segfaults? 1142344236 M * coocoon is the code false or something in the system 1142344245 M * coocoon sorry 1142344258 M * Bertl well, could be both actually 1142344268 M * coocoon ok thanx :-( 1142344290 M * Bertl where and when do you get that? 1142344466 M * coocoon i got it if i try to start pkgtool in slack10.0 "/sbin/pkgtool: line 331: 20736 Segmentation fault --> " http://pastebin.com/601267 at line 27, it has something to to with chmod 1142344478 M * coocoon 1777 1142344481 M * coocoon at temp 1142344589 M * Bertl very likely a pkgtool 'bug' 1142344606 M * coocoon i think it comes from rcM script http://pastebin.com/601272 there will be made a check 1142344616 M * Bertl check with strace -fF and if that isn't conclusive use gdb 1142344616 M * coocoon # Ensure basic filesystem permissions sanity. 1142344654 M * Bertl well, that comment is wrong, imho 1142344682 M * Bertl it should be # Ensure filesystem permission insecurity :) 1142344702 M * coocoon ah ok 1142344714 M * Bertl but let me check that ... sec 1142344753 M * Bertl nah, sorry, it's adding the sticky bit, which should be fine 1142344788 M * coocoon but is 1777 right 1142344802 M * coocoon can i set anything other or documented it out 1142344805 M * Bertl doesn't that work inside a guest? 1142344808 M * coocoon when it comes from there 1142344820 M * Bertl try it on the guest command line 1142344827 M * coocoon i do not know 1142344831 M * coocoon hm i will 1142344839 M * coocoon but how ;-) 1142344853 M * Bertl you have a running guest? 1142344864 M * coocoon yes 1142344879 M * Bertl then try: vserver enter 1142344896 M * Bertl after that, you do 6. 1142344898 M * Bertl chmod 1777 /tmp /var/tmp 1142344904 M * Bertl (ignore the 6.) 1142344916 M * coocoon ok 1142345011 Q * matta Quit: Download Gaim: http://gaim.sourceforge.net/ 1142345033 M * coocoon i have done this before without /var/tmp, with it it only doesn't work, i also get message /dev/pts/1: Operation not permitted during enter 1142345044 M * coocoon -only +also 1142345075 M * Bertl that is okay 1142345080 M * coocoon good 1142345086 M * coocoon i am happy 1142345103 M * Bertl but the /var/tmp part seems not okay 1142345130 M * Bertl so chmod 1777 /tmp /var/tmp doesn't work, but 1142345135 M * Bertl chmod 1777 /tmp does? 1142345144 M * coocoon no also not 1142345153 M * Bertl so none of them work? 1142345157 M * coocoon yes 1142345174 M * Bertl very interesting ... 1142345246 M * Bertl you get something like this: 1142345247 M * Bertl chmod: changing permissions of `/tmp': Operation not permitted 1142345295 M * coocoon no sorry, I thought u mean another function :-) 1142345309 M * coocoon it works fine, but my problem isn't be solved 1142345336 M * Bertl well, I didn't say it comes from there, you reported that not working, right? :) 1142345358 M * coocoon u r right 1142345382 M * coocoon i though about it to tell yes 1142345384 M * Bertl and both commands you uplaoded seem to work fine 1142345407 M * Bertl and if I take it, they even work fine for you, right? 1142345417 M * coocoon no new are good news? 1142345425 M * coocoon which commands 1142345574 J * DataCompBoy ~datacompb@217.8.236.1 1142345631 M * Bertl welcome DataCompBoy! 1142345646 M * Bertl coocoon: http://pastebin.com/601272 1142345711 M * DataCompBoy Hello, All! 1142345790 M * DataCompBoy Can someone describe me, how to create new vserver that will have only hardlinks to my base system, with copy-on-write ? 1142345808 M * DataCompBoy kernel-patch 2.1.0 installed on linux-2.6.15-i686-smp 1142345818 M * Bertl usually you don#t want to do that for several reasons 1142345834 M * Bertl what you might want to do is: 1142345845 M * coocoon bertl: this was part of rcM i have uploaded 1142345855 M * Bertl DataCompBoy: - make a copy of your base system and clean that up 1142345877 M * Bertl DataCompBoy: - use that 'template' to create new unified guests from 1142345927 M * DataCompBoy so, I need to create template vserver, and then clone it when needed? 1142345930 M * Bertl now you might ask, why not use the basesystem, if you have CoW 1142345930 M * Bertl here are some (good) reasons: 1142345936 M * DataCompBoy Any way to centralized upgrade of them at-all ? 1142345950 M * coocoon i have seen a skeleton tool which copied all files from the system right 1142345957 M * Bertl DataCompBoy: okay, I postpone the reasons :) 1142345962 M * coocoon or all progs 1142345975 M * Bertl DataCompBoy: this depends on the distro you are using 1142346002 M * Bertl DataCompBoy: for example rpm based distros can be updated with vrpm (batch update to several/all guests) 1142346024 M * Bertl similar tools exist for yum and apt-rpm IIRC 1142346076 M * DataCompBoy host: debain-etch, guests: debian-etch too:) 1142346185 M * DataCompBoy pffr, where to see documentation onto util-vserver tools? at /usr/share/doc/util-vserver -- nothing; at manpages -- nothing... from linux-vserver.org also can't find anything... 1142346233 M * Bertl http://linux-vserver.org/alpha+util-vserver 1142346280 J * shedi ~siggi@tolvudeild-200.lhi.is 1142346313 M * Bertl btw, there is documentation in the util-vserver package, but I don't know if debian installs it 1142346343 M * Bertl welcome shedi! 1142346345 M * DataCompBoy http://dev.call2ru.com/doc/util-vserver -- that's all that it installs 1142346373 M * shedi hello Bertl 1142346420 M * Bertl DataCompBoy: well, I'd say, time to send a message to the debian maintainers :) 1142346433 M * DataCompBoy ok:) 1142346463 M * Bertl ls util-vserver-0.30.210-shiny/man/ 1142346463 M * Bertl Makefile-files chcontext.8 rebootmgr.8 vps.8 vserver-copy.8 vserver.8 1142346466 M * Bertl chbind.8 distrib-info.8 reducecap.8 vpstree.8 vserver-stat.8 vtop.8 1142346479 M * Bertl ls util-vserver-0.30.210-shiny/doc/ 1142346479 M * Bertl Makefile-files compatibility.html configuration.css configuration.xml 1142346482 M * Bertl compatibility-xhtml.xsl compatibility.xml configuration.dtd configuration.xsd 1142346485 M * Bertl compatibility.dtd configuration-xhtml.xsl configuration.html intro.txt 1142346491 Q * gudi Quit: Verlassend 1142346532 M * Bertl and IIRC, there is more docu which is built on the fly 1142346592 M * DataCompBoy any way to create vserver that will live on virtual iface? 1142346705 M * Bertl what's that? virtual iface? 1142346750 M * DataCompBoy virtual network iface:) 1142346778 J * pzYsTorM schak@dslc-082-082-169-036.pools.arcor-ip.net 1142346791 M * DataCompBoy pfrr... also it will good to have "ip rule xid AAA ..." matching:) 1142346829 M * Bertl wb pzYsTorM! 1142346837 J * mmouse ~mmouse@office.haefft.de 1142346850 M * mmouse hi 1142346866 M * Bertl DataCompBoy: why would you want to have a guest on a virtual interface? 1142346873 M * Bertl hey mmouse! 1142346902 M * DataCompBoy I will run every vserver on its own virtual iface, and manage routing between them with iproute2 + iptables 1142346991 M * Bertl iproute2 and iptables do not work on/require 'virtual' interfaces 1142347004 M * Bertl they are based on IPs 1142347036 M * Bertl but you can put each guest on a (possibly separate) dummy interface 1142347080 M * DataCompBoy what you call with "dummy interface" ? 1142347219 M * Bertl dummy0 1142347277 M * mmouse anybody knows why "traceroute" results in "raw socket: Operation not permitted"? raw_icmp capability should be set by default?! 1142347303 M * DataCompBoy Bertl: ooo! thanks, will see that 1142347324 M * mnemoc mmouse: you need an special and evil capability to do traceroute 1142347352 Q * shedi Quit: Leaving 1142347412 A * mmouse likes doing evil things ;-) 1142347430 M * DataCompBoy ooops... there no distribution "etch" for debootstrap creation?! 1142347455 M * coocoon there is if u use the latest debootstrap 1142347462 M * coocoon rpm 1142347466 M * coocoon or deb 1142347468 M * DataCompBoy of course latest -- i'm on debian-etch :) 1142347468 M * coocoon sorry 1142347480 M * DataCompBoy vserver template build -m debootstrap --hostname template --interface dummy0:192.168.1.0/24 -- -d testing -- --exclude=$REMOVE_PACKAGES 1142347484 M * coocoon look in the pool 1142347493 M * DataCompBoy root@orange:~# ls /var/lib/vservers/template/ 1142347494 M * DataCompBoy dev etc proc 1142347503 M * DataCompBoy nothing more than dev etc proc 1142347504 M * DataCompBoy :( 1142347510 M * Bertl try -d etch (and if necessary, specify the url) 1142347518 M * DataCompBoy tried. 1142347520 M * mmouse mnemoc: ok, which caps to set? 1142347547 M * DataCompBoy and etch and testing provide identical results... 1142347579 M * Bertl what url did you try? 1142347581 M * mnemoc mmouse: CAP_NET_RAW, but it's *evil* 1142347589 M * coocoon vserver DebianEtch build -m debootstrap -- -d etch -m ftp://ftp.at.debian.org/debian/ ... blablah 1142347611 M * DataCompBoy coocoon: moment, i 1142347624 M * DataCompBoy 'll try... just thinks it have looked at my /etc/apt/sources.list:) 1142347627 M * coocoon i i mean and so on 1142347654 M * DataCompBoy how to remove vserver ? rm -rf /etc/vservers/$vs /vservers/$vs ? 1142347671 M * Bertl yep, that's basically a good approach 1142347683 M * Bertl make sure to stop it before, though 1142347685 M * DataCompBoy why not vserver-remove ? :) 1142347713 M * Bertl you have to ask Enrico (the guy maintaining the tools) for that, I've posted a request about half a year ago :) 1142347728 M * mmouse mnemoc: i see. looks like it's that evil that it even isn't on the Caps+And+Flags wiki page 1142347732 M * DataCompBoy ooo! ok, i'll make bash script for me:) 1142347750 M * Bertl daniel_hozac: btw, this and the clone stuff would be really useful, do you see any chance to add this sooner or later? 1142347789 M * Bertl anyway ... have to leave now ... so have fun and good luck 1142347795 M * DataCompBoy coocoon: the same effect. debootstrap is not runned. 1142347798 N * Bertl Bertl_oO 1142347836 M * Bertl_oO DataCompBoy, coocoon: hint: double check the debootstrap version 1142347888 M * DataCompBoy Version: 0.3.3 1142347889 M * coocoon this is the latest debootstrap_0.3.3_all.deb from here ftp://ftp.at.debian.org/debian/pool/main/d/debootstrap/ 1142347889 M * DataCompBoy latest 1142347907 M * DataCompBoy coocoon: I have already installed debootsrap 0.3.3. 1142347934 M * coocoon must say i have also downloaded the tar file and with make and mak install after that it wors 1142347958 M * coocoon i do not know why but it works after that 1142347972 M * coocoon ftp://ftp.at.debian.org/debian/pool/main/d/debootstrap/debootstrap_0.3.3.tar.gz 1142348023 M * DataCompBoy may be something need to be put into /etc/vservers/.distributions/ ? 1142348080 M * coocoon no it will installed in the system 1142348087 M * coocoon +be 1142348115 M * DataCompBoy coocoon: I have installed version of debootstrap. I have used it to build crootjails already several times 1142348209 M * coocoon DataCompBoy: i have no idea I have done it also first install the deb package but it only works after i also compiled the tar file, sorry i do not know 1142348256 M * DataCompBoy what "which debootstrap" say to you? 1142348290 M * coocoon hm i only use it for installing vserver, have tried several things after that it works 1142348294 M * coocoon ;-) 1142348332 M * coocoon have u tried it 1142348579 Q * pagano_ Ping timeout: 480 seconds 1142349059 Q * nox Read error: Connection reset by peer 1142349319 J * nox ~nox@noxlux.de 1142349512 M * DataCompBoy coocoon: http://vserver.13thfloor.at/Stuff/newvserver have you seen that? 1142349649 M * DataCompBoy after changeing "grep '^s_context:[^0-9]0$';" to "grep '^VxID:[^0-9]0$'; 1142349654 M * DataCompBoy it start working :) 1142349733 J * pagano ~pagano@lappagano.cnaf.infn.it 1142349746 M * coocoon have never used that 1142349777 M * DataCompBoy i'm waiting, hope template will be created fine:) 1142350214 M * coocoon bertl: i am asking myself why pkgtool failled at slackware10.X version, it doesn't depend on my guest installation, could it be that the kernel is too actual, because of slackware works with 2.4 1142350269 M * coocoon my kernel is 2.6 1142350315 M * coocoon because of the segment error 1142350999 J * Smutje_ ~Smutje@xdsl-87-78-63-240.netcologne.de 1142351104 Q * Smutje Ping timeout: 480 seconds 1142351104 N * Smutje_ Smutje 1142351140 J * matta ~matta@c-68-32-239-173.hsd1.pa.comcast.net 1142351201 M * daniel_hozac Bertl_oO: the guy who posted the delete patch to the mailing list also submitted it in savannah. i've incorporated it into my packages. 1142351265 M * daniel_hozac Bertl_oO: and clone is on my list :) 1142351312 J * JimmyGulp ~james@ns0.esagroup.co.uk 1142351453 M * DataCompBoy ugh.... I have installed it -- but network is configured strangely 1142351464 M * DataCompBoy Should I set up ipaddr at host too for guest? 1142351482 M * daniel_hozac unless you've touched nodev in the interface directory, util-vserver will set it up. 1142351502 M * DataCompBoy not touched. but won't work. 1142351507 M * DataCompBoy at host I have: 1142351508 M * daniel_hozac meaning? 1142351515 M * DataCompBoy moment... 1142351665 M * DataCompBoy http://pastebin.com/601431 1142351732 M * JimmyGulp howdy, I've got vserver running (2.6 kernel, util-vserver 0.20.210), with a few legacy vservers running. I'm wondering how I get vserver-stat to display the name of each vserver running. At the moment, they're blank. 1142351821 M * coocoon bertl: ok it depends on my installation and on the others 1142351922 M * mmouse datacompboy: add a file "name" with some name in it to the interface dir, otherwise ifconfig won't display the interface 1142351960 M * DataCompBoy ifconfig show interface, but doesn't show its IP 1142351963 M * mmouse datacompboy: and don't use .0 as ip address, its the network ip 1142351977 M * DataCompBoy oooh! you right, my miss :) 1142352008 M * mmouse ifconfig only displays the parent interface not the virtual one the guest uses 1142352102 J * wam ~wigwam@proxy1.msh.de 1142352198 M * wam Hi! I'm installing vserver 202 on a new server on kernel 2.6.15.6 and I'm having a little problem that I didn't have on my other server: When I call vserver-stat or vserver name start, I get "WARNING: can not access /proc/uptime. Usually, this is caused by procfs-security." Every user can call "cat /proc/uptime". I didn't find the answer in the faq... What could this be? 1142352218 J * shedi ~siggi@tolvudeild-200.lhi.is 1142352222 M * wam does vserver need write-permission to uptime? 1142352235 M * JimmyGulp wam, search for vprocunhide in the FAQ 1142352261 M * wam JimmyGulp: I thought I'd need this to unhide the proc inside the vservers? Or IS this my bug? 1142352322 M * wam damnit. thanks! 1142352330 M * JimmyGulp :) 1142352814 J * lilalinux_ ~plasma@dslb-084-058-217-176.pools.arcor-ip.net 1142352941 P * wam 1142353114 M * DataCompBoy umm... how to set up an default route? 1142353154 M * DataCompBoy root@template:/# route add default gw 192.168.1.250 1142353155 M * DataCompBoy SIOCADDRT: Operation not permitted 1142353177 M * mmouse datacompboy: set it up on your host. it's visible to the guest. at least for me ;-) 1142353180 M * DataCompBoy and at interfaces/* doesn't see how to set it up on host 1142353190 Q * lilalinux Ping timeout: 480 seconds 1142353191 M * DataCompBoy I need different gateway for guest:) 1142353245 M * DataCompBoy agh, found in memory. 1142354133 J * mkhl ~mkhl@200-153-153-148.dsl.telesp.net.br 1142354229 Q * _chris Remote host closed the connection 1142354416 M * DataCompBoy people, for look of iptables on host, packets generated by vhosts is OUTPUT or FORWARD ? 1142354551 J * Viper0482 ~Viper0482@p54976824.dip.t-dialin.net 1142354715 Q * shedi Quit: Leaving 1142354967 J * ntrs ~ntrs@68-188-37-15.dhcp.stls.mo.charter.com 1142355077 M * DataCompBoy how I must set up virtual machine, so it will everytime run machine in chbind ? 1142355687 M * daniel_hozac what? 1142355996 J * lilalinux__ ~plasma@dslb-084-058-195-114.pools.arcor-ip.net 1142356141 M * DataCompBoy ok, i have fixed network config. 1142356155 M * DataCompBoy one question left: how to select what vservers should start on reboot? 1142356233 M * daniel_hozac echo default > /etc/vservers//apps/init/mark 1142356245 M * daniel_hozac and make sure the vservers-default initscript is set to run. 1142356297 M * phreak`` Bertl_oO: pong 1142356304 M * DataCompBoy daniel_hozac: many thanks! 1142356375 Q * lilalinux_ Ping timeout: 480 seconds 1142356662 Q * pagano Quit: Leaving 1142356786 M * DataCompBoy hmm... 1142356794 M * DataCompBoy vserver-copy work with only legacy API? 1142356803 M * DataCompBoy E: vserver-copy: Vserver file "/etc/vservers/template.conf" does not exist 1142356954 M * SiD3WiNDR yeah 1142356973 M * DataCompBoy so, how to clone vserver then? 1142356987 M * DataCompBoy vserver ... build -m copy ? 1142357452 M * DataCompBoy root@orange:/etc/vservers# vserver br build -m copy 1142357454 M * DataCompBoy ld.copy: No such file or directory 1142357459 M * DataCompBoy /usr/lib/util-vserver/vserver-build: line 130: /usr/lib/util-vserver/vserver-bui 1142357472 M * DataCompBoy looks like copy method only documented -- not shipped 1142357633 M * DataCompBoy o, found solution: 1142357635 M * DataCompBoy http://www.micropp.se/vserver/ 1142358158 J * lilalinux_ ~plasma@dslb-084-058-224-045.pools.arcor-ip.net 1142358535 Q * lilalinux__ Ping timeout: 480 seconds 1142358871 M * DataCompBoy -1 1142358872 Q * DataCompBoy Quit: Õðÿï 1142359412 Q * ntrs Read error: Connection reset by peer 1142360409 J * cmantito ~gphreak@c-68-37-52-88.hsd1.nj.comcast.net 1142360814 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1142361011 J * lilalinux__ ~plasma@dslb-084-058-197-058.pools.arcor-ip.net 1142361039 J * alexx ~alexx@proxy.ikse.net 1142361390 Q * lilalinux_ Ping timeout: 480 seconds 1142362125 Q * gerrit Ping timeout: 480 seconds 1142362803 J * shedi ~siggi@inferno.lhi.is 1142363655 J * derjohn ~derjohn@80.69.37.19 1142363715 J * lilalinux_ ~plasma@dslb-084-058-208-084.pools.arcor-ip.net 1142364090 Q * lilalinux__ Ping timeout: 480 seconds 1142364256 Q * lilalinux_ Remote host closed the connection 1142364644 J * gerrit ~gerrit@bi01p1.co.us.ibm.com 1142364794 J * liquid3649_ ~Viper0482@p54976824.dip.t-dialin.net 1142364795 Q * liquid3649_ Quit: 1142364935 J * doener_ ~doener@i5387E549.versanet.de 1142365334 Q * doener Ping timeout: 480 seconds 1142366481 Q * Hollow Remote host closed the connection 1142367222 M * mmouse is there an issue with setting an locks rlimit for a guest? 1142367254 M * mmouse or an maximum value or something like that? 1142367900 J * Hollow ~hollow@home.xnull.de 1142368211 Q * matta Quit: Download Gaim: http://gaim.sourceforge.net/ 1142368235 J * matta ~matta@c-68-32-239-173.hsd1.pa.comcast.net 1142368319 Q * Viper0482 Remote host closed the connection 1142368337 M * daniel_hozac mmouse: shouldn't be. why? 1142368485 M * mmouse daniel: cat 500 > /etc/vservers/nnn/rlimit/locks 1142368500 M * daniel_hozac s/cat/echo/, right? 1142368520 M * mmouse results in "vc_set_rlimit(): no such file or directory" at vserver startup 1142368524 M * mmouse daniel: yes, sorry 1142368536 M * mmouse other rlimits are working 1142368617 M * daniel_hozac hmm. 1142368629 M * daniel_hozac locks should be supported. 1142368663 M * daniel_hozac what kernel? 1142368696 M * daniel_hozac how odd. 1142368705 M * mmouse 2.6.15.6-vs2.0.2-rc10 1142368717 M * daniel_hozac oh sorry, no. 1142368723 M * daniel_hozac LOCKS isn't valid either. 1142368767 M * mmouse daniel: um? doesn't work with this kernel? 1142368803 M * daniel_hozac doesn't work with any kernel. 1142368810 M * mmouse ok :) 1142368822 M * daniel_hozac guess you'll have to ask Bertl_oO why that isn't valid :) 1142368905 M * mmouse doesn't really matter. if it doesn't work, I won't set this file. 1142368913 M * mmouse thanks 1142368920 M * daniel_hozac i believe it should work, it's just listed as a valid limit. 1142368936 M * daniel_hozac +isn't, obviously. 1142369198 M * tokkee http://tokkee.org/cgi-bin/collection.cgi/vserver-49152 ... this is a sample output of my collectd-vserver-plugin that was announced on the mailing list a couple of days ago ;-) 1142369255 M * daniel_hozac you should use static contexts :) 1142369282 M * daniel_hozac that's cool though. 1142369288 J * FireEgl Atlantica@Atlantica.Tcldrop.Com 1142369307 M * tokkee daniel_hozac: What does static context mean? 1142369375 M * daniel_hozac an xid that will be the same over reboots. 1142369391 M * daniel_hozac i.e. between 2 and 49151. 1142369452 M * tokkee How do I set it? 1142369531 M * daniel_hozac echo > /etc/vservers//context 1142369556 M * tokkee Alright... I probably have to reboot for the change to take effect?! 1142369574 M * daniel_hozac you probably should've stopped it before setting it :) 1142369627 M * tokkee Ok... thx ;-) 1142370083 M * mmouse tokkee: thats pretty cool stuff 1142370100 M * tokkee mmouse: thx :-) 1142370180 J * Aiken ~james@tooax7-134.dialup.optusnet.com.au 1142370202 M * tokkee If anybody has an idea what else should be collected please tell me ;-) 1142370309 M * mmouse tokkee: do you have a script to generate html+png from the rrd database, or is something like this included in collectd? 1142370428 M * tokkee mmouse: The collection.cgi script is included in collectd 1142370451 M * tokkee mmouse: It creates the html+png for all collected data... 1142370624 M * mmouse tokkee: thats fine. I'll try it myself... 1142370762 J * coocoon ~coocoon@p54A06472.dip.t-dialin.net 1142370771 M * coocoon morning 1142371365 J * AnGeL4U asdfghjklk@pc09.romarnia.roman.rdsnet.ro 1142371369 M * AnGeL4U hellow 1142371377 M * AnGeL4U Bertl_oO are u there ? 1142371386 M * daniel_hozac oO == out of office ;) 1142371395 M * AnGeL4U i see 1142371401 M * AnGeL4U are u in he`s office? 1142371416 M * daniel_hozac no. 1142371432 M * AnGeL4U are u in linux-vserver staff ? 1142371441 M * AnGeL4U need me best infos :P 1142371442 M * daniel_hozac AFAIK, there is no "staff". 1142371450 M * AnGeL4U i know 1142371454 M * AnGeL4U here is one project 1142371474 M * AnGeL4U what version of vserver are the best for 2.4.32 kernel? 1142371480 M * AnGeL4U and if can i put grsec too 1142371623 M * daniel_hozac i guess 1.2.11-rc1 would be your best bet. 1142371633 M * daniel_hozac probably won't apply cleanly to 2.4.32 though. 1142371843 M * daniel_hozac and as for grsec combos, you'll have to check the wiki. 1142371975 Q * mmouse Quit: connection reset by beer 1142371986 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1142372004 M * AnGeL4U wiki ? 1142372007 M * AnGeL4U what`s that 1142372014 M * AnGeL4U i`m fool :p explain me a bit 1142372152 M * daniel_hozac www.linux-vserver.org 1142372364 J * coocoon ~coocoon@p54A06472.dip.t-dialin.net 1142372564 J * spd1snd ~spd1snd@68-232-131-226.chvlva.adelphia.net 1142372569 Q * anonc Ping timeout: 480 seconds 1142373299 Q * AnGeL4U Quit: 1142373336 J * f_ ~f_@83-215-237-1.seek.stat.salzburg-online.at 1142373353 Q * f_ Quit: 1142373514 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1142374084 J * coocoon ~coocoon@p54A06472.dip.t-dialin.net 1142374432 Q * spd1snd Quit: spd1snd 1142374711 N * Bertl_oO Bertl 1142374719 M * Bertl evening folks! 1142374734 M * coocoon hello 1142374764 M * coocoon bertl: i have nade it sorry for nerving have found a template with right scripts 1142374771 M * Bertl hey coocoon! 1142374936 M * harry q; should i put my grsec + vserver patches on the wiki? 1142374942 M * daniel_hozac yes. 1142374951 M * harry any preferred places? 1142374966 M * harry (/me thinks it's bloody damn hard to find the docs you are looking for on the wiki atm :s) 1142374968 M * daniel_hozac the place with all the other grsec+vserver merges would seem like a good candidate. 1142374977 A * harry looking for them as we speak :) 1142374985 M * daniel_hozac tools and patches, IIRC. 1142375012 M * daniel_hozac maybe you should move all the grsec links to a separate block. 1142375035 A * harry not a good website maker, so i'm kinda... scared to do that :) 1142375076 M * daniel_hozac just look at the other paragraphs on the page. 1142375226 M * harry there we go :) 1142375232 M * Bertl harry: and make use of the preview option 1142375261 M * harry hahaaaaaaa... i'm on the internet :p 1142375281 M * harry just added myself at the bottom of the list :) 1142375356 M * harry When I was ten, to reboot the computer you held it upside down over your head and shook it. Oh wait, that was 'Etch-a-Sketch' . Holy crap, how times have changed! 1142375360 M * harry haha :) 1142375385 M * Bertl you are 12 now? :) 1142375415 M * harry hehje 1142375434 M * harry you could double that, and then you're close :) 1142375622 M * Aiken Bertl in test.sh with make -j99, that had the poor machine using 1 1/3 gig of swap 1142375753 M * harry Aiken: try make -j 1142375753 M * harry :p 1142375801 M * Bertl Aiken: hmm, okay, again not what we were looking for 1142375803 M * Aiken I am using make -j2 to get a sensible value first 1142375818 M * Bertl yeah, that's much better, how did the fork tests turn out? 1142375836 M * Aiken http://pastebin.com/602540 1142376098 M * Bertl yeah, funny, same as on x86_64, inside the guest the test is better :) 1142376117 Q * matta Read error: Connection reset by peer 1142376132 M * Aiken last night I managed to get test.sh to run with make -j9, in the guest was about 3 minutes slower 1142376141 M * Aiken host time was around 10 minute 1142376146 M * Aiken also used make allnoconfig 1142376164 M * Bertl daniel_hozac: did you get my message regarding tools and clone/remove? 1142376165 M * Aiken have just started that run a gain 1142376191 M * Bertl Aiken: 3 mintues by 10 minute total, that'd be 30% 1142376221 M * Aiken I'll have definate figures in < 1/2 hour 1142376231 M * Bertl phreak``: ping again? :) 1142376297 J * matta ~matta@c-68-32-239-173.hsd1.pa.comcast.net 1142378729 M * Aiken Bertl in this run the kernel build took 2 1/2 minutes longer as a guest and used more cpu http://pastebin.com/602617 1142379305 M * Bertl that's quite a difference 1142379397 M * Bertl does this make use of swap space, I assume so? 1142379435 M * Aiken no swap 1142379474 M * Bertl impressive ... we must have quite some unaligned access there, I'd say 1142379522 Q * doener_ Quit: leaving 1142379558 M * Aiken no syslog messages about anything unaligned 1142379560 M * Bertl I will add some magic to track this down, do you think you could do a few tests tomorrow or on thursday with that? 1142379562 M * daniel_hozac Bertl: didn't you get my responses? ;) 1142379570 M * Aiken I can do that 1142379572 M * Bertl daniel_hozac: probably not :) 1142379593 M * Bertl daniel_hozac: but read it now, tx! 1142379615 M * Bertl Aiken: do you know if oprofile works on alpha? 1142379632 M * daniel_hozac don't know if you saw it, but why is locks not a valid rlimit? 1142379650 M * Aiken would not have a clue 1142379673 M * Bertl daniel_hozac: IIRC, we have no limit checks there yet, or if we do, we just haven't updated the valid check :) 1142379700 M * Aiken having a look 1142379705 M * daniel_hozac $ grep -c vx_locks_avail fs/locks.c 1142379705 M * daniel_hozac 1 1142379721 M * Bertl Aiken: hmm, you could compile a kernel (one of the older ones, which still work) with oprofile enabled (in the kernel) 1142379726 M * Aiken the docs say alpha 1142379736 M * Bertl daniel_hozac: then we should add it to the valid limits, no? 1142379764 M * Bertl daniel_hozac: probably others could benefit from a short check too, maybe something automated? 1142379769 M * daniel_hozac IMHO, yeah. 1142379809 M * daniel_hozac a check that all the limits are used? 1142379815 M * Bertl Aiken: okay, so once compiled with oprofile, just get the oprofile rpm/deb/tar and contact me ... 1142379857 M * Bertl daniel_hozac: something which basically tells us where we have accounting (cres) and where we have limits (avail) and which ones are listed as valid 1142379883 M * Bertl all those with limits should be listed as valid 1142379929 M * daniel_hozac ok, i'll see what i can do. 1142380066 M * Bertl nothing terribly fancy, just a quick hack 1142380600 M * Bertl hmm, seems 2.6.16-rc6 doesn't even boot in QEMU :) 1142380610 M * daniel_hozac RLIMIT_MSGQUEUE seems to be implemented too. 1142380642 M * Bertl arch/i386/kernel/setup.c:1603: warning: implicit declaration of function `check_acpi_pci' 1142380649 M * Bertl maybe somebody forgot something :) 1142380658 Q * matta Quit: Download Gaim: http://gaim.sourceforge.net/ 1142380662 M * daniel_hozac lol, looks like it.