1119836830 J * _mcp ~hightower@wolk-project.de 1119836830 Q * mcp Read error: Connection reset by peer 1119836834 N * _mcp mcp 1119840788 M * cods what's the point of VXF_STATE_SETUP and VXF_STATE_INIT ? (I'm trying to understand --endsetup and --initpid options of vcontext) 1119842397 J * eXplasm explasm@p549F799C.dip.t-dialin.net 1119842675 J * Aiken_ ~james@tooax6-039.dialup.optusnet.com.au 1119842990 Q * Aiken Ping timeout: 480 seconds 1119842994 Q * neofutur Ping timeout: 480 seconds 1119843308 Q * eXplasm Quit: Verlassend 1119843368 J * neofutur ~neofutur@neofutur.net 1119843787 J * Doener` ~doener@p54874294.dip.t-dialin.net 1119844223 Q * Doener_zZz Ping timeout: 480 seconds 1119844451 Q * sannes Quit: aiaiaia 1119845375 J * meebey_ meebey@booster.qnetp.net 1119845375 Q * meebey Read error: Connection reset by peer 1119845459 Q * monrad Ping timeout: 480 seconds 1119850865 Q * _are_ Quit: bbl 1119851312 Q * Aiken_ Ping timeout: 480 seconds 1119851880 J * Aiken ~james@tooax6-039.dialup.optusnet.com.au 1119853855 M * cods Isn't "vcontext --create --xid 42 vattribute --secure /bin/sh" supposed to remove all capabilities and forbid me to ping, chroot,.. ? 1119854993 M * Doener` it removes all 'dangerous' capabilities 1119855022 M * Doener` chroot is still allowed, and ping is allowed through a special context capability 1119855081 N * Doener` Doener 1119855107 Q * mcp Read error: Connection reset by peer 1119855122 J * mcp ~hightower@wolk-project.de 1119856709 J * sukria ~sukria@sargon.lncsa.com 1119857289 J * erwan_ho ~erwan@konilope.dyndns.org 1119857347 Q * erwan_ho Remote host closed the connection 1119857773 Q * Aiken Quit: Leaving 1119859788 J * prae ~prae@ezoffice.mandriva.com 1119860339 J * virtuoso ~s0t0na@80.253.205.251 1119860515 M * virtuoso hey 1119860571 Q * virtuoso Quit: 1119860621 J * virtuoso ~s0t0na@80.253.205.251 1119860768 Q * virtuoso Quit: 1119860885 J * virtuoso ~s0t0na@80.253.205.251 1119861299 Q * Eyck Read error: Connection reset by peer 1119861828 Q * id Ping timeout: 480 seconds 1119862128 J * erwan_taf ~erwan@81.80.43.67 1119862373 J * id ~id@relax-media.softwarezentrum.de 1119862597 M * Doener morning virtuoso 1119862994 J * tbenita ~tbenita@tbenita.net1.nerim.net 1119863005 M * tbenita hi all 1119863066 M * tbenita on a gentoo vserver 2.6, I can't get my iptables forward trafic from my vserver to an external ip 1119863118 M * TheSeer that's not a forward but an output rule 1119863129 M * tbenita I should add a default gateway into the vserver that is different of the default gateway of the MAIN server 1119863161 M * TheSeer hmm?? 1119863165 M * TheSeer that sounds bad.. 1119863208 M * tbenita mmm... In fact I wanted to rewrite the same rules that I would have into a private network 1119863227 M * tbenita but it looks like it's not the good way to do it 1119863263 M * tbenita the vserver has only a private ip 1119863282 M * tbenita the MAIN server has both private and public ip 1119863292 M * tbenita some vservers have another public ip 1119863347 M * TheSeer okay.. so what you need is dnat/snat 1119863400 M * TheSeer here's what you do 1119863407 M * TheSeer ethX: public ip 1119863415 M * TheSeer dummy0: private ip MAIN 1119863432 M * TheSeer you bind the alias ip for the vserver on dummy0 1119863459 M * TheSeer you then define iptable rules to dnat/snat stuff from/to dummy0 1119863472 M * TheSeer at least if you want to run services within that vserver 1119863478 M * TheSeer otherwise you just use nat on dummy0 1119863567 M * TheSeer understood? :> 1119863594 M * tbenita yep ;) 1119863598 A * tbenita is re-trying 1119864441 M * tbenita from the MAIN server, I can't ping the external ip from dummy0 1119864467 M * TheSeer ? 1119864469 M * tbenita ping -I dummy0 ext_ip lose all packets 1119864530 M * tbenita i added : 1119864534 M * tbenita iptables -t filter -A FORWARD -i dummy0 -j ACCEPT 1119864550 M * tbenita iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE 1119864578 M * tbenita and echo 1 > /proc/sys/net/ipv4/ip_forward 1119864612 M * TheSeer if you ping "from" dummy0, shouldn't be the filter be -o dummy0 ? 1119864672 M * Doener tbenita: the forward chain is not used, as the traffic originates from the same host 1119864691 M * Doener you need SNAT/DNAT rules in OUTPUT and POSTROUTING 1119864790 M * Doener ehr, s/OUTPUT/PREROUTING/ 1119864855 M * Doener Jun 05 01:29:57 Bertl iptables -t nat -A PREROUTING -d 172.16.1.1 -j DNAT --to 172.16.0.1 1119864855 M * Doener Jun 05 01:29:57 Bertl iptables -t nat -A POSTROUTING -s 172.16.0.1 -j SNAT --to 172.16.1.1 1119865099 J * rt ~rt@195.246.161.1 1119865103 M * rt hi 1119865118 M * Doener welcome rt 1119865135 A * Doener .oO( rs, rt, so when does ru join? ;) 1119865177 M * rt i am going to migrate my vservers to 1.9, is there any example-configuration available? 1119865190 M * tbenita Doener, in this case you map 2 ips together ? 1119865230 M * Doener rt: with the alpha tools you can easily get one 1119865266 M * Doener check "vserver - build --help" for the stuff you can configure, then build a skeleton vserver (configuration + minimal fs) 1119865308 M * Doener just make sure that you don't use any name that is in-use, for testing use sth. like "vserver conftest build -m skeleton *" 1119865332 M * Doener you _can_ still use your old config, but you won't get the new features then ;) 1119865350 M * Doener also, you probably want 2.0-rc4, not 1.9.x ;) 1119865385 M * rt ok, i ll have a look at the alpha tools... i don t think, that i need the new features ;-) 1119865402 M * Doener then you can keep using your old tools + config 1119865405 M * rt hmm, rc4 sounds nice 1119865412 M * Doener hmm... or did you mean a kernel config? 1119865434 M * rt no, the new "vserver.conf" 1119865443 M * Doener ok 1119865492 M * Doener IIRC some folks have been talking about some conversion tool on the mailing list, a few weeks(?) ago, so checking the archives may be worth a try 1119865521 M * rt ok, i ll check the list 1119865532 M * rt please stand by ;-) 1119865546 M * Doener tbenita: i just know the basics ;) but it looks like... maybe postrouting is enough if you only want outgoing connections... 1119865566 M * tbenita ok thx Doener and TheSeer 1119865592 M * Doener you're welcome 1119866611 M * rt Doener, if i want to create a minimal debian vserver (a zunzuncito ;-) the syntax would be: 1119866644 M * rt vserver zunzuncito build -m debootstrap --hostname zunzuncito --netdev eth0 --interface 192.168.1.237/24 ? 1119866858 M * Doener you should add a static context id (dynamic ones cause problems with some features, f.e. with disk limits) and at the end, add: -- -d {sarge,etch,sid} 1119866881 M * Doener debootstrap needs to know which distro you want, and the auto-detection of the current system is broken IIRC 1119866960 M * Doener tbenita: pinging a local address _always_ goes via lo, no matter to what interface the address is assigned 1119866996 M * maharaja Doener: is v2 compatible to 1.9.5 ? 1119867015 M * maharaja so: can i upgrade without any fear of breaking things? :) 1119867022 M * maharaja like 1.3 -> 1.9 did 1119867056 M * id maharaja, should be no problem 1119867493 M * maharaja k 1119867494 M * maharaja thank you 1119868239 Q * albeiro Ping timeout: 480 seconds 1119868277 J * albeiro ~albeiro@graffias.estrefa.pl 1119868396 Q * sukria Remote host closed the connection 1119868913 P * erwan_taf Leaving 1119868953 J * _mcp ~hightower@wolk-project.de 1119869082 Q * tbenita uranium.oftc.net unununium.oftc.net 1119869082 Q * prae uranium.oftc.net unununium.oftc.net 1119869082 Q * mcp uranium.oftc.net unununium.oftc.net 1119869082 Q * tchan uranium.oftc.net unununium.oftc.net 1119869082 Q * stephenM uranium.oftc.net unununium.oftc.net 1119869082 Q * case uranium.oftc.net unununium.oftc.net 1119869082 Q * hiaslbot uranium.oftc.net unununium.oftc.net 1119869082 Q * xbing uranium.oftc.net unununium.oftc.net 1119869082 Q * flock uranium.oftc.net unununium.oftc.net 1119869082 Q * maharaja uranium.oftc.net unununium.oftc.net 1119869082 Q * dsoul uranium.oftc.net unununium.oftc.net 1119869082 Q * TheSeer uranium.oftc.net unununium.oftc.net 1119869082 Q * Vudumen uranium.oftc.net unununium.oftc.net 1119869082 Q * Zoiah uranium.oftc.net unununium.oftc.net 1119869082 Q * daniel_hozac uranium.oftc.net unununium.oftc.net 1119869082 Q * gregster uranium.oftc.net unununium.oftc.net 1119869082 Q * kevinp uranium.oftc.net unununium.oftc.net 1119869082 Q * SiD3WiNDR uranium.oftc.net unununium.oftc.net 1119869082 Q * Bertl_oO uranium.oftc.net unununium.oftc.net 1119869082 Q * pattieja uranium.oftc.net unununium.oftc.net 1119869085 N * _mcp mcp 1119869161 J * Bertl_oO ~herbert@212.16.62.52 1119869165 J * gregster ~gregor@greart.de 1119869177 J * dsoul darksoul@pingu.ii.uj.edu.pl 1119869187 J * tchan ~tchan@c-24-13-81-164.hsd1.il.comcast.net 1119869193 J * SiD3WiNDR luser@bastard-operator.from-hell.be 1119869234 J * xbing ~nb@dsl081-044-121.lax1.dsl.speakeasy.net 1119869266 J * Zoiah Zoiah@matryoshka.zoiah.net 1119869334 J * maharaja maharaja@ipax.at 1119869339 J * TheSeer ~theseer@border.office.salesemotion.net 1119869340 J * case ~case@donpanic.faveve.uni-stuttgart.de 1119869499 J * tbenita ~tbenita@tbenita.net1.nerim.net 1119869531 J * prae ~prae@ezoffice.mandriva.com 1119869544 J * kevinp ~kevinp@ny.webpipe.net 1119869611 J * daniel_hozac ~daniel@h56n2fls32o829.telia.com 1119869640 J * pattieja ~pattieja@adsl-69-153-174-41.dsl.stlsmo.swbell.net 1119869640 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il 1119869659 J * Vudumen vudumen@perverz.hu 1119869696 Q * tbenita Quit: 1119869894 M * rt ok, the zunzuncito is up and running ;-). but the server has not the ip adresse i have configured :-/. when i want to set it i get 1119869895 M * rt SIOCSIFADDR: Permission denied 1119869895 M * rt SIOCSIFFLAGS: Permission denied 1119869895 M * rt SIOCSIFADDR: Permission denied 1119869895 M * rt SIOCSIFFLAGS: Permission denied 1119869897 M * rt SIOCSIFFLAGS: Permission denied 1119870893 M * rt Doener, any ideas? 1119871384 J * sukria ~sukria@213.223.184.193 1119871518 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com 1119871981 J * eXplasm explasm@p549F799C.dip.t-dialin.net 1119872454 M * Doener rt: how did you check that is has no ip address assigned? 1119872481 M * Doener you didn't specify an alias, thus it won't have one, which means that ifconfig is unable to show that address 1119872495 A * Doener .oO( i should really write a wiki page about that... ) 1119872510 M * Doener using "ip" from iproute, you should see the address 1119872524 M * Doener syntax is "ip addr show" or just "ip a" 1119872579 M * Doener if you want an alias you can add a file called "name" into /etc/vserver//interfaces/0/ and put the alias in there (don't forget a newline at the end, the tools need that) 1119872623 M * Doener when you build the vserver and want an alias to be setup directly you need sth. like --netdev alias=eth0:10.0.0.1/8 1119872653 M * Doener that would give you eth0:alias as name for 10.0.0.1 1119872685 M * Doener --netdev hansi=eth0:192.168.0.100/24 would give you eth0:hansi for address 192.168.0.100 and so on... 1119872720 M * Doener HTH, got to go now, back later 1119872726 N * Doener Doener|gone 1119873187 Q * rs Quit: rs 1119873960 Q * nox Quit: leaving 1119874005 J * nox ~nox@noxlux.de 1119874324 Q * eXplasm Ping timeout: 480 seconds 1119876210 J * rs ~rs@ATuileries-153-1-32-154.w82-123.abo.wanadoo.fr 1119876429 J * ruuth VooDoo@topas.informatik.uni-ulm.de 1119877444 J * eXplasm explasm@p549F67D1.dip.t-dialin.net 1119878463 J * jonsmel ~jscottorn@209.33.206.3 1119879456 M * jonsmel morning everyone 1119879553 M * id moin 1119880097 J * eXplasm2 ~explasm@p549F73F0.dip.t-dialin.net 1119880262 Q * eXplasm Ping timeout: 480 seconds 1119881550 M * id is it possible to use vunify with jfs ? 1119883865 M * jonsmel Hi Bert, you here 1119883936 M * maharaja damn, my server crashes every few minutes 1119883942 M * maharaja 2.6.11.9-vs2.0-rc4 1119883949 J * explasm__ explasm@p549F5149.dip.t-dialin.net 1119883950 M * maharaja and there is no msg on the serial console 1119883956 M * maharaja any guess? 1119883957 Q * eXplasm2 Ping timeout: 480 seconds 1119883974 N * Bertl_oO Bertl 1119884002 M * Bertl jonsmel: somewhat .. :) 1119884017 M * Bertl maharaja: hmm, don't remember we had a 2.6.11.9-vs2.0-rc4 1119884022 M * jonsmel :) 1119884094 J * monrad ~monrad@213083190130.sonofon.dk 1119884110 M * Bertl evening monrad ! 1119884120 M * monrad hey 1119884121 M * Bertl id: does jfs support the xattributes? 1119884125 M * jonsmel bert: Do you think you will have a few mins to help me troubleshoot sometime 1119884141 M * Bertl yes, I'm available in a few minutes ... 1119884147 M * jonsmel ok, thnx 1119884194 M * maharaja Bertl: indeed. i patched a 2.6.11.9 with the 2.6.11.11 patchset 1119884200 M * maharaja anyways, what shall i do? 1119884209 M * maharaja i always crash when working inside a vserver 1119884212 M * Bertl how does it crash? 1119884221 M * maharaja the last two times, i was using vim and did a :x 1119884248 M * maharaja bevore that, vserver 1.9.5 used to crash while disk io (copy 200-400mb of data) 1119884252 M * maharaja it simply freezes 1119884266 M * maharaja no netwerk connectivity, no keyboard input 1119884273 M * maharaja but i still see the screen 1119884280 M * maharaja ocasionally, i get a kernel oops via the serial console 1119884286 M * maharaja but thats seldom 1119884365 M * maharaja i turned on some of the debugging stuff for the kernel 1119884368 M * maharaja but not all 1119884386 M * maharaja need any .config or any other information? 1119884401 M * maharaja before that, i thought it is a grsec <-> vserver problem 1119884414 M * maharaja but now, i did no grsecurity patch 1119884420 M * maharaja so i'm kind of puzzled 1119884448 M * rt to enable fakeinit, i have to put "fakeinit" to /etc/vservers/name/flags , right? 1119884543 M * Bertl maharaja: sounds to me like a hardware issue 1119884589 M * maharaja Bertl: testme.sh 000, 031 and 202 fail 1119884616 M * maharaja testme.sh-0.12 1119884617 M * Bertl that's not a good sign :) 1119884646 M * maharaja so .. :) 1119884664 M * Bertl please upload the complete output to pastebin.com or so 1119884981 M * maharaja of the testme.sh 1119884982 M * maharaja right? 1119885008 M * maharaja mhm, do i need new tools for v2? 1119885039 M * Bertl yes, of the testme.sh output ... 1119885067 M * Bertl rt: well, depends on what you consider 'fakeinit' (it's a little confusing) 1119885071 M * maharaja http://pastebin.com/304650 1119885097 M * maharaja # chcontext true && echo 1 1119885097 M * maharaja vcontext: vc_xidopt2xid(""): No such file or directory 1119885122 M * maharaja i use util-vserver: 0.30.205; Apr 12 2005, 15:02:32 1119885164 M * Bertl 0.30.205 is an unfortunate choice ... 1119885165 M * jonsmel brb bert, let me know when you are ready, thnx 1119885186 M * maharaja why? :) 1119885187 M * Bertl jonsmel: okay, let me know when youre back, I should have some time then ... 1119885205 M * maharaja anyways, ill switch to 207 1119885206 M * Bertl maharaja: because they where broken IIRC ... 1119885224 M * maharaja yes, the .sh script was broken 1119885233 M * maharaja but i fixed the small error and submitted a patch (iirc) 1119885667 J * sannes ~ace@cm-84.118.217.070.chello.no 1119885766 M * Bertl welcome sannes! 1119885813 M * maharaja any clue why "chcontext true" breaks? 1119885819 M * sannes hi there Bertl :) 1119885821 M * maharaja or better: is broken 1119885928 M * sannes Bertl : if you do funky namespace stuff, is there anything you shouldn't be doing? 1119886175 M * sannes limited to what vserver and fstab can do .. 1119886617 M * Bertl hmm, no, actually they should work pretty the same as on the host 1119886641 M * Bertl because we are not doing any different as the usual linux host does 1119886656 M * Bertl (we just use more namespaces than the typical linux system) 1119886710 J * zaikxtox ~xcvxcvx@200.80.231.196 1119886738 M * Bertl welcome zaikxtox! 1119886763 M * zaikxtox hello. i have set up a vsever on a debian box using a howto. everything looks fine.. but when i use ssh to enter my vserver it uses my root server filesystem! 1119886766 M * zaikxtox hello Bertl 1119886771 M * DaPhreak lo Bertl, #vserver ;) 1119886796 M * Bertl zaikxtox: you have to limit your host's sshd to a subset of ip addresses 1119886809 M * Bertl (best done with the Listen* directives in the sshd config) 1119886813 M * DaPhreak zaikxtox: you should check your hosts sshd_config if it listens on all ips (ListenAddress) 1119886813 M * Bertl hey DaPhreak! 1119886832 M * zaikxtox Bertl, oh. it looks reasonable. thanks :) 1119886950 M * zaikxtox fine 1119886971 M * zaikxtox and... how can i install packages inside the vserver without an ssh connection??? 1119886972 M * zaikxtox :D 1119886974 M * zaikxtox chroot? 1119886983 M * Bertl no 1119886991 M * Bertl you have some options there 1119886999 M * zaikxtox oh. 1119887003 M * Bertl - you can use 'vserver enter' 1119887018 M * Bertl - you can use 'vserver (su)exec ...' 1119887032 M * zaikxtox fine. i wil try the "enter" 1119887039 M * Bertl - you can use the vserver management commands 1119887058 M * Bertl (which, depending on the distro, will do some package installation/removal for you) 1119887064 M * zaikxtox you are right Bertl 1119887070 M * zaikxtox really, thanks a lot 1119887079 M * Bertl yw 1119887173 M * zaikxtox i can now start to play :) 1119887200 M * Bertl have fun! 1119887227 M * Bertl jonsmel: around now? 1119887603 M * jonsmel ok, I am back now as well 1119887614 M * Bertl k 1119887615 M * maharaja bbl - compiling 2.6.11.12-vs2.0-rc4 and util-vserver 0.30.207 1119887644 M * jonsmel Did you get my emails about what I had found last week 1119887655 M * jonsmel any thought's on that? 1119887829 M * Bertl I did not completely udnerstand the issue ... 1119887852 M * jonsmel sorry 1119887854 M * Bertl but to me it looks like you get problems once the ip is locked 1119887874 M * Bertl so I would suggest to do some simple tests with your 'working' and vserver patched setup 1119887894 M * Bertl does that sound reasonable? 1119888029 M * Bertl (btw, I'm pretty sure the mtab file is not what causes your issues) 1119888191 M * jonsmel ok 1119888191 M * jonsmel np 1119888196 M * jonsmel let me know what to try 1119888260 M * Bertl okay, first, make a directory on your lustre fs, and everything we do will happen on the host 1119888274 M * jonsmel ok 1119888278 M * Bertl we need two ips, the one the host uses, and the other which the guest used 1119888314 M * jonsmel they don't need to be public though right 1119888319 M * Bertl no 1119888327 M * Bertl the first few tests here: 1119888345 M * jonsmel ok 1119888346 M * Bertl (I assume you cd-ed into the dir) 1119888348 M * jonsmel i have a test dir 1119888351 M * jonsmel yes 1119888364 M * Bertl chcontext --xid 666 touch x.test 1119888387 M * jonsmel ok that worked 1119888387 M * Bertl lsxid x.test 1119888414 M * jonsmel lstat(): value too large for defined data type 1119888452 M * Bertl hmm, that's interesting, please make an strace -fF of that one, and upload it somewhere 1119888465 M * Bertl showattr x.test 1119888552 M * Bertl setattr --iunlink x.test 1119888581 M * jonsmel same error for showattr x.test 1119888610 M * Bertl okay, let's fallback to some simpler tools then ... 1119888665 M * jonsmel ok 1119888700 M * Bertl http://vserver.13thfloor.at/Experimental/TOOLS/vxid-0.03.tar.bz2 1119888712 M * Bertl http://vserver.13thfloor.at/Experimental/TOOLS/vflags-0.02.tar.bz2 1119888741 M * Bertl just unpack and do 'make' 1119888759 Q * prae Quit: Execute Order 69 ! 1119888821 M * jonsmel both done 1119888835 M * Bertl ./vxid x.test 1119888954 M * jonsmel xid=-1208373664, flags=0x01000000, mask=0x00060000, ../x.test 1119888979 M * Bertl okay, did it report any errors or so? 1119888988 M * jonsmel that is all that reported 1119889033 M * Bertl okay, fine .. let's try 1119889038 M * jonsmel well, hold on 1119889046 M * jonsmel the system is not responding now 1119889069 M * Bertl ah? same as you observed? 1119889111 M * jonsmel looks like i will have to reboot it 1119889125 M * jonsmel have you ran into it then? 1119889227 M * Bertl lets see if this command will lock up your machine again ... 1119889235 M * Bertl if so, we know where to look for it ... 1119889251 M * jonsmel ok, i'll let you know when it is back up 1119889433 Q * rs Quit: rs 1119889523 Q * zaikxtox Quit: Abandonando 1119889715 M * jonsmel ok 1119889729 M * jonsmel nothing reported on screen and it is not froze yet 1119889752 M * Bertl so the command works sometimes and sometimes not? 1119889766 M * jonsmel it would seem so 1119889776 M * Bertl that's not good .... 1119889809 M * Bertl what tests did you do with the lustre setup (without vserver) 1119889818 M * jonsmel so what if i unmount that partition and remount then try again 1119889825 M * jonsmel would that help at all 1119889850 M * jonsmel I have not had anyproblems creating, removing, touching files 1119889872 M * Bertl any tests regarding xattributes or >16 bit uid/gid? 1119889881 M * jonsmel no, 1119889899 M * Bertl hmm, maybe we should test first with a lustre only setup then? 1119889925 M * jonsmel ok 1119890017 M * Bertl your setup includes at least two hosts, right? 1119890026 M * jonsmel not right now 1119890029 M * jonsmel just one 1119890040 M * Bertl ah, okay, so 'lustre' local for now? 1119890047 M * jonsmel yes 1119890075 M * jonsmel well, it is not setup using the local script provided by them 1119890095 M * jonsmel it is using a script for multiple nodes, but only itself is setup 1119890124 M * Bertl okay 1119890125 M * jonsmel but of course I can change that if we want to 1119890133 M * jonsmel and use the local script 1119890174 M * Bertl well, let's not change too many things at once ... 1119890196 M * jonsmel right 1119890208 M * Bertl and you might consider setting up some QEMU images for each kernel ... (unless you want to recompile/reinstall every time) 1119890214 M * jonsmel ps... i am also getting a second node ready 1119890237 M * jonsmel no, I will seperate them 1119890242 M * jonsmel diff images that is 1119890318 M * jonsmel oh, crap i forgot i have to leave for a bit but i will be back in about 2 hrs will you be around about then 1119890344 M * Bertl probably ... 1119890364 M * jonsmel ok, i will check when i get back by that time i will have 2 boxes setup as well. 1119890379 M * jonsmel what I might do is use the one I have now as the one with lustre and vserver 1119890385 M * jonsmel and the second one as just lustre 1119890399 M * Bertl yes, might help ... 1119890404 M * jonsmel so that way we can test quicker if needed 1119890407 M * jonsmel right 1119890419 M * jonsmel it will be ready by the time i am back 1119890421 M * jonsmel thanks 1119890426 M * Bertl yw, cya! 1119890428 M * jonsmel laterzzz 1119890435 M * FaUl will somone of you folks go to WTH btw? 1119890559 J * rs ~rs@Laubervilliers-151-13-4-57.w82-127.abo.wanadoo.fr 1119890607 M * Bertl evening rs! 1119890614 M * rs hi 1119890719 M * FaUl hey rs 1119890727 M * FaUl hey bertl 1119890853 M * FaUl .oO( obvously not ) 1119891653 Q * explasm__ Remote host closed the connection 1119891905 J * eXplasm explasm@p549F5149.dip.t-dialin.net 1119892040 M * Bertl wb eXplasm! 1119892317 N * Doener|gone Doener 1119892396 M * Bertl evening Doener! 1119892404 M * Doener evening Bertl 1119892420 M * Bertl how are you? everything fine? 1119892484 M * Doener yep, finally the "new" car is "zugelassen" (don't know which translation of that word would fit here...) 1119892509 M * DaPhreak "angemeldet" ? ;) 1119892521 M * Doener ehr, yes! 1119892529 M * Doener now we just need a translation for that one ;) 1119892569 M * Doener weather is also fine, not too hot... (too cold is impossible when it comes to my room in the summer time ;) 1119892607 M * DaPhreak well, licensed, permitted, registered *g* pick one ;) 1119892741 M * Bertl excellent! what's new in linux-vserver land? 1119892774 M * Bertl btw, read your 'request' to include linux-vserver into the comparison chart 1119892909 M * Doener not much news, been a littly busy 1119893048 M * Bertl so how will the summer be, except hot? :) 1119893097 M * Bertl I would really like to make ngnet happen pretty soon ... but I guess I need some help with testing and somesuch ... 1119893134 M * Doener still didn't find time to look into the tc stuff you suggested to be useful for ngnet :( 1119893177 M * Bertl well, I'm not talking about the past, I'm looking to the future ... 1119893227 M * Doener hm, but if we want to use tc I need to know it ;) at least to some extend 1119893337 M * Bertl yeah, probably, but I guess we can get there by learning it by using/doing it :) 1119893462 M * Bertl as far as I can tell, the 2.0 release will not change anymore (kernel side) 1119893480 M * Bertl well, I'm pretty sure we get a bunch of bug reports once it is released 1119893536 M * Doener hehe 1119893563 M * Bertl I'm currently not in the mood to hack on the userspace tools, and some folks already started to play with them and/or the syscalls :) 1119893595 M * Doener yep, seen that, i'm not eager to hack/search my way through them either 1119893630 M * Doener don't know what it is, but somehow they get harder to read instead of easier... 1119893632 M * Bertl nevertheless I'm collecting ideas for a generalized command/control daemon on the host, which does the complete guest management (replacing the active part of the tools, not the build part) 1119893666 M * Bertl (as I consider this a better approach long term than the current) 1119893702 M * Doener yep, i planned on writing such a thing about a year ago, but obviously i forgot about it ;) 1119893731 M * Doener would be helpful when working on a web-based interface 1119893773 M * Bertl yes, well, if you prefer to work on something like that, I'm not fixed on anything right now ... 1119894042 M * Doener hm, I'll try to produce something of poc quality ;) 1119894400 M * Bertl okay, sounds good ... then I'll finish the 2.0 release for now and we wait for the feedback :) 1119894510 J * erwan_ho ~erwan@konilope.dyndns.org 1119894516 M * Bertl evening erwan_ho! 1119894524 M * erwan_ho hey Bertl 1119895085 M * Doener Bertl: is the syscall_new.h ready for use? 1119895114 M * Bertl it should be .. I used it in the recent hack tools 1119895144 M * Bertl (and they seem to work well on various archs) 1119895167 M * Bertl if you encounter any issues, please let me know 1119895214 M * Doener ok, just wondered about the name 1119895225 M * Bertl yeah, will rename it to syscall.h 1119895239 M * Bertl the bloat version is basically obsolted 1119895414 M * Bertl *obsolteted 1119896684 Q * sukria Quit: see you 1119896812 Q * ruuth Quit: Nettalk6 der Freeware IRC-Client 1119897785 M * Bertl back later, (dinner time) 1119897789 N * Bertl Bertl_oO 1119897905 J * knoppix_ ~knoppix@dsl-213-023-144-072.arcor-ip.net 1119898342 J * Aiken ~james@tooax6-203.dialup.optusnet.com.au 1119898794 J * scott101 ~scott101@adsl-63-205-64-73.dsl.snfc21.pacbell.net 1119898968 M * scott101 hey guys, I folled the centos4 guide to get a vserver up, ran test.sh all good. but when I try to get the damn guestOS of centos up and running I hit this error: Config Error: Error accessing file for config... (here is a pastebin: http://pastebin.ca/16000 ) 1119899006 M * scott101 google search on the issue showed a IRC log with the same issue, but I couldn't find any answers in that log..., any ideas? 1119899474 M * eXplasm is full routing editing possible on a vserver? 1119899652 Q * rs Ping timeout: 480 seconds 1119899653 M * eXplasm btw, whats a postive argument for the vserver compared to the "user mode linux" 1119899991 J * _are_ ~are@dsl-084-056-130-115.arcor-ip.net 1119900145 M * daniel_hozac eXplasm: far less overhead. 1119900228 J * shuri shuri@dsl.speedline209.66.electronicbox.net 1119900277 M * eXplasm user mode linux is the better way for developers, i think 1119900308 M * daniel_hozac depends on the developers, i suppose. 1119900504 P * shuri 1119900665 J * are|lunch foobar@dsl-084-056-149-120.arcor-ip.net 1119900805 Q * _are_ Read error: Connection reset by peer 1119901046 J * Jani ~mail@G9d46.g.pppool.de 1119901049 M * Jani *waves* Hi all 1119901500 N * are|lunch _are_ 1119901503 M * _are_ hi 1119902168 M * jonsmel Bertl: I am back now and have the two boxes setup as well. 1119902182 M * jonsmel let me know when you are available 1119902226 N * Bertl_oO Bertl 1119902258 M * Bertl hey Jani! 1119902290 M * Bertl hey jonsmel, I'm back! 1119902298 M * jonsmel cool 1119902312 M * jonsmel i have the second box setup running their local script 1119902359 M * Bertl excellent! 1119902375 M * Bertl scott101: I get invalid ID with that url 1119902424 M * Bertl jonsmel: okay, let's do some chattr tests on the non-vserver machine 1119902458 M * Bertl like 'touch x.test' 'chattr +i x.test' 'lsattr x.text' 1119902469 M * maharaja bertl: installed the new tools and it testme.sh-0.12 works without a flaw 1119902474 M * maharaja and installed a new kernel ;) 1119902490 M * maharaja so, lets see what crashes this one will bring 1119902492 M * Bertl good 1119902512 M * jonsmel ok 1119902522 M * Bertl maharaja: did you enable some kernel debug stuff? 1119902531 M * jonsmel have ----i------------ x.test 1119902543 A * Jani has now also get SuSe to work with the vs2.0-rc4 and the lastest utils 1119902603 M * Bertl Jani: cool, wanna share what the issues were and how you solved them? 1119902645 M * Bertl jonsmel: okay, now let's try the same on the l+v host (and do a manual sync afterwards, on both) 1119902700 M * jonsmel same result from lsattr 1119902715 M * jonsmel what do you mena manual sync 1119902720 M * Bertl 'sync' 1119902721 M * jonsmel oops, mean 1119902734 M * jonsmel oh ok, done 1119902755 M * Bertl you have the strace from the previous runs (we did)? 1119902773 M * jonsmel i do not, strace froze up the session 1119902789 M * jonsmel i can try it again if you like 1119902791 M * Bertl okay, then let's try again with the vxid and strace 1119902830 M * Bertl do you have some console (serial or real) for that machine where you can see kernel messages? 1119902860 M * jonsmel I am ssh'd in 1119902877 M * jonsmel now, you want this run on the l+v box correct 1119902892 M * Bertl also, could you please provde a kernel patch (of your current krnel source tree) against vanilla 2.6.12? 1119902916 M * Bertl yes, the l box will not say much, except that the syscall is not supported 1119902936 M * jonsmel ok, forgive me, what was the strace command again 1119902941 A * sannes is celebrating his first full hour on vserver with 2.6 kernel 1119902952 M * Bertl congrats sannes! 1119902967 M * Jani Bertl: The onliest issue was that I have missread a number ;) 1119902989 M * Bertl sounds good too :) 1119903068 M * sannes thank you! :) I hope I can say something about a week in 6 days and 23h .. :) atleast it is a record thus far .. :) 1119903125 Q * scott101 Quit: 1119903277 M * Bertl jonsmel: strace -fF -o vxid.trace vxid ... 1119903497 M * jonsmel it freezes when I run strace -fF -o vxid.trace vxid ../x.test 1119903507 M * jonsmel is that incorrect 1119903513 M * jonsmel my command that is 1119903540 M * Doener grr... recv returns 0 when a non-blocking socket has no data and sets errno to EAGAIN... i'd expected it to return -1... stupid thing ;) 1119903577 M * Bertl sockets are tricky, should look into select and friends 1119903606 M * Bertl select(2) and poll(2) that is 1119903621 M * SNy pay attention to the details 1119903634 M * SNy mainly the timeout WRITE parameter 1119903640 M * SNy that gets reset 1119903651 M * Doener yep, got beej's tutorials on the next virtual desktop :) unfortunately those are written using HPUX, and on HPUX recv seems to return -1 in the said case 1119903678 M * Bertl k, probably depends on the library ... 1119904027 M * Bertl jonsmel: couldn't parse your messages, but it would be really beneficial if you could hook up some serial console, to log kernel messages, and no ssh is not a good option :) 1119904063 M * jonsmel ok, have one 1119904130 M * Aiken what about a 2nd ssh session tail -f /var/log/messages ? 1119904148 M * Bertl Aiken: doesn't help when the kernel panics/oopses ... 1119904164 M * Bertl you'll see ... nothing :) 1119904176 M * Aiken oh, that kind of message. the bad type of message 1119904194 M * jonsmel Can't get strace to finish though 1119904198 M * Bertl lol, yeah, the evil final message ... 1119904219 M * Bertl jonsmel: okay, so you got a console or kernel oops? 1119904310 M * Bertl (or maybe both?) 1119904335 M * jonsmel console 1119904336 M * jonsmel sorry 1119904358 M * Bertl k, and it doesn't write anything, just sit there? 1119904374 M * jonsmel yes 1119904381 M * Bertl did it write some stuff into the trace file? did you put the trace file on the lustre fs? 1119904411 M * Bertl do you have the delta patch (vanilla -> your kernel) somewhere? 1119904427 M * jonsmel sorry, not yet. 1119904438 M * Bertl np, just let me know when you are there ... 1119904465 M * jonsmel ok 1119904473 M * jonsmel what diff command do you want to catch it all 1119904481 M * jonsmel i have the two src dirs now 1119904485 M * Bertl best would be to do: 1119904502 M * jonsmel and, do you want it of l+v or just lustre 1119904503 M * Bertl cp -la src src-copy 1119904509 M * DaPhreak `diff -NrupP --minimal old.dir new.dir` 1119904511 M * Bertl (for both sources) 1119904527 M * Bertl then do 'make mrproper' inside each of the copies 1119904539 M * Bertl then the diff DaPhreak mentioned 1119904546 M * DaPhreak heh ;) 1119904566 Q * neofutur Quit: leaving 1119904569 A * DaPhreak sometimes reads Bertl's diffs 1119904578 M * Bertl amazing :) 1119904591 J * neofutur ~neofutur@neofutur.net 1119904596 M * Bertl wb neofutur! 1119904599 Q * neofutur Quit: 1119904601 M * DaPhreak yeah .. *G* and i'm slowly getting an idea whats happening there 1119904638 J * neofutur ~neofutur@neofutur.net 1119904642 M * Bertl so you can take over development pretty soon, yeah? *G* 1119904679 M * DaPhreak lol not for sure ;) 1119904693 M * DaPhreak well, just wait 2-10 years *g* maybe then 1119904723 M * Bertl when linus released his first EULA ... 1119904736 M * eXplasm is ip routing table editing possible on a vserver? 1119904746 M * DaPhreak on the host, yes 1119904754 M * DaPhreak but not from inside of a vserver 1119904760 M * eXplasm damn 1119904769 M * DaPhreak but i guess ngnet will change that 1119904781 M * Bertl well, you can give the required caps to the guest 1119904801 M * DaPhreak NET_ADMIN ? 1119904805 M * Bertl but it will allow the guest to control _all_ the routing tables 1119904806 M * eXplasm how 1119904823 M * Bertl i.e. not just his own ... 1119904844 M * Doener DaPhreak: if you keep asking Bertl often enough, you'll learn a great deal ;) 1119904859 M * Doener just don't scare him away, I still need him to ask questions myself *g* 1119904862 M * Bertl eXplasm: this is part of the bcapabilites (see config file) 1119904866 M * DaPhreak eXplasm: give the vserver the required capability via bcapabilities .. 1119904867 M * Bertl Doener: thanks! 1119904873 M * eXplasm kk 1119904919 M * DaPhreak heh, naah ... he doesn't seems to be scared Doener, since he's still here .. *runofdoenersrevenge* 1119904929 M * DaPhreak s/run/runs/ 1119904970 A * Doener starts throwing dabbers at DaPhreak 1119904993 M * DaPhreak nah .. only joking ;P damn its again bed-time ;( 1119905045 M * Bertl night then ... (have a scary one :) 1119905099 M * DaPhreak thanks Bertl ;) good (whatever `date +'%Z'` says) then :) 1119905100 M * Doener g'night DaPhreak 1119905109 M * jonsmel that diff file is huge 1119905129 M * jonsmel 500 mb is that what you want 1119905133 M * Bertl hmm, check with 'lsdiff' that you do not include the object files or such 1119905159 M * Bertl 500 mb for vserver + lustre sounds excessive 1119905169 M * Doener looks like a missing "make distclean" or a kernel-in-kernel source 1119905173 M * jonsmel does to mee too 1119905178 M * Bertl that's more like the entire kernel 1119905207 M * Bertl btw, you can also pack up the kernel source if that is easier for you :) 1119905212 M * jonsmel i did make mrproper with in the two sources 1119905217 M * jonsmel then rand the diff cmd 1119905241 M * Bertl as I said, check with lsdiff 1119905363 M * jonsmel is lsdiff another program i would have to install, i don't have it 1119905374 M * Bertl hmm, diffutils? 1119906116 Q * knoppix_ Quit: Verlassend 1119906500 M * jonsmel hmm, maybe I am totally missing something, i installed diffutils and lsdiff is still not there 1119906530 M * Bertl hmm, sorry my fault: 1119906531 M * Bertl rpm -qf `which lsdiff` 1119906532 M * Bertl patchutils-0.2.30-1mdk 1119906555 M * jonsmel ps... I am running on Debian 1119906583 M * Bertl yeah, but the packages are called similar ... 1119906591 M * jonsmel yep, installing now 1119906629 M * jonsmel now, I am lsdiffing the two src dirs correct 1119906649 M * Bertl no, you use lsdiff to list info about the patch 1119906727 M * jonsmel ah 1119906732 M * jonsmel found out what it was 1119906741 M * jonsmel fixed, only 3.3 mb now 1119906754 M * Bertl much better :) 1119906836 M * jonsmel it is on that site I gave you before now 1119906852 M * jonsmel under the filename 2.6.12.diff 1119906860 M * maharaja Bertl: i enabled some debuggin stuff, yes 1119906884 M * Bertl jonsmel: I'll upload it on vserver.13thfloor.at then ... 1119906909 M * jonsmel oh, that is for lustre with vserver correct 1119906920 M * Bertl hope so? 1119906927 M * jonsmel yes it is sorry 1119906934 M * jonsmel i was making sure that is what you wanted 1119906963 M * maharaja Bertl: my current config is at http://raoul.bhatia.at/~raoul/config-2.6.11.12-vs2.0-rc4-p4b 1119906986 M * Bertl I assume it is vanilla 2.6.12 -> 2.6.12-vs2.0-rc4-luste1.4.2.1 right? 1119907071 M * jonsmel sorry, reversed 1119907078 M * Bertl np 1119907082 M * jonsmel I can rerun the diff if you want 1119907094 M * Bertl no, thanks, we know how to specify -R :) 1119907095 M * Doener Bertl: hm, any idea what the --migrate-self option of vcontext does? looks strange to me, as the VCMD_ctx_create already does the migration... and it seems to migrate into the process' own context?! 1119907112 M * Doener (the latter one if for --migrate-self, not VCMD_ctx_create) 1119907120 M * Doener s/if/is/ 1119907135 M * Bertl no, the migration is only half way in 1119907135 M * jonsmel I am still having issues getting strace to run, let me make sure the command is correct.... 1119907140 J * rs ~rs@imhotep.rhapsodyk.net 1119907144 M * Bertl evening rs! 1119907163 M * jonsmel strace -fF -o vxid.trace /vxid-0.03/vxid x.test 1119907164 M * Bertl jonsmel: try it on an unpatched host, just replace the vxid ... with sleep 1119907166 M * rs re 1119907180 M * Bertl jonsmel: command looks fine 1119907193 M * Bertl if somebody is interested: http://vserver.13thfloor.at/Stuff/LUSTRE/delta-2.6.12-vs2.0-rc4-lustre1.4.2.1.diff.bz2 1119907199 M * jonsmel unpatched without vservers or a straight vanilla kernel 1119907220 M * Doener Bertl: hm, --migrate-self seems to just set the namespace 1119907270 M * Doener "half way in" means? *feels lost again* 1119907271 M * Bertl yes, from the API point of view, the migration ends with clearing the setup flag 1119907303 M * Bertl you do, create, setup stuff, change things, clear setup flag and you are locked in 1119907312 M * jonsmel ok, got results from the box with vservers patched in 1119907330 M * maharaja Bertl: if you've got time to check my debug settings, send me a note (e.g. query) 1119907336 M * maharaja off to bed now 1119907364 M * Bertl k 1119907369 M * jonsmel It errors on screen get_iattr:: function not implemented 1119907408 M * jonsmel xid= 0, flags=0x00000000, mask=0x00000000, x.test 1119907418 M * Doener Bertl: ok, that's what --endsetup does, --migrate-self just sets the namespace... strange stuff ;) 1119907425 M * jonsmel there is a page of stuff within vxid.trace 1119907464 M * rs Bertl: lustre support for linux-vserver ?\ 1119907483 M * rs is it doing well ? 1119907487 M * Bertl rs: well, we are looking into it, but it doesn't look good 1119907500 M * rs why ? 1119907504 M * Bertl the main issue seems to be that lustre can not be compiled into the kernel 1119907508 M * jonsmel I have posted the vxid.trace on that site under that filename 1119907545 M * Bertl rs: it requires a bunch of 'dubious' kernel aptches, and then still builds everything into a bunch of modules 1119907592 M * Bertl and you have to do very complicated setup to get it running it seems 1119907604 M * Bertl but jonsmel can probably tell you more about it 1119907610 M * jonsmel actually going to 2.6.12, it is pretty quick 1119907626 M * jonsmel the earlier kernels are harder 1119907632 Q * erwan_ho Quit: Leaving 1119907651 M * jonsmel It took me about 20 mins to get the second box running 1119907675 M * jonsmel mostly compiling the kernel 1119907692 M * Doener Bertl: btw, you said something about a bug in the rss/anon macros in the 2.6.12-vs2.0-rc4 patch, is there a patch fixing that? and what is that bug? 1119907702 M * jonsmel but yes I agree with Bert that they should have made it more user frendily 1119907712 M * Bertl Doener: huh? 1119907742 M * Doener Jun 23 17:18:06 Doener most significant are the rss/anon(?) account macros 1119907742 M * Doener Jun 23 17:18:23 Bertl yep, including a bug, which is already fixed, IIRC 1119907770 M * Doener seems like i misinterpreted/misread that? 1119907815 M * Doener (that's from the discussion of changes from 2.6.11->2.6.12 regarding the vserver patch) 1119907834 M * Bertl ah, yeah, I emant the kernel patches ... 1119907853 M * Bertl (IIRC, you submitted a patch :) 1119907881 M * Doener ah, ok, yep, that was included in rc5 IIRC 1119908044 M * Bertl jonsmel: what are the stamp-* thingies? 1119908091 M * jonsmel where at 1119908103 M * jonsmel ... where are you seeing them 1119908191 M * Bertl they appear among others in the kernel source dir :) 1119908199 M * Bertl (with your patch reversed :) 1119908232 M * Bertl probably a side effect of the quilt/whatever 1119908251 M * jonsmel could be, i'm not sure what they would be there for 1119908261 M * jonsmel i'm just looking at it now 1119908373 M * Bertl okay, here is a real delta now ... 1119908374 M * Bertl http://vserver.13thfloor.at/Stuff/LUSTRE/delta-2.6.12-vs2.0+lustre.diff.bz2 1119908386 M * FaUl what is lustre? 1119908395 M * Bertl distributed filesystem 1119908397 M * FaUl ah 1119908399 M * FaUl nice 1119908464 M * jonsmel you know stamp-* might have to do with make-kpkg 1119908471 M * jonsmel for debian 1119908482 M * jonsmel not sure, but that is what it looks like 1119908554 M * Bertl http://vserver.13thfloor.at/Stuff/LUSTRE/delta-2.6.12-vs2.0-lustre.diff.bz2 1119908570 M * Bertl okay, this is now a delta from vs2.0(rc4) to the lustre patched version ... 1119908635 M * jonsmel that is what the stamp-* are, for make-kpkg for debian images 1119908647 M * jonsmel sorry, I can remove them and rediff if you want 1119908671 Q * nox uranium.oftc.net neutron.oftc.net 1119908671 Q * Hunger uranium.oftc.net neutron.oftc.net 1119908759 M * Bertl and here is the automagically processed diff to look at: http://www.13thfloor.at/~herbert/delta-2.6.12-vs2.0-lustre.diff.hl 1119908789 M * Bertl ah, I missed the debian dir :) 1119908890 J * Hunger Hunger.hu@levnor.hu 1119908896 M * jonsmel so did i 1119908899 M * Bertl http://www.13thfloor.at/~herbert/delta-2.6.12-vs2.0-lustre2.diff.hl 1119908949 M * jonsmel What are you thinking our next direction is to go 1119908973 M * Doener ok, got a really stupid daemon, that listens on a unix domain socket, plus a client that simple reads from stdin and forwards input to the daemon 1119908974 M * Bertl well, I'll look at the kernel aptch/changes, and will comment/fix what I see 1119908974 M * jonsmel btw, you still have to install lustre to get the modules 1119908982 M * jonsmel once you have the patched kernel 1119908993 M * Bertl but you're on your own with the modules ... 1119909004 M * Doener doing "create1000" i get a pretty much unconfigured context with a single process :) 1119909017 M * jonsmel right, I was just mentioning it for others, just in case 1119909023 M * Bertl probably the module code needs some changes too ... 1119909040 M * jonsmel ah 1119909043 M * Bertl Doener: being a newline or cr? 1119909050 M * jonsmel let me know if I can do anything 1119909052 M * jonsmel to help 1119909071 M * jonsmel I do have to run for today, Internet outages that need fixed. 1119909081 M * jonsmel be back tomorrow 1119909093 M * Bertl k, cya! 1119909099 M * jonsmel thanks for the help bert 1119909103 M * jonsmel later 1119909106 M * Bertl yw 1119909133 M * Doener hitting enter, client uses fgets for now... 1119909158 J * nox ~nox@noxlux.de 1119909160 Q * Aiken uranium.oftc.net jupiter.oftc.net 1119909160 Q * eXplasm uranium.oftc.net jupiter.oftc.net 1119909160 Q * sannes uranium.oftc.net jupiter.oftc.net 1119909160 Q * monrad uranium.oftc.net jupiter.oftc.net 1119909160 Q * flock uranium.oftc.net jupiter.oftc.net 1119909160 Q * pattieja uranium.oftc.net jupiter.oftc.net 1119909183 M * Bertl i.c. guess it would be good to make most 'commands' line oriented (but with the arguments at one line) 1119909187 J * Aiken ~james@tooax6-203.dialup.optusnet.com.au 1119909187 J * eXplasm explasm@p549F5149.dip.t-dialin.net 1119909187 J * sannes ~ace@cm-84.118.217.070.chello.no 1119909187 J * monrad ~monrad@213083190130.sonofon.dk 1119909187 J * pattieja ~pattieja@adsl-69-153-174-41.dsl.stlsmo.swbell.net 1119909187 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il 1119909190 M * Bertl welcome nox! 1119909220 M * Doener yeah, just playing around now to get a feeling for it... 1119909224 M * daniel_hozac and please use readline for input ;) 1119909235 M * Bertl Doener: sure, no critizism ... 1119909257 M * Bertl daniel_hozac: I think it should working something like this: 1119909269 M * Bertl daemon: listening for connections (unix/port) 1119909302 M * Bertl client: opening connection to daemon and submitting commands, queries, and receiving results, confirmations ... 1119909317 M * Bertl (of course the client can support readline :) 1119909331 M * Doener the unix socket currently gets text feeded, guess I'll use some structs later and have the parser and stuff in a client that may be used like some administration thingy 1119909436 M * Bertl okay have to leave now, back in an hour or so .. 1119909444 N * Bertl Bertl_oO 1119909447 M * Doener cya 1119909515 Q * cods Quit: brb 1119909974 Q * Hollow Quit: /dev/zero 1119910012 J * Hollow ~Hollow@home.xnull.de 1119910260 J * are|lunch foobar@dsl-084-056-141-025.arcor-ip.net 1119910261 M * FaUl der von dem avr-rechner sollte auf jedenfall tun 1119910264 M * FaUl oh, wwn 1119910348 M * FaUl sollte nu tun 1119910351 M * FaUl narf 1119910460 M * Doener heh 1119910676 Q * _are_ Ping timeout: 480 seconds 1119912945 N * Bertl_oO Bertl 1119912958 M * Bertl back now ... 1119913089 M * Doener ah damn, i'm an idiot... 1119913102 M * Doener the semantics of revc() _are_ like those on hpux... 1119913148 M * Doener gcc just suggest to put parantheses in: if (ret = recv() > 0) (of course)... 1119913199 M * Doener and of course, i put too many of them in there, i had: if ((ret = (recv() > 0))) 1119913247 Q * eXplasm Quit: Verlassend 1119913339 M * Bertl lol 1119913347 M * Bertl well, happens ... 1119913371 M * Doener yeah, but on the positive side, i feel like getting into a code rush... 1119913389 M * Doener didn't have that since last summer, and i really enjoy it :) 1119913417 M * Bertl well, can I look at the current state/code? 1119913448 M * Bertl I might have some suggestions if you like to get them ... 1119913492 M * Doener not much to see ;) but I'll upload it anyways 1119913651 M * Doener http://www.13thfloor.at/~doener/vserver/tools/vservd.tar.bz2 1119913655 M * Bertl tx 1119913660 M * Doener pretty experimental, not really clean code ;) 1119913668 M * Bertl np with that 1119913678 M * Doener you have been warned ;) 1119913693 M * Doener but it has a makefile :) 1119913698 M * Bertl OMG, I've been warned *G* :) 1119913866 M * Bertl okay, keep rushing ... 1119913876 M * Doener btw, I'm looking into select/poll now, in case that you wonder 1119913894 M * Bertl do you need some (known working) example codeß 1119913902 M * Bertl s/ß/?/ 1119913907 A * Doener didn't rush yet (except reading stuff about userspace/ipc programming) 1119913941 M * Doener http://www.ecst.csuchico.edu/~beej/guide/net/html/advanced.html#select 1119913960 M * Doener that's what i'm working with atm 1119914265 M * Bertl ah, yes, you find examples all over the internet ... 1119914586 Q * rs Read error: Connection reset by peer 1119914603 J * are|dinner foobar@dsl-084-056-137-250.arcor-ip.net 1119914697 J * rs ~rs@imhotep.rhapsodyk.net 1119914851 J * Johnsie ~john@acs-24-154-32-12.zoominternet.net 1119914859 M * Bertl welcome Johnsie! 1119914863 M * Johnsie Hi. :) 1119914918 M * Johnsie I'm still new to the vserver thing... 1119914940 M * Johnsie A friend is trying to get OpenVPN to work... 1119914980 M * Bertl sounds good ... 1119914986 M * Johnsie And when he brings up the VPN on a VPS (virtual private server), it won't ping down the other side. 1119914991 Q * are|lunch Ping timeout: 480 seconds 1119915014 M * Johnsie I kind of have an idea of how he's setting things up, but do you have any tips, warnings, suggestions, etc? 1119915044 M * Johnsie It works from the host... 1119915056 M * Johnsie Rather, the non-chroot environment does fine. 1119915058 M * Bertl well, havent tried OpenVPN yet, but I guess it does require something like tun/tap 1119915066 M * Johnsie From a VPS standpoint, it doesn't. 1119915074 M * Johnsie Okay. 1119915147 M * Johnsie I'll see what happens and report back. ;) 1119915150 M * Bertl if so, you have to do some tricks to make it work 'securely' inside a guest 1119915155 M * Johnsie So, how are things here? 1119915158 M * Johnsie Okay. 1119915198 M * Bertl things are fine here ... if you have any issues you want to resolve, just try to explain them to us ... ;) 1119915210 M * Johnsie haha 1119915211 M * Johnsie Alright. 1119915223 M * Johnsie Well, the gentleman who's setting this stuff up is a brainiac. 1119915240 M * Johnsie I told him I'd fish around and see if anyone else has been down this lane with OpenVPN. 1119915252 M * Johnsie Between your suggestions and his brains, I'm sure he'll figure it out. 1119915263 M * Johnsie I'm still learning. 1119915269 M * Johnsie I sure love the vserver concept... 1119915281 M * Johnsie I can deploy/teardown a system in about four minutes. 1119915286 M * Johnsie If that. 1119915355 M * Bertl sounds good, so maybe you should try it, and let me know what the 'result' is ... maybe with an strace or something like that, which shows what it tries and where it doesn't succeed ... 1119915379 M * Johnsie I'll do that. 1119915383 M * Johnsie I know he's plugging away at it now. 1119915394 M * Johnsie I leave that kind of work to the experts. 1119915396 M * Johnsie hahah 1119915412 M * Bertl maybe he should just fire up an irc client and join us then :) 1119915415 M * Johnsie Oh, thanks for the help about a month ago with the bind mount issue. 1119915435 M * Bertl you're welcome (if I was involved ;) 1119915451 M * Johnsie We went with the Gentoo 2.6.11.9-vs2.0-rc1 kernel...works great. 1119915460 M * Johnsie Yes, you were the gent I talked to. 1119915477 M * Johnsie I think we were using 1.9.4 before that. 1119915542 M * Johnsie Yeah, he comes by now and again...nickname "Beave". 1119915567 M * Bertl okay ... 1119915587 M * Johnsie Oh, we did manage to get Asterisk working too. 1119915609 M * Johnsie We had a problem where it wouldn't share the "ztdummy" device, which acts as a timing source... 1119915612 M * Bertl excellent, IIRC some folks (maybe you?) reported that some time ago 1119915621 M * Johnsie I may have, not sure. 1119915630 M * Johnsie My memory is like Swiss cheese. 1119915641 M * Bertl what are we talking about, again? 1119915650 M * Johnsie Well... 1119915674 M * Johnsie On Asterisk on a VPS environment, since it's one kernel, it wouldn't allow the use of the "ztdummy" timing source... 1119915683 M * Johnsie ztdummy is needed to run conferencing, music on hold, etc. 1119915708 M * Johnsie So, we wrote up a little script to trick it into working within a chroot environment... 1119915717 M * Johnsie Since it needs access to /dev stuff. :) 1119915726 M * Jani Is it possible to install a vserver with a Gentoo minial system? Like -d sarge -m ftp://ftp.debian.org/debian just for Gentoo? 1119915777 M * Bertl Jani: not that I know of, but I guess gentoo folks are working on something like this ... 1119915809 M * Bertl Jani: debian has the advantage of an existing tool (debootstrap) which does this ... 1119915853 M * Jani *hmms* 1119915864 M * Bertl Johnsie: yes? was it published? 1119915877 M * Johnsie Not yet? 1119915880 M * Johnsie I'm not sure. 1119915888 M * Johnsie That can be done, however. 1119915920 M * Johnsie Do you folks run like a collective vserver wiki or anything? 1119915938 M * Bertl yes, the main linux-vserver site is a wiki :) 1119915942 M * Johnsie Ooo. 1119915963 M * Johnsie Well if you have a section for such things, we'd be glad to tuck it in there. 1119915970 M * Bertl there are pages for stuff like application issues and such ... 1119915976 M * Johnsie Oh, excellent. 1119916038 M * Bertl http://linux-vserver.org/ProblematicPrograms 1119916042 M * Bertl (for example) 1119916606 A * Jani goes to bed now. "Good night alltogether." 1119916617 M * Bertl night Jani! 1119916654 Q * Jani Quit: 1119916755 J * shuri shuri@64.235.209.226 1119916782 M * Bertl evening shuri!